Interactions between large felids and humans in zoos: animal, keeper and visitor perspectives. Monika Szokalski. Doctor of Philosophy (Psychology)

Size: px
Start display at page:

Download "Interactions between large felids and humans in zoos: animal, keeper and visitor perspectives. Monika Szokalski. Doctor of Philosophy (Psychology)"

Transcription

1 Interactions between large felids and humans in zoos: animal, keeper and visitor perspectives By Monika Szokalski BPsych (Hons) A thesis submitted for the degree of Doctor of Philosophy (Psychology) School of Psychology, Social Work and Social Policy Division of Education, Arts and Social Sciences December 2014 Supervisor: Dr Carla Litchfield Associate supervisor: Dr Wendy Foster i

2 ii

3 Contents List of Figures List of Tables.... Glossary. Summary Publications Declaration... Acknowledgements... vi viii xi xiii xv xvi xvii Chapter One: Enrichment for large felids in zoos: Human interaction as a notable gap in the literature Introduction Literature search method Results Discussion of literature findings Conclusion Aims and structure of this thesis Chapter Two: The evolving large felid-human relationship: The need to promote positive interactions in zoos Introduction A snap-shot of the large felid-human relationship in early history Flashing forward to the modern large felid-human relationship: changes in the 20 th century Large felids in modern captive environments Conclusion Chapter Three: General methods and research site overview Introduction Exploring zoo keeper experiences with large felid interactions Exploring an interactive experience from animal perspectives Exploring an interactive experience from zoo visitor perspectives Study sites for studies Two and Three: Zoos South Australia Conclusion 140 Chapter Four: What can zoo keepers tell us about the handling and training of 141 iii

4 large felids in zoos? 4.1 Introduction Methods Results Discussion Conclusion 163 Chapter Five: The short-term behaviour of large felids during behind-thescenes zoo visitor tours: A case study Introduction Method Results Discussion Conclusion 201 Chapter Six: Exploring zoo visitor responses to interactive versus traditional large felid experiences Introduction Method Results Discussion Conclusion 255 Chapter Seven: Discussion and conclusion Summary of thesis Limitations to the scope of the thesis Implications for stakeholders and directions for future research Conclusion 276 References Appendix A Contribution of authors to each publication Appendix B Published version of Chapter One. 295 Appendix C Key methodological details of some past zoo animal studies Appendix D Key methodological details of some past zoo visitor studies 321 Appendix E Maps of Adelaide Zoo and Monarto Zoo Appendix F Published version of Chapter Four 332 iv

5 Appendix G Online questionnaire used in Chapter Four Appendix H Sample questionnaire from Chapter Four data Appendix I Published version of Chapter Five Appendix J Pooled data for analysis in Chapter Five 388 Appendix K Survey materials used in Chapter Six 392 Appendix L Pooled data for analysis in Chapter Six. 403 v

6 List of Figures Figure 1.1. A proposed model of human-animal interactions and their consequences for human-animal relationships in zoos 52 Figure 2.1. An image of a lion being released in to a King s animal park at Mesopotamia, featuring as part of a relief on a palace wall Figure 2.2. Image of tigers pulling a lion prince in a cart as part of one of Hagenbeck s circus acts.. 78 Figure 3.1. Top: Adelaide Zoo tiger exhibit (front view from main visitor viewing area); Bottom: Examples of Adelaide Zoo tiger exhibit signage Figure 3.2. Top: Adelaide Zoo lion exhibits (left one of the three smaller exhibits; right the larger exhibit); Middle and bottom: Examples of Adelaide Zoo lion exhibit signage Figure 3.3. Top: Monarto Zoo cheetah exhibit (view from visitor viewing platform); Bottom: Examples of Monarto Zoo cheetah exhibit signage Figure 3.4. Top: Monarto Zoo lion exhibit (front view from visitor viewing platform); Bottom: Examples of Monarto Zoo lion exhibit signage Figure 5.1. Mean percentage of scans in which the lions, cheetahs and tiger were observed in active species-typical behaviour on non-tour and tour days, in all time sessions (pre, during, and post). 180 Figure 5.2. Mean percentage of scans in which lions, cheetahs and tiger were observed in inactive species typical behaviour on non-tour and tour days, in all time sessions (pre, during, and post) Figure 5.3. Mean percentage of scans in which lions, cheetahs and tiger were observed in pacing behaviour on non-tour and tour days, in all time sessions (pre, during, and post) Figure 5.4. Mean percentage of scans in which lions, cheetahs and tiger were observed in each proximity category (close, moderate, distant) to humans (keepers and visitors) in the during tour session on tour days Figure 6.1. Open-ended knowledge responses per category made by post-tour and post-exhibit visitors, as a percentage of total comments vi

7 Figure 6.2. Open-ended emotional responses to seeing animals/s per category made by post-tour and post-exhibit visitors, as a percentage of total comments Figure 6.3. Open-ended emotional responses to learning about animals/s per category made by post-tour and post-exhibit visitors, as a percentage of total comments Figure 6.4. Mean frequency of self-reported performance of 10 conservation behaviours prior to their visit by pre-tour and pre-exhibit visitors. 237 Figure 6.5. Mean percentage of post-tour and post-exhibit visitors reporting willingness to engage in 10 conservation behaviours after their visit Figure 6.6. Open-ended suggestions per category made by post-tour and post exhibit visitors regarding ways to assist engagement in speciesspecific conservation behaviours, as a percentage of total comments 240 Figure 6.7. Open-ended responses per category made by post-tour and postexhibit visitors regarding the best aspect/s of their visit, as a percentage of total comments Figure 6.8. Open-ended responses per category made by post-tour and postexhibit visitors regarding the worst aspect/s of their visit, as a percentage of total comments. 243 Figure 6.9. Open-ended responses per category made by post-tour and postexhibit visitors regarding other desired aspect/s of their visit, as a percentage of total comments. 244 vii

8 List of Tables Table 1.1. Summary of some key similarities and differences between the seven large felid species Table 1.2. Search terms used to locate peer-reviewed journal articles related to enrichment for large felids in zoos 11 Table 1.3. Summary of key details of the 16 large felid enrichment studies included in this review (published in the peer-reviewed literature between ).. 13 Table 2.1. Summary of key details of behind-the-scenes large felid zoo visitor tours available in all Australian ZAA member organisations (as per individual zoo website details and correct as of July 2014). 82 Table 2.2. Summary of key details of large felid attacks in captivity (including zoos, circuses, exotic animal farms, and private ownership) derived from a selection of case reports in the peer reviewed, English-language medical literature (published between ). 90 Table 3.1. Summary of key methodological details of three peer-reviewed studies, published between , of keeper-animal interactions in zoos involving zoo keepers as participants Table 3.2. Summary of key methodological details of four peer-reviewed studies, published between , of captive dolphins involved in behind-the-scenes visitor tours Table 3.3. Summary of key methodological details of five peer-reviewed studies, published between , of the impact of zoo visitors on large felids 110 Table 3.4. Summary of key methodological details of three peer-reviewed studies, published between , of the impact participating in behind-the-scenes tours on zoo visitors 118 Table 3.5. Summary of key methodological details of nine peer-reviewed studies, published between , of the impact of participating in interactive/interpretive experiences on zoo visitors 121 Table 3.6. Summary of key details of tiger and lion exhibits, signage, keeper talks and behind-the-scenes tour at Adelaide Zoo 133 viii

9 Table 3.6. Summary of key details of cheetah and lion exhibits, signage, keeper talks and behind-the-scenes tours at Monarto Zoo Table 4.1. Self-reported rates of handling and training as a percentage of participants (n = 86) working with each of the seven large felid species Table 4.2. Percentage of participants (n = 86) who agreed that each type of handling and training is most beneficial for the animals, keepers, and visitors Table 4.3. Median (IQR) participant-reported benefit ratings of each handling practice (hands-on contact, protected contact, hands-off contact, and training) for each user group (large felids, keepers, and visitors). 150 Table 4.4. Categories of open-ended responses made by participants regarding the benefits and limitations of each handling practice and training for large felids 151 Table 4.5. Categories of open-ended responses made by participants regarding the benefits and limitations of each handling practice and training for zoo keepers 153 Table 4.6. Categories of open-ended responses made by participants regarding the benefits and limitations of each handling practice and training for zoo visitors 154 Table 5.1. Key background details of participants (n = 7) included in this study. 170 Table 5.2. The nine condensed behaviour categories (with details of included behaviours) used when conducting animal behavioural observations during this study 174 Table 5.3. Mean number of on-exhibit observation scans used in data analysis for each individual animal per observation condition (Pre, During, Post) on non-tour and tour days Table 5.4. Results of main and mixed effects of linear mixed models analysis for active species typical, inactive species typical, and pacing behaviours for lions and cheetahs 179 Table 5.5. Total number of observation sessions with at least one incident of Feeding, Aggression, Interact conspecific, Interact human, Other, and Out-of-sight behaviours observed in the lions, cheetahs and 185 ix

10 tiger across the three observation sessions (Pre, During, and Post) on non-tour (NTD) and tour days (TD).... Table 5.6. Results of main and mixed effects of linear mixed models analyses for proximity to humans for lions and cheetahs 186 Table 6.1. Participant response rate for each individual visitor condition in this study Table 6.2. Key demographic characteristics (n and %) and chi-square results of tests for differences between tour (n = 121), pre-exhibit (n = 164), and post-exhibit (n = 161) visitors Table 6.3. Mean ± standard deviation self-rated knowledge scores for participants in each condition Table 6.4. Mean ± standard deviation percentage of correct knowledge responses for participants in each condition each condition. 231 Table 6.5. Mean ± standard deviation attitude toward species scores for participants in each condition Table 6.6. Mean ± standard deviation on the three attitudes toward conservation items for participants in each condition Table 6.7. Mean ± standard deviation self-reported frequency of conservation behaviours (general and species specific) for pre-tour and pre-exhibit visitors Table 6.8. Mean ± standard deviation number of conservation behaviours (general and species specific) that post-tour and post-exhibit visitors proposed they will engage in after their visit 238 Table 6.9. Mean ± standard deviation expectation and evaluation ratings made by visitors in each condition. 240 x

11 Glossary [Note: some of the definitions included here have been developed for inclusion in studies conducted as part of this thesis, and published in some instances. If the definitions have been published, or adapted from another source, the source is cited.] Behind-the-scenes tour - an experience in which zoo visitors (usually) pay an additional fee to be taken on a tour of on- and/or off-exhibit animal enclosures (usually led zoo keepers and, possibly, zoo volunteers). Such experiences may include opportunities to feed animals and assist with other keeper duties (such as shifting animals to/from night quarters, and adding enrichment to enclosures). Environmental enrichment - the addition and/or removal of particular objects or stimuli from a zoo animal's environment, or any change to the structure of an animal's environment or husbandry practices, for the purpose of providing variability to the environment, increasing behavioural diversity, and/or reducing engagement in potentially negative behaviours and/or states (e.g., stereotypy) in order to improve physical fitness and psychological wellbeing (adapted from Mason et al., 2007; Swaisgood & Shepherdson, 2005). Hands-off contact - situations where there is no contact between animal and human (zoo keeper and/or zoo visitor), beyond that which is necessary, such as feeding and shifting between areas (e.g., enclosures) or viewing from public vantage points (Szokalski et al., 2013). Hands-on contact - physical contact between animal and human (zoo keeper and/or zoo visitor), with no protective barrier, including human/s entering animal enclosures with animal present, or taking the animal for walks (e.g., on leads/harnesses; Szokalski et al., 2013). Large felid - overarching term given to the seven largest extant species of the felidae family: the cheetah (Acinonyx jubatus), jaguar (Panthera onca), leopard (Panthera pardus), lion (Panthera leo), puma (Puma concolor), snow leopard (Panthera uncia), and tiger (Panthera tigris). Also known as big cats or large carnivores. xi

12 Natural - used to denote any factor (e.g., feeding routines, space requirements, plant substrates, presence of humans, etc.) associated with the environment from which a particular species originates (i.e., the 'wild'). Protected contact - contact between animal and human (zoo keeper and/or zoo visitor) with a protective barrier between the two (e.g., interacting through a mesh fence; Szokalski et al., 2013). Species-typical behaviour - the repertoire of behaviours characterising how a species behaves in its natural habitat. Also known as species-specific behaviour (adapted from Hosey et al., 2009). Stereotypic behaviour - a repetitive behaviour, which may result from frustration, attempts to cope with sub-optimal environments, or central nervous system dysfunction. Examples include pacing, swaying, and fur plucking (adapted from Mason, 2006). Training - any planned and targeted procedure whereby a zoo keeper encourages the performance of specific behaviours in an animal (through principles of operant conditioning, such as positive reinforcement; Szokalski et al., 2013). Zoological organisation (zoo) - an accredited animal keeping facility, housing a collection of animals, which is a member of an organisation (such as the Zoo and Aquarium Association [ZAA], the British and Irish Association of Zoos and Aquariums [BIAZA], and the Association of Zoos and Aquariums [AZA]), and has as its aims conservation, education, welfare, and/or husbandry (to the exclusion of other forms of captivity, such as private animal ownership and circuses; adapted from Hosey et al., 2009). xii

13 Summary In their natural habitats, relationships between large felids and humans are characterised by increasing conflict. In contrast, a rise in deliberate interactions between people and these animals has recently been seen in Western zoological organisations. In zoos, large felids are being handled in one of three ways: hands-on, protected and hands-off contact. They are also being trained, and are involved in various interactive experiences. By their nature, these animals are apex predators, designed specifically for hunting and consuming prey, and therefore pose realistic threats to people. Moreover, interacting with humans is largely unnatural for these predominantly solitary species. Despite this knowledge and the increasing occurrence of interactions in the zoo environment, there is a paucity of information about the influence of these interactions on those involved. This thesis addresses this void by exploring how large felids and humans might benefit from, or be adversely affected by, interacting with each other in the zoo environment. To do this, the thesis begins with a review of both the enrichment literature for large felids highlighting human interaction as an underdeveloped area of research and the historical relationship between these animals and humans. A series of three studies then follows, exploring this issue from the perspective of zoo keepers, zoo animals and zoo visitors. Given the lack of information on this topic, Study One involved an online, worldwide survey of zoo keepers, exploring the types of handling they practice with their large felids, whether they practice training, and what their opinions of these different methods are. Findings revealed that protected contact was the most frequently used handling method, particularly with lions, tigers and cheetahs, and training was practiced by the majority of participants, with all large felid species. Consistent with this pattern of use, these practices were reported as being most beneficial for keepers, animals, and visitors alike. In contrast, concerns were raised regarding the potential for danger and for wrong messages to be sent to zoo visitors when hands-on contact is used. Interactions between zoo keepers and large felids make possible a range of interactive experiences for zoo visitors. As such, the remaining two studies in this thesis explored the responses of large felids and visitors to one form of interaction behind-the- xiii

14 scenes tours at Zoos South Australia. To do this, Study Two explored the behaviour of three groups of large felids lions, cheetahs, and tiger before, during and after zoo visitor tours, as well as on non-tour days. The findings revealed some changes in behaviour particularly inactivity and pacing in association with the tours, but these were not necessarily indicative of compromised well-being. To complement this, Study Three surveyed visitors who participated in these tours, compared to visitors who only viewed the animal exhibits. The study examined changes and differences in knowledge, attitudes, emotions and proposed conservation behaviour among these visitors. Overall, the data revealed that behind-the-scenes tours are meeting the Zoo s aim of providing a more educational, emotionally engaging, and conservationinspiring visitor experience. Combined, the results of these studies are discussed in light of the benefits and limitations of large felid-human interactions in zoological organisations. Suggestions for how these findings can be utilised by other facilities worldwide are discussed, along with directions for future research. xiv

15 Publications The list of publications resulting from this thesis includes: Szokalski, M.S., Litchfield, C.A., & Foster, W.K. (2012). Enrichment for captive tigers (Panthera tigris): Current knowledge and future directions. Applied Animal Behaviour Science, 139, 1-9. Szokalski, M.S., Foster, W. K., & Litchfield, C.A. (2013). Behavioral monitoring of big cats involved in behind-the-scenes zoo visitor tours. International Journal of Comparative Psychology, 26, Szokalski, M.S., Litchfield, C.A., & Foster, W.K. (2013). What can zookeepers tell us about interacting with big cats in captivity? Zoo Biology, 32, The following presentation was given based on this thesis: Szokalski, M.S. (2011). My PhD experience at Zoos South Australia: Exploring human-big cat interactions in captivity. Australian Veterinary Association Annual Conference, May, Adelaide, Australia. xv

16 Declaration This thesis presents work carried out by myself and does not incorporate without acknowledgment any material previously submitted for a degree or diploma in any university; to the best of my knowledge it does not contain any materials previously published or written by another person except where due reference is made in the text; and all substantive contributions by others to the work presented, including jointly authored publications, are clearly acknowledged. The contribution of each author to each of the publications generated from this PhD can be found in Appendix A. Monika Sofie Szokalski December 2014 xvi

17 Acknowledgments There are various people to whom I owe thanks for the help they have provided throughout my research journey. Firstly, I must thank my two inspiring supervisors, Dr Carla Litchfield and Dr Wendy Foster. Carla thank you for giving me this opportunity, and for always welcoming my ideas with support and encouragement. I am extremely grateful to have been given the freedom to explore an issue that I am so curious and passionate about. Wendy thank you for being an endless source of information and for providing me with an insight to the zoo world that would not have otherwise been possible. Combined, you have both taught me so much. This research would not have been achievable without the help and support of the Royal Zoological Society of South Australia and, in particular, the carnivore keepers. To Jason Hakof, Anna Bennett, Michelle Lloyd, and the rest of their teams thank you for encouraging this research and providing support when needed. Similarly, I wish to thank the visitor centre staff at Monarto Zoo and the volunteer staff at both locations, who not only assisted with data collection for one of my studies, but provided me with insightful conversations during my time at the sites. I am particularly grateful to the countless participants in my research - to the zoo keepers who took the time out of their busy schedules to complete my survey, and to the zoo visitors who were willing to forfeit some of their leisure time to contribute to this research. I owe equal thanks to the beautiful animals I was fortunate enough to observe during this research. It was an absolute privilege and a humbling experience to spend so much time in their presence. I owe huge thanks to my fellow PhD candidates Hayley Tindle, Dr Elissa Pearson, Jillian Ryan, Dr Renae Hayward, Amy Reynolds, and Dr Bradley Smith. At various stages, and in your own ways, you have provided me with more support, motivation and encouragement than I could have asked for. It has been a pleasure to work with each of you and to watch you succeed. I must also thank some of the academic and support staff in the School of Psychology, Social Work and Social Policy at the University of South Australia, who have provided me with casual teaching work, encouraged my research, and assisted with administration tasks - Amanda McGroder, xvii

18 Lorelie Ball, Mark Kohler, Cathy Balfour, Janet Bryan, and Rob Ranzijn. I would also like to thank the staff in the interlibrary loans department for the various sources that they have speedily tracked down for me over the years. To end, I must thank my family and friends for their continued support. To my Mum, Bron, and siblings, Hannah and Eddie thank you for always encouraging my academic endeavours and for providing me with a happy balance to my research work. Lastly, I cannot express more thanks to my ever-supportive husband, Ty Ferguson. Thank you for sharing a genuine interest in my research, for being proud of my efforts, and for always being willing to assist wherever possible. There have been plenty of days that would have been far harder without your tireless positive encouragement. I am so fortunate to have someone who loves animals and zoos the way I do sharing this research experience with you has made it all the more enjoyable. xviii

19 1 Enrichment for large felids in zoos: Human interaction as a notable gap in the literature 1 in order to significantly improve current enrichment practices, and consequently the welfare of captive animals, it is necessary to quantify current practices. J.M. Hoy, P.J. Murray, and A. Tribe (2009, p. 2) The rationale for this thesis to explore large felid-human interactions in zoological organisations emerged from the recognition that these practices are increasing, and are often argued to be enriching for the animals involved. However, compared to other forms of enrichment and husbandry practices, this topic has featured less in the scientific literature for these species. As such, this chapter will introduce the seven large felid species and provide a review of what is known to date about improving their welfare, through environmental enrichment, in the zoo environment. It will lead to the conclusion that, although increasingly occurring, we currently understand little about how interactive practices with humans might influence the well-being of these animals Introduction Large felids, or big cats as they are more commonly known, is the overarching term given to the seven largest extant species of the felidae family: the cheetah (Acinonyx jubatus), jaguar (Panthera onca), leopard (Panthera pardus), lion (Panthera leo), puma (Puma concolor), snow leopard (Panthera uncia), and tiger (Panthera tigris). In their natural home ranges, these animals are found on all continents except Australasia and Antarctica, occupying a variety of habitats and differing in typical home range size (see 1 An adaptation of this chapter has been published as: Szokalski, Litchfield and Foster (2012). Enrichment for captive tigers (Panthera tigris): Current knowledge and future directions. Applied Animal Behaviour Science, 139, 1-9. See Appendix B. 1

20 Table 1.1 for a comparison of key features for each species). Despite some differences in physical colouring and size, these species have a relatively similar appearance and share core behaviour traits (Guggisberg, 1975). Whilst also differing in conservation status (from the puma being of least concern, to the critically endangered Sumatran tiger [ssp. sumatrae]) these animals are all facing varying degrees of conflict with humans, particularly over natural resources. Ultimately, human activity is responsible for their decline (Loveridge, Wang, Frank, & Seidensticker, 2010). In addition to in situ efforts, zoos and wildlife parks play a crucial role in the conservation of these animals. Although it is not known exactly how many of these animals are currently housed in zoos worldwide, with some felid species believed to exist in higher numbers in captivity (zoos and other forms of exotic animal ownership) than in their natural home ranges (such as the tiger; Nyhus, Tilson, & Hutchins, 2010), it is timely to explore what is known to date about the best practices for enriching the lives of these animals in zoos. This will assist with moving towards a better understanding of how to optimise their well-being in this environment and to identify areas for future research. This is pertinent, given that zoos may likely be their only safeguard from extinction in the future The importance of optimal zoo environments Recently, there has been a greater focus on the role that zoos can play in animal conservation, particularly through maintaining collections of animals (Conde, Flesness, Colchero, Jones, & Scheuerlein, 2011; Hutchins & Smith, 2003; Kirkwood, 2003; Rustiati & Chanin, 2000; Tribe & Booth, 2003). While the zoo environment may protect against many of the factors contributing to species declines in natural habitats (e.g., poachers), it may not be an ideal environment for all animals. Therefore, it is important to consider the impact of existing in this environment on the behaviour of animals for two key reasons. Firstly, given that natural selection may engineer a population to fit its captive environment, ensuring minimal genetic change in zoohoused animals is an important consideration (Kirwood, 2003; Mason, Clubb, Latham & Vickery, 2007). Failure to do this may result in captive-adapted strains of animals that are potentially less suited to the natural (i.e., 'wild') environment, creating significant problems if zoo populations are used as behavioural models of wild 2

21 Table 1.1. Summary of some key similarities and differences between the seven large felid species (adapted from IUCN, 2012; Macdonald, Loveridge, & Nowell, 2010; Macdonald, Mosser, & Gittleman, 2010). Species Region Mean home range (km 2 ) Cheetah Africa/Asia M = * F = * Jaguar Americas M = 79.0 F = 50.0 Leopard Africa/Asia M = F = Lion Africa/Asia M = 82.7* F = 82.7* Puma Americas M = 83.0 F = 29.0 Snow leopard Asia M = F = Tiger Asia M = 67.0* F = 21.0* Mean weight (kg) M = 41.4 F = 35.9 M = F = 66.9 M = 53.0 F = 30.5 M = 189.6* F = 117.7* M = 58.9 F = 30.7 M = 40.7 F = 34.3 M = F = Mean head/body length (mm) M = 1225 F = 1245 M = 1565 F = 1304 M = 1340 F = 1143 M = 1949* F = 1711* M = 1347 F = 1179 M = 1210 F = 1173 M = 2300 F = 1663 Sociality Typical prey Conservation status: major threats Solitary (some M = social) Small-medium ungulates Vulnerable: Habitat loss and fragmentation; conflict with humans; depleting prey. Note that two subspecies (Northwest African and Asiatic cheetah) are critically endangered. Solitary Large ungulates Near threatened: Persecution by humans over livestock predation; competition with humans over prey Solitary Ungulates Near threatened: Habitat loss; depleting prey; Social Medium-large ungulates Solitary Small-medium ungulates illegal skin/body part trade; persecution by humans over livestock predation. Note that three subspecies (Javan, Arabian, and Amur leopard) are critically endangered. Vulnerable: Conflict with humans over life and livestock. Least concern: Habitat loss/fragmentation, human poaching of prey base; persecution by humans over livestock predation Solitary Ungulates Endangered: Depleting prey; illegal skin/bone trade; conflict with local humans; lack of conservation capacity, policy and awareness Solitary Large ungulates Endangered: Depleting prey; illegal poaching; Note: M = male; F = female; * indicates maximum figures found (e.g., for lion weights, findings were given for populations in both eastern and southern Africa). home range loss; conflict with farmers. Note two subspecies (South China and Sumatran tiger) are critically endangered. 3

22 counterparts, if animals are part of zoo-breeding programs, or if they are to be reintroduced to the wild (Kirkwood, 2003; Mason et al., 2007). Taking a simplistic example, in their wild habitats, animals must acquire their own food (whether directly as individuals or through communal hunting/foraging/consumption) in order to survive, yet this is often provided for in zoos (although the method of provision will often vary), which may mean that many food-acquisition behaviours are not required. If repetitive generations of a species are housed in zoos and lack the opportunity to perform food-acquisition behaviours, it could be the case that such behaviours become lost from the animal's repertoire (e.g., through physiological changes, and motivation), which has obvious consequences for survival if animals of this species are released from zoos. However, this concern is grounded in the notion that zoos exist to conserve species from extinction. Counter to this argument, from the perspective of individual animal welfare, genetic change may not be problematic as it may indicate that an animal/group of animals has experienced such changes by way of adaptation to the zoo environment. To the best of this author's knowledge, it is usually not the case that zoo-housed large felids are released to their natural habitat and therefore such an issue is unlikely to be problematic for the individual animal. In this sense, and if the purpose of zoos is to provide optimal welfare for the animals in their collections, then behavioural and genetic change may actually be beneficial to optimal welfare in this environment. Moreover, with some species now extinct in the wild, if their only 'habitat' in the future is a zoo setting, then adapting to this environment is beneficial beyond the individual animal. Therefore, the second reason for needing to preserve natural behaviours is out of consideration for an animal s psychological well-being (Hutchins & Smith, 2003). This concern is based on the premise that animals have the capacity for conscious awareness of pleasant and/or unpleasant feelings, and that they can therefore suffer when welfare is poor (Kirkwood, 2003). Behaviour can be an indicator of the presence of these feelings and responses to the zoo environment, and it is thus frequently employed as a means to assess and identify both positive and diminished 4

23 well-being (Hill & Broom, 2009). Positive well-being may be demonstrated when an animal engages species-typical behaviours (that is, in the repertoire of behaviours characterising how a species behaves in its natural habitat; Hosey et al., 2013). The presence of these behaviours may indicate that the zoo environment is catering for the needs of the animal to the extent that it allows opportunities to engage in these behaviours, despite the difference in environments. Conversely, diminished wellbeing is often, but not exclusively, assessed through explorations of stereotypic behaviour Stereotypic behaviour in zoos While some animals may adjust well to zoo-living, the heavy focus in past literature on ways to improve the well-being of zoo animals suggests that various animals experience (or at least are perceived to experience) negative impacts associated with existing in zoos. These negative effects of zoo-living have been widely acknowledged (Mason, 2010). Among other measures, such as reproductive success, there is a general consensus that the limitations of a zoo environment can lead to the performance of stereotypic behaviour in many species (Mason et al., 2007; Mellen & Sevenich MacPhee, 2001; Shyne, 2006; Swaisgood & Shepherdson, 2005). Mason (2006) suggests that stereotypic behaviour is repetitive behaviour resulting from: frustration (either due to internal states induced by the captive environment and/or external cues); attempts to cope; and/or central nervous system dysfunction (based on a past, early rearing environment). Common examples of such behaviours across species include pacing, swaying and fur plucking. While species, and indeed individuals within species, will likely differ in their expression of these behaviours, there is some evidence to suggest that large felids may be more prone to such behaviour than other species. In particular, Clubb and Mason (2007) highlight that those carnivores which typically occupy large ranges in the wild tend to fare worse in zoos and are much more vulnerable to welfare problems. Large felids generally occupy large home ranges (see Table 1.1) and a widespread problem facing zoos is the presence of stereotypic behaviour in these animals (Mason et al., 2007). In combining data from numerous observational studies, Clubb and Mason (2007) report varying rates of stereotypic pacing (defined as, repetitive walking movement along a 5

24 particular path, for no apparent purpose) for large felids, with it being the sole stereotypic behaviour observed in these animals: 7% for snow leopards, 11% for leopards, 12% for pumas, 16% for tigers, 21% for jaguars, 25% for cheetahs, and 48% for lions Knowledge of wild requirements and contrasts in zoos While it is recognised that wild and captive settings (including zoos) can be considered to exist along a continuum, in terms of their features and the impact these might have on animals, there are some ways in which zoo environments differ notably from the natural habitats in which large felids are found. It is also acknowledged that there are some ways in which captive environments may be considered "better" for individual animals, such as protection against poachers, readily available food, and health care monitoring. As such, although the wild may not always be an optimal environment, it can provide a guide for understanding the types and diversity of behaviours engaged in by a particular species, and the stimuli associated with these. It can also provide some indicators for the type of resources a species requires (e.g., food-related, spatial needs, etc.). As indicated in Table 1.1, most large felids are solitary species (with the exception of lions living in prides and some male cheetahs living in coalitions) and occupy large home ranges in their natural habitats. Although there is some variation between species in daily activity and hunting pattern, most of these felids tend to be inactive during the day, but will also hunt opportunistically at any time (Sunquist & Sunquist, 2002). These animals are carnivorous, typically requiring a stable population of smalllarge sized ungulate prey (Macdonald, Loveridge, & Nowell, 2010; Macdonald, Mosser, & Gittleman, 2010). In zoos, however, the environment for large felids is vastly different in these regards. Large felids are often housed with, or in close proximity to, conspecifics (as well as heterospecifics), and spatial and financial constraints appear to restrict enclosure size, resulting in significantly smaller territories in which to exist. Further, the absence of a live prey base and, instead, the provision of pre-prepared diets often fails to allow for the performance of their natural 6

25 hunting behaviours and energy expenditure (see Section 1.4). Lastly, while this is likely to vary between zoos, large felids may be expected to be active and visible during zoo opening hours when visitors are present (as evidenced by attempts to encourage active behaviour through the provision of enrichment devices on-exhibit during these hours). This is problematic since these species are often inactive in zoos, spending much of their time either out of visibility or resting/sleeping (Powell, 1995; Van Metter et al., 2008). Although inactive behaviours are part of a large felid s species-typical behavioural repertoire, it may be considered to detract from the zoo visitor experience (i.e., the animal may be perceived as boring ) and therefore zoos have an obligation to address this by encouraging active behaviour during visitation hours. There are some additional features of zoos that contribute to the unique nature of this environment. Notably, the zoo environment is one that is characterised by a large number of unfamiliar humans (Hosey, 2005). The relatively constant presence, and (in many instances) very close proximity, of people (zoo keepers and zoo visitors) is different from what these animals experience in their habitat countries ranges (discussed throughout Chapter Two). Additionally, by the very nature of captivity (that is, to keep animals enclosed in spaces they cannot move from, and under routines that are largely determined by people), these and other animals are likely to lack the control to behave and engage in activities with the freedom with which they would in their natural habitats (of course, there is consideration for this not always being the case, with people encroaching on natural habitats, poaching animals, etc.). Given these differences, which may restrict engagement in species-typical behaviours, it is not surprising that these animals may display stereotypic behaviour in the zoo environment (for a more detailed discussion of the impact of such factors, see Morgan & Tromborg, 2007). However, it is recognised that other factors such as an individual animal s personality and history can also contribute to these behaviours (Hill & Broom, 2009). 7

26 Environmental enrichment Stereotypic behaviour in zoo-housed large felids has not been ignored by the zoo community, and numerous efforts have been employed to not only reduce this behaviour but also to encourage greater engagement in species-typical behaviours. It appears that the primary means for doing so has been through use of environmental enrichment, an approach which has been well-established as successful in improving animal well-being (for more in-depth coverage of enrichment success, see de Azevedo, Cipreste, & Young, 2007; Swaisgood & Shepherdson, 2005). Environmental enrichment has been defined in a variety of ways, and its specific aims tend to vary depending on the context and species with which it is explored. However, some central tenets of this approach include its overarching aim to provide variability to an animal's environment by providing behavioural opportunities that were not previously available (Swaisgood & Shepherdson, 2005). In doing so, it often aims to increase animal engagement in behaviours displayed in their natural environments (but which may not have otherwise been necessary in the zoo environment), to increase behavioural diversity (that is, the range of behaviours engaged in), and to reduce engagement in potentially negative behaviours and states, such as stereotypy and boredom (Mason et al., 2007). Enrichment may also aim to increase an animal's use of its environment/space (Young, 2003). As such, enrichment targets both physical and psychological well-being. Usually, enrichment is achieved through the addition and/or removal of particular objects or stimuli from an animal's environment, and/or by making changes to husbandry practices (Mason et al., 2007; Swaisgood & Shepherdson, 2005). As such, there are numerous types of enrichment and these can be categorised in various ways. Using two relatively recent and comprehensive reviews of enrichment literature (Swaisgood & Shepherdson, 2005) and practices (Hoy et al., 2009), enrichment can be thought of as including (but not limited to) the following categories: feeding (e.g., manipulation of food provision), novel objects (e.g., provision of 'toys'), sensory (including auditory, olfactory and visual; e.g., addition of natural or artificial odours), structural/environmental (e.g., altering a physical aspect of the enclosure), enclosure rotations (e.g., shifting animals between enclosures), social (e.g., alterations to group 8

27 composition), and human-animal interactions (e.g., positive reinforcement training). In this way, enrichment does not necessarily include items or stimuli found in natural home ranges, but rather those which can encourage engagement in species-typical behaviours Environmental enrichment for large felids Past reviews of enrichment literature encompassing various species tend to suggest that large felids may be quite well-represented in this body of research. For example, in a review of zoo-based enrichment studies (n = 25) published in Zoo Biology, Applied Animal Behaviour Science and Animal Welfare between 1990 and 2003, Swaisgood and Shepherdson (2005) found the felid family to be the most frequently represented (n = 9). Similarly, de Azevedo et al. (2007) explored the enrichment literature (n = 744) available on The Web of Science database, between 1985 to 2004, across a range of animal housing settings (including laboratories, farms, and zoos), finding that carnivores were the fourth most commonly studied taxamonic group (n = 41). While no specific data were presented, it was stated that carnivores were the most represented animals within the zoo based literature and, although this information is not specific to large felids, when coupled with Swaisgood and Shepherdson's (2005) finding, it provides an additional indication of their representation in zoo enrichment literature. Finally, in an international (13 countries), multi-institutional (60 zoos) survey of individuals working with zoo animals (n = 238), Hoy et al. (2009) found that carnivores were the second highest group with regard to quantity of enrichment provided per day (26.4% of respondents), second to primates (50.6%). This adds further support to the importance of studying the outcomes of enrichment with these species, given that they may receive enrichment quite frequently. While none of these data are specific to large felids per se, these findings provide some evidence to indicate that there may exist a moderately sized body of evidence regarding the provision of enrichment and evaluation of its impact with regard to these animals. Based on what is known about how frequently these animals may display stereotypic behaviour (i.e., pacing), and what this behaviour potentially represents (i.e., compromised welfare; see Section 1.1.2) it appears that such studies are warranted and that synthesis of this information is needed. 9

28 Aim of this literature review Given the importance of improving the well-being of zoo-housed large felids, the purpose of this review was to explore what is presently known in the peer-reviewed literature about environmental enrichment for these animals, with the aim of synthesising this information to identify gaps in our knowledge and potential areas for future research Literature search method The literature included in this review was collected cumulatively throughout this researcher's candidature ( ). Strategies for accessing peer-reviewed literature included searching common zoo journals, such as Zoo Biology, Applied Animal Behaviour Science, Animal Welfare, Applied Animal Welfare Science, as well as databases such as Google Scholar and PsycArticles. Different combinations of a variety of search terms were used to identify potentially relevant articles (for common search terms, see Table 1.2). No limits were placed on the dates of publication and, where necessary (i.e., in databases only, not journal websites), results were restricted to peer-reviewed studies, as this review sought to include those of highest academic quality and rigour 2. Peer-reviewed studies were also located by manually searching the reference lists of those studies found via the previously described searches. Studies were included in this review if they met the inclusion criteria of involving at least one large felid species (i.e., multi-species studies could be included), at least one enrichment technique (i.e., multi-enrichment technique studies could be included), and were conducted in a zoo (i.e., studies of circus, laboratory, or privately owned felids were excluded). Only original research reports were included (i.e., review articles were excluded). Studies which investigate large felid welfare (e.g., the 2 Information from the "grey" literature - conference papers and anecdotal reports - are not the primary target of this review, however some of this information has been used to complement the discussion of the scientific literature. Like the scientific literature, such information was accessed cumulatively over this researcher's candidature, particularly through Google searches, reference lists of peer-reviewed studies, and journals such as The Shape of Enrichment. 10

29 influence of zoo factors on welfare/well-being) were excluded, unless the author/s specifically describe their research as enrichment-focused. Potential articles were screened by reading the title and/or abstracts, and many were excluded based on this information alone. In instances where more information was required, full-text versions of the articles were read before a final decision regarding inclusion/exclusions was made. Table 1.2. Search terms used to locate peer-reviewed journal articles related to enrichment for large felids in zoos Search element Animal/s Enrichment Zoo Welfare Search terms large felid*, big cat*, feli*, tiger*, lion*, cheetah*, snow leopard*, leopard*, jaguar*, puma*, mountain lion* and/or cougar* enrich*, feed*, food, scen*, exhibit*, novel object*, sensory, human*, keeper*, visitor* and/or enclosure* zoo* and/or captiv* stereotyp*, well-being, welf*, species-typical, behavio*, and/or husbandry Critical appraisal process Articles meeting the inclusion criteria (see Section 1.2.) were first read repeatedly for clarity of topic and understanding. Next, all articles were subjected to a critical appraisal process. After searching previous zoo-based review articles, no specific critical appraisal tools relevant to this field of research were identified. Critical appraisal tools widely used in other fields, such as health sciences, were considered; however, no single tool was considered appropriate (given that many of these tools focus on specific study designs, e.g. randomized control trials, and include appraisal elements that are less relevant to the present body of research). As such, the critical appraisal process used here was adapted based on combining key appraisal elements from: (1) Swaisgood and Shepherdson's (2005) enrichment literature review; and (2) common criteria used across the Critical Appraisal Skills Program's (CASP, 2013) suite of critical appraisal tools, commonly used in the health science disciplines. This resulted in the following information being collected and documented for each included article: (1) sample size and species of animal/s (recorded as the number of animals included in data analysis rather than the total sample size of the study, as not 11

30 all studies included data from all animals observed; also recorded only for the large felid species as some studies also included other species), as well as study location (including whether it was a single or multi-site study); (2) enrichment type (e.g., feeding, olfactory, etc.); (3) study design (within-groups: before/after - AB, or repeated treatment - ABA, ABAB, etc.; between-groups; correlational); (4) measures of enrichment success (e.g., behaviour), as well as reliability of measures; (5) data collection period (duration of study period and duration of conditions); and (6) data analysis (type of analysis and justification for use) Results A total of 16 peer-reviewed articles, published between 1990 and 2011, were located for inclusion in this review 3. Key details of these studies are presented in Table 1.3 (in instances where multiple enrichment strategies are tested within a single article, the article has been categorised and summarised separately for each strategy). 3 Since this was not indented to be a systematic review, and the literature base was added to over a number of years, it is not possible to provide data on how many potentially relevant studies were located, nor the proportion of the final number of included studies in comparison to potential studies. 12

31 findings # Study Table 1.3. Summary of key details of the 16 large felid enrichment studies included in this review (published in the peer-reviewed literature between ) Enrichment category Enrichment type Sample (zoo/s/ region/s) Novel food Carcasses 1.4 cheetah (San Diego Wild Animal Park/USA) Carcasses 1.2 leopard 1.1 lion 1.3 snow leopard (Toledo, Potawatomi & Binder Park zoos/usa) Live fish & bones 1.2 lion 1.1 tiger (Zoo Atlanta/USA) Experimental design and condition details Between-groups: (1) Ground meat commercial diet (n = 2) (2) Partial or whole carcasses (n = 3) ABA ( 7 reps/animal): (A) Baseline: no enrichment, off-exhibit (1 day, before first enrichment only) (B) Enrichment: calf carcasses, off-exhibit (1 day) (A) Baseline: no-enrichment, on-exhibit Lions - AB (on- and offexhibit): (A) Baseline: no enrichment (4wks) (B) Enrichment: bones (2x/wk for 4wks) Tigers - ABBA (on-exhibit): (A) Baseline: no enrichment (B) Enrichment: bones (2x/wk for 4wks) (B) Enrichment: fish (2x/wk for 4wks) (A) Baseline: no enrichment (4wks) Approx. study duration (months) & data collection period/s Study duration: 2mths Data collection: 31 randomly chosen feeding days (duration/frequency of obs = NS) Study duration: 3.5mths Data collection: 2hrs offexhibit baseline (5min intervals/animal); 2hrs/day during feeding on enrichment days (5min intervals/animal); 4x/day (5hrs) on post-enrichment baseline days (1min intervals/ 10mins/exhibit) (total obs days/condition = NS) Study period: 4mths Data collection: data collected during 3 different time periods/day; 6 x 1hr obs/animal/condition (and /exhibit for lions) (1min intervals); total 540hr data Outcome measures Duration and/or frequency of various feeding behaviours Proportions of behaviour: Feeding Natural Stereotypic behaviour Inactive Active Hiding Proportions of behaviour: Pacing Non-stereotypic activity (locomoting or active) Resting Consumption Standing Not visible pacing Key behaviour speciesspecific details N/A Yes* Bond & Lindenburg (1990) Yes* Yes* McPhee (2002) Yes* Yes Bashaw et al. (2003) 13

32 findings # Study Table 1.3. Continued Enrichment category Feeding routines Enrichment type Frozen fish & bones (jaguar fish only) 'Gorge and fast' feeding Temporal and spatial feeding variation on a randomized schedule Temporal & spatial variation of food presentation on fixed, continuous schedule Sample (zoo/s/ region/s) cheetah 1.0 jaguar lion puma 1.2 tiger (Montgomery Zoo/USA) 1.4 lions (Topeka Zoo/USA) cheetah (Fota Wildlife Park/Ireland) 2.10 cheetah (FotaWildlife Park/Ireland) Experimental design and condition details ABA: (A) Baseline: 4wks (B) Enrichment: fish or bones (jaguar = fish only) daily for 4-8 consecutive days (A) Baseline: 1 day (7 days post-enrichment) AB: (A) Baseline: predictable, fixed, small, feed 6x/week for 2wks (B) Enrichment: greater proportions of food offered less predictably and frequently 3x/wk for 2wks AB (with enrichment only): (A) Starve days (B) Feed days AB: (A) Baseline: 8 days no enrichment (B) Enrichment: temporal feeding (time of provision) or spatial feeding (place of provision) 8x repetitions of each ABA: (A) Baseline: no enrichment (20 days) (B) Enrichment: temporal feeding (time of provision) or spatial feeding (place of provision) for 5 days (A) Baseline: no enrichment (14 days) Approx. study duration (months) & data collection period/s Study duration: 3mths Data collection: all animals - 2x 30min obs (1min intervals) for 10 days baseline; 1 st, 2 nd and last day of enrichment; 1 day postenrichment Study duration: 2.5mths (4yrs for weight loss) Data collection: 15min continuous obs per animal throughout day for 10-20x/wk for the first and last 2wks of the study period Study duration: 40 days Data collection: 4hrs obs/exhibit/day (5min intervals); 8 days/baseline, 8 days/enrichment type Study duration: 5mths Data collection: 5days obs/condition - 2hr 25min obs/exhibit/condition (5min intervals) Outcome measures Proportions of behaviour: Active Inactive Stereotypic Proportions of behaviour: Active Stand Rest Pace Not visible Digestibility Weight loss Proportions of behaviour: Locomotion Vigilance Other Exploratory Inactive Pacing Proportions of behaviour: Locomotion Vigilance Other Exploratory Inactive Pacing pacing Key behaviour speciesspecific details Yes* Yes* Skibiel et al. (2007)** Yes* Yes* Altman et al. (2005) Yes* Yes* Quirke & O'Riordan (2011a)** Yes Yes Quirke & O Riordan (2011b)** 14

33 findings # Study Table 1.3. Continued Enrichment category Food presentation Enrichment type Feeding boxes Meat hidden in hessian sack Sample (zoo/s/ region/s) 1.1 tiger (Zurich Zoo/Switzerland) 2.1 jaguar (Marwell Zoological Park/UK) Experimental design and condition details Part 1 - Solitary housing AB1B2 (3 consecutive days/condition; 8x replications): (A) Baseline: conventional feeding (incl. 1 stave day) (B1) Conventional feeding (B2) Enrichment: Box feeding Part 2 - Pair housing AAB1AAB2(1 day per condition; 8 replications, with box/conventional feeding reversed for each) (A) Baseline: conventional feeding (A) No feed day (B1) Box feeding (A) Baseline (A) No feed day (B2) Conventional feeding ABA (6 consecutive days/condition): (A) Baseline: no enrichment (B) Feeding enrichment (A) Baseline: no enrichment, one week post-enrichment Approx. study duration (months) & data collection period/s Part 1 - Solitary housing Study duration: 5mths Data collection: No data collected during first baseline; 6hr obs (continuous) on 3 rd day of second baseline and enrichment phases/animal (8x 9day repetitions) Part 2 - pair housing Study duration: 2mths Data collection: No data collected during baseline or no food days; 5.5hr obs (continuous) on conventional and box feeding days (8x 6day repetitions) Study duration: 2mths Data collection: 4.5hr obs/animal/day for 6days/condition (15s intervals) Outcome measures Durations of behaviour: Stereotyped pacing Locomotion Resting Sleeping Others Latency to open boxes Duration and frequencies of behaviour: Rest Pace Groom Move Stationary Desirable Other pacing Key behaviour speciesspecific details Yes* Yes* Jenny & Schmid (2002) Yes* Yes* Charlton (1998) 15

34 findings # Study Table 1.3. Continued Enrichment category Enrichment type Novel objects Metal kegs (tigers only), cardboard boxes (lions only), frozen blood balls Sample (zoo/s/ region/s) Moving bait 1.1 cheetah (Scottish National Zoological Park, UK) Frozen fish balls and hanging logs 2.0 lion 1.3 tiger (Smithsonian National Zoological Park/USA) 3.1 lion (NS/USA) Experimental design and condition details ABA (10 consecutive days/condition): (A) Baseline: whole rabbit feeding (B) Enrichment device containing whole rabbit (A) Baseline: whole rabbit ABA (repeated 3x/enrichment type - random order): (A) Baseline: 30-60min preenrichment (B) Enrichment (A) Baseline: 1 day postenrichment AB: (A) Baseline (no enrichment) (B) Enrichment or log (log only provided to 1.1 lions) Approx. study duration (months) & data collection period/s Study duration: 2mths Data collection: Obs (30s intervals) 1hr pre-feed, 20mins during feed, and 1hr post-feed for all conditions (10 days during baseline and enrichment each; 5days post-enrichment baseline) Study duration: 2.5mths Data collection: 30min obs/animal/condition (30s intervals); 3x repetitions of obs for each enrichment Study duration: 1mth Data collection: 2x 30min obs/day (20s intervals) (1baseline; 1enrichment); total 24 days obs total (number of days per enrichment type = NS) Outcome measures Proportions of behaviour: Sleeping Resting Resting alert Standing Walking Sprinting Playing Feeding Grooming Affiliation Sniffing Observation Other Frequency of sprinting Enclosure use Proportions of behaviour: Locomotion Rest Sit Sleep Swim One-zero occurrence: Discrete behaviours (e.g., climb, defecate) Object-directed behaviours (e.g, bite, carry) Behavioural diversity indices Proportions of behaviour: Resting Alert Standing/Locomoting Not visible Occurrence of behaviour: Licking/gnawing Paw manipulation Sniffing/flehmen Face rubbing/back roll Enclosure use pacing Key behaviour speciesspecific details N/A Yes* Williams et al. (1996) N/A Yes* (but analysed with scent enrichment) Van Metter et al. (2008)** N/A Yes* Powell (1995)** 16

35 findings # Study Table 1.3. Continued Enrichment category Enrichment type Olfactory Oryx faeces on randomized schedule Oryx faeces on fixed, continuous schedule Spices (cinnamon, chilli powder & cumin) Spices (musk cologne, peppermint extract, allspice & almond extract) Sample (zoo/s/ region/s) cheetah (Fota Wildlife Park/Ireland) 2.10 cheetah (FotaWildlife Park/Ireland) cheetah 1.0 jaguar lion puma 1.2 tiger (Montgomery Zoo/USA) 3.1 lion (zoo = NS/USA) Experimental design and condition details AB: (A) Baseline: 8 days no enrichment (B) Enrichment: oryx faeces added to enclosure 8x repetitions ABA: (A) Baseline: no enrichment (20 days) (B) Enrichment: oryx faeces added to enclosures for 5 days (A) Baseline: no enrichment (14 days) ABA: (A) Baseline: 4wks (B) Enrichment: spices daily for 5-9 consecutive days (A) Baseline: 7days postenrichment (1 day) AB (repetitions = NS): (A) Baseline (no enrichment) (B) Enrichment (spices) Approx. study duration (months) & data collection period/s Study duration: 1.5mths Data collection: 4hrs obs/exhibit/day (5min intervals); 8 days/baseline, 8 days/enrichment type Study duration: 5mths Data collection: 5days obs/condition - 2hr 25min obs/exhibit/condition (5min intervals) Study duration: 3mths Data collection: all animals - 2x 30min obs (1min intervals) for 10 days baseline; 1 st, 2 nd and last day of enrichment; 1 day postenrichment Study duration: 1mth Data collection: 2x 30min obs/day (20s intervals) (1baseline; 1enrichment); total 24 days obs total (number of days per enrichment type = NS) Outcome measures Proportions of behaviour: Locomotion Vigilance Other Exploratory Inactive Pacing Proportions of behaviour: Locomotion Vigilance Other Exploratory Inactive Pacing Proportions of behaviour: Active Inactive Stereotypic Proportions of behaviour: Resting Alert Standing/Locomoting Not visible All-occurance of behaviour: Licking/gnawing Paw manipulation Sniffing/flehmen Face rubbing/back roll Enclosure use pacing Key behaviour speciesspecific details Yes Yes* Quirke & O'Riordan (2011a)** Yes Yes Quirke & O Riordan (2011b)** Yes* Yes* Skibiel et al. (2007)** N/A Yes* Powell (1995)** 17

36 findings # Study Table 1.3. Continued Enrichment category Enrichment type Zebra dung & squash (scented with cinnamon and vanilla) Synthetic pheromone (Feliway ) Exhibits Exhibit rotation Sample (zoo/s/ region/s) 2.0 lion 1.3 tiger (Smithsonian National Zoological Park/USA) 2.6 snow leopard (Edinburg Zoo, Cat Survival Trust, Dudley Zoo & Port Lympe Wild Animal Park/UK) 2.0 tiger a (Louisville Zoological Garden/USA) Experimental design and condition details ABA (repeated 3x/enrichment type - random order): (A) Baseline: 30-60min preenrichment (B) Enrichment (A) Baseline: 1 day postenrichment ABA: (A) Baseline (2 days) (B) Enrichment (3 days) (A) (Baseline (2 days) Between-groups: (1) Rotational housing (n = 1) (2) Single exhibit housing (n = 1) Approx. study duration (months) & data collection period/s Study duration: 2.5mths Data collection: 30min obs/animal/condition (30s intervals); 3x repetitions of obs for each enrichment Study duration: 3mths Data collection: 5-10x 30min obs/animal/day (1min intervals) for 7 days Study duration: 3yrs Data collection: Continuous recording (10min sessions); number of obs/animal = NS (1949 obs total but included other animals and conditions) Outcome measures Proportions of behaviour: Locomotion Rest Sit Sleep Swim One-zero occurrence: Discrete behaviours (e.g., climb, defecate) Object-directed behaviours (e.g, bite, carry) Behavioural diversity indices Proportions of behaviour Rest Alert Occupied Active Scent related Social Pace Durations of behaviour: Eating Locomotion Sationary alert Resting Object manipulation Stereotypy Spray Other Out of sight Location use pacing Key behaviour speciesspecific N/A Yes* (but analysed with novel object enrichment) details Van Metter et al. (2008) No Yes Macri & Patterson- Kane (2011) No Yes* White et al. (2003) 18

37 findings # Study Table 1.3. Continued Enrichment category Social enrichment Enrichment type Exhibit rotation Enclosure complexity (structurally enriched v. barren) Pre-existing housing Sample (zoo/s/ region/s) 1.0 tiger (Louisville Zoological Garden/USA) 10.6 leopard (Thiruvananthapuram Zoo, Aringar Anna Zoological Park, Shri Chamarajendra Zoological Gardens & Guindy Children's Park/India) 0.15 tiger (5 zoos/uk; 2 zoos/germany; 1 zoo/france; 1 zoo/netherlands) Experimental design and condition details AB: (1) Exhibit 1 (2) Exhibit 2 (both on-exhibit enclosures and similar size/structure) Between-groups (for on- and off-exhibit enclosures separately): (1) Enriched enclosures (2) Barren enclosures Correlation (for on- and offexhibit enclosure separately): Enclosure complexity (barren; partially barren; those with sleeping platforms/trees; or those with trees, clumps of bamboo and/or water bodies) & use of enriched areas of exhibit Between-groups: (1) Solitary housing with neighbouring conspecifics (n = 6) (2) Pair housing with neighbouring conspecifics (n = 3) (3) Pair housing without neighbouring conspecifics (n = 6) Approx. study duration (months) & data collection period/s Study duration: 3yrs Data collection: Continuous recording (10min sessions); number of obs/animal = NS (1949 obs total but included other animals and conditions) Study duration: 5mths Data collection: 2 days obs (5min intervals) on- (10hrs/day) and 2 days offexhibit (6hrs/day) per animal Study duration: 3mths Data collection: 30min obs (1min intervals) at various times over 5 consecutive days (total18hr)/animal Outcome measures Durations of behaviour: Eating Locomotion Stationary alert Resting Object manipulation Stereotypy Spray Other Out of sight Location use Proportions of behaviour: Rest Active Stereotypic pacing Enclosure use Proportions of behaviour: Locomotory (incl. pacing) Scent-marking Olfactory-response Ingestive Social Other Proportions of activities: Lying Siting Standing Moving Out of view pacing Key behaviour speciesspecific details N/A Yes* White et al. (2003) No Yes* Mallapur et al. (2002) Yes* Yes* De Rouck et al. (2005) Note: # = studies may have included other behavioural outcome measures, but these are the most frequently reported behaviour categories; * = statistically significant difference reported (for at least one individual/group of animals in the study, and at least one behaviour category related to pacing or species-specific behaaviour; ** = study appears in table twice due to exploring multiple types of enrichment 19

38 Summary and critical appraisal of included articles A summary of the six appraisal elements is provided here. In order to provide an inclusive review of the literature, no studies have been excluded from this review on the grounds of quality after undergoing critical appraisal, however these elements should be considered when interpreting results Sample Across studies, samples of large felids ranged from two to 16 animals (mean n = 7.25; median n = 5.5). Swaisgood and Shepherdson (2005) highlight the importance of large sample sizes for increasing external validity and the ability to conduct statistical analyses on collected data. It is difficult to determine whether the sample sizes in this body of literature are sufficient, however the mean sample size here was greater than that reported in Swaisgood and Shepherdson's (2005) review of enrichment for various species (median = 4; range 1-17) and therefore may be sufficient. Two-thirds of these studies have focused on a single large felid species, with the remaining portion including multiple felid species, both large and small (additionally, White et al.'s (2003) study also included non-felid species). Tigers, lions and cheetahs appear more frequently in this body of literature (in eight, six and five of the studies respectively) than leopards, snow leopards, and jaguars (2 studies each). Pumas are the least represented, appearing in only one of these studies. Reasons for this discrepancy are unknown - it may be attributed to zoos housing more of the former species, or a heavier focus on the welfare of these animals in the literature and therefore a greater perceived need to explore ways to enrich them. Much of the enrichment research for these species has been conducted in the United States, with none conducted in the Australasian region. The majority of the studies were conducted at single zoos, with four conducted at multiple zoos, usually within the same region. Being conducted at multiple zoos has typically allowed for larger sample sizes within these studies, and for greater external validity of findings. While this does not 20

39 discount the importance of those studies conducted at single sites, the findings should be generalised with caution Enrichment type A variety of enrichment types were explored. The majority of these efforts have focused on enrichments that target natural feeding and hunting-related behaviours: novel food items (n = 4 investigations); feeding routines (n = 3); methods of food provision (n = 3); novel objects (n = 2). Similarly, olfactory enrichment received a notable focus (n = 6). Less attention has been paid to enclosures (n = 3) or social housing (n = 1), and no studies investigated social enrichment in the form of human (keeper and/or visitor) interaction. The focus on feeding-type enrichments is not surprising, given the primary motivation of these animals to hunt and consume food. Similarly, the importance of patrolling and scent-marking behaviours for these animals is reflected in the focus on olfactory stimulation Study design Within-groups designs were most common. Five investigations (Altman et al., 2005; Bashaw et al., 2003; Powell, 1995; Quirk et al., 2011a; White et al., 2003) used a pre/post (or AB) type design to compare a range of enrichment techniques, but particularly alterations to feeding. While this design is useful for comparing enrichment to baseline conditions, the potential for confounding variables in different conditions to hinder interpretation of results is problematic (particularly when these are not controlled for in data analysis). A greater number of investigations (Bashaw et al., 2003; Charlton, 1998; Jenny & Schmid, 2002; Macri & Patterson-Kane, 2011; McPhee, 2002; Quirke et al., 2011b; Skibiel et al., 2007; Van Metter et al., 2008; Williams, 1996) used withingroups, repeated-treatment (e.g., ABA) type designs, allowing for increased internal validity and confidence in study findings regarding the influence of enrichment. Between-groups designs were used in four of the studies (Bond & Lindenburg, 1990; De Rouck et al., 2005; Mallapur et al., 2002; White et al., 2003) with no patterns in the type of enrichment investigated. While this design can be useful, it can also hinder conclusions if the 'control' and 'experimental' groups differ (e.g., regarding variables such as species, gender, age, etc.) and if these possible confounds are not accounted for. 21

40 Finally, one study (Mallapur et al., 2002) used a correlational design, to explore the relationship between enclosure complexity and exhibit use. Being a large, multiinstitutional study, this design is more appropriate than it would have been had a smaller sample been used, but does not allow for conclusions about causation (or the impact of an enrichment) to be determined Measures of enrichment success Behaviour observation was the sole method used the evaluate enrichment in all studies, with the exception that Altman et al.'s (2005) feeding enrichment study also evaluated digestibility of food and changes in animal weight. Regarding behaviour observations, studies varied in terms of which behaviours they focused on; this is not surprising given that different forms of enrichment will likely target different behaviours depending on the aim of the intervention. For example, studies of feeding enrichment included behaviours relevant to food acquisition and consumption, while studies of olfactory enrichment included scent-marking behaviours. Overall, there was a trend for studies to use increases in various species-typical behaviours, and decreases in pacing behaviour (as a potentially stereotypic behaviour), as indicators of enrichment success. However, while all studies included at least one species-typical behaviour, not all included pacing (Bond & Lindenberg, 1990; Powell, 1995; Van Metter et al., 2008; Williams et al., 1996). Of those that included pacing as a measure (i.e., the 12 remaining studies), various terms were used to describe this (including stereotypic/stereotyped pacing, stereotypy/stereotypic behaviour, and pace), but inspection of descriptions in-text confirmed that all terms referred to a similar form of behaviour. No other behavioural measures of compromised welfare/enrichment success were included and pacing was the sole stereotypic behaviour documented. Location use was also measured in some studies (Mallapur et al., 2002; Powell, 1995; White et al., 2003; Williams et al., 1996). Studies varied in the type of behaviour observations conducted, but focal animal recording using instantaneous sampling was used most frequently (n = 11 studies). While there are some limitations to this method (particularly in terms of its inability to document all occurrences of behaviour and durations of behaviour), it appears appropriate to the aims of these studies, which typically sought to understand whether 22

41 an enrichment influences the proportion of behaviour/s engaged in by an animal. With the majority of studies in this review using this method, comparability of studies is improved. Some studies also explored durations of behaviour (e.g., Altman et al., 2005; Bond & Lindenberg, 1990), consistent with a focus on whether enrichment can alter bouts of behaviour. The focus on behaviour in all of these studies is not surprising, given one of the primary aims of enrichment being to alter behaviour (see Section 1.1.2), and is consistent with past enrichment reviews finding it to be primarily evaluated by behaviour change across a range of species in zoos, as well as other captive settings (de Azevedo et al., 2007; Swaisgood & Shepherdson, 2005). So, although it is fitting to use behavioural markers to evaluate enrichment since the process seeks to increase species-typical behaviour and reduce stereotypies, enrichment also seeks to improve psychological well-being (Swaisgood & Shepherdson, 2005) and it has been suggested that well-being should be measured by at least two of the following criteria: behaviour, health, reproduction and longevity (Laule & Desmond, 1998). Therefore, with the exception of Altman et al.'s (2005) study, caution should be taken when judging the findings of this body of work as behaviour only offers one perspective on well-being and welfare. Four of the studies mentioned the use of intra-rater reliability analyses where multiple observers collected behavioural data (durations of observation sessions used for this varied between studies). Altman et al. (2005), Bashaw et al. (2003), and Skibiel et al. (2007) reported on percentage of agreement between observers, with inter-rater analyses in each study achieving at least 88.5% agreement. In contrast, White et al. (2003) reported the mean correlation coefficient between observers, of Despite the different approaches taken to report inter-rater reliability, these values represent high agreement between observers and add strength to the reliability of data collected within each of these studies. While it may be assumed that the remaining studies were conducted by single observers (though not all authors stated this), the inclusion of a similar measure of reliability can be useful for determining whether the single observer is accurately recording behaviour and how well defined the behaviour categories are 23

42 (for further discussion, see Martin & Bateson, 2007). Therefore the remaining studies could have been strengthened in this regard Data collection period All studies reported study duration and there is variation across studies regarding the study and data collection period, ranging from one month (Powell, 1995) to three years (White et al., 2003). Excluding White et al.'s (2003) study, as data were not actually collected continuously throughout this time, rather every summer, the remaining 15 studies had a mean duration of 3.1 months (range 1 to 7 months; median = 3 months). Absolute duration is not particularly informative, considering that some studies implemented multiple enrichments during the single time-frame and/or observed multiple animals at more than one organisation, therefore absolute time for individual animal involvement is lower than the overall time for the study. Regardless, whether or not study duration is appropriate relates more to the individual aims of each study. With studies typically seeking to explore short-term changes in behaviour, studies of this duration seem appropriate. Similarly, data collection duration varied across studies (often this was less than the duration of enrichment implementation). Not all studies reported duration of each condition (e.g., Bond & Lindenberg, 1990) and studies varied in whether they reported the number of days or number of hours of data collected per animal/condition. This makes determining whether enrichment and data collection periods are sufficient. Perhaps what is more important is consistency in the amount of data collected in each condition within studies, which was similar in most cases (except that some studies had a longer baseline phase, which is necessary to gain a more comprehensive understanding of pre-enrichment behaviour) Data analysis Most studies included a distinct, and clearly described, data analysis section. However, some did not and/or failed to include sufficient information to understand how observation were converted to and included in analysis, or to justify why the particular analysis method was chosen (e.g., Altman, 1998; Bond & Lindenburg, 1990; Mallapur et al., 2002; White et al., 2003). All studies conducted tests for statistical significance, with none presenting descriptive statistics or individual animal statistics alone. For the 24

43 within-groups studies, non-parametric alternatives (e.g., Friedman's ANOVA, Wilcoxon signed-rank test, Kruskal-Wallis test, etc.) were commonly used and most authors attributed this to small samples sizes and/or non-normally distributed data. Where parametric tests were used, these were well-justified, and transformations were conducted if necessary (Charlton, 1998; McPhee, 2002; Van Metter et al., 2008). While this use of non-parametric alternatives is a cautious approach given violations of assumptions of parametric tests, these tests do not appear to account for the nonindependence of data, which characterises repeated treatment designs. Another problem with some of these studies is that data were often pooled for some or all animals, without justification for whether this was appropriate (e.g., Bashaw et al., 2003; Van Metter et al., 2008). This could be problematic, considering that differences between individual animals (e.g., age, sex, housing conditions, etc.) may account for differences in behavioural responses, but cannot be controlled for with these tests. Regarding the small number of between-groups studies (n = 4), a combination of nonparametric (Bond & Lindenburg, 1990; Mallapur et al., 2002; White et al., 2003) and parametric (De Rouck et al., 2005; White et al., 2003) tests were employed. Again, while it is cautious to use non-parametric alternatives considering small sample sizes and non-normally distributed data, these tests do not allow for the control of confounds inherent in between-groups analyses and therefore should be interpreted with caution. In contrast, De Rouck et al.'s (2005) study employed parametric analyses and adjusted for confounds (e.g., weather conditions; opportunities to engage in particular behaviours) by only including data from observations in which animal conditions were similar; the authors also transformed their data prior to analysis, making parametric testing more appropriate Discussion of literature findings Findings from the literature are discussed herein according to the following key enrichment categories: feeding and hunting; olfactory; enclosures; and social housing (see Table 1.3). This review takes a narrative format, whereby the results of the studies 25

44 are discussed within the context of what is presently known about these elements of zoo-living for these species Enrichment techniques targeting species-typical feeding and hunting behaviours Given that one of the primary motivations of large felids is to hunt and consume prey (Guggisberg, 1975), it is not surprising that the majority of documented environmental enrichment efforts have addressed these behaviours. As shown in Table 1.1, these animals primarily consume ungulate prey (of varying sizes), although all large felids will also take other prey opportunistically (Macdonald, Loveridge, & Nowell, 2010). While some individual differences are likely to exist, these felids all follow a similar hunting pattern (Kitchener, Valkenburgh, & Yamaguichi, 2010; Macdonald, Loveridge, & Nowell, 2010). Over 20 years ago, Lindburg (1988) drew upon field data of tiger, lion, leopard and cheetah feeding behaviour and proposed the shared use of the following foraging tactics: locating (including travel and detecting), capturing (including stalking, scavenge hunting, coursing, and ambushing), killing (including disabling and dispatching), and processing (including pre-feeding activity, parceling, ingesting and caching). Further, Kitchener et al. (2010) discuss how large felids all possess similar morphological features that make these processes possible. As such, they are designed, and specifically adapted, for a range of hunting and consumption behaviours. Despite this readily available knowledge about the importance of hunting and consuming prey for these species, zoo-based feeding routines for large felids have historically involved daily feeds of pre-prepared diets (e.g., meat that has been removed from the bone); with a greater emphasis placed on nutritional value, rather than natural food (Law et al., 1997; McPhee, 2002; Mellen & Shepherdson, 1997). Moreover, in many Western countries such as Australia the provision of live vertebrate prey (including fish) to captive animals is prohibited under legislation (see Department of Agriculture, 2009, p. 24, S 4.2). Processed diets do not require natural foraging tactics, energy expenditure, or much use of appendages, and this can have various physiological implications (Bond & Lindenburg, 1990; Duckler, 1998; Law et al., 1997; Quirke, 26

45 O'Riordan, & Zuur, 2012; Skibiel et al., 2007). For example, Duckler (1998) found differences in the skull formation of wild and zoo-housed tigers and lions, concluding that this could be due to differences in the use of head and neck muscles during hunting and feeding. Worldwide feeding guidelines (e.g., WAZA, 2005) do not advocate for pre-prepared diets and instead suggest that zoo-housed large felids should be provided with whole or partial carcasses, with skin intact. To reflect this knowledge, enrichment efforts have therefore focused on stimulating natural hunting/feeding behaviours and attempting to replicate, to the extent possible, feeding as it would occur in natural habitats. This has been achieved through the provision of novel food items, changes to feeding routines and food provision, as well as novel toys/objects to stimulate hunting behaviours Novel food items as enrichment Four of the 16 studies (Bashaw et al., 2003; Bond & Lindenburg, 1990; McPhee, 2002; Skibiel et al., 2007) explored novel food items as an enrichment for each of the large felid species. Results show that items such as animal bones, carcasses (whole and partial), and fish (frozen and live) can contribute to an increase in species-typical hunting behaviours, from capturing, mock killing and fur plucking, through to consumption. For example, Bond and Lindenburg (1990) explored the duration and frequency of behaviours observed during feeding sessions of carcass-fed cheetahs (n = 2) compared to those fed commercial diets (n = 3) at San Diego Wild Animal Park, reporting the former to spend more time smelling their food, performing more bites per swallow, more time using molars to slice food, and more time feeding overall, compared to the latter. However, when pooling data for analysis, the criteria for significant findings in this study was set at p = Therefore, while the researchers (Bond & Lindenburg, 1990) suggest that carcasses better meet the psychological and nutritional needs of cheetahs, since they account for factors such as diet consistency, texture, complexity, temperature, palatability and variability, their conclusions are limited by results that would usually not be considered significant (p < 0.05 is a more universally accepted cut-off). The study also fails to provide a detailed methodology and it is unclear how many feeding sessions were observed per animal (the paper states that 31 feeding days were observed, but it is unclear whether this is per animal, or for all 27

46 animals combined); therefore, whether sufficient data were collected to address the study aim and perform statistical analysis cannot be determined. Despite these limitations, some support for Bond and Lindenburg's (1990) study is provided by McPhee (2002) finding increased proportions of both feeding and other natural behaviours when leopards (n = 3), lions (n = 2) and snow leopards (n = 4) were provided with whole carcasses compared to baseline feeding of their regularly provided processed diets. With the study involving seven repetitions of the baseline and enrichment phases for each animal, the internal validity of the findings is increased, as is the external validity given that this was a multi-institutional study. Similarly, Skibiel et al. (2007) found all species to show a trend for increased active behaviour when provided with bones and frozen fish compared to a four-week baseline phase, which translated to a significant increase when data for all species were pooled for statistical analysis. While conducted with multiple species, each animal only received the enrichment condition during one phase and therefore repetitions of provision may have produced more reliable results. In addition, decreases in pacing have also been observed with novel food items in the three studies which explored this behaviour (McPhee, 2002; Bashaw et al., 2003; Skibiel et al., 2007). For example, Bashaw et al. (2003) found a trend for a 50% reduction in pacing with the provision of live fish to captive tigers, from 60% at baseline (six hours of observation at different times of day over a four week period) down to 30% on the day of enrichment (which was repeated eight times over a further four week period); this result was sustained two days after provision. Although nonsignificant, this is clearly a noticeable decrease for these individual animals and may have positive changes for their individual welfare. As recognised by the authors (Bashaw et al., 2003), the lack of significant findings may be associated with the low power associated with the relatively sample size (n = 2), rather than with any flaws in the study design (considering that multiple repetitions of the enrichment were conducted). Adding some support to Bashaw et al.'s (2003) finding is Stark's (2005) conference proceeding, which discusses how pacing was completely abolished on the day of enrichment with the provision of calf carcasses to a tigress (provision was 28

47 repeated four times, at least two-weeks apart). Of note, however, this individual paced at a rate of 9% of baseline scans (recorded during three, 20 minute observation periods per day for 13 days), demonstrating a less drastic reduction in this behaviour than that observed by Bashaw et al. (2003). Moreover, the external validity of this finding is low given the focus on a single participant in this comparison (the study also involved two female cubs, but neither paced at all during any study condition). Additionally, while McPhee (2002) found decreased pacing for some individual animals, when pooling data for all participants pacing behaviour did not differ with the provision of feeding enrichment. McPhee (2002) suggests that this lack of difference might be associated with relatively low pacing in these animals to begin with. However, no rates of pacing are provided (only the direction of differences and significance levels are reported), hindering readers from determining the accuracy of this conclusion Altered feeding routines and methods of food display as enrichment Research has also explored alterations to food provision. In an attempt to replicate the natural environment, where felids do not always hunt (or hunt successfully) each day, some zoos do not feed their large felids daily, as suggested in readily available - albeit dated - guidelines for housing these animals (Dierenfeld, Bush, Phillips, & Montali, 1994). Further, it is believed by some that regular and/or predictable feeding schedules can be linked to stereotypy and/or food anticipatory activity in carnivores (Bassett & Buchanan-Smith, 2007; Carlstead, 1998). Whether or not this represents compromised welfare in these species is yet to be supported; however there is some support for the link between pacing and predicable feeding. A recent observational study of 112 cheetahs housed in 88 exhibits at nine international zoos (Quirke et al., 2012) aimed to determine factors that may contribute to the prevalence of pacing in these animals. Findings revealed that pacing was observed in 85 of the enclosures, accounting for 10.8% of scans, and that this behaviour was higher in those animals experiencing a predictable feeding routine (defined as the provision of food within the same hour each day). Within the available enrichment literature, three studies have focused on altered feeding routines for lions and cheetahs. Altered feeding cycles and methods of food display may 29

48 be considered enriching given the variation provided, with the aim being to simulate feeding in natural habitats where food is unlikely to be available on a predictable schedule. Altman et al. (2005) investigated this by gradually altering the regular feeding routine of lions (n = 5) over a 10-week period, from feeding six days/week at a set time of day, to feeding three days/week at non-predictable times and on varied days. In this study, the type of food (pre-prepared, ground horse-meat with nutritional supplements) and weekly amount of food were kept constant. Comparisons of behaviour during the first two weeks of the study to the last two weeks showed a significant reduction in pacing (from 4% to 2% of scans). Initially, no changes in active behaviour were found; however, upon analysing individual behaviours, the authors found a significant increase in appetitive behaviours (from 6% to 9% of scans). While some caution should be taken when interpreting these results, as the authors did not detail their data anlaysis process (making it unclear which data went into analysis and whether this was pooled for all animals), the study also explored non-behavioural outcomes. The finding of greater digestibility, decreased food intake, and weight loss (to levels comparable with data for wild lions) with the enrichment, adds support to the behavioural findings regarding the benefits of this technique. Another facet, which may influence behaviour, is temporal (time) and spatial (location) variation of feeding (Bassett & Buchanan-Smith, 2007). This has been explored in two studies of cheetahs, by Quirke and O'Riordan (2011a, 2011b), who implemented feeding enrichment whereby the usual feeding time and feeding location was altered. One study did this on a fixed schedule (Quirke & O'Riordan, 2011b) and the other randomised the enrichment implementation (Quirke & O'Riordan, 2011a). While the scheduled enrichment (Quirke & O'Riordan, 2011b) only revealed trends for decreases in pacing (from 13.7% at baseline to 7.7% and 10.1% with temporal and spatial variation respectively), the randomised enrichment program (Quirke & O'Riordan, 2011a) found significant decreases in proportions of this behaviour (from 0.08 at baseline to 0.02 with temporal and 0.08 with spatial variation). No significant differences were found in locomotory or exploratory behaviours in either study with either type of feeding enrichment. However, 'other' behaviours (including feeding and affiliation) significantly increased with temporal feeding on a random schedule (from 0.07 at baseline to 0.17). Despite the slightly different composition of animals, with 30

49 these studies being conducted at the same zoo there is stronger support for the importance of randomised enrichment schedules. The effect of food presentation on tigers, lions, cheetahs and jaguars has also been explored in three studies, with success found when food has been presented in feeding boxes and hidden in hessian sacks, and when moving bait has been provided (Charlton, 1998; Jenny & Schmid, 2002; Williams et al., 1996). For example, Williams (1996) provided moving bait to cheetahs (n = 2) who had not previously experienced hunting enrichment, by attaching their daily feed (a whole rabbit carcass per individual) to a suspended pulley system. The study found that the frequency of sprinting behaviour tripled (values not specified) and proportions of scans engaged in observation behaviour significantly increased during the enrichment provision (5%) compared to baseline (2%) feeding (whole rabbit carcasses not on lure). Interestingly, however, proportions of feeding behaviours were actually lower in the enrichment phase (~8% of scans) compared to baseline (~12%) and post-enrichment (~15%). The authors suggest that this may be associated with the moving bait simulating natural hunting whereby there is a need to consume food quickly to avoid it being taken by other animals. Given that the same size and type of food was provided in each condition, this seems like a feasible explanation. In another example, Jenny and Schmid (2002) explored the impact of feeding boxes (whereby food was placed in various boxes and could only be accessed if opened by the animal) on one male and one female Amur tiger (Panthera tigris altaica), compared to pre-existing conventional feeding (with food being provided at the same time and location of exhibit each day, except for on fasting days; the frequency or predictability of this was not specified). Both individuals demonstrated a decrease in pacing with the addition of feeding boxes, but there were differences for each individual when housed solitarily and paired: although the female demonstrated a significant reduction in pacing durations under both conditions, the male only demonstrated a significant reduction in duration of this behaviour during box feeding when housed with the female. It may be that the presence of the female influenced the male's response to the feeding enrichment; however, since there were no comparisons between solitary and pair 31

50 housing without enrichment, this cannot be determined. Moreover, the study design was such that the two trials were conducted separately, with the pair housing condition only occurring after the male and female had mated. The authors propose the lack of change in pacing in solitary confinement to be associated with the animal recently being separated from mother and siblings and moved to this zoo. Additionally, this study showed that both tigers displayed an increase in durations of species-typical behaviours during the experimental conditions, with the female exhibiting an increase in sleeping, and the male showing an increase in active behaviours directed at the feeding boxes. Given the repeated-treatment design used in this investigation, and that the animals were habituated to the enrichment prior to the trials beginning, the study findings are deemed to be valid. Meehan and Mench (2007) discuss how opportunities for problem solving, such as this one explored by Jenny and Schmid (2002), can enhance captive animal welfare through providing opportunities for animals to use their cognitive skills (such as processing, storing, retrieving and acting upon information). In the case of Jenny and Schmid's (2002) study, the animals were required to learn associations between their behaviours and the consequence of obtaining food, thereby rely upon cognitive processes, to access the enrichment. In support of this finding, food presentation has been found to similarly influence captive small felid behaviour (for discussion, see Shepherdson, Carlstead, Mellen, & Seidensticker, 1993). Although the overall finding from these studies is that novel food presentation is a successful enrichment method for zoo-housed large felids, it should also be considered that individual animals may respond to this (and other) forms of enrichment in different ways. For example, while Charlton (1998) found the provision of hessian sacks containing food to significantly decrease pacing frequency and duration in one captive jaguar (compared to conventional feeding days before and after enrichment, whereby the same type of food - horse or beef meat on or off the bone - was placed on exhibit flooring), two others in the study actually displayed an increase in pacing frequency and duration during enrichment (although this result was non-significant, but may be attributed to the small sample size being unable to detect effects). The results of this study are strengthened by the repetition of conditions on three occasions, and by allowing the animals to habituate to the enrichment before the trials began, ensuring that any novelty associated with the technique was diminished. 32

51 Novel toys/objects as enrichment To compensate for the lack of live prey and the subsequent lack of hunting-related energy expenditure, zoos have been creative in developing novel play devices to encourage these natural behaviours in their large felids. A number of devices have been added to exhibits, including: artificial prey in the form of plastic rabbit replicas and bird sounds; cardboard box animals; burlap sacks hung from trees that move when attacked ; ropes for tugging (a tiger-tug ); barrels containing stones for sound; watermelons that roll away when chased; metal kegs; and frozen blood balls. However, experimental exploration of such objects appear to have only featured in two peerreviewed sources, focusing on tigers and lions (Powell, 1995; Van Metter et al., 2008) and the findings presented are limited, particularly since both studies involve multiple enrichment items within the one study phase. Powell (1995) compared lion behaviour in holding areas with 30 minutes of enrichment compared to 30 minutes earlier in the day with no enrichment. The study involved presentation of different enrichment items, including frozen blood balls and hanging logs (as well as scents, discussed in Section 1.4.2). With the provision of frozen blood balls, a significant increase was found for scans in which object-directed behaviour (e.g., lick/gnaw, and paw manipulation) was observed in adults and cubs, as well as increases in stand/locomote and sniff/flehmen for cubs. This is not unexpected given the item is intended to stimulate such behaviours. Similarly, significant increases in paw manipulation behaviour were seen with the hanging logs when provided to the adults (not provided to cubs). Changes in proportions of other species-typical behaviours (e.g., rest and alert) were not observed. While these findings are positive, no mention of stereotypic behaviours were made (it is unclear whether this behaviour was present in these animals), and the lack of a post-enrichment condition fails to show whether this form of enrichment has any further influence (positive, negative or neutral) on behaviour. Moreover, the study does not detail how many repetitions of this enrichment item were provided, limiting the understanding of how valid the findings are. The findings of Van Metter et al.'s (2008) study also provided limited insight into this form of enrichment. These authors implemented a combination of novel objects (metal 33

52 kegs, cardboard boxes and frozen blood balls) and scent enrichments (zebra (Equus quagga) dung and scented squash) to tigers (n = 2) and lions (n = 3). Despite some strengths of the study (such as randomised and repeated implemention of each enrichment), the study pooled and analysed data from all enrichment types compared to pre- and post-enrichment baselines and did not present data with respect to each enrichment individually. However, it did compare behavioural diversity between enrichments, finding this variable to be significantly lower with the provision of blood balls compared to scents (no mention of kegs/boxes is made). As such, despite the study being well designed, the analysis fails to provide sufficient detail about individual enrichments, which is important if we are to have a better understanding of how these species might respond to different forms of enrichment. Stereotypy was not explored in this study due to no reported observations of this behaviour. The success of novel objects appears to be more readily discussed in anecdotal accounts (in both peer-reviewed and non-peer-reviewed sources) as well as in conference proceedings for various large felid species (Hare & Karrand, 1998; Markowitz & LaForse, 1987; Poulsen & Miller, 1996; Roynon, 2000; Stark, 2005; Tresz, 1997). For example, Hare and Jarrand (1998) anecdotally discuss how burlap sacks (filled with either branches and vegetation, or meat, and suspended from trees) encouraged natural felid hunting behaviours, including stalking, approaching, and biting, when provided to tigers at San Diego Zoo. In addition to this relatively limited understanding of novel object enrichment, gender differences have been found in response to such items. In a study primarily focused on investigating the potential to determine snow leopard personality, rather than enrichment, Gartner and Powell (2011) observed the behaviour of these animals in response to various novel objects (some examples include a round buoy, a trash can, and drain piping). They found that male snow leopards (n = 4) spent significantly more time engaged in exploratory behaviour, scent marking and visiting the objects than females (n = 5). More research is needed to determine whether a gender difference exists in response to novel toy enrichment, the factors that may account for it, and whether this is common across all large felid species. 34

53 Olfactory stimulation as enrichment One of the key ways in which large felids communicate with each other is through olfactory behaviours, such as rubbing objects with body parts (including head, cheek, chin, and neck), claw raking, urine spraying and ground scraping with hind feet (Clark & King, 2008; Kitchener, 2000; Sunquist & Sunquist, 2002). These behaviours serve multiple purposes, such as bringing animals together (e.g., via the communication of an animal s reproductive status), avoiding contact, or separating individuals by means of an animal signaling its territory (Sunquist & Sunquist, 2002). Scent marking behaviour is particularly useful for these animals, given their spatially and temporally dispersed lifestyles (Sunquist & Sunquist, 2002). However, in many zoos where felids constantly occupy the same space, there may be little need for scent-marking and territorial behaviours, due to limited olfactory changes in the environment. As such, some focus has been placed on stimulating natural territorial behaviours through olfactory enrichments in zoos through the addition of scents to an animal s enclosure. Such studies have focused on all large felids, with the exception of leopards. The addition of scents, such as perfumes, spices, and faeces from prey animals, have been found to encourage natural communicatory and territorial behaviours, and to increase large felid stimulation in zoos (Powell, 1995; Quirke & O'Riordan, 2011b; Skibiel et al., 2007; Thomas, Balme, Hunter, & McCabe-Parodi, 2005; Van Metter et al., 2008). While the studies by Powell (1995) and Van Metter et al. (2008) provide some evidence for this, their findings are difficult to interpret for the reasons discussed in Section Information from other peer-reviewed sources can provide more support for this form of enrichment. Skibiel et al. (2007) compared the presentation of spices (cinnamon, chilli powder and cumin) on the behaviour of tigers, lions, cheetahs, jaguars, and pumas (as well as ocelots [Leopardus pardalis]). For all animals combined, it was found that the observed proportion of active behaviours significantly increased from baseline (15.64% to 28.02%) with the provision of spices, while the proportion of scans engaged in pacing significantly decreased (26.8% to 10.85%). However, although the enrichment condition was compared to baseline behaviour, the baseline condition involved the removal of any pre-existing enrichment items. While it is clear that this allowed the researchers to observe the effects of the new enrichment items, if typical husbandry of these animals includes enrichments then these should have remained 35

54 during baseline to avoid over-inflating the success attributed to the new enrichments. Regardless, anecdotal accounts also support these findings, with observed behaviours including extensive flehmen responses (scent inhalation), paw scrapes, grass pulling and digging with the presentation of olfactory stimuli (for examples, see Baker & Campbell, 1997; Tresz, 1997). Again, the two studies by Quirke and O'Riordan (2011a, 2011b; discussed in Section ) show that cheetah responses to olfactory enrichment (oryx faeces) can differ depending on whether the enrichment is presented in a scheduled or random manner. In particular, while one study showed no significant difference in proportions of exploratory behaviour from baseline to enrichment when the enrichment was presented continuously for five days (2.9% compared to 3.1%; Quirke & O'Riordan 2011b), the other showed that when randomising the presentation of enrichment (such that it was randomly provided on eight days out of a total of 40, including days of other enrichment and no enrichment) there was a significant increase in proportions of this behaviour on enrichment days (from 0.01 on baseline days to 0.05 on enrichment days; Quirke & O'Riordan 2011a). While this may lend support to randomised enrichment schedules, baseline proportions of this behaviour were lower in the latter study. Many of these aforementioned scents are not related to items/stimuli that would usually be found in these animals' natural habitats, and Wells' (2009) review of sensory enrichment in captivity (including, but not limited to, zoos) suggests that olfactory stimuli do not necessarily need to be found in a particular animal's natural environment in order to be a successful form of enrichment. There is some evidence to support this with respect to large felids. Although not an enrichment study per se, Thomas et al. (2005) investigated whether captive cheetahs were attracted to perfume and cologne as part of a study involving the use of scents to attract wild large felids to hair traps (to collect hair samples from wild cats) in South Africa. These researchers tested 24 commercially available scents (perfumes/colognes) on a cheetah sibling pair at the Bronx Zoo and found that both investigated most of these, taking just over 10 minutes to come into contact with them and usually exploring the scents for around one minute each. Both displayed sniffing, rubbing and spraying behaviours in response to the 36

55 scents. Similar reactions were anecdotally observed in tigers, lions, leopards and snow leopards at the same zoo (Thomas et al., 2005), and have been also been documented elsewhere in lions (Powell, 1995). Interestingly, when used in the natural habitat to attract large felids, this method was essentially unsuccessful, with only four of 50 scentbaited traps collecting hair samples (Thomas et al., 2005). The fact that their wild counterparts did not investigate the scents as readily may be a result of various factors, such as trap-shyness, or neophobic responses towards objects or smells associated with humans. Nevertheless, the different response of the two groups of animals to the same scents lends support to the use of olfactory stimuli in zoos, to promote interest in an environment which may lack the olfactory complexity of natural habitats. In another study, Macri and Patterson-Kane (2011) investigated the use of Feliway (a synthetic pheromone used with domestic cats) on the behaviour of zoo-housed snow leopards. Data were pooled and analysed separately for those animals house solitarily (n = 4) and those housed socially (n = 4). No significant behavioural changes were found, however some trends were reported: while pacing remained relatively stable for the solitary animals, it increased on enrichment days for the social animals (although in all conditions pacing remained <5% of scans); for both groups, active behaviour remained at approximately 10% of scans before and during enrichment, but reduced to approximately 5% of scans on post-enrichment days; scent-marking and olfactory behaviours increased during the enrichment days (but did not raise beyond 3% of scans in any condition). Rather than the enrichment not being successful, there are various possible reasons for the non-significant findings. While it is understandable that scan sampling was used to document active and pacing behaviours, with the purpose of this enrichment to increase scent-marking and olfactory behaviours, which may occur relatively quickly (e.g., spraying), it is surprising that the authors did not use a more sensitive observation method (such as all-occurrence recording or shorter intervals). As such, the non-significant findings may be attributed to the methodology rather than the enrichment not being successful. Alternatively, the data were analysed separately for socially- and solitarily-housed animals. With only four animals per condition and one phase of enrichment, increasing the sample size and including repeated presentations of the enrichment may have increased power to detect effects. As such, it remains unclear as to whether or not conspecific scents are enriching (or stressful) for zoo-housed felids 37

56 (or whether there are species differences in pheromones). Anecdotal accounts suggest that other felid scents promote natural territorial behaviours and stimulate interest in conspecifics (for example, see Tresz, 1997), but there is a clear need for additional scientific research, particularly since scents from other felids may be easy to obtain in zoos housing multiple animals of the same species. Conversely, given that most large felids are solitary, conspecific scents may also have the potential to cause stress to these animals and therefore more evidence is needed before such an approach can be recommended Enclosures as enrichment Aspects of an enclosure, such as its size and complexity, could be altered as a form of enrichment. Whilst differing depending on the region, social structure of the individual species, and prey abundance, large felids typically occupy large home ranges in their natural habitats (Sunquist & Sunquist, 2002). In zoos, however, this range is drastically reduced, especially in urban facilities (as opposed to those in open-range settings). Restricted space may decrease the opportunity for these animals to perform territorial behaviours to the extent that they would in their natural habitats and the proportion of time devoted to this (assuming that a smaller territory in zoos requires less time to patrol and scent mark). This has been substantiated by some studies of large felid welfare. For example, in an exploration of the effect of zoo environments on various felid species (n = 19), Lyons et al. (1997) found a significant positive correlation between enclosure size and average movement, even after controlling for animal body size. More recently, Bashaw et al. (2007) found a trend for increased activity and decreased pacing when tigers (n = 2) were housed in a larger exhibit (957m 2 ) compared to a smaller exhibit (149m 2 ). Lastly, in developing a model for predicting pacing prevalence, Quirke et al. (2012) investigated the behaviour of cheetahs at nine different facilities in Africa, Canada and the United Kingdom, finding that the probability of pacing behaviour being observed decreased as enclosure sizes increased. However, specific sizes of enclosures and levels of pacing were not reported, making it difficult to apply these findings to an understanding of 'ideal' enclosure sizes for this species. 38

57 The influence of size might only hold true in enclosures that are not particularly complex. For example, Clubb and Mason (2007) discuss the importance of understanding what having a large home range actually means for an animal that is, determining which behaviours a large home range allow animals to perform and how these can be encouraged in zoos when a larger space is not available. As seen with feeding enrichment (Section 1.4.1), elements of feeding can be changed to simulate natural feeding behaviours, without actually providing live food. As such, it may be that changes to enclosures are more important than the absolute size/space available to these animals. This has been explored with respect to enclosure rotation and complexity in the large felid enrichment literature Enclosure rotations Enclosure rotation is a process whereby several animals may occupy a number of exhibits simultaneously or serially, and are moved from one to the next either within a single day or on different days. Rotating animals in such a way is thought to more accurately resemble the natural environment, where other species are present and animals have interconnected home ranges and pathways (Coe, 1997). This can allow for various environmental stimuli to be encountered, such as physical variation and the scents of other animals (White et al., 2003). Mellen and Shepherdson (1997) recommend this practice for felids and anecdotally note how a pair of Amur tigers housed at Metro Washington Park Zoo, Portland, appeared to display more activity after the introduction of a rotation system compared to before. Scientific investigations of this form of enrichment, however, have tended to focus on primates (Coe, 1997). It appears that only one study has included a large felid species, with White et al. (2003) comparing the behaviour of a Sumatran tiger rotated through four exhibits (simultaneously housing orangutans [Pongo pygmaeus], Malayan tapirs [Tapir indicus], siamangs [Hylobates syndactylus] and babirusa [Babyrussa babirusa]) to a Siberian tiger housed in a solitary exhibit. Findings revealed no significant differences in most behaviour categories, including stereotypic pacing, but the two tigers did differ in time spent resting and engaged in other behaviours (including species-typical behaviours, such as spraying and grooming). The traditionally housed 39

58 tiger rested significantly more than the tiger on rotational display (52.20% of scans compared to 37.31% of scans), while other behaviours were significantly higher for the rotational tiger (5.67% compared to 1.69%), indicating the potential for enclosure rotation to increase active, species-typical behaviours. While a promising finding, with only two participants, the data analysis could not account for age and gender differences between the two animals (which may have contributed to different behavioural responses) and the authors failed to report how much, or how often, data were collected from each individual. As such, it may be that the data were not equal and a comparison of this nature may not be appropriate. Moreover, the limited external validity due to the small sample size means that more research is needed here, with the inclusion of other large felid species too. Considering that many zoos are likely to have multiple large felid enclosures, this could be explored with relative ease (provided that shifting animals in such a way is possible within any one facility). The study (White et al., 2003) also compared the Sumtran tiger's behaviour when rotated between two different exhibits during two consecutive summers. A significant increase in locomotion and decrease in resting were found over time, and a significant different in stationary alert behaviour between exhibits. With the two enclosures being almost identical in size and features, this tells us little beyond the idea that this animal behaved differently in the two environments Enclosure complexity Despite what is known to date about the influence of enclosure size and space on large felid welfare, only one study has explored the influence of structurally enriched enclosures. Mallapur et al. (2002) documented the behaviour of 16 singularly housed leopards across four Indian zoos, varying in on-exhibit enclosure complexity (from barren enclosures, to enclosures with trees, bamboo, and/or water bodies), and found that those individuals in the more enriched enclosures displayed significantly more active behaviour. Further, those individuals with more enriched enclosures spent more time in the enriched areas than non-enriched areas of the enclosures, indicating their use of the enriched features. No differences in pacing were found. The large sample across multiple zoos increases the external validity of the findings, however more experimental-type studies (rather than explorations of pre-existing difference) are needed to confirm these results. Again, such research could be useful for updating 40

59 minimum exhibit furnishing standards. However, it is also acknowledged that making changes throughout exhibits is likely to be more costly and time intensive than implementing a single enrichment device, and this may account for the lack of studies of this nature Social stimulation as enrichment Given that feeding and territorial behaviours are dominant in large felids in their natural habitats, it is not surprising that much of the literature has focused on enrichment efforts to encourage these. Of lesser recognition is social enrichment for these animals. Swaisgood and Shepherdson (2005) highlight that solitary species typically fare worse in zoos than more social species due to lack of access to, or lack of benefit from interacting with, conspecifics. Although it is recognised that appropriate social stimulation, both within and between species, may be considered enriching as a result of increased arousal and the opportunity for interactions (Swaisgood & Shepherdson, 2005), this has not been explored in sufficient depth for large felids. Most large felids, with the exception of the lion and some male cheetah, are solitary in nature, usually only interacting for mating purposes and when females rear cubs (Sunquist & Sunquist, 2002). Despite this, not all cats are necessarily antisocial. For example, there have been reports of additional conspecific interactions in wild tigers, such as group hunting (Kitchener, 2000; Tilson et al., 1997). Shoemaker et al. (1997) indicate that large felids do not necessarily have to be housed singularly, and details of housing style in enrichment studies suggest that many typically solitary cats are often housed socially in zoos (examples include: Lyons et al., 1997; Quirke & O'Riordan, 2011b; Skibiel et al., 2007; Williams et al., 1996). While the use of social housing is probably aimed at saving space and the expense of building separate enclosures, as suggested by Law et al. (1997), it is not clear at present whether or not this could be enriching for large felids. It has been argued that appropriate group housing for social species is one of the best forms of enrichment for them (Young, 2003) and, as such, it is perhaps not surprising that the social housing of lions and male cheetahs has not been 41

60 addressed in the literature (since these are social animals and it may be assumed that housing them socially should be standard practice, not an enrichment per se). To date, only one zoo-based study appears to have explored social housing from an enrichment perspective for large felids. In a multi-institutional investigation of preexisting housing routines, de Rouck et al. (2005) found that, compared to solitaryhoused female tigers, their pair-housed counterparts performed a wider variety of behaviours, which were primarily interactive in nature. Further, those that were pairhoused and had no neighbouring tigers paced significantly less than those (either single or paired) with neighbouring tigers. Indeed, other researchers (Bashaw et al., 2007; Miller, Bettinger, & Mellen, 2008) have found that being housed in close proximity to conspecifics could be a cause of pacing in tigers and explored the use of visual barriers between neighbouring zoo exhibits to reduce this behaviour (with contrasting findings regarding success of this technique). Changes in behaviour of socially housed female tigers (n = 6) have also been explored in a longitudinal study by Miller and Kuhar (2008). However, with the purpose of this study being to investigate changes in social behaivour over time, it is not particularly useful for determining whether or not social housing is enriching. Social housing has also been explored with cheetahs in a non-zoo, captive setting. Wielebnowski et al. (2002) found significant decreases in grooming, and increased pacing and flehmen responses, when female cheetahs were housed in pairs at White Oak Conservation Centre in North America. Notably, in 10 of the 12 pairings, the individuals only paced when pair-housed. Further, most pairs displayed more agonistic behaviours than affiliative ones. There is a clear need for more research in this area, since suboptimal social groups are recognised as a potential source of stress for zoo-housed animals (for more detail, see Morgan & Tromborg, 2007; Price & Stoinski, 2007). Further, the influence of such housing on physiological factors should also be considered. For example, in addition to testing the influencing of pair housing on the behaviour of female cheetahs, Wielebnowski et al. (2002) also examined faecal oestradiol, progestogen and corticoid levels in the individuals. Of interest, they found that five of the six pairs demonstrated prolonged anoestrus (periods of sexual inactivity) during pair-housing, suggesting that 42

61 unnatural housing for this species may have negative implications for breeding success. This is particularly problematic for cheetahs, since they are notorious for experiencing breeding difficulties in zoos (Marker-Kraus, 1997). Moreover, research with smaller felids has revealed negative impacts of social housing, such as reduced reproductive success in pair-housed animals and higher levels of pacing (Law et al., 1997; Mellen & Shepherdson, 1997; Swanson et al., 2003). Given that large felids may be housed socially in many zoos (see Section 1.1.3), additional research to substantiate the efficacy of this practice is pertinent Human interaction: possible enrichment or negative influence on welfare? None of the located studies explored social enrichment in the form of human interaction. This is not surprising, considering the solitary nature of many of these species, and the apparent absence of positive interactions in natural home ranges (discussed further throughout Chapter 2). However human interaction has previously been considered a form of enrichment (see Section 1.1.2) and, given that there is no clear consensus about the success of the aforementioned types of enrichment, it warrants attention. Large felids and humans have been interacting both within and outside of zoos and other forms of captivity throughout history (discussed in more depth in Chapter Two). Although the potentially positive benefits of large felid-human interaction were first raised by Hediger (1964, 1968), research to date has largely appeared to ignore the influence that people might have on the welfare of these animals. For the purpose of this discussion, large felid-human interaction will refer to those interactions taking place between the animals and either zoo keepers (familiar humans) or zoo visitors (unfamiliar humans). Although keepers are a constant factor in zoo animals environments, human interaction has not been scientifically explored as an enrichment technique, despite it having the potential to be enriching. Further, little is understood about the influence that zoo visitors might have on these animals. It is both interesting and concerning that, despite the constant presence of people in the lives of large felids 43

62 housed in zoos, the literature has ignored the influence humans may have on them. Understanding the impact - be it positive, negative, or neutral - of human presence/interaction on large felid welfare will be important for providing an optimal environment for zoo animals Familiar humans Mellen and Shepherdson (1997) highlight that keepers play a significant role in the welfare of zoo-housed felids. However, despite the roles they play in maintaining animals, there is a lack of information about how zoo keepers might be enriching for large felids. There are two key ways in which they might achieve this: (1) through different handling styles and (2) through positive reinforcement training Handling style At present, keepers appear to be engaging in three levels of handling with their large felids, defined here as: a) Hands-on contact: physical contact between large felids and keepers, with no protective barrier, including keepers entering large felid enclosures with animals present, or taking the animals for walks (e.g., on leads/harnesses); b) Protected contact: contact between large felids and keepers with a protective barrier between the two (e.g., interacting through a mesh fence); c) Hands-off contact: situations where there is no contact between large felids and keepers, beyond that which is necessary, such as feeding and shifting between areas (e.g., enclosures). The benefits of any of these practices have not been specifically explored with large felids, although there are aspects of large felid-keeper interactions discussed in the anecdotal literature (Dorfman, 2005; Poulsen & Miller, 1996). For example, documented interactions have occurred at the Calgary Zoo, where a tiger-tug was used as a play device between tigers and keepers through protective fencing (Poulsen & Miller, 1996). Poulsen and Miller (1996) explain that this was a source of increased interest for the tigers, with individuals that did not previously show interest in keepers 44

63 actively soliciting play when keepers were present after the provision of the tiger tug. Despite being anecdotal, this example of tigers voluntarily initiating contact highlights a positive benefit of protected contact interactions for the stimulation of these animals by providing opportunities to engage in behaviours and activities that were previously unavailable Training Employing either a protected or hands-on approach also allows keepers to directly engage in training with their large felids. Training, defined here as any planned and targeted procedure whereby a zoo keeper encourages the performance of specific behaviours in an animal, has long been practiced with many species (particularly nonhuman primates) and may be considered enriching (Claxton, 2011). Based on principles of operant conditioning, it usually takes the form of positive reinforcement training, where animals are rewarded (rather than punished) for performing a desired behavioural response (Laule & Desmond, 1998). This process therefore relies on the voluntary cooperation of the subject (Bassett & Buchanan-Smith, 2007). In what appears to be the only empirical documentation of training for large felids, Broader et al. (2008) describe the use of positive reinforcement training with a seven year old female snow leopard at Potter Park Zoo in the United States. The purpose of this training was to condition the animal to receive transabdominal ultrasound to monitor pregnancy, without the use of anesthesia. The authors describe in detail the steps in the training process (from habituating the animal to a holding enclosure, through to finally being able to perform the ultrasound), which took months to complete. In addition to these details, which would be of use to others wanting to start a training program with their large felids, the authors stress the importance of some key factors about the process which were thought to contribute to success. Firstly, identifying a reward which the animal is highly motivated to work toward is essential to maintaining its interest. Secondly, keepers play a vital role in the training process, in that being able to build a trusting relationship with the animal prior to training was thought to make it possible for keepers to monitor the disposition of the animal and determine whether she was willing to cooperate. Thirdly, maintaining accurate records 45

64 of the training sessions and the communication of information between keepers is important in terms of ensuring consistent knowledge for all involved in the animal s training. Since this individual animal had no prior experience with training, these findings are useful for highlighting how to initiate a successful training program, as well as providing evidence for the benefits of training for improving animal health care. However, what this paper does not indicate is whether or not training could be enriching. Other than Broader et al. s (2008) study, any other information about training practices with large felids appears to be primarily anecdotal. Searches of zoo websites will also reveal that various institutions worldwide are practicing training with their felids (Australian examples include Australia Zoo and Dreamworld). Further, both Medina (2005) and Thomas's (2004) conference proceedings discuss the use of public tiger training sessions at the Bronx Zoo, where visitors are able to watch keepers training tigers through mesh fencing. Countless video footage of such practices is also available on websites such as YouTube. The literature heavily supports training practices with non-human primates, in both zoo and laboratory settings (Bassett & Buchanan-Smith, 2007; Bloomsmith & Else, 2005; Bloomsmith, Stone, & Laule, 1998; Colahan & Breder, 2003; Laule, Bloomsmith, & Schapiro, 2003; Reinhardt, 2003; Savastano, Hanson, & McCann, 2003; Schapiro, Bloomsmith, & Laule, 2003). However, aside from one brief anecdotal mention of a training program for a tiger at Portland Zoo (Mellen & Shepherdson, 1997) and the findings described by Broader et al. (2008), whether the same benefits apply to large felids remains unknown Combining the influence of handling and training In an international, multi-institutional survey of enrichment practices with zoo housed mammals, Hoy et al. (2009) found that over 80% of respondents (n = 238) reported human-animal interaction to be either important or very important for the animals. Although research has yet to explore this for large felids, studies with other species indicate that zoo animals may benefit in a variety of ways from increased interactions and training with their keepers/carers. Primarily, the literature highlights that animals may benefit in terms of improved physical well-being, since keepers are better able to 46

65 monitor their animals for health and medical problems, resulting in earlier problem detection and treatment (Hosey & Melfi, 2012). In addition, increased physical wellbeing can be achieved with a reduced need for anaesthesia during medical treatment which, along with increased trust in keepers, may decrease the stress faced by animals during such procedures (Broader et al., 2008; Lambeth et al., 2006; Laule et al., 2003; McKinley et al., 2003; Phillips et al., 1998). On a similar note, both Mellen and Shepherdson (1997) and Swaisgood and Shepherdson (2005) discuss how increased keeper interaction can contribute to increased reproductive success in small felids. Whether or not these findings apply to large felids has not yet been explored. Moreover, if these benefits do apply, it may be that they are positive for welfare but not necessarily enriching (as per the definition in Section 1.2). These practices may improve psychological well-being, since a stronger bond between animal and keeper (which may be established over time with repeated, positive interactions during training and other interactive encounters) may reduce stress associated with living in the zoo environment (Bayne, 2002; Hosey & Melfi, 2012). This is evidenced by both behavioural and physiological measures. For example, various authors (Mellen, 1998; Mellen & Shepherdson, 1997; Swaisgood & Shepherdson, 2005) have discussed how time with a caretaker is negatively associated with pacing levels in small felids. Moreover, lower chemical stress levels have been found in hand-reared individuals compared to their mother-reared counterparts (Latham & Mason, 2008). Wielebnowski et al. (2002) explored this in clouded leopards (Neofelis nebulosa) housed at more than one dozen North American facilities, finding a negative relationship between animal stress levels (measured through cortisol levels) and the number of hours keepers spend with these individuals per week, and a positive relationship with the number of keepers caring for an animal. In addition, there is the potential to offer animals more choice and control over their environment (Bassett & Buchanan-Smith, 2007), particularly through training where they learn about the consequences of their behaviour and how they can manipulate their environment to obtain what they want (Laule & Desmond, 1998). This is not to say that animals have choice over the behaviours engaged in during training (as these would usually be predetermined, targeted behaviours planned by zoo keepers), but that they have choice to engage in training as an activity. However, it should also be considered that, if rewards 47

66 are offered (food or otherwise), then perhaps animals have limited choice and instead are encouraged to participate. Regarding the learning (and conditioning process) associated with training, this activity could also enhance the cognitive stimulation of animals since they are required to process the information they receive and use this information to make future choices (Laule et al., 2003; Laule & Desmond, 1998; Reinhardt, 2003; Mellen & Shepherdson, 1997; Swaisgood & Shepherdson, 2005). Interactive handling and training also have the potential for increased social stimulation through formation of bonds between large felids and keepers. Although many of these species are solitary (see Table 1.1), they may still benefit from the social aspects of interacting with humans. For example, given that their primary means of communication is through scent marking and scent inhalation behaviour (Clark & King, 2008; Kitchener 2000; Sunquist & Sunquist, 2002), the additional sensory stimulation involved in keeper interactions and training could be beneficial. Interactive practices may improve behavioural stimulation, since training can be used to encourage behaviours exhibited in natural environments, many of which are redundant in zoos (Kirkwood, 2003). For example, at Australia Zoo (Queensland, Australia), tigers are encouraged to climb trees to obtain food rewards (personal observation). Without the food incentive, this behaviour would not normally be needed in the zoo environment since food is usually provided routinely and does not need to be hunted. Training can also be used to reduce negative behaviours, such as aggression and stereotypy (Laule et al., 2003; Prescott & Buchanan- Smith, 2003; Savastano et al., 2003). However, this potential for training to be enriching in regards to encouraging engagement in speciestypical behaviour would likely be limited to the type of behaviour being trained. For example, the animal benefits associated with training an a cheetah to 'chase' a lure to stimulate hunting behaviour are likely to be greater than training a cheetah to open its mouth to assist with dental checks. It is clear that large felid-keeper interaction has the potential to be enriching, however, since the vast majority of the literature supporting these benefits has been conducted 48

67 with non-human primates (which differ in physiology, behaviour, and needs from large felids), there is a timely need to explore whether such benefits do indeed apply to large felids, as well as to bring to light any potential dangers, or otherwise negative implications, for these animals. Moreover, enrichment is just one way that these practices might be beneficial and therefore these effects need to be better understood Unfamiliar humans Human visitors are another constant factor in the zoological environment, but it has been argued that since these people only spend a short amount of time at exhibits, it is unlikely that animals form relationships with individual visitors (Hosey, 2008). As such, much of the literature has tended to focus on the negative influence that visitors have on zoo animals, commonly referred to as the visitor effect. The visitor effect has been widely acknowledged, with Davey (2005, 2007) highlighting that various visitor characteristics including presence, density, activity, size and position/proximity can be linked to changes in animal behaviour, as well as physiology. Again, research to this end has primarily focused on non-human primates, to the exclusion of most other species, and Davey (2007) highlights that the visitor effect is not the same for all animals. In support, after an extensive review of the literature, Hosey (2008) argues that zoo-housed felids are less susceptible than primates to the visitor effect in terms of their behavioural responses. Thus there is a need for more research to explore this with large felids. The handful of studies that have directly explored the influence of visitors on large felids offer conflicting ideas 4. Two decades ago, O Donovan et al. (1993) found no significant changes in the behaviour of female cheetahs and their cubs (n = 14) in response to visitor presence. More recently, Margulis, Hoyos and Anderson (2003) found no variation in the activity of six captive felid species (n = 14; including lion, leopard, tiger and snow leopard; as well as clouded leopard and fishing cat [Felis viverrinus]) in response to the presence or absence of visitors, concluding that these 4 For detail regarding the methodologies employed in these studies, see Table

68 particular animals did not appear to respond to visitor disturbance or attempts to engage with them. On the other hand, Mallapur and Chellam (2002) found that leopards (n = 14) housed at four zoos in southern India displayed significantly higher levels of activity and significantly lower levels of resting on days when human visitors were absent compared to days when they were present. Similarly, Sellinger and Ha (2005) found that both visitor density (number of visitors) and intensity (behaviour of visitors) significantly influenced the behaviour of two captive jaguars (male and female) at Woodland Park Zoo, Seattle. In particular, these variables were associated with time spent out-of-sight by both animals, and visitor intensity was associated with pacing in the female. Further, although non-significant, aggressive behaviour in the male showed a trend for association with changes in visitor density and intensity. Recently, Maia, Volpato and Santos (2012) explored the behavioural responses of two female puma to visitor noise and numbers at a Brazilian zoo, with both animals displaying increased inactivity with higher visitor noise levels. These authors also contribute to this growing body of research by revealing that this effect varied by time of day, such that behaviour only showed this variation in the mornings (and not the afternoons). At present, with these studies offering conflicting ideas, there is not enough information to accurately determine how large felids respond to zoo visitors. Like other variables in the zoo environment, it is clear that visitors have the potential to influence the behaviour of these animals. However, it will take much more research to determine how they do this and what the nature of their influence is. Despite this limited understanding of the influence of visitors at exhibits, zoos are increasingly introducing interactive visitor tours, with visitors entering behind-the-scenes animal areas. Such tours are made possible by the increased large felid-keeper interactions and are likely to be established in order to encourage greater support for conservation among the visiting public and to increase revenue, but it is not known how the animals might respond to this (discussed further in Chapter Five). It will be important to explore this, since forced proximity to humans has been argued to be potentially stressful for zoo animals (Morgan & 50

69 Tromborg, 2007). Conversely, it is also possible that the variability provided might be stimulating and, therefore, enriching The combined influence of familiar and unfamiliar humans Combined, familiar and unfamiliar humans might influence the overall response that a zoo animal has to people. Hosey (2008) has explored this in depth and proposes a theoretical model for the interaction of these experiences (see Figure 1.1), based on understandings of the human-animal relationship from farm animal literature. Although not specific to large felids, this model is important in acknowledging the way people combined (familiar and unfamiliar) might influence a zoo animal. Hosey (2008) theorises that the relationship an animal has with unfamiliar humans will influence its relationship with familiar humans, and vice versa. For example, an animal who typically has positive interactions with keepers and visitors is likely to exhibit low fear of humans, display confidence around people (familiar or unfamiliar), and may be enriched by them. Conversely, those who have typically negative interactions with keepers and visitors are more likely to experience a high level of fear and stress in the presence of humans. Hosey (2008) suggests that the way an animal responds is also largely dependent on the type of species and its fear of humans, as well as enclosure design, extent of early handling, and temperament/personality. Although this model will not be tested specifically in this thesis, it is useful in highlighting why it is so vital to explore the large felid-human relationship in the zoo setting and why it is essential to explore animal responses to both familiar and unfamiliar humans. Further, Hosey (2008) stresses that this model could be useful in optimising the experience for all involved in such interactions. 51

70 IMAGE REMOVED IN ELECTRONIC VERSION OF THESIS DUE TO COPYRIGHT REGULATIONS. PLEASE REFER TO HOSEY (2008). Figure 1.1. A proposed model of human-animal interactions and their consequences for human-animal relationships in zoos. Sourced from Hosey (2008, p. 118, Figure 1) Elsevier Conclusion This review has indicated that there is currently a moderate body of literature surrounding environmental enrichment for zoo-housed large felids, yet the focus has been placed on tigers, lions and cheetahs, and the majority of efforts have centred on feeding and hunting strategies. Since it is believed that felids habituate quickly to novel conditions (Mellen & Shepherdson, 1997), it is pertinent to continue to evaluate the effectiveness of as many enrichment techniques for these animals as possible so that these can be employed by zoos. If large felid populations are limited to captivity in the future, it is imperative that the information surrounding how best to enrich them in zoos is updated. This is paramount given the potential negative effects of zoo living (see Section 1.1.2). Additional scientific research of current enrichment and management practices for these species could greatly contribute to this area, and a greater focus is needed on the underdeveloped areas. In particular, we know that large felids can benefit from feeding, hunting and olfactory techniques, and, to a lesser extent, appropriate housing conditions; we now need to focus our attention on social enrichment, particularly in the form of human interaction. Given apparent increases in practice, we can no longer ignore the influence of people on animal well-being in the zoological environment, and the potential benefits and limitations need to be investigated. 52

71 1.6. Aims and structure of this thesis This review of the large felid enrichment literature has led to the conclusion that one of the notable gaps in knowledge concerns interactions with humans. Given the contrast between this lack of information and the apparent increasing occurrence of such interactions (discussed in Chapter Two), the overarching aim of this thesis is to begin to explore how large felids and humans might benefit from, or be adversely affected by, interacting with each other in the zoo environment. It is recognised that zoos serve multiple user groups primarily, their animals, their zoo keepers, and the visiting public and therefore this thesis seeks to explore large felid-human interactions from these three perspectives. In particular, this research aims to understand: what zoo keepers can tell us about interacting with large felids and what their perceived benefits and limitations of interactions are, not only for themselves, but also for their animals and visitors; how large felids respond (behaviourally) to interacting with people, particularly during zoo visitor tours; and what impact such interactions have on zoo visitors, specifically in terms of meeting zoo goals of enhancing knowledge, attitudes, emotions and proposed conservation behaviour. Addressing these three perspectives was deemed essential to being able to provide a balanced and holistic understanding of the topic, as well as to be able to generate useful feedback to zoos currently using, or looking to adjust their use of, these different practices. In this light, a key aim of this program of research is to publish the findings of the studies in international peer-reviewed journals, in order to make a widespread contribution to this underdeveloped field. This has been achieved, with three chapters published in: Applied Animal Behaviour Science (a variation of Chapter One), Zoo Biology (Chapter Four) and the International Journal of Comparative Psychology (Chapter Five). Chapter Six and an alternative version of Chapter One are currently in preparation. As a result of writing much of this thesis for publication, many of the chapters have been structured and presented as manuscripts herein, which results in some overlap of information in places (particularly in the introductory and concluding sections). 53

72 2 The evolving large felid-human relationship: The need to promote positive interactions in zoos. Felids are widely depicted in art, from Stone Age petroglyphs and cave paintings to more modern depictions of cats as art, reminding us that humans have interacted with felids for as long as we have been humans. A.J. Loveridge, S.W. Wang, L.G. Frank and J. Seidensticker (2010, pp ) The previous chapter reviewed the environmental enrichment literature for large felids, highlighting the need for research to explore large felid-human interactions in zoos. The zoo setting is, however, just one environment in which these interactions occur. In order to place this thesis within the wider context, this chapter provides an overview of the large felid-human relationship 5 in various settings. The chapter begins with a brief overview of what is known about large felids in human history. It then compares and contrasts the current relationship between large felids and humans in felid home countries and in Western 6 settings, in order to reinforce the need to understand how best to promote positive interactions in all shared environments and, in particular, the role that zoos play in this Introduction Fossil evidence indicates that the earliest lion-like cat existed in East Africa during the Late Pliocene, some 5.0 to 1.8 million years ago (Barnett, Yamaguchi, Barnes, & 5 For the purpose of this discussion, the term 'relationship' will primarily be used to highlight relationships between large felids and humans as species (i.e., rather than as relationships between an individual large felid and human). In this context, a relationship can be both positive and negative in nature. 6 From hereon in, unless otherwise specified, the word 'Western' is used to denote developed countries 54

73 Cooper, 2006), and Sunquist and Sunquist (2002) suggest that cave paintings, from over years ago, show that the earliest lions existed alongside our earlier human ancestors. This historical relationship was largely characterised by humans using large felids for purposes such as hunting. More recently, with the expansion of the human population and the subsequent overlap of large felid and human territories, both have come into increasing contact with each other, and the nature of interactions between the two has evolved. Guggisberg (1975, p. 291) surmises that in his physical contacts with wild cats, man has either been a prey, a competitor, or an exploiter, and this can be seen when exploring the early/ancient history of the relationship between the two. This chapter will address the historical relationship between large felids and people in this way, through an exploration of the threats these animals pose to humans and the exploitation of large felids by people, particularly through hunting and entertainment. Almost forty years ago, Guggisberg (1975) suggested that the newest state of the relationship between the two is one of fascination, and this appears to hold true today, becoming particularly noticeable when examining the rise of safari-style tourism in large felid home countries, increased conservation research efforts and, perhaps most significantly, when exploring large felids in Western forms of captivity. In particular, the rise of the modern zoo in the late 1800s saw opportunities for the general public - not just the royal, wealthy, or elite to interact with these animals outside of their natural home ranges. Thus, the 20 th Century saw a noticeable divergence occurring between interactions with these animals in the wild and those interactions in captivity (including both zoos and circuses). This leads to the current state of this relationship, which continues to see a striking contrast between the relationships in these two different environments A snap-shot of the large felid-human relationship in early history In what is recorded about their earliest history, it appears large felids were both highly revered and exploited by people. This is evident when examining their role in key ancient societies, particularly those of Egypt, Mesopotamia, Asia, Greece and Rome, and, to a lesser extent, central and South America. In essence, encountering large felids appears to have been reserved almost exclusively for royalty and the wealthy elite, with the appeal of the unknown giving rise to interest in such animals as gifts, which were 55

74 used to show signs of influence and wealth. As Jennison (2005, p. xi) highlights, few presents which could be made to Kings or other rulers were so cheap, so effective, or so highly appreciated as a gift of wild animals. In support, many historical artefacts reveal that large felids were often kept as pets in private collections. Despite these eras indicating that large felids may have been exploited for human purposes, inherent in this is that these animals appear to have experienced varying degrees of training and, in a sense, taming. This is important, since it may highlight that these animals have experienced this form of interaction with humans throughout their history. A brief summary of this relationship in key historical periods is provided here. The information presented in Sections is largely derived from peerreviewed sources (i.e., journal articles and edited texts). Despite being scholarly sources, it is important to note that often these (and any secondary sources used) rely on art works and other artefacts to evidence the arguments made. While it is recognised that art reflects an interpretation (both of the artist and the viewer of the art), in the absence of other forms of recorded communication on the topic, it appears necessary and appropriate that such information be used. However, some caution should be taken when considering this information. Additionally, where secondary sources have been used, these are largely written by those with expertise in the area, such as renowned felid experts or historians (e.g., 'classic' felid texts, such as those by Guggisberg and Sunquist & Sunquist; and recent books about large felids in India, Exotic Aliens and Tiger Fire, by wildlife conservationist, Thapar). Lastly, with the majority of large felid natural habitats being in countries where English is not the first language, it is possible that much of what is known about historical relationships has been captured in works published in languages other than English. In this way, the content of this chapter is limited to a perspective gained from English materials only. In light of these limitations, the information herein provides only a brief overview of large felid-human interactions, and the examples used to evidence these interactions cannot be assumed to hold true for all large felid species, or for all interactions and relationships, during these times. 56

75 Egypt According to Kisling (2001), the ancient Egyptians were passionate about keeping animals. Guggisberg (1975) also attests that animals frequently appeared in their culture. Examples include the cheetah portraying a goddess (Marker-Kraus, 1997), and pharaohs depicted as lion-bodied sphinxes which were seen to guard the Nile and the pyramids (Guggisberg, 1975). It has been said that felids were regarded as highly sacred in this era (Kalof, 2007). A recent archeological finding provides some confirmation of this, with an article in Nature describing the 2001 discovery of a once-mummified lion skeleton in the Egyptian tomb of Maïa (Callou et al., 2004). Lion and cheetah images were also used heavily in art, including the decoration of tombs and rock temples (Sunquist & Sunquist, 2002). Numerous royal animal collections containing these large felids were maintained by pharaohs in this region, and were later maintained by conquering Greek Ptolemaic and Roman rulers (Labuschange & Walker, 2001). Many of these animals were said to have been tamed and domesticated (although specific methods for this are not described), and were primarily used to meet human needs, such as to assist with hunting (Kisling, 2001; Sunquist & Sunquist, 2002). The cheetah was commonly used for this purpose, with early records of cheetah training coming from the 17 th and 18 th Egyptian dynasties ( BC; Sunquist & Sunquist, 2002), a tradition continued by parallel and subsequent societies. For example, art works from the Renaissance period (from the mid-1400s) reveal how cheetahs were carried out into the hunting fields on horse-back, perched on a pillow behind the handler. They were then released to run down hares and roe-deer (Guggisberg, 1975, p. 266). Egyptian royalty were also said to have hunted large felids for sport (particularly leopards and lions; Kisling, 2001). As an example, Guggisberg (1975) describes how, in the first 10 years of his reign, the Pharaoh Amenophis III ( BC) reportedly killed 102 lions Mesopotamia Occupying the area within what we now know as Iraq, Mesopotamia is another example of an ancient society in which large felids featured. Kisling's (2001) account of early zoos suggests that lions were often kept in Mesopotamian gardens and, similar to those in Egypt, some royalty kept tame lions as pets, whilst others used them for hunting, 57

76 fighting and exhibitory purposes. These animals had largely been exported to this region from ancient Indian and Chinese cultures, and were given to Kings as gifts (Dalley, 1993; Kisling, 2001). Again, evidence of the keeping of these animals can be seen in artefacts from this time (see Figure 2.1; Dalley, 1993). In particular, Marker-Kraus (1997) asserts that the earliest record of the cheetah s history with humans dates back to 3000 BC with an image of a leashed cheetah depicted on an official Sumerian seal. The cheetah being leashed in this image further indicates that the desire for people to tame and train these animals for hunting purposes occurred very early in human history (Guggisberg, 1975). IMAGE REMOVED IN ELECTRONIC VERSION OF THESIS DUE TO COPYRIGHT REGULATIONS. PLEASE REFER TO KISLING (2001). Figure 2.1. An image of a lion being released in to a King s animal park at Mesopotamia, featuring as part of a relief on a palace wall. Sourced from (Kisling, 2001, p. 10) CRC Press Ancient India and China Thapar, Thapar and Ansari's (2013) recent book has challenged the previously held belief that lions and cheetahs were native species in India. Through an extensive analysis of historical accounts and artefacts, these authors demonstrate that early accounts of large felids in the Indian subcontinent were restricted to tigers and leopards. Moreover, their account suggests that lions and cheetahs were imported from Africa and Persia over five centuries or more by Indian royal families (e.g., for menageries and 58

77 hunting). In a review of this book, Packer (2013) agrees that the lions of Gir are exceptionally tame, and he concludes that the biology of large felids in India must be reassessed. Regardless of these recent views, various authors have described the way people and large felids encountered each other in ancient India and China. Kisling (2001) suggests that, since many of the large felids (e.g., tigers and leopards) are native to these regions, these ancient societies were active in trading these animals to the elite in other regions (Kisling, 2001). In addition, there are indicators of positive relationships between large felids and humans in these eras. For example, during the Vedic period in India (from about 3,000 BC), carvings of animals - including tigers - have been found on archaeological relics, with Walker (2001b, p. 253) suggesting that this art reflects an affectionate and respectful partnership. Further, the Chinese were said to see the tiger as sign of power and strength (Sunquist & Sunquist, 2002), a belief that continues today. In contrast, as in other ancient cultures, these animals (particularly lions and leopards) were used for fighting in ancient Asia - either with humans or each other - and cheetahs were regularly used for hunting (Kisling, 2001). They were also held in early forms of captivity (Schafer, 1968), and apparently were frequently maintained in large numbers, which also continued through later societies. Walker (2001b) provides an interesting account of how, in more modern times, Indian emperors in the 1500s s kept, trained and nurtured thousands of animals, particularly cheetahs. Significantly, a Mughal Emperor, Akbar, is said to have collected at least 9,000 khasa (imperial cheetah) during his reign (Walker, 2001a) Ancient Greece and Italy The ancient Greeks apparently imported lions in the thousands from north Africa as personal pets, for menageries and for public spectacles (Sunquist & Sunquist, 2002). Jennison (2005) suggests that foreign animals began to appear in Roman games somewhere near the start of the second century BC, and much of the recorded history of lions, tigers, and leopards during this time highlights that their primary purpose was for entertainment in the gladiatorial arena (Kisling, 2001; Kyle, 2001). Although the Italians were said to have initiated gladiatorial shows, the use of new animal species and animal care methods came from the Greeks, with these animals often exhibited as tamed and trained creatures (Jennison, 2005). In order for animals to perform in this arena, Jennison (2005) suggests that the training/taming methods used were remarkably similar to those used much later by European animal trainers in the early 1900s. While 59

78 Jennison's (2005) view here can only be based on his interpretation of what is recorded about this early history, he was in a somewhat informed and experienced position to make such a claim, given that, during the late 1800s/early 1900s, his family owned Belle Vue Zoological Gardens in England, of which he also became director 7. Again, such activities were most common among the wealthy (Guggisberg, 1975; Kisling, 2001) but apparently occurred in almost all Roman cities (Kalof, 2007), made possible by the close proximity of Italy to Africa and the control Rome had over north Africa in the second century (Jennison, 2005). The use of large felids for these purposes saw significant numbers of them subsequently killed as they fought against each other and against people (Kyle, 2001; Sunquist & Sunquist, 2002). For example, Guggisberg (1975) describes an instance where 51 tigers were apparently killed as part of a marriage celebration Northern, Central and South America Much less appears to be documented in scholarly sources about the historical relationship between people and large felids native to Northern, Central and South America namely, the jaguar and the puma. However, the limited information available does suggest a long-standing relationship similar to that experienced by large felids on the other continents. Again, artefacts provide some evidence of this, such as 5,000 year old sculptures found in Peru, displaying jaguars (Sunquist & Sunquist, 2002). This species was apparently central to religion and culture in this region, often associated with thunder, lightning and rain in South America, and seen as a symbol of power by the Incas; it also played a key role in hunting rituals (Sunquist & Sunquist, 2002). Although historical records prior to arrival of Europeans are virtually non-existent, the puma appears to have a long history of association with humans. The puma featured in Native American myths, legends, ceremonies and a petroglyph (rock engraving) in Arizona dated between AD depicts an image of a puma (Gill, 2010). Some authors suggest that this large felid did not feature as predominantly in early images, legends and mythology in comparison to other carnivores in the region, such as wolves (Kellert, Black, Rush, & Bath, 1996). However, it appears to have featured in some 7 Note that although the publication date for this source is 2005, it is a re-print of the original text published in

79 stories, such as that of Maldonada - a girl who became lost on a search for food and was protected by a puma, which emphasised the species having a kind temperament, rather than being an animal to fear (Guggisberg, 1975) Flashing forward to the modern large felid-human relationship in the 20 th century What is known about the relationship between large felids and humans from Antiquity through to the Enlightenment periods (17 th & 18 th Centuries) appears to be characterised by these animals being seen as a sign of wealth and status, reserved mostly for the elite (Kalof, 2007). It is also reported that they continued to be used for purposes such as hunting, particularly the cheetah (Sunquist & Sunquist, 2002). Again, this represents a paradoxical relationship, with both positive (e.g., respect) and negative (e.g., exploitation) elements. As such, the start of the 20 th Century appears an appropriate place to re-examine and redefine the large felid-human relationship, since interactions with these animals were no longer limited to the elite. This period saw dramatic increases in human populations (Swim, Clayton, & Howard, 2011). This forced people and animals into closer contact, and resulted in increased conflict over resources (Inskip & Zimmermann, 2009), having severe impacts on wild large felid populations (i.e., see current status of these animals in Table 1.1). In turn, this appears to have contributed to a primarily negative large felid-human relationship in their home countries. Consequently, the discussion will now describe the differing nature of the relationship during this phase including the killing of large felids (as a result of sport hunting, retaliation in response to livestock and human depredation, and illegal poaching), as well as human fascination with them (demonstrated by the rise of safari-type tourism and research associated with the birth of modern wildlife conservation in the early 1900s (Ray, 2005)). This period also saw the emergence and growth of the modern zoo, with increased opportunities for the average person, particularly in urban areas, to encounter a large felid (among many other species). As such, the discussion will then shift to address the state of the relationship in captivity (including circuses and zoos), which reinforces both the exploitation of, and fascination with, these animals. 61

80 The killing of large felids Perhaps the most significant and widely discussed relationship between large felids and people is that both have been responsible for killing each other, often referred to as human-big cat conflict. Although more pronounced for the larger of the cats, this conflict is thought to affect over 75% of the existing felid species (Inskip & Zimmermann, 2009; Loe & Roskaft, 2004). This can be categorised in three key domains: sport hunting; retaliatory killing in response to depredation on humans and livestock; and illegal poaching for fur and body parts. Although these actions occurred in what is known about their earlier historical relationships (see Section 2.2), such killing appears to have been more heavily documented in scientific and popular media reports in recent times The rise of sport/trophy hunting The trophy hunting of large felids has occurred for thousands of years (Loveridge et al., 2010), however the increase in this activity as a sport appears to have been most pronounced from the mid-1800s through to the mid-late-1900s (see Thapar, 2013). From a human perspective, this activity was again dominated by the sign of wealth and status that it afforded. One of the most notorious examples of a highly targeted large felid is the tiger, which was regarded as a pest species in some of its natural regions during the mid-1800s to mid 1900s and hunted to the brink of extinction. For example, Tilson et al. (2004) cite other primary and secondary evidence to discuss how, in the early 1950s, there were reported to be over 4000 South China tigers existing in their natural range, but that this number drastically reduced (to approximately animals by the early 1980s), largely as a result of government 'anti-pest' campaigns (under the rule of Mao Zedong) which supported uncontrolled hunting of these animals. During their field survey of eight reserves in China between 2001 and 2002, these authors found no evidence of these animals existing in these regions (Tilson et al., 2004). Similarly, the increasing prevalence and widespread use of firearms was responsible for eliminating the Asiatic lion from all of its prior regions, except India (Nyhus & Tilson, 2010; Sunquist & Sunquist, 2002). However, as mentioned in Section 2.2.3, India may not have been the origin of these animals. It was not until later in the 1900s, after significant damage had already been done, that large felids began to obtain 62

81 protected status. Today, the majority of the large felid species (cheetah, jaguar, leopard, snow leopard and tiger) and some subspecies (Asiatic lion - Panthera leo persica; and Eastern and Central American puma subspecies - Puma concolor coryi, costaricensis, and couguar) are listed under CITES Appendix I, giving these animals protected status against commercial international trade (CITES, 2013). While this sport typically lacked regulation or control in the past, today it appears to have become more restricted and some suggest that it might actually benefit animal conservation (Loveridge et al., 2010). If managed properly, and the number of animals hunted is controlled and sustainable (e.g., through permits and quotas), various authors have argued that trophy hunting can significantly raise financial revenue, particularly in Africa with lions, leopards and cheetahs (Baldus, 2004; Balme et al., 2010; Loveridge et al., 2010; Marker, Dickman, Mills, & Macdonald, 2010). In particular, Mbaiwa (2004) highlights how tourism hunting can generate income for local communities, using Botswana as an example where local people can benefit if hunting concession areas are sub-leased and annual wildlife hunting quotas are sold to safari hunting companies. Termed community-based safari hunting, this approach is not without its drawbacks but - if local communities are empowered to learn the necessary skills to play key roles in such processes - it is thought to be sustainable (Mbaiwa, 2004). If uncontrolled, however, sport hunting may be highly problematic. Loveridge et al. (2010) point out that there is often a higher demand for adult males (perhaps due to the size and features of their pelts; e.g. lion manes) which can result in infanticide (since if the dominant lion in a pride is removed, succeeding males will often kill any offspring in the pride in order to be able to mate with the lionesses). In this way, trophy hunting can remain detrimental to large felid conservation, regardless of whether it decreases in occurrence Illegal poaching Loveridge et al. (2010, p. 186) highlight how human cultures have always prized products derived from large carnivores. Thus, the killing of large felids for the use of both their fur and/or body parts in traditional medicines is another example of the exploitative nature of human relationships with these animals. 63

82 Fur trade Historically, each of the large felids have been reportedly hunted for their pelts (Loveridge et al., 2010). Guggisberg (1975) suggests that the earliest record of this occurring is a painting of humans wearing leopard skins dating back to 6,500 BC, with subsequent art work continuing to show the use of large felid pelts in all ancient cultures. Although such items have reportedly been used and worn in various eras, it was not until the mid-1900s that the widespread desire for animal fur products emerged (Guggisberg, 1975). A prime example was the growing demand for tiger skins at the end World War II, when increased weaponry and transport sophistication reportedly made the task of obtaining pelts much easier (Sunquist & Sunquist, 2002). To further highlight the proposed increase in large felids killed for this trade, approximately 4000 jaguars were killed in South America for their skins in the early 1800s (Sunquist & Sunquist, 2002); later, in the early-mid 1960s, 15,000 jaguar skins were extracted for the fashion trade from the Brazilian Amazon (Loveridge et al., 2010). Although strict legislation has lessened the demand for these items in Europe and the USA today, there are still parts of the world where markets for these products exist (such as Eastern Europe and China; Loveridge et al., 2010). Currently, clothing made from leopard and tiger skins is said to remain popular among the wealthy in China and Tibet (Loveridge et al., 2010) Traditional medicines Humans have also prized other large felid body parts, particularly for their proposed medicinal benefits. The tiger is well-known for its use in traditional medicines and this is largely contributing to the decline of this species, despite legislation and bans imposed worldwide to eliminate illegal trade (Damania et al., 2008; Guggisberg, 1975; Loveridge et al., 2010; Tilson & Nyhus, 1998). Sunquist and Sunquist (2002) suggest that tiger bone has been used in such medicines for over 1000 years, having multiple uses in 25 different forms of medicinal drugs. These authors report that, in 1985, 110 pharmaceutical factories produced such medicines (Sunquist & Sunquist, 2002). These authors also report that, at the time of publishing their book (2002), traders were paying poachers over US$300/kg for tiger bone, and that snow leopard bones were reportedly 64

83 used as a substitute (Sunquist & Sunquist, 2002). With this source being over 10 years old, this cost may be very different today, but more recent evidence to support this appears to be limited. There is some evidence for the current use of tiger bone. For example, a face-to-face interview study of 1880 adults residing in seven Chinese cities explored tiger consumption behaviour (Gratwicke et al., 2008). Collected in 2007, the data reveal that almost half (43%) of the sample reported ever using some form of product allegedly containing tiger body parts. In particular, 38% of the sample reported using tiger bone plasters (that is, plasters applied externally to alleviate aches and pains) and 6.4% reported using tiger bone wine. For both products, over half of the participants (60% and 52% respectively) had reportedly used these in the past two-years. Main reasons for using both of these products included: to treat traumatic injury, to cure rheumatism, to replenish calcium, as an anti-inflammatory, to treat fragile bones, and to improve sexual capacity. While it is not known whether these products actually contained tiger bone, and although the results are limited to urban and not rural Chinese adults, this provides some indication of the use of tiger parts in China today. In addition to bone, there are multiple other uses for tiger body parts: the fat as a tonic for rheumatism and an aphrodisiac; the flesh for strength; the collarbone as a charm against evil; the heart for strength, courage and cunning; and the brain as a cure for laziness and acne (Sunquist & Sunquist, 2002). There is a notable lack of evidence to indicate that tiger body parts possess medicinal properties, and so challenging (mis)perceptions about the use of this animal in this way is part of the task being undertaken by the Global Tiger Initiative across the tiger s various home countries (Global Tiger Initiative Secretariat, 2013) Retaliatory killing Retaliatory killing in response to large felid depredation (on livestock and people) is also responsible for decreases in the population of these animals and a relationship with humans that is largely characterised by conflict. Depredation usually occurs when felids 65

84 disperse into human-dominated areas (particularly boundaries of protected areas), when people enter felid habitats (for purposes such as hunting, food, leisure, and tourism), when the natural prey population is depleted, and/or when felids are too sick, injured, or old to take their usual prey (Loveridge et al., 2010; Michalski et al., 2005; Sunquist & Sunquist, 2002). Although it has been argued that tracking the nature of this conflict is difficult, primarily due to incomplete depredation records (Nyhus & Tilson, 2004), this topic has been quite widely discussed in both the peer-reviewed and secondary literature, as outlined in the following sections Depredation on livestock In a systematic review of primary and secondary human-felid conflict literature (published between 1979 and 2007), Inskip and Zimmermann (2009) report that felid attacks on livestock have occurred more frequently than attacks on humans, with such attacks being reported to more commonly occur by the larger felid species. Due to their lack of anti-predator response behaviours, domestic livestock are thought to be more easily killed than similarly-sized wild prey (Palmeira et al., 2008). Not surprisingly, with the increasing overlap between large felid and human living areas (as a result of human population growth and expansion), depredation of livestock by each of these cats has commonly been reported across their home ranges (Sunquist & Sunquist, 2002). For example, Patterson et al. (2004) report on livestock depredation in ranches adjoining Tsavo National Parks, Kenya. It was found that, over a four year period, lions were responsible for 86% of attacks (total number of attacks = 312). The economic toll of the loss to livestock as a result of lions over this period was US$30,127. In comparison, the toll as a result of depredation by cheetahs was US$203. Given that data for this study were collected between 1996 and 1999, it is unclear what the financial impact of depredation would be in this, and other, regions today. Regardless, this information is useful for highlighting differences in rates and costs of depredation between large felid species. In a chapter in an edited text, Cavalcanti et al. (2010) synthesise the findings from several published and unpublished interview studies conducted by the authors between 2000 and 2008 to explore human-jaguar conflict in the Pantanal and the Amazon. With relation to depredation, some key findings were that the majority of those surveyed 66

85 reported depredation problems caused by jaguars, and that this species is the most detrimental to human livelihoods. In particular, greater losses were found for those with smaller ranches/herds, with the greatest reported loss being calculated at over US$18,000 for one ranch over a one year period, resulting from the loss of 80 calves (from a herd of 2000). Being a synthesis of various studies, it is not stated whether/how these losses were substantiated (it is implied that such losses were reported by ranchers and, given the negative attitudes to jaguars held by these people, as discussed within the paper, it could be that this loss is over-reported). Similarly, a more recent study explored the costs and benefits of sport hunting of leopards in a reserve in Mozambique (Jorge et al., 2013). As one component of this study, a survey of villagers (N = 158) found that, over a two year period (2009 and 2010), leopards reportedly depredated 11 goats, resulting in a total loss of US$440 to six households (~US$73/household). While this rate of depredation and financial cost may appear low, the authors highlight that the average household income was <US$100 per month and therefore this loss would have a noticeable impact. Differences in rates of depredation have also been found to vary by season. For example, Patterson et al. (2004) found greater rates of depredation by lions during the wet season. Additionally, Palmeira at al. (2008) explored seasonality in their exploration of cattle depredation by puma and jaguars in a cattle ranch in centralwestern Brazil (between 1998 and 2003). In contrast to Patterson et al. (2004), these authors found that while seasonal management of cattle resulted in more livestock being born during the wet season (to take advantage of food abundance at this time), rates of depredation were not associated with monthly rainfall, nor with births of calves. Abundance of prey and large felids is another factor likely to influence rates of depredation. A recent study of snow leopard depredation on livestock suggests that both relative abundance of large felids and their ungulate prey accounted for higher livestock depredation, indicating that prey recovery programs must be accompanied by increased protection of livestock (Suryawanshi et al., 2013). While the extent of the damage varies from region to region (Sunquist & Sunquist, 2002), it has obvious consequences for livestock owners (primarily, financial declines; see Palmeira et al., 2008) and it has been suggested that there is often no compensation for this loss (Baldus, 2004). 67

86 Depredation on people Large felids have not only been known to attack livestock but also humans and, according to Loveridge et al. (2010, p. 172), perhaps the most striking relationship people have with felids is that large felids occasionally prey upon, kill, or injure people. Again, as long as large felids and people have interacted, humans have sometimes become their prey. Guggesberg (1975) supports this by describing how a lion is shown to be grabbing a human by the throat in an 8 th Century ivory panel. However, whether or not this was the creator's reflection of something that actually happened cannot be determined. Guggesberg (1975) also suggests that tigers, lions and leopards have preyed upon people more commonly than other felid species. More recent accounts lend support to this. Various peer-reviewed papers and reports detail these occurrences, yet these are often limited to specific regions (e.g., see Baldus, 2004; Treves & Naughton-Treves, 1999), making it difficult to understand how prevalent depredation on people actually is for the different species (or altogether, for that matter). Some have attempted to collate this information. For example, Loe and Roskaft's (2004) review article sought to determine rates of attacks on humans by large carnivores (not limited to felids) by integrating scientific literature with interviews of wildlife managers, researchers, and park personnel (n = unspecified). These authors reported that, during the 20 th Century, estimated rates of attacks for large felids were as follows: 12,599 tiger attacks (in Bangladesh, China, India, Indonesia, Malaysia, Myanmar, Nepal, Russia, Singapore, Thailand, and Vietnam); 840 leopard attacks (in India, Nepal, South Africa, and Uganda); 552 lion attacks (in India, South Africa, Tanzania, Uganda, and Zambia); and 18 puma attacks (in Chile, Canada, and USA). While it is unclear how the literature search was conducted, or how interview participants were recruited, this appears to be relatively comprehensive attempt to understand many of these attacks have occurred. These findings are substantiated by Loveridge et al.'s (2010) more recent chapter in an edited text, which provides a thorough account of reported incidents in the wild, highlighting that tigers and lions are the most dangerous of these animals and their attacks on humans are the most widely publicised, followed by leopards. Attacks by 68

87 cheetahs and snow leopards do not appear to have been documented. In addition, Guggisberg (1975) has argued that, outside of Africa and Asia, despite many 'horror stories' of unprovoked puma attacks in North America, these are largely unsubstantiated or include incidents where the animal of concern was sick or injured. Similarly, Conforti and de Azevdeo (2003) argue that there have not been any reliable reports of attacks by jaguars on people (other than those which occur in situations where the animals were provoked, e.g., poaching situations). As with killing livestock, large felids are thought to attack people either due to forced proximity, decreased physical condition, or a combination of both (Guggisberg, 1975; Sunquist & Sunquist, 2002). As such, 'one-off' killing might occur when people are unintentionally encountered. Some large felids are also thought to become man-eaters, responsible for large numbers of human death (Guggisberg, 1975). Man-eaters are thought to be those animals who selectively attack humans, even when vulnerable livestock are available (Sunquist & Sunquist, 2002). Moreover, Gurung et al. (2008) describe tigers who have killed two or more people as 'serial human-killers'. Consistent with livestock depredation, man-eating is reported to be more common among tigers and lions and, to a lesser extent, leopards (Sunquist & Sunquist, 2002). Some infamous examples are the man-eating lions of Tsavo in the late 1800s who were proposed to have killed over 100 humans, a tigress who killed 25 humans in Northern India before being shot by Corbett (famous British hunter-turned-conservationist), and a leopard in the mid-1800s who is said to have killed in excess of 200 people over a three year period (Guggisberg, 1975). More recent discussions in the literature suggests that methods for such attacks include lions entering huts at night time, either forcing their way through the grass/mud walls, or jumping on the roofs (Baldus, 2004; Yamazaki & Bwalya, 1999). Some understanding has been added as to when such attacks might occur. For example, a retrospective analysis of 450 lions attacks occurring in Tanzania between 1888 and 2009 (based on government reports and confirmed via interviews with survivors and victims' families) showed that lions tended to attack people on dark or moonless nights, and in the days following the full moon when they are most hungry (Packer et al., 69

88 2011). Other research has explored the circumstances under which such attacks have occurred. For example, a study of tiger attacks in and around Chitwan National Park in Nepal between , found that of the 88 human deaths attributed to attacks by 36 tigers, almost half occurred when people were collecting fodder for their livestock, and the majority occurred within 1km of the buffer zone around the park (Gurung et al., 2008). Greater understanding of the context in which large felid attacks occur is likely to be important for minimising future occurrences (e.g., through education/awareness programs for individuals living in these areas) Responses by humans to depredation It is not difficult to understand that depredation can lead to negative perceptions of, and attitudes toward, large felids among local people (for more detailed discussions, see Baldus, 2004; Oli, Taylor, & Rogers, 1994; Suryawanshi et al., 2013). It is also not surprising that people sometimes retaliate as a result of threats to loss of livestock, family members, or themselves. The financial hardship associated with a large felid attack can be severe and it appears that many people in home countries are willing act to eliminate the problem that is, eliminate the responsible animal (Loveridge et al., 2010). In the past, relevant governments advocated for this (in some instances, bounties were even offered), with the animals considered vermin/pests, subsequently resulting in significant large felid culling (Sunquist & Sunquist, 2002). In addition, persecution largely continues because, in many regions, there are neither the funds nor staff to enforce legislation, even if it does exist (Sunquist & Sunquist, 2002). Means to eliminate problem individuals include shooting, spearing, trapping or poisoning (Baldus, 2004; Sunquist & Sunquist, 2002; Yamazaki & Bwalya, 1999). Retaliatory killing can have negative consequences for the particular animal population, as many other animals are often killed before the offending one is (Goodrich, 2010; Loveridge et al., 2010). Presently, what is reported suggests that compensation for such loss is often inadequate or non-existent (around US$30-50 in Tanzania; Baldus, 2004; Loveridge et al., 2010), but a number of potential solutions to the issue of depredation have been widely discussed and some have been met with success. Potential solutions include: design of conservation policy and building management capacity; lethal control or translocation 70

89 of problem felids; improved protection and husbandry of livestock; compensation for losses; alternative livelihoods, benefit sharing and stakeholder/local participation; and zonation of land use (for detailed discussions of these approaches, see Goodrich, 2010; Loveridge et al., 2010). It has been argued that unless more attention is focused on how to reduce large felid-human conflict, it will only increase if other conservation initiatives are successful in increasing numbers of free-ranging large felids, therefore making such efforts redundant (Goodrich, 2010; Tilson & Nyhus, 1998). Given the diversity of habitats in which these animals exist, there is not likely to be a 'one-sizefits-all' approach to reducing depredation; instead, it is more likely that a multi-strategy approach, developed with the input of local people and tailored to the specific needs of those living in a particular location, would be most effective. Such a strategy should also focus on what the benefits of these approaches are for local people and how they could be encouraged to partake. For example, one study explored perceptions of jaguars and pumas held by residents living around Iguacu National Park, Brazil (Conforti & de Avezedo, 2003). Through a survey of 75 households, it was found that, when asked if they would be willing to alter their current livestock husbandry practices to prevent depredation, over 70% agreed, particularly if financial support was provided to assist with this. Perhaps one starting point, which underlies the various strategies for reducing depredation, is a better understanding of human attitudes towards these animals and the 'problems' they cause. This understanding could be used to better direct anti-depredation initiatives. For example, although it has been suggested that acceptance for free-ranging large felids in areas where they come into contact with humans is typically quite low (Baldus, 2004), some evidence has been provided to indicate that this is associated more with the perceived, rather than actual, threat posed by the animal (Conforti & de Avezedo, 2003). Further, some studies suggest that not all local people have negative attitudes towards these animals, despite the actual risk of depredation, and many are supportive of conserving the species (Gurung, 2008; Kellert et al., 1996; Lagendijk & Gusset, 2008). This is thought to be associated with various factors, such as how people have interacted with the animals historically (e.g., attitudes towards pumas appear to be quite ambivalent due to their lack of significance in history compared to jaguars and wolves; Kellert et al., 1996), and individual education level (with more favourable 71

90 attitudes to predators associated with higher education level; Lagendijk & Gusset, 2008). Regardless of these findings, depredation is still a defining characteristic of the large felid-human relationship in home countries Fascination with large felids in home countries Despite the typically negative relationships described within this discussion so far, Guggisberg (1975) suggests that people are also fascinated with large felids. This is illustrated, although not exclusively, by human interest in safari-type tourism, as well as by increased field research of large felids and their conservation Wildlife tourism and safaris Peoples fascination with large felids is clearly evidenced by the rise of wildlife tourism and safaris, and the willingness people have to travel to reserves with the hope of catching a glimpse of one of these animals (Guggisberg, 1975). Wildlife tourism is notoriously popular in Africa. For example, although a somewhat dated statistic (yet still relevant when considering modern interactions), Kenya has been said to attract over one million international tourists annually for wildlife safari and nature-based tourism (Akama & Kieti, 2003). There is also more recent evidence to suggest the popularity of these animals compared to other species. For example, a study conducted over a four-day period in 2005, at Kenya's Amboseli National Park, tracked 70 tourist vehicles for 30mins each, with a particular focus on viewing time with reference to different species (Okello, Manka, & D'Amour, 2008). It was found that, although tourist vehicles only spent 28.5% of their time actually stopped to observe animals, lions were observed for the greatest proportion of this time (28.6%) compared to any other species. Cheetahs were the fifth most popularly observed animals (accounting for 7.3% of stopping time). Although this study is limited to a short duration of observation days at one national park, which may mean that potential differences in animal visibility at different times (e.g, different seasons) would not have been captured, it is useful for highlighting the popularity of these species. Similarly, in India, more than one million people are reported to visit tiger reserves every year (Karanth & DeFries, 2011). Wildlife tourism is seen as highly important as it can serve to fuel the expansion of 72

91 large felid habitats (Karanth & Chellam, 2009). It also has the potential for these flagship species to contribute to wider biodiversity conservation, since they often exist in, and rely on, protected areas which are shared with other species (Walpole & Leader- Williams, 2002). It has been argued that for wildlife tourism to be locally beneficial and attractive, the income generated must benefit the associated communities (Walpole & Leader- Williams, 2002). To a degree, it does seem to be presently meeting these objectives. For example, in a 2006 survey of 20 South African based wildlife tour operators, Spenceley (2007) found that nearly all reported having a positive impact on local communities, in terms of providing economic benefits through employment and using local services/products, as well as to local education, health and conservation initiatives. If local communities are able to see and experience such financial benefits, tourism can play an important role in changing their attitudes towards living with large felids which, in turn, can promote a positive relationship and contribute to aiding their conservation (Walpole & Leader-Williams, 2002). Although wildlife tourism has long existed in Africa and is promoted by the state, its occurrence in South Asia is much more recent, and less is known about how it can benefit local communities (Sekhar, 2003). Of note, Sekhar (2003) explored attitudes of local people towards wildlife tourism in Sariska Tiger Reserve, India, through the use of a close-ended questionnaire (administered to 180 households across 18 villages). Only one quarter of respondents reported benefiting (directly or indirectly) from tourism, with the key benefit being associated with the sale of milk and vegetables to restaurants that catered for tourists. Regardless, almost 70% of respondents were supportive of conservation (with those that did benefit showing significantly more positive attitudes to tourism), but 62% reported wanting to benefit more, particularly through employment and income generated from marketing local products. Consequently, Sekhar (2003) argues that tourism in South Asia needs to be improved, with a greater focus on the benefits to local people in order for it to be sustainable and successful. Achieving this could serve to improve the large felid-human relationship in this region. 73

92 Research efforts towards conservation Bulbeck (2005) highlights how, over the past century, we as humans have recognised that we owe animals the protection of their environment. Reflective of this, and of human fascination with large felids (Guggisberg, 1975), research has increasingly focused on the conservation of these animals in the wild, representing yet another shift in the large felid-human relationship. Uncovering and understanding felid behaviour in natural habitats is said to have become easier, through advances in methods such as radiotelemetry and satellite telemetry, camera trapping, molecular scatology and hair sampling (Sunquist & Sunquist, 2002; Wang & Macdonald, 2009). Thus, we have seen research into the origin (i.e., physical and genetic) and diversity of these species (Barnett et al., 2006), their ecology (Harmsen et al., 2010), and their presence and density in home countries (Gros, 1998; Janecka et al., 2008; Johnson et al., 2006; Linkie et al., 2003; Xiaofeng et al., 2011). Research has also explored the large felid-human relationship in varying contexts, such as studies of tourists in game parks (e.g., Okello et al., 2008), local human attitudes toward large felids (e.g., Cavalcanti et al., 2010; Kellert et al., 1996), and how these animals respond to living closer to humans (e.g., Kerley et al., 2002). Conservation awareness and increased scientific specialisation in more recent times (Kisling, 2001) could be largely responsible for increasing research efforts to conserve large felids. With the aforementioned literature largely published by scientists and researchers working in Western or developed countries (as per the author details listed on publications), this could initially be assumed to reflect a Western relationship with these animals, rather than that of people living in habitat countries (with the exception of pumas in North America). However, given that only Englishlanguage publications are included here, it is likely that numerous non-english language papers have also been published on similar topics Large felids in modern captive environments In contrast to the conflict between large felids and humans in various felid home countries, people in Western countries appear to display a fascination with these species, evidenced by their active seeking out of interactions with these animals in zoos (e.g. close encounters or behind-the-scenes experiences). As highlighted throughout 74

93 Section 2.2, large felids have long featured in various forms of captivity. More recently, the general public has been given access to these, and other, animals through both zoos and circuses, both within and outside of felid home countries 8. Exploring interactions between large felids and humans in captivity highlights further the fascination people have with these animals and historical accounts of past interactions can contribute to our understanding of how present-day interactions evolved Large felids in circuses Circuses are said to have allowed people to view exotic animals well before zoos were available to the general public (Nyhus et al., 2010). It is reported that they emerged during the 16 th 18 th centuries in the form of travelling exhibitions around Europe, and included various wild animal species (Hosey et al., 2009). Historically, multiple sources have suggested that tigers and lions featured more than the other large felids in circuses; and, to a lesser extent, so did leopards (Guggisberg, 1975; Tait, 2012). Circuses are important in understanding the large felid-human relationship since, although there is much debate and controversy surrounding the ethics of using animals in circuses (Tait, 2012), the inclusion of them in this arena tells us much about their relationships with humans and the ways in which people have trained them. However, scientific data of circus practices appears to be lacking and, where it does exist, appears to highlight the negative impacts of circus life on animal welfare (e.g., see Iossa, Soulsbury, & Harris, 2009; Nevill & Friend, 2006). As such, information regarding the way large felids and humans have interacted in this arena appears limited to what has been written by and about well-known circus 'trainers'. For the purpose of this section, syntheses of this information (in the form of books - e.g., Rothfells, 2002; Tait, 2012) can be used to gain an overview of these interactions but it should be acknowledged that these sources are limited by the authors' interpretations of the information used. While the trustworthiness and rigour of such information is limited given that it has not been peer-reviewed, these sources are useful in providing insight when more scientific information is lacking. 8 There are also large felids housed in zoos and other captive institutions in home countries, but this chapter will focus on Western captive environments, as these are most closely linked to the definition of zoos used within this thesis and the setting of the current research. 75

94 Nyhus et al. (2010) argue that the animal trainers in the circus are some of the most experienced at handling these animals, highlighting the importance of understanding some of their methods used to encourage (or force) the performance of largely unnatural behaviours (i.e., those behaviours that are outside of a species usual behavioural repertoire). Of note, Tait (2012, p. 12) explains how training conditioned movement and developed the capacity of animal performers to withstand the artificial circumstances of being close to another species (genera) in ways that defied their inclinations. Although not an animal behaviour 'expert' (rather, a Professor in the field of theatre and drama), Tait's (2012) interpretation appears credible given what is widely known about the types of behaviours exhibited, and conditions experienced, by circus animals. In drawing on extensive, primary evidence to support her claims, Tait (2012) also describes how circus trainers had their origins in early menageries, where tamers entered small animal enclosures and trained certain behaviours, most recognisably the act of putting one s head inside a lion s open jaw. Multiple images of large felids in circuses show them engaged in these unnatural behaviours, and it is not surprising that there was a growing movement in animal rights campaigns from the 1970s onwards (Tait, 2012). However, the ability for trainers to encourage animals to perform such behaviours is worthy of recognition. Before it was widely accepted that wild animals had the capacity for both individuality and subjectivity, circus animal trainers recognised this, and such trainers were required to closely understand the behaviour and discrete responses of their animals in order to avoid injury (Tait, 2012). Again, this claim put forward by Tait (2012) appears justified in that it could be logically assumed that working with animals such as large felids would require this understanding (given their size, physical features, etc.). Perhaps the most influential person to change the way circus animals were trained was German-born, Carl Hagenbeck (who also influenced modern zoo design; Section 2.4.2). In his book, Beasts and Men, Hagenbeck (1909) describes his experiences of training large felids, among other exotic species (e.g., elephants), in the late 1870s. Hagenbeck (1909) describes his concerns regarding common training methods at the time (i.e., punishment) and his preference for techniques more aligned with what we would now refer to as positive reinforcement training (see Section ). In short, he documents his approach as one involving sympathy for the animal, patience, and 76

95 recognition of the unique characteristics and aptitudes of the individual; as well as one involving reinforcement of desirable behaviours through the use of food rewards, rather than punishment and fear. Hagenbeck (1909) also recognised the importance of trainer characteristics for the success of interactions, such as intelligence and a love of animals, and that a trainer's behaviour could influence that of the animal. He further describes his first attempt with this training method, when he trained four from a collection of 21 lions, in (and commenting that these four had the "necessary talent for the work"; p. 32). Through such methods, Hagenbeck was apparently able to train large felids to perform an array of unnatural acts in his show, such as tigers and lions riding tricycles, tigers rolling on globes, and panthers and leopards on a seesaw to name but a few (Rothfels, 2002). Perhaps his biggest 'feat' was a crowned lion being pulled along on a chariot by tigers and with dogs as footmen (see Figure 2.2). Tait (2012) draws on accounts by other circus trainers 9 to highlight how Hagenbeck s ideas were adapted and expanded by those to follow, such as Frank Bostock, who believed that animals were childlike and should thus be treated like children - not punished, but led by an authoritative, parental figure; Alfred Court, who recognised the Pavlovian conditioning that occurs with trained animals, and their responses to cues (such as the uniform of a trainer) meaning they required less instruction to perform behaviours than was once thought; and Clyde Beatty, who incorporated up to 20 lions and 20 tigers in one act - a staged aggressive confrontation between himself and the animals. While Tait's (2012) interpretation of these sources might not be accurate, what is important here is the notion that there were different methods for training animals in circuses and that different behaviours were encouraged. The purpose of this discussion has not been to advocate for circus practices; however, regardless of one s view of the circus, what we can take away from it is that people have long interacted closely with these animals, and other species, outside of home countries. 9 Unfortunately, efforts were made by the University of South Australia s interlibrary loans departments to access these original sources, however, these were not located in any of the libraries (Australia-wide) which the university has access to; therefore this section of information is limited to Tait's (2012) interpretation of the sources. 77

96 This is useful in illustrating that the modern zoo is not the first place in which large felids and people encountered each other outside of home countries, nor the first place in which these animals were 'trained' (although training may have different meanings and outcomes in these different contexts). IMAGE REMOVED IN ELECTRONIC VERSION OF THESIS DUE TO COPYRIGHT REGULATIONS. PLEASE REFER TO ROTHFELS (2002). Figure 2.2. Image of tigers pulling a lion prince in a cart as part of one of Hagenbeck s circus acts (from Rothfels, 2002, p. 154, Figure 37) The Johns Hopkins University Press Large felids in zoos As highlighted in Sections 2.1 and 2.2, people have reportedly kept wild animals since the first attempt at domestication (i.e., around 10,000 BC; Kisling, 2001) and, although zoos have their origins in menageries and similar institutions dating back at least 4000 years, it was not until the 19 th Century that the modern zoo was said to have been born (Hosey et al., 2009). This era saw a shift in audiences, with the general public having access to these places and exotic animal viewing no longer being reserved privately for the royal and wealthy elite (Hosey et al., 2009; Kisling, 2001). From what has been recorded about early zoos, as long as there have been such collections, large felids have usually featured in them; for example, the Tower Menagerie of London housed lions, leopards and tigers in the early 1700s (Keeling, 2001). 78

97 As zoos evolved from ancient menageries to the more naturalistic places they are today, the modern zoo appears to have undergone many changes and seen shifts in the way people interact with many species of animal, not just the large felids. Whilst much has been written about the history of the modern zoo (for example, see Baratay & Hardouin- Fugier, 2002) few records detail practices used to handle and train large felids in this environment. However, we can derive some information from a wider exploration of the evolution of zoos and the place of large felids in this. In their historical account of zoos, Baratay and Hardouin-Fugier (2002, p. 173) describe how, in the 1800s, large felids were seen as popular because they symbolized wildness and cruelty and encapsulated both the fear of nature and the satisfaction of having overcome it. Further, the public s relationships with zoo animals at this time were based on attraction and repulsion, curiosity and fear (Baratay & Hardouin-Fugier, 2002, p. 181). These ideas should be viewed with some caution however, as although written by academics in the field of history, this work is a translation of the original French version of the text and therefore some meaning may be added that is not true to the original account. This era apparently saw zoos adopting training and taming practices from the circus in order to attract the public (Baratay & Hardouin-Fugier, 2002), with the animals housed in barred and barren cages, indicating the power humans held over them (Davies, 2005). At this time, animals were typically housed alone, with one individual used to represent its whole species (Davies, 2005). By contrast, the 1900s saw barred cages being replaced with more naturalistic exhibits (Kisling, 2001), often containing more than one animal of each species (Davies, 2000). The design of zoos during this time is suggested to have been influenced by Hagenbeck, whose vision of more open-style exhibits began to become a reality (see Hagenbeck, 1909), as well as by Hediger, who encouraged the improvement of animal welfare in the mid-1900s (Baratay & Hardouin-Fugier, 2002). This time saw bars being replaced with glass viewing panels, and barren exhibits replaced with ones that more accurately reflected an animal s natural habitat (Baratay & Hardouin-Fugier, 2002). Zoos also began to be seen as places of education (Baratay & Hardouin-Fugier, 2002). Although it does not appear to be documented, it could be assumed that the lack of information about interactions with large felids at this time 79

98 indicates a shift in which the increase in naturalistic exhibit styles saw declines in direct interactions with these animals. As a pioneer in advances in animal husbandry and management in the mid-1900s, Hediger (1964, 1968) was one of the first to document ideas about the way animals and people see each other in the captive environment (including zoos and circuses). As raised in Section 1.4.5, however, decades later our understanding of how large felids interact with people is limited. These animals remain popular features of zoos worldwide, but we know little about how they interact with people. Today, zoos remain highly popular, with WAZA (2005) suggesting that over 700 million people visit zoos in the WAZA network on an annual basis. As such, there is the potential for a large proportion of the public to currently experience large felids within such captive environments Large felids in Australian zoos today and available interactions Zoos have been operating in Australia since the late-1800s (de Courcy, 2001) and today, as they have historically, house the full range of large felid species (with the exclusion of the jaguar; Hibbard, Ford, & Hoggs, 2012). A recent estimate suggests that between 8-10 million people visit Australian zoos every year (Smith, 2012). Moreover, more people are said to visit zoos in Australia each year than they do any other type of cultural entertainment (with the exception of cinemas; ARAZPA, 2009). Although we may currently have a limited understanding of how large felids and people interact in this environment, we can begin to explore this area by acknowledging currently occurring interactive practices. As in other Western countries, visitors to Australian zoos can have a range of experiences with these animals. In addition to viewing naturalistic exhibits and reading educational signage at exhibits, a search of Australian zoo websites revealed that, as of July 2014, visitors can pay to participate in various behind-the-scenes large felid tours at multiple organisations in Australia (see Table 2.1 for experiences available at organisations which are ZAA members). Most of these tours involve tigers, but also include lions and cheetahs, and one tour involves snow leopards. Tours with the other large felid species (leopards, cougars, and jaguars) do not 80

99 appear to be available. However, some experiences, such as 'Zoo Keeper for a Day' at National Zoo and Aquarium, may include morning tea with cougars. Despite the lack of documentation of interactive practices, investigating the nature of tours available today (at least from what is available on zoo websites) indicates that there is a degree of handling and training occurring between keepers and the large felids that they care for, with an equal mix of hands-on and protected contact experiences offered to zoo visitors (see Table 2.1). Of note, all experiences with cheetah are handson in nature and some tiger experiences are hands-on, whilst all lion experiences take a protected-contact form. Although minimal detail is provided on zoo websites, most tours appear to involve visiting the off-exhibit animal areas, feeding (or watching the animals being fed by keepers), and taking photographs. Some tours also include a professional photographer to capture the experience. These tours are all relatively expensive compared to zoo entry fees, and vary in price and duration between organisations. Perhaps the willingness of visitors to pay a high financial cost to participate in these experiences indicates their desire to interact with the animals, which contrasts notably with some of the relationships between local people and large felids in home countries (Section 2.3.1). Combined, this information indicates that there is a demand for such tours, with both zoos and their visitors promoting them. This represents a notable shift from what is recorded about earlier zoos, where visitors appear to have only been able to view animals in their enclosure, and it raises many questions about the benefits and limitations of such interactions. 81

100 Table 2.1. Summary of key details of behind-the-scenes large felid zoo visitor tours available in all Australian ZAA member organisations (as per individual zoo website details and correct as of July 2014). Organisation/Location Species Interaction level Guest animal feeding Professional photographs specified Cost (AUD) Duration Max group size Minimum age Max Tour frequency Altina Wildlife Park Lion Protected contact Yes No $60/guest 30 min 4 guests 16 yrs 2 x weekly (NSW) 1 Australia Zoo (QLD) 2 Cheetah Hands-on No Yes $150/guest 30 min 4 guests 15 yrs 1 x daily No Yes $60/group 15 min 2 guests 15 yrs 5 x daily Cheetah Hands-on (photo session) Tiger Snow Lion and Hands-on (photo No Yes $300/1-2 guests 15 min 4 guests 15 yrs 1 x daily session) $400/3-4 guests Tiger Hands-on (walk with No Yes $400/1-2 guests 30 min 4 guests 15 yrs 1 x daily animal) $500/3-4 guests Tiger (cubs) Hands-on No Yes $150/guest 30 min 3 guests 10 yrs 1 x daily Billabong Koala and Protected contact Yes No $50/guest Not 4 guests 16 yrs 2 x daily Wildlife Park 3 leopard specified Darling Downs Zoo Protected contact Yes No $80/guest 15 min Not 8 yrs 3 x daily (QLD) 4 tiger specified Dreamworld (QLD) 5 Tiger Protected contact Yes Yes $55/guest 15 min 6 guests 13 yrs 1 x daily Tiger Hands-on (photo session) Tiger Hands-on (walk with animal) No Yes $345/1 guest $395/2 guests $445/3 guests $495/4 guests $545/5 guests $595/6 guests 15 min 6 guests 10 yrs 3 x daily Mon, Wed, Fri, Sat; 4 x daily Tues, Thurs, Sun No Yes $695/group 60 min 4 guests 15 yrs 6 x weekly Mogo Zoo (NSW) 6 Tiger Protected contact Yes No $120/group 20 min 2 guests 16 yrs Not specified White lion Protected contact Yes No $120/group 20 min 2 guests 16 yrs Not specified 82

101 This table displays experiences for individual visitors, although it is recognised that some zoos also offer larger group experiences (e.g., Taronga Zoo currently offers a Big Cat Dinner for Table 2.1. Continued. Organisation/Location Species Interaction level Guest animal feeding Professional photographs specified Cost (AUD) Duration Max group size Minimum age National Zoo and Cheetah Hands-on No No $175/guest 15 min 2 guests 12 yrs 4 x daily Aquarium (ACT) 7 White lion Protected contact Yes No $50/one guest $80/2 guests $120/3-4 guests Tiger Protected contact Yes No $50/one guest $80/2 guests $120/3-4 guests Max Tour frequency 15 min 4 guests 12 yrs 1 x daily 15 min 4 guests 12 yrs 1 x daily Perth Zoo (WA) 8 Lion Protected contact No No $35/guest 20 min 6 guests 14 yrs 2 x weekly Zoos South Australia Protected contact Yes No $145/guest 60 min 6 guests 12 yrs 3 x weekly Tiger and (Adelaide Zoo; SA) 9 lion Zoos South Australia 10 Cheetah Hands-on No No $170/guest 60 min 4 guests 16 yrs 3 x weekly Lion Protected contact Yes No $145/guest 60 min 6 guests 10 yrs 4 x weekly Zoos Victoria (Melbourne Tiger Protected contact No No $73/adult Zoo; VIC) 11 $60/child Zoos Victoria (Werribee Lion Protected contact No No $80/adult Open Range Zoo; VIC) 12 $70/concession $65/child Taronga Western Plains Tiger Protected contact Yes No $59/guest 20 min Not Zoo (NSW) 13 specified 25 min 8 guests 15 yrs 1 x daily 50 min 8 guests 10 yrs 1 x weekly w.altinawildlife.com/portals/0/brochures/behind%20the%20scenes%20-%20wild%20encounters%20brochure%20edited.pdf; yrs 1 x daily 20 people). It also contains those tours which are specific to one or more large felid species, but it is recognised that many zoos also conduct tours of multiple animals (e.g., Keeper for a day tour at Zoos SA). 83

102 Potential benefits of interacting with large felids in zoos As discussed in Section 1.4.5, interacting with people through appropriate training and handling practices has the potential to improve the existence of zoo-housed animals. Although empirical evidence is still needed to understand how large felids respond to human interaction in the zoo environment, it is also important to acknowledge how such interactions might be of benefit to the zoo keepers who work with these animals and to the public who visit them Benefits to zoo keepers While there is limited literature suggesting any potential benefits to zoo keepers of interacting with and/or training large felids specifically, some can be derived from research with other species. Possibly the most widely discussed benefit for any person working with animals in an increased contact and/or training capacity (in both zoos and also laboratory settings) is the potential for a strong bond and/or relationship to be formed between animal and human (Bayne, 2002; Carlstead, 2009; Hosey & Melfi, 2012; Laule et al., 2003; McKinley et al., 2003; Prescott & Buchanan-Smith, 2003; Savastano et al., 2003). Suggested reasons for this improved relationship include: the increased frequency and contact spent with the animal, and the opportunity to get to know the animal as an individual; the large amount of time that individual animals spend living in particular facilities; the dependency of the animals on staff for their needs; and the ability to witness and appreciate the benefits that a training program has on the well-being of the animals (Bayne, 2002; Savastano et al., 2003). In addition to the bonds formed, increased handling and training also have the potential to improve ease of performance of duties by zoo keepers (Hosey & Melfi, 2012). Not ignoring the investment of effort and time involved in establishing and maintaining a contact and/or training program (Prescott & Buchanan-Smith, 2003), a successful program may improve the duties of keepers in many ways, particularly in relation to easier movement of the animals (since trained animals, or animals familiar with their keepers, are more likely to shift more easily between enclosures, whether on- or offexhibit; Colahan & Breder, 2003). Consequently, it could be suggested that more efficient moving/shifting can contribute to better enclosure maintenance, since keepers 84

103 can access enclosures more frequently and readily for purposes such as cleaning and the addition of enrichment items. Moreover, handling and training may improve the care of the animals, since animals that are trained and have a closer bond with their keepers may be more likely to engage in routine veterinary checks more readily (resulting not only in less stress for the animals, but also less stress for their keepers; McKinley et al., 2003). Combined, these factors could lead to increased job satisfaction. In particular, in a paper describing the development of an operant condition training program for various New World primates, Savastano et al. (2003) discuss how the program not only allowed keepers at the Bronx Zoo to build rapport with the animals, but that it was more enriching and rewarding for them, since they got to know the animals on a more individual basis, interact positively with them, and were able to appreciate the benefits that training has for the animals' well-being. Although this represents the authors' opinions only and was not systematically measured, it does provide some insight into the possible benefits of such training practices. Whether or not these same benefits apply to interactions with large felids is yet to be explored Benefits to zoo visitors (and hence zoo conservation initiatives) There are various ways in which zoo visitors might benefit from increased interactive practices with large felids. Indirectly, it could be argued that if large felids benefit from interactions with keepers in the ways described in Section (e.g., increased psychological well-being, etc.), then zoo visitors may be more likely to see large felids that are engaged in species-specific behaviours and activity budgets that do not indicate boredom or stress (e.g., stereotypy). This might then enhance the visitor experience and improve visitor understandings of the natural behaviour of these animals. More directly, handling and training allow a number of interactive experiences for the visiting public 10 (see Table 2.1). According to Nyhus et al. (2010), people have a desire to be close to, and even touch, animals. In this way, zoos could act as a substitute for wild animal encounters (Mason, 2000), providing visitors with a place to enjoy interactions in a 10 It is acknowledged that handling and training of large felids are not pre-requisites for all interactive experiences to occur, but it is suggested that these practices can allow zoos to offer more interactive experiences. For example, if a keeper trains the tigers under their care then there is an opportunity for the zoo to offer public training sessions. Similarly, if cheetahs have been hand-reared then there is an opportunity for zoos to offer 'hands-on' behind-the-scenes tours. 85

104 managed and (hopefully) safe way that does not contribute to risks for wild populations (e.g., disease transmission and loss of fear of humans). Providing visitors with a more enjoyable experience can also assist zoo conservation initiatives. Falk et al. (2007) argue that by offering interactive, interpretive and educational experiences whereby visitors can encounter live animals, zoos have a noticeable impact on the visiting public. This impact will be explored in more depth in Chapter Six, where it is argued that there is a lack of information about how visitors respond to experiences with large felids, either during interactive encounters or when viewing them more passively in their regular exhibits. Research with other captive species, in both wildlife encounters and zoo settings, suggests that visitors who engage in more interactive encounters report enhanced satisfaction with their experiences, and display increased knowledge, pro-environmental attitudes and long-term behaviour change intentions (Zeppel & Moulin, 2008; Visscher, Snider & Van Stoep, 2008). These positive changes may also apply to experiences with large felids in the zoo environment. Certainly, the potential for such changes has likely facilitated many zoos to offer interactive experiences. The Australian Animal Welfare Standards and Guidelines: Exhibited Animals, which are followed by ZAA organisations, strive for interactive programs in which visitors have a positive experience and have an enhanced appreciation of animals (Department of Agriculture, Forestry & Fisheries, 2009, p. 36). It therefore appears timely to explore this, particularly considering the existence of negative perceptions of large felids in their natural habitats, in stark contrast to captivity where these animals possess broad public appeal (Ray, 2005). Allowing the public to visit and interact with large felids in zoos also has the potential to raise funds for the conservation of these animals. For example, Christie (2007) highlights that between 1998 and 2005, zoos contributed to 61.2% of the total funds spent on Sumatran tiger conservation. Thus, although it is difficult to measure the conservation value of wildlife experiences, largely because we do not know what visitors do once they leave a zoo (discussed further in Chapter Six), the financial contribution people make by simply attending an interactive experience may be substantial. Such contributions may be put toward in-situ conservation programs, such 86

105 as those targeted to reduce human-big cat conflict as a result of depredation (e.g., see Australia Zoo, 2013). Measuring this financial impact would also be important to understanding the success of large-felid human interactions in zoos Potential dangers of interacting with large felids in zoos Injuries/fatalities Just as in their habitat countries, large felids can have the potential to be dangerous in zoos, particularly in situations where hands-on contact is permitted. This is not to say that protected contact is not dangerous, but that where there is no barrier between large felid and human there is clearly greater potential for a dangerous interaction to occur. It does not appear possible to obtain credible information regarding the risk of 'attacks' on people by these animals in zoos (and other captive settings), largely due to there not appearing to exist a place where this information is regularly collated. However, some insight can be gleaned from the human medical and surgical literature, which has documented a number of such cases. A basic Google Scholar search (using search phrases such as 'big cat attacks in captivity') for peer-reviewed, English-language literature (along with subsequent searching of reference lists of relevant articles) revealed 17 papers, published between 1989 and 2013, describing instances were humans have been attacked by large felids in captive settings. Key details of this literature are displayed in Table 2.2. This literature base is by no means exhaustive, nor would it be reflective of all attacks that have occurred during the time period of the studies, as many occurrences would likely be unreported. Some reasons for lack of reporting might be attributed to the nature of medical and surgical work possibly leaving limited time for scientific publication, or the lack of perceived benefits to the scientific publishing of this information. Additionally, not all such attacks would result in medical attention (for numerous reasons) and therefore would not have the opportunity to be documented. Nonetheless, this literature is included here to provide some understanding about these occurrences. Primarily, these publications have taken a case-study format, whereby the authors (usually medical doctors, according to the author details provided) present an overview of the injuries sustained by a person, and the medical/surgical response to treat them 87

106 (where possible/necessary). To summarise this literature, Table 2.2 indicates that a large proportion of this body of work focuses on attacks by tigers (accounting for 11 of the 18 cases reported). Only three cases are reported involving lions and leopards, one by a jaguar, and none by the remaining large felid species (cheetahs, pumas and snow leopards). While this is reflective of what is known about large felid attacks in their home countries (see Section ), it is unclear from this literature alone why more of these cases involve tigers and it certainly can not be said that tigers attack people in captivity more than other large felids - among multiple other reasons, it may be that these species are kept in captivity at a greater proportion than the others, or it may be that when they have attacked they have inflicted more serious damage requiring medical attention. Regarding the setting, the majority (n = 11) of these instances have occurred in unregulated forms of captivity (circuses, private ownership, etc.), compared to zoos (however it is unknown whether these zoos are accredited or not). This may imply that there is less risk of attacks in zoos, but this cannot be substantiated. More of these cases have been authored in the USA (n = 8) compared to any other region, primarily in unregulated settings, and this may tell us something about the need for legislation (or reinforcement of legislation) to reduce the keeping of these animals there outside of regulated zoos. Injuries outweigh fatalities in these case studies (12 cases compared to 6, respectively), perhaps because, although some injuries would eventuate to fatalities, injuries would require medical attention whereas fatalities might occur on-site and therefore hospitalisation for treatment would not be required. If the purpose of this literature is to provide other medical professionals with evidence to support treatment after such attacks, then it is not surprising that injuries are more commonly reported. Regardless, this body of literature indicates the serious consequences involved. Examples of injuries sustained include cranial fractures (e.g., see Dabdoub et al., 2013; Emami et al., 2012) and lacerations to various parts of the body, including neck (e.g., see Anderson et al. 2008) and back (e.g., see Capitini et al., 2002). Wound infection has also been reported (e.g., see Isotalo et al., 2000). This is important for indicating the damage that can be inflicted by these animals, not only as a result of their natural hunting instincts, but also as a result of their sheer size (in comparison to a human) and other physical features (such as teeth, claws, etc.). These features may make these animals better equipped to 88

107 cause serious harm compared to some other species and highlights why interacting with them can pose risks. There is no clear bias towards basic demographics of the human (i.e., gender, age) involved. However, within each facility type there are some patterns. In particular, none of the zoo cases involve children, yet more frequently involve zoo keepers. This might imply that zoo keepers are at greater risk than zoo visitors, which would be logical considering the closer contact they have with animals (not necessarily for the purpose of interacting, but by the nature of the work they do - cleaning, feeding, etc.). Perhaps what is most important from this literature is an understanding of the circumstances surrounding these attacks. However, given the aforementioned focus of these studies (to document injuries and treatment), this body of literature only provides brief information about the preceding events (and often fails to indicate where this information has come from; it could be assumed that it is derived from patient/others' accounts upon hospital presentation). Despite this limitation, Table 2.2 highlights that human error or choice has resulted in these incidences (e.g., unlocked animal cages, entering an animal's space, attempting to touch an animal). In this way, recommendations cannot be drawn regarding how to 'safely' interact with these animals, other than that these situations may have been prevented by modification to human behaviour (e.g., ensuring cages are secure, not entering an animal's space, not attempting to touch an animal). Media reports also provide some insight into the occurrences of these dangers and many video snippets are freely available on the internet (for example, YouTube searches of big cat attacks will show a number of incidences and the behaviours used by these animals during such encounters). It does not appear that any recent sources have attempted to synthesise this information. However, some insight can be gained from a review, conducted by Nyhus et al. (2003), of media reported tiger attacks from 1998 to Although somewhat dated now, the search revealed articles from 30 international media sources, reporting on 59 incidences of injury/fatalities inflicted by tigers. The majority of these events (n = 27) occurred in the US. Similar to the aforementioned medical and surgical literature, most (75%) of these occurred in cases where the animals were privately owned, particularly 89

108 Table 2.2. Summary of key details of large felid attacks in captivity (including zoos, circuses, exotic animal farms, and private ownership) derived from a selection of case reports in the peer-reviewed, English-language medical literature (published between ). Species Captive facility (Region*) Country Human Circumstance Outcome Country/Authors Fatality Kohout et al. (1989) Tiger Zoo Australia 33yr old female zoo keeper Person entered animal cage (authors unsure as to whether the woman knew the animal was not secured or if she entered deliberately) Zoo Ireland 19yr old female Intoxicated person jumped zoo fence to feed Injury Murphy et al. (2007) animal by putting hand through enclosure fencing Cleaning animal's cage Injury Schiller et al. (2007) Zoo USA 37yr old female zoo worker Circus USA 10yr old male Person reached into tiger s cage in attempt to pat the animal Injury Clark et al. (1991) Exotic animal farm USA 36yr old female Cleaning tiger cage (two tigers involved) Injury Anderson et al. (2008) Exotic animal farm USA 28yr old male Cleaning animal's cage Injury Wiens & Harrison employee (1996) Fatality Chapenoire et al. (2001) Private ownership Canada 11yr old boy Trainer was moving the animal (via a leash) to a Injury Chum et al. (2011) viewing area for photography Private animal reserve USA 7yr old female Animal broke through chain-link fence Injury Capitini et al. (2002) Injury Emami et al. (2012) Private ownership France 49yr old male Person entered animal space, apparently to calm the animal (whom he had hand-reared) NS (possibly circus) Germany 28yr old 'tamer' (gender = NS) Person slipped and fell during an evening show (three tigers involved) NS Canada 35yr old trainer Occurred during a training session Injury Isotalo et al. (2000) Fatality Bock et al. (2000) Lion Zoo Germany Male (age unspecified) Person believed to have entered lion enclosure overnight (considered suicide) and found deceased by a zoo keeper Private ownership USA 49yr old male Intoxicated person put hand into lion s cage to pat animal (lion privately owned by man s neighbour) Circus Bolivia 8yr old boy Animal escaped from its unlocked cage during a show Injury Lazarus et al. (2001) Injury Dabdoub et al. (2013) Leopard Zoo Czech Republic 26yr old male Animal escaped from unlocked cage Fatality Hejna (2010) Zoo USA 55yr old zoo volunteer Person entered the animal s cage to feed it Injury Bahram et al. (2004) Private ownership USA 2yr old female Child's father s pet leopard escaped from its Fatality Cohle et al. (1990)^ cage Animal escaped its cage Fatality Cohle et al. (1990)^ Jaguar Zoo South America 30yr old zoo keeper Note: * = location of primary author, rather than region where instance occurred, as most papers did not specify the latter; NS = not stated ; ^ = study appears twice as included cases of multiple large felid attacks 90

109 when humans were too close to tigers (e.g., when viewing them, handling/moving them, being photographed with them, or feeding them). Fewer incidents (n = 4) occurred as a result of a tiger escaping from its cage. Outside of the USA, these attacks were more frequently experienced by visitors at zoo/animal parks, again primarily when humans were in very close proximity to the animals. These differences in the location of incidences (private ownership versus zoos) between the USA and those in other countries may be attributed to pet ownership of large felids being legal some states of the former. While Nyhus et al. s (2003) review provides some information about the circumstances under which such incidences have occurred, it is likely that many more go unreported in the media and so the results cannot be considered conclusive. Clearly, however, there is the potential for negative interaction between large felids and people in the zoo environment. Beyond the obvious impact this can have on humans, these incidents may result in euthanasia of the animals (Chapenoire, Camiade, & Legros, 2001) Perceptual dangers Apart from the possibility of physical risks, interacting with large felids may send negative messages to zoo visitors. Clearly, these are dangerous animals that typically do not interact positively with people in their natural habitats. However, in seeing keepers interact with them, or having an opportunity to interact with a large felid themselves, visitors may leave the zoo with misperceptions about these animals. This could have negative impacts on willingness to assist in conservation and, worse, it could result in visitors wanting to purchase a large felid as a pet. In Australia, it is not legal to keep exotic animals as domestic pets, however, it is widely recognised that it is legal to do so in some parts of the USA (Born Free USA, 2013). It has been estimated that approximately one fifth of visitors (over three million) to Australian zoos are international tourists (ARAZPA, 2009); while it is unknown exactly where these visitors come from, this figure indicates that Australian zoos are spreading their messages to a large number of international tourists each year. Therefore, these zoos need to be aware that their interactive programs have implications, and research is needed to determine what messages visitors are leaving the zoo with. 91

110 The responsibility for Western zoos to set a good example While some (e.g. anti-zoo lobbyists) may argue that zoos do not contribute to the conservation of wild species, this thesis takes the view that accredited zoos can make a positive contribution to conservation, both directly (e.g., through captive breeding programs and the provision of funds to support wildlife preservation in the field) and indirectly. In the latter sense, accredited zoos can be seen to have a responsibility to set a 'good' example of animal keeping in order to encourage better practices elsewhere (and to educate the public about what 'good' practices are in order to encourage them not to support practices that fall short of this). This has particular importance for unofficial forms of captivity, as well as the portrayal of large felids in the media Large felids in unofficial forms of captivity Western zoos have a responsibility to set a good example for large felid-human interactions elsewhere, since not all captive institutions make significant contributions to conservation. It has been argued that in many less developed countries wildlife is still considered more valuable as a commodity in the marketplace than as a conservation resource in the jungle (Walker, 2001a, p. 215). Tigers are a prime example of this form of tourism, with thousands kept captive in Asia (Nyhus et al., 2010). Perhaps the most renowned of these interactive organisations is Tiger Temple, in Thailand. Originally established as a rescue centre for poached tigers, this facility sees hundreds of visitors per day interacting hands-on with tigers (Tiger Temple, 2008). Photographs taken at the facility, freely available through Google searches, show the animals chained, exposed to heat and in close proximity to each other, with visitors of all ages (infants through to adults) interacting with them in a hands-on way and engaging in various interactive behaviours (such as sitting on the animals). Images also show the use of force (and punishment) to condition and control the animals behaviour. A non-scientific report of an investigation of the organisation (Care for the Wild International, 2008, p. 3) concluded that, despite the initial good intentions of the facility, the animals are poorly housed and badly maltreated to make them compliant and perform for visitors, that the facility has become a breeding centre to produce and keep tigers solely for the tourists and therefore the Temple s benefit, and that it participates in illegal international trafficking. Such conclusions suggest that the breeding program at the centre does not contribute to species conservation in the wild. Although this is just one view of the 92

111 facility, this example is a reminder that visitors can interact with large felids in various forms of captivity outside of Western countries. Although appearing to have been studied scientifically, the dominance of Western tourists in photographs promoting these organisations (as per Google searches), suggests that people in developed countries could be driving the demand for such interactions. Australian and other Western zoos have a responsibility to set a good example in this regard, so it is even more pressing that we begin to understand how such interactions can be conducted in a way that is not damaging to either the large felids or humans involved Large felids in the media In addition to the direct interactions that people can have with large felids, these animals also feature quite frequently in the Western media, presenting another way in which people encounter them. Examples have been both encouraging and discouraging of interactions. A prime example of the former is when, in 2008, video footage of a reunion between two young men and their once pet lion, Christian, became a YouTube sensation (available at Christian was reportedly purchased from Harrods department store in London in 1969 by two Australians, who then 'released' him in Kenya to avoid him being kept in a zoo (Bourke & Rendall, 2009). The footage of the reunion shows an encounter one year after Christian s release, in which he recognises his former owners and engages in affiliative behaviour with them (e.g., head-rubbing, patting, etc.). Although this is quite a unique example of large felid pet ownership, the positively-appearing reunion could contribute to the general public believing that these animals are tame and can be easily handled. Other media representations can serve to substantiate these views. For example, in early 2013 a commercial was aired on Australian television promoting cosmetic company, Maybelline s, Cat Eyes mascara. In addition to images of various large felids throughout the advertisement, it ended with an actress cuddling a tiger and lion cub. Such portrayals may skew the public's view about the dangerousness of these animals, the unnatural nature of interacting with them in this way, and the implications this could have for animal welfare. 93

112 Other portrayals in the media show the dangers of exotic pet ownership. A leading example is an incident that has been dubbed, the Ohio Massacre. In 2011, Zanesville resident, Terry Thompson, was alleged to have released his collection of numerous exotic animals (including not only large felids, but also bears, wolves, and primates) from their cages on his property before taking his own life (Australian Broadcasting Commission [ABC], 2013). What resulted was the shooting of the majority of these animals by local law enforcement personnel, reportedly as a means to reduce the danger posed to local people and the apparent lack of facilities in which to house these animals even if they could be caught (ABC, 2013). This event sparked public outcry around the world and, regardless of one s view of the way this incident was handled, this is a devastating reminder of what can go wrong when animals, such as large felids, are maintained as pets. As mentioned in Section 2.3.2, zoos might be responsible for encouraging such pet ownership, but are also in a position to educate the public otherwise. With potentially negative implications for animals and people alike, these exemplars of large felid-human interactions in the mainstream media accentuate the need for Western zoos to acknowledge responsibility for setting a good example with regard to the handling of these animals Conclusion In the history of large felid-human interactions, which have occurred for as long as the two have co-existed, the practice of direct interactions in zoological organisations, in their current form, appears relatively new. Indeed, according to available documentations of the history of zoos, only in the last century have the general public outside of large felid home countries been able to see, let alone interact with, these animals. The way large felids and humans interact in home countries differs in many ways from the way the two experience each other in various forms of captivity. This chapter has highlighted that visitors to Australian zoos can have a range of interactive experiences with large felids and that these interactions could be considered unnatural compared to those taking place in large felid home countries. Thus, with an increase in interactions in the zoo setting, the focus of this thesis is to explore these interactions. Primarily, the impetus for this is out of concern for the welfare of the animals. Additionally, there is a need to explore how the humans involved are influenced; as 94

113 such this thesis equally seeks to understand how familiar (zoo keepers) and unfamiliar (zoo visitors) humans respond to such interactions. As a research area in its infancy, the collection of systematic data regarding interaction between these three groups will help to shed light on this issue, bringing some understanding of the welfare and conservation value of large felid-human interactions in zoos. This is particularly important given that these animals are flagship species and have the potential to encourage and support the conservation of other species and ecosystems (Loveridge et al., 2010). The following chapter provides a summary of methodologies used in past research to explore humananimal interactions in zoos and a rationale for the methodologies used in the present thesis. 95

114 3 General methods and research site overview It is up to us humans not to abuse and destroy this mutual trust between man and the big cats which has emerged after millennia of confrontation and exploitation. C.A.W. Guggisberg (1975, p. 308) The first two chapters of this thesis provided reviews of both the enrichment literature for large felids, and an overview of the relationships they have shared with humans in various environments. It was highlighted that very little is understood about current interactive practices with these animals in Western zoological organisations. This chapter summarises the methodologies employed in peer-reviewed investigations of human-animal interactive practices in zoos to date 11. An overview of the specific methodologies employed to meet the thesis aims follows. A brief description of the Royal Zoological Society of South Australia (Zoos SA) the organisation at which the majority of the research was conducted and its large felid collections, is also provided. 11 The literature included in this chapter has been accumulated over the course of this research degree ( ) as well as through searching the reference lists of key review articles on this topic (e.g., Claxton, 2011; Fernandez et al., 2009; Hosey, 2008). This is not intended to be an all-inclusive, systematic review of human-animal interaction research methodology, but rather an overview of the majority of studies on this topic. Moreover, it is recognised that human-animal interactions have been explored in many contexts (e.g., farms and laboratories); however, aligning with the focus on interactions in zoos in this thesis, only zoo-based studies are included here. 96

115 3.1. Introduction As Guggisberg (1975) urges, we as humans have a responsibility to ensure a positive relationship with large felids. This can extend beyond interactions in home countries to zoological organisations, as is the focus of this research. Doing this successfully will require research which considers the perspectives of all parties involved in the relationship/interaction. Therefore, the central perspective underpinning this thesis is the acknowledgement that zoos serve multiple user groups (typically, but not exclusively, the animals, their keepers and zoo visitors) and that managing their conflicting needs is a complex task, since any changes in animal husbandry or management will undoubtedly influence not only the animals, but also the zoo keepers and visiting public involved (Mason, 2000; Morgan & Hodgkison, 1999; Shettel-Neuber, 1988; Wilson et al., 2003). In this way, if zoo design and management are to be optimised, we must consider these three perspectives. To address the thesis aims (see Section 1.6), the research is comprised of three studies. In embarking on this thesis, it was recognised that there are multiple ways in which large felids and humans interact in the zoo environment, including those that involve zoo keepers (e.g., feeding, training, health checking, enclosure maintenance), those that involve visitors (e.g., watching, feeding, photographing), and those that involve both (e.g., information talks and interactive experiences). Combined, these various interactions pose multiple questions which appear to have remained unanswered in the literature to date, and only a few of which could be addressed within the timeframe and resources of a PhD program. Given the limited research in this field, it was deemed most appropriate to contribute to this area by exploring pre-existing practices, rather that setting out to experimentally test new practices (such as, determining how long it takes to train a lion to present body parts). Consequently, a logical starting point was to ask zoo keepers how they interact with their large felids and to seek their opinions of the benefits and limitations of different interactive practices (Study One). Following this, one form of interaction behind-the-scenes zoo visitor tours - was investigated from animal (Study Two) and visitor (Study Three) perspectives. This form of interaction was selected for exploration as it is a prime example of a practice made possible by large felid-keeper contact, and has implications for animals and visitors 97

116 alike. Since each phase of the research involved different procedures and techniques, it is important to understand from the outset why each approach is appropriate and how the different approaches will complement each other towards an overall conclusion at the completion of this thesis Exploring zoo keeper experiences with large felid interactions (Study One) The first phase of this research sought to explore the nature of interactions between large felids and keepers, as well as to uncover keeper opinions of different interactive practices. This was based on the rationale that, being an area of research in its infancy, zoo keepers would be able to not only provide valuable perspectives that could be used to understand some basic information about interactions (such as which levels of interaction are most common with each species and how much time is spent interacting) but also to provide subjective opinions which could spark future, scientific investigations. Guidance for the design of this study was sought from other investigations of zoo keeper interactions with animals Methodologies employed in human-animal interactive research: zoo keepers Comparative to zoo animal and zoo visitor research (see Sections 3.3. and 3.4 respectively), very little zoo-based research involves zoo keepers. While it is recognised that researchers have sought zoo keeper involvement in some studies, such as personality assessment of animals (e.g., Gartner & Powell, 2011), behavioural assessment of animals (e.g., Carlstead & Brown, 2005; Wielebnowski, 1999), enrichment evaluation (e.g., Hoy et al., 2009), and exhibit evaluation (e.g., Shettel- Neuber, 1988; Wilson et al., 2003), there is a dearth of literature exploring humananimal interactions from the perspective of zoo keepers. Additionally, to the best of this researcher's knowledge, no peer-reviewed literature has specifically explored zoo keepers' opinions of handling and training practices for any species, let alone large felids. 98

117 Only three peer-reviewed studies, published between 2007 and 2012, appear to exist which explore human-animal interactions by involving zoo keepers as participants. The key methodological details of these studies are summarised in Table 3.1. Given the differing aims of each study, it does not appear appropriate to collate them to draw a conclusion about how such research should be conducted. Instead, the studies can be described individually to highlight the relevance of their methodology to our understanding of how human-animal interactions could be explored from zoo keeper perspectives. Of the three studies, those conducted by Carlstead (2009) and Hosey and Melfi (2012) are most similar to the present topic, with their aims of exploring keeperanimal relationships in zoos and human-animal bonds with zoo animals respectively. Both studies used self-report survey methodologies (with the addition of behaviour observations in Carlstead's (2009) study), with purpose-developed questionnaires. Carlstead's (2009) questionnaire used rating scale items of animal behaviour and a combination of closed and open-ended questions to assess keeper demographics and animal husbandry, with reference to four species. These measures were combined with behaviour observations of interactions, and used to test specific hypotheses to further understand factors influencing the development of relationships between keepers and animals. Hosey and Melfi (2012) used a combination of closed- and open-ended questions to understand whether zoo professionals form bonds with any animals, whether these bonds can be predicted by participant demographic characteristics, and what participants' views are regarding these bonds and their potential benefits. The use of a combination of closed and open-ended questions appears appropriate, given the exploratory nature of each of these studies (i.e., the closed-ended questions allowed specific information to be obtained, and the open-ended questions allowed for elaboration). While using validated measures of the constructs of interest may have strengthened the quality of information obtained from these studies, the ability to do so is not possible when these do not exist. Both studies enhanced their generalisability to zoos internationally through recruiting participants from multiple countries. While the approach for doing so was not described in Carlstead's (2009) study, keepers (n = 48) were recruited from 46 zoos (n = 82). Hosey and Melfi (2012) describe the use of a convenience sampling method to recruit participants (n = 130) at various zoo-related 99

118 Table 3.1. Summary of key methodological details of three peer-reviewed studies, published between , of keeper-animal interactions in zoos involving zoo keepers as participants. Study focus Human sample/ region Keeper-animal Zoo keepers (n = relationships in zoos 82)/46 international zoos Human-animal bonds formed by zoo professionals Effects of personality of keepers and tigers on their interactive behaviour Zoo professionals (n = 130, including n = 53 zoo keepers)/various regions Male zoo keepers (n = 7)/ Dreamworld (AUS) Species of animal Design Method Instrument/Outcome measures* Study details Carlstead (2009) Black rhinoceros, cheetah, maned wolf, great hornbill Various species Crosssectional Correlational Medium: Self-report questionnaire and behaviour observations Participation time: NS Sampling method: NS Medium: Self-report questionnaire Participation time: NS Sampling method: Convenience sampling at three international events Bengal tigers Correlational Medium: Self-report questionnaire and behaviour observations Participation time: NS Sampling method: Convenience sampling of keepers working at the zoo Questionnaire: - Animal behaviour (8 items - rating scale) - Keeper background and caretaking behaviour (10 items - closed and open-ended) - Based on past research: No Behaviour observations: interactions between keepers and animals Questionnaire: - Demographics - Participant views of relationships (5 rating scale statements) - Whether a bond had been developed/with which species, what sort of contact was had with the animal and how frequent the contact was for different interaction categories (e.g. feeding and cleaning) - How the participant and the animals benefit from the bond (open-ended) -Based on past research: No Questionnaire: - Animal personality (Adjectives adapted from previous animal personality research - rating scale) - Keeper personality (Revised NEO Personality Inventory - rating scale) -Based on past research: Yes Behaviour observations: interactions between keepers and animals Hosey & Melfi (2012) Phillips & Peck (2007) * = materials/measures details primarily relate only to the survey components of the relevant studies (see studies for more detail regarding other methods; e.g., behaviour observations); NS = not stated 100

119 events. Since these events varied in topic nature, this appears an appropriate method for recruiting participants with diverse animal bond backgrounds. In an investigation of a different nature, Phillips and Peck's (2007) study explored the effects of keeper and tiger personality on their interactive behaviour. This lent to the dual methodology of participants (n = 7) completing a self-report personality questionnaire (using the previously validated, Revised NEO Personality inventory to assess keeper personality), as well as a purpose-developed animal personality measure (developed from past animal personality research); and behaviour observations of interactions between the keepers and animals (n = 7). This study is limited in generalisabity due to its small sample size and being conducted at only one organisation; however, as shown in Table 2.1, Dreamworld is one of only two organisations in Australia where hands-on contact is practiced with tigers and therefore the findings might be quite representative of zoos in this region. While a useful and innovative approach to exploring human-animal interactions, this method does not align with the aims of the current study (i.e., to investigate keepers' opinions of interactive practices) and is of limited use in this context Rationale for Study One methodology: Zoo keeper survey With so few studies exploring human-animal interactions from keeper perspectives, guidance from past research regarding how to conduct this aspect of the current research was limited. However, given the exploratory nature of the study (Section 3.2), some aspects of this research, along with more general research design principles were employed when designing Study One (Chapter Four). This study aimed to explore current (at the time of study) use of interactive practices with large felids in zoos, and, more importantly, keeper opinions of these practices. As such, a cross-sectional design was decided upon. As used by Carlstead (2009) and Hosey and Melfi (2012), as well as in keeper evaluation studies (Hoy et al., 2009; Shettel-Neuber, 1988; Wilson et al., 2003), a survey method using a self-report questionnaire was considered the most logical approach, given that this study sought to explore keeper opinions. Although limited by the potential for misinterpretation (of 101

120 questions by participants, and of answers by researchers) and biased responses, the selfreport survey method is considered to have many benefits, such as efficiency (allowing for data to be collected from many participants at one point in time, in their own time) and anonymity (which may encourage more honest responses) (Portney & Watkins, 2009). Alternative methodologies - such as focus groups or in-depth interviews - were also considered; however, given the relatively niche participant group (i.e., large felid keepers), such a method was deemed to be not feasible for multiple reasons. For example, focus groups could have been conducted locally with Zoos SA's large felid keepers, however it is likely that keepers working at the same organisation and implementing the same practices could have similar perspectives and that this would likely limit the generalisability of findings to other zoos. To counter this, focus groups could have been conducted at multiple zoos in Australia, but funding for this (i.e., travel expenses) was not available. Telephone interviews could have been an alternative, however this would have required a considerable time investment from zoo keepers during working hours. Other studies involving interviews of zoo keepers have been limited to smaller sample sizes (comparative to Carlstead (2009) and Hosey & Melfi's (2012) self-report survey studies) at single sites (e.g, n = 33 zoo staff in Shettel- Neuber's (1988) investigation of opinions toward different exhibit types; and n = 23 staff in Wilson et al.'s (2003) evaluation of a panda conservation centre). For these reasons, a self-report questionnaire was deemed most appropriate. Given the aim of surveying an international sample of large felid keepers, an online survey format chosen. Online surveys are considered advantageous due to their ease of completion (Moss, Plowman, Hosey, Stevenson, & Whitehouse-Tedd, 2013). To maximise participation, it was decided that a dual approach would be taken, whereby all major, international online zoo keeper forums (determined through prior knowledge and internet searches), and all Australian and New Zealand zoos housing large felids at the time of study (determined based on those listed on the Zoo and Aquarium Association website and checked via individual zoo website searches), would be contacted and asked to provide the web-link to their members and staff (respectively). Ten zoo keeper forums (Animal Keepers Association of Africa, Australasian Society of Zoo Keeping, American Association of Zoo Keepers, Association Francophone de Soigneurs 102

121 Animaliers, Association of British and Irish Wild Animal Keepers, Zoo Biology, Australasian Zoo Keeping, Asociacion Iberica de Cuidadores de Animales Salvajes, and Berusfverband der Zootierpfleger, Stichting De Harpij), nine Australian zoos (Australia Zoo, Cairns Wildlife Safari Reserve, Dreamworld, Mogo Zoo, National Zoo, Perth Zoo, Taronga Zoo (2 properties), Zoos South Australia (2 properties), and Zoos Victoria (2 properties with large felids)) and five New Zealand zoos (Wellington Zoo, Auckland Zoo, Hamilton Zoo, Orana Wildlife Park, and Zion Wildlife Gardens) were selected for approaching. Alternatives used in other research, such as Hoy et al. (2009) and Hosey and Melfi's (2012) approach of distributing surveys at conferences/training sessions, were considered. However, this would have required considerable costs (i.e. travel expenses and conference registration) and would have limited the sample to those keepers who were also able to attend such events. Given the particular focus on large felid keepers, not zoo keepers in general, the potential sample at such events may have been small (or smaller than the possible population accessed via online methods). For example, while Hosey and Melfi's (2012) study explored human-animal bonds with any species (i.e. not limited to felids or to zoo keepers) and had a sample of 130 respondents, only 53 of these were zoo keepers. Although this is only one guide, this might offer an indication of how many keepers attend conferences and training sessions in comparison to other zoo professionals (e.g., researchers, managers, etc.). In the absence of past literature exploring keeper opinions of interactive practices, a self-report questionnaire was purpose designed for this study. Discussed further in Section 4.2.2, a 133-item questionnaire was generated, comprising of: 1. Demographic and work-experience questions (adapted based on the questions asked by Carlstead (2009)); 2. Closed and open-ended questions regarding current use of handling and training practices with each large felid species; 3. Closed and open-ended questions regarding opinions of different handling and training practices for animals, keepers, and visitors, developed based on converting current research findings regarding the benefits of different 103

122 interactive practices into statements, to be responded to on a rating scale. To further illuminate benefits and limitations of interactive practices, numerous open-ended responses were also included. Closed and open-ended questions each have their benefits and limitations, a combination of both was seen as the most useful way to understand rates and opinions of current practice, as well as a more in-depth understanding of the benefits and limitations of each. This is similar to the approach taken in past zoo keeper research (e.g., Carlstead, 2009; Hosey & Melfi, 2012; Wilson et al., 2003) Exploring an interactive experience from animal perspectives (Study Two) The second phase of this research explored behind-the-scenes tours from animal perspectives (Chapter Five). Specifically, it sought to understand how large felids might be influenced, if at all, by their involvement in such tours. This was based on the rationale that although various zoos in the Australian region conduct protected and hands-on contact tours with large felids (see Table 2.1), there has yet to be any investigation of the impact of these on the animals involved. Guidance for the design of this study was sought from studies of other animals in behind-the-scenes tours, and from studies exploring the impact of visitors on large felids at zoo exhibits Methodologies employed in human-animal interactive research: zoo animals Studies exploring animal involvement in behind-the-scenes tours Despite the many and varied species involved in behind-the-scenes tours worldwide (as can be seen when searching zoo websites), it appears that peer-reviewed research of the involvement of animals in such tours is limited to captive dolphins, with four studies published between 1995 and 2005 (Brensing & Linke, 2003; Kyngdon, Minot, & Stafford, 2003; Samuels & Spradlin, 1995; Trone, Kuczaj, & Solangi, 2005; see Table 3.2;). As seen in Table 3.2, three of these are single-site studies; the majority include no more than five animal participants, with the exception of Kyngdon et al.'s (2003) multisite study of 22 dolphins. The majority of these studies have explored the impact of 104

123 participating in tours on animal behaviour, using within-groups, repeated conditions designs. This appears appropriate given the broad aims of exploring differences in behaviour within the same animals associated with participation in tours. While Brensing and Linke's (2003) study was concerned with animal responses to different groups of visitors, with these different groups representing different conditions, the remaining studies included additional, non-tour conditions to explore the impact of tours, such as conditions immediately before and/or after tours, as well as non-tour days. In this way, the tour sessions essentially represented 'experimental/treatment' conditions, while the non-tour sessions represented 'control/no visitor' conditions. Bishop, Mitchell and Hosey (2013) advocate for the use of 'no visitor' conditions as baselines from which to compare other levels of visitor impact. In the case of these studies, such designs have allowed conclusions to be drawn regarding the short- and/or long-term influence of tours more strongly than if behaviour was only observed during the tours. All four studies used behaviour observation methodologies. While Brensing and Linke (2003) used video recording to measure differences in specific behavioural attributes (such as swim speed) in tour sessions with different categories of visitors, the remaining studies used direct observation to explore the impact of tours on differing behaviour categories. Behaviours of interest varied across studies, but most also included additional outcome measures, such as use of refuge areas (where visitors are not able to interact with animals), and proximities between animals and/or between animals and visitors. Among these studies, varied sampling rules (e.g. ad libitum sampling) and recording rules (e.g. continuous recording) were followed. The studies also varied in their duration (ranging from 5-9mths), and the frequency of data collected. 105

124 Table 3.2. Summary of key methodological details of four peer-reviewed studies, published between , of captive dolphins involved in behind-the-scenes visitor tours Study focus Sample/Zoo Design Outcome measure/s Sampling method (recording medium, sampling rules, and Whether dolphins can distinguish between different groups of humans (adults, children <12yrs, children <12yrs with disabilities) during tours Impact of visitor tours on animal well-being Dolphins (n = 5)/Dolphins Plus (USA) Common dolphins (n = 3)/Marineland (New Zealand) Within-groups Repeated conditions Within-groups Repeated conditions Behaviour: - Distance to other dolphins and humans - Swim speed - Contact frequency with visitors - Contact distance to visitors Behaviour: - Use of refuge area - Frequencies and location of surfacing - Behaviour change (categories: other, touch, aggressive, submissive, abrupt, play) recording rules*) Video recorded behaviour obs^: - Focal animal, all-occurrence sampling during tour sessions Direct behaviour obs: - Focal animal sampling using two methods: 1) instantaneous time sampling of position and behaviour at 30s intervals before (>15min), during (~45min), after tours (~20min) for 3mths; 2) alloccurrence sampling of surfacing behaviour during tour sessions with and without swimming visitors for 3mths. Study duration/ data collection 9mth study period: x 30min sessions (no humans = 30 sessions, swimwith-program = 30 sessions, therapy program = 23 sessions) 6mth study period: x tour/non-tour sessions Study details Brensing & Linke (2003) Kyngdon et al. (2003) 106

125 Table 3.2. Continued Study focus Sample/Zoo Design Outcome measure/s Sampling method (recording medium, sampling rules, and Impact of visitor tours on animals during tours with and without trainer (keeper) control Short- (daily) and long-term (monthly) impact of visitor tours on animal behaviour Bottlenose dolphins (n = 22) /Dolphin Quest, Dolphin Plus, Theater of the Sea, and Dolphin Research Centre^^(USA) Dolphins (n = 5)/Marine Life Oceanarium (USA) Within-groups Repeated conditions Within-groups Repeated conditions Behaviour: - Social interactions with visitors or dolphins (categories: aggressive, submissive, sexual, abrupt, neutral/affiliative) - Behavioural states (categories: social vs non-social activities; activities under direct trainer control v not under direct trainer control) - Proximity to other dolphins (within 1m) - Details (age/sex) of visitors within 1m Behaviour: - Social behaviour (solitary and dolphin-dolphin) - Behavioural events (play, swimming, motor movements, orienting, sexual, aggressive, resting, miscellaneous) recording rules*) Direct behaviour obs: - Instantaneous focal animal sampling at 30s intervals during tours (mean duration range 17.7min min) and free time (mean duration range 19.2min min). - Continuous, all-occurrence sampling of social interactions during swims - Ad libitum sampling of unusual events and social interactions of non-focal animals Direct behaviour obs: - One-zero focal animal sampling for 10s periods every 5mins, before (90mins) and after (90mins) tours, and equivalent time during non-tour month. Study duration/ data collection 5mth study period: - 107hrs tour time and 94hrs free time across all organisations 6mth study period: - 12days obs/month (2mths without tours; 4 mths with tours) - Approx. 36hrs data/mth Study details Samuels & Spradlin (1995) Trone et al. (2005) * Recording medium = direct observation, video recording, etc.; sampling rules = ad libitum, focal, scan, or behaviour sampling; recording rules = continuous recording, or time sampling (including instantaneous sampling or one-zero sampling) (Martin & Bateson, 2007); ^ obs = observation; ^^ note this study has been included because although one site is a research centre, the remaining three are classified as zoos as per this thesis definition. 107

126 From this small body of literature, it is difficult to reach a consensus regarding the 'gold standard' method for measuring the impact of behind-the-scenes tours on animal behaviour, beyond the use of behaviour observations of the same animals across multiple conditions. It appears that decisions regarding the 'size' of the study is determined by the facilities available and the number of animals housed within them. In particular, it may be the case that organisations house a larger number of the species than those which are involved in interactive programs (as is true at both Australia Zoo and Zoos SA, personal observation), and this will determine the number of animals available at a particular site. It is also likely that decisions regarding the specifics of the behaviour observation method and the study duration are determined by factors such as the aims of the study, the number of animals, the size of exhibits, and the number of observers. Such decisions can be made during the pilot testing phase of a study (Bishop et al., (2013). Being different species to large felids (i.e., at a very basic level, dolphins are typically more social species which reside in water), it does not appear appropriate to use the behaviour categories from these studies to guide the present investigation. However, some of the features, such as recording of social/interactive behaviour and proximity to others, would likely be universal when exploring the influence of interactive tours, to allow for an understanding of how animals interact with and/or respond to people during such activities. To gain further guidance for directing a study of the impact of behind-the-scenes tours on large felids, some information can be gleaned from the broader literature exploring the impact of people on felids (small and large) in zoos Studies exploring the impact of humans on large felids in zoos A host of peer-reviewed research has explored the impact of humans on various species of zoo animals, particularly primates, investigating the influence of zoo visitor factors more frequently than zoo keeper factors. Typically, this research has involved behaviour change as the primary outcome measure, however a small number of studies have explored hormonal changes as indications of stress responses (Carlstead & Brown, 2005; Davis et al., 2005; Desportes et al., 2007; see Appendix C). A summary of the key details of this literature has been provided in Appendix C, however only those studies involving felids have been included within this chapter, given the focus on 108

127 understanding how best to measure how these animals may responds to participating in activities involving people. Just as the dolphin interactive literature was considered of limited value to guide large felid behaviour observations, so too is the human-impact literature on primates. Table 3.3 displays the key methodological details of what appear to be the only five peer-reviewed studies, published between 1993 and 2012, exploring felid responses to visitors in zoos. It does not appear that any studies have explored felid responses to keepers in this environment. A focus has been placed on visitor presence/absence, visitor density (i.e., numbers at an exhibit), visitor noise (i.e., how loud visitors are), visitor intensity (i.e., how active visitors are), and combined visitor factors. All of these studies were conducted at single sites (with the exception of Mallapur & Chellum's (2002) investigation at four Indian zoos), with sample sizes ranging from two to 14 animals. Those studies with the larger sample sizes were either conducted at multiple sites (Mallapur & Chellam, 2002), included multiple species (Margulis et al., 2003), or mother and cub groupings (O'Donovan et al., 1993). Each of these studies typically focused on one species of large felid only, except for Margulis et al.'s (2003), which included multiple large felids as well as one smaller felid. All species of large felids are represented in this literature. Given the aims of these studies being to explore the impact of different visitor factors on animals, it appears appropriate that all studies used a within-subjects design, with at least two conditions, depending on the purpose of the study. All studies investigated animal behaviour as their outcome measure. There is some variation across studies in terms of the individual behaviours explored, but these typically involve broad categories such as active and inactive species-typical behaviour (all used these categories except for Sellinger & Ha, 2005). Some studies involve other distinct categories, such as vocalisation (Maia et al., 2012; Margulis et al., 2003), social behaviour (by cubs; O'Donovan et al., 1993) and basic physiological needs (Maia et al., 2012). Measures of adverse behaviour include pacing (Mallapur & Chellam, 2002; Margulis et al., 2003; Sellinger & Ha, 2005), as well as reactive behaviour (such as slapping, spitting or hissing; O'Donovan et al., 1993), agonistic threat displays (Maia et al., 2012), and 109

128 Table 3.3. Summary of key methodological details of five peer-reviewed studies, published between , of the impact of zoo visitors on large felids Study focus (human element) Visitor presence/absence Visitor presence/absence + density Visitor presence/absence + density + noise Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Indian leopards (n = 14)/4 zoos (India) Lion (n = 1), amur leopard (n =2), amur tiger (n = 2), snow leopard (n = 6), clouded leopard (n = 2), fishing cat (n = 1)/Brookfield Zoo (USA) Cheetah (n = 14)/Fota Wildlife Park (Ireland) Within-groups: -Two conditions (visitor presence, visitor absence) Within-groups: - Two presence conditions (visitor presence, visitor absence) - Four density conditions (0, 1-10, 11-20, >20 visitors) Within-groups: - Two presence conditions: visitor presence, visitor absence - Four density conditions: 1-5, 6-10, 11-15, >15 visitors - Three noise conditions: none, low, medium Behaviour: - Activity: running, climbing, jumping, standing, walking, rubbing/rolling - Resting: sleeping, sitting, sitting erect, lying down - Stereotypy: pacing Behaviour: - State categories: inactive (asleep, alert, self-groom, out-ofview); active (forage, locomote, affiliative) - Event categories: vocalize, fight - Other behaviour: pacing Behaviour: - Adult categories: active (locomotor activity), inactive (resting, recumbent, sleeping) - cubs: active (locomotor activity, eating, suckling), inactive (resting, recumbent, sleeping), social (play, auto-or allogrooming) - React behaviour (slap, spit or hiss) - Location within exhibit rules*) Direct behaviour obs^: - Scan sampling using instantaneous sampling at 5min intervals, between 0600 and 1800hr Direct behaviour obs: - Scan sampling of behaviour states at each enclosure (n = 7) for 1min each on fixed route every 10mins (between ) - Ad libitum recording of event and pacing behaviours Direct behaviour obs: - Scan sampling using instantaneous scan sampling of behaviour and location at 30s intervals over 10min period per group (n = 5) (between and hrs) - Ad libitum recording of 'react' behaviour Study duration/ data collection 5mth study period: - 3 days per zoo: 2 control days (10hrs) in on-exhibit with visitors present; 1 day onexhibit with visitors absent (10hrs) - 120hr total obs 4mth study period: - 2hr sessions ( ), biweekly for 2mths (25hrs total); follow-up: 1hr sessions (between 1200 and 1400), 1-4x/wk - 13hrs total obs 2mth study period: - number of days/hours and total obs = NS Study details Mallapur & Chellum (2002) Margulis et al. (2003) O'Donovan et al. (1993) 110

129 Table 3.3. Continued Study focus (human element) Visitor density + noise Jaguars (n = 2)/Woodland Park Zoo (USA) Visitors - combined impact of multiple factors (Visitor noise and density by time of day) Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording rules*) Puma (n = 2)/Basque dos Jequitibas Zoo (Brazil) Within-groups: - Five density conditions: 1-10, 11-20, 21-30, 31-40, >41 visitors - Five noise conditions: quiet, low, moderate, high, extreme Within groups - Three noise conditions: high, low, silence - Density: counts Behaviour: - Categories: pacing, aggression, social, nonvisible, other (incl. skip pace) Behaviour: - Categories: locomotion, inactivity, basic physiological needs, cleaning and maintenance, agonistic threat, exploratory behaviour, vocalizations, direct gaze, refuge area (i.e. out of sight) Direct behaviour obs: - Continuous focal animal sampling over 30min period Video recorded behaviour obs: - Continuous recording for 2x 3hr periods ( hr and hr)/day Study duration/ data collection 8mth study period: - Approx. 8hrs/wk, with 2-4hr obs per visit at different times across different days (between 0700 and 2030) - 230hr total obs 6wk study period: - 2x 1week obs periods - 83hrs total obs * Recording medium = direct observation, video recording, etc.; sampling rules = ad libitum, focal, scan, or behaviour sampling; recording rules = continuous recording, or time sampling (including instantaneous sampling or one-zero sampling) (Martin & Bateson, 2007); ^ obs = observation. Study details Sellinger & Ha (2005) Maia et al. (2012) 111

130 aggression/fighting (Margulis et al., 2003; Sellinger & Ha, 2005). One study also observed animal locations within the exhibit (O'Donovan et al., 1993). The majority of the studies employed direct observation of behaviour, although Maia et al. (2012) used video recordings. Similar to the dolphin behind-the-scenes studies (see Section ), these studies varied in the specifics of their sampling method. The three studies with the largest sample sizes used scan sampling (Mallapur & Chellam, 2002; Margulis et al., 2003; O'Donovan et al., 1993). While these studies did not observe all included animals at once, this method seems appropriate given that there was usually more than one animal being observed at a time (making continuous recording of behaviour difficult if conducted directly, rather than by video-recording). Given that these were the studies with the largest sample sizes, this method appears the most appropriate. The two studies with the smaller sample sizes (n = 2; Maia et al., 2012; Sellinger & Ha, 2005) used continuous recording to provide a more complete overview of animal behaviour. Again, similar to the dolphin behind-thescenes studies, the duration of these studies varied. Study periods ranged from six weeks to eight months, with total durations of behavioural observations ranging from 13 to 230 hours. Interestingly, a study with one of the largest sample sizes (Margulis et al., 2003) collected data for the shortest period and a study with one of the smallest sample sizes (Sellinger & Ha, 2005) collected data for the longest. As such, it is again difficult to determine a 'gold standard' approach to measuring the behaviour of these species, and this is likely dependent on the aims of the study and resources available. As Bishop et al. (2013) suggest, it would seem logical that an important consideration is how the data relates to the study question. For example, while the total number of hours of observational data was relatively large (120hrs) in Mallapur and Chellum's (2002) investigation of the impact of visitors at leopard exhibits, this study involved 14 animals observed for three days each: two with visitors present and one with visitors absent. Considering the various factors that may fluctuate on a daily basis and potentially contribute to animal behaivour change (e.g., number of visitors on different days of the week, weather conditions, etc.), this small number of days may fail to capture such changes. However, if similar findings were observed across the multiple sites, then the inclusion of numerous sites increases the validity of the findings. 112

131 Rationale for Study Two methodology: Animal behaviour study The methodology used in Study Two (Chapter Five) was guided by combining information from studies of behind-the-scenes tours, and of the impact of visitors on felids at zoo exhibits. The overarching purpose of this study was to explore the potential behavioural responses of large felids associated with participation in behindthe-scenes tours. As such, a sample of large felids (lions, tigers, and cheetahs) involved in two different tour types (hands-on and protected contact) at Zoos SA was chosen (discussed further in Section 3.4). This sample was representative of the types of tours currently involving large felids in the Australian region (see Table 2.1), and of the most popular species of large felids housed in Australian zoos (Hibbard et al., 2012). The population of tour animals housed at Zoos SA included 10 lions, three tigers and three cheetahs. Although it was intended that all animals would be included in the study, logistical reasons prevented this (discussed in Section 5.2.1) and therefore the final sample included three lions, one tiger and three cheetahs. Falling in the mid-range of sample sizes for the dolphin tour (Table 3.2) and large-felid visitor (Table 3.3) studies, a sample of this size was deemed sufficient. As used in the aforementioned tour and visitor impact studies (see Sections and ), a single-zoo, within-groups, repeated conditions design was deemed most appropriate to explore the potential influence of participating in tours on the animals. To increase robustness of the potential findings, it was decided that comparisons would be made during tour days (that is, before, during and after tours), as well as at the same time on tour and non-tour days. In this way, the tour session represented an 'experimental'/'intervention' condition, while the other sessions acted as 'control'/'no human' conditions. Behaviour observation was the chosen methodology. Behavioural assessment is considered an inexpensive measure and a useful tool for measuring animal welfare (Hill & Broom, 2009). Not only has behaviour observation been used in all of the aforementioned literature, enabling the present study to contribute to this body of work in a comparable way, but it was also considered the most appropriate method 113

132 within the resources available for this program of research. Two alterative (or complementary) methods were considered: physiological measures and keeper rating. Physiological measures can be used to provide an indication of changes in animal welfare, and Hill and Smith (2013) highlight how the quantification of glucocorticoids is a commonly used measure. Such an approach has been used in human impact studies with other species in zoos; for example: urinary cortisol in Columbian spider monkeys (Davis et al., 2005); fecal cortisol in Black rhinoceros (Carlstead & Brown, 2005); and blood cortisol in Harbour porpoises (Desportes et al., 2007). However, these methods were considered inappropriate in the context of the present research for various reasons. Any one of the possible media (e.g., urine, feces, saliva, blood) would require collection by zoo staff, which would have conflicted with the aim of this research to have a minimal impact on zoo staff time and routines, and would have added another 'interaction' variable to the research. Collection of samples via saliva and blood would have required a more direct interaction (either with or without a physical barrier) between animal and zoo staff member, which would have added another interaction variable/factor to the data. Fecal sampling would likely have been the least invasive option, and would have required little interaction between animal and keeper for collection (beyond shifting animals from enclosures to collect samples). However, the time lag between interactive experiences and when cortisol changes in blood would appear in faeces would have made it difficult to determine immediate changes associated with interactions. Given the short-term focus of the study, using this method as a measure of welfare was deemed inappropriate. It is recognised, however, that such an approach would be useful in longer-term studies to assess the impact of human interaction (e.g., see Desportes et al. (2007) for an example of the use of this method to explore the role of husbandry training in reducing stress associated with invasive procedures in Harbour porpoises). Keeper rating of animal behaviour was also considered. Some studies have shown the potential for keepers to accurately report the behaviour (Wielebnowski, 1999) and personality traits (Gartner & Powell, 2011) of large felids by comparing these subjective measures with behavioural observations. However, during preliminary 114

133 discussions with Zoos SA staff prior to this project, keepers highlighted their interest in this study for the reason that they are usually not present to observe animal behaviour before and/or after tours (due to performing other tasks, such as food preparation and maintenance of other enclosures, at these times). As such, while keeper rating may have been useful for providing an account of what large felids do during tours, it would have been unlikely to allow for an accurate comparison of their behaviour at the other times of interest. With behaviour decided upon as the measure of choice, considerations were made regarding the medium in which this would occur, the behaviours of interest, and the specific sampling method. For practical reasons, given the size and shapes of the enclosures, direct behaviour observation was used, rather than video recording (as numerous cameras would have been required to capture the exhibit space). For comparability to large felid studies (see Table 3.3), key behavioural outcomes to consider included both active and inactive species-typical behaviours as measures of positive welfare, pacing and aggression as measures of potentially compromised welfare, and social interactions and proximities to humans as indicators of interactive behaviour. These were further refined during preliminary observations at the exhibits (see Section 5.2.4). Similarly, preliminary observations, rather than previous literature, dictated the sampling and recording methods used. With the aim of the study being to explore changes in animal activity budgets across the different conditions, scan sampling using instantaneous recording was considered most appropriate. This allowed for the frequencies of behaviour of all involved animals to be observed and recorded. Given that the number of animals observed at any given time was no larger than three, it was possible to document behaviour at relatively short time intervals (30s). Had there been a much larger number of animals, a focal sampling approach may have been more appropriate. Conversely, had there been fewer animals (or the potential for video recording to capture the full exhibits), continuous recording of each animal may have been possible. If the focus of the study had been different, for example to provide a more detailed understanding of what these animals do during the tours, then perhaps focal sampling with continuous recording would have been more appropriate (with such a method, durations of 115

134 behaviours could have been documented, along with more subtle interactions that may be missed when using instantaneous recording). The time frame selected for this study was a three-month period. Since previous literature varied in duration, this duration was based primarily on the aims of the study being to document short-term changes in behaviour. Had long-term changes been the focus (for example, to track how animal responses to participating in tours change as an animal ages, or how starting an interactive program alters the behaviour of previously non-tour animals), then a longer study duration would have been used. It was decided that conducting observations on one tour and one non-tour day for each animal group per week over this period would capture sufficient variation in changes to the zoo environment usually experienced by these animals (e.g., different weather conditions, different zoo keepers conducting tours, different structure of tour activities, etc.), as well as changes within the animals (e.g., illness, or days when animals may experience different levels of interest in participating in tours). Alternating the days of the week that each animal group was studied also allowed the data to include variation in visitor numbers (i.e., days when both tours and regular visitation were more and less popular - weekends and week days) Exploring an interactive experience from visitor perspectives (Study Three) To complement Study Two, the third phase of this research explored behind-thescenes tours from visitor perspectives at Zoos SA (Chapter Six). In particular, it sought to explore the impact of participating in such tours on visitor outcomes (such as learning and attitude change). This gap in the literature was identified based on the rationale that although there have been numerous studies of interactive zoo experiences (see Section 3.4.1), and numerous investigations of the impact of zoo visits in general (see Section 3.4.2), studies exploring the impact of participating in behind-the-scenes tours with any species, let alone large felids, are lacking. As such, guidance for the design of this study was primarily sought from survey studies of visitors participating in interactive zoo experiences, as well as from the broader literature of the impact of zoo visits. 116

135 In particular, the literature included here explores how past studies have surveyed zoo visitors, in order to understand the processes used to do this and the outcome measures employed. It is recognised that other methodologies can be utilised to explore visitor experiences, such as behaviour observations of visitors at zoos to explore: circulation patterns (Davey & Henzi, 2004); duration and viewing behaviours (Adelman, Falk, & James, 2000; Davey, 2006a); time spent engaging with interpretive elements (Ross & Gillespie, 2009); and differences in stay time and/or other at-exhibit behaviours to determine preferences for/popularity of different species (da Silva & da Silva, 2007; Margulis et al., 2003; Moss & Esson, 2010; Ward et al., 1998). While such methodologies can be very useful for understanding what visitors do during their visits, and are likely to be more rigorous than asking participants to self-report their visiting behaviours, they do no allow for documentation of changes in more internal visitor outcomes (such as learning). Similarly, it is recognised that survey methodologies have been used to explore other visitor constructs, such as: motivations for visiting (Morgan & Hodgkinson, 1999; Packer & Ballantyne, 2002); the impact of signage on animal visibility (Bashaw & Maple, 2001); and perceptions of the role of zoos (Reade & Warran, 1996). However, these again are outside the scope of the present focus on the impact of a visit on visitor outcomes and are not reviewed here Methodologies employed in human-animal interactive research: zoo visitors Studies exploring behind-the-scenes zoo visitor experiences To date, it appears that only three peer-reviewed studies have explored the influence of behind-the-scenes tours on zoo visitor outcomes. Key details of these studies are provided in Table 3.4. Between these studies (Curtin, 2006; Curtin & Wilkes, 2007; Weiler & Smith, 2009), two main approaches were taken; therefore, these are discussed separately herein. 117

136 Table 3.4. Summary of key methodological details of three peer-reviewed studies, published between , of the impact participating in behind-the-scenes tours on zoo visitors Study focus (zoo element) Recollections of wild and captive swimwith-dolphin program Post-experience cognitive dissonance after swim-withdolphin program Number of interpretive experiences with lions NS = Not stated Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/ data collection Phenomenol NS Curtin ogy (2006) Visitors, age = NS (n = 14)/multiple worldwide sites (NS) Adult visitors (n = 10)/multiple worldwide sites (NS) Adult visitors (n = 288)/Werribee Open Range Zoo (AUS) Phenomenol ogy Predictive correlational Medium: Interview (form = NS) Time point/s: Post-experience (2mths - 3yrs) Location: N/A Participation time: NS Sampling method: Purposive (advertisement placed on a university website) Medium: Interview (form = NS) Time point/s: Post-experience (2mths - 3yrs) Location: N/A Participation time: NS Sampling method: Purposive (advertisement placed on a university website) Medium: Self-report questionnaire Time point/s: Once only Location: Exhibit exit Participation time: 10mins Sampling method: Random Qualitative interview: - Questions = Qualitative - specifics NS (but primary question was 'What is it like to swim with dolphins?') - Based on past research: Yes Qualitative interview: - Questions = NS - Based on past research: Yes Questionnaire: - Cognitive, affective, and behavioural outcomes of interpretation (Interpretation and Evaluation Tool Kit - rating scale and yes/no questions) - Based on past research: Yes Study details NS Curtin & Wilkes (2007) 3 mth study period: - holiday and non-holiday periods, weekdays and weekends Weiler & Smith (2009) 118

137 The two studies by Curtin (2006) and Curtin and Wilkes (2007) make use of the same data set. In each study, a phenomenological (i.e., qualitative) approach was employed to understand the 'lived experience' of swimming with dolphins (in captive and wildlife settings in Curtin's (2006) study, and in captivity only in Curtin & Wilkes' (2007) study). This research was not conducted at the experience sites; rather, a purposive sampling approach was taken whereby participants were recruited via a university website. While useful for gaining a sample of participants who visited various sites (i.e., rather than sampling visitors at a single organisation), this adds variation to the results and may be considered problematic given the relatively small sample sizes. However, given that these studies are qualitative in nature, sample size is less of a concern. The qualitative approach used is heavily justified within each article, however, the authors provide little information regarding the types of questions asked during interview; this can be accounted for by the discussion-based nature of qualitative interviewing. As a guide for the present research, this work is prohibitive as it offers limited information regarding how to measure specific visitor outcomes. In a different study, Weiler and Smith (2009) used a predictive correlational design to explore whether the number of interpretive experiences (including behind-the-scenes tours) visitors (n = 288) had with lions at one Australian zoo influences their cognitive, affective and behavioural outcomes. Through an exit survey, visitors completed a 10 minute, previously validated survey instrument - the Interpretation Evaluation Tool Kit (Ham & Weiler, 2005) - to explore cognitive, affective and behavioural outcomes. While designed specifically for nature-based experiences, the instrument is not specific to zoos (i.e., it is designed for use in a range of interpretive wildlife settings), or to particular species. While this is useful for comparing research of interpretive experiences in general, it is less useful for exploring specific objectives of the site of interest (e.g., specific learning objectives of particular experiences, or measurement of specific behaviour change intentions). It also contains outcomes that may not be addressed at all types of interpretive experiences (such as appreciation of indigenous connections to nature). 119

138 Combined, these studies provide little guidance regarding methodologies to explore the influence of behind-the-scenes tours on visitor outcomes in the context of the present research. As such, the broader interactive/interpretive literature was turned to Studies exploring interactive/interpretive zoo visitor experiences A moderate body of literature has explored interactive/interpretive zoo visits - that is, opportunities that allow visitors to do more than simply viewing a zoo exhibit, by providing at least one additional element to the visit. The key methodological details of nine such peer-reviewed studies, published between 1998 and 2014, are displayed in Table 3.5. These studies vary in their focus and include numerous interactive/interpretive elements, such as keeper talks and training sessions, interpretive touch-tables, videos, and zoo theatre. Of these nine studies, samples were typically comprised of adult visitors. Sample sizes ranged from 67 to 639 participants (mean = 286; excluding Broad and Weiler's (1998) study as participants were described as groups, not individuals, and therefore total number could not be determined). While none of the studies report power analyses to determine whether samples were sufficient to detect effects, the samples appear appropriate to the aims of each study and the analyses conducted. All but two of these were single site studies, which again appears appropriate considering the aims of these studies to compare the impact of interactive experiences (i.e., the 'experimental' condition) to traditional exhibit viewing (i.e., the 'control' condition), and the availability of these different levels of experiences in the same zoos. Similarly, those that were multi-site studies were also appropriate, given that Ballantyne et al.'s (2011) study was interested in experiences at various tourism sites, and that Broad and Weiler (1998) were interested in different exhibition styles for one species. 120

139 Table 3.5. Summary of key methodological details of nine peer-reviewed studies, published between , of the impact of participating in interactive/interpretive experiences on zoo visitors Study focus (zoo element) Interpretive marine wildlife experiences in the 'wild' and in zoos Traditional versus interpretive tiger exhibits^ Touch-table at bearded vulture exhibit versus regular exhibit viewing Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/ Study data collection details NS Ballantyne et al. (2011) Adult visitors (n = 240)/Four unnamed organisations (AUS) Visitors - aged = NS (n = 138 groups)/western Plains Zoo and Dreamworld (AUS) Adult visitors (n = 639)/Tierpark Goldau (Switzerland) Qualitative, crosssectional Betweengroups Within- and betweengroups Longitudinal Medium: Web-based survey Time point/s: 4mths post-visit Location: N/A - online Participation time: NS # Sampling method: NS (part of larger study of visitors surveyed at wildlife experiences) Medium: Interview Time point/s: Once only Location: At exhibits Participation time: NS Sampling method: NS Medium: Self-report survey Time point/s: Immediately before and after experience, plus 2mth follow-up* (medium = NS) Location: At exhibit and follow-up Participation time: NS Sampling method: NS *but only reports on the second two time points Questionnaire: - Memories and responses to visit (open-ended questions about the experience) - Based on past research: No Questionnaire: - Cognitive, affective and behavioural learning (number and type of questions = NS) - Based on past research: No Questionnaires: Post-visit: - Knowledge gain (perceived and actual; openended, rating scale and multiple choice questions) Follow-up: - As above, plus knowledge-seeking behaviour since visit (question types = NS) - Based on past research: Yes (and expert involvement) 14 day study period: - different times/days/ public holidays 6wk study period at zoo: - 20 days (plus follow-up) Broad & Weiler (1998) Lindeman- Matthies & Kramer (2005) 121

140 Table 3.5. Continued Study focus (zoo element) Interactive palm oil education campaign Zoo theatre versus traditional viewing at polar bear exhibit^ Interpretive clouded leopard program versus traditional exhibit viewing^ Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/ data collection Adult visitors (n = Betweengroups, Medium: Self-report questionnaire 2yr study period: 403)/Melbourne Time point/s: 6mths before, 6mths into, - 8 days (one Zoo (AUS) Longitudinal 12mths into, and 6mths post campaign weekday and Location: Orangutan exhibit exit one weekend Participation time: NS day per time Sampling method: NS point) Adult visitors (n = 300)/Central Park Zoo (USA) Adult visitors (n = 112)/Point Defiance Zoo and Aquarium (USA) Betweengroups Betweengroups Medium: Survey (NS) Time point/s: Once only directly after experience at exhibit (n = 150 with theatre; n = 150 without theatre) Location: Polar bear exhibit Participation time: 1.5-2mins Sampling method: Random Medium: Self-report survey Time point/s: Once only Location: around zoo (interpretive element) and at exhibit Participation time: NS Sampling method: Random Questionnaire: - Knowledge about orang-utans (multiple choice quiz) - Attitudes toward orang-utans (rating scale; Animal Attitudes Scale) - Support for palm oil labelling (yes/no questions) - Behaviour - past behaviour, willingness to change future behaviour, perceived impact of zoo visit on future behaviour, and signing of petition in support of palm oil labelling (yes/no responses) - Subjective norm regarding orang-utan conservation (rating scale) - Experience satisfaction (rating scale questions) - Based on past research: Yes Questionnaire: - Knowledge gain - perceived and actual, as well as identification of where knowledge was gained (types of questions = NS) - Satisfaction and enjoyment with experience (types of questions = NS) Based on past research: No Questionnaire: - Cognitive/affective learning (2x open-ended questions) - Attitudes regarding quality of life and care of the animals (rating scale, yes/no and open-ended questions) - Based on past research: Partially 5mth study period: - various times/days NS (but conducted at various times/days of week) Study details Pearson et al. (2014) Penn (2009) Povey & Rios (2005) 122

141 Table 3.5. Continued Study focus (zoo element) Interpretive bird of prey presentation Interpretive versus exhibit elephant viewing Interpretive training, fact only training, and training with no presentation at black rhinoceros exhibit Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/ data collection Adult visitors (n = Within- and 2mth study 175)/Healesville betweengroups, period at zoo: Sanctuary (AUS) - stratified by Longitudinal times/days/ holiday periods (plus follow-up) Visitors aged > 10yrs (n = 355)/Zoo Atlanta (USA) Children visitors - 5 th graders (n = 67)/Potter Park Zoo (USA) Betweengroups Betweengroups Medium: Interview Time point/s: At zoo and follow-up (5-6mths later; those who indicated action intentions (n = 38)) Location: Zoo exit and telephone interview Participation time: 5 mins Sampling method: Random Medium: Survey NS Time point/s: Once only Location: Zoo exit Participation time: NS (several minutes) Sampling method: Random Medium: Quiz Time point/s: Once only Location: NS (immediately after experience) Participation time: NS Sampling method: Convenience sampling of classes on school excursion (all students asked to participate) NS = Not stated; ^ studies included other, non-interpretive or non-survey elements which are not relevant for inclusion here Questionnaire: Post visit: - Knowledge of conservation behaviour recycling and removing road kill from the road (yes/no and open-ended questions) - Behavioural intentions (current and future) (yes/no and open-ended questions) Follow-up: - Behaviour since visit (yes/no and open-ended questions) - Based on past research: No Questionnaire: - Current and prior experience with elephants and elephant interpretive exhibits (multiple-choice questions) Behaviour: - Conservation behaviour (signing a petition for/against ivory trade; and returning a solicitation card, either upon exit or post to zoo after visit) Based on past research: Partially Quiz: - Knowledge (3x open-ended questions) - Based on past research: Yes 2wk study period: - daily at varying times of day 3wk study period: - 1x class/wk Study details Smith et al. (2008) Swanagan (2000) Visscher et al. (2009) 123

142 Among these papers, study designs vary according to the aims of each investigation, but it is not surprising that between-groups designs are most common. With this literature primarily exploring different experience types by opportunistically surveying visitors during their visit, it would not likely be possible or feasible to survey the same participants after different experiences. However, between-groups designs may make it difficult to draw conclusions about the impact of a different experience, since other differences between participant groups might contribute to differences in the outcome/s of interest. To account for this, it is important to consider ways in which participants might differ (e.g., age, gender, previous zoo/animal experience) and control for these (e.g., Pearson et al. (2014) controlled for differences between groups at each time point). The majority of studies were conducted at one point in time (i.e., after the particular experience), which is sufficient to determine the immediate/short-term impact of an experience. Two studies (Lindemann-Matthies & Kamer, 2006; Smith, Broad, & Weiler, 2008) also included follow-up elements, whereby visitors were surveyed some months after their visit to determine longer-term effects. The use of follow-up components is a credit to these studies, adding to the strength of conclusions drawn regarding the impact of interactive experiences and an understanding of what visitors do after their visit (although caution should be taken when interpreting results from self-reported, rather than actual, behaviour change data as it may suffer from reporting bias). In contrast, given the inherent difficulties in drawing causality between a zoo experience and post-visit outcomes, such as knowledge and behaviour (due to the various other activities/experience people may engage in), it is not surprising that the majority of studies did not include a follow-up component. A survey method was used in all investigations, with questionnaires being completed either by the participant (in person or online) or via interview (either in person or over the telephone for follow-up studies). Usually, these were conducted at the zoo exhibit of interest, after an experience had taken place. Most studies did not mention the time taken for visitors to participate, and those that did reported varied times (1.5-2 mins for Penn (2009); 5 mins for Smith et al. (2008); 10 mins for Weiler and Smith 124

143 (2009)). Reporting of participation time can be important for understanding what a reasonable questionnaire length might be, particularly in terms of the potential burden placed on zoo visitors. Given that visitors (usually) pay to enter a zoo, it should be important for researchers to ensure that minimal time is taken away from their visiting experience (as this may leave them with a negative perception of their experience and may deter them from visiting again). While there do not appear to exist any guidelines for how much is 'too much', of those three studies reporting this information, the time investment does not appear excessive. Given the nature of the research environment, all zoo-based studies rely on convenience sampling to obtain participants, yet the specific method for sampling participants failed to be mentioned in four of the studies and so it is unclear whether any sampling biases occurred. Five studies described random sampling procedures (such as approaching every n th visitor) which, when coupled with sampling over a variety of times and days, provides the opportunity to capture a sample reflective of the wider zoo attendee characteristics and increases generalisability of findings. With the exception of Ballantyne et al. (2011), all researchers reported some details of their study duration and number of data days. All varied in this regard, with no patterns evident other than that many specified that data were collected over various times of day and days of the week, including holiday and non-holiday periods. Again, this improved the representativeness of the samples. Perhaps most important to the present study is the types of outcomes measures used in this past literature. As might be expected, given the differing aims of these studies, there is variation across all studies regarding the outcome variables and measures used to investigate these, as well as in level of detail provided about these measures. Many studies investigated multiple outcomes, with the most common being: knowledge gain, attitudes, and conservation behaviour. With the exception of Ballantyne et al.'s (2011) study of various marine wildlife experiences, the remaining studies focused on single species and therefore the outcome measures were largely specific to these animals (e.g., knowledge of black rhinoceros; Visscher et al., 2009). In addition, the studies reveal no validated measures of these constructs, with the 125

144 exception of Pearson et al. (2014) using an adaptation of the Animal Attitudes Scale and applying this to orang-utans. As such, these studies indicate that validated measures of these constructs in this setting are not presently available, and that instead it is likely that the design of questionnaire items is dictated by the focus of the individual study. In the absence of such measures, it would be expected that elements of past research be drawn upon when purpose-designing measures and seven of the studies reported doing this to some degree. Despite this lack of information, a trend across the literature is for the use of both closed- and open-ended questions to explore these constructs (e.g., multiple choice questions to assess knowledge, followed by open-ended questions regarding additional information learnt). This appears an appropriate and logical approach, as it can allow for specific comparisons to be made (through closed questions) and for more depth to be obtained to explain any differences (through open-ended questions). In particular, open-ended questions appear appropriate when asking visitors what they have learnt or behaviours that they might be willing to engage in (as many of these might be missed if using a pre-determined list). Of the studies measuring knowledge gain, studies explored both perceived and actual changes in knowledge, through multiple choice and open-ended questions. Close-ended knowledge questions appear appropriate if the purpose is to determine whether visitors have learnt specific information (e.g., as in Pearson et al., 2014), whereas open-ended questions appear more appropriate when seeking to understand what visitors learnt (e.g., as in Lindeman-Matthies & Kramer, 2005). Studies exploring attitude change typically relied on rating scale questions regarding attitudes toward the particular species (e.g., Pearson et al., 2014), and the quality of life and care of the animals (Povey & Rios, 2005). Studies exploring behaviour largely sought to understand the types of conservation behaviours visitors were engaging in before their visit and their willingness to engage in behaviours in the future, through yes/no (e.g., Pearson et al., 2014) and/or open-ended questions (e.g., Smith et al., 2008). Additionally, Swanagan's (2000) study included on-site opportunities for engaging in conservation behaviour (petitions and solicitation cards). 126

145 All studies primarily used a self-report method to measure their constructs. Although appropriate for understanding internal constructs, self-report methods can be open to biases and inaccuracies, and caution should be taken when interpreting results. For some variables, such as knowledge gain, this method appears most appropriate and less susceptible to inaccuracies (other than that correct answers could be guessed when using a closed-question format, particularly if an 'unsure' option is not provided). In contrast, the results of studies using this method to measure behaviour change, by asking visitors to report past/current/future intentions, should be viewed with caution. Particularly when conducting research in zoos, where it often expressed (e.g., through signage, keeper talks, etc.) that pro-conservation behaviours are encouraged, participants might demonstrate a positive reporting bias. In some instances, it may be possible to document actual behaviour change (e.g., Swanagan, 2000). This may provide a more accurate measure of participants' willingness to engage in pro-conservation behaviour. However, the specific nature of some conservation behaviours (e.g., signing a petition) makes them potentially easier to measure, whereas for more general, on-going behaviours (such as recycling) a selfreport measure is likely to be more feasible, despite the inherent limitations Studies exploring general zoo visitor experiences While the literature discussed in Section can provide some guidance for the exploration of interactive zoo experiences, a key limitation of this body of research is the lack of consistency with regard to how visitor outcomes (such as knowledge, attitudes, and behaviour) may be measured. As such, the broader zoo visitor literature, which explores the impact of zoo visits in general, was also explored to provide further insight here. A total of 15 such studies, published between 1992 and 2013 were accessed and explored. Since the purpose of exploring this literature was to gain further insight into the outcome measures used, other methodological details are not described in detail here (see Appendix D). To briefly summarise, however, these studies investigated a range of elements; some use correlational and predictive correlational designs to explore how certain variables predict visitor outcomes (e.g., Bitgood, 1992; Pearson, Dorrian, & Litchfield, 2013), while others use combined within- and between-groups designs to explore changes in visitor outcomes before 127

146 and after visits (e.g., Adelman et al., 2000). Typically, these were single zoo studies of adult visitors, with samples ranging from 30 to 1000 participants (mean n = 531). As with the interactive studies, a survey method was used in all investigations (often conducted via interview, but also by participant self-completion). These were conducted at various locations around the zoo (depending on the focus of study). Seven studies reported participation time, ranging from 5-30 minutes for single surveys, and up to 35 minutes for participants who competed both pre and post instruments. Regarding visitor outcome measures specifically, similar to the interactive literature, outcomes such as knowledge, attitudes and behaviour are common across these studies. Unfortunately, however, limited consistent information is available regarding how to measure these constructs. Like the literature discussed in Section , a range of closed- and open-ended questions are usually purpose-designed to match the focus of the study (e.g., the particular species) and none of these studies focused on large felids. An additional visitor outcome in this body of literature is the study of visitor emotions, explored through self-rated emotions on pre-determined scales and/or open-ended responses (Clayton, Fraser, & Saunders, 2009; Luebke & Matiasek, 2013; Myers, Saunders, & Birjulin, 2004). As such, this literature confirms the trend for visitor impact research to be tailored to the aim of the particular study, and for outcome measures to be developed that align with this Rationale for Study Three methodology: Zoo visitor survey The methodology used in Study Three (Chapter Six) was guided by a combination of information from studies of behind-the-scenes tours, interactive experiences, and general zoo visits. The key purpose of this study was to explore whether behind-thescenes tours are effective in altering zoo visitor outcomes. As such, a range of large felid interactive experiences offered at Zoos SA (detailed in Section 3.4) were chosen as the tour/'experimental' conditions, and regular viewing of large felid exhibits was chosen as the 'control' condition. Given that this one organisation offered different 128

147 levels of experiences of interest, collecting data at additional zoos was not necessary to fulfil the aims of the research. While the previously discussed interactive studies typically used between-groups designs to compare visitor responses to different types of experiences, a dual design was chosen for this study. With the study aiming to explore the impact of participating in a behind-the-scenes tour on visitor outcomes, a within-groups approach was taken by surveying participants before and after their experience. Further, to explore whether such tours are more effective than simply viewing an exhibit, a between-groups approach was taken to survey regular exhibit visitors either before or after viewing a large felid exhibit. In this way, conclusions could be drawn not only about the experiences in isolation, but about the impact of tours with reference to typical experiences. The sample chosen was adult visitors, given that behind-the-scenes tours primarily involve adults. This was also based on the desire to contribute to the current body of literature using a comparable sample (since the majority of the literature described in Section includes adults only and children may differ from adults in the way they experience zoos). Determining an appropriate sample size was difficult, given that past studies do not report power analyses or size of effects, and because these studies measure similar outcomes using different approaches. Guidance was sought from a leading Australian tourism researcher, who suggested that a minimum of 30 participants per condition is usually considered sufficient for a study comparing different visitor conditions, but that 100 participants per condition would be expected in a thesis focusing solely on visitor surveys (Packer, Personal Communication, 2010), which is consistent with past interactive studies (Table 3.5). As such, 100 participants for the interactive and regular exhibit conditions were aimed for (and exceeded). A random sampling approach was taken for all exhibit visitors, while all tour visitors were invited to participate, given that the small number of potential participants on each tour would have resulted in only one participant per tour session (and therefore significantly increased the data collection period). 129

148 As is common in this body of research, a self-report survey format was chosen. Not only was this considered less intrusive for zoo visitors, but also more time-efficient. In particular, while interviewing visitors at exhibits may have been feasible, when surveying zoo visitors before a tour, having one researcher available to interview participants would have meant only one visitor could be surveyed per session, as visitors typically arrived no earlier than minutes before their experience. Similarly, interviewing participants after a tour would have meant that the remaining tour visitors would have been required to wait to be interviewed. This was considered to be disruptive to their experience. In the absence of consistent past literature exploring visitor outcomes following interactive and regular zoo visits with respect to large felids 12, a self-report questionnaire was purpose designed for this study. Discussed further in Section 6.2.3, two questionnaires were generated (one pre-visit and one post-visit), with wording adapted to match the type of experience (interactive or exhibit) and species encountered (lions, tigers, or cheetahs). Based on the type and format of questions asked by studies listed in Tables 3.4 and 3.5, in brief the questionnaires contained the following: 1. Demographic and work-experience questions; 2. Multiple choice, rating scale, and open-ended questions concerning knowledge of the animals; 3. Rating scale questions concerning attitudes towards the animals and conservation in general; 4. Open-ended questions regarding emotions experienced when seeing and learning about the animals (post-experience questionnaire) 12 Although two studies explored large felid experiences (Broad & Weiler, 1998; Weiler & Smith, 2009), these offered little insight for designing the present questionnaire items. In their investigation of traditional versus interpretive tiger exhibits, Broad and Weiler (1998) failed to provide information about their measures. While Weiler and Smith (2009) used the Interpretation Evaluation Tool Kit to explore the impact of the number of lion experiences on cognitive, affective and behavioural visitor outcomes (see Section ), the scale is not specific to the species of interest. 130

149 5. Rating scale questions regarding current/proposed future engagement in conservation behaviours (species-specific and general); 6. Rating scale questions regarding experience expectations/satisfaction, and open-ended questions regarding positive and negative aspects of the visit (post-experience questionnaire only). In line with those past studies which provided information of the time taken to complete their questionnaires, and a requirement of the participating organisation (Zoos South Australia) being to keep the survey brief so as not to disrupt the visitor experience, the instrument used in this study was designed to take approximately 15 minutes to complete (as confirmed by peer-checking of the instrument). Data were collected over a nine-month period (between September April 2011), during regular zoo hours on a variety of weekends and weekdays, as well as public and school holidays. This allowed for a diverse and representative sample of zoo visitors to be approached Study site for Studies Two and Three: Zoos South Australia With an understanding of the rationale for methodologies used within this thesis, a description of the study sites used in Studies Two and Three is provided here to add context to the research chapters (Chapters 5 and 6 respectively). Zoos South Australia (Zoos SA) is a non-profit conservation charity that undertakes field-based conservation work as well as runs two zoological parks in the Adelaide region: Adelaide Zoo and Monarto Zoo. A description of the large felid facilities and visitor experiences at each are discussed herein (additional animal husbandry details are discussed in Section 5.2.2) Adelaide Zoo Adelaide Zoo is situated approximately 2 km, or a 15 minutes walk, from the Adelaide CBD. Covering eight hectares, the zoo houses nearly 300 native and exotic species (mammals, birds, reptiles and fish; for complete zoo map, see Appendix E). 131

150 Opening to the public in 1883, the zoo is the second oldest zoo in Australia. Like other urban zoos, visitors explore this facility by foot, either at their own leisure or as part of guided Walkabout tours that depart half-hourly throughout the day (these tours are conducted by zoo volunteers, who take visitors through various sections of the facility). During data collection, Adelaide Zoo housed three unrelated adult tigers: a male, Tuan, and two females, Kimiri and Assiqua. Due to tigers being solitary species, these animals were housed in individual enclosures, both on- and off-exhibit. Adelaide Zoo also housed three adult lions - a male, Mujambi, and a sibling pair of females, Yizi and Amani during data collection. The male was always housed separately from the females (who were usually housed together) as he experiences a medical condition making him prone to seizures and therefore vulnerable around other lions. In addition to viewing these large felids in their exhibits, visitors could watch keeper talks at exhibits (which usually include animal feeding), or pay an additional fee to go on a behind-the-scenes tour. Details of the exhibits and visitor experiences are discussed in Table 3.6; images of the tiger exhibit and signage are shown in Figure 3.1, and of the lion exhibit and signage in Figure

151 On-exhibit signage Two panels of information >10 panels of information Keeper talks Daily at 2pm (at main exhibit): Behind-the-scenes tour Big Cat Encounter (Protected contact) Table 3.6. Summary of key details of tiger and lion exhibits, signage, keeper talks and behind-the-scenes tour at Adelaide Zoo On-exhibit enclosure/s and features Tigers Lions Two adjacent enclosures with separate viewing areas: 1) Main enclosure (usually housing Tuan*): - Glass viewing area - Fenced viewing area with a moat - Large amount of naturalistic foliage, making animal visibility difficult at times 2) Smaller enclosure (usually housing Kemiri): - Smaller viewing area - Platform overlooking exhibit - Approx. 15mins duration - Information provided about species in wild and zoo - Usually involves the provision of an enrichment or feeding item - Usually better visibility, with animal closer to viewing areas, and increased animal activity - Keepers available to answer visitor questions Four adjacent exhibits (one larger and three smaller) with continuous viewing area: - Mesh fencing - Limited vegetation making visibility very clear Four days/week (Tues, Thurs, Sat, Sun; plus public holidays) at 1.40pm: - Approx. 15mins - Information provided about species in wild and zoo - Usually involves feeding through a chute (or sometimes enrichment) - Keepers available to answer visitor questions - 3 days/week (Tues, Thurs, Sun) - 1hr duration - $145/pp (less for Zoos SA members) - 6 visitors/session (minimum age 12 years) - Toured first through the tiger on- and off-exhibit facilities (15mins per tiger; viewed from outside enclosures), then through the lion on- and off-exhibit facilities (15mins total; viewed from within one of the enclosures with animals removed) - Informed about the individual animals and species in the wild - Feeding through mesh fencing - Photographs through mesh fencing * Assiqua is usually housed off-exhibit in the tiger holding facility, but occasionally has access to one of the on-exhibit enclosures (at this time, Tuan or Kemiri would be housed off-exhibit). 133

152 Figure 3.1. Top: Adelaide Zoo tiger exhibit (front view from main visitor viewing area); Bottom: Examples of Adelaide Zoo tiger exhibit signage. Monika Szokalski 134

153 Figure 3.2. Top: Adelaide Zoo lion exhibits (left one of the three smaller exhibits; right the larger exhibit); Middle and bottom: Examples of Adelaide Zoo lion exhibit signage. Monika Szokalski 135

154 Monarto Zoo Monarto Zoo is an open-range facility, situated approximately 65 km from the Adelaide CBD (a one-hour drive by car). Opening to the public in 1993 and covering 1,000 hectares of land, the zoo houses both native and exotic, primarily threatened species (for complete zoo map, see Appendix E). Visitors can explore the park by foot, traveling along the various walking tracks connecting the different exhibits (approximately 10km in total), or they can catch the Zu-loop shuttle bus, transporting them through/past various exhibits and stopping at the different wildlife viewing platforms (approximately 1 hour for a full loop of the park). The bus departs the visitor centre regularly, at 30 minute intervals during the week and 15 minute intervals on weekends, public and school holidays. Volunteer zoo guides are present at each of the viewing platforms to answer visitor questions and are also present on each bus to narrate as visitors are driven through the park. During data collection, Monarto Zoo maintained two coalitions of male cheetahs on display to the visiting public (as well as females which were housed separately, offexhibit). One coalition was the focus of this research (as the other coalition and females were not involved in interactive tours). This coalition contained three adult brothers Askari, Skukuza, and Tsotsie born at the zoo in 2004, housed together. Monarto Zoo also housed a pride of two unrelated adult male lions (Leroy and Inkosi), four adult females (siblings, Tiombe and Kibira, their half-sister, Zalika, and unrelated Kiamba), and one female cub (Jahzara, daughter of Kiamba), housed in varying combinations (with the exception that the mother and cub were housed separately). Like at Adelaide Zoo, in addition to viewing the exhibits, visitors at Monarto Zoo could also witness daily keeper talks and, for an additional cost, participate in various behind-the-scenes encounters. Details of the exhibits and visitor experiences are discussed in Table 3.7; images of the cheetah exhibit and signage are shown in Figure 3.3, and of the lion exhibit and signage in Figure

155 On-exhibit signage 6 panels of information 5 panels of information Keeper talks Daily at 3pm (at main exhibit): Behind-the-scenes tour Cheetah Interactive' (Hands-on contact): Table 3.7. Summary of key details of cheetah and lion exhibits, signage, keeper talks and behind-the-scenes tours at Monarto Zoo On-exhibit enclosure/s and features On-exhibit enclosure viewing Cheetahs Lions Three adjacent enclosures: 1) Main enclosure (usually housing Askari, Skukuza & Tsotsie): - Platform overlooking exhibit - Touch tables available when zoo volunteer present - Naturalistic vegetation making for limited visibility depending on animal location 2) Other enclosures: - Contain similar vegetation to main exhibit and animals somewhat visible from platform Either from viewing platform or via bus (driven inside one exhibit and adjacent to the others) - Approx. 30mins duration - Information provided about the hands-on approach with these individual animals, as well as information about species in general - Visitors witness either animal feeding, or informal positive reinforcement training session - Keepers available to answer visitor questions - 3 days/week (Mon, Thurs, Sat) at 9.45am - 1hr duration - $170/pp visitors (minimum age 16yrs) - Visitors enter main exhibit with animals present - Introduced to each animal - Informed about species in wild and in zoo - Visitors witness positive reinforcement training, and animal feeding (on Saturdays) - Visitors able to pat animals and have photos taken with them One main exhibit: - Platform overlooking exhibit - Touch tables available when zoo volunteer present - Naturalistic vegetation - Limited visibility given size of exhibit (~8ha) depending on animal location Either from viewing platform or via bus (driven inside the exhibit) Daily at 12.30pm: - Approx. 30mins duration - Overall carnivore talk - not specific to lions (also covering painted dogs and hyenas) - Information provided about species in wild and zoo - No lion feeding - Keepers available to answer visitor questions 'Lions at Bedtime' (Protected contact): - 3 days/week (Tues, Fri, Sat, Sun) at 3.15pm - 1hr duration - $125/pp visitors (minimum age 10yrs) - Visitors assist keepers 'calling' lions into night quarters - Feed lions through feeding chute in mesh fencing - Informed about the animals - Photographs through mesh fencing - At time of collecting data, a 3.5mth old cub was housed off-exhibit able to be viewed 137

156 Figure 3.3. Top: Monarto Zoo cheetah exhibit (view from visitor viewing platform); Bottom: Examples of Monarto Zoo cheetah exhibit signage. Monika Szokalski 138

157 Figure 3.4. Top: Monarto Zoo lion exhibit (front view from visitor viewing platform); Bottom: Examples of Monarto Zoo lion exhibit signage. Monika Szokalski 139

158 3.4. Conclusion This chapter has presented a summary of the methodologies used in similar humanzoo animal interaction research, providing a rational for the methodologies used for each study included within this thesis. In addition, the large felid facilities and visitor experiences at Zoos South Australia have been described, providing an overall backdrop to understand and interpret the individual studies. The chapter to follow will now describe the first study conducted as part of this thesis. 140

159 4 What can zoo keepers tell us about the handling and training of large felids in zoos? 13 I have had many friends among lions, tigers and panthers, which were no harder to handle than pet dogs. Moreover their affection is very enduring, and survives long after they have found another home. C. Hagenbeck (1909, p. 98) The previous chapters highlighted that little is understood about the benefits and limitations associated with large felid-human interactions in zoos. Therefore, this chapter describes a study which investigated this issue by asking zoo keepers how they interact with their large felids and what they think of different interactive practices Introduction Recently, there has been a shift in the way large felids are managed, with various zoos breaking the barrier between these animals and their keepers. Traditionally, large felids were housed solitarily in barren concrete cages (see Section 2.4.2). Today, housing conditions are more naturalistic, with exhibits more accurately reflecting the species natural habitats, and keepers spending more time with their animals. As identified in Section , current keeper interaction with large felids can be 13 This chapter has been published as: Szokalski, M.S., Litchfield, C.A., & Foster, W.K. (2013). What can zookeepers tell us about interacting with big cats in captivity? Zoo Biology, 32, See Appendix F. 141

160 categorised into three key levels of contact: hands-on, protected, and hands-off contact. In addition, some keepers may also train these animals (see Section ). Given the present lack of understanding of the consequences of engaging in these practices, keepers are in a valuable position to shed light on outcomes of these techniques Potential benefits of interactions for large felids, keepers, and visitors Research related to other species, primarily non-human primates, indicates that captive animals may benefit from both increased interactions with their keepers/carers and training. As discussed in Section , animals managed in more interactive ways have been found to experience improved physical well-being, as it can allow for increased health monitoring and early detection of problems, and increased psychological well-being, as animals may form stronger bonds with keepers/carers, and may be given greater choice and control over their environment, cumulatively resulting in reduced stress. Other potential benefits include cognitive stimulation, as a result of processing and using information gained during interactions and training, and social stimulation through the interactions. Additionally, behavioural stimulation may occur, as increased contact and/or training programs may provide opportunities for animals to engage in a range of behaviours from their natural behavioural repertoire, some of which are often redundant in the zoo environment (such as sprinting for cheetahs). Finally, these practices may reduce negative behaviours, such as aggression and stereotypy. Similarly, as discussed in Section , information derived from research with other species suggests that keepers may benefit from interactions with large felids, particularly as a result of the strong bond and relationship that may be formed. Incrased interactions may potentially assist the performance of zoo keeper duties, such as by making it easier to shift/move animals, allowing better enclosure maintenance and improved care of the animals. Combined, these factors could lead to increased job satisfaction. 142

161 Finally, as outlined in Section , visitors may also benefit from large felidkeeper interactions, particularly if the animals involved engage in more speciesappropriate behaviours and demonstrate less boredom and/or stereotypy. These animals may be more favourably perceived, potentially improving the visitor experience and the opportunity for visitor education. Such practices can also allow for public training sessions, keeper talks, and interactive animal encounters, which may increase the visitor-animal connection, as well as present opportunities to deliver educational messages in a more enjoyable and emotionally engaging manner (explored further in Section 6.1). This may contribute to a greater influence on visitor conservation behaviours, such as signing petitions in support of animal conservation Potential limitations of interactions The possible limitations associated with large felid-keeper interactions must be recognised (see Section ), the most obvious being the potential for physical injuries/fatalities to humans. Like all animals, large felids can be unpredictable and their sheer size, coupled with their natural hunting instincts and carnivorous nature, pose realistic risks to people and animal euthanasia may also result. Further suggested limitations include the initial time investment required for keepers to establish a handling and/or training program, and that interacting with humans in this way typically does not occur in natural habitats and may have negative behavioural and welfare consequences. There is also the possible risk of inappropriate messages being sent to the visiting public, particularly regarding the dangerous nature of these animals and the mis-perceptions that such interactions might encourage Aims of the current study Despite their crucial role in maintaining captive collections, keepers are rarely, but increasingly, included in zoological research, with various studies demonstrating the potential for keepers to contribute to the scientific literature (see Section 3.2.1). As an understanding of large felid-keeper interaction practices is in its infancy in the scientific literature, keepers are well-placed to provide vital information and to direct future research in this area. 143

162 The current study addressed the following questions: 1. How common are the different handling methods (hands-on, protected and hands-off) and training with large felids? 2. How beneficial do keepers perceive each of the three handling methods and training to be for large felids, themselves, and zoo visitors? 3. What are the additional keeper-perceived benefits and limitations of each of these practices? 4. How do the benefits/limitations of handling differ from the benefits/limitations of training? 4.2. Methods Recruitment and participants A self-report questionnaire was completed by 86 large felid keepers (61 female, 23 male, 2 unspecified), who were recruited through six of the 10 major international online zoo keeper forums (Australasian Society of Zoo Keeping, American Association of Zoo Keepers, Association of British and Irish Wild Animal Keepers, Zoo Biology, Australasian Zoo Keeping, and Asociacion Iberica de Cuidadores de Animales Salvajes), as well as through via seven of the 14 zoos/wildlife parks in Australia and New Zealand with large felid collections (Zoos South Australia, Zoos Victoria, Taronga Zoo, Cairns Wildlife Safari Reserve, Wellington Zoo, Auckland Zoo, and Hamilton Zoo; for full lists of forums and organisations contacted, Section 3.2.2). The remaining four forums and seven zoos either did not respond or did not consent to participate. An individual response rate could not be determined based on it not being known how many large felid keepers are members of each forum or employed at each zoo. Similarly, a completion rate could not be determined, as the online survey system did not offer this function. Respondents ranged in age, with 37% aged between 20 and 29 years; 31% aged years; 27% aged years; 4% aged years; and 1% aged 60+. The majority of participants were employed in public facilities (n = 63), with a further 19 in private facilities (4 other/unidentified). Approximately half worked in the United States 144

163 (51%), 16% in Australia, 11% in New Zealand, and 11% in the United Kingdom (11% from various other countries; 1 unspecified). Most participants were working in keeper roles (n = 50), with a further 26 holding senior keeper positions, and 7 holding higher positions (such as zoo manager or director; 3 unspecified). Mean years' of employment in current role was 6.29 (SD = 6.54), with a minimum of 2 months and a maximum of 33 years. A large proportion of participants (69%) had previously worked in the zoo industry, with a range of 8 months to 35 years experience (M = 7.65 years, SD = 1.66) Questionnaire A self-report questionnaire was generated in TellUs2 a University of South Australia online survey software application (KnowledgeSouth Survey IT Version 2.0, 2001). In addition to demographic details, the following information was sought (all sections of the questionnaire used have been described here; see Appendix G for the complete version of the questionnaire in its online format): Determining current handling practices and use of training: Participants were asked which species of large felid/s they currently work with as an individual keeper, what level of handling they practice with them, and whether or not they practice training. Participants were also asked to indicate, on average, how much daily time they spend in contact with, and/or training, the animals. Basic opinions of the different handling practices and training: Participants were asked how strongly they agreed that the current handling and training practices employed at their institution allow the needs of large felids, keepers, and visitors to be simultaneously met. They were asked which of the three levels of contact they believe to be best for each of the three groups, and whether or not training is beneficial to each of the three groups (yes/no/unsure response). Information specific to the different levels of handling and training: To allow a more in-depth understanding of participant opinions, respondents were asked to rate how 145

164 strongly they agree (where 1 = strongly disagree and 5 = strongly agree) that each level of handling and training contributes to five different benefits for large felids, keepers, and visitors (these benefits were derived from the literature about these practices with other species, see Sections , , and ; limitations were not assessed due to a lack of information about these in the literature): 1. Benefits to large felids: improved physical well-being; improved psychological well-being; improved cognitive stimulation; improved social stimulation; improved behavioural stimulation. 2. Benefits to keepers: improved relationship with large felids; improved movement of large felids (e.g., between enclosures); improved enclosure maintenance; improved job satisfaction; improved care of large felids (e.g., veterinary checks). 3. Benefits to visitors: improved entertainment; improved knowledge of large felids; improved enjoyment; improved connection with large felids; improved information about large felid conservation methods. In order to identify additional perceived consequences of these practices, participants were asked to describe any other benefits or limitations associated with each handling type and training in open-ended questions Procedure Prior to commencing, this research was approved by the University of South Australia Human Research Ethics Committee (project number P111/10), as well as by each of the zoo keeper organisations and zoos involved. Between July 2010 and February 2011, a link to the web-based questionnaire was posted on the six zoo keeper forums and ed to the seven consenting Australian and New Zealand zoos for internal distribution to staff (see Section 4.3.1). Interested participants then accessed and completed the questionnaire online (either during work or personal time; this was not determined). The questionnaire took approximately 30 minutes to complete. The 146

165 online survey system allowed the results to be automatically stored within the system for access by the researcher Statistical analysis Basic descriptives were examined to understand the use of the different handling methods and training, as well as respondents outright opinions of these. To determine which practice (hands-on, protected, and hands-off contact, and training) was viewed as most beneficial for each of the three groups (large felids, keepers, and visitors), ratings for each group were pooled to generate a mean benefit rating, ranging from 1 to 5 (where 1 = strongly disagree and 5 = strongly agree with the benefits to each group). The responses to these statements were not normally distributed, and both log and square root transformations failed to normalise the data. Therefore, these scores were subjected to Friedman s ANOVA analyses (one per group - animals, keepers, visitors), with Wilcoxon Signed-Ranked post hocs (Siegal & Castellan, 1998) in SPSS version 17 (all post hoc comparisons were conducted with a Bonferroni correction applied, corrected α = 0.008). In the absence of previous studies of this nature, prospective power analysis was difficult to calculate. As such, although somewhat controversial (Thomas, 1997), a retrospective power analysis was conducted. This analysis indicated that with the present sample (n = 86), with four levels of withingroups comparisons, sufficient power was achieved to detect small effects (f = 0.13, 1-β = 0.80, α = 0.05) in all statistical analyses. Given the exploratory nature of the open-ended questions regarding participantperceived benefits and limitations of each handling practice, and of training, for the three groups, these results were not subjected to statistical analysis. Instead, these qualitative data were organised into categories of statements (similar to Hosey & Melfi, 2012) by reading the responses and grouping these into 'themes'. The number of statements within each category was tallied to provide an indication of how frequently such comments were made. 147

166 4.3. Results How commonly performed are the three different handling methods and training with large felids? Results revealed that 19.8% (n = 17) of participants practice hands-on contact, 87.2% (n = 75) practice protected contact and 20.9% (n = 18) practice hands-off contact with their large felids. A further 77.9% (n = 67) practice training. These rates were then explored individually for each species, as shown in Table 4.1, with protected contact being practiced by the majority of participants with all species and training being practiced by at least half of the participants with all species. More variation appears to exist regarding hands-on and hands-off contact, with an evident trend for more keepers to report engaging in hands-on contact with cheetahs compared to other species, and more engaging in hands-off contact with leopards and jaguars than the other species. Table 4.1. Self-reported rates of handling and training as a percentage of participants (n = 86) working with each of the seven large felid species. Species (n = number of keepers working Hands-on Protected Hands-off Training with species) Tiger (n = 65) 9.23% 89.23% 18.46% 69.23% Lion (n = 61) 6.56% 81.97% 18.03% 62.30% Cheetah (n = 60) 42.31% 57.69% 11.54% 50.00% Leopard (n = 27) 7.41% 70.37% 29.63% 51.85% Jaguar (n = 12) 0.00% 83.33% 33.33% 50.00% Cougar (n = 24) 16.67% 83.33% 16.67% 50.00% Snow leopard (n = 21) 9.52% 90.48% 14.29% 71.43% Keepers reported spending a mean maximum of 1.94 hours (SD = 1.46; range = 0.33 to 6.0 hours) engaged in hands-on contact per day (with all animals), 1.70 hours (SD = 1.47 in protected contact (range = 0.17 to 8.0 hours), and 1.09 hours (SD =.99) in training (range = 0.17 to 4.0 hours). Some responses were excluded due to not being able to confidently calculate the number of hours (four for protected contact and four for training). 148

167 What are keepers basic opinions of the three different levels of handling and training? Respondents were asked to rate how strongly they agreed that the current handling and training practices used at their institution allow for the combined needs of the animals, keepers and visitors to be met. The majority of participants (74.5%) were in agreement that their institution did meet these needs (54.7% agreed and 19.8% strongly agreed). They were also asked to indicate which level of handling and training they believed is best for the three aforementioned groups. As shown in Table 4.2, there is a clear trend for training to be regarded as the most beneficial practice for all three user groups ( 88%). Protected contact is reported as similarly beneficial for animals and keepers, but less so for visitors. Additionally, few participants ( 12%) reported hands-on and hands-off contact as beneficial for any group, with the exception that over 40% of participants rated hands-off contact as beneficial for zoo visitors. Table 4.2. Percentage of participants (n = 86) who agreed that each type of handling and training is most beneficial for the animals, keepers, and visitors. Hands-on contact Protected contact Hands-off contact Training Animals 12% 80% 8% 99% Keepers 7% 84% 9% 99% Visitors 7% 50% 43% 88% How strongly do keepers agree with the potential benefits of each level of handling and training? To provide a more thorough understanding of keeper opinions of the different handling methods, participants were asked to rate their agreement with various statements about the benefits of each level of handling for each of the three groups (large felids, keepers, and visitors). Mean benefit ratings of each handling practice (hands-on contact, protected contact, hands-off contact, and training) for each of the three groups are presented in Table 4.3. Freidman's ANOVA analyses revealed a significant difference in rating scores of handling practices within each user group: large felids, χ 2 (3) = , p < 0.05; keepers, χ 2 (3) = , p < 0.05; visitors, χ 2 (3) = 98.06, p < Post-hoc analyses are individually discussed for each group herein. 149

168 Table 4.3. Median (IQR) participant-reported benefit ratings of each handling practice (hands-on contact, protected contact, hands-off contact, and training) for each user group (large felids, keepers, and visitors). Handling practice Hands-on Protected Hands-off Training Large felids 3.0 ( ) 4.0 ( ) 2.6 ( ) 4.2 ( ) Keepers 3.0 ( ) 4.0 ( ) 2.2 ( ) 4.6 ( ) Visitors 3.0 (2.2 to 3.7) 3.7 (3.0 to 4.0) 2.6 (2.0 to 3.0) 4.0 (3.6 to 4.5) Note: Benefit rating score range = 1-5 (with a higher score indicating greater agreement that the practice is beneficial) Keeper opinions of handling and training for large felids Participants perceived training to be more beneficial to large felids than hands-on (T = 10, p <.001, r = -.53), protected (T = 13, p <.001, r = -.38), and hands-off contact (T = 4, p <.001, r = -.57). Protected contact was perceived as more beneficial than both hands-on contact (T = 14, p <.001, r = -.42) and hands-off contact (T = 7, p <.001, r = -.52). There was no significant difference between the perceived benefit ratings of hands-on and hands-off contact (P = 0.032) Additional benefits and limitations for large felids. A number of participants provided details of additional benefits and limitations of the three levels of contact and of training for large felids. As seen in Table 4.4, participants reported a number of limitations associated with hands-on and hands-off contact, and more benefits associated with protected contact and training. In particular, the most frequently made comments were that hands-on contact is an unsafe practice (28 comments), followed by protected contact being a safer practice (12 comments). For each method, participants also noted that the benefits/limitations will depend on other factors, such as the individual animal (for example, species, size, and history), the set-up of the facility and housing style, the staff involved, the type of program and reinforcement used, as well as other enrichment/stimulation offered to the animal. 150

169 Table 4.4. Categories of open-ended responses made by participants regarding the benefits and limitations of each handling practice and training for large felids Benefits Limitations Hands-on Better health/medical care (n = 3) Unsafe and leads to animal euthanasia (n = 28) Greater trust/bond with keeper (n = 2) Unnecessary (n = 9) Higher breeding success (n = 1) Influences natural behaviour (n = 7) Improved welfare (n = 1) Removes choice and control (n = 3) Has negative consequences (e.g., punishment; n = 3) Reduces breeding success (n = 2) Removes mother rearing (n = 1) Gives animal wrong impression of humans (n = 1) Loss of dignity for animal (n = 1) Limits animal s exhibit access (n = 1) Protected Safer (n = 12) Potentially reinforces unnatural behaviour (n = 1) Better training and veterinary care (n = 7) More beneficial for keeper than animal (n = 1) Reduces stress and aggression (n = 7) Increases choice and control (n = 7) Provides cognitive stimulation (n = 5) Encourages natural behaviour (n = 2) Enjoyable for animal (n = 1) Decreases boredom (n = 1) Allows for conspecific interaction (n = 1) Gives animal more space (n = 1) Improves overall animal welfare (n = 1) Hands-off Gives animal choice (n = 2) Can lead to negative behaviours (e.g., stereotypy; n = 8) Allows natural animal behaviour (n = 1) Unnecessary (n = 6) Beneficial method for rehabilitation animals (n = 1) Lack of bond between animal and keeper (n = 4) Lack of health care and behavioural monitoring (n = 5) Lack of enrichment (n = 3) Training Better health/medical care (n = 8) Can be a source of stress for animal (n = 1) Stimulating and enriching (n = 6) Could be considered unnatural (n = 1) Allows stronger bond between animal and keeper (n = 4) Potentially reinforces unnatural behaviour (n = 1) Can reduce negative behaviours (e.g., stress; n = 3) Improve physical, mental, emotional well-being (n = 3) Note: n = number of comments made. 151

170 Keeper opinions of handling and training for keepers. Participants perceived training to be more beneficial for keepers than hands-on (T = 5, p <.001, r = -.53), protected (T = 14, p <.001, r = -.37), and hands-off (T = 1, p <.001, r = -.60) contact. Protected contact was perceived as more beneficial than both handson contact (T = 18, p <.001, r = -.39) and hands-off contact (T = 3, p <.001, r = -.57), and hands-on contact was perceived as more beneficial than hands-off contact (T = 20, p <.001, r = -.35) Additional benefits and limitations for keepers. Whilst fewer comments were made about the benefits/limitations of each practice for keepers compared to the animals, participants similarly noted more limitations for hands-on contact than the other methods (see Table 4.5). Additionally, like the openended responses regarding these practices for large felids, the most frequently made comment was that safety is a benefit of protected contact for keepers (12 comments). Participants also noted that the benefits/limitations of any practice for keepers will depend on the qualification/experience of the individual keeper, the species of cat, compliance of the animal, the purpose of the particular practice, the type of reinforcement used, and the quality of the program Keeper opinions of handling and training for visitors. Participants perceived training to be more beneficial to visitors than hands-on (T = 11, p <.001, r = -.51), protected (T = 14, p <.001, r = -.36), and hands-off contact (T = 4, p <.001, r = -.58). Protected contact was perceived as more beneficial than both hands-on contact (T = 23, p <.001, r = -.30) and hands-off contact (T = 10, p <.001, r = -.47), and hands-on contact was perceived as more beneficial than hands-off contact (T = 27, p =.003, r = -.21). 152

171 Table 4.5. Categories of open-ended responses made by participants regarding the benefits and limitations of each handling practice and training for zoo keepers Benefits Limitations Hands-on Improves keeper s knowledge of animal behaviour (n = 1) Unnecessary (n = 5) Stimulating for keeper (n = 1) Dangerous (n = 3) Allows strong connection between keeper and animal (n = 1) Potentially puts pressure on keepers (n = 1) Protected Safer (n = 12) Potentially dangerous (n = 2) Allows keeper to bond with animal (n = 6) Allows keeper to provide better health/medical care (n = 4) Challenging/stimulating for keeper (n = 2) Interesting/engaging for keeper (n = 1) Improves keeper well-being (n = 1) Teaches keeper to communicate with animal (n = 1) Less stressful for keeper (n = 1) Can be practised by most keepers (n = 1) Hands-off Safe (n = 4) May lead to job dissatisfaction/frustration (n = 1) Still able to achieve things with this approach (n = 2) Does not impact job satisfaction (n = 1) Poorer provision of health/medical care (n = 1) Training Strengthened relationship with animal and subsequent positive Unsafe (n = 2) impact on work (n = 5) Rewarding and gives keeper pride in work (n = 3) Some keepers may be unsuitable to practice it (n = 1) Allows keeper to provide better health/medical care (n = 3) More about meeting keeper needs than animal needs (n = 1) Safe (n = 1) Results in happier/better keepers (n = 1) Note: n = number of comments made. 153

172 Additional benefits and limitations for visitors. As shown in Table 4.6, a number of comments were made by participants about the benefits and limitations of the three levels of contact and training for visitors. Of note, the most frequent comment made was that hands-on contact has the potential to send wrong messages to zoo visitors (39 comments); this was the most frequent benefit/limitation comment made by participants regarding any of the practices for any of the three user groups (animals, keepers and visitor). Participants also noted that benefits and limitations depend on the overall educational program of a facility and the way information is communicated to visitors, the individual handler and the training used, whether the handling/training is conducted on- or off-exhibit, how safe the program is, and that no one type has an impact on a visitor and his/her learning. It was also noted that any level of contact should only be practiced by keepers, not visitors. Table 4.6. Categories of open-ended responses made by participants regarding the benefits and limitations of each handling practice and training for zoo visitors Benefits Limitations Hands-on More enjoyable experience (n = 4) Sends wrong message about dangerousness of animal and encourages pet ownership (n = 39) Improves opinion of animals (n = 2) Unnecessary (n = 6) Potential to increase education and conservation support (n = 2) Dangerous (n = 2) Could result in fear if a visitor has a negative experience (n = 1) Protected Sends more accurate message about animal (n = 7) Sends inaccurate message about animals (n = 4) High education value (n = 6) Not as beneficial as hands-on (n = 1) Allows visitor to connect with animal (n = 4) Possible disease transmission from animal to visitor (n = 1) Safer (n = 3) Interesting for visitors (n = 3) More natural (n = 2) May encourage financial support (n = 1) Hands-off Sends more accurate message about animal (n = 6) Often no opportunity for keeper talks (n = 2) Seeing animals in this form should be sufficient for visitors (n = 1) Animal not as interesting (n = 2) No opportunity for connection between visitor and animal (n = 1) Training Sends accurate message (n = 3) Sends inaccurate message (n = 2) Educational (n = 3) May encourage financial support (n = 1) Encourages connection between visitor and animal (n = 1) Note: n = number of comments made. 154

173 4.4. Discussion The findings from this study appear to have provided the first published information regarding the use of, and opinions regarding, different handling and training practices with large felids in zoos. Although protected contact is the most popular handling method reportedly used with large felids among this sample, some keepers reported that they employ a combination of two or more methods, and the majority reported using training. These participants work with a range of large felid taxa, often more than one, but most commonly with tigers and lions. Cheetahs, leopards, cougars and snow leopards are worked with at a moderate level, and fewer keepers reported working with jaguars. Protected contact is used most frequently by these participants with all species, whereas the use of hands-on and hands-off contact varies between species. Notably, over 40% of these keepers who work with cheetahs practice hands-on contact with them, compared to a maximum of 17% for keepers working with the other large felid species; and hands-off contact is used more with leopards and jaguars. This may be attributed to the nature of each individual species, such as the cheetah generally being less confrontational than the lion, for example, and thought to be more easily tamed compared to other large felids (Marker-Kraus, 1997). Similarly, half of all keepers reported that they practice training with cheetahs, leopards, jaguars, and cougars, and more (60-70%) with tigers, lions, and snow leopards. Further, respondents indicated that they spend more time engaged in hands-on and protected contact compared to training. This is not unexpected, since training is usually a more formal procedure, likely to occur for set periods of time (e.g., training an animal to present a body part for inspection might be something that is practiced for 10 minutes per day), whereas contact only may occur more frequently (e.g., in a hands-on exhibit, keepers might sit and pat the animal for an undetermined amount of time and this might occur intermittently throughout the day). Whilst specific to the sample studied, these findings provide a starting point for understanding the current global use of these interactive practices with large felids. Reflecting its high use among this sample, protected contact was perceived as the most beneficial handling method for all three groups (large felids, keepers, and visitors), both when participants were asked outright and in the benefit rating scale. Few participants 155

174 (around 10%) reported hands-on to be most beneficial for all three groups when asked outright; and few believed hands-off to be most beneficial for large felids (8%) and keepers (9%), whereas around 45% believed it to be best for visitors. Although handson contact received a higher rating than hands-off on the rating scale for all three groups, this is somewhat misleading since median scores for hands-on were 3 (out of 5) for each, representing neutral/unsure. Thus, whilst hands-off is not viewed as being beneficial to any group based on the scale responses, it does not necessarily mean that hands-on is perceived any more favourably (based on this information alone). The open-ended questions highlighted potential additional benefits and limitations of each technique. Protected contact received high benefit ratings and primarily positive comments for each of the three groups. Some of these benefits - such as it being considered safe for large felids, providing them with better health care, as well as reducing negative behaviour, increasing choice, allowing animals to bond with keepers, and improving cognitive stimulation - are similar to those found in previous research exploring interactions and bonds between people and captive animals (but not necessarily protected contact per se; Baker, 2004; Hosey & Melfi, 2012). Likewise, for keepers, protected contact was considered safe, allowing them to develop a bond with their animals, and improving the provision of health/medical care; again, consistent with prior research of interactions (Bayne, 2002; Hosey & Melfi, 2012). Protected contact was believed to send accurate messages about animals to zoo visitors and to have a high educational value, also reflective of the quite extensive literature suggesting that visitors enjoy seeing active animals (Margulis et al., 2003) and that interactive programs may successfully increase visitor attitudes, knowledge, and conservation behaviour (discussed in Section 6.1). Few comments were made about limitations of protected contact for any group. These findings strengthen the results from the benefit rating scale, while providing additional insight into the potential positive outcomes of this approach. Responses to the rating scale showed that keepers were somewhat ambivalent about the benefits of hands-on contact. The majority of open-ended comments suggest that participants typically believe this practice is not a positive one, with comments 156

175 reflecting how dangerous hands-on contact can be, both for animals and keepers; one respondent explained,...if a cat does attack a keeper, that animal will often end up shot. This is not fair to the cats as they are acting on natural instincts and should not have been put in the position to be able to attack anyone. Despite many participants raising this issue and the strong sentiment with which it was communicated, none mentioned specific incidents where this occurred. Regardless, large felid attacks do occur in captivity, often with serious consequences (as discussed in Section ). Respondents also believed hands-on contact negatively influences the natural behaviour of large felids, and is unnecessary since protected contact can yield similar benefits (for example, a close keeper-animal bond) without the added risk. This supports Carlstead s (2009) finding that relationships between animals and keepers are more favorable when a keeper interacts through a barrier as opposed to entering the animal s enclosure. Participants most frequently noted that hands-on contact sends the wrong message to visitors, implying that it is safe to interact closely with large felids, thereby potentially encouraging pet ownership; as one participant explained: What the public walks away from is a skewed image of what these cats are and do. They see trainers touch them and then they think that all cats can be touched. They then want one with out (sic) thought to the costs, care and betterment of the species. Of note, most of the respondents work in the USA, where it is legal in many states to keep exotic felines as pets, potentially accounting for the large number of participants expressing this view. These consequences have been previously discussed in the literature (Kreger & Mench, 1995; Nyhus et al., 2003). Despite these concerns, a small number of respondents expressed that hands-on contact allows a strong large felid-keeper bond, assists with medical care, and provides an enjoyable experience for visitors. More research is needed to explore these benefits. While respondents rated hands-off contact as not beneficial for all three groups, the open-ended responses were mixed. For large felids, this approach was considered limiting, leading to negative behaviours (e.g., stereotypy), with keepers being a source of stress for animals under such circumstances. One respondent explained, Having an animal which is managed hands off in captivity and in constant close proximity of keepers/visitors can cause aggressive, defensive and stressed animals. Further, participants highlighted that hands-off contact does not strengthen bonds or improve 157

176 health/medical care. In contrast, despite hands-off being reported as not beneficial for keepers and visitors on the rating scales, the open-ended responses actually raised more benefits than limitations (albeit the overall number of comments was very low compared to the other practices). For keepers, participants discussed it as safe and sending an accurate message to visitors about what large felids are really like. These findings suggest that what may be beneficial for one group is not necessarily so for another, and zoos must be aware of this. Almost all participants agreed that training is beneficial for large felids and keepers, with 88% agreeing that it is for visitors. This agreement was higher than that for any of the three handling practices. Rating scale and open-ended responses substantiated these findings. For large felids, respondents saw training as beneficial to veterinary/medical procedures, and as stimulating and enriching, similar to discussions about such practices with other species (as discussed in Section ). Consistent with findings in primate research (Bayne, 2002; Savastano et al., 2003), participants highlighted that training significantly strengthens the keeper-animal bond, positively impacting a keeper s work. One participant articulated this well: Once you form that bond with the animal you are training, you are inspired to do more for the animal because it is important to you therefore you clean better, you give more enrichment, you get involved in conservation, you learn to enjoy talking to the public about your animals.... Finally, like with protected contact, participants primarily discussed how training sends an accurate message to zoo visitors about the animals and the importance of caring for them, and that it is therefore also educational. Minimal negative comments were made about this practice. These findings add support to the rating scales and provide additional information about the benefits of training. In this study handling and training were treated as separate entities whereas, in reality, training must be accompanied by some level of handling (not always vice versa). The findings allowed these effects to be isolated, revealing that for all three groups (animals, keepers, and visitors) training is viewed as most beneficial. For large felids, this is not surprising, given the potential enrichment value of training (as discussed in Section ), and its structured nature allowing aspects of animal well-being, such as 158

177 cognitive stimulation, to be addressed. In contrast, contact only can be less structured and may not be targeted specifically to improve animal well-being. Respondents highlighted that the benefits and limitations of each practice depend on various other factors. For large felids, individual factors (gender, age, species, history, and personality) were thought to influence how much each animal benefits from these practices. This is consistent with Carlstead (2009), who highlighted that species differences occur in the components that contribute to the nature of interactions between animals and their keepers, and with an animal s affinity to their keeper. Similarly, participants discussed how individual keeper factors will determine how much they benefit, with some suggesting that not all keepers are suited to practice each handling method and/or training. This notion has been explored in a previous Australian study, which investigated the effects of male keeper (n = 7) and Bengal tiger (n = 7) personalities on their respective behaviours in a hands-on facility (Phillips & Peck, 2007). Using the well-known, Revised NEO Personality Inventory (Costa & McCrae, 1992), Phillips and Peck (2007) demonstrated that the keepers' self-reported personalities differed from standardised average personality ratings for adult males in a number of ways - in particular, these keepers were above average in the dimensions of activity, impulsivity, excitement, and positive emotions; yet were less trusting and altruistic than average, and had a lower sense of duty and order. In addition to these differences, Phillips and Peck (2007) explored correlations between keeper personality traits and their interactive behaviours, with some key findings being that those keepers whose personalities were more angry in nature engaged in less patting behaviour with the tigers, those who were more conscientious spent less time playing with the animals, and those who rated higher on neuroticism engaged in less interactions with the animals. Clearly, the generalisability of these findings is limited by the study's small sample size, the inclusion of only male keepers, and the focus on hands-on interactions only. However, this study provides some preliminary support for the idea that zoo keepers who engage in interactive practices might differ personality-wise from other adults, and not only might these differences make them better suited to the nature of their work, but that differences 159

178 between keepers might influence the way that they interact with their animals. This lends support to the finding in the present study, with some participants believing that not all zoo keepers are suited to interactive practices, and that keeper suitability could influence the outcomes of interactions for all involved. Finally, participants noted that the overall educational approach of a facility impacts the experience of zoo visitors. Likewise, prior research has indicated that individual factors - such as prior experience and reasons for visiting the zoo - can also influence how visitors experience and respond to zoos (discussed further in Section 6.1). While direct conclusions cannot always be drawn between a particular zoo practice and its consequences (be they positive, negative or neutral), it is important to consider these variables when interpreting these results and when addressing the experience of animals, keepers and visitors. This study has placed a heavier focus on identifying the benefits of different handling and training practices. However, a number of limitations have also been raised by keepers and, in the interest of promoting optimal well-being for zoo-housed animals, these are of concern. For example, a hands-on approach is thought to influence natural cat behaviour and remove choice and control from the animal. Similarly, a hands-off approach has been thought to lead to negative behaviours and reduced health care. Each of these factors is associated with reduced welfare. While the welfare impacts of zoo living vary between species, and indeed among individual animals, it is widely recognised that various factors in the zoo environment may lead to diminished wellbeing (Basset & Buchanan-Smith, 2007; Morgan & Tromborg, 2007; see also Section 1.1.3), and human interaction cannot be considered an exception. Little is understood about how humans might negatively (or positively) influence large felid welfare, with only a small number of studies offering conflicting findings about the influence of unfamiliar humans (zoo visitors; see Section ), and nothing scientific documented about the influence of familiar humans (zoo keepers). These findings have provided a starting point for future research in this regard. 160

179 Limitations of the research Some limitations of this research must be considered. Firstly, although sufficient for detecting small effects, the sample size of the study is not particularly large (n = 86). While it is not possible to determine the potential population of zoo keepers working with large felids world wide (see Section 4.2.1), it is likely that a sample of this size has limited generalisability on an international scale. Nonetheless, efforts were made to capture as many participants as possible through using an online medium (see Section 4.2.3). Stemming from this, the sample size meant that responses could not be statistically compared for individual regions and this may have skewed some of the data. For example, in some states of the USA it is legal to keep exotic animals as pets. Given that half of the respondents in this sample were working in the USA, this could account for the large number of participants who were concerned about the message that interactive practices send to zoo visitors, particularly since these comments frequently related to exotic pet ownership. In order to maximise best practice world-wide, more indepth investigations of keepers employed in different regions would allow a more comprehensive understanding of the issues that each region is facing and thus ensure that zoo practice is tailored specifically to that particular area. In the same way, the small sample size meant that the analyses could not control for the type of handling practiced by each participant. This is problematic, since it is intuitive that the type of interactive experiences keepers have will influence their opinions of the different practices and this could not be accounted for. In this study, participants reported practicing multiple levels of handling with different species (for example, one might have reported protected contact with tigers and lions, but hands-on contact with cheetahs). Coupled with the small sample size, this resulted in it not being feasible to draw out these effects. This could be achieved in a larger scale study. Secondly, although the open-ended questions sought keepers opinions of both the benefits and limitations of the different practices, the rating scale questions only assessed benefits. Whilst participants could obviously disagree with these potential benefits, a more robust understanding of practice could have come from including rating scale questions about limitations. This was not done, since the literature about interactive practices with other species (which was used to form the basis of this study) focused almost exclusively on benefits. Additionally, the rating scale items were all 161

180 positively worded. This was done given the brief nature of these items (five per user group) and negatively worded items would be used in larger inventory (e.g., see Section 4.4.2) to reduce response bias. Despite these aspects of the scale items, participants had the opportunity to explore limitations these practices in the open-ended responses. Indeed, a number of potentially negative consequences were raised to complement the scale responses, with the two most frequently reported responses (by far) being the potential for wrong messages to be sent to zoo visitors and the potential for danger associated with hands-on interactive practices. As such, these possible weaknesses of the scale items have not hindered participants' opportunity to express positive and negative opinions. Finally, the subjective nature of the study (exploring perceptions, not actual outcomes) means that conclusions cannot be drawn regarding the actual benefits and limitations of these handling and training practices. However, in the absence of information about the consequences of interactions for any of the parties (animal or human) involved, the purpose of this study was to seek zoo keeper opinions of their current interactive practices. In doing so, this study has provided a useful framework for measuring keeper opinions, both quantitatively (through the benefit rating scales) and qualitatively (through open-ended questions). Conducting objective explorations to substantiate these initial results will be an important next step Suggestions for future zoo keeper research regarding animal handling and training With one of the overarching aims of this study being to provide a foundation for future research regarding the potential benefits and limitations of interacting with large felids in zoos, the findings from the present study can be used to understand the types of questions we should be asking zoo keepers in the future. A sample questionnaire, generated through combining the suggested benefits and limitations of each handling practice into a list of positively and negatively worded rating scale items, can be found in Appendix H. While the psychometric properties of such an instrument would need to be established prior to use, this sample inventory provides a clear example of how the open-ended findings from the present study could be substantiated to further our 162

181 understanding of the benefits and limitations of interacting with large felids in zoos. In this way, the sample questionnaire is not limited to large felids and could be tested with other species. Indeed, it could be used to test keeper opinions of practices between species, to further our understanding of which species might be better suited to different types of interactive practices. The use of such a questionnaire would ideally be coupled with more objective measures of benefits and limitations before a complete picture is provided Conclusion The participants in this study have provided a significant new understanding about the possible consequences of different handling practices and training with large felids. While these findings are not particularly surprising, given that the two most beneficial practices (protected contact handling and training) are engaged in most frequently by these participants, these findings have provided a basis upon which future, objective research can build. As a research area in its infancy, there is a clear need for the benefits and limitations raised here to be substantiated. As has been done for primates, behavioural and physiological data is required to further understand whether the influence of protected contact and/or training on large felids is positive, negative, or neutral. And, while the sample size was sufficient for this initial investigation and for producing small effects, a larger number of respondents, with a more even spread using each practice, could provide a better understanding through more robust results. Given that this survey relied on self-report measures, some objective methods to investigate the benefits and limitations to keepers would add weight to these findings. Finally, independent measurement of the impact of these different methods on zoo visitors is also required, and could be achieved through observational and survey research of visitors who encounter these different practices. Overall, zoo keepers have provided valuable information which will assist in directing and complementing future research on this somewhat controversial topic. With these perspectives in mind, the following chapter now explores how large felids respond (behaviourally) to interacting with people during behind-the-scenes tours at one Australian zoological organisation. 163

182 5 The short-term behaviour of large felids during behind-the-scenes zoo visitor tours: A case study 14 The presence of people at zoo exhibits may be perceived by nonhuman animals as either a negative influence, a form of enrichment, or simply a changing variable that has no effect. G. Davey (2007, p. 169) The previous chapter explored current rates of handling and training engaged in by zoo keepers with their large felids, revealing that protected contact is the most common handling technique and that training is practiced more often than not among the sample. These interactions make possible a range of interactive experiences for zoo visitors, often based on the idea that having an opportunity to interact with an animal may enhance the visitor experience and, subsequently, zoo conservation initiatives. However, the influence of such experiences on the animals involved appears to have been neglected in the literature to date. As such, this chapter presents the findings of a case study which explored the behaviour of large felids during behind-the-scenes zoo visitor tours at one Australian zoological organisation Introduction The primary goal of modern zoos is to protect animals from extinction. This is addressed not only through captive breeding programs, research, and in-situ 14 This chapter has been published as: Szokalski, M.S., Foster, W.K., & Litchfield, C.A. (2013). Behavioral monitoring of big cats involved in behind-the-scenes zoo visitor tours. International Journal of Comparative Psychology, 26, See Appendix I. 164

183 conservation work but also through the education and engagement of the visiting public. In order to combine the needs of animals (optimal well-being) and visitors (primarily, recreation and education), as well as zoo staff, recent decades have seen a shift in the way animals and visitors interact and encounter each other. Today s exhibits are becoming increasingly naturalistic (Fernandez et al., 2009), aiming to address the psychological and physical needs of the animals, whilst providing visitors with a more realistic experience. Further breaking the barrier to traditional exhibit viewing, zoo visitors are being offered various interactive experiences including behind-the-scenes tours of off-exhibit zoo facilities, entering animal enclosures, touching animals, and having photographs taken with them. Although there is currently a paucity of information about the influence of these interactive animal experiences on visitors, research tends to suggest that the education and conservation value of experiences that allow visitors to do more than simply view an animal s exhibit (such as witnessing keeper talks) may be high (for discussion, see Section 6.1). However, the influence of these experiences upon the animals is rarely considered in the literature (see Section 3.2). In order to support the continued involvement of animals in these programs, and to make suggestions about how such programs might be altered, it is vital to understand how participation might influence animal well-being. Of particular interest is the use of larger, potentially more dangerous animals in these encounters: namely, large felids. At present, zoos world-wide offer various large felid encounters for zoo visitors and the opportunity for people to engage in these appears to be increasing. Across the Australian zoo network, there are currently (as of July 2014) 23 such tours on offer to zoo visitors, with 14 of these being protected contact and nine hands-on in nature (Table 2.1). What makes this particularly controversial is that these animals rarely have direct physical contact with humans in the wild, and many of the recorded interactions (portrayed in both the scientific literature and the media) document injuries and fatalities sustained by humans as a result of attacks from these animals (see Sections and ). Further, these species are typically solitary, with the exception of lions and, to a lesser extent, some male cheetahs (Macdonald, Loveridge, & Nowell, 2010). Therefore, regular contact and interaction with zoo visitors could result in negative repercussions for these animals. 165

184 Currently, information regarding the influence of visitors on zoo-housed animals has focused almost exclusively on visitors at exhibits, as opposed to interactive tours (see Section 3.3). As discussed in Section , visitors could be considered a positive (e.g., enriching), negative or neutral influence on zoo animals. To date, however, the majority of the visitor-impact research has focused on primates (for an overview of areas addressed and methodologies used in these studies, see Appendix C; for reviews of this literature, see Claxton, 2011; Hosey, 2008). Given the obvious differences between these species and large felids, we cannot necessarily generalise this information to the latter, as other authors agree (Davey, 2007). Additionally, in reviewing research on the effects of unfamiliar humans on different taxonomic groups, Hosey (2008) concluded that, compared to primates, felids are much less susceptible (behaviourally) to the presence of zoo visitors. Only five studies appear to have investigated the influence of visitors in relation to large felids, through behaviour observations, and the results of such studies are conflicting. As detailed in Section , two of these investigations have shown no variation in cheetah, lion, leopard, snow leopard and tiger behaviour in response to visitor presence or absence (Margulis et al., 2003; O'Donovan et al., 1993). In contrast, others have shown variation in the behaviour of leopards, jaguars and puma in association with visitors (Mallapur & Chellum, 2002; Sellinger & Ha, 2004; Maia et al., 2012). These changes typically relate to alterations in levels of activity and inactivity, rather than pacing or other stereotypic/adversive behaviours. However, this does not necessarily mean that visitors are not having an adverse impact on the welfare of these animals, since studies to date have been limited to behavioural observations, with a lack of supporting physiological data. Moreover, with these studies exploring the influence of visitors at exhibits, they only offer limited insight into the potential influence of tours given the differences between these two visitor variables. For example, while visitors at exhibits might be more unpredictable (i.e., in terms of stay time, density, noise, and behaviours), visitors on tours might be more predictable. In particular, tours are likely to be under zoo keeper/staff control, occurring at set times of the day, on certain days of the week, and for set durations, with maximum numbers of visitors at any one time (see Table 2.1 summary of tour details in Australian zoos). Keepers/staff are also likely to exert more control over the behaviours visitors engage in during tours compared to at 166

185 exhibits. These differences in visitor predictability during tours may mean that behavioural responses of large felids to these experiences (if they respond at all) might differ from how they respond to visitors at exhibits and that the small body of literature to date is insufficient to guide our understanding of the influence of visitor tours on large felid behaviour. In addition to the dearth of explorations specifically focused on the influence of interactive tours on the behaviour of these species, very little is known about the potential visitor effect during close encounters involving any species, with aquariumhoused dolphins participating in swim-with-dolphin programs being the primary focus of such research in the peer-reviewed literature (Brensing & Linke, 2003; Kyngdon et al., 2003; Samuels & Spradlin, 1995; Trone et al., 2005). Overall, these studies have concluded that the behaviour of these animals is not being adversely influenced by their participation in such experiences, both in the short- and long-term. Since dolphins are social animals, these findings cannot be generalised to other, less social species like large felids. What these studies do highlight, however, is that behavioural analysis of animals in such programs can be an efficient and effective method for measuring animal welfare, just as it has been for large felids in both enrichment-related (see Table 1.3) and visitor-impact (see Table 3.2) studies Aims of the Current Study At present, large felid tours in Australian zoological organisations are taking the form of either protected or hands-on contact (Table 2.1). As outlined in Chapter Three, Zoos SA (including Adelaide Zoo and Monarto Zoo) is one Australian zoological organisation currently offering large felid encounters to their visitors, including both a protected contact tour of the tiger and lion facilities at a city zoo (see Section 3.4.1), and a handson cheetah tour at an open-range zoo (see Section 3.4.2). It therefore serves as an ideal place in which to begin to explore the potential impact of such practices on the animals involved. 167

186 This case study sought to empirically examine the behaviour of three felid species (tiger, lion, and cheetah) during these behind-the-scenes zoo visitor tours, in order to determine whether any behavioural changes were exhibited by the animals that may indicate altered welfare. Given the lack of previous explorations of this nature, and the conflicting results of visitor impact studies with large felids to date, this case study was an exploratory one and no specific hypotheses were tested. Instead, the study sought to document changes in behavioural patterns. Consistent with previous studies of large felid behaviour in response to zoo visitors (see Section ) and to enrichment (see Section ), key behavioural indicators were species-typical behaviours and pacing. In particular, species-typical behaviours (both active and inactive) were used as indicators of positive well-being, since the expression of these behaviours might indicate that animals are not being adversely affected by the tour to the extent that they are not prevented from behaving in ways that are typical of the species. Pacing, a potential stereotypic behaviour, indicated the possibility of stress or compromised wellbeing. Aggression was also explored; although this is a natural behaviour, its presence if directed at humans during tours may also indicate a negative, or stress, response. In addition, the study focused on documenting the nature of interactions between these animals and humans (including both keepers and visitors), since little is understood about how large felids respond to people and the behaviours they engage in during interactive tours. As an indirect measure of interaction, estimated physical proximities of large felids to humans (keeper/s and visitor/s) during tours were investigated. Again, specific predictions about proximities were not made Method This study received ethics approved from the Institute of Medical and Veterinary Sciences (South Australia; project number 29/11) Participants The participants involved were three separate groups of adult large felids currently participating in interactive tours: one Sumatran tiger 15 (Panthera tigris sumatrae) and 15 Although there are three tigers at Adelaide Zoo (see Section 3.4.1), only the male was observed due to the two females alternating between being housed on-display and off-display for the tours, limiting data consistency 168

187 three African lions (Panthera leo leo) involved in the protected contact tour at Adelaide Zoo 16, and three African cheetahs (Acinonyx jubatus) involved in the hands-on tour at Monarto Zoo (see Table 5.1 for individual details). The tiger and lions had been involved in the protected contact tour for approximately the last four years, whilst the cheetahs had been participating in hands-on visitor experiences since they were cubs (eight years) Participant Housing and Feeding Routines Adelaide Zoo As outlined in Section 4.2.1, all tigers at Adelaide Zoo are housed solitarily and the individual of interest is typically housed in the same on-exhibit enclosure on a daily basis 17. Overnight, this tiger is given access to an off-exhibit enclosure and is then put on-exhibit in the mornings the timing of this varies depending on enclosure maintenance, but is usually around 8.15am on tour days and before zoo opening hours (9.30am) on non-tour days. The on-exhibit enclosure is approximately 225m 2 and contains a moat, a pool, and various naturalistic plants and substrate. There are a number of shaded and retreat locations, as well as logs and rocks for elevation. 16 Similarly, although data collection also involved the lions at Monarto Zoo (see Section 3.4.2) initially, this was unfortunately ceased due to changing patterns in the display of the on-exhibit animals. Husbandry practices at this time meant that the animals were housed on-exhibit in varying combinations and not in a regular pattern. Moreover, a cub had recently been born and was in the process of being introduced to the other lions, which added further changes to the usual routine of the animals. Whilst these practices are optimal for the animals, it would have resulted in very small amounts of data for each individual lion during the data collection time frame, as well as additional confounding variables, which might have influenced the results. For example, although no data analysis has been conducted, preliminary observations suggested that there was individual variation in behaviour before the tours depending on the combination of animals housed on-exhibit at a given time. 17 Occasionally, one of the two female tigers is given access to this enclosure and the male is either housed off-exhibit or in another on-exhibit enclosure. 169

188 Table 5.1. Key background details of participants (n = 7) included in this study. Species Names (Gender) D.O.B Place of birth Sumatran tiger Tuan (M) Lisbon Zoo, Portugal African lion Mujambi (M) Mogo Zoo, Australia Yizi (F) & Auckland Amani (F) a Zoo, New Zealand African Skukuza (M), Monarto Zoo, cheetah Tsotsie (M) & Australia Askari (M) b Note: a = lionesses born in same litter; b = cheetah born in same litter. Approximate age of arrival at Zoos SA 8.5yrs 2yrs 1yr N/A Rearing history Mother Mother Mother Human The lion facility at Adelaide Zoo is comprised of four adjacent enclosures (three are ~35 m 2 each in size and the other ~170 m 2 ). Fences connecting the enclosures can be opened to provide the animals with access to more than one enclosure at a time; usually, the fences between the three smaller enclosures remain open (totalling ~100 m 2 ). Each enclosure contains mostly dirt-based flooring, with various trees and/or small plants in each, and raised platforms or rocks for elevation. The larger exhibit also contains a raised, grassy mound and a large tree for climbing. Since the male lion experiences a medical condition, which has made him prone to seizures in the past and therefore vulnerable around other animals, he is housed separately from the two females. The animals are maintained in the on-exhibit enclosures throughout the day and night. During exhibit maintenance on non-tour day morning s (usually between 9.00am am), they are switched between the smaller and larger enclosures (usually giving them two days in each exhibit at a time). This typically requires the male lion to be kept in a lock-away area whilst the females are shifted, which results in him being off-exhibit during these times. In order to replicate feeding patterns in natural habitats (see Section ), where large felids may not feed daily, the tiger and lions are typically fed on Tuesdays, Thursdays, Saturdays, Sundays and public holidays. Whilst the proportion of food provided to each animal varies according to their individual weights, horse meat forms a major part of their diet. They are also fed chicken quite regularly, with kangaroo, rabbit, 170

189 turkey and goat provided when available. Meat is usually given as partial carcasses, with Big Cat supplement added. These animals are usually fed in their on-exhibit enclosures, at around 2.30pm for the tiger and 2.00pm for the lions. During the tours, these animals are also fed a small portion of meat (e.g., a turkey neck), as part of the experience involves visitors having the opportunity to feed an animal (either using their hands or a pair of tongs to place the meat through the mesh fencing) Monarto Zoo As outlined in Section 3.4.2, this zoo currently houses two groups of male display cheetahs (as well as breeding females off-exhibit), however only one of these is involved in visitor tours. The on-exhibit facilities are comprised of three adjacent exhibits. The main exhibit (which has a visitor viewing platform attached) is where the three subjects are housed each day, as well as most nights of the week. This exhibit is approximately 4500 m 2 and contains various naturalistic plants and substrate, providing various shaded locations for the animals. It also includes a small hut filled with hay, a low platform, a large tree log and raised platform for climbing, as well as a small drinking pond. Several nights per week, the animals are housed in one of the other onexhibit enclosures, which contains similar features as their regular exhibit, but is roughly seven times larger. When this occurs, the three cheetahs are then shifted to their regular exhibit in the morning, usually before zoo opening hours (9.30am). Due to these animals being hand-raised 18, they have regular physical contact with their keepers, who enter the enclosure for routine maintenance on a daily basis. Usually this occurs before 10.00am on both tour and non-tour days. A naturalistic feeding pattern is also adopted at this zoo, whereby the cheetahs are usually only fed five days per week (with Monday and Thursday being non-feed days). Their diet includes portions of beef, horse, kangaroo and rabbit. Typically, each 18 To the best of this researcher's knowledge, the cheetahs were not hand-reared for the primary purpose of 'facilitating' their inclusion into behind-the-scenes tours/interactive programs. Their mother became ill with an infection shortly after the cubs' birth and so a decision was made by the carnivore team to remove the cubs from their mother to prevent the infection spreading to the cubs. It was intended that the cubs would be returned once the infection cleared, however when this was attempted the cubs were rejected by the mother. As such, the cubs continued to be hand-raised and the potential for their involvement in the tours was later realised. 171

190 individual is fed kg of meat each (including bone), but occasionally they will be given a kg kangaroo carcass to consume together. Keepers like to avoid a set feeding time for these animals and, whilst they are often fed at around 3.00pm during a keeper talk, feeding can occur at any time of day. When tours are conducted on feeding days, these animals are usually given their daily feed during the tour Tour Details Protected contact tour at Adelaide Zoo This tour is conducted up to three mornings (Tuesday, Thursday and Sunday) per week. It starts prior to the zoo opening and runs for approximately one hour (8.45am am), during which up to six visitors (minimum age 12 years) are toured first through the tiger on- and off-exhibit facilities, and then through the lion on- and offexhibit facilities. When visiting the tiger facilities, visitors view the animals from outside of the enclosures. Visitors view the tiger of interest housed in his regular onexhibit enclosure, where he is also located before and after the tour. When visiting the lion facilities, one of the smaller lion exhibits is vacated of animals and visitors enter this enclosure, so that they can view the male lion on one side and the female lions on the other in their regular on-exhibit enclosures, where they are also housed before and after the tour. The tour group generally remains with each individual tiger for 15 min, and with the lions combined for another 15 min. Visitors are introduced to each of the individual animals, learn about their life at Adelaide Zoo, have a chance to feed them through mesh fencing, and take photographs of them. To maintain visitor safety, visitors are asked by the keeper to remain approximately 1m away from the exhibit fencing at all times. The only exception to this is when feeding an animal, during which time the keeper approaches the fence with the visitor and the visitor places a piece of meat through the exhibit fencing. Visitors are instructed not to place their hand through the fencing, to let go of the food once the animal has taken it, and to step back once they have done this. One keeper and one zoo volunteer are always present throughout the tour. 172

191 Hands-on tour at Monarto Zoo This tour is conducted up to three mornings (Monday, Thursday and Saturday) per week, for approximately one hour (10.00am am). During this encounter, up to four visitors (minimum age 16 years) are taken into the cheetah exhibit with the animals present (with no protective barrier between them), where they meet the individual cheetahs, learn about their hand-rearing situation, and witness some positivereinforcement training and animal feeding (on feeding days only). They are also able to pat the animals and have photographs taken with them. The cheetahs remain in this exhibit before and after the tour. Being a hands-on tour, a number of measures have been established to maintain visitor safety and these are explained to visitors before they enter the cheetah enclosure. They are instructed to remain standing at all times, and not to crouch down to the animal s level. They are also instructed not to approach the animals unless asked to do so by a keeper, and are told to remain still and to refrain from touching a cheetah if one approaches them. When a keeper does allow visitors to pat the animals, they are instructed not to touch the cheetahs faces, nor to look the animals directly in the eyes. To promote safe interactions, the cheetah are positively reinforced (with a combination of verbal praise, patting, and various food pastes) when they are in close proximity to visitors. Moreover, two keepers are always present on this tour: one to conduct the tour and the other to monitor activities of the animals and visitors (e.g., to ensure that visitors are not leaving the tour group). As a precautionary measure, keepers also carry spray bottles filled with water, which may be used to deter any potentially dangerous animal behaviour (anecdotal accounts from keepers indicate that this has never needed to be used during a visitor tour) Behaviour Observations Behaviour was coded according to an ethogram containing categories of felid behaviour; definitions were adapted from past felid research (Skibiel et al., 2007; Umapathy et al., 2007; Van Metter et al., 2008; Wielebnowski & Brown, 1998). These definitions were further refined based on behaviour observed during preliminary 173

192 observations, prior to formal data collection commencing (these observations were not included in the analysis). Individual behaviours were then condensed into behaviour categories for data analysis (Table 5.2). Estimated proximity between animals and people (keepers and zoo visitors), was recorded by visual inspection and divided into three categories: close (< 2 m), moderate (2-5 m), and distant (> 5 m). Any other occurrences and possible extraneous variables, such as the presence of neighbouring conspecifics, or keepers at exhibits (outside of tour times), were also noted informally on the data record sheets to assist with data interpretation. Table 5.2. The nine condensed behaviour categories (with details of included behaviours) used when conducting animal behavioural observations during this study. Category Inactive Active speciestypical Feeding Conspecific interaction Human interaction Pacing Aggression Out-of-sight Other Included behaviours Sit, stand, or lay whilst not engaged in any other behaviour (animal may be vigilant or not vigilant during any of these behaviours) Walk, climb, manipulate object, scratch ground/grass, crouch, sniff self, scratch self, groom self, vocalise, or any olfactory behaviour (e.g., spray, flehmen) Any consumption behaviour (e.g., eating, dragging food) Any direct affiliative interactive behaviour with a conspecific, including head rub, sniff, paw at, play, present or other sexual behaviours, groom (behaviours may be either direct or through fence, and may or may not include affiliative vocalisations) Any direct affiliative interactive behaviour with a human (keeper and/or visitor), including taking meat from (including other behaviours, such as stretching up to take food, or vocalising), being patted (whilst engaged in any other behaviour, such as sit, stand or eat; may include being rewarded), engage in training (such as sit on command), or routine health checks (such as being weighed on scales) Repetitive walk or trot around enclosure without an apparent goal (displaying at least two repetitions of the same pattern of movement; may or may not include vocalisation) Banging on enclosure doors/fencing, striking at another animal (may or may not include aggressive vocalisations, such as growling or hissing) Any instance where the animal/its behaviour is not visible to observer Any behaviour not listed above (e.g., defecate, urinate, drink, regurgitate, or engage with specific enrichment device) Procedure The study used an observational design, with data (animal behaviour and proximity) being collected at 30s intervals according to instantaneous scan sampling methods (Martin & Bateson, 2007). Although this measure of behaviour can be limited by its inability to capture occurrences and durations of all behaviours, the shortest possible sampling interval (based on the number of animals observed at any given time, and 174

193 adjusted during preliminary observations) was chosen to maximise the behavioural information captured. Where multiple animals were observed at one time (i.e., the lion and cheetah exhibits), these animals were always observed in the same order at each interval. All observations occurred when animals were in their on-exhibit enclosures. Observations were usually collected for one hour before each tour (with the exception that only half an hour of pre-tour data was collected for the tiger, due to husbandry needs and the tour commencing close to the start of the work day), the duration of the tour (approximately 15 min for the tiger, 15 min for the lions, and one hour for the cheetahs), and the hour after the tour. As such, observations typically spanned from 8.15am 10am for the tiger, 8.30am 10.45am for the lions, and 9:00am 12:00pm for the cheetah. Behaviour was then observed for the same time periods on non-tour days (Mondays, Wednesdays and Fridays for the tiger and lions; Tuesdays, Wednesdays and Fridays for the cheetah). The data collection period spanned from October to December 2011; typically, one tour and one non-tour day were observed per animal group per week over this three month period 19. This allowed for 12 tour and 12 non-tour days of data for each individual animal, with the exception that only 10 tour days were collected for the tiger due to not enough tours being conducted during the study period and changes to the on-exhibit management of this animal toward the end of the study. This researcher conducted all behaviour observations. To test the reliability of the ethogram and proximity estimates, an inter-rater assessment was conducted by a second, independent observer. Mean agreement with this second observer was 95% for behaviour and 91% for proximity, achieved in one hour of simultaneous data collection for each species. 19 Although this was the aim, it was not always possible to conduct one tour and one non-tour day each week for each group, due to factors such as heavy rain and tours not running due to no visitor bookings. 175

194 Analysis To determine proportions of behaviours performed by each animal, counts were totalled for each individual animal within each condition (pre, during and post sessions on both non-tour and tour days). A mean value for the percentage of scans that each individual animal spent engaged in each behaviour category was then generated. Where an individual was off-exhibit for any part of an observation period, these scans were excluded from the total number of scans for that period, such that behaviours were only calculated as proportions of on-exhibit scans. The mean number of on-exhibit scans used in data analysis for each animal (per condition) are displayed in Table 5.3. Due to these differences in number of scans within each individual across conditions (particularly for the lions and tiger), caution should be taken when interpreting the results for the within-day (non-tour and tour) comparisons. Table 5.3. Mean number of on-exhibit observation scans used in data analysis for each individual animal per observation condition (pre, during, post) on nontour and tour days. Non-tour days Tour days Pre During Post Pre During Post Lions Mujambi Yizi Amani Cheetah Skukuza Tsotsie Askari Tiger Tuan N/A N/A N/A Note: Means calculated from number of scans on 12 non-tour and 12 tour days for each lion and cheetah; and 10 tour days for the tiger. Data were then pooled for each species prior to statistical analysis. Visual inspection of means for individual animals for the three main behaviour categories (active speciestypical, inactive-species typical, and pacing) confirmed that pooling of data in this way was appropriate (for graphically presented means, see Appendix J). Given the serial correlation across repeated measures data collected for each individual animal, linear mixed models analyses were conducted to investigate behavioural differences across conditions as a function of the independent variables (day (non-tour/tour), and time (pre/during/post)) for the lion (n = 3) and cheetah groups (n = 3). To appropriately 176

195 account for within and between-groups effects, models specified subject ID as a random effect (van Dongen et al., 2004). Separate linear mixed models analyses were conducted for the following behaviour categories: active species typical, inactive species typical, and pacing. Where appropriate, post-hoc pairwise comparisons were conducted. While there are problems inherent in utilising parametric analyses with small sample sizes, the use of this test was chosen due to its ability to account for serial correlation (which is not possible with non-parametric analyses). The tiger's data has been excluded from all statistical analyses for two reasons: firstly, this animal was often off-exhibit (due to husbandry and enclosure maintenance) until later in the morning on most non-tour days, meaning that the data collection period was typically briefer on these days 20 compared to tour days. It was thought that entering the exhibit at a later time on non-tour days would likely influence the behaviours engaged in during the different time blocks. Secondly, it was deemed inappropriate to conduct a linear mixed models analysis with only one participant. Instead, the tiger's tour day data has been presented graphically and visual trends are reported for the behaviours active species-typical, inactive species-typical, and pacing. The mean percentage of scans spent engaged in all other behaviour categories (feeding, aggression, interact conspecific, interact human, other, and out-of-sight) was either too low or non-existent across the different conditions (pre, during, and post on non-tour and tour days) for all animals, rendering it not feasible to conduct linear mixed models analyses on these differences. Moreover, because comparing means with such low frequency data is problematic, the total number of observation sessions during which each of these behaviour categories were observed are presented instead. Linear mixed models analyses were also conducted to explore pooled estimated proximity between animals and humans (keepers and visitors) in the during tour 20 Mean number of scans for each session on non-tour days (based on 10 non-tour day observations): pre = 40 (range = 0-68 scans), during = 22 (range = 0-24 scans), and post = 118 (range = scans) 177

196 sessions on tour days for each species (again, visual inspection of means for individual animal proximities confirmed that pooling of data was appropriate; see Appendix J). Fixed factors entered into the model were distance (close: <2m/moderate: 2m- 5m/distant: >5m), and humans (keeper/visitor), with subject ID specified as a random effect. Where appropriate, post-hoc pairwise comparisons were conducted. These analyses were conducted separately for lions (n = 3) and cheetahs (n = 3). Again, the tiger's data was deemed inappropriate for inclusion in this analysis, but visual trends are reported. All behavioural and proximity analyses were based on 12 non-tour and 12 tour days for each individual lion and cheetah, and for 10 tour days for the tiger. All statistical analyses were conducted in SPSS version 20 and were considered significant at the p < 0.05 level Results The effects of tours on active species-typical, inactive speciestypical, and pacing behaviour As shown in Table 5.4, there were main effects of day (tour, non-tour) on proportions of observations of inactive and pacing behaviour, and of time (pre, during, post) on activespecies typical, inactive species-typical, and pacing behaviour, for lions. There were also mixed effects of day x time on proportions of all three behaviours for this group of animals. For cheetahs, there were main effects of day on proportions of active speciestypical and inactive species-typical behaviours, and main effects of time on proportions of all three behaviour categories. There were also mixed effects of day x time on active species typical and inactive behaviour only. Post-hoc analyses are described individually for each behaviour category. 178

197 Table 5.4. Results of main and mixed effects of linear mixed models analysis for active species typical, inactive species typical, and pacing behaviours for lions and cheetahs. Active species-typical Inactive species-typical Pacing df F df f df f Lions Day (tour, non-tour) 1, , ** 1, ** Time (pre, during, post) 2, * 2, ** 2, ** Day * Time 2, * 2, ** 2, ** Cheetahs Day (tour, non-tour) 1, * 1, * 1, Note: * = p<0.05; ** = p<0.01 Time (pre, during, post) 2, ** 2, ** 2, ** Day * Time 2, ** 2, ** 2,

198 Pre During Post Pre During** Post** Pre During Post Mean % of scans Changes in active species-typical behaviour Mean proportion of scans observed in active species-typical behaviour remained below 20% for all species in any condition. Differences between conditions and significant post hoc comparisons for proportions of this behaviour are shown for each species in Figure 5.1. For lions, across time comparisons indicate that there were significant differences in active species-typical behaviour on both non-tour and tour days, with greater proportions of this behaviour observed in the pre session compared to post on non-tour days, and greater proportions in the during session compared to both pre and post on tour days. For cheetahs, there were differences between days, such that greater proportions of active scans were observed in the during and post sessions on tour days compared to non-tour days. Across times, greater observations of this behaviour were made in the pre session compared to during and post on non-tour days, and greater in the pre and during session compared to post on non-tour days, for cheetahs. For the tiger, there was a trend within tour days for greater scans of active species-typical behaviour in the pre and post session compared to during. 100 pre>post* during>pre*, post** pre>during**, post** pre**, during**>post Non-tour Tour 20 0 Lions Cheetahs Tiger Figure 5.1. Mean percentage of scans in which the lions, cheetahs and tiger were observed in active species-typical behaviour on non-tour and tour days, in all time sessions (pre, during, and post) [Note: Bars indicate standard deviation; significant between day differences are marked on the x-axis labels, and significant within day differences are noted above the bars, where * = p<.05 and ** = p<.001] 180

199 Changes in inactive species-typical behaviour Proportions of inactive species-typical behaviour ranged from 22% - 87% of scans for all species in any condition. Differences between conditions and significant post hoc comparisons for proportions of this behaviour are shown in Figure 5.2. For lions, between day comparisons revealed greater proportions of inactive species-typical behaviour in the during session on non-tour days compared to tour days. Across times, these animals showed differences on tour days only, such that they were more active in the pre and post session compared to during, and more active in the post compared to pre session. Cheetahs displayed more inactive species-typical behaviour on tour days compared to non-tour days in the pre session, but more in the during and post session on non-tour days compared to tour days. Across times, these animals showed more inactive in the during and post sessions compared to pre on non-tour days, but more in the post session compared to pre and during on tour days. For the tiger, there was a trend for less inactive species-typical behaviour in during session compared to pre and post on tour days Changes in pacing behaviour Proportions of pacing behaviour ranged from zero to 42% of scans for all species in any condition. Differences between conditions and significant post hoc comparisons for proportions of this behaviour are shown for each species in Figure 5.3. For lions, between day comparisons indicate greater proprotions of scans observed pacing on tour days in the pre and during sessions, but greater in the post session on non-tour days. Across time, differences were only found for tour days, with greater scans in the during tour session. For cheetahs, differences were only found across times, such that there was more pacing in the pre session compared to during and post on both non-tour and tour days. The tiger showed a trend for increased pacing from the pre (~5%) to post (~15%) sessions on tour days. 181

200 Mean % of scans post>pre**, during**; pre>during** during**,post**>pre post>pre**, during* Non-tour Tour 20 0 Pre During** Post Pre* During** Post* Pre During Post Lions Cheetahs Tiger Figure 5.2. Mean percentage of scans in which lions, cheetahs and tiger were observed in inactive species typical behaviour on non-tour and tour days, in all time sessions (pre, during, and post). [Note: Bars indicate standard deviation; significant between day differences are marked on the x-axis labels, and significant within day differences are noted above the bars, where * = p<.05 and ** = p<.001] 182

201 Mean % of scans 100 during>pre**, post**; pre>post** pre>during**, post** pre>during**, post** Non-tour Tour 20 0 Pre** During** Post* Pre During Post Pre During Post Lions Cheetahs Tiger Figure 5.3. Mean percentage of scans in which lions, cheetahs and tiger were observed in pacing behaviour on non-tour and tour days, in all time sessions (pre, during, and post). Bars indicate standard deviation. [Note: Bars indicate standard deviation; significant between day differences are marked on the x- axis labels, and significant within day differences are noted above the bars, where * = p<.05 and ** = p<.001] The effects of tours on all other behaviour categories Observations of the behaviour categories of feeding, aggression, interact conspecific, interact human, other, and out-of-sight were typically low, or non-existent, across all conditions. The total number of observations with an incident of each of these behaviour categories in each condition are shown in Table 5.5. Trends in feeding behaviour appear relatively consistent, with increased incidents of observations in the during condition on tour days. Incidents of aggression were very low for all species, except for lions who demonstrate a trend for increased aggression in 183

202 the during tour session on tour days. For interact conspecific behaviours, the lions showed a decreasing trend in the during session on both non-tour and tour days, and the cheetahs showed a trend for a decrease in the during session on non-tour days only. This behaviour almost never occurred for the tiger. While incidences of interact human were similarly low for the lions and tiger, and showing a trend for a slight increase in the during session within tour days, for cheetahs there was a trend for increased incidents this behaviour category in the pre session on non-tour days and during session on tour days. Lions tended to show lower incidences of other behaviour in the during session on both non-tour and tour days. Cheetahs demonstrated a trend for lower incidences of other behaviour on non-tour days, and a decrease in the post session on tour days. Incidences of other behaviour for the tiger were moderate and decreased in the during session on tour days. Lastly, while the lions were almost never out-of-sight, the cheetahs and tiger showed quite high incidences of this behaviour in all sessions. In particular, cheetahs showed a trend for more incidences in the pre session on both nontour and tour days, and the tiger showed a moderate decrease in out-of-sight in the during session on tour days. 184

203 Table 5.5. Total number of observation sessions with at least one incident of Feeding, Aggression, Interact conspecific, Interact human, Other, and Out-of-sight behaviours observed in the lions, cheetahs and tiger across the three observation sessions (Pre, During, and Post) on non-tour (NTD) and tour days (TD). Feeding Aggression Interact conspecific Interact human Other Out-of-sight NTD TD NTD TD NTD TD NTD TD NTD TD NTD TD Lions Pre During Post Cheetahs Pre During Post Tiger Pre During Post Max. number of possible sessions 185

204 Estimated proximities to humans during tours For both lions and cheetahs, there was no significant main effect of humans (keeper, visitor) on estimated proximities (Table 5.6). However, there was a main effect of proximity (close = <2m, moderate = 2m-5m, distant >5m) for both. There was also a two-way effect of human x proximity for cheetahs. Table 5.6. Results of main and mixed effects of linear mixed models analyses for proximity to humans for lions and cheetahs. df f Lions Humans (keepers/visitors) 1, Proximity (close/moderate/distant) 2, ** Humans * Proximity 2, Cheetahs Humans (keepers/visitors) 1, Proximity (close/moderate/distant) 2, ** Humans * Proximity 2, ** Note: * = p<0.05; ** = p<0.01 Differences between conditions and significant post hoc comparisons (where appropriate) for estimated proximities are shown for each species in Figure 5.4. For lions, within human comparisons revealed that for both keepers and visitors, these animals were observed to spend more scans in distant proximity (~ 65%), compared to close (~ 10%) and moderate (~ 25%), and more scans in moderate proximity compared to close. For cheetahs, within human comparison revealed similar differences for both keepers and visitors, such that these animals were observed to spend more scans in distant proximity (~ 57%) compared to close (~ 30%) and moderate (~ 12%), and more scans close proximity compared to moderate. The tiger showed a trend for very similar proximities to keepers and visitors, with more scans observed in close (~53%) compared to distant (~31%) and moderate (16%). 186

205 Mean % of scans distant>close**, moderate**; moderate>close* distant>close**, moderate**; moderate>close* distant>close**, moderate**; close>moderate** distant>close**, moderate**; close>moderate* Keepers Visitors 20 0 Close Moderate Distant Close Moderate Distant Close Moderate Distant Lions Cheetahs Tiger Figure 5.4. Mean percentage of scans in which lions, cheetahs and tiger were observed in each proximity category (close, moderate, distant) to humans (keepers and visitors) in the during tour session on tour days. [Note: Bars indicate standard deviation; significant differences within distances to humans and visitors are noted above the bars, where * = p<.05 and ** = p<.001] 187

206 5.4. Discussion To date, no peer-reviewed research has explored how large felids respond, behaviourally or otherwise, to interactive zoo visitor tours. The present case study sought to investigate whether short-term behaviour differs in lions and tigers involved in a protected contact tour, and in cheetahs in a hands-on tour, with observations conducted on non-tour days acting as controls. Observed behavioural profiles were primarily comprised of active species-typical, inactive species-typical, and pacing behaviour for all three species. There were some fluctuations in these behavioural categories across conditions, as would be expected with the addition of a stimulus to an animal's exhibit. However, these fluctuations do not necessarily appear to be associated with the tours per se. Findings are discussed separately for the two tour types herein Changes in behaviour with the protected contact tour Despite the lions and tiger participating in the same tour, results were analysed at the species level, and therefore findings are discussed individually here. As highlighted in Section 5.2.6, some caution should be taken when interpreting results between conditions within non-tour and tour days, considering the time difference in observation sessions (i.e., ~1hr pre, 15min during, 1hr post for lions; and ~30min pre, 15min during, and 1hr post for the tiger). As such, although proportions of observation sessions are reported here (not absolute durations of behaviour), these differences should still be considered Lions When combined, active species-typical and inactive species-typical behaviours accounted for a large proportion of behaviour in any condition for the lions. Given that these categories comprise of behaviours within the natural repertoire of this species, this could be indicative of tours not compromising well-being. Although it is somewhat difficult to make comparisons between the behaviour of lions (and other species) in zoos, largely due to the relatively small data set presently available for this species, and the differences in environments and methodologies with which this data are collected, other studies of zoo-housed lions suggest that inactivity commonly accounts for a large 188

207 proportion of their behaviour in this environment. For example, Bashaw et al.'s (2007) study sought to explore the influence of different environmental variables on the behaviour of large felids in zoos; combining mean proportions of behavioural observations for lions (n = 3) and tigers (n = 2), it was found that inactivity comprised approximately 70% of scans throughout the day, whereas activity accounted for approximately 10%. Given that these data were pooled for lions and tigers, literature specific to lions can provide further insight into this animal's behaviour. In Margulis et al.'s (2003) study of the influence of felid activity on human interest, animal behaviour was categorised as either active or inactive; the sole lion in the study was observed to be active for less than 20% of scans, implying the dominance of inactivity in this animal. However, data collection was limited to midday periods during summer months, which may contribute to lowered activity (due to the heat at this time) and therefore this finding should be viewed with caution. Despite this limitation, this finding is supported by the baseline data collected in Van Metter et al.'s (2008) enrichment study which found lions (n = 2) to sleep for over 80% of scans. While sleeping may not be considered a particularly reliable behaviour category (given that it is difficult to determine whether an animal is actually asleep and not just resting), this data adds support to the idea that lions are often inactive during day-time hours. So, although there are difficulties inherent in comparing these data, it becomes clear that inactivity is a dominant behaviour observed in these animals, which is confirmed by what is known about their natural activity patterns in home countries (Guggisberg, 1975). As such, the finding of high proportions of inactive behaviours for lions across most conditions in the present study is consistent with past research. However, some changes were observed in the proportions of behaviour of these animals across conditions, particularly that inactive species-typical behaviour significantly decreased and pacing behaviour significantly increased in the during session on tour days. While inactivity has been established as a commonly observed behaviour among this species, decreases have been seen during the provision of enrichment items in other studies (e.g., Van Metter et al., 2008). The decrease in inactivity here can probably be accounted for by the rise in pacing within this condition, as well as by the trend for increased feeding behaviour. The fluctuation in pacing behaviour may be of some concern. Pacing is a common form of stereotypic behaviour observed in large felids (see 189

208 Section 1.1.2) and, depending on circumstances, its presence can be suggestive of compromised animal well-being (Carlstead, 1998). In this study, lions demonstrated the greatest proportion of pacing on tour days in the during tour session, as well as more pacing before the tour than after. This trend was different from non-tour days, where these animals showed no changes in pacing across times, but did show lower proportions of pacing in the pre and during on non-tour days, and more post on non-tour days. One possible reason for this change is that the lions' pacing may be associated with them being fed during the tours, supported by the trend for a higher incidence of sessions in which feeding behaviour occurred in the during condition on tour days. While frequencies of feeding were too low across all conditions to allow statistical analysis, pacing prior to feeding has been observed at other feeding times at this zoo (personal observation), as well as in lions at Monarto Zoo who are also fed on behindthe-scenes tours (these lions were originally included in this study, but data collection was ceased due to changes in group composition both on- and off-exhibit, as explained in Section 5.2.1). Indeed, research indicates that animals may pace in association with feeding and Carlstead (1998) explains that temporally predictable feeding schedules have been linked to stereotypy in some carnivores. Bassett and Buchanan-Smith (2007) also argue that food anticipatory activity characterised by increased arousal and activity is typically seen in animals which are fed on a regular schedule and there is some evidence to support this for lions. Although not explored as anticipatory activity, greater observations of pacing behaviour were observed in Lyons et al.'s (1997) study of preexisting feeding regimes, with a decrease in pacing from 65% before feeding to 15% after feeding in lions (n = 2) fed daily. Since the lions in this tour are regularly fed (albeit a small amount of food) during the tours, this pacing could be a form of food anticipatory activity, rather than stereotypy. That these animals paced less in the pre and during tour blocks on non-tour days compared to tour days, and minimally in the posttour block on both days, adds some support to this argument, since feeding usually only occurred in the during condition on tour days. Before drawing a concrete conclusion about the causes of this pacing, it would be important to observe the lions' behaviour on tours without feeding, as well as their feeding behaviours at other times of day. Moreover, further research into the differences in the purpose of pacing behaviour in 190

209 different contexts/in response to different stimuli could help to accertain whether such behaviour, as displayed here, is anticipatory; moreover, further research could help to determine whether this anticipation is positive or negative in nature. Given that predictability might contribute to reduced welfare in zoo animals (Morgan & Tromborg, 2007), some efforts could be made to minimise the lions' pacing behaviour if this is considered problematic by ZoosSA staff. This may be deemed problematic on the grounds that pacing may be considered stereotypic and possibly compromising welfare; and/or on the grounds that it may not be desirable for visitors to observe this behaviour (e.g., Miller (2012) offers preliminary evidence to indicate that support for tiger conservation decreases when visitors see these animals pacing). If deemed problematic, efforts to minimise this behaviour could include altering the order in which these tours are conducted, to reduce the predictability associated with feeding at this time (for a more detailed discussion of the negative effects of predictable feeding, see Bassett & Buchanan-Smith, 2007). Another alternative could be to introduce a reliable feeding signal (such as a buzzer) to indicate to the animals when they will be fed (Bassett & Buchanan-Smith, 2007). The rationale behind this is that by knowing exactly when they will be fed, the lions could wait for this cue rather than rely on other cues (in this instance, the visual, auditory and olfactory cues that a tour group is present, which precede feeding). In particular, this may aid in reducing the spike in pre-tour pacing observed on tour days compared to non-tour days Tiger Although it was not possible to conduct statistical analyses for the tiger's behaviour, the trends observed on tour days for this animal warrant some attention. Like the other two species, this animal primarily engaged in active species-typical and inactive speciestypical behaviours across all three conditions. Again, this is consistent with Bashaw et al.'s (2007) aforementioned finding (see Section ) and with findings from baseline phases of enrichment studies with this species (e.g., Bashaw et al., 2003; Jenny & Schmid, 2002; Skibiel et al. 2007; Van Metter et al., 2008). A trend for increased proportions of pacing were observed from the pre to post observation sessions, such that the highest proportions of this behaviour were observed after the tour had concluded. 191

210 Pacing after the tour could indicate a possible short-term welfare issue. In all sessions, this animal predominantly paced along the perimeters of the enclosure, which are shared with female tigers, indicating that perhaps the behaviour is related to the uncontrolled stimulatory access of neighbouring conspecifics, rather than tour-induced 'stress'. Although the tiger s behaviour was not included in the comparison of tour and non-tour days in this study (meaning that comparisons cannot be made regarding whether this behaviour occurs more frequently on tour/non-tour days), pacing in the same locations was observed in this individual in the morning sessions on non-tour days in this study, and throughout the day in an earlier, unpublished study by this researcher. In addition, although this individual would sometimes engage in a few bouts of pacing along the fence where tours were conducted as the tour group was arriving, he would usually then lay by the fence (often chuffing) until he was fed. Others have witnessed pacing in response to conspecifics. For example, de Rouck et al.'s (2005) multi-institutional study of female tigers (n = 15) found those who were pair housed and had no neighbouring tigers paced less (5% of active scans) compared to those who were either pair or solitary housed with neighbouring tigers (>20% of active scans). The use of visual barriers to block the view of neighbouring conspecifics has been previously explored for its potential to reduce this behaviour (Bashaw et al., 2007; Miller et al., 2008). Indeed, a visual barrier has now been put in place at this exhibit since this study was conducted, but research has not yet explored how influential this has been in reducing pacing behaviour Interactive behaviour and proximity to humans during the protected contact tour Given the absence of past research exploring interactive behaviour between these species and people in a protected contact capacity, it is not possible to compare this aspect of the present findings to current knowledge for these species Lions Incidents of the interact human behaviour category were negligible for lions, although there was a trend for more of this behaviour in the during session on tour days, which is to be expected given the lack of human interaction these animals have outside of tour 192

211 times. By definition, all interactive behaviours were affiliative in nature, although these animals did demonstrate some aggressive behaviours at the start of the tours, primarily banging on the doors at the back of their enclosures (with or without vocalisations) when the tour group had arrived. This behaviour was also observed at other times when keepers walked through these areas alone. As with pacing, formal documentation of this behaviour at other times of day would be needed to determine whether this is a response to the presence of the tour group, or an anticipatory behaviour related to keepers being in the area and the possibility of feeding. Other than this, no direct or indirect aggressive behaviours were observed. Being a protected contact tour, there is obviously limited opportunity for direct physical interactions, either positive or negative, to take place. The estimated proximity data indicate that lions spent the majority of their during tour scans in distant proximity >5m to both keepers and visitors. When the animals were in close and moderate proximity, this was usually prior to and/or during feeding time, after which they would retreat to other parts of their enclosures Tiger Like the lions, incidents of observations of the interact human category were negligible for the tiger (occurring in less than half of the during tour sessions, and not occurring at all in the pre and post observation sessions). When this behaviour did occur, it was usually whilst visitors or keepers were feeding this animal, and no aggressive behaviours were observed at any time. Dissimilar to the lions, however, this animal showed a trend for spending the majority of tour scans in close proximity to the tour group. As such, this data suggest that visitors on the protected contact tour are likely to see the tiger 'up close' for a greater portion of the tour than they would see the lions. Like the lions, this animal would often lay directly by the fence until he was fed, after which time he would move away to consume the food and not return. Obviously, this data cannot be used to draw concrete conclusions about why the animals preferred to be close to people at times, it appears that they only remained close for the purpose of obtaining food (i.e., a conditioned response). Conducting tours without feeding would again be useful in determining how this might influence proximity to keepers and visitors for both the lions and tiger. 193

212 Changes in behaviour with the hands-on tour Similar to the animals in the protected contact tour, high proportions of inactive speciestypical behaviour were observed in all conditions by the cheetahs in the hands-on tour (range ~40% - 85% across all conditions). Again, this is consistent with the limited available information regarding how these animals behave in zoos. For example, although related to adult females and male and female cubs, O'Donovan et al. (1993) found inactivity to comprise almost 90% of scans, regardless of visitor presence or absence at exhibits. It is also consistent with the natural activity pattern of this species (Guggisberg, 1975) and could be indicative of positive well-being. The decrease in inactivity on tour days in the during and post sessions could be accounted for by the increased activity, as well as by trends for greater incidences of feeding and interact human behaviours. Regarding activity, these animals demonstrated an increase in active species-typical behaviour in the during and post session on tour days, compared to nontour days, which may indicate the potential for tours to be enriching for these animals. For example, the presence and movement of people within their enclosure might allow for variation to the environment (e.g., scents), which could contribute to increased olfactory behaviours. Similar increases in activity have been shown for this species with the provision of novel food, objects, and scents in Skibiel et al.'s multi-species enrichment study (2007), as well as with variations to feeding regimes and the provision of olfactory enrichment in Quirke and O'Riordan's (2011a, 2011b) investigations. This finding is inconsistent with O'Donovan et al.'s (1993) finding for no changes in active behaviour in cheetahs when visitors were present or absent, but this difference may be attributed to the study exploring visitors at exhibits, rather than in exhibits (like the present study). No differences were found in proportions of pacing on tour and non-tour days, and similar differences were observed within days, such that on both tour and non-tour days pacing was at its highest in the pre tour session. So, although pre-tour pacing could indicate an anticipatory-type response, the lack of difference between observations of this behaviour on non-tour and tour days may discount this argument and may suggest that participating in tours does not influence this behaviour. Alternatively, perhaps this 194

213 behaviour is a conditioned response to the time of day, but not days of the week, that tours start. However, more support can be given to the former argument by considering the context of this pacing. Like the tiger, it is proposed that the cheetahs' pacing may be related to the presence of neighboring conspecifics, since (although not analysed statistically) pacing primarily occurred along the boundary of their exhibit shared with conspecifics, either when the other cheetah/s were being shifted in or out of the adjacent enclosure, or when keepers were performing routine maintenance prior to the other cheetahs being given access to the exhibit. This happened on both non-tour and tour days. Notes taken during data collection indicate that these activities usually occurred in the mornings, roughly half an hour before the tour condition start time. Being territorial animals, this may be an expression of natural cheetah patrolling behaviour, and is consistent with a multi-county, multi-institutional study of 112 cheetahs, where the probability of observing pacing in these animals significantly increased when cheetahs had the ability to view conspecifics in an adjacent enclosure (Quirke et al., 2012). Visual barriers may again be useful here (if this behaviour is deemed problematic) but, given the size of the enclosures at this open-range zoo, this is probably not feasible. Although not significant, the trend for decreased observations of pacing in the during tour session on tour days (from approximately 10% of scans to zero) may add some support to the potentially enriching effect of tours, with this activity possibly providing an alternative source of interest than neighbouring conspecifics Interactive behaviour and proximity to humans during the hands-on contact tour Given the hands-on nature of their tours, and the opportunity for direct interactive behaviours to occur, the cheetahs showed a trend for higher incidences of observations engaged in human interaction in the during tour sessions. Although not investigated individually in this study, these behaviours typically involved the animals sitting with and being patted by keepers and visitors, as well as engaging in training with keepers. No aggressive or otherwise antagonistic behaviours were displayed by the cheetahs toward humans during the tour (nor during any other condition), indicating no obvious threats of danger to any parties, despite the physical contact involved. This is similar to O'Donovan et al.'s (1993) finding of no difference in reactive behaviours (e.g., ground slapping, spitting, or hissing) in cheetahs when visitors were present or absent. 195

214 In addition to these direct interactive behaviours, the proximity data can also contribute to an understanding of indirect interactions. Despite the proportion of scans spent engaged in direct interactions, these animals were in close proximity (< 2 m) to keepers for approximately 35% of scans, and to visitors for 25%. As such, they remained close for notable proportions of tour scans, despite not interacting physically the whole time. However, they also spent the majority (almost 60%) of during tour scans in distant proximity to keepers and visitors, indicating that despite being in a situation where direct physical interaction was possible, the animals might show a preference for being further away from the tour group. These findings differ from those of O'Donovan et al. (1993), who observed that cheetahs tended to use the front sections of their exhibit (i.e., those closest to visitor viewing areas) more than the middle or back sections regardless of whether visitors were present or absent, implying a lack of change in location in reaction to visitor presence. The difference in findings in the present study is likely attributed to the hands-on nature of this tour, with visitors actually entering the exhibit Positive Aspects of the Zoos SA Interactive Tours Although these findings do suggest that interactive tours may influence the type and proportion of behaviours observed in these large felids, the finding of relatively high proportions of species-typical behaviours (both active and inactive) across conditions is positive, and the occurrences of pacing, which might be regarded as problematic for welfare, might be explained by factors not specific to the tours. As such is important to draw attention to some of the potentially positive factors of the tours. Firstly, predictability has been said to be an important factor in the captive environment, depending on the circumstance (Morgan & Tromborg, 2007). All of the study animals have been participating in these tours for a number of years and the nature of the tours have remained relatively consistent, which may have resulted in a high degree of predictability. Although different visitors participate in the tours, the predictability that tours do not pose threats to the animals may be important here. Another factor is that none of the animals are forced to interact with visitors or to engage in specific behaviours, but are positively reinforced for their participation. This reinforcement can be direct (such as cheetahs receiving verbal praise for sitting on command whist visitors pat them), or simply feeding all animals at the end of a tour. As 196

215 discussed Section , positive reinforcement training has been well-argued as the most successful and appropriate training technique, and has been suggested to be beneficial to animal welfare in the zoo setting, since it increases control and predictability over aversive events (Bassett & Buchanan-Smith, 2007; Laule & Desmond, 1998). Further, these animals have the opportunity to retreat if desired. Forced proximity to humans has been highlighted as potentially stressful for zoo-housed animals (Morgan & Tromborg, 2007), and there are many features in each of the enclosures which could be used by the animals to remove themselves from view and potentially reduce any stress associated with visitor presence. Indeed, the lions and cheetahs in this study spent most of their tour time in distant proximity to keepers and visitors. This is more so important in the cheetah tour, where visitors actually enter the exhibit. In many instances, the animals were not in close or moderate proximity to humans and would often interact for between five to 10 minutes before walking away and laying down in the shade, away from the group. When the animals did this, they were never followed by keepers or visitors. Like predictability, having retreat spaces has been found to be important in interactive programs with other species, and in moderating animal behaviour and improving animal welfare (Anderson et al., 2002; Samuels & Spradlin, 1995). One final factor is that these tours are always conducted by zoo keepers. Whilst it is unlikely that tours of this nature would ever occur without an experienced zoo keeper present, a study of dolphins (n = 22) at four facilities in the United States found that potentially risky behavioural interactions were lower in sessions where trainers had direct control over the interactions between dolphin and visitor, compared to those without direct trainer control (Samuels & Spradlin, 1995). The presence of a familiar keeper will undoubtedly influence how the animal responds since there is likely to be an existing relationship. It may be that having a keeper present mediates the interactions, making it a more positive experience. Indeed, in reviewing studies of interactions between humans and zoo animals, Claxton (2011) suggests that it may be possible for an animal s interactions/relationship with unfamiliar humans (e.g., zoo visitors) to be influenced by the relationship that they have with familiar humans (e.g., zoo keepers). In the case of the present study, the lack of observed aggressive behaviour toward humans, and direct affiliative interactions between cheetahs and visitors specifically 197

216 (see Section 5.4.4), might be indicative of this effect, with the animals behaving in ways that they would otherwise behave with their keepers. Similarly, the observation of similar aggressive responses of lions when tour groups were arriving at their enclosure and when keepers were arriving at non-tour times (see Section ), might also be indicative of this effect. Alternatively, the animals in this study may not have been responding to visitor presence at all, since the animals appeared to respond to the tour group in the same way as they would when keepers were present at other times of the day (for example, becoming alert upon keeper arrival, or pacing before a feed) Limitations, Confounding Variables and Directions for Future Research When combining results from each of the tour types, it is evident that these tours were having some influence on animal behaviour. This is consistent with the small amount of previous research that has shown large felid behaviour to vary in association with zoo visitation at exhibits (Maia et al., 2012; Mallapur & Chellam, 2002; Sellinger & Ha, 2005). The findings from this study extend this past research by providing an understanding of the potential influence of closer visitor interactions on animal behaviour. However, drawing conclusions about which aspects of the tours contributed to alterations in behaviour is difficult. There are always likely to be a number of confounding variables in zoological studies, many of which are unavoidable if the purpose of the research is to observe animals behaving according to their regular routines (and if it is not possible for researchers to request alterations to zoo routines). This often makes drawing conclusions about behaviour change quite challenging, particularly when small sample sizes restrict the ability to control for such differences in statistical analysis (due to the fact that this would reduce power to detect effects). As such, some of these issues are discussed here and suggestions for how they might influence the results are provided, as well as how they might be addressed in future research. Firstly, since there was a difference in the feeding routines of these animals on tour and non-tour days, it would be important for future studies to consider keeping this potentially confounding variable consistent across conditions, in order to more fully 198

217 understand the influence of tours. This was not possible in this study, since the objective was to observe the animals behaviour according to their regular routines, which therefore meant not wanting to make any changes to their usual husbandry or management. Moreover, it is suggested that it might be possible that the animals here responded to the tour group in the same way that they do to keepers alone, at other times of day (see Section 5.4.5). It would be imperative for future research to document animal responses to keepers at various times throughout the day, to determine whether such a similarity does exist and to draw more informed conclusions about animal responses to participating in tours. Hosey (2008) argues that both the quality and quantity of animal interactions with keepers differ from those with visitors; therefore, separating these influences would also be important. Moreover, Phillips and Peck (2007) found that tiger keepers differ in their individual interaction style with their animals (discussed in Section 4.4), so understanding if and how large felids respond to different keepers/keeping styles should be addressed in order to maximise experiences. Although two different types of tours (protected contact and hands-on) were explored in this study, the behaviour of both groups of animals was not statistically compared and so conclusions about which tour might have a greater influence on behaviour, and welfare, cannot be made. Comparing different tour types was not the purpose of this study, and such a comparison was deemed inappropriate here, since there are various other differences in the histories and current husbandry of these two groups of animals which could account for behavioural differences. Addressing the influence of varying types of tours could be made possible at organisations where the same species participate in different types of tours. Along these same lines, whether animals have been hand-reared and the amount of prior interaction that they have had with humans could account for behavioural differences during tours, so it would be useful to compare the same species with different levels of interaction, to determine whether any interspecies differences exist. This could allow zoological organisations to understand which large felid species are most suited to the different types of tours. It has been proposed that various factors will account for a particular animal s response to humans such as species, extent of handling in early life, animal personality and temperament (Davey, 2007; Hosey, 2008). As such, the investigation of how individual factors relate to 199

218 interactive tours would also be useful in being able to predict which animals would be suitable for involvement in such programs. As with many studies of zoo/aquarium-housed animals (see Section 3.3.1), this one was limited to a small number of subjects, at one organisation, making findings difficult to generalise. However, as indicated in Table 2.1, the animals involved and the types of tours conducted were reflective of those in the broader Australian zoo network. In order to generalise findings beyond these individuals, future research should consider a greater number of animals, at multiple institutions, as well as other large felid species. Moreover, the generalisability of these findings is limited in that the individual animals involved here have participated in the tours for a number of years (four years for the tiger and lions, and most of their lives eight years - for the cheetahs) and have likely adapted to interactive tours to some degree. Additionally, being hand-raised, the cheetahs have daily physical interaction with their keepers, providing them with further opportunities to adjust to interacting with people. In this way, it was only possible to examine immediate, short-term behaviours associated with tours. It would be important for future research to examine long-term changes in animals who are not yet involved in such a program. In terms of gauging the influence of tours, although the use of multiple measures of well-being have been previously advocated for (e.g., see Hill & Broom, 2009), behaviour was deemed the most efficient method here (for rationale for the use of behaviour as a measure, see Section 3.3.2). Since the study was interested in immediate effects, and because many other factors in the zoo environment may contribute to stress, a corticosteroid analysis (or similar) did not appear suitable here, but may be of value to a longer-term investigation focusing on the establishment of tours with new animals. An alternative, non-invasive method for exploring welfare implications could be the use of heart rate monitors, which have been shown to be effective means for measuring heart rate variability as an indicator of stress in farm animals (for a review of the utility of these devices, see von Borell et al., 2007). Obviously, the logistics and practicalities of fitting such a device to large felids would need to be considered, but this could be an option for animals that are managed handson. As a result of focusing solely on behaviour, the present findings do not provide a complete picture of animal responses to tours, and therefore caution should be taken when interpreting these results. 200

219 Beyond the behaviour and welfare of animals involved in interactive tours, future research should also address the safety of zoo visitors during such experiences. Although no direct aggressive or otherwise antagonistic behaviours towards humans were observed in this study, there is a wealth of literature to indicate that large felids can cause both injuries and fatalities to people, in their natural home ranges (see Section ) and in various forms of captivity (see Section ). Indeed, when surveying zoo keepers about their opinions of interacting with large felids (see Chapter Four), it was found that safety was a major concern among participants. A number of these keepers discussed human safety as an issue when a hands-on style is adopted; they also expressed the concern that if an attack does take place, the animal involved is usually euthanised, highlighting the negative consequences for both people and animals when interacting. Typically, participants explained that a protected contact style is safer and more appropriate than hands-on. For these reasons, it will be important for research to continue to investigate the behaviours displayed by these animals during tours, and how the safety of both humans and animals might be preserved or compromised Conclusions The results of this study provide a starting point for understanding how participation in interactive zoo visitor tours might influence animal behaviour, with some indicators to suggest a potential negative influence on the lions and tigers in a protected contact tour, but a potentially enriching effect on cheetahs in a hands-on tour. The research also supports the continued use of behavioural monitoring as an effective means to understand animal welfare. Differences in proportions of behaviours namely, active species-typical, inactive species-typical, and pacing - were observed between the conditions in this study, however, due to the multiple confounding variables inherent in this type of research, it is not possible to draw concrete conclusions about how animal welfare might be influenced by these tours. In particular, it is important to acknowledge that it cannot be determined, from this investigation alone, which aspects of the tours are responsible for the changes observed. This will need to be explored in future research. Moreover, since the presence of pacing behaviour can be indicative of compromised well-being, it will be of importance for future research to address this in order to support the continued involvement of large felids in interactive programs. However, it is suggested that the overall conditions under which these tours occur such as predictability, allowing the animals choice to participate, the use of positive 201

220 reinforcement, and the provision of retreat spaces appear quite positive and should be encouraged elsewhere. There is clearly a need for more research in this area and it is hoped that this study will promote the investigation of large felids involved in interactive tours at other zoological facilities. Ferndandez et al. (2009) argue that those interactions which meet the entertainment and education aims of visitors, whilst also being helpful for animal welfare, should be encouraged. As such, the thesis now turns to zoo visitors, with the next chapter exploring how visitors at Zoos SA respond to these interactive tours. 202

221 6 Exploring zoo visitor responses to interactive versus traditional large felid experiences In order to achieve their mission objectives and also to be financially viable, zoos must not only admit the public but must also give them an enjoyable and positive experience based upon seeing and perhaps even interacting with active, healthy animals. G. Hosey (2005, p. 112) The previous chapter explored the behaviour of large felids during interactive zoo visitor tours at Zoos SA. The present chapter complements this by exploring how zoo visitors respond to these same experiences, fitting with the overall aim of the thesis being to explore the impact of interactive practices on different user groups. As acknowledged in Section 3.1, animal management and husbandry do not concern animals in isolation and can have various impacts on the visiting public. In this light, the present chapter first describes what is known to date about the influence of zoo visits (both traditional and interpretive) on the public. It then describes a study which explored differences in how visitors respond to interactive and traditional large felids experiences, particularly in terms of knowledge, attitudes, emotions, and behaviour change intentions Introduction As highlighted in Section 2.4.2, today s zoos vary significantly from those of the past, with a greater focus placed on their role as informal learning settings and their high conservation potential. Presently, the rationale for many zoos housing collections of endangered species is for these animals to directly contribute to conservation efforts where appropriate, particularly by acting as ambassadors for their wild counterparts (WAZA, 2005). This is based on the anticipation that an experience with these animals 203

222 will increase visitor knowledge of the animals and the issues they face, improve visitor attitudes towards them, and ultimately encourage behaviours that will aid in the conservation of the species (WAZA, 2005). Although the exact mechanisms through which these outcomes occur are complex and not entirely clear, there is a growing body of literature which suggests that a zoo visit can have a positive influence on a person s knowledge, attitudes, and/or behaviour (see Section 3.4.1). There is also growing acknowledgement of the emotional aspect of zoo visits and the role that this can play in enhancing the visitor experience (again, see Section 3.4.1) Beyond traditional exhibit viewing Despite the need for more research to better understand the knowledge-attitudebehaviour link in this informal learning setting, there is emerging support for the use of interactive programs in zoos, with a general consensus that in order to optimise knowledge gain, both the cognitive and affective components of learning must be targeted (Ballantyne & Packer, 2005; Clayton et al., 2009). That is, coupling factual information with other means to engage the emotions of learners to enhance knowledge gain (Ogden et al., 2004; Povey & Rios, 2005). Indeed, a number of studies have shown that visitors do not spend much time viewing exhibit signage (Bowler et al., 2012; Clayton et al., 2009; Ross & Gillespie, 2008) this is problematic given that signage is the key way in which information is usually available to visitors at an exhibit. To address this, zoos have been increasingly creative in designing enhanced visitor experiences, beyond simply viewing exhibits, including video and interactive touchscreens, touch-tables, the presence of keepers and/or volunteers at exhibits for visitors to talk to, designated keeper talks, public training/interactive sessions, and zoo theatre. Some of these techniques have been suggested to increased visitor stay time (Broad & Weiler, 1998; Harley et al., 2010; Povey & Rios, 2005). For example, in a review paper summarising the results of research conducted at Disney's, The Seas, Harley et al. (2010) report that visitors spent more than four times longer watching bottlenose dolphins (Tursiops truncatus) when the animals were engaged in an interpretation session than when they are not (but what these stay times actually were was not stated). As such, it is apparent that these interactive practices can present a greater opportunity in which to captivate and educate visitors, and the literature is beginning to provide 204

223 evidence for the positive influence of interactive experiences. As described in Section 3.4.1, this literature typical involves survey methodology and the findings from such studies in relation to the four commonly investigated outcome measures - knowledge, attitudes, emotions, and behaviour - are discussed herein, within the context of what is known about these variables in response to traditional exhibit viewing. More details about the methods used in these studies can be found in Section Changes in visitor knowledge associated with traditional and interactive experiences Zoo visits present an opportunity for informal learning and the literature has begun to explore knowledge gain amongst the visiting public, particularly in the short-term via surveys immediately after visits. However, measuring learning in such settings is difficult. Learning is a cumulative process, occurring frequently in different contexts throughout the duration of a person s life, resulting in variation in the knowledge bases that visitors arrive at zoos with (Clayton & Brook, 2005; Ballantyne & Packer, 2005; Falk & Adelman, 2003; Heimlick & Storsdieck, 2007; Knapp & Yang, 2005; Lukas & Ross, 2005). In addition, despite being presented with the same learning materials (e.g., exhibit signage), zoo visitors may encounter these materials in different ways, depending on individual differences. Moreover, it has been shown that visitors have different motivations for, and interests in, visiting (e.g., education versus entertainment) and this can influence their learning (Ballantyne et al., 2007; Falk & Adelman, 2003; Falk & Dierking, 2000; Falk et al., 2007; Packer & Ballantyne, 2002). Other factors, such as who an individual visits the zoo with, can also influence their learning opportunities. For example, in a tracking study of visitor (n = 338) behaviour at Lincoln Park Zoo's Regenstein African Journey exhibit, Ross and Gillespie (2009) found that visitors without children spent a greater proportion of their visit looking at signage than those with children, presenting different opportunities to engage with learning. Similarly, those who spent more time engaged in social interactions spent less time engaged with signage. While these data provide support for differences in learning opportunities, they were not linked to learning outcomes. 205

224 Another consideration is that factors related to the animals themselves may influence visitor interest and hence learning. Margulis et al. (2003) found that visitor interest (as determined by the number of visitors at, and time spent viewing, the exhibits) in seven felid exhibits at Brookfield Zoo was higher when the animals were more active. Also using an observational approach to record the proportion of visitors who stopped at exhibits, and their exhibit visit duration, Moss and Esson (2010) concluded that mammals were most interesting to visitors at Chester Zoo. Whilst documentation of visitor behaviour can be used to draw conclusions about visitor interest, this method alone does not account for why certain animals might elicit more interest. Although not contributing to this understanding regarding differences in interest toward different species, a self-report survey at multiple Australian zoos found visitor (n = 240) satisfaction with orangutan activity to significantly predict visitor knowledge (Pearson et al., 2013). Given these various factors, it has been proposed that a combination of individual and zoo-based factors will contribute to the learning experience (Broad & Weiler, 1998). Regardless of these complexities, studies of typical zoo visits have demonstrated the potential for zoological organisations to contribute to learning, particularly at exhibits (e.g., Adelman et al., 2000; Falk et al., 2007; Lukas & Ross, 2005; Mallapur et al., 2008). A handful of studies have explored knowledge gain in association with interactive experiences, including interactive elements at exhibits, as well as keeper talks and zoo theatre. Lindeman-Matthies and Kramer (2005) explored the influence of a manned touch-table at a bearded vulture (Pogona vitticeps) exhibit on visitor learning at a Swiss zoo. Utilising a self-report questionnaire, the authors explored knowledge regarding the biology, ecology and conservation of bearded vultures. When asked to list the topics learnt about during their visit, 47% of those who visited the exhibit when the touchtable was present (n = 313) reported learning about the species, compared to 13% of those who viewed the exhibit when the touch-table was absent (n = 326). Similarly, those in the test group were able to more accurately answer questions specific to bearded vultures. A two-month follow-up with a sub-group of these visitors (n = 113 touch-table, n = 93 non-touch table) found these results to be sustained. These differences were found despite only 47% of participants in the touch-table condition actually interacting with the touch table. A stronger understanding of the impact of this 206

225 feature may have been gained if the researchers tweaked the methodology to only include visitors in this condition if they had interacted with the touch-table. In a similar study of interactive elements, Pearson et al. (2014) conducted an evaluation of the 'Don't Palm Us Off' campaign at Melbourne Zoo. During a year-long campaign, a one-minute video was played on a constant loop at the orang-utan exhibit, educating visitors about the impact of palm oil on orang-utan habitat destruction, and the use of palm oil products in Australia. Interactive touch screen computers were also located at the exhibit for visitors to interact with. Groups of adult visitors (n = 403) at the exhibit were invited to complete a self-report questionnaire at different time points: 6 months before the campaign began; 6 and 12 months in to the campaign; and 6 months after the campaign ended. No changes were found in overall visitor knowledge of orang-utans across the different time-points, but participants more often correctly identified palm oil as the product most threatening the survival of the species during and after the campaign. The finding for this knowledge to be sustained after the campaign had ended may be associated with the campaign being advertised outside of the zoo (e.g., through free-to air television advertisements and social media), supported by the fact that the majority of visitors surveyed (approx. 90% across conditions) resided in Australia. Keeper talks/presentations have also had a comparable influence on knowledge gain. In a study of adult visitors, Povey and Rios (2005) explored the influence of an interpretive clouded leopard program at Point Defiance Zoo and Aquarium, whereby keepers walk these animals around the zoo, offering visitors a chance to encounter them spontaneously and ask questions. Through observation and documentation of visitor viewing time and information-seeking behaviour, it was found that a greater proportion (45%) of visitors who engaged with the interpretive element (n = 75) displayed information-seeking behaviours (namely, asking questions), compared to only 25% of visitors who viewed the exhibit (n = 75) whose sole information-seeking behaviour was reading exhibit signage. Additionally, a second set of visitors was surveyed after their experience (n = 55 interpretive; n = 57 exhibit), using two open-ended questions to explore learning. Unfortunately, however, only vary sparse information is provided in the article, with the key finding being that interpretive visitors reported more factual 207

226 information (i.e. something they had learnt) while the exhibit visitors reported more observational information (i.e. something they had seen). Adding a more comprehensive measure of knowledge gain (e.g., a quiz like that described by Pearson et al., 2014) would have added strength to the conclusions drawn from this study (i.e., that interpretation is an effective way to deliver educational messages in zoos). In a similar study, but of school children, Visscher et al. (2009) explored the role of interpretation to encourage learning about black rhinoceros (Diceros bicornis) training at Potter Park Zoo. The study involved three conditions: (1) a training session, coupled with keeper interpretation which verbally and physically engaged the audience (n = 20); (2) a training session with a fact only presentation and no attempt to engage the audience (n = 21); (3) a control condition whereby participants watched the training with no keeper presentation/interpretation (n = 26). Participants in the interpretive condition demonstrated more accurate knowledge immediately after the session than those in the other conditions, and those in the fact-only condition demonstrated more accurate knowledge than those in the control. Whilst this is promising, and the researchers made appropriate use of a non-parametric statistical analysis (given the small sample sizes per condition), these findings are based only a very brief, three-item quiz. Given that the quiz assessed content conveyed in conditions one and two only, it is not surprising these groups were shown to learn more than the control. Comparable to results from keeper-talk studies, Penn (2009) surveyed adult visitors (n = 150) immediately after they viewed and engaged with an actor dressed as a polar bear a form of zoo theatre at Central Park Zoo s polar bear exhibit, finding that they felt they had learnt more than those who viewed the exhibit without this interpretive element (n = 150). Although the type of knowledge questions asked were not specified, this study did report on differences in the type of knowledge gained. It was found that visitors in the interactive condition learnt about the natural history of the species, whilst visitors who simply viewed the exhibit reported learning associated with what they saw (e.g., that polar bears play with toys). This is useful for highlighting how the content of knowledge gained may differ depending on the nature of the visitor experience. However, some caution should be taken when interpreting these results, as these results 208

227 were obtained from a brief (<2 minute) survey and it is questioned how much data can be gained in such a short time-frame. In a study involving multiple interpretive elements, Broad and Weiler (1998) compared visitor learning at a traditional tiger exhibit (n = 74 groups), with minimal signage, to that at a theme park-based display (n = 64 groups), with various interpretive elements (including touch-tables and keeper presentations). Through qualitative interviews with groups of visitors (ranging from 1-12 individuals in each; ages unspecified) at the exhibits, it was reported that those experiencing the latter exhibit recalled a greater quality and quantity of cognitive (facts about tigers), affective (feelings about tigers) and behavioural (conservation) information immediately after their visit than those who viewed the traditional exhibit. Moreover, the majority of visitors at both sites commented that handler/keeper talks were responsible for their learning. However, these findings have limited support given that details of the methodology used to elicit this information were not provided. Whilst suffering from some shortfalls, such as limited number of questions used to assess knowledge and lack of longer term follow ups, these studies suggest that there is potential for interactive and interpretive experiences to enhance opportunities for zoo visitor learning, at least in the short-term Changes in visitor attitudes associated with traditional and interactive experiences Just as zoos hope to encourage learning, they also provide opportunities to foster appreciation, care and concern for nature among their visitors (WAZA, 2005). However, like knowledge gain, understanding and measuring attitudes in this environment is a complex task. By their very nature, zoo visitors might already possess, or at least report, quite positive attitudes toward animals (Clayton et al., 2009), and changes in attitudes may also differ depending on an individual s motivation for visiting (Falk et al., 2007). Individual factors, such as education level, self-rated knowledge, actual prior knowledge, and reading about animals for leisure, have also been shown to be associated with more positive attitudes towards animals and their conservation (Bitgood, 1992; Lukas & Ross, 2005). 209

228 In comparison to knowledge, less is known about the influence of a zoo visit on attitudes, as explored through either pre/post studies or between visitors in different conditions, despite a general belief that a zoo experience does contribute to changes in this domain. The peer-reviewed work of Adelman, Falk and colleagues (Adelman et al., 2000; Falk & Adelman, 2003) assesses the impact of the overall experience of visiting the National Aquarium in Baltimore, with pre/post/follow-up survey findings indicating that visiting can contribute to stronger attitudes in favour of conservation. Similar findings are reflected in an AZA report by Falk et al. (2007), in which a similar survey methodology was used to explore the overall impact of a visit on over 5500 visitors at 12 AZA-accredited zoos. While findings indicate that visitors have positive attitudes towards the role that zoos and aquariums play in conservation and education, the results have unfortunately not been published in the peer-reviewed literature. In contrast to these studies, which investigated general conservation attitudes, Lukas and Ross (2005) found no differences in attitudes toward African apes at Lincoln Park Zoo in Chicago when utilising a previously used attitude scale to compare visitors entering (n = 500) and exiting (n = 500) the exhibits. Given the paucity of attitude-related studies, it is difficult to compare the outcomes from each and to make a conclusion about the impact of zoo visits in this regard. However, other than methodological differences between the studies, the differences in findings might be accounted for by the focus of the research - the former studies related to attitudes to conservation (which might be sparked multiple times, at various exhibits, throughout the duration of one's visit) compared to attitudes to specific species (which might only be sparked once during a visit). Furthermore, studies of attitude change associated with interactive zoo experiences have been limited. Povey and Rios (2005) aforementioned (Section 6.1.3) observational exploration of the interpretive clouded leopard display at Point Defiance Zoo and Aquarium also included six rating-scale questions to assess attitudes (both specific to the animal and to conservation in general). Differences were found on two items only, with visitors who engaged with the interpretive element (n= 55) reporting stronger beliefs regarding both the quality of life and care that the animals received, compared to those who viewed the exhibit with no interpretation (n = 57). The former visitors also showed a trend towards having higher interest in clouded leopards than the latter, however relatively high attitudes (~4 on a 5-point scale) were reported across the items 210

229 regardless of condition. In the absence of pre- and post- experience data, it is difficult to determine causality between animal display type and attitudes. Moreover, it would have been interesting if this study had conducted the observational and survey components with the same visitor samples, in order to draw links between visiting behaviour and attitudes (e.g., whether those who sought more information reported more positive attitudes). Pearson et al.'s (2014) aforementioned (see Section 6.1.3) study of visitors at Melbourne Zoo used the previously validated, Animal Attitudes Scale (10 items), adapted to orangutans, to explore attitude change throughout the campaign. While visitors across all conditions reported very positive attitudes (>42 on the 50-point scale), those visitors in the latter three conditions all demonstrated more positive attitudes than those at baseline. This finding held true after controlling for visitor satisfaction with animal activity levels, which was found to relate to more positive attitudes. This is a promising finding, adding support to the role of interactive experiences in instilling positive attitudes towards animals. In a study using a different methodology - qualitative, post-experience telephone interviews - the work of Curtin (2006) and Curtin and Wilkes (2007) provides some further insight into attitude change related to participation in swim-with-dolphin programs. In particular, a key finding of one of these studies was that participants reported negative attitudes regarding the keeping of such animals in captivity (Curtin & Wilkes, 2007). Although not intending to explore attitudes specifically, this is an interesting finding as it highlights the potential for interactive experiences to impart attitudes that are not supportive of zoos (which is problematic for encouraging return visits). However, the very small sample size (n = 10), with participants who had visited a diverse range of aquariums worldwide, makes the findings difficult to generalise. There is a clear need for more research in this area before it can be understood how interactive experiences influence attitudes; at present, there is limited support to suggest that these experiences have the potential to contribute to positive attitude change. 211

230 Changes in visitor emotions associated with traditional and interactive experiences Some researchers have suggested that a zoo visit can enhance visitor emotions (e.g., both through stimulating emotional responses and fostering a sense of connection to animals), and that this might play an important role in increasing knowledge, attitudes, and/or behaviour change intentions (Ballantyne & Packer, 2005; Clayton et al., 2009; Luebke & Matiasek, 2013; Myers et al., 2004; Vining, 2003). In reviewing the literature related to conservation learning in free-choice settings (including zoos and aquariums), Ballantyne et al. (2007) argue that one of the key factors contributing to successful learning includes opportunities to engage visitors emotionally, however this has yet to receive notable support in the empirical literature. There are various ways that this might hold true. For example, if a visitor has an experience whereby they feel very strong emotions of concern for the survival of a species, they may be more likely to engage in a conservation behaviour that they perceive will assist the animal (e.g., donating to a conservation charity). On a different level, the emotion felt during an experience might influence a visitor's later recollections of that experience and subsequently their future behaviour. For example, if a visitor feels happy during their experience, they may remember the visit fondly and this might encourage them to visit again (another behaviour which aids conservation, due to zoo entry funds and opportunities for further conservation learning). These are just some examples to highlight importance of understanding emotional experiences in zoos, both in terms of the types of emotions felt and the valence of these emotions. Visitors have been found to experience a range of emotions including respect, happiness, peacefulness, connection, privilege, relaxation, interest, and caring when encountering various zoo animals (Ballantyne et al., 2011; Clayton et al., 2009; Marseille et al., 2012; Myers et al., 2004). Two of these studies have explored the concept of emotion in response to a general zoo visit (Clayton et al., 2009; Myers et al., 2004). Like changes in knowledge and attitudes, emotional responses have been shown to differ depending on other factors, such as the species experienced. For example, Myers et al. (2004) conducted a between-groups study of adult visitors (n = 279) who were asked to self-report the emotions they felt when viewing either the okapi (Okapia johnstoni), snake (multiple, unspecified species) or gorilla (species unspecified) exhibits 212

231 at Brookfield Zoo. Participants rated how they felt about the species on 17 uni-polar scales, and also reported on evaluative (e.g., happy/sad) and arousal (e.g., focused/distracted) emotions on seven bipolar scales. While there were no differences in emotional response to different species on seven items (respect, wonder, sympathy, surprised, concerned, angry, embarrassed), visitors who viewed okapi and gorillas demonstrated higher emotional ratings for 'sense of beauty', 'peaceful', 'special privilege', 'caring' and 'attracted' compared to those who viewed snakes. Conversely, those viewing snakes showed higher ratings of 'fearful' and 'disgust' compared to those viewing the other two species. Similar to what is known about knowledge and attitudes, this study also found that emotional responses were influenced by other factors, such as participant emotionality and relation to the animal (e.g., similarity of animal to human), as well as demographic variables, including age and pet ownership. In another investigation, Clayton et al. (2009) explored emotion as one variable, among others, that links to conservation support among visitors (n = 206) at four different locations within Cleveland Metroparks Zoo. To do this, three dichotomous emotions (tense/relaxed, happy/sad, interested/bored) were measured on a self-report rating scale. Whilst limited to a very small subset of emotions, the study did find positive correlations between emotions of happiness and interest and a reported sense of connection to animals. Additionally, those visitors who felt interested were more likely to report wanting to know more about the animal/species. Given the limited focus on emotions in the general zoo visitor literature, it is not surprising that the exploration of emotional differences between traditional exhibit viewing and interactive animal experiences appears to be limited to one study. Through a web-based survey distributed four-months after visitors (n = 240) attended one of four marine wildlife experiences (witnessing turtles nesting and hatchlings emerging; whale watching from boats in close proximity to the animals; typical aquarium and marinebased theme park exhibit viewing) in Queensland, Australia, Ballantyne et al. (2011) used four open-ended questions to understand visitor recollections of their experience. A qualitative analysis of the data identified emotional affinity (that is, reporting of emotional responses to the experiences or emotional connections to the animals) as one 213

232 of four key themes raised by participants. Emotions included feelings related to what visitors saw, as well as a sense of empathy and emotional connection to the animals. A quantitative analysis of the percentage of respondents from each experience who reported emotional affinity showed that this was significantly higher for those who had a non-captive experience compared to a captive one (turtle nesting = 29%; whale watching = 62%; aquarium = 17%; marine theme park = 21%). As such, in addition to viewing animals in their natural habitat, the potential for closer encounters with these wild animals may have contributed to a heightened emotional experience. Although it might seem logical that a closer/interactive encounter could generate a stronger emotional response (either positive or negative), more research is needed to determine the type and valence of emotions experienced by visitors who interact with zoo animals, and how these emotions might differ from visitors who only view animals in exhibits Changes in visitor behaviour associated with traditional and interactive experiences A desire to promote pro-conservation behaviours amongst zoo visitors is widely recongised as one of the ultimate aims of many zoological organisations. As with attitudes, knowledge and emotion, however, there are complexities associated with understanding exactly how zoos promote behaviour change. Engagement in a behaviour can be influenced by a variety of factors, including one s perceived ability to perform the behaviour (Ajzen, 2001; Saunders, Brook, & Myers, 2006) and one s self-identity (Fielding, McDonald, & Louis, 2008). Outside of zoo-based research, other studies of behaviour change suggest that situational variables (like socio-demographics) and psychological variables (such as motivations and subjective norms) can influence one s behaviours and/or behaviour intentions (Barr, 2007; Corraliza & Berenguer, 2000). Connectivity to nature (as measured, for example, by rating scale questions asking visitors to report how connected they feel to the natural world or to a particular species) has also been shown to have a positive relationship with environmental concern and environmental behaviour/behavioural intentions in various studies (Clayton et al., 2009; Dutcher et al., 2007; Hinds & Sparks, 2007; Skibins & Powell, 2013; Skibins, Powell, 214

233 & Hallo, 2013). Indeed, Mobley et al. (2009) found that environmental concern is a stronger predictor of environmental behaviour than knowledge or attitudes. Although it seems intuitive that greater knowledge and more positive attitudes and emotions would contribute to pro-conservation behaviour, the links between these constructs are still unclear in the zoo setting, perhaps because these connections are not always direct and are mediated by a variety of other factors (Ajzen, 2001; Ogden et al., 2004; Povey & Rios, 2005). Recently, however, Pearson et al.'s (2013) multi-zoo study provided evidence for a significant association between knowledge, attitudes and behavioural intentions. Studies of behaviour change among zoo visitors usually rely on measures of selfreported behaviour change intentions, rather than actions, and typically tend to focus on immediate, short-term intentions, rather than being longitudinal in design (see Section 3.4.1). Where longer term follow up has been undertaken, findings often reveal that behaviour change has not been long-lived (e.g., see Adelman et al., 2000). This may be associated with the ease of performing conservation behaviours, with Stoinski et al. (2002) finding that significantly more visitors (44% versus 2%; n = 93) at Zoo Atlanta performed conservation-related behaviour (signing a petition card related to the bushmeat crisis) when the option to do so was provided on-site (return the card whilst still at the zoo), as opposed to off-site (post the card to the zoo after visit). Behaviour change intentions have also been shown to vary depending on how much a visitor engages in a particular behaviour when they enter the zoo. For example, in a study exploring the impact of visiting a Conservation Station at Disney's Animal Kingdom, Dierking et al. (2004) surveyed visitors (n = 702) before and after their visit, as well as at follow-up (two to three months post-visit). Using their self-reported pre-visit behaviour, visitors were categorised into appropriate stages of behaviour change, according Prochaska's Model of Behaviour Change (widely used in the human health field; categories include: pre-contemplation, contemplation, preparation, action, and maintenance). A significant association was found between stage of behaviour change at entry and self-reported behaviour, such that those in the contemplation (thinking about changing behaviour) and preparation (committed to changing behaviour/making plans 215

234 to change) stages reported greater intentions to engage in behaviours after their visit. However, these changes were not sustained at follow-up. To date, an understanding about the influence of interpretive visits on conservation behaviour change is limited to three studies (Pearson et al., 2014; Smith et al., 2008; Swanagan, 2000). As with the other outcome variables (e.g., knowledge), this has typically involved asking visitors to self-report their behaviour. For example, Smith et al. (2008) explored behaviour change intention among visitors (n = 175) to a bird presentation at an Australian zoo, finding that 54% of participants reported that they would either start an action or increase their commitment to action toward bird conservation (particularly recycling or removing road kill from the road when driving). In a six-month follow-up telephone survey with 38 participants, three reported starting an action and 16 increased their commitment to an action since their visit. While this does indicate the potential for changes toward longer-term, post-visit conservation behaviour, this refers only to self-reported behaviours, with actual behaviour change not measured. It might also be the case that those visitors who agreed to participate in the follow-up were those who had a greater interest in bird conservation. Had it been possible to include all initial visitors, this proportion may have been lower. Similarly, Pearson et al.'s (2014) evaluation of the 'Don't Palm Us Off' campaign asked visitors to report their past conservation behaviour and willingness to change future behaviour (yes/no responses). A rise in self-reported conservation behaviours was found across the campaign. For example, visitors were asked if they had intentionally purchased palm-oil free products in the last 12 months, with increasing agreement as the study progressed (from 18.7% of visitors pre-campaign, to 38.9% of visitors postcampaign). Whilst a promising finding, perhaps visitors felt obliged to agree with this, given the explicit anti-palm oil messages of the campaign. Visitors were also given the opportunity to complete a petition card during the campaign, with almost 75% of visitors at 6 and 12 months reporting either having done this or intending to do this. Again, this finding is promising, but it must be considered that this behaviour was selfreported. While it is stated that the researcher made observations to confirm this behaviour, no detail was reported regarding how these observations were conducted. 216

235 Additionally, since the petition cards were not available during the pre- and postcampaign phases, conclusions could not be made regarding the impact of the program on this behaviour. Compared to these studies, which explore self-reported behaviour, Swanagan (2000) incorporated an on-site action component when investigating visitor responses to an interactive elephant (species unspecified) demonstration and/or bio-fact program at Zoo Atlanta, compared to visitors who only viewed the exhibit. Using an exit survey, visitors (n = 355) were asked to self-report their past and current experiences with elephants and were subsequently given solicitation cards, which, if returned, would indicate their willingness to support elephant conservation. Overall, 18.3% of participants returned these, with those who had a higher level of interaction with elephant exhibits (past and/or present) having a greater return rate. While this finding is promising, it is not stated how the researcher placed visitors into the different experience categories (five categories were listed - highest, high, undetermined, low and lowest). Therefore, understanding which interactive elements correspond with willingness to engage in this form of conservation behaviour is unclear. Although more research is needed, and caution should be taken when interpreting selfreported behavioural outcomes, these studies provide some support for the idea that interactive experiences can assist in increasing visitor behaviour change/intentions Interactive experiences with large felids With the exception of the studies by Curtin (2006), Curtin and Wilkes (2007), and Povey and Rios (2005), the aforementioned literature has focused on interactive visitor experiences at exhibits, rather than direct encounters with zoo animals. Moreover, these studies have rarely focused on large felids (with the exception of Broad & Wieler, 1998). There appears to have been only one exploration that has included a direct interactive experience with a large felid species, conducted by Weiler and Smith (2009). However, the focus of this study was on comparing the number of interpretive 217

236 experiences that visitors (n = 288) had with lions at Werribee Open Range Zoo, Australia, rather than comparing outcomes for visitors who had different experiences (including viewing the exhibit, listening to a keeper talk, interacting with a volunteer guide, watching an actor thematic interpretation, and attending a behind the scenes tour to visit the off-exhibit facilities and feed the lions). As might be expected, this study found that the more interpretation visitors interacted with, the higher their reported cognitive (e.g., thought provocation), affective (e.g., attitudes towards nature conservation) and behavioural (e.g., desire to participate in additional interpretive experiences) outcomes, as measured through the previously validated, 29-item Interpretation and Evaluation Tool Kit (see Ham & Weiler, 2005). Exploring ways to enhance the visitor experience with large felids is important from the perspective of conserving the species and the role that a zoo visit might play in encouraging visitors to engage in conservation behaviours in support of these animals. Although often among the most popular zoo animals, carnivores can often be found to be inactive, out-of-sight, or pacing in this environment (Powell, 1995). Therefore, although many visitors may have the opportunity to view large felid exhibits, the opportunity to tap into the affective component of visitor learning may be missed if these animals are inactive or out of view. Indeed, research has found that visitors show higher interest when felids are active (Margulis et al., 2003) and may be less interested in supporting conservation when they are seen pacing (Miller, 2012). Moreover, with large felids being exotic to Australia, zoos present the only local opportunity for people in this country to engage with these animals (except for travelling circuses), making it even more pertinent that we understand how to maximise the visitor experience if we are to encourage visitors to support these animals (e.g., through financial support of conservation funds, or decreased purchasing of products threatening natural habitats, such as palm oil). In this way, exploring the effectiveness of such programs is important for the conservation of the species. As large felids are often flagship species with the potential to assist the conservation of other animals (Ballantyne et al., 2007), the evaluation of such programs can also provide insight into conservation impacts beyond these species. Given the potential limitations and negative outcomes associated with increased large felid-human encounters (e.g., see Section ), it is vital to explore whether these experiences are having the desired influence on visitors, in order to 218

237 provide support for continued animal involvement, and to offer suggestions for improvements to current practice Aims A variety of Australian zoological organisations currently offer interactive large felidvisitor experiences (see Table 2.1), and the Australian Government Department of Agriculture, Forestry and Fisheries (2009) supports the use of interactive programs provided that visitors are offered a positive experience and leave with an improved appreciation of the animal they were interacting with. While it would be expected that a more interactive experience would have a more positive impact on these visitor outcomes, as detailed throughout Section 6.1, there is insufficient information available in the literature to date to make this prediction across all outcome variables. As such, this study takes an exploratory approach to investigate the impact of a behind-the-scenes tour. The overarching aim of this study was to explore whether interactive large felid experiences can be used to meet the proposed goals of zoos to increase knowledge about these species, encourage positive attitudes towards these species and their conservation, impart a positive emotional experience, and contribute to pro-conservation behavioural intentions. To do this, a within and between groups approach was taken, to allow for differences to be compared not only before and after an interactive large felid experience, but also to compare interactive experiences to traditional large felid exhibit viewing. Furthermore, a secondary aim of this study was to evaluate the visitor experience, in terms of whether it meets visitor expectations and how it could have been improved. This provides opportunities to make recommendations about successful experiences with large felids, which could be incorporated by other zoos, and to shed light on areas for improvement. 219

238 6.2. Method This study was approved by the University of South Australia Human Research Ethics Committee (project number P110/10) Sites and experiences As outlined in Section 3.4, this research took place at Zoos SA's two sites Adelaide Zoo (AZ; a city zoo) and Monarto Zoo (MZ; an open-range zoo). Participants either engaged in an interactive tour (including both tigers and lions at AZ; or cheetahs or lions at MZ), or viewed one of the large felid exhibits at either zoo (tiger or lion exhibits at AZ; or cheetah or lion exhibits at MZ) from the public access area (see Tables 3.6 and 3.7 for experience details) Participants A sample of 446 visitors participated in this study across all conditions: 121 visitors were involved in a pre- and post-interactive tour survey (AZ lion and tiger interactive, n = 41; MZ cheetah interactive, n = 40; MZ lion interactive, n = 40); 164 visitors were surveyed before viewing a large felid exhibit (AZ tiger, n = 42; AZ lion, n = 41; MZ cheetah, n = 39; MZ lion, n = 42) and 161 after viewing a large felid exhibit (AZ tiger, n = 40; AZ lion, n = 41; MZ cheetah, n = 37; MZ lion, n = 43). Participant response rates are displayed in Table 6.1. Participation was completely voluntary and participants were free to withdraw at any time. 21 It was initially intended that visitors would also be surveyed after watching a keeper talk at each of the exhibits. Whilst this would have been ideal in being able to compare a range of visitor experiences, this proved to be quite challenging logistically as visitors would often only watch part of a talk before leaving the area. These areas were also particularly crowded, making it difficult for one researcher to survey the required number of participants within the project timeline. Moreover, the talk at the Monarto Zoo lion exhibit tended to focus more on the other carnivore species (African painted dogs and hyenas), with little information given about lions specifically. Therefore, this study focused on interactive experiences and traditional exhibit viewing only. 220

239 Materials Two self-report questionnaires (pre-visit and post-visit) were generated to measure visitors knowledge of large felids, attitudes to large felids and their conservation, emotional reactions, and proposed conservation behaviour, as well as to evaluate their experiences. As discussed in Section 3.4.1, there is a lack of reputable scales for measuring these domains in the zoo visitor literature and the majority of studies tend to respond to this by purpose-designing survey items based on their unique study aims. As such, this approach was taken here and the literature outlined in Table 3.5 was used to determine the type and structure of questions commonly asked. Each questionnaire took approximately 10 minutes for participants to complete. A separate version of the questionnaire was created for each of the three species (tiger, lion and cheetah) and for each experience (tour and exhibit). For simplicity, the animals in the questionnaires will be referred to the species herein. In addition to demographic details, all items and questions related to analysis in this study are discussed here (for complete questionnaires, see Appendix K). Table 6.1. Participant response rate for each individual visitor condition in this study. Condition Approached Returned Response rate Lion and tiger interactive (AZ) * 96% Pre-tiger viewing (AZ) % Pre-lion viewing (AZ) * 95% Pre-cheetah viewing (MZ) * 80% Pre-lion viewing (MZ) % Post-tiger viewing (AZ) % Post-lion viewing (AZ) * 81% Post-cheetah viewing (MZ) % Post-lion viewing (MZ) * 53% Note: AZ = Adelaide Zoo; MZ = Monarto Zoo; * = numbers of returned questionnaires for these conditions are higher than the number of participants actually included in the data analysis due to missing data/incomplete questionnaires. A response rate could not be included for the MZ tours (cheetah and lion) since some of these questionnaires were distributed by zoo staff and response records were not kept Pre-visit questionnaires Species knowledge: To determine how much participants thought they knew about the species before their visit/encounter, they were asked to subjectively rate how knowledgeable they felt they were about the species ( How would you rate your knowledge about the species before visiting the zoo today, where 1 = I know nothing about the species and 5 = I am an expert on the species ). To assess actual knowledge, they were then asked five questions about the species. The content of these questions 221

240 was developed based on this researcher participating in the behind-the-scenes tours and viewing the signage at each exhibit, and subsequently selecting content that overlaps between the two, to ensure that visitors at all experiences had the potential to obtain this information. For the first two questions, visitors were asked to tick the major threats to the survival of the species and the main prey items of the species (of which there were multiple correct answers). For the last three questions, visitors were asked to correctly select the conservation status of the species, the social structure of the species, and the home country/ies of the species (of four choices, one was correct). Each of the five knowledge questions included an unsure response to reduce guessing. A combination of multiple choice questions containing both one and more than one correct responses have been used by other researchers (e.g., Lindemann-Mattheis & Kamer, 2005) and was deemed to be most appropriate given that many facts about large felids do not have one correct answer only. An overall knowledge score for each participant was generated by dividing the total number of correct selections by the total number of possible correct selections for all five questions (where a correct selection = selecting a right answer or not selecting a wrong answer); with a higher percentage reflecting more accurate knowledge. This was done to gain a more accurate understanding of visitor knowledge, by accounting for those participants who might have selected multiple responses for each answer if they were unsure. Attitudes toward species: To determine how participants felt about the species before their visit, they were asked four attitude-related questions. Participants were asked to rate, on a 5-point Likert-type scale, how strongly they like the particular species, how important the species is to them, how concerned they are about the species becoming extinct, and how important it is that the species is saved from extinction (where 1 = negative attitude and 5 = positive attitude). Scores for these four items were totaled to give an overall attitude toward species rating (possible range 5-20, with 20 representing the most positive attitude). This scale has good internal reliability (Cronbach s alpha = 0.84). Attitudes toward species conservation: Three rating-type questions (on a scale of 1-5, where 1 = negative attitude and 5 = positive attitude) were used to assess participants attitudes towards conservation ( How important is the conservation of nature to you? ) and their perceived level of control over conservation of the species ( How strongly do you believe that you can personally contribute to the conservation of the species? and 222

241 How strongly do you believe that people combined can contribute to the conservation of the species? ). These items were not combined for analysis, since each addresses a different construct of individual interest. Conservation behaviour: To assess performance of conservation behaviour, participants were asked how often (never, rarely, sometimes, often, or always) they engage in 10 different conservation behaviours: five general behaviours (such as attempt to minimise energy consumption ) and five species-specific behaviours (such as inform others about the species and their plight for survival ). Mean frequency of performance was totaled for the five behaviours in each of the two groups, giving a maximum possible score of 25 (with a higher score indicating more frequent performance) for general and specific conservation behaviours. Visit expectations: To determine participant expectations of their visit, participants were asked to rate their agreement with four statements: whether the experience would be enjoyable, whether they would learn about the particular species, whether the exhibit/encounter would be entertaining, and whether they would learn about species conservation (all on a 5-point scale, where 1 = strongly disagree and 5 = strongly agree). Since each individual item addresses a different construct, these items were not combined Post-visit questionnaire The post-visit questionnaire was identical to the pre-visit questionnaire, except that relevant questions were asked retrospectively (e.g., the wording of I think I will enjoy visiting this exhibit was changed to I enjoyed visiting this exhibit ). Some additional questions were added to various sections of this questionnaire: Species knowledge: To determine a change in knowledge, in addition to the subjective and multiple choice questions, participants were also asked (in open-ended questions) to recall two new things that they learnt about the species during their visit. Emotions: To gauge the emotional responses of visitors, two open-ended questions were included which asked participants to describe how seeing the species made them feel and how learning about the species made them feel. Although some previous studies 223

242 (e.g., Myers et al., 2004) have used rating-type questions to assess visitor emotion, by asking them to rate the level to which they have experienced a selection of emotions, open-ended questions were used here to determine the types of emotional responses felt since it is presently unknown which emotions are experienced by visitors during interactive encounters. Conservation behaviour: Instead of asking participants how often they engage in the different conservation behaviours, they were asked to indicate whether they would be more likely to begin/continue to engage in the same list of 10 behaviours after their visit (yes/maybe/no). Scores were again totaled separately for the general and speciesspecific behaviours, by totaling the number of behaviours for which a participant responded with yes to engaging in (maximum score of 5 for both groups of behaviours, with a higher score indicating a greater number of behaviours). To uncover barriers to performing conservation behaviour, visitors were also asked (in an openended question) to describe anything that might help them to engage in more conservation behaviour. Visit evaluation: In order to understand more about what visitors like and dislike from the different experiences, participants were asked three open-ended questions about their visit (best aspects, worst aspects, and anything else they would have liked to experience) Procedure Visitors were approached either before and/or after participating in an interactive encounter or visiting a large felid exhibit at either zoo from September 2010 through April Data collection occurred during regular zoo hours, over a variety of weekdays and weekends, as well as school and public holidays. All tour visitors who arrived at least 10 minutes prior to their tour starting time were invited to participate in the pre-visit survey before going on their tour; these visitors where then met after their tour to complete the post-visit questionnaire. These questionnaires were number coded to match the before and after responses. Pre-exhibit visitors were approached away from the large felid exhibits at both zoos and were required to have not yet viewed one of these exhibits; post-exhibit visitors were approached as they were leaving the large 224

243 felid exhibits and were required to have been seen to be viewing the exhibit. In this way, two separate groups of regular exhibit visitors (pre- and post-exhibit) were included in this study. Whilst all tour visitors were asked to participate (due to small numbers of visitors on the tours), only one visitor per group of pre- and post-exhibit visitors were asked to participate (to avoid similar responses). Visitors in these conditions were randomly selected - after the first visitor to cross an imaginary line had been approached for the session, all subsequent visitors were approached according to the next visitor to cross the imaginary line. Pre-tour and pre-exhibit questionnaires were required to be completed on-the-spot, whilst post-tour and post-exhibit visitors had the opportunity to return their questionnaires to a box at the zoo exits if they did not wish to finish the survey immediately following their experience. In all instances, participants were briefly informed about the nature of the study and, if they agreed to participate, were given a questionnaire and pen to complete it. All participants were required to be adults (i.e., at least 18 years of age) and speak English. All data were collected by this researcher, with the exception of the assistance of another researcher (fellow PhD student) for two days at the beginning of the study at AZ, and the assistance of zoo staff at MZ for the lion and cheetah interactive experiences Data analysis Since the overarching purpose of this investigation is to compare the experience of participating in an interactive large felid tour compared to viewing a large felid exhibit, data for these two experiences were combined regardless of species viewed (see Section 6.2.2). Visual inspection of variables for each individual species experience was conducted to confirm pooling of data within each major condition (pre-tour, post-tour, pre-exhibit, and post-exhibit; see Appendix L). Prior to data analysis, all data were screened visually and statistically both for missing variables and for normality. Any missing variables were checked against the original participant questionnaire sheets, as some may have been missed during the data entry stage. Where this was the case, these data were entered. No process was undertaken to replace data for item responses that were truly missing (i.e., item/s left blank by the participant). Since the majority of the outcome variable scores rely on tallying responses to multiple items (e.g., knowledge; see Sections and ), where a participants failed to answer a particular item, 225

244 their data were removed from the analysis related to that variable, but not from the study overall (for this reason, n values differ between, and hence have been reported for, all analyses). Mean values for each outcome variable were then screened for normality to determine appropriateness of conducting parametric analyses. Mean values were all normally distributed, except for the variable 'attitudes toward species', which demonstrated a notable negative skew (as discussed in Section , reflecting and applying a square root transformation normalised the data). Given that different visitors participated in the tour, pre-exhibit, and post-exhibit conditions, chi square analyses were conducted prior to subsequent analyses to determine whether there were significant differences between sample characteristics (as reflected in Table 6.2), to allow for control of these variables in subsequent analyses. To assess the influence of participating in a behind-the-scenes large felid tour on visitors knowledge about species, attitudes toward species, attitudes to conservation, and frequency/number of conservation behaviours, any differences in means within tour visitors (pre- and post-tour conditions) were analysed using paired sample t-tests (since the same participants were used in these conditions). To assess the influence of visiting a large felid exhibit on regular exhibit visitors knowledge about species, attitudes toward species, and attitudes to conservation, any differences in mean outcome variable scores between regular exhibit visitors (pre- and post- exhibit) were analysed using ANCOVAs, controlling for gender (since different participants were used in each condition and Table 6.2 reveals that these two groups differed on this demographic variable). ANCOVAs were similarly used to assess the difference between means for both groups of post-experience participants (post-tour versus post-exhibit) on knowledge about species, attitudes toward species, attitudes to conservation, and number of proposed conservation behaviours. In these analyses, gender, country of residence (Australia or overseas), zoo membership (member versus non-member) and reasons for visit (entertainment, learning, and time with people) were all controlled for (again, since Table 6.2 reveals that these two groups differed on these variables; prior experience with species was not controlled for since this was thought to be related to zoo 226

245 Table 6.2. Key demographic characteristics (n and %) and chi-square results of tests for differences between tour (n = 121), preexhibit (n = 164), and post-exhibit (n = 161) visitors. Tour visitors Pre-exhibit visitors Post-exhibit visitors n % n % n % χ 2 for tour v pre-exhibit visitors χ 2 for tour v post-exhibit visitors χ 2 for pre- v post-exhibit visitors Gender Male % % % * 8.25* Female % % % Age <20yrs 6 5.0% 5 3.0% 8 5.0% yrs % % % 30-39yrs % % % 40-49yrs % % % 50-59yrs % % % >60 yrs % % % Australian resident % % % 6.46* 7.94* 0.13 Zoo member % % % * 2.47 Zoo visit frequency First ever visit 4 3.3% 3 1.8% 0 0% Every few years % % % At least once/yr % % % At least once/6 mths % % % At least once/mth 3 2.5% 7 4.3% % > once/mth 3 2.5% 2 1.2% 4 2.5% Reason for visit Entertainment % % % 7.10* 10.33* 0.36 Learning % % % 21.22* 28.18* 0.62 Time with people % % % 18.84* 17.56* 0.03 Time with nature % % % Relaxation % % % Other % % %

246 Table 6.2. Continued. Tour visitors Pre-exhibit visitors Post-exhibit visitors χ 2 for tour v pre-exhibit visitors χ 2 for tour v post-exhibit visitors χ 2 for pre- v post-exhibit visitors Prior experience with species Other zoo % % % * 0.10 Behind-the-scenes tour 3 2.5% % % * 1.08 Hands-on tour 4 3.3% 9 5.5% % Seen in circus % % % Seen in sanctuary % % % Never seen before 7 5.8% 8 4.9% 5 3.1% Other 3 2.5% 6 3.7% 8 5.0% Pet owner % % % Education <High school % 4 2.4% 4 2.5% High school % % % Diploma % % % Trade 2 1.7% 3 1.8% 2 1.2% Tafe/certificate 4 3.3% % 9 5.6% Undergraduate % % % Postgraduate 7 5.8% % 4 2.5% Note: * = p <

247 membership, on the basis that zoo members may be more likely to have had more past experiences than non-members). Simple correlations were used to determine relationships between self-rated knowledge and knowledge scores for each participant group. Rates of conservation behaviour were compared between pre-tour and preexhibit visitors using ANCOVAs and controlling for country of residence and reasons for visit (entertainment, learning, and time with people). All statistical analyses were conducted in SPSS version 20 and were considered significant at the p <0.05 level. Only significant differences are reported (i.e., any non-significant p values are not reported). Similar to Chapter Four (see Section 4.2.4), prospective power analysis could not be calculated, given the absence of this information in previous zoo visitor literature and the use of unique outcome measures. As such, a retrospective power analysis was conducted. For any within-groups comparisons for tour visitors (n = 121), sufficient power was achieved to detect small effects (f = 0.13, 1-β = 0.80, α = 0.05). For the between-groups comparisons for pre-exhibit (n = 164) and post-exhibit (n = 161) visitors, sufficient power was achieved to detect small to moderate effects (d = 0.28, 1-β = 0.80, α = 0.05). Lastly, the between-groups analyses for tour (n = 121) and pre-exhibit (n = 164) visitors, and tour (n = 121) and post-exhibit (n = 161) visitors, sufficient power was achieved to detect small to moderate effects for both sets of comparisons (d = 0.30, 1-β = 0.80, α = 0.05). All open-ended responses were read and categorised into key themes, which were then totaled and the proportion of each response category was generated (similar to Wilson et al., 2003; Woods, 2002). Some participants responses included more than one comment and therefore fell into more than one category, hence in all instances frequencies were calculated based on number of responses in each category divided by the total number of comments (rather than number of participants). Responses obtained their own category if at least 5% of comments fell under the same theme; any remaining responses were placed into an other category. No statistical analyses were conducted for any of the open-ended questions, but trends in these response categories are reported based on the frequency of responses. 229

248 6.3. Results Demographic characteristics When combining visitors in each condition, over half of the sample was female, with almost 50% aged either or years. Participants typically reside in Australia, are pet owners, and approximately 1/3 were members of a zoo. The majority of participants reported visiting a zoo once every few years. Additional demographic characteristics and differences between participants in each condition are shown in Table Knowledge Self-rated knowledge Visitors in all score themselves as having moderate knowledge of the species (maximum score = 5; Table 6.3). Differences were observed within tour visitors (such that these visitors reported higher self-rated knowledge after their visit compared to before, t(116) = , p < 0.001), and between post-tour and post-exhibit visitors (such that post-tour visitors had higher self-rated knowledge than post-exhibit visitors, F 1, 265 = , p = 0.001). Table 6.3. Mean ± standard deviation self-rated knowledge scores for participants in each condition. Pre-tour Post-tour Pre-exhibit Post-exhibit Self-rated knowledge 2.52 ± ± 0.89 a* 2.82 ± ± 0.92 b* (n = 117) (n = 117) (n = 163) (n = 157) Note: a = pre-/post-tour comparison; b = post-tour/post-exhibit comparison; * = p <.001. n = number of participants who completed this item. Potential score range = 1 5, with a higher score indicating greater self-rated knowledge Responses to multiple choice questions The mean percentage of correct knowledge responses are presented for each condition in Table 6.4 (maximum = 100%). Regardless of condition, participants scored relatively highly on this variable. Scores only differed within the tour condition, such that these visitors had more accurate knowledge after their visit than before, t(110) = , p <

249 Table 6.4. Mean ± standard deviation percentage of correct knowledge responses for participants in each condition each condition. Pre-tour Post-tour Pre-exhibit Post-exhibit Knowledge score ± ± 9.49 a* ± ± (n = 116) (n = 115) (n = 162) (n = 157) Note: a = pre-/post-tour comparison; * = p <.001. n = number of participants who completed this item. Potential maximum = 100%, with a higher percentage indicating more accurate knowledge Correlations between self-rated and actual knowledge Weak correlations between self-rated knowledge and percentage of correct answers existed among post-tour, r =.25, p <.05, pre-exhibit, r =.19, p <.05, and post-exhibit, r =.19, p <.05, visitors only Responses to open-ended knowledge question A greater proportion of visitors in the post-tour group (83%) reported learning at least one new fact from their visit compared to post-exhibit participants (59%). Whilst there was a trend for some variation in the level of detail provided in responses across individual participants (e.g., for cheetahs, run fast compared to Cheetahs can run up to 100 km per hr ), visitors in both groups reported learning a range of new information (Figure 6.1). Regardless of condition, the majority of responses focused on behaviour of the species in the wild (e.g., Male lions eat first, followed by females followed by cubs ). Almost 20% of participants in both groups reported conservation related knowledge, primarily including the status of the species (e.g., They are critically endangered ), population size/features (e.g., Less than 500 wild tigers exist in Sumatra ), and threats to survival (e.g., Palm oil plantations threaten tigers via habitat destruction ), whilst others also commented on conservation efforts (e.g., Farmers are being paid to allow cheetah on their property aids conservation ). Further, information specific to the animals at Zoos SA (e.g., They love scented enrichment ) was the third most frequent knowledge category for both groups. 231

250 Percentage of responses 35% 30% 25% 20% 15% 10% 5% 0% Post-tour Post-exhibit Figure 6.1. Open-ended knowledge responses per category made by post-tour and post-exhibit visitors, as a percentage of total comments (261 comments by post-tour visitors; 194 by post-exhibit visitors) Attitudes Attitudes toward species Attitudes towards species were very positive irrespective of condition (pre-tour, M = 18.50, SD = 1.59; post-tour, M = 18.96, SD = 1.42; pre-exhibit, M = 18.00, SD = 2.12; post-exhibit, M = 18.57, SD = 1.69; maximum = 20) and there was a significant negative skew in this variable. Consequently, the data was reflected and a square root transformation applied to normalise the data (possible range = 1 16, with a lower score now representing a more positive attitude). Table 6.5 indicates that post-tour visitors had more positive attitudes after their visit compared to before, t(117) = 4.264, p < 0.001; post-exhibit visitors had more positive attitudes than pre-exhibit visitors, F 1, 311 = 5.71, p = 0.017; and post-tour visitors had more positive attitudes than post-exhibit visitors, F 1, 268 = 5.861, p =

251 Table 6.5. Mean ± standard deviation attitude toward species scores for participants in each condition. Attitude toward species Pre-tour Post-tour Pre-exhibit Post-exhibit 1.51 ± ± 0.45 a** 1.64 ± ± 0.51 b*c* (n = 120) (n = 119) (n = 161) (n = 160) Note: a = pre-/post-tour comparison; b = pre-/post-exhibit comparison; c post-tour/post-exhibit comparison; * = p <.05, ** = p <.001; n = number of participants who completed this item. Potential score range = 1-16, with a lower number representing a more positive attitude toward species Attitudes toward conservation Visitors in each condition reported high attitudes towards conservation, moderate attitudes toward their perceived ability to personally assist large felid conservation, and high attitudes toward their perceived ability for people combined to assist large felid conservation (maximum score for each statement = 5; Table 6.6). Tour visitors reported higher attitudes towards the importance of conserving nature in general after their visit compared to before, t(115) = -2.50, p = 0.014, as well as a higher belief in one s ability to personally assist big cat conservation, t(114) = -7.07, p < 0.001, and a higher belief in the ability for people combined to assist big cat conservation, t(113) = -3.64, p < Between regular exhibit visitors, pre-exhibit visitors had a higher belief in one s ability to personally assist big cat conservation compared to post-exhibit visitors, F 1, 312 = 2.864, p = When comparing attitudes between post tour and post regular visitors, the former had a higher belief in one's ability to personally assist big cat conservation, F 1, 264 = , p < 0.001, and a higher belief in the ability for people combined to assist big cat conservation than the latter, F 1, 264 = 7.572, p < Table 6.6. Mean ± standard deviation on the three attitudes toward conservation items for participants in each condition. Importance of conservation of nature in general Perceived ability for individual to assist large felid conservation Perceived ability for people combined to assist large felid conservation Pre-tour Post-tour Pre-exhibit Post-exhibit 4.49 ± ± 0.58 a* 4.52 ± ± 0.67 (n = 121) (n = 116) (n = 164) (n = 159) 3.36 ± ± 0.81 a** 3.62 ± ± 0.86 c** (n = 120) (n = 116) (n = 164) (n = 158) 4.18 ± 0.66 (n = 119) 4.41 ± 0.64 a** (n = 116) 4.12 ± 0.66 (n = 163) Note: a = pre-/post-tour comparison; b = pre-/post-exhibit comparison; c = post-tour/post-exhibit comparison; * = p <.05, ** = p <.001; n = 4.16 ± 0.73 b*c* (n = 158) number of participants who completed each item. Potential score range = 1-5, with a higher score indicating a more positive attitude for each item 233

252 Emotion Emotions related to seeing the animal/s Almost all (95%) of the post-tour and 80% of the post-exhibit visitors commented on how seeing the animal/s made them feel. Both groups reported similar emotions (Figure 6.2), with general positive emotions being the most frequent category, which including a variety of one-word responses such as happy, excited, wonderful, great, and terrific. The second most frequent comment related to positive emotions toward the animal/s, such as being in awe of them (e.g., Amazed at their power and strength, as well as being awestruck by them ). Both groups reported similar frequencies of negative emotions (e.g., Sad that we are destroying so many species, that the safest place is often far removed from natural range, sizes and habitat ). Post-exhibit visitors also reported negative emotions related to the animal/s being in captivity (e.g., Sad to see them in such a confined area. I always think of the male lion as a proud animal in charge of his pride. This feeling was not imparted today ), but comments of this nature were not reported by the post-tour participants. A greater proportion of responses by the post-tour visitors included feelings of privilege and gratefulness (e.g., Privileged to be so close to such powerful, special creatures ). Remaining categories in which there were minimal differences between participants in the two conditions included feeling interest/intrigued (e.g., Interested in tigers ), happy about the animals condition (e.g., great to see them in a great environment ), and scared/intimidated (e.g., Intimidated by their size/power ). The other category included a variety of emotions, such as a sense of pride in the work of Zoos SA, and excitement at the prospect of seeing large felids on a future visit to home countries. 234

253 Percentage of responses 50% 45% 40% 35% 30% 25% 20% 15% 10% 5% 0% Post-tour Post-exhibit Figure 6.2. Open-ended emotional responses to seeing animals/s per category made by post-tour and post-exhibit visitors, as a percentage of total comments (173 comments by post-tour visitors; 199 comments by post-exhibit visitors) Emotions related to learning about the animal/s Again, the majority (90%) of post-tour visitors commented on how learning about the species made them feel, compared to 51% of the post-exhibit participants. Despite this difference, the emotions described by both groups were similar in nature and were largely positive (Figure 6.3). For both groups, emotions related to feelings of increased knowledge were most frequently reported (e.g., More informed about them and the problems in their habitats ), followed by general positive emotions, like happiness and excitement. A slightly greater proportion of visitors in the post-tour group reported negative emotions associated with the current status and plight of the animals, such as anger towards the factors that are threatening them and sadness at their declining numbers (e.g., Sad mainly due to there (sic) environment changing due to us cutting down forests/jungles ); however, this group also more frequently reported emotions related to feelings of positivity towards conservation. There were minimal differences between participants in the remaining emotion categories, which included feeling interested/intrigued (e.g., Interested in what the keeper was saying ), privileged/grateful (e.g., Lucky to have access to the experience ), positive toward 235

254 Percentage of responses animal (e.g., impressive animal ), and positive toward conservation (e.g., Encouraged that we can help save them ). The other category included emotions such as surprise at what was being learnt, and enthusiasm about seeing the animals more often. 35% 30% Post-tour Post-exhibit 25% 20% 15% 10% 5% 0% Figure 6.3. Open-ended emotional responses to learning about animals/s per category made by post-tour and post-exhibit visitors, as a percentage of total comments (120 comments by post-tour visitors; 100 comments by post-exhibit visitors) Behaviour Entry level self-reported conservation behaviour. Figure 6.4 shows the self-reported frequency of performance of all ten conservation behaviours by visitors arriving at the zoo (pre-tour and pre-exhibit), indicating little variation between groups. When combining the frequencies of general and speciesspecific behaviours, both pre-tour and pre-exhibit visitors reported that they engage in general conservation behaviours at a moderate frequency and species specific behaviours at a low frequency (score range = 5-25; see Table 6.7). Between groups analyses revealed a difference only for species specific behaviours, such that pre-exhibit visitors reported engaging in these behaviours more frequently than pre-tour visitors, F( 1, 251 ) = 9.889, p = Within groups comparisons revealed that both groups of visitors engage in general conservation behaviours more frequently than species specific 236

255 Mean frequency of performance conservation behaviours: pre-tour visitors, t(115) = 22.19, p <.001; pre-exhibit visitors, t(147) = 14.63, p <.001. Table 6.7. Mean ± standard deviation self-reported frequency of conservation behaviours (general and species specific) for pre-tour and pre-exhibit visitors. Pre-tour General conservation behaviours ± 3.35 (n = 118) Species specific conservation behaviours ± 3.07 (n = 117) Note: a = pre-tour/pre-exhibit comparison; * = p <.05. ; n = number of participants who completed each item. Potential score range = 5-25, with a higher score indicating more frequently performed behaviours. Pre-exhibit ± 3.30 (n = 146) ± 3.77 a* (n = 142) 5 4 Pre-tour Pre-exhibit Figure 6.4. Mean frequency of self-reported performance of 10 conservation behaviours prior to their visit by pre-tour and pre-exhibit visitors (where 1= never and 5 = always) 237

256 Exit level self-reported conservation behaviour Figure 6.5 shows the variation in the percentage of leaving visitors (post-tour and postexhibit) that reported a willingness to engage in each of the ten conservation behaviours after their visit. Visual inspection of the individual behaviours indicates that a greater proportion of post-tour visitors reported a willingness to engage in each of the behaviours compared to post-exhibit visitors (see Figure 6.5). This was most pronounced for recycling, informing self about large felids, and informing others about large felids. Between groups comparisons were conducted to investigate differences in proposed number of conservation behaviours to be engaged in by both group of postexperience (post-tour or post-exhibit) visitors. As shown in Table 6.8, both groups reported that they would engage in a moderate number of general and species specific conservation behaviours after their visit (maximum number of behaviours = 5). Between group comparisons revealed that, compared to post-exhibit visitors, post-tour visitors reported that they would engage in more general, F( 1, 253 ) = , p <0.001, and species specific behaviours, F( 1, 247 ) = , p < Within groups comparisons revealed that both groups of visitors proposed to engage in more general conservation behaviours than species specific conservation behaviours: post-tour visitors, t(109) = 4.682, p <.001; post-exhibit visitors, t(142) = 4.91, p <

257 Mean percentage of visitors 90% 80% Post-tour Post-exhibit 70% 60% 50% 40% 30% 20% 10% 0% Figure 6.5. Mean percentage of post-tour and post-exhibit visitors reporting willingness to engage in 10 conservation behaviours after their visit Table 6.8. Mean ± standard deviation number of conservation behaviours (general and species specific) that post-tour and post-exhibit visitors proposed they will engage in after their visit. Post-tour General conservation behaviours 3.34 ± 1.38 (n = 114) Species specific conservation behaviours 2.75 ± 1.34 (n = 111) Post-exhibit 2.66 ± 1.58 a* (n = 149) 1.95 ± 1.40 a* (n = 146) Note: a = post-tour/post-exhibit comparison; * = p <.001; n = number of participants who completed each item. Potential number of behaviours = 1-5, with a higher score indicating a greater number of behaviours. 239

258 Percentage of responses Suggestions for enhancing species specific conservation behaviour In an open-ended response, 37% of post-tour and 32% of post-exhibit participants made suggestions about things that could assist them to engage in more species specific conservation behaviours (Figure 6.6), with the importance of increased education/information/awareness being the most frequent response (e.g., Continued exposure to information regarding what can be done at a basic level to assist with purchasing of products that are not made from items sourced from native habitats ). Other suggestions related to visiting zoos (e.g., Being closer to a zoo ), time (e.g., Having more time ), and money (e.g., Having more money would allow me to take more time off to volunteer ). The other category included various ideas, such as becoming a volunteer and living closer to the natural habitats of these species. 80% 70% Post-tour Post-exhibit 60% 50% 40% 30% 20% 10% 0% More education/ information/awareness Visiting zoos More time More money Other Figure 6.6. Open-ended suggestions per category made by post-tour and post-exhibit visitors regarding ways to assist engagement in species-specific conservation behaviours, as a percentage of total comments (51 comments by post-tour visitors; 56 comments by post-exhibit visitors) Experience evaluations Expectations and evaluations Expectations and evaluations of experiences were high regardless of condition (range = 1-5, with a greater score representing a higher expectation/evaluation). Differences 240

259 between visitors can be seen in Table 6.9. Tour visitors reported higher satisfaction on two of the four after their visit compared to their expectations before their visit: enjoyment, t(119) = , p < 0.001; entertainment, t(116) = , p < Posttour visitors also reported higher satisfaction than post-exhibit visitors on all four items: enjoyment, F( 1, 261 ) = , p <0.001; learning about species, F( 1, 255 ) = , p <0.001; entertainment, F( 1, 255 ) = , p <0.001,; learning about conservation, F( 1, 253) = , p < However, post-exhibit visitors reported lower satisfaction than pre-exhibit visitor expectation on three items: learning about species, F( 1, 303 ) = , p <0.001; entertainment, F( 1, 299 ) = , p <0.001; learning about conservation, F( 1, 300) = , p < Table 6.9. Mean ± standard deviation expectation and evaluation ratings made by visitors in each condition. Pre-tour Post-tour Pre-exhibit Post-exhibit I will/did enjoy visiting the exhibit 4.79 ± 0.47 (n = 121) 4.95 ± 0.22 a** (n = 120) 4.48 ± 0.70 (n = 163) 4.44 ± 0.68 c** (n = 152) I will/did learn new things about the species 4.68 ± 0.49 (n = 119) 4.69 ± 0.60 (n = 116) 4.05 ± 0.90 (n = 162) 3.70 ± 1.05 (n = 150) I will be/was entertained 4.62 ± ± 0.29 a** 4.32 ± ± 0.97 (n = 119) (n = 119) (n = 161) (n = 147) I will/did learn about 4.50 ± ± ± ± 0.99 species conservation (n = 119) (n = 118) (n = 163) (n = 146) b**& c** b**& c** b** & c** Note: a = pre-/post-tour comparison; b = pre-/post-exhibit comparison; c = post-tour/post-exhibit comparison; * = p <.05, ** = p <.001; n = number of participants who completed each item. Potential score range = 1-5, with a higher score indicating more positive expectations/evaluations Qualitative evaluations Almost all (98%) of post-tour visitors commented on a best aspect of their visit, compared to 70% of post-exhibit viewers. Figure 6.7 shows some differences in this response. For post-tour visitors, being in close proximity to the animal (e.g., That the lion was close to the fence ) and animal feeding (e.g., Feeding the lions ) were the highlights. For post-exhibit visitors, seeing a particular behaviour (e.g., When one female [lion] stalked a bird ) was the most frequently reported highlight. Other best aspects included the nature of exhibit (e.g., I like the exhibit ), seeing a particular behaviour, activity (e.g., Seeing them move around ), visibility (e.g., That they were visible ) and the information provided (e.g., Some information presented on lion conservation ). Comments in the other category included a range of aspects, such as good photographic opportunities and participants children enjoying their experience. 241

260 Percentage of responses 60% 50% 40% 30% 20% 10% 0% Post-tour Post-exhibit Figure 6.7. Open-ended responses per category made by post-tour and post-exhibit visitors regarding the best aspect/s of their visit, as a percentage of total comments (note: 166 comments by post-tour visitors; 134 comments by post-exhibit). Whilst 60% of post-exhibit visitors commented on a worst aspect of their visit, only 1/3 of post-tour visitors did so (Figure 6.8). For the post-tour visitors, the most frequent negative comment was that there was not enough time spent with the animals (e.g., Not long enough ). These participants were also concerned about not being able to be closer to the animals (e.g., [The cheetahs] were too far away ), and insufficient interaction (e.g., I couldn t give one a cuddle ). Participants in this group also made a number of individualistic comments (placed in the other category), such as the smell of the animals and/or their food. For the post-exhibit visitors, the most frequent comments were the poor visibility of animals (e.g., Could not see the tiger at all ), concern for the animals being in captivity (e.g., Lion enclosures do not seem very stimulating for the lions, enclosures are small ), and animal inactivity (e.g., They weren t moving much ). These visitors also commented on there not being enough animals (e.g., Only one tiger out and about this morning ). Another difference between the two groups is that the post-tour visitors also commented on the weather/environmental conditions (e.g., It started to rain at the end ) more than the post-exhibit visitors. 242

261 Percentage of responses 45% 40% Post-tour Post-exhibit 35% 30% 25% 20% 15% 10% 5% 0% Figure 6.8. Open-ended responses per category made by post-tour and post-exhibit visitors regarding the worst aspect/s of their visit, as a percentage of total comments (44 comments by post-tour visitors; 117 comments by post-exhibit visitors). A lower number of visitors in the post-tour group (22%) made comments about what else they would have liked to experience, compared to post-exhibit viewers (44%). As shown in Figure 6.9, post-tour visitors reported being most interested in being offered more information (e.g., Information about other species of tiger ) and in being able to touch the animals (e.g., Being able to pat them ), whilst post-exhibit visitors primarily reported wanting to witness an animal feeding session (e.g., Seeing them being fed would have been exciting ), to see particular behaviours (e.g., See them run ), and better visibility (e.g., If they were a little more visible ). Remaining categories for which there were minimal differences between participant conditions include the ability to see cubs (e.g., Tiger cubs which would demonstrate Adelaide Zoo s contribution to growing/maintaining the population ) and better activity (e.g., seeing them more active ). The other category included various comments, such as suggested improvements to exhibit styles to enhance the visitor experience. 243

262 Percentage of responses 35% 30% 25% 20% 15% 10% 5% Post-tour Post-exhibit 0% Animal feeding More information Touch animal See particular behaviour Better visibility To see cubs Better activity Other Figure 6.9. Open-ended responses per category made by post-tour and post-exhibit visitors regarding other desired aspect/s of their visit, as a percentage of total comments (30 comments by post-tour visitors; 85 comments by post-exhibit visitors) Discussion There is a growing body of research to support the importance of a zoo experience for the visiting public. To date, however, there has been a lack of information about how behind-the-scenes interactive tours fit within this literature and what the impact of these experiences might be. With such tours having the potential for negative impacts on the animals involved (such as increased stress, as discussed in Section ), it is pertinent to explore whether tours are meeting zoo goals of providing an enjoyable and educational experience for visitors. The current study addressed this by investigating the influence of behind-the-scenes tours on self-reported visitor knowledge, attitudes, emotion and proposed behaviour. Overall, the findings have provided some initial - albeit limited - support for the benefits of such tours, both for visitors and for zoo initiatives. 244

263 Knowledge Knowledge was assessed through the dual method of multiple choice questions, as well as an open-ended question asking participants to list two new things they had learnt about the species. Responses to the former revealed an increase in knowledge within tour visitors (pre-tour compared to post-tour), indicating that tours may be successful in promoting knowledge gain. Although a significant difference was found, the percentage of correct responses only increased by 4% after tours compared to before, which does not translate into a very meaningful difference considering that only five multiple choice questions were asked. With this knowledge measure only assessing a relatively small amount of visitor learning, these results should be interpreted with caution. Further, contrary to findings of other studies investigating interpretive zoo experiences (Broad & Weiler, 1998; Lindemann-Matthies & Kamer, 2005; Penn, 2009; Povey & Rios, 2005; Visscher et al., 2009), there were no differences between post-tour and postexhibit visitors, suggesting that perhaps these tours are not maximising their opportunity to educate visitors. Ross and Gillespie (2008) have found that visitors spend more time looking at animals rather than engaging with interpretive elements and perhaps this is the case during interactive tours where, given the notable amount of money they have paid for the experience, visitors may be more likely to focus on the animal/s rather than anything else. Complementing the multiple choice responses, over 20% more post-tour visitors (85%) than post-exhibit visitors (59%) were able to report at least one new fact learnt during their experience, suggesting that perhaps tour visitors absorbed different information during tours than what was measured by the five questions. The new information that either group of visitors reported typically focused on the behaviour of the animals in the wild, followed by conservation-related facts. It is interesting that the same proportion of comments by visitors in each condition (approximately 16%) related to conservation knowledge. Whilst this is an important finding, as it highlights that visitors can learn about conservation from either experience, it does indicate that perhaps tours need to do more to facilitate learning in this regard. Although it has been acknowledged there are difficulties inherent in measuring learning in free-choice contexts, such as zoos (Ballantyne & Packer, 2005), it is widely recognised that knowledge can play in important role in contributing to behaviour change (e.g., Pearson et al., 2013) and 245

264 support for protecting animals (Clayton et al., 2009), so it will be important for future research to explore how to enhance the provision of this information during interactive experiences Attitudes Visitors in all conditions reported very positive attitudes toward large felids. Analyses revealed an increase in positive attitudes between pre- and post-tour visitors, between pre- and post-exhibit visitors, and between post-tour and post-exhibit visitors, indicating that either experience with a large felid can result in increased attitudes, but even more so for those participating in an interactive tour. These findings are similar to those of Povey and Rios (2005). Whilst this appears to be a very positive finding in support of interactive tours, particularly considering the important role attitudes can play in behaviour change (Ajzen, 2001), attitudes were already very high to begin with among these participants (which is likely to be expected in a purely zoo-based sample; Clayton et al., 2009) and so the change found was quite minimal. Whilst this may not translate into a very meaningful difference, any increase in attitudes is a positive finding. Being a preliminary investigation of interactive large felid experiences, there are some methodological issues that could be addressed if future research aims to validate the present findings. In particular, a validated attitude scale, such as the well-known New Ecological Paradigm or Animal Attitudes Scale, may be more sensitive in detecting change in attitudes and reducing reporting bias. However, such a scale was not employed in this study due to the absence of validated large felid-specific attitude measures (as discussed in Section 3.4.1), as well as a desire to keep each section of the questionnaire brief (so as not to burden zoo visitors or interrupt their visit). Findings were mixed in regard to visitor attitudes toward conservation. Of note, tour visitors showed an increase in all three items after their experience compared to before. Moreover, post-tour visitors reported more positive attitudes toward their ability to personally assist large felid conservation and the ability for people combined to assist large felid conservation compared to post-exhibit visitors. This suggests that the tours contributed to visitors feeling more positive about the role that people can play 246

265 individually and collectively in large felid conservation. This is consistent with the findings of Balmford et al. (2007) who show that a zoo visit has the potential to allow visitors to have a better idea about the difference that they can make to conservation. Since perceived behavioural control is also thought to contribute to behaviour change (Ajzen, 2001), this finding is promising. As with attitudes to species, however, scores on these variables were all quite high to begin with, so any change detected was minimal and findings should be interpreted with caution Emotions The open-ended responses reveal some unique findings about the range of emotions experienced by zoo visitors. Past research has typically ignored what zoo visitors feel (with the exception of the work by Ballantyne et al., 2011; Clayton et al., 2009; Marseille et al., 2012; Myers et al., 2004), yet in the present study almost all tour visitors (95%) reported feeling an emotion related to seeing the animal/s, along with 80% of regular exhibit visitors, indicating that visitors to zoos do experience emotions related to their experience and that these might be more prevalent during interactive experiences. Visitors in both conditions largely reported general positive emotions (such as happy or excited). Additionally, tour visitors more frequently reported feeling privileged/grateful for their experience, as well as specific positive emotions towards the animals. Such emotions are similar to those reported in response to range of species (e.g., marine species, snakes, primates, and polar bears) in past research (Ballantyne et al., 2011; Clayton et al., 2009; Marseille et al., 2012; Myers et al., 2004). In a similar trend, more post-tour visitors (90%) reported how they felt when learning about the animals, compared to only half of the post-exhibit visitors. This is consistent with the lower knowledge scores and open-ended responses for post-exhibit visitors compared to post-tour visitors, and with the enhanced potential for visitors to learn during tours. Given the small number of visitors permitted on the Zoos SA tours and the presence of at least one zoo keeper and volunteer guide, these visitors have a greater opportunity in which to ask questions about the species than do those at exhibits. Over 50% of the comments made by visitors in either group were that they felt more 247

266 knowledgeable or interested after their experience, which adds further support for the capacity for a zoo experience to encourage visitor learning. The remainder of reported emotions were largely positive in nature, with only around 10% of comments made by either group reflecting negative emotions (such as anger at what the visitor was learning or sadness for the animals). This is likely to reflect the type of information available on signage for exhibit visitors (usually relating to more neutral information, such as feeding and activity patterns in the wild) and information presented by keepers on tours (which is usually more personalised to the individual animals, such as their characteristics and behaviours), as reflected in the open-ended knowledge responses. The trend toward greater reporting of positive compared to negative emotions related to seeing and learning about large felids is consistent with past research. For example, Clayton et al. (2009) discuss how close encounters have the potential to personalise animals, which may promote empathy among zoo visitors. However, these authors caution that personalising animals may contribute to visitors seeing them as less representative of the particular species which, in turn, might reduce concern for the species (Clayton et al., 2009). In this way, if zoos are to successfully encourage conservation action among their visitors, it would seem logical that they also need to devote a significant proportion of attention to the state of the species in the wild in order to impart feelings of concern. Although striking this balance between connection and fear/worry is difficult (Vining, 2003), it has been argued that informal learning experiences need to target the emotional aspects of learning in order to be successful, particularly since emotional arousal can motivate learning (Ballantyne & Packer, 2005; Clayton et al, 2009). For example, Myers et al. (2004) found a positive correlation between emotional reaction (including, but not limited to, caring, love, and sense of connection) to viewing animals and a visitor s desire to save the species. Similar findings have been found in other informal learning settings (Falk & Gillespie, 2009). In this way, the findings of the present study tend to indicate that tours at Zoos SA are enabling an emotional connection to be established, although it is yet to be determined how this connects to conservation behaviour change. 248

267 Behaviour One of the key limitations of this study was that the measurement of the behaviour items was different for the pre- and post-experience questionnaires, since preexperience visitors were asked to rate how frequently they engaged in each behaviour prior to their visit, whilst post-experience visitors were asked to rate how likely they would be to engage in each behaviour after their visit. This meant that the frequency of behaviours could not be compared between pre- and post-visitors. Regardless of this limitation, the findings revealed that, irrespective of condition, general conservation behaviours are reportedly more popular than large felid specific conservation behaviours. This finding is not particularly surprising, given the relative ease of the former (such as recycling) compared to the relative difficulties of the latter (such as volunteering with a large felid conservation project). Understanding how to encourage behaviour change related to large felids will be important if tours are seen to be successful. Participants reported some barriers to engaging in these behaviours. Although only approximately one third of participants (37% post-tour and 32% postexhibit) responded to this aspect of the questionnaire, the majority of these individuals discussed the importance of having more knowledge or information about what they can be doing, indicating that they, too, are aware of the link between knowledge and behaviour. Moreover, the need for more time and money were also commented on, indicating some additional barriers to behaviour change. Of further importance was that post-tour visitors reported that they would engage in a greater number of both general and large felid specific conservation behaviours after their experience compared to post-exhibit visitors. Although this refers to proposed rather than actual behaviour, this finding is promising as it indicates that tours are communicating a message about the need for more action to assist large felid conservation. It would be useful for future research to explore this further, possibly with a measure such as a behaviour diary to track what individuals do when they leave the zoo. Although such a method would still rely on participants truthfully reporting their behaviours, it could provide a more informed understanding of which behaviours are performed. Alternatively, an opportunity to perform a behaviour prior to leaving the zoo, such as signing a petition to support conservation (e.g., see Pearson et al., 2014; Stoinski et al., 2002; Swanagan, 2000) could achieve this aim. 249

268 More research is needed regarding actual behaviour and the long-term nature of any potential changes. Ballantyne and Packer (2005) have argued that many changes in conservation behaviour often fail to be sustained, and that this could in part be due to the lack of opportunity to perform them. Understanding how better to both provide visitors with opportunities for conservation behaviour and remove the aforementioned barriers would be important. For example, one method could be to provide visitors with options about behaviours that they could engage in (Ballantyne et al., 2007). With an understanding that knowledge is a barrier, coupled with the fact that Australian zoo visitors are far removed (spatially) from the source of large felid problems (which has been previously recognised as a barrier to conservation behaviour; Ballantyne et al., 2011), perhaps one avenue for improvement could be to provided visitors with an information sheet containing suggestions for behaviours and how these could be achieved (such as details of a large felid conservation fund and how time/money can be donated to it). Another suggestion could be to continuously provide information after the visit. Hughes, Packer, and Ballantyne (2011) found that providing visitors (n = 100 families) with post-visit action resources (access to materials as well as online support and contact via s and forums) after a wildlife experience resulted in them being significantly more likely to report a change in conservation behaviour (frequency of picking up litter) compared to a control group (n = 100 families) who received no support. Moreover, it might also be the case that visitors feel overwhelmed after a visit to a zoo given the varied number of conservation threats facing animals, and this may contribute to the difficulty in deciding which behaviours to perform. Perhaps it would be useful for zoos to focus on a small number of specific behaviours to educate visitors about during the tours Experience evaluations Measuring visitor satisfaction is important to understanding how experiences can be improved (Mason, 2000), particularly since visitors are vital to the operation and success of zoos (Hosey et al., 2009). In the present study, all visitors had quite high expectations and evaluations of their experiences. Some key findings were that posttour visitors rated higher enjoyment and entertainment regarding their experience than 250

269 they did pre-tour, as well as higher evaluations on all items compared to post-exhibit visitors. Moreover, post-exhibit visitors reported lower evaluations on most items compared to pre-exhibit visitors expectations, indicating that the exhibits are not meeting visitor expectations. Combined, these findings add support to behind-thescenes tours being both educational and enjoyable for visitors, and hence meeting the aims of zoos to provide a positive visitor experience in this regard. The open-ended responses about positive and negative visit aspects are additionally useful in highlighting what visitors wish to gain from their experience (whether interactive or at exhibits) and whether these desires align with what zoos are currently doing/what is possible to do. Consistent with evaluations of other wildlife experiences (Woods, 2002), post-tour visitors most frequently reported that being in close proximity and/or being able to interact with the animal/s was the highlight of their visit, followed by the opportunity to feed an animal. In contrast, seeing the animal/its behaviour was the highlight for post-exhibit visitors. These findings are quite logical, since tours provide opportunities for closer interactions, whilst exhibits offer opportunities for animal viewing. Moreover, almost all post-tour visitors (98%) commented on the best aspect of their exhibit, compared to 70% of post exhibit visitors, which could indicate greater satisfaction among the former. A small proportion (30%) of post-tour visitors reported a worst aspect of their visit, with the most frequent comment being that they wanted more time with the animals. In many ways, this is a positive finding since it implies that visitors enjoyed spending time with the animal/s and wanted more, which is consistent with findings from captive dolphin interactive experiences (Curtin & Wilkes, 2007). In comparison, over half (60%) of post-exhibit visitors reported on the worst aspect of their experience, with comments frequently related to poor animal visibility and activity, as well as concern for the animal/s being in captivity. Such comments are also not unexpected, since large felids can be inactive and/or out of sight in zoos (Powell, 1995; Van Metter et al., 2008), yet activity and good visibility have been reported as two of the key elements visitors want when they view zoo animals (Margulis et al., 2003; Whitworth, 2012). Perhaps one way to improve this could be to provide signage at exhibits to explain to visitors that resting 251

270 is a typical behaviour for these species; some suggestions about where the animals might be hiding could also be useful in encouraging visitors to try to locate them. These strategies may reduce disappointment associated with poor activity and visibility (Woods, 1998). Moreover, both zoos in this study conduct regular keeper presentations at the exhibits, which do provide visitors with opportunities to see the animals more active and in greater proximity. These presentations are advertised to visitors through zoo brochures (collected at the entry gate) and/or signage at the exhibits, so there are strategies in place already to tackle these suggested areas for improvement. A small proportion (20%) of post-tour visitors reported other things that they would have liked to experience during their visit. In itself, this could again indicate that most of these visitors were satisfied with their experience. The main comment made by these participants included wanting more information on the tours, which is very positive since it is suggestive of a desire to learn more and is consistent with past research on visitor evaluations of whale tours in natural habitats (Luck, 2003). However, 25% of comments made by post-tour visitors in the current study related to a desire to touch the animals. Although this comment relates to only a handful of post-tour visitors (most likely those who did not attend the cheetah tour as hands-on contact is available to those visitors), and may be reflective of a desire to be close to the animals and a strong liking of them, it is important for zoos to be cautious of the message they are sending to their visitors about the dangerousness of these animals (as reported by keepers in Table 4.5). For example, more emphasis could be placed on explaining to visitors that a protected contact approach is taken because these animals have not been hand-reared and possess natural hunting instincts, which could be detrimental to both the humans and animals involved in an interaction. Moreover, it could be explained that a protected contact approach is taken to preserve natural large felid behaviour (as much as is possible in the zoo environment). In comparison, almost half of the post-exhibit visitors made comments about other things they would have liked to experience, indicating that not as many of these visitors were satisfied with their experience compared to tour visitors. Over 25% of these responses related to a desire to see the animals feeding, which has long been a highlight for visitors (Kalof, 2007). Moreover, consistent with the reported negative aspects of their experience, these visitors also commented that they would like better animal visibility and to see particular behaviours. Again, these factors 252

271 are being provided for during keeper talks (with the exception that feeding is not available at the lion exhibit at Monarto Zoo). Perhaps there needs to be greater promotion of keeper talks at these zoos to ensure that the experience of visitors at exhibits is maximised Limitations/future directions The limitations of this study should be acknowledged. As discussed in Section 3.4.1, a key limitation to furthering our understanding of the influence of zoo visits is the lack of valid and reliable measures of zoo visitor outcomes (e.g., knowledge). As such, like many previous zoo visitor studies, the findings of this study are limited by the use of purpose-designed items. In the absence of appropriate items specific to large felid experiences, this study used multiple question styles (open and closed responses) to capture as complete an understanding of outcomes as possible. The items are also limited in that the questionnaire was intended to be brief (so as not to interfere with the visitors' experiences), and therefore more valid findings may come from longer scales. For example, a better understanding of the type of knowledge gained could come from a scale with more than five items. Another consideration when interpreting the results is that data were pooled for visitors to each large felid experience or exhibit. Whilst this was done for the purpose of increasing the sample sizes within tour and exhibit conditions to allow for statistical comparison (as discussed in Section 6.2) to meet the aims of the study, the findings must be taken with caution as there might be differences with each species, experience and exhibit type which could influence how visitors respond. Within the time frame and resources of a PhD program, it was not possible to collect more data in order to be able to separate the analyses for each species and experience type because data collection was a lengthy process. For example, a small number of visitors were permitted on each tour (maximum of 6, depending on the tour type) and often many of these would arrive just prior to their tour starting (i.e., within a few minutes), making them unable to participate due to time constraints. As such, it took approximately three months of distributing surveys on every available tour in order to collect 40 participants in each of 253

272 these conditions. A larger-scale project with greater funding would result in more data being collected, allowing analysis at the individual tour/exhibit level for each species. Another limitation is that the present study surveyed two separate groups of visitors before and after attending a large felid exhibit. This was conducted to meet the requirements of Zoos SA to provide minimal intrusion on visitor experiences, as it was acknowledged that visitors usually only spend very small amounts of time viewing exhibits (for example, Clayton et al. (2009) found that mean stay time at a range of exhibits at various zoos was seconds). Consequently, it was not possible to survey the same participants before and after a typical exhibit visit. Although any demographic differences between the two groups were controlled for in all between-groups statistical analyses, having the same participants in each condition would have allowed for a more accurate comparison of the experience of individual visitors and stronger conclusions about the influence of interactive tours. This approach is consistent with past zoo visitor research (see Section 3.4.1), but is something that could be improved on in future investigations. Perhaps the most significant limitation of this study, and most important direction for future research, is that conservation behaviour change after the visit was not measured or predicted from the variables assessed. The key purpose of this study was to assess the value of interactive large felid tours on zoo visitors in terms of knowledge, attitudes, emotions and proposed behaviour change intentions, since these are the variables most commonly explored in zoo visitor studies (see Section 3.4.1). To this end, the present study has provided some preliminary evidence to show that an opportunity to participate in interactive experiences can have a positive influence on zoo visitors. However, since one of the ultimate goals of zoos is to encourage engagement in conservation behaviours among the visiting public, it would be useful for future research to predict conservation behaviour from these four variables. In order to do this, subsequent research would benefit from using validated measures of the four constructs in order to determine whether such tours are having the desired impact. This might also assist in addressing the potential for a positive reporting bias which may have been evident for many of the variables in this study. Although there is an absence of validated measures 254

273 specific to large felids (see Section 3.4.1), the findings from this study - particularly regarding the type of information visitors learn and the emotions they feel - could be used to develop such measures. Moreover, it would be pertinent that future research employs measures to gauge actual behaviour change before drawing conclusions about the overall efficacy of behind-the-scenes tours. Without this information, there will be an incomplete picture of the role of tours and limited justification for their continued practice. Since learning is a cumulative process and a single zoo visit represents just one part of an individual s lifelong learning experience (Ballantyne & Packer, 2005), understanding the impact of interactive tours will require a broader picture of what visitors have done before their visit and what they do once they leave. A more comprehensive understanding of past experiences that visitors have had with the species (and possibly other animals) and what types of prior learning they have engaged in, as well as tracking post-zoo learning, would allow for a greater recognition of the role of behind-the-scenes tours. For example, it may be that knowledge gain immediately after tours is quite limited, as visitors may be more interested in focusing on the animal/s. This focus could be responsible for a heightened emotional experience and a connection to the animal/species, which could prompt self-directed learning after the visit. If this is the case, then zoos could direct their attention toward how best to encourage this connection. One avenue which was not explored in the present study could be to determine which level of contact with an animal (hands-on or protected) is most successful in facilitating this connection. Understanding these effects would require a longer-term study, with various follow-up phases, which could be possible with greater research resources Conclusion This study has provided promising findings to support the use of behind-the-scenes large felid tours to encourage knowledge gain, attitude change, emotion stimulation, and proposed conservation behaviour change among the visiting public. The findings suggest that an interactive experience can be more beneficial in meeting zoo aims 255

274 compared to traditional exhibit viewing. Whilst it will be essential for future research to validate these findings, this study does indicate that the experiences currently offered at Zoos SA are meeting both the needs of zoos and those of visitors. This chapter presented the final research component of this thesis; to follow is an overall summary of findings and discussion. 256

275 7 Discussion and conclusion Ultimately we may know enough about the dynamics of human-animal interactions in the zoo context to be able to ensure the best welfare of the animals while still providing a positive and rewarding experience for both the people who work with those animals and the zoo-visiting public on whom most zoos depend. G. Hosey (2008, p. 123) The previous chapter discussed the final phase of research conducted as part of this thesis. This chapter provides a summary of the thesis and the key findings obtained, a discussion of problems encountered and lessons learned, as well as implications and suggested directions for future research 7.1. Summary of thesis A narrative review of the scientific environmental enrichment literature was initially conducted (Chapter One) to highlight the current state of knowledge regarding keeping large felids in zoos. This investigation revealed that, whilst we know quite a lot about enrichment techniques that target natural feeding and hunting-related behaviour, some newer areas of practice, and hence research, are less well understood. One such area is the role of human interaction, with it currently unclear whether this might be a positive, neutral, or negative experience for zoo-housed large felids. In response to this lack of knowledge, the following chapter (Chapter Two) provided an overview of is known about large felid-human interactions and relationships in human history, as well as more recently in order to demonstrate how this has evolved over time and how it is characterised today. The contrast between the way people interact with these animals in 257

276 home countries and in Western forms of captivity was discussed, with a particular focus on Australian zoos. The information indicated that although people appear to be increasingly interacting with large felids in zoos, such interactions are particularly unnatural for the animals, and have a number of potential benefits and limitations for the zoo keepers and visiting public involved. This highlighted the importance of exploring these interactions in zoos from various perspectives. The next chapter (Chapter Three) offered an overview of past research methodologies used to investigate human-animal interactions in the zoo setting from the perspective of zoo keepers, animals, and zoo visitors. To explore how those involved in large felid-human interactions (i.e., the keepers, animals, and visitors) respond to such experiences, a series of three studies was conducted. Study 1 Zoo keeper survey In Chapter Four, a cross-sectional, web-based survey of zoo keepers (n = 86) provided the first scientifically documented investigation of handling and training practices used specifically with large felids in zoos worldwide. Key findings of this study were: Protected contact was the most commonly used handling method (compared to hands-on and hands-off contact) among this sample, and training was used more often than not. Participants rated protected contact as more beneficial than hands-on and handsoff contact for large felids, keepers and visitors. Perceived benefits of protected contact were that it allows a bond to form between keeper and animal, and has higher educational benefits for zoo visitors. Perceived limitations of hands-on contact were that it can be dangerous for cats and keepers, and sends the wrong message to zoo visitors. Perceptions of hands-off contact were mixed, with some suggesting that it is a safer practice for keepers and may send more accurate messages to visitors, while others suggested that it could lead to negative behaviours in the animals. 258

277 Training was rated as being more beneficial for all three parties than any of the handling methods, with respondents believing it to be of particular benefit to the health care of large felids, allowing a strong bond between animal and keeper to develop, and sending a positive message to zoo visitors. These findings provide a basis for future research, by uncovering various subjective opinions, which can now be objectively tested. Study 2 Zoo animal observations Chapter Five presented a three-month case study of three large felid species (tiger, n = 1; lion, n = 3; cheetah, n = 3) at Zoos SA, providing the first investigation of short-term behaviour in large felids during interactive zoo visitor tours. Key findings of the study were: Inactive species-typical behaviour comprised the majority of observation sessions per species, but tended to decrease during tours. Fluctuations in pacing were observed within and between non-tour and tour days, such that the lions demonstrating more pacing in the tour session, the cheetahs demonstrated less during the tour session, and the tiger more after the tour session. For all species, feeding behaviour was commonly observed during tours. While some aggressive behaviour was demonstrated by the lions, no aggressive, or otherwise antagonistic behaviours, were observed to be directed at humans by the cheetahs or tiger. During the tour sessions, all species exhibited fluctuations in estimated proximities to humans (keepers and visitors), with the lions and cheetahs predominantly observed in distant proximity and the tiger in close proximity. Combined, these findings indicate that large felid behaviour may be influenced by their participation in interactive tours, but that these changes may be associated with other zoo factors (such as feeding and neighbouring conspecifics) and are not necessarily indicative of compromised well-being. Adding to the few published studies which 259

278 suggest that these animals display behavioural changes in response to zoo visitors at exhibits, this study has highlighted a number of important avenues for further understanding the influence of visitor tours on these animals. Study 3 Zoo visitor survey Chapter Six presented a within and between groups, pre/post survey of Zoos SA visitors. It provided the first investigation of zoo visitor responses to participating in behind-the-scenes large felid experiences (n = 121) compared to traditional exhibit viewing (n = 325). Key findings of the study were: Tour visitors demonstrated increases in knowledge, attitudes to species, and attitudes toward conservation after their experience compared to before. Although post-tour visitors did not rate more highly on the multiple choice knowledge items compared to post-exhibit visitors, more of the former visitors were able to report new facts that they had learned compared to the latter. Post-tour visitors also rated higher than post-exhibit visitors on attitudes to the study species and some attitudes to conservation. Both post-tour and post-exhibit visitors reported emotions related to seeing and learning about the animals, with more comments made by tour visitors, perhaps indicating a more emotive experience. Post-tour visitors reporting that they would engage in a greater number of general and large felid-specific conservation behaviours after their visit compared to post-exhibit visitors. Post-tour visitors rated their experience as more educational, enjoyable, and entertaining than post-exhibit visitors; they also described more positive and less negative aspects of their experience, and provided fewer suggestions for how the experience could have been improved These findings extend the present body of research regarding interactive experiences by indicating that behind the scenes tours can provide a heightened opportunity for zoos to 260

279 educate, inspire and encourage their visitors in favour of large felid conservation, and are hence useful in meeting zoo goals in this regard Limitations to the scope of the thesis As with any research, there were various limitations encountered throughout this PhD program. Limitations to the specific studies have been discussed in the relevant chapters (see Section for the keeper study, for the animal study, and for the visitor study). In addition to these, several overarching limitations were encountered which have restricted the scope of the thesis and the associated findings. Firstly, the practice of large felid-human interactions is an under-explored area of study (see Section 1.4.5) and so determining which avenues to investigate required balancing depth and breadth. It was recognised from the outset that any exploration of these practices needed to consider not only the perspective of the animals, but also that of zoo keepers and visitors, since all of these groups are key zoo stakeholders. In this way, the thesis aimed to provide a snap shot of each perspective. Although successful in meeting this aim, the focus of each study was decided based on the time and financial constraints of a PhD program of research, and the thesis has not been able to provide an in-depth exploration from each perspective. Had the thesis taken one perspective only (e.g., evaluation of interactive experiences for zoo visitors), more about this perspective could have been understood. Regardless, insights have been contributed from each perspective, with two of the three studies disseminated in the peer-reviewed literature, and grounding for future research has been provided. Stemming from the approach to address multiple perspectives is that this thesis does not fall within a single discipline of research. Understanding and investigating each perspective required an exploration of various bodies of literature, including animal behaviour, conservation psychology, human-animal history, and zoo-based education. As such, it was again not fitting with the aims of the thesis to explore any one of these bodies of literature in as much depth as would be done in a thesis which focuses on one of these alone. 261

280 Furthermore, a major restriction on the depth of research undertaken was the limited available funding. Whilst the online nature of the keeper survey (Chapter 4) meant that travel for data collection was not required, expanding the animal (Chapter 5) and visitor (Chapter 6) studies to incorporate more than one zoo would have required this. For example, replicating the animal study would have involved at least a 3-month interstate stay, yet the available funding would not have covered the associated travel/accommodation expenses. While the impact of the thesis findings could have been greater with additional funding, the size of animal and human samples obtained from Zoos SA are comparable to those in other peer-reviewed studies (see Sections and respectively). Finally, another limitation of the thesis is its exploratory, rather than experimental, focus. Unlike other, more established areas of research, the purpose of this thesis was not to test particular theories, models, or assumptions, to contest previous findings, or reveal inconsistencies. While experimental research can allow for a stronger understanding of a particular topic, an exploratory approach was taken for two key reasons. Firstly, the lack of scientific knowledge regarding large felid-human interactions in zoos at the start of this researcher's candidature indicated that "ground work" was not yet available in this field. For example, it was not known how commonly keepers engage in different types of handling and training with their large felids. Therefore, the thesis aimed to evaluate select current practices (see Section 1.6), to provide a much-needed starting point for understanding how large-felids and humans (keepers and visitors) might benefit from, or be adversely influenced by, interacting with each other in the zoo environment. Secondly, undertaking experimental research at the zoo sites in this researcher's region (Zoos SA) would not have been possible due to a requirement of conducting research at Zoos SA being that no changes be made to existing animal-keeping practices, keeper routines, or visitor experiences. As such, an exploratory focus, using previously employed methodologies, was chosen. Each of these limitations point towards the need for multidisciplinary, interuniversity/zoo collaborative research to investigate this topic. Unlike some schools/disciplines, this researcher was not part of a collaborative research group and 262

281 therefore all aspects of the research were driven and conducted by this researcher alone. Working towards forming a nation-wide collaborative research group to explore the topic of human-animal interactions in zoos could help to address limitations in future research, such as relatively small samples of large felids in each zoo. Some suggested areas of future research, which would be best addressed through a collaborative approach, are discussed in Section 7.3. Each of these limitations and lessons learned have contributed to this researcher's understanding of conducting research within an applied, real-world setting which requires striking a balance between what is needed in the field and what is possible within present constraints Implications for stakeholders and directions for future research Notwithstanding the limitations of this program of research, findings from the three studies have a number of implications for different stakeholders and have provided a springboard for future studies in the field of large felid-human interactions in the zoo environment Implications for large felids and directions for future research Given that the primary impetus for conducting this research was the lack of information about how interactive practices might influence the behaviour (and hence, well-being) of large felids, it seems apt to discuss the implications of this thesis on the animals first. Section introduced the idea that interacting with people (zoo keepers and/or zoo visitors) may have the potential to be a positive experience for large felids (e.g., it may be enriching). The opinions of zoo keepers described in Chapter Four support this, highlighting that a degree of handling in particular, protected contact might benefit animal well-being (particularly in terms of physical well-being, psychological wellbeing, cognitive stimulation, social stimulation and behavioural stimulation), which is consistent with past research exploring human-animal relationships among other species (Section ). In stating that hands-off contact is not beneficial these participants 263

282 highlight that, although large felids are typically solitary in the wild (Table 1.1) and do not appear to interact positively when they come in to contact with people in their home countries (Section 2.3.1), they may be negatively influenced by a lack of human interaction in the zoo environment. Moreover, participants were cautious when discussing hands-on contact, particularly expressing concern regarding the possibility of euthanasia for cats who attack a person. Considering these views, it appears that finding an appropriate level of handling is a delicate task. Indeed, keepers were conscious of stating that many benefits and limitations of these different levels of interaction will depend on the individual animal, keeper, and circumstances, and so this raises a vast possibility of research questions that will need to be answered before we can understand how best to manage and enrich large felids through interactions with their keepers. Possible avenues will need to involve assessing the welfare (behavioural and hormonal) of large felids managed with the three different types of handling. Ideally, this would include a multi-institutional study of each species managed with each type of handling. If a large enough sample could be obtained, it may be possible to control for individual differences (e.g., animal ages, gender, rearing history, etc.) as well as differences in management practices (e.g., enclosure size and features, feeding regimes, access to exhibits, etc.) allowing more concrete conclusions to be drawn about the influence of handling styles. It may then be possible to have some measure of prediction about which individual animals might respond best to each level of handling. This could go a long way to improve the management of these animals in zoos, since it could mean that only those which are most suited to these practices are included in them. Section also highlighted that training could be beneficial for large felids, as it has been for other species. The findings in Chapter Four revealed that over half of the participating keepers train each of the seven large felid species, indicating a high prevalence of this practice. This has provided some empirical evidence for what was thought to be the case (based on information on zoo websites and details of available interactive tours at various zoos; see Section ). Although the study did not investigate which behaviours are trained or for what purpose training occurs, keepers reported it to be very beneficial to animal well-being (more so than any of the handling practice). This is significant since, although zoo professionals have previously been 264

283 found to believe that human interaction plays a very important role in mammal enrichment (Hoy et al., 2009), we have not yet known what they think specifically about large felids in this regard. Various authors have commented on the importance of keeper understandings of their animals in particular, they are usually experienced with, and knowledgeable about, the individuals they work with and have developed sound intuition about them (Melfi, 2009; Whitham & Wielebnowski, 2009). As such, although subjective, keepers are well-placed to offer their opinions, and the findings from Chapter Four are important for furthering our understanding of how training may benefit large felids and directing areas for future research. As has been done with primates (e.g., Savastano et al., 2003), a variety of directions could be taken to advance knowledge here, including how long it takes to train particular behaviours, which rewards are most motivating and effective, and which behaviours are most commonly trained (to date, only one study appears to have done this for a large felid species; see Broader et al., 2008). Having further validation and guidelines for implementing training and handling may allow for more managed practice, and continued monitoring will allow optimal animal well-being to be aimed for. Section raised the issue that unfamiliar humans (namely, zoo visitors) also have the potential to influence the welfare of large felids. Although limited to a very small body of research, it was discussed how past research has observed large felid behaviour to differ in different visitor conditions at exhibits (Maia et al., 2012; Mallapur & Chellam, 2002; Sellinger & Ha, 2005). This suggests that visitors could also influence the behaviour of these animals elsewhere, such as in off-exhibit housing, and under different circumstances, such as tours. The present research addressed this gap by investigating large felid behaviour associated with one type of closer interactive experience available to zoo visitors behind-the-scenes tours. On the grounds of observation sessions typically primarily comprising on inactive and active species typical behaviours, and suggestions that the observed pacing may be associated with factors not specific to the tours (e.g., feeding), it was argued that tours do not appear to be having a negative influence on the welfare of these animals. However, firm conclusions were difficult to draw from the findings - due to the numerous extraneous variables, which are often inherent in zoo-based research. In Section 5.4.5, it was discussed how it is possible that many of the features of the tours at Zoos SA serve to 265

284 protect against compromised well-being (e.g., predictability, retreat spaces, and feeding). An important direction for future research would be to understand if, and how, each of these factors serve to protect against negative welfare impacts. This could be made possible in a large, multi-institutional study, in which the use of a larger sample size could allow for the impact of these factors to be extracted. Since interactive large felid tours are offered at numerous Australian zoos (Table 2.1), there is the potential for this work to be conducted. Based on these preliminary findings, it is clear that more research is needed to facilitate our ability to improve the experience of large felids participating in tours. In particular, Melfi (2009) has argued for the importance of examining measures of optimal wellbeing, suggesting that an absence of poor well-being indicators is not enough to make inferences about good well-being. Therefore, in the interest of taking a proactive approach to enhancing zoo animal well-being and welfare (Hill & Broom, 2009), it is essential that future research explores how to maximise the potential for these experiences to be positive for the animals. For example, if future objective research finds training to be enriching for large felids, perhaps tours could involve a more formal training component. This could have the dual outcome of providing additional stimulation to cats during the tours, whilst also maximising the opportunity to showcase natural behaviours to zoo visitors. Although training is presently conducted informally and intermittently during the Zoos SA tours, the inclusion of a more formal training component could allow the influence of this to be measured. This would lend itself to an experimental design in which animal outcomes are documented on tours with and without training. Other ways that welfare during interactions could be improved is to understand how the characteristics of different levels of contact (in terms of factors such as time spent interacting, type of interactive behaviours, and proximity to humans) influence animal behaviour and hence well-being. If welfare cannot be enhanced through interactive practices, even in the absence of evidence of compromised wellbeing, then perhaps it is not appropriate to include such animals in tours. Future research will also need to investigate the different levels of handling for each species. One of the limitations of Chapter Five was that, due to differences in animal 266

285 histories, management and tour duration, it was deemed inappropriate to conduct analyses to compare the protected contact tour and hands-on contact tour (see Section 5.4.6). Understanding how each species responds to the different levels will allow an understanding of which type of handling is most appropriate for each. This would again be possible through a multi-institutional study. Indeed, all cheetah tours in Australia currently are hands-on in nature and there are also some hands-on tiger experiences (see Table 2.1), so there are a range of organisations that could be involved in such a study. Moreover, it is likely that zoos do not use their complete collections in tours; they may also house other animals of the same species and these could be involved in explorations of handling and training. For example, although Monarto Zoo practices hands-on contact with some of its cheetahs, others are interacted with through a protected contact approach (personal observation). An important outcome of such research would be to determine whether a protected contact approach is more beneficial than a hands-on one. Keepers opinions expressed in Chapter Four were largely concerned about the safety risks of a hands-on approach and, therefore, being able to show that benefits to animals can be obtained without needing to physically touch them would allow for safer practice (and therefore less risk of human injury/fatality and animal euthanasia). Part of understanding large felid behavioural responses to different levels of human interaction and training will require enhanced knowledge of the influence of specific human behaviours on these animals. For example, during the cheetah tour at Monarto Zoo, visitors are specifically instructed by zoo keepers/volunteers not to engage in particular behaviours that are thought to be confrontational to the animals (e.g., crouching down to the animal's level, making direct eye contact, touching the animal's face or ears; personal observation). As such, it is clear that there is some acknowledgment of the impact that human behaviour can have on animal responses, but documented evidence about this would be useful for ensuring that we minimise the impact we have on animals if we are to continue to interact with them. Some research has begun to recognise this. For example, in a combined analysis of cheetah and maned wolves (n = 115) behaviour, Carlstead (2009) found these animals to display increased aggression when their keepers locomoted or made unexpected noises when calling the animals. Given the limited literature presently available about large felid responses to 267

286 humans, some (Claxton, 2011; Hosey, 2008) have argued that these animals might be less susceptible to the visitor effect than other species, such as primates. However, we must place a greater emphasis on understanding how differences in human behaviour can influence the well-being of large felids, particularly if we are to work towards a positive and proactive approach to enhancing welfare. Another future direction that cannot be overlooked is the need for additional methods to complement behavioural data. Many animal welfare studies employ behavioural observations as their sole measure of animal well-being, particularly when evaluating enrichment (see Table 1.3). This is fitting, since enrichment often seeks to increase animal activity, increase active behaviours, and reduce stereotypic behaviour (Swaisgood & Shepherdson, 2005). The use of behavioural indicators is also fitting with the aims of studies exploring animal responses to visitors, which often focus on shortterm changes (see Section 3.3.1). However, Laule and Desmond (1998) suggest that psychological well-being should be measured by at least two of the following criteria: behaviour, health, reproduction, and longevity. Therefore, it is apparent that future research must take into account more than just the behaviour of the animals. One potential method to allow for a more complete measurement of the influence of human interaction could be to couple behavioural data with a biochemical marker of stress, such as a cortisol analysis (for an example of a combined behavioural and cortisol analysis of captive female cheetah, see Wielebnowski et al., 2002). Other measures, which may be more difficult and costly to obtain, include heart and respiratory rates as indicators of stress responses (Hill & Broom, 2009). The rationale for not including such measures in the thesis is discussed in Section Moreover, an enhanced understanding of animal characteristics and suitability of individuals for engagement in interactive practices with keepers or visitors will be important. One potential avenue is to understand more about felid personality and/or temperament. Various authors have acknowledged that individual animals will respond to the same stimuli in their own, unique ways (Carlstead, 2009; Hill & Broom, 2009) and perhaps personality/temperament is one contributor to this. In a review by Gartner and Weiss (1995), it was identified that most felid personality research has been 268

287 explored in domestic cats. Those studies which have explored personality in zoo felids (Gartner & Powell, 2012; Phillips & Peck, 2007; Wielebnowski, 1999; Wielebnowski et al., 2002) have identified a range of personality traits evident in snow leopards, cheetahs, tigers and clouded leopards, through both surveys and behavioural observations. These include, but not exclusively, active, vigilant, aggressive, calm, curious, playful, dominant, social and shy. Using personality as an indicator of suitability to interactive practices could enhance well-being and contribute to a proactive approach to the inclusion of animals which are handled and trained. For example, those individuals which are found to display traits of fearfulness and aggression in the presence of people could be excluded from such practices Implications for zoo keepers and directions for future research Section raised the idea that a degree of handling and training with large felids has the potential to be beneficial for zoo keepers, and the findings presented in Chapter Four provide support for this (in terms of improved relationship with animal, ease of movement of animal, enclosure maintenance, job satisfaction, and care of animals). Although the zoo-based literature to date has largely ignored zoo keepers, the findings presented in this thesis support those of Hosey and Melfi (2012) who found that zoo professionals (including zoo keepers) report various benefits of having a bond with their animals, including easier training and husbandry, and work which is more enjoyable and emotionally rewarding. If future research is able to show, more objectively, that the handling and training of large felids can assist with keeper duties and their workplace enjoyment, then this has important implications for optimising their work. Moreover, Melfi (2009) has argued that keepers will be more motivated with their work if they can see the effectiveness of a particular change. This can have similar benefits for others involved in the care of the animals, such as veterinarians who may be more easily able to perform their procedures without the need for anesthesia. Using more objective outcome measures, such as job productivity, will allow associations between practices and outcomes to be determined. Additionally, the participants in Chapter Four commented that the benefits of each practice will depend on the individual keeper and their skills; therefore, exploring keeper styles and individual practices, and linking this to animal and keeper outcomes, will allow for a better understanding of how best they 269

288 can interact with their animals. Indeed, Philips and Peck (2007) found interactive behaviour between tigers and keepers at Dreamworld, Australia, to be more related to keeper personality than cat personality (discussed further in Section 4.4). The findings from the animal study in Chapter Five are also important to keepers since they highlight what is working well and offer suggestions for approaches which could be incorporated into their work. To date there is a lack of information to guide training with large felids in the scientific literature (although it is acknowledged that there are likely to be guidelines for this within individual institutions). With the exception of the snow leopard training study (Broader et al., 2008), there is a clear need for more scientific research to guide the practice of keepers in this regard. This could include, for example, understanding which rewards motivate certain animals and how long it takes to train particular behaviours. Whilst objective research would be ideal, the method used in this thesis of surveying zoo keepers could similarly be used to extract more detail about what they do when handling and training. In a similar way, the visitor study findings presented in Chapter Six can also be useful in educating keepers about how what they do during tours can improve the experience and therefore maximise their potential to contribute to conservation. For example, it was found that Zoos SA could be making more of their opportunity to educate visitors on tours about specific conservation behaviours that they could engage in. In this way, keepers could incorporate more of this information into their tours. As previously established (see Section 3.2.1), keepers are often ignored in the zoo-based literature. Whilst the questions asked of them as part of this thesis were deemed important from a researcher s perspective, an important direction for future research would be to work more closely with keepers to understand what, if anything, they would like to know regarding large felid handling and training (from animal or human perspectives). This would allow a more bottom-up approach to such research and is deemed necessary in order for any results to be beneficial for keepers and for future research to have the greatest impact on practice. If it is of interest to them, and within their time constraints, having a better understanding of animal responses (as discussed in Section 7.3.1) could eventually lead to quicker measures that could be implemented 270

289 by zoo staff (either keepers or volunteers). Whitham and Wielebnowski (2009) advocate for keeper assessment of animal welfare, describing the use of Welfare Score Sheets (containing approximately 10, 5-point scale items) for various species at Brookfield Zoo, which provide weekly scores of animal well-being. These authors validated such a measure through matching with behavioural and fecal glucocorticoid monitoring, finding it to be a useful long-term monitoring option. Such a measure could be applied in zoos using interactive practices with their large felids and could provide longitudinal data. This would be important for complementing research data, which is often only limited to short periods of time (relative to the lifetime of animals/their involvement in interactive practices). Melfi (2009) also argues for the involvement of keepers in evidence-based research, on the grounds that they are in a position to measure interventions and their consequences Implications for zoo visitors and directions for future research The visitor experience needs to be considered here from two different perspectives: what visitors want for themselves, and what zoos want for visitors. From the first perspective, zoos depend on the public for financial support and therefore have an obligation to provide them with experiences that meet their reasons for visiting. This can be a complicated task, particularly when the needs of animals and visitors are often very different. For example, in the case of large felids, these animals are often inactive during day light hours (Powell, 1995; Van Metter et al., 2008), yet visitors often report a preference for active animals (Margulis et al., 2003). However, the findings from Chapters Five and Six have provided some support for interactive tours meeting the needs of both of these parties. This is a promising finding for visitors because, although further exploration is needed, there is some preliminary evidence for zoos to support such practices and continue using them. The findings from Chapter Six show that visitors who interact with large felids at Zoos SA are satisfied with their experience (see Section 6.3.6). They reported enjoying the opportunity to be in close proximity to and/or interact with the animals, and also the opportunity to witness animal feeding. A very small proportion of these participants reported an aspect of their experience that they were unhappy with, however, those that did primarily reported wanting to spend more time with the animals. Perhaps this is an indication of their enjoyment of the 271

290 experience and the desire to continue to interact further. Some tour participants reported that they would like to have received more information during the experience, as well as to touch the animals (presumably accounted for by those on the protected contact tours, as all visitors on the hands-on cheetah tour were able to pat the animals). These evaluations are important, since they can inform not only Zoos SA but also other zoos about what visitors want and how these desires can be achieved through interactive tours. Indeed, tours provide for the desires (e.g., to witness animal feeding) of regular visitors surveyed as part of this thesis. These findings suggest that tours can be a successful way to meet visitor desires for close interactions with animals. Visitors Australia-wide are paying notable amounts of money (see Table 2.1) to engage in these activities, so ensuring that they are satisfied is important, both in terms of encouraging them to promote their experience to other members of the public and, similarly, in terms of encouraging them to revisit the zoo. The findings from this study have shown that, similar to past zoo visitor research (see Section 3.4), survey methodology can be a useful way to gauge visitor satisfaction and, in order to continue to track this, zoos could conduct brief questionnaires after their tours (possibly through the assistance of zoo volunteers). From the perspective of what zoos want such as knowledge, attitude, emotional and behaviour change the findings presented in Chapter Six have provided some initial support for the role of behind-the-scenes tours to achieve this. This is important since there is an absence of information on the impact of such tours with any species in zoos (see Section ). Since this appears to be the first study of visitor responses to behind-the-scenes experiences with large felids, future research should address some of the limitations revealed in this initial investigation. In particular, given that many zoos aim to encourage their visitors to engage in conservation action/behaviour (as evidenced by their mission statements), it will be necessary to understand whether, how, and to what extent this is encouraged by participation in large felids tours. Doing this in the present research would have required more robust measures of each outcome variable. In the interest of keeping the questionnaire brief whilst attempting to address the four outcomes, measures of each were limited in length and validated scales were not available for use (see Sections and 6.2.3). The development of psychometrically validated scales will be an important area of research and could allow 272

291 for a model to be tested regarding how knowledge, attitudes and emotions interact to contribute to behaviour change. In order to justify risking the possible negative implications for animal well-being associated with visitor tours, zoos must be able to justify these experiences from visitor perspectives and this will need to be compared to other experiences, such as keeper talks. If talks are having the same impact, but involve less direct interaction between animals and visitors, then there may not be a need for tours (aside from revenue raising). Another important direction is comparing the level of contact to meet visitor outcomes. To date, some evidence has indicated that a greater number of interpretive experiences can lead to more pronounced changes in visitor outcomes (e.g., knowledge) (Weiler & Smith, 2009), but it is currently unclear which types of interpretive large felid programs are most effective. The present research combined hands-on and protected-contact experiences into one data set, to allow for a larger sample to be able to conduct statistical analyses and to compare to regular exhibit visitors. However, this did not allow for the impact of different levels of experience to be understood, so this would be a necessary direction for future research, particularly in terms of understanding whether visitors need to touch (or otherwise physically interact with) an animal in order to have the greatest impact on conservation outcomes. Indeed, it could be argued that in hands-on situations, where visitors have paid a relatively large sum of money to touch an animal, they may be likely to be more focused on doing this than listening to a keeper (and hence opportunities for behaviour change might be missed). Also, some zoos in Australia do offer very brief encounters (e.g., 15 minutes; see Table 2.1) with the sole purpose of visitors patting and having their photograph taken with a large felid. It is unlikely that this presents an opportunity for visitor learning. Conversely, it might be that even though visitors may not gain conservation knowledge during these brief experiences, the opportunity to connect with the animal could inspire them to engage in learning after their visit. The use of a follow-up survey of visitors at different intervals (e.g., three months, six months and one-year) could allow for this to be investigated. Finally, it would be essential for future research to explore visitor understandings of interactive practices with large felids. Indeed, Kreger and Mench (1995) urge that we 273

292 understand what visitors interpret about animal training. In Chapter Four, keepers were primarily concerned about the potential to send wrong messages to visitors through interactions. Future research could address this by asking visitors questions such as: a) why do you think keepers interact with large felids in zoos?; b) do you think interacting with felids benefits them?; c) in what ways do you think these interactions benefit large felids?; d) do you think large felids make good pets?; e) does interacting with a large felid make you want one as a pet? This will allow for an understanding of how interactive practices influence the opinions of zoo visitors, particularly in terms of pet ownership, and will allow zoos to adjust and target their practices and messages accordingly. For example, if seeing keepers pat a tiger during a keeper talk encourages visitors to want a tiger as a pet, then perhaps all direct interactions should occur offexhibit where visitors cannot witness it. In Australia, exotic animals cannot be legally kept as pets, but there are other parts of the world where they can and this is something that zoos need to be wary of (see Section ). If it is determined that visitors are walking away with misconceptions about the animals, then zoos will need to look at either adapting their practices or providing more information to visitors about why interactions are occurring. Having scientific evidence to support these practices will add weight to this information and will allow zoos to work more successfully towards tackling the various portrayals in recent media which may serve to encourage interactive practices outside of zoos (e.g., pet ownership; see Section ) Implications for zoos in general and directions for future research Whilst the scope of this thesis has focused exclusively on large felids, the methodologies used and investigations conducted could be useful for exploring interactions between humans and various other species in the zoo environment. Although primates have already focused quite heavily in the training literature, in both zoo and laboratory settings (see Section ), the exploration of human-animal interactions in zoos is a relatively new area of study. This thesis has reinforced the importance of exploring animal husbandry and management processes from the perspectives of animals, keepers and visitors, in order to provide a more holistic understanding of the associated benefits and limitations and to improve practice in ways to benefit all key stakeholders. From the freely available information on zoo websites, 274

293 zoos appear to offer experiences for interaction with multiple species, not just large felids, and so an important future direction would be to conduct cross-taxa research, exploring interactions from the three perspectives covered in this thesis. This would allow zoos to have a better understanding of which animals are most suited to human interaction and how people (keepers and visitors) respond to this. Past reviews have revealed that interactive research is largely limited to primates (see Claxton, 2011; Hosey, 2008) and so there is much that remains to be known about interactions with the full range of species housed in zoos. If the primary purposes of interactive experiences are to raise revenue through the visiting public and encourage conservation behaviour change among these individuals, then an important avenue for future research would be to explore human interactions with a range of species living in the same region and threatened by the same factors. For example, deforestation for palm oil plantations is a major threat to the survival of multiple species in Indonesia, including the tiger, rhinoceros, orangutan and elephant. It could be the case that an interactive experience with any one of these species is equally inspiring to encourage behaviour change among zoo visitors. If behavioural and physiological data on animal behaviour supports this, and if it has benefits for zoo keepers, then the most suited species could be used in interactive programs and the welfare of those deemed unsuited can be preserved. Undertaking research of this nature will require a multi-institutional collaborative project, involving numerous zoos and university research groups Implications for large felid populations and humans in home countries Section extensively highlighted the conflict occurring between large felids and humans in home countries. In addition to financial support, which may be provided through revenue raised by zoos as a result of interactive experiences, the findings of this thesis can have some implications for promoting positive relationships outside of Western zoos. Challenging this largely negative relationship in home countries is a complex task, however, a better understanding of the role that interactive programs can have on zoo visitors could be applied to animal housing facilities in home countries to promote a better understanding of the importance of large felids to biodiversity among local people. Of course, it would be naïve to think that an experience like this would impact the attitudes, and hence behaviour, of people whose livelihood can be threatened 275

294 by these animals (see Section ) in the same way that it influences people in Western zoos, or that people in home countries have access to zoos/other animal keeping facilities. However, if revenue from these activities in Western zoos can be used to provide support for local people in home countries (for example, through increased compensation for losses associated with livestock depredation, or through the provision of measures to increase safety of people and livestock), then perhaps interactions can serve as a secondary method to enhance our understanding of the value of these animals among people with whom they share their natural environment. Combined, this could contribute to a more positive large felid-human relationship Conclusion Although appearing to be expanding quite rapidly in practice, the study of large felidhuman interactions in zoos is a research area is in its infancy. This thesis has provided a starting point for investigating this phenomenon from the perspectives of the different parties involved the animals, keepers and visitors. As Guggisberg (1975) advocated, as humans we are responsible for encouraging and sustaining a positive relationship with large felids. Although many of us are quite far removed (spatially) from these animals in their home countries, understanding how positive relationships can exist in zoos is vital, particularly if we are to maintain the optimal welfare of large felids and promote their conservation to the general public, and also if we are to promote best practice outside of Western zoological organisations. The studies discussed in this thesis have provided some much-needed groundwork for future research in this area. In bringing this thesis to a close, it is important to acknowledge that, at least for the foreseeable future, large felids will continue to be housed in zoos, keepers will continue to care for them, and the public will continue to visit them. With this is mind, I wholeheartedly share in Hosey s (2008) ambition that we will one day understand enough about how best to meet these often-conflicting needs in a way that will allow zoos to provide optimal experiences for their animals, their keepers and the visiting public. 276

295 References Adelman, L. M., Falk, J. H., & James, S. (2000). Impact of National Aquarium in Baltimore on visitors' conservation attitudes, behavior, and knowledge. Curator: The Museum Journal, 43, Ajzen, I. (2001). Nature and operation of attitudes. Annual Review of Psychology, 52, Akama, J. S., & Kieti, D. M. (2003). Measuring tourist satisfaction with Kenya's wildlife safari: A case study of Tsavo West National Park. Tourism Management, 24, Altman, J. D. (1998). Animal activity and visitor learning at the zoo. Anthrozoos, 11, Anderson, U. S., Benne, M., Bloomsmith, M. A., & Maple, T. (2002). Retreat space and human visitor density moderate undesirable behavior in petting zoo animals. Journal of Applied Animal Welfare Science, 5, Bahram, R., Burke, J. E., & Lanzi, G. L. (2004). Head and neck injury from a leopard attack: Case report and review of the literature. Journal of Oral and Maxillofacial Surgery, 62, Baker, K. C. (2004). Benefits of positive human interaction for socially-housed chimpanzees. Animal Welfare, 13, Baker, W. K., & Campbell, R. (1997). Enriching the pride: Scents that make sense. The Shape of Enrichment, 6, 1-3. Baldus, R. D. (2004). Lion conservation in Tanzania leads to serious human-lion conflicts: With a case study of a man-eating lion killing 35 people. Tanzania Wildlife Discussion Paper, 41, Ballantyne, R., & Packer, J. (2005). Promoting environmentally sustainable attitudes and behaviour through free-choice learning experiences: What is the state of the game? Environmental Education Research, 11, Ballantyne, R., Packer, J., Hughes, K., & Dierking, L. (2007). Conservation learning in wildlife tourism settings: Lessons from research in zoos and aquariums. Environmental Education Research, 13, Ballantyne, R., Packer, J., & Sutherland, L. A. (2011). Visitors' memories of wildlife tourism: Implications for the design of powerful interpretive experiences. Tourism Management, 32, Balme, G. A., Hunter, L. T. B., Goodman, P., Ferguson, H., Craigie, J., & Slotow, R. (2010). An adaptive management approach to trophy hunting of leopards (Panthera pardus): A case study from KwaZulu-Natal, South Africa. In D. W. Macdonald & A. J. Loveridge (Eds.), Biology and conservation of wild felids (pp ). Oxford: Oxford University Press. Baratay, E., & Hardouin-Fugier, E. (2002). Zoo: A history of zoological gardens in the West (O. Welsh, Trans.). London: Reaktion Books. Barnett, R., Yamaguchi, N., Barnes, I., & Cooper, A. (2006). The origin, current diversity and future conservation of the modern lion (Panthera leo). Proceedings of the Royal Society B, 273, Barr, S. (2007). Factors influencing environmental attitudes and behaviors: A U.K. case study of household waste management. Environment and Behavior, 39, Bashaw, M. J., Bloomsmith, M. A., Marr, M. J., & Maple, T. L. (2003). To hunt or not to hunt? A feeding enrichment experiment with captive large felids. Zoo Biology, 22, Bashaw, M. J., Kelling, A. S., Bloomsmith, M. A., & Maple, T. L. (2007). Environmental effects on the behavior of zoo-housed lions and tigers, with a 277

296 case study of the effects of a visual barrier on pacing. Applied Animal Behaviour Science, 10, Bashaw, M. J., & Maple, T. L. (2001). Signs fail to increase zoo visitors' ability to see tigers. Curator, 44, Bassett, L., & Buchanan-Smith, H. M. (2007). Effects of predictability on the welfare of captive animals. Applied Animal Behaviour Science, 102, Bayne, K. (2002). Development of the human-research animal bond and its impact on animal well-being. Institute for Laboratory Animal Research Journal, 43, 4-9. Birke, L. (2002). Effects of browse, human visitors and noise on the behaviour of captive orangutans. Animal Welfare, 11, Bishop, J., Mitchell, H., & Hosey, G. (2013). Studies on the effects of human visitors on zoo animal behaviour. In J. Bishop, G. Hosey & A. Plowman (Eds.), Handbook of zoo research, guidelines for conducting research in zoos (pp ). London: BIAZA. Bitgood, S. (1992). The impact of a zoo visit on attitudes: A preliminary report on interaction effects. Visitor Behavior, 7, Blaney, E. C., & Wells, D. L. (2004). The influence of a camouflage net barrier on the behaviour, welfare and public perceptions of zoo-housed gorillas. Animal Welfare, 13, Bloomsmith, M. A., & Else, J. G. (2005). Behavioral management of chimpanzees in biomedical research facilities: The state of the science. Institute for Laboratory Animal Journal, 46, Bloomsmith, M. A., Jones, M. L., Snyder, R. J., Singer, R. A., Gardner, W. A., Liu, S. C., & Maple, T. L. (2003). Positive reinforcement training to elicit voluntary movement of two giant pandas throughout their enclosure. Zoo Biology, 22, Bloomsmith, M. A., Stone, A. M., & Laule, G. E. (1998). Positive reinforcement training to enhance the voluntary movement of group-housed chimpanzees within their enclosures. Zoo Biology, 17, Bond, J. C., & Lindenburg, D. G. (1990). Carcass feeding of captive cheetahs (Acinonyx jubatus): The effects of a naturalistic feeding program on oral health and psychological well-being. Applied Animal Behaviour Science, 26, Born Free USA. (2013). Summary of state laws relating to private possession of exotic animals. from Bowler, M. T., Buchanan-Smith, H. M., & Whiten, A. (2012). Assessing public engagement with science in a university primate research centre in a national zoo. PLoS ONE, 7. doi: /journal.pone Brensing, K., & Linke, K. (2003). Behavior of dolphins towards adults and children during swim-with-dolphin programs and towards children with disabilities during therapy sessions. Anthrozoos, 16, Broad, S., & Weiler, B. (1998). Captive animals and interpretation - A tale of two tiger exhibits. Journal of Tourism Studies, 9, Broader, J. M., MacFadden, A. J., Cosens, L. M., Rosenstein, D. S., & Harrison, T. M. (2008). Use of positive reinforcement conditioning to monitor pregnancy in an unanesthetized snow leopard (Uncia uncia) via transabdominal ultrasound. Zoo Biology, 27, Bulbeck, C. (2005). Facing the wild: Ecotourism, conservation and animal encounters. London: Earthscan. Callou, C., Samzun, A., & Zevie, A. (2004). A lion found in the Egyptian tomb of Maia. Nature, 427,

297 Care for the Wild International. (2008). Exploiting the tiger: Illegal trade, animal cruelty and toursts at risk at the Tiger Temple: Care for the Wild International. Carlstead, K. (1998). Determining the causes of stereotypic behavior in zoo carnivores: Towards appropriate enrichment strategies. In D. J. Shepherdson, J. D. Mellen & M. Hutchins (Eds.), Second Nature: Environmental enrichment for captive animals (pp ). Washington: Smithsonian Institution Press. Carlstead, K. (2009). A comparative approach to the study of keeper-animal relationships in the zoo. Zoo Biology, 28, Carlstead, K., & Brown, J. L. (2005). Relationships between patterns of fecal corticoid excretion and behaviour, reproduction, and environmental factors in captive black (Diceros bicornis) and white (Ceratotherium simum) rhinoceros. Zoo Biology, 24, Chamove, A. S., Hosey, G. R., & Schaetzel, P. (1988). Visitors excite primates in zoos. Zoo Biology, 7, Chapenoire, S., Camiade, B., & Legros, M. (2001). Basic instinct in a feline. The American Journal of Forensic Medicine and Pathology, 22, Charlton, N. (1998). The effects of an enrichment device on the behaviour of captive jaguars. Ratel, 25, CITES (2013). Appendices I, II, and III. From Clark, F., & King, A. J. (2008). A critical review of zoo-based olfactory enrichment. In J. L. Hirst, R. J. Beynon, S. C. Roberts & T. D. Wyatt (Eds.), Chemical signals in vertebrates II (pp ). New York: Springer. Claxton, A. M. (2011). The potential of the human-animal relationship as an environmental enrichment for the welfare of zoo-housed animals. Applied Animal Behaviour Science, 133, Clayton, S., & Brook, A. (2005). Can psychology help save the world? A model for Conservation Psychology. Analyses of Social Issues and Public Policy, 5, Clayton, S., Fraser, J., & Saunders, C. D. (2009). Zoo experiences: Conversations, connections, and concern for animals. Zoo Biology, 28, Clubb, R., & Mason, G. J. (2007). Natural behavioural biology as a risk factor in carnivore welfare: How analysing species differences could help zoos improve enclosures. Applied Animal Behaviour Science, 102, Coe, J. C. (1997). Entertaining zoo visitors and zoo animals: An integrated approach. Paper presented at the AZA Convention Proceedings., Bethesda, MD. Cohle, S. D., Harlan, C. W., & Harlan, G. (1990). Fatal big cat attacks. The American Journal of Forensic Medicine and Pathology, 11, Colahan, H., & Breder, C. (2003). Primate training at Disney's Animal Kingdom. Journal of Applied Animal Welfare Science, 6, Conde, D. A., Flesness, N., Colchero, F., Jones, O. R., & Scheuerlein, A. (2011). An emerging role of zoos to conserve biodiversity. Science, 331, Conforti, V. A., & de Avezedo, F. C. C. (2003). Local perceptions of jaguars (Panthera onca) and pumas (Puma concolor) in the Iguacu National Park area, South Brazil. Biological Conservation, 111, Cook, S., & Hosey, G. R. (1995). Interaction sequences between chimpanzees and human visitors at the zoo. Zoo Biology, 14, Corraliza, J. A., & Berenguer, J. (2000). Environmental values, beliefs, and actions: A situation approach. Environment and Behavior, 32, Costa, P. T., & McCrae, R. R. (1992). Neo PI-R Professional Manual. Lutz, FL: Psychological Assessment Resources. 279

298 CASP. (2013). CASP Checklists. From Curtin, S. (2006). Swimming with dolphins: A phenomenological exploration of tourist recollections. International Journal of Tourism Research, 8, Curtin, S., & Wilkes, K. (2007). Swimming with captive dolphins: Current debates and post-experience dissonance. International Journal of Tourism Research, 9, da Silva, M. A. M., & da Silva, J. M. C. (2007). A note on the relationships between visitor interest and characteristics of the mammal exhibits in Recife Zoo, Brazil. Applied Animal Behaviour Science, 105, Dalley, S. (1993). Ancient Mesopotamian gardens and the identification of the Hanging Gardens of Babylon. The Garden History Society, 21, Damania, R., Seidensticker, T., Whitten, T., Sethi, G., Mackinnon, K., Kiss, A., & Kushlin, A. (2008). A future for wild tigers. Washington, D.C.: World Bank. Davey, G. (2005). The "Visitor Effect". Zoos' Print Journal, 20, Davey, G. (2006a). An hourly variation in zoo visitor interest: Measurement and significance for animal welfare research. Journal of Applied Animal Welfare Science, 9, Davey, G. (2007). Visitors' effects on the welfare of animals in the zoo: A review. Journal of Applied Animal Welfare Science, 10, Davey, G., & Henzi, P. (2004). Visitor circulation and nonhuman animal welfare: An overlooked variable? Journal of Applied Animal Welfare Science, 7, Davies, G. (2005). Virtual animals in electronic zoos: The changing geographics of animal capture and display. In C. Philo & C. Wilbert (Eds.), Animal spaces, beastly places. London: Routledge. Davis, N., Schaffner, C. M., & Smith, T. E. (2005). Evidence that zoo visitors influence HPA activity in spider monkeys (Ateles geoffroyii rufiventris). Applied Animal Behaviour Science, 90, de Azevedo, C. S., Cipreste, C. F., & Young, R. J. (2007). Environmental enrichment: A GAP analysis. Applied Animal Behaviour Science, 102, de Courcy, C. (2001). Zoological gardens of Australia. In V. N. Kisling (Ed.), Zoo and aquarium history: Ancient animal collections to zoological gardens (pp ). Hoboken: CRC Press. De Rouck, M., Kitchener, A. C., Law, G., & Nelissen, M. (2005). A comparative study of the influence of social housing on the behaviour of captive tigers (Panthera tigris). Animal Welfare, 14, Department of Agriculture, Forestry, & Fisheries. (2009). Australian animal welfare standards and guidelines: exhibited animals (Version 5 - Stakeholder review draft). Retrieved from data/assets/pdf_file/0006/762846/draft-animalexhibit-standards.pdf. Desportes, G., Buholzer, L., Anderson-Hansen, K., Blanchet, M.-A., Acquarone, M., Shephard, G.,... Siebert, U. (2007). Decrease stress; train your animals: The effect of handling methods on cortisol levels in harbour porpoises (Phocoena phocoena) under human care. Aquatic Mammals, 33, Dierenfeld, E. S., Bush, M., Phillips, L., & Montali, R. (1994). Nutrition, food preparation and feeding. In R. Tilson, G. Brady, K. Traylor-Holzer & D. Armstrong (Eds.), Management and conservation of captive tigers (2nd ed.). Minnesota: Minnesota Zoo. 280

299 Dierking, L. D., Adelman, L. M., Ogden, J., Lehnhardt, K., Miller, L., & Mellen, J. D. (2004). Using a behavior change model to document the impact of visits to Disney's Animal Kingdom: A study investigating intended conservation action. Curator: The Museum Journal, 47, Dorfman, L. (2005). Bonding with mother-reared cheetahs. Animal Keepers' Forum, 32, Duckler, G. L. (1998). An unusual osteological formation in the posterior skulls of captive tigers (Panthera tigris). Zoo Biology, 17, Dutcher, D. D., Finley, J. C., Luloff, A. E., & Buttolph Johnson, J. (2007). Connectivity with nature as a measure of environmental values. Environment and Behavior, 39, Falk, J. H., & Adelman, L. M. (2003). Investigating the impact of prior knowledge and interest on aquarium visitor learning. Journal of Research in Science Teaching, 40, Falk, J. H., & Gillespie, K. L. (2009). Investigating the role of emotion in science center visitor learning. Visitor Studies, 12, Falk, J. H., Reinhard, E. M., Vernon, C. L., Bronnenkant, K., Deans, N. L., & Heimlich, J. E. (2007). Why zoos & aquariums matter: Assessing the impact of a zoo visit. Silver Springs, MD: Association of Zoos & Aquariums. Fernandez, E. J., Tamborski, M. A., Pickens, S. R., & Timberlake, W. (2009). Animalvisitor interactions in the modern zoo: Conflicts and interventions. Applied Animal Behaviour Science, 120, 1-8. Fielding, K. S., McDonald, R., & Louis, W. R. (2008). Theory of planned behaviour, identity and intentions to engage in environmental activism. Journal of Environmental Psychology 28, Gartner, M. C., & Powell, D. (2011). Personality assessment in snow leopards (Uncia uncia). Zoo Biology, 29, Gill, R. B. (2010). To save a mountain lion. In M. Hornocker & S. Negri (Eds.), Cougar: Ecology and Conservation (pp. 5-16). Chicago: University of Chicago Press. Glatston, A. R., Geilvoet-Soeteman, E., Hora-Pecek, E., & van Hooff, J. A. R. A. M. (1984). The influence of the zoo environment on social behavior of groups of cotton-topped tamarins, Saguinus oedipus oedipus. Zoo Biology, 3, Global Tiger Initiative Secretariat. (2013). Global tiger recovery program implementation implementation plan: Washington, D.C.: World Bank. Goodrich, J. M. (2010). Human-tiger conflict: A review and call for comprehensive plans. Integrative Zoology, 5, Gros, P. M. (1998). Status of the cheetah Acinonyx jubatus in Kenya: A field-interview assessment. Biological Conservation, 85, Guggisberg, C. A. W. (1975). Wild cats of the world. London: David and Charles. Gurung, B. (2008). Tiger human conflicts: Investigating ecological and sociological issues of tiger conservation in the buffer zone of Chitwan National Park, Nepal. Unpublished PhD Thesis. University of Minnesota. Minnesota, USA. Retrieved from Gurung, B., Smith, J. L. D., McDougal, C., Karki, J. B., & Barlow, A. (2008). Factors associated with human-killing tigers in Chitwan National Park, Nepal. Biological Conservation, 141, Hagenbeck, C. (1909). Beasts and men (H. S. R. Elliot & A. G. Thacker, Trans.). London: Longmans, Green, and Co. 281

300 Ham, S., & Weiler, B. (2005). Interpretation and evaluation tool kit: Methods and tools for assessing the effectiveness of face-to-face interpretive programs. Brisbane, Australia: Sustainable Tourism Cooperative Research Centre. Hare, V. J., & Karrand, P. (1998). Artificial prey that fights back (and other tales of tiger enrichment). The Shape of Enrichment, 7, 1-4. Harley, H. E., Fellner, W., & Stamper, M. A. (2010). Cognitive research with dolphins (Tursiops truncatus) at Disney's The Seas: A program for enrichment, science, education and conservation. International Journal of Comparative Psychology, 23, Harmsen, B. J., Foster, R. J., Silver, S. C., Ostro, L. E. T., & Doncaster, C. P. (2010). The ecology of jaguars in the Cockscomb Basin Wildlife Sanctuary, Belize. In D. W. Macdonald & A. J. Loveridge (Eds.), Biology and conservation of wild felids (pp ). Oxford: Oxford University Press. Hediger, H. (1964). Wild animals in captivity (G. Sircom, Trans.). New York: Dover. Hediger, H. (1968). The psychology and behaviour of animals in zoos and circuses (G. Sircom, Trans.). New York: Dover. Heimlick, J. E., & Storksdieck, M. (2007). Think Piece. Changing thinking about learning for a changing world. Southern African Journal of Environmental Education, 24, Hejna, P. (2010). A fatal leopard attack. Journal of Forensic Sciences, 55, Hibbard, C., Ford, C., & Hoggs, C. (2012). Australasian species management programs: Regional census and plan (22nd ed.). Sydney, Australia: Zoos and Aquarium Association. Hill, S., & Smith, T. (2013). Monitoring adrenal response in zoo animals as an indicator of welfare. In J. Bishop, G. Hosey & A. B. Plowman (Eds.), Handbook of zoo research, guidelines for conducting research in zoos London: BIAZA. Hill, S. P., & Broom, D. M. (2009). Measuring zoo animal welfare: Theory and practice. Zoo Biology, 28, Hinds, J., & Sparks, P. (2007). Engaging with the natural environment: The role of affective connection and identity. Journal of Environmental Psychology, 28, Hosey, G. (2008). A preliminary model of human-animal relationships in the zoo. Applied Animal Behaviour Science, 109, Hosey, G., & Melfi, V. (2012). Human-animal bonds between zoo professionals and the animals in their care. Zoo Biology, 29, Hoy, J. M., Murray, P. J., & Tribe, A. (2009). Thirty years later: Enrichment practices for captive mammals. Zoo Biology, 29, Hutchins, M., & Smith, B. (2003). Characteristics of a world-class zoo or aquarium in the 21st century. International Zoo Yearbook, 38, Inskip, C., & Zimmermann, A. (2009). Human-felid conflict: A review of patterns and priorities worldwide. Oryx, 43, IUCN. (2012). The IUCN Red List of Threatened Species. Version from Janecka, J. E., Jackson, R., Yuquang, Z., Diqiang, L., Munkhtsog, B., Buckley-Beason, V., & Murphy, W. J. (2008). Population monitoring of snow leopards using noninvasive collection of scat samples: A pilot study. Animal Conservation, 11, Jennison, G. (2005). Animals for show and pleasure in ancient Rome. Philadelphia, PA: University of Pennsylvania Press. 282

301 Jenny, S., & Schmid, H. (2002). Effects of feeding boxes on the behavior of stereotyping Amur tigers (Panthera tigris altaica) in the Zurich Zoo, Zurich, Switzerland. Zoo Biology, 21, Johnson, A., Vangkhamheng, C., Haedemark, M., & Saithongolam, T. (2006). Effects of human-carnivore conflict on tiger (Pantheris tigris) and prey populations in Lao PDR. Animal Conservation, 9, Kalof, L. (2007). Looking at animals in human history. London: Reaktion Books. Karanth, K. K., & DeFries, R. (2011). Nature-based tourism in Indian protected areas: New challenges for park management. Conservation Letters, 4, Karanth, K. U., & Chellam, R. (2009). Carnivore conservation at the crossroads. Oryx, 43, 1-2. Keeling, C. H. (2001). Zoological gardens of Great Britain. In V. N. Kisling (Ed.), Zoo and aquarium history: Ancient animal collections to zoological gardens (pp ). Hoboken: CRC Press. Kellert, S. R., Black, M., Rush, C. R., & Bath, A. J. (1996). Human culture and large carnivore conservation in North America. Conservation Biology, 10, Kirkwood, J. K. (2003). Welfare, husbandry and veterinary care of wild animals in captivity: Changes in attitudes, progress in knowledge and techniques. International Zoo Yearbook, 28, Kisling, V. N. (2001). Ancient collections and menageries. In V. N. Kisling (Ed.), Zoo and aquarium history: Ancient animal collections to zoological gardens. Hoboken: CRC Press. Kitchener, A. C. (2000). Are cats really solitary? Lutra, 43, Kitchener, A. C., Van Valkenburgh, B., & Yamaguchi, N. (2010). Felid form and function. In D. W. Macdonald & A. J. Loveridge (Eds.), Biology and conservation of wild felids (pp ). Oxford: Oxford University Press. Knapp, D., & Yang, L. (2005). A phenomenological analysis of long-term recollections of an interpretive program. Journal of Interpretation Research, 7, Kohout, M. P., Percy, J., WSears, W., & Yeo, J. D. (1989). Tiger mauling: Fatal spinal injury. Australian and New Zealand Journal of Surgery, 59, Kreger, M. D., & Mench, J. A. (1995). Visitor animal interactions at the zoo. Anthrozoos, 8, Kyle, D. G. (2001). Spectacles of death in ancient Rome. London: Routledge. Kyngdon, D. J., Minot, E. O., & Stafford, K. J. (2003). Behavioural responses of captive common dolphins Delphinus delphis to a 'swim-with-dolphin' programme. Applied Animal Behaviour Science, 81, Labuschange, W., & Walker, S. (2001). Zoological gardens of Africa. In V. N. Kisling (Ed.), Zoo and aquarium history (pp ). Hoboken: CRC Press. Lagendijk, D. D. G., & Gusset, M. (2008). Human-carnivore coexistence on communal land bordering the Greater Kruger area, South Africa. Environmental management, 42, Lambeth, S. P., Hau, J., Perlman, J. E., Martion, M., & Schapiro, S. J. (2006). Positive reinforcement training affects hematologic and serum chemistry values in captive chimpanzees (Pan troglodytes). American Journal of Primatology, 68, Latham, N. R., & Mason, G. J. (2008). Maternal deprivation and the development of stereotypic behaviour. Applied Animal Behaviour Science, 110, Laule, G., & Desmond, T. (1998). Positive reinforcement training as an enrichment strategy. In D. J. Shepherdson, J. D. Mellen & M. Hutchins (Eds.), Second 283

302 nature: Environmental enrichment for captive felids (pp ). Washington, DC: Smithsonian Institution Press. Laule, G. E., Bloomsmith, M. A., & Schapiro, S. J. (2003). The use of positive reinforcement training techniques to enhance the care, management, and welfare of primates in the laboratory. Journal of Applied Animal Welfare Science, 6, Law, G., Macdonald, A., & Reid, A. (1997). Dispelling some common misconceptions about the keeping of felids in captivity. International Zoo Yearbook, 35, Lazarus, H. M., Price, R. S., & Sorensen, J. (2001). Dangers of large exotic pets from foreign lands. The Journal of Trauma, Injury, Infection, and Critical Care, 51, Lindburg, D. G. (1988). Improving the feeding of captive felines through application of field data. Zoo Biology, 7, Lindemann-Matthies, P., & Kamer, T. (2006). The influence of an interactive educational approach on visitors' learning in a Swiss zoo. Science Education, 90, doi: /sce Linkie, M., Martyr, D. J., Holden, J., Yanuar, A., Hartana, A. T., Sugardjito, J., & Leader-Williams, N. (2003). Habitat destruction and poaching threaten the Sumatran tiger in Kerinci Seblat National Park, Sumatra. Oryx, 37, Loe, J., & Roskaft, E. (2004). Lare carnivores and human safety: A review. Ambio, 33, Loveridge, A. J., Wang, S. W., Frank, L. G., & Seidensticker, J. (2010). People and wild felids: Conservation of cats and management of conflicts. In D. W. Macdonald & A. J. Loveridge (Eds.), Biology and conservation of wild felids (pp ). Oxford: Oxford University Press. Luck, M. (2003). Education on marine mammal tours as agent for conservation - but to tourists want to be educated? Ocean & Coastal Management, 46, Luebke, J. F., & Matiasek, J. (2013). An exploratory study of zoo visitors' exhibit experiences and reactions. Zoo Biology, 32, Lukas, K. E., Barkauskas, R. T., Maher, S. A., Jacobs, B. A., Bauman, J. E., Henderson, A. J., & Calcagno, J. M. (2002). Longitudinal study of delayed reproductive success in a pair of white-cheeked gibbons (Hylobates leucogenys). Zoo Biology, 21, Lukas, K. E., & Ross, S. R. (2005). Zoo visitor knowledge and attitudes toward gorillas and chimpanzees. The Journal of Environmental Education, 36, Lyons, J., Young, R. J., & Deag, J. M. (1997). The effects of physical characteristics of the environment and feeding regime on the behavior of captive felids. Zoo Biology, 16, Macdonald, D. W., Loveridge, A. J., & Nowell, K. (2010). Dramatis personae: An introduction to the wild felids. In D. W. Macdonald & A. J. Loveridge (Eds.), Biology and conservation of wild felids (pp. 3-58). Oxford: Oxford University Press. Macdonald, D. W., Mosser, A., & Gittleman, J. L. (2010). Felid society. In D. W. Macdonald & A. J. Loveridge (Eds.), Biology and conservation of wild felids (pp ). Oxford: Oxford University Press. Macri, A. M., & Patterson-Kane, E. (2011). Behavioural analysis of solitary versus socially housed snow leopards (Panthera uncia), with the provision of simulated social contact. Applied Animal Behaviour Science, 130,

303 Maia, C. M., Volpato, G. L., & Santos, E. F. (2012). A case study: The effect of visitors on two captive pumas with respect to time of the day. Journal of Applied Animal Welfare Science, 15, Mallapur, A., & Chellam, R. (2002). Environmental influences on stereotypy and the activity budget of Indian leopards (Panthera pardus) in four zoos in southern India. Zoo Biology, 21, Mallapur, A., Qureshi, Q., & Chellam, R. (2002). Enclosure design and space utilization by Indian leopards (Panthera pardus) in four zoos in Southern India. Journal of Applied Animal Welfare Science, 5(2), Mansour, A. A. H., Zakaria, A.-H., & Fraser, A. F. (2000). Effect of enclosure quality on reactivity and welfare of captive soemmerring's gazelle (Gazella soemmerringii). Journal of Applied Animal Welfare Science, 3, Margulis, S. W., Hoyos, C., & Anderson, M. (2003). Effect of felid activity on zoo visitor interest. Zoo Biology, 22, Marker, L., Dickman, A. J., Mills, M. G. L., & Macdonald, D. W. (2010). Cheetahs and ranchers in Namibia: A case study. In D. W. Macdonald & A. J. Loveridge (Eds.), Biology and conservation of wild felids (pp ). Oxford: Oxford University Press. Marker-Kraus, L. (1997). History of the cheetah Acinonyx jubatus in zoos International Zoo Yearbook, 35, Markowitz, H., & LaForse, S. (1987). Artificial prey as behavioral enrichment devices for felines. Applied Animal Behaviour Science, 18, Marseille, M. M., Elands, B. H. M., & van den Brink, M. L. (2012). Experiencing polar bears in the zoo: Feelings and cognitions in relation to a visitor's conservation attitude. Human Dimensions of Wildlife, 17, Martin, P., & Bateson, P. (2007). Measuring behaviour: An introductory guide (3rd ed.). New York: Cambridge University Press. Mason, G. (2006). Stereotypic behaviour in captive animals: Fundamentals, and implications for welfare and beyond. In G. Mason & J. Rushen (Eds.), Stereotypic behaviour in captive animals: Fundamentals and applications for welfare (2nd ed.). Oxfordshire, UK: CAB International. Mason, G., Clubb, R., Latham, N., & Vickery, S. (2007). Why and how should we use environmental enrichment to tackle stereotypic behaviour? Applied Animal Behaviour Science, 102, Mason, G. J. (2010). Species differences in responses to captivity: Stress, welfare and the comparative method. Trends in Ecology and Evolution, 25, Mason, P. (2000). Zoo tourism: The need for more research. Journal of Sustainable Tourism, 8, Mbaiwa, J. E. (2004). The socio-economic benefits and challenges of a communitybased safari hunting tourism in the Okavango Delta, Botswana. The Journal of Tourism Studies, 15, McKinley, J., Buchanan-Smith, H. M., Bassett, L., & Morris, K. (2003). Training common marmosets (Callithrix jacchus) to cooperate during routine laboratory procedures: Ease of training and time investment. Journal of Applied Animal Welfare Science, 6, McPhee, M. E. (2002). Intact carcasses as enrichment for large felids: Effects on onand off-exhibit behaviors. Zoo Biology, 21, Medina, M. (2005). The Bronx Zoo's tiger mountain: An exhibit as enrichment. Paper presented at the Seventh International Conference on Environmental Enrichment, New York. 285

304 Meehan, C. L., & Mench, J. A. (2007). The challenge of challenge: Can problem solving opportunities enhance animal welfare? Applied Animal Behaviour Science, 102, Melfi, V. A. (2009). There are big gaps in our knowledge, and thus approach, to zoo animal welfare: A case for evidence-based zoo animal management. Zoo Biology, 28, Mellen, J., & Sevenich MacPhee, M. (2001). Philosophy of environmental enrichment: Past, present, and future. Zoo Biology, 20(3), Mellen, J. D. (1991). Factors influencing reproductive success in small captive exotic felids (Felis spp.): A multiple regression analysis. Zoo Biology, 10, Mellen, J. D., & Shepherdson, D. J. (1997). Environmental enrichment for felids: An integrated approach. International Zoo Yearbook, 35, Michalski, F., Boulhosa, R. L. P., Faria, A., & Peres, C. A. (2005). Human-wildlife conflicts in a fragmented Amazonian forest landscape: Determinants of large felid depredation on livestock. Animal Conservation, 9, Miller, L. J. (2012). Visitor reaction to pacing behavior: Influence on the perception of animal care and interest in supporting zoological institutions. Zoo Biology, 31, Miller, L. J., Bettinger, T., & Mellen, J. D. (2008). The reduction of stereotypic pacing in tigers (Panthera tigris) by obstructing the view of neighbouring individuals. Animal Welfare, 17, Mobley, C., Vagias, W. M., & DeWard, S. L. (2010). Exploring additional determinants of environmentally responsible behaviour: The influence of environmental literature and environmental attitudes. Environment and Behavior, 42, Morgan, J. M., & Hodgkinson, M. (1999). The motivation and social orientation of visitors attending a contemporary zoological park. Environment and Behavior, 31, Morgan, K. N., & Tromborg, C. T. (2007). Sources of stress in captivity. Applied Animal Behaviour Science, 102, Moss, A., & Esson, M. (2010). Visitor interest in zoo animals and the implications for collection planning and zoo education programmes. Zoo Biology, 29, Moss, A., Plowman, A., Hosey, G., Stevenson, M., & Whitehouse-Tedd, K. (2013). Surveys and questionnaires. In J. Bishop, G. Hosey & A. Plowman (Eds.), Handbook of zoo research, guidelines for conducting research in zoos (pp ). London: BIAZA. Murphy, I. G., Dempsey, M. P., & Kneafsey, B. (2007). Tiger bite in captivity. European Journal of Plastic Surgery, 30, Myers, O. E., Saunders, C. D., & Birjulin, A. A. (2004). Emotional dimensions of watching zoo animals: An experience sampling study building on insights from psychology. Curator, 47, Nyhus, P. J., & Tilson, R. (2010). Panthera tigris vs Homo sapiens: Conflict, coexistence, or extinction. In R. Tilson & P. J. Nyhus (Eds.), Tigers of the world: The science, politics, and conservation of Panthera tigris (pp ). London: Academic Press. Nyhus, P. J., Tilson, R., & Hutchins, M. (2010). Thirteen thousand and counting: How growing captive tiger populations threaten wild tigers. In R. Tilson & P. J. Nyhus (Eds.), Tigers of the world: The science, politics, and conservation of Panthera tigris (pp ). London: Academic Press. 286

305 Nyhus, P. J., Tilson, R. L., & Tomlinson, J. L. (2003). Dangerous animal in captivity: Ex situ tiger conflict and implications for private ownership of exotic animals. Zoo Biology, 22, O'Donovan, D., Hindle, J. E., McKeown, S., & O'Donovan, S. (1993). Effect of visitors on the behaviour of female cheetahs Acinonyx jubatus and cubs. International Zoo Yearbook, 32, Ogden, J., Routman, E., Vernon, C., Wagner, K., Winsten, K., Falk, J.,... Reinhard, E. (2004). Inspiring understanding, caring and conservation action: Do we or don't we? Communique, Okello, M., Manka, S. G., & D'Amour, D. E. (2008). The relative importance of large mammal species for tourism in Amboseli National Park, Kenya. Tourism Management, 29, Oli, M. K., Taylor, I. R., & Rogers, M. T. (1994). Snow leopard (Panthera uncia) predation on livestock: As assessment of local perceptions in the Annapurna Conservation Area, Nepal. Biological Conservation, 68, Packer, C., Swanson, A., Ikanda, D., & Kushnir, H. (2011). Fear of darkness, the full moon and the nocturnal ecology of African lions. PLoS ONE, 6, e doi: /journal.pone Packer, J., & Ballantyne, R. (2002). Motivational factors and the visitor experience: A comparison of three sites. Curator, 45, Palmeira, F. B. L., Crawshaw Jr., J. G., Haddad, C. M., Ferraz, K. M. P. M. B., & Verdade, L. M. (2008). Cattle depredation by puma (Puma concolor) and jaguar (Panthera onca) in central-western Brazil. Biological Conservation, 141, Patterson, B. D., Kasiki, S. M., Selempo, E., & Kays, R. W. (2004). Livestock predation by lions (Panthera leo) and other carnivores on ranches neighboring Tsavo National Parks, Kenya. Biological Conservation, 119, Pearson, E. L., Dorrian, J., & Litchfield, C. A. (2013). Measuring zoo visitor learning and understanding about orangutans: evaluation to enhance learning outcomes and to foster conservation action. Environmental Education Research, 19, Pearson, E. L., Lowry, R., Dorrian, J., & Litchfield, C. A. (2014). Evaluating the conservation impact of an innovative zoo-based educational campaign: 'Don't palm us off' for orang-utan conservation. Zoo Biology, 33, Penn, L. (2009). Zoo theater's influence on affect and cognition: A case study from the Central Park Zoo in New York. Zoo Biology, 28, Phillips, C., & Peck, D. (2007). The effects of personality of keepers and tigers (Panthera tigris tigris) on their behaviour in an interactive zoo exhibit. Applied Animal Behaviour Science, 106, Phillips, M., Grandin, T., Graffam, W., Irlbeck, N., & Cambre, R. (1998). Crate conditioning of bongo (Tragelaphus eurycerus) for veterinary and husbandry procedures at the Denver Zoological Gardens. Zoo Biology, 17, Portney, L. G., & Watkins, M. P. (2009). Foundations of clinical research: Applications to practice. New Jersey: Pearson. Poulsen, E., & Miller, L. (1996). Got a tiger by the tug. The Shape of Enrichment, 5, Povey, K. D., & Rios, J. (2005). Using interpretive animals to deliver affective messages in zoos. Journal of Interpretation Research, 7, Powell, D. M. (1995). Preliminary evaluation of environmental enrichment techniques for African lions (Panthera leo). Animal Welfare, 4,

306 Prescott, M. J., & Buchanan-Smith, H. M. (2003). Training nonhuman primates using positive reinforcement techniques. Journal of Applied Animal Welfare Science, 6, Price, E. E., & Stoinski, T. S. (2007). Group size: Determinants in the wild and implications for the captive housing of wild mammals in zoos. Applied Animal Behaviour Science, 103, Quirke, T., & O'Riordan, R. M. (2011a). The effect of a randomised enrichment treatment schedule on the behaviour of cheetahs (Acinonyx jubatus). Applied Animal Behaviour Science, 135, Quirke, T., & O'Riordan, R. M. (2011b). The effect of different types of enrichment on the behaviour of cheetahs (Acinonyx jubatus) in captivity. Applied Animal Behaviour Science, 113, Quirke, T., O'Riordan, R. M., & Zuur, A. (2012). Factors influencing the prevalence of stereotypical behaviour in captive cheetahs (Acinonyx jubatus). Applied Animal Behaviour Science, 142, Ray, J. C. (2005). Large carnivorous animals as tools for onserving biodiversity: Assumptions and uncertainties. In J. C. Ray, K. H. Redford, R. S. Steneck & J. Berger (Eds.), Large carnivores and the conservation of biodiversity (pp ). Washington: Island Press. Reade, L. S., & Waran, N. K. (1996). The modern zoo: How do people perceive zoo animals? Applied Animal Behaviour Science, 47, Reinhardt, V. (2003). Working with rather than against macaques during blood collection. Journal of Applied Animal Welfare Science, 6, Ross, S. R., & Gillespie, K. L. (2009). Influences on visitor behavior at a modern immersive zoo exhibit. Zoo Biology, 28, Roynon, J. (2000). It's a cat's life! The effectiveness of simulated prey on a pair of Sumatran tigers (Panthera tigris sumatrae). Paper presented at the Proceedings of the 2nd Annual Symposium in Zoo Research, London. Rustiati, E. L., & Chanin, P. (2000). Saving the Sumatran tiger: The important role of captive animals in leading the fieldwork. Paper presented at the 2nd Annual Symposium in Zoo Research, Federation of Zoological Gardens of Great Britain and Ireland, London. Samuels, A., & Spradlin, T. R. (1995). Quantitative behavioral study of bottlenose dolphins in swim-with-dolphin pograms in the United States. Marine Mammal Science, 11, Saunders, C. D., Brook, A. T., & Myers, O. E. (2006). Using psychology to save biodiversity and human well-being. Conservation Biology, 20, Savastano, G., Hanson, A., & McCann, C. (2003). The development of an operant conditioning training program for new world primates at the Bronx Zoo. Journal of Applied Animal Welfare Science, 6, Schafer, E. H. (1968). Hunting parks and animal enclosures in ancient China. Journal of the Economic and Social History of the Orient, 11, Schapiro, S. J., Bloomsmith, M. A., & Laule, G. E. (2003). Positive reinforcement training as a technique to alter nonhuman primate behavior: Quantitative assessments of effectiveness. Journal of Applied Animal Welfare Science, 6, Schiller, H. J., Cullinane, D. C., Sawyer, M. D., & Zietlow, S. P. (2007). Captive tiger attack: Case report and review of the literature. The American Surgeon, 73,

307 Sekhar, N. U. (2003). Local people's attitudes towards conservation and wildlife tourism around Sariska Tiger Reserve, India. Journal of Environmental Management, 69, Sellinger, S. L., & Ha, C. J. (2005). The effects of visitor density and intensity on the behavior of two captive jaguars (Panthera onca). Journal of Applied Animal Welfare Science, 8, Shepherdson, D. J., Carlstead, K., Mellen, J. D., & Seidensticker, J. (1993). The influence of food presentation on the behavior of small cats in confined environments. Zoo Biology, 12, Shettel-Neuber, J. (1988). Second and third generation zoo exhibits: A comparison of visitor, staff, and animal responses. Environment and Behavior, 20, Shoemaker, A. H., Maruska, E. J., & Rockwell, R. (1997). Minimum husbandry guidelines for mammals: Large felids. New York: American Association of Zoos and Aquariums. Shyne, A. (2006). Meta-anayltic review of the effects of enrichment on stereotypic behavior in zoo mammals. Zoo Biology, 25(4), Siegal, S., & Castellan, N. J. (1998). Nonparametric statistics for the behavioral sciences (2nd ed). New York: McGraw-Hill. Skibiel, A. L., Trevino, H. S., & Knaugher, K. (2007). Comparison of several types of enrichment for captive felids. Zoo Biology, 26, Skibins, J. C., & Powell, R. B. (2013). Conservation caring: Measuring the influence of zoo visitors' connection to wildlife on pro-conservation behaviors. Zoo Biology, 32, Skibins, J. C., Powell, R. B., & Hallo, J. C. (2013). Charisma and conservation: charismatic megafauna's influence on safari and zoo tourists' pro-conservation behaviors. Biodiversity and Conservation, 22, Smith, L., Broad, S., & Weiler, B. (2008). A closer examination of the impact of zoo visits on visitor behaviour. Journal of Sustainable Tourism, 16, Spenceley, A. (2007). Responsible tourism practices by South African tour operators: Survey results from participants at the 2006 Tourism Indaba. International Centre for Responsible Tourism - South Africa. Retrieved from b.pdf Stark, B. (2005). The use of carcass feeding to enhance animal welfare. Paper presented at the Seveneth International Conference on Environmental Enrichment, New York. Stoinski, T. S., Allen, M. T., Bloomsmith, M. A., Forthman, D. L., & Maple, T. L. (2002). Educating zoo visitors about complex environmental issues: Should we do it and how? Curator, 45, Sunquist, M. E., & Sunquist, F. (2002). Wild cats of the world. Chicago, IL: University of Chicago Press. Suryawanshi, K. R., Bhatnagar, Y. V., Redpath, S., & Mishra, C. (2013). People, predators and perceptions: Patterns of livestock depredation by snow leopards and wolves. Journal of Applied Ecology, 50, Swaisgood, R. R., & Shepherdson, D. J. (2005). Scientific approaches to enrichment and stereotypies in zoo animals: What's been done and where should we go next? Zoo Biology, 24, Swanagan, J. S. (2000). Factors influencing zoo visitors' conservation attitudes and behavior. Journal of Environmental Education, 31,

308 Swanson, W. F., Johnson, W. E., Cambre, R. C., Citino, S. B., Quigley, K. B., Brousset, D. M.,... Wildt, D. E. (2003). Reproductive status of endemic felid species in Latin American zoos and implications for ex situ conservation. Zoo Biology, 22, Swim, J. K., Clayton, S., & Howard, G. S. (2011). Human behavioral contributions to climate change: Psychological and contextual drivers. American Psychologist, 66, Szokalski, M.S., Litchfield, C.A., & Foster, W.K. (2013). What can zookeepers tell us about interacting with big cats in captivity? Zoo Biology, 32, Tait, P. (2012). Wild and dangerous performances: Animals, emotions, circus. New York: Palmgrave Macmillan. Thapar, V. (2013). Tiger fire: 500 years of the tiger in India. New Delhi: Aleph. Thapar, V., Thapar, R., & Ansari, Y. (2013). Exotic Aliens: The lion and the cheetah in India. New Delhi: Aleph. Thomas, L. (1997). Retrospective power analysis. Conservation Biology, 11, Thomas, P. (2004). The Bronx Zoo's tiger mountain: Enrichment at the forefront of exhibitry, Association of Zoos and Aquariums Annual Conference. Thomas, P., Balme, G., Hunter, L., & McCabe-Parodi, J. (2005). Using scent attractants to non-invasively collect hair samples from cheetahs, leopards and lions. Animal Keepers' Forum, 7, Thompson, V. D. (1989). Behavioral response of 12 ungulate species in captivity to the presence of humans. Zoo Biology, 8, Tilson, R., & Nyhus, P. (1998). Keeping problem tigers from becoming a problem species. Conservation Biology, 12, Tilson, R., Siswomartono, D., Manansang, J., Brady, G., Armstrong, D., Traylor- Holzer, K.,... Nyhus, P. (1997). International co-operative efforts to save the Sumatran tiger. International Zoo Yearbook, 35, Tresz, H. (1997). Providing enrichment at no cost. The Shape of Enrichment, 6, 1-4. Tribe, A., & Booth, R. (2003). Assessing the role of zoos in wildlife conservation. Human Dimensions of Wildlife: An International Journal, 8, Trone, M., Kuczaj, S., & Solangi, M. (2005). Does participation in dolphin-human interaction programs affect bottlenose dolphin behaviour? Applied Animal Behaviour Science, 93, Umapathy, G., Sontakke, S. D., Srinivasu, K., Kiran, T., Kholkute, S. D., & Shivaji, S. (2007). Estrus behavior and fecal steroid profiles in the Asiatic lion (Panthera leo persica) during natural and gonadotropin-induced estrus. Animal Reproduction Science, 101, Van Metter, J. E., Harringer, M. D., & Bolen, R. H. (2008). Environmental enrichment utilizing stimulus objects for African lions (Panthera leo leo) and Sumatran tigers (Panthera tigris sumatrae). BIOS, 79, Vining, J. (2003). The connection to other animals and caring for nature. Human Ecology Review, 10, Visscher, N. C., Snider, R., & Vander Stoep, G. (2009). Comparative analysis of knowledge gain between interpretive and fact-only presentations at an animal training session: An exploratory study. Zoo Biology, 28, Walker, S. (2001a). Zoological gardens of Asia. In V. N. Kisling (Ed.), Zoo and aquarium history (pp ). Hoboken: CRC Press. Walker, S. (2001b). Zoological gardens of India. In V. N. Kisling (Ed.), Zoo and aquarium history: Ancient animal collections and zoological gardens (pp ). Hoboken: CRC Press. 290

309 Walpole, M. J., & Leader-Williams, N. (2002). Tourism and flagship species in conservation. Biodiversity and Conservation, 11, Wang, S. W., & Macdonald, D. W. (2009). The use of camera traps for estimating tiger and leopard populations in the high altitude mountains of Bhutan. Biological Conservation, 142, Ward, P. I., Mosberger, N., Kistler, C., & Fischer, O. (1998). The relationship between popularity and body size in zoo animals. Conserv Biol, 12, WAZA. (2005). Builging a future for wildlife - The World Zoo and Aquarium Conservation Strategy. Retrieved from Weiler, B., & Smith, L. (2009). Does more interpretation lead to greater outcomes? An assessment of the impacts of multiple layers of interpretation in a zoo context. Journal of Sustainable Tourism, 17, White, B. C., Houser, L. A., Fuller, J. A., Taylor, S., & Elliot, J. L. L. (2003). Activitybased exhibition of five mammalian species: Evaluation of behavioral changes. Zoo Biology, 22, Whitham, J. C., & Wielebnowski, N. (2009). Animal-based welfare monitoring: Using keeper ratings as an assessment tool. Zoo Biology, 28, Whitworth, A. W. (2012). An investigation into the determining factors of zoo visitor attendances in UK zoos. PLoS ONE, 7. doi: /journal.pone Wielebnowski, N., & Brown, J. L. (1998). Behavioral correlates of physiological estrus in cheetahs. Zoo Biology, 17, Wielebnowski, N. C. (1999). Behavioral differences as predictors of breeding success in captive cheetahs. Zoo Biology, 18, Wielebnowski, N. C., Fletchall, N., Carlstead, K., Busso, J. M., & Brown, J. L. (2002). Noninvasive assessment of adrenal activity associated with husbandry and behavioral factors in the North American Clouded leopard population. Zoo Biology, 21, Wielebnowski, N. C., Zielger, K., Wildt, D. E., Lukas, J., & Brown, J. L. (2002). Impact of social management on reproductive, adrenal and behavioural activity in the cheetah (Acinonyx jubatus). Animal Conservation, 5, Williams, B. G., Waran, N. K., Carruthers, J., & Young, R. J. (1996). The effect of a moving bait on the behaviour of captive cheetahs (Acinonyx jubatus). Animal Welfare, 5, Wilson, M. K., A., Poline, L., Bloomsmith, M., & Maple, T. (2003). Post-occupancy evaluation of Zoo Atlata's giant panda conservation center: Staff and visitor reactions. Zoo Biology, 22, Woods, B. (1998). Animals on display: Principles for interpreting captive wildlife. Journal of Tourism Studies, 9, Woods, B. (2002). Good zoo/bad zoo: Visitor experiences in captive settings. Anthrozoos, 15, Xiaofeng, L., Yi, Q., Diqiang, L., Shirong, L., Xiulei, W., Bo, W., & Chunquan, Z. (2011). Habitat evaluation of wild Amur tiger (Panthera tigris altaica) and conservation priority setting in north-eastern China. Journal of Environmental Management, 92, Yamazaki, K., & Bwalya, T. (1999). Fatal lion attacks on local people in the Luangwa Valley, Eastern Zambia. South African Journal of Wildlife Research, 29, Young, R. J. (2003). Environmental enrichment for captive animals. Oxford: Blackwell Science. 291

310 Zeppel, H., & Muloin, S. (2008). Conservation benefits of interpretation on marine wildlife tours. Human Dimensions of Wildlife: An International Journal, 13,

311 Appendix A Contribution of authors to each publication 293

312 This section provides a statement by the PhD candidate (Monika Szokalski) describing the contribution of each author to each of the four publications/manuscripts generated from this thesis. Monika is the first author of all publications and manuscripts as a result of her primary contribution to the work. The criteria for authorship used are defined in Section 5 of the following document: Australia Government. (2007). Australian code for the responsible conduct of research: Revision of the joint NHMRC/AVCC statement and guidelines on research practice. Canberra: Attorney General s Department. Available online at: Contribution of each author: Chapter Conception and design of project 1 M.S., C.L., W.F. 4 M.S. C.L., W.F. 5 M.S. C.L., W.F. Analysis of research data Interpretation of research data Drafting significant parts of the work Critically revising work so as to contribute to the interpretation M.S. M.S. M.S. M.S., C.L., W.F. M.S. M.S., W.F., M.S. M.S., C.L., C.L. W.F. M.S., W.F. M.S., W.F., M.S. M.S., C.L., C.L. W.F. All collaborating authors agree to their contribution to each publication as stated in the table above. Monika Szokalski (M.S.) Carla Litchfield (C.L.) Wendy Foster (W.F.) Note: As supervisors of Monika s PhD, Carla Litchfield and Wendy Foster have been involved throughout all stages of the research process. However, in the case of publications, their contributions would fall into the categories described in the above table. 294

313 Appendix B Published version of Chapter One 295

314 Contents ltsls available at SeiVerse ScienoeOuect Applied Animal Behaviour Science jourl'lal homep ag e: www. el iievie r.eo1'l'l/l oca te/a pplani1'l'i Review Enrichment for captive tigers (Panthera tigris): Current knowledge and future directions Monika S. Szokalski " '. Carla A Litchfield'. Wendy K. f oster St'Jioot Of h)'(ho~. So(iol I\~~~ So(iol l'lii'i(y. (fii1'1i"sil)'(l/so;:l'9 Ausooliu. Cl'O b ) (O<!Sf:""riO" Art. Z«IS S0111A A~C~ll. fiomt- kocid. M\"_,ick SA 500Q. A;:J.tr(il.fo Adi'l'oidt. SA soo I, A11StJO!lt1 A R T I C L! l t<.' F O A ~ STR A C T Arl'.i(.lr hisl07:!\c~p(t'o 21 ft~l'ully 2C 12 Av~il;rl;f(" Ofllint 16 M.al(h 2Ct2 l<()'..vrd1. l'ttl.li"lt Cm id'lment O!f,n.'IOty ti~rielt :.'at'n! S«i.al tnrie:lt1~ So d.:i!!'lousi~ Hum.:.i'l tn1et tlon ~ytboobg)o:.ti.,.."u bc'!ng: fllviroru'oc'nt.ol t nrk hm('l'lt is ~ «immon ;)pptoocb for ;)d-ot t ssing st<":reoiypk bc-ll;)\'loor1n C.tpL~\".w:mab. Li'.ke ni.w) li.g cats. lit;.m (A:l11thmt 1.(!;rit).:1e re1ww1ted f1ji1 their ~t~ typk p.:dng. )<et 1el.:.1ivd y l [(t~e is l:nown.:.bout o;iti:n.tl enridlmei:t ror this speeies.gh-e1i tht l.:..."t:e p11jipo1tion ut ti:nt wild tigtrs s;iffld e1'1!;.iged i1i lrnnti:ig.!e\"di:ig..:.nd tenitoi:ial bel:.:.viouts, rese.:1 ch to d.tte I:.!$ fi>t'ustd a!rm)sl primarily ot: enridlmei:t devici:ll.:.i:d ltd! niqi.tes ll:.:.! s1.imul.:.te the$e'. Suettss has been!c>u:id i1i 1m,~I food item$, a!te1ed!e~i:ii roorioo ;)l'ld fooel displ.ays,.lnd no\'<'i (O}'Sfobjens: olf.mory stim\ll.ittlon vi;) fhe inrroc:hk tion of novc'l S«nu.uid cnd os'l.lrt rot.ltions:.uid incrt.;,sc:s 1n cnc ooure sillc'. In «intr;)sl.. 1:n1e <attention h;)s b«o p.aid (0 so(i;)) cnrw:tunc~n(, Althootll v.lnous iooo house tigers 1n sod ;)I groups. t~ limited titcr.orutt in~'t'stiz,;tlng lhis offers cooflk u nz «iocluoons bout the eftiects IX'1 t?l\".:.1:~1:1als. Fu1tlw1. h1w1an tige1 fo te.ttt.i0:1 has been ig..-iored in the lite,... atuie. de~p!t e its inc1easing occur.e11'r. Vii th nwre 14;en \"xistirq; in t'j;itivity Llia:i fo t.l:e wild, it if impe1ath~ tlw;t uur under~t.:...-idi no. ortq;e, ei:!'ielmwnt is exp:u\tled, partkul:.rly with 1eg.:.1d to tliese unde1deve'.klped.:1eas. Q)ntMtS 1. IJ'ltroc:ilJCOOO " ' " ' " ' " ' " ' " ' " ' "... ' " ' " ' 1.1. ~ It(('(! t0 pr<'scrvt n;)tur;)i bch;)vkxjrs..... i i 1.2. ~ tff«tsof c<a;iovity OO bcll;)vloor.....,,,,,,,,,,,,,..,..,..,.. i 13, 'Vlild versus 'c.apd vc' ugcr bch.iviour,..,..,...,,,,,,,,,,,,,, i 2. E11vi11Ji1::1wnt.:I enridlmei:t!oi:'c.:.pth-t tigers l 3. E1:rid ::1wnt tetliniquest.i.:rt;eti.-ig t:atur.:.l feed: 1t:.t1:.d huntirq; be-lw.vio11.. "'S l 3.1. Novel food itern<l as e1idchment l 3.2. Alteml reeding roulints.wd mrt?lod:s of rood disp.\ly <is tniidlmrnt Novel roys/objects ;as ennd'lmcn(,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,..,..,..,..,.., , (>!fi<tory stim1jl.ation <1'.S ennctuncn(,,,,,,..,..,..,..,..,..,...,,,,,,,, 4 4, 1. /\ddlrion o4 S«nts.lnd ol>..i<'<ts to tiger cnckls'l.in:'.s,,,,,,,,,,,,,,,,,..,..,..,..,..,...,, 4 4,2, t ndo5ur(': rot tloq'is,,.,,.,,..,.,,.,,.,,.,,.,... " S s. Ei:d osures as e1idchment s.1. tnd!t>sl!.res.tte s cor~spon:l!ng ~I?«. Tel: Gl «-5!14: ax: GI 3 lll0' E>'l:'di a~ Momb5%0bfsber\ln.U.Mu~ (M.S. S:.ob!ski). o IGll 159 1/S see froi'l1 r.:'l;au er o 20l 2 lllscvler a v. A!! ~'l'lrs ttstr.'l;'cl. doi: li,l).:.;ipl.:ii'l!r.:'l-20 l UY..l

315 6. Sod~ snmolaoon <is cnricll.jn('nt,,,, ,..,..,..,..,..,..,,,..,..,,, ,..,..,..,..,..,..,..,..,......, 6.1. Sod.ill housing,,,,,,,,, ,..,..,..,..,..,'" ",,,,,,,,, ,..,..,..,..,..,..,..,..,.., !. Homan lntera(oon <is enricll.jn('nt..,..,..,..,.., ,, ,..,..,..,..,..,..,..,..,...,..,.. ;; 6.2.L Pllsit.ive reinr-0rceme1:t tr.i.ining U11Sb:.t:dry sty~i.' Mtthudok>gic.d coosidl.'r.i!ians!oi:" foluil' ~l'ardl C«teusi<oiu S A<koowlOOgicmenu ,(fcrcnctS JntJ'Odu(tion t he dger (f'llnrhero rfgris) is the largest living cat (Sonquht, 20 10) and one of the best-known large mammals (lr.ternational Union for Conservation of Nature. t 996). Sld l~'. 1heir numbers in the \\'ild are declining signifi(antl/. with a1 least three subsped es beoomingextinct sin«tht i 9'10s (Seidensticker et al ). In particular. 1he Som.lttan tiger (J'. dgris.rumarrae) is now Us1ed ascriticallyencangered. \vi th around 300 individuals proposed to be li\'in~i n the wild(seidens1ickeret <1L 20 10). In their natural habilat (various and disperse Asian range states) tigers <1re facing serious conflict \Yith humans over resources. \vith huw n <1cti..,ity largely causing their decline. In addition to in situ efforts, zoos<1nd wildlife parj.:s play a crucial role in the conservation of this species. It is believed that 1here are no\.v more tigers in cap1ivity - including zoos<1nd circuses - than in the \vild (:-..'yhus et al ). Therefore. it appears timely 10 explore what is kno'l.vn to date <tbout 1he best practices fo r enriching tigers in zoos. in order 10 ob1ain a grea1er understanding of how 10 improve their lives in c<1pti\'hy and 10 identify are<ts for future research. This is i:en inent. given tha1 captivity may likely be their only saftgu.ard from extinction in the ne<ir future. I. l. The need ro preserve nantr0i behaviours Re«ndy, there has been.l grea1er focus on the role that zoos anc other captive institutions can play in the conserv.n ion o: endangered species (Hutchins <1nd Smith, 2003: Kirk\, oo:.t 2003: Rusti.ati <1nd Chanin. 2000) Given the high numbe r:: o(tigcr:: held in c.;iptiviry ( Ng >"ll'ld Ncm.Of.l ). it is imponant 10 con.sider 1he impact of 1his on the beha..,iocr of the animals for two reasons. firstly. given 1hat nau ral selection may 'engineer' a population 10 fi t its capth't environment. \ve must ensure minimal genetic change in caplive animals (Kirk\, ood. 2003: ~1ason et al_ 2007). i 1i.s can result in captive-.adapted strains that may be less suited to the n.atural environmen1, creating significant problems if zoo populations<1re used to research 1he beh<\'iour of wild coun1erpans. if animals <1re pan of captive-breeding programs. or if they are 10 be reintroduced 10 the \vild (Kirkv.'OOd, 2003: Mason et al 2007). The second re<tson is out of considera1ion for an <1nima1"s psychological \\'ell-being (Hutchins and Smith. 2003). 'l'his corurn is based on the premise that animals have 1he c<tp.:cit>' for conscious <1wareness of p!e<tsant and/or unple<ts.;nt feelings. and tha1 they can therefore suffer \Vhen v.rtlfare i.s poor (Kirk\, ood. 2003). Behaviour c<tn be <-n indicator of the presence of these feelings. <1nd it is lherefore emplo)'ed as a means 10 assess and identify botl diminished and positive \\'eu-being. i.2. The tffecrs of captivity on behaviour 'l'he negative e!tec1s of captivity have been \videly <.cknowledged in the literature (e.g.. Mason ) anj!here i.s agenerajconsensus that the I imitations of a captiw tnvironment can!e<td to 1he performantt of stereot)'pic bch.lviour in m.lny $pe<ic:; p.1.;i:;on ct.;ii., :!00'1: Melle.., <.nd Se\'enich Maei)hee. 2001: Shyne. 2006: s,vaisgood an :! ~.hepherd.so n. 2005). In particular. Clubb and f\~aso n (200?) highlight that those animals \\'hich typ kall~ occupy latg? home ranges in the wild tend to fare v.'orse in cap1ivity <.nd are much more vulnerable to v.-elfare problems.. 'l'igers <.re one such spedesand a common problem facing zoos is!he presence of stereotypic pacing in these a nimal s (~1aso1 N aj ). In combining data from various observa1ionil nudies. Clubb and ~1 ason (2007) repon 1hat c<tptive adult tigers spend t6%(median) of behaviour scans engaged h nereotypic pacing. L3. ~Vild" \.'trs us capri~-e tiger bthaviojjr Not onl>' does 1he tiger's elusive. wide-ranging natur? make it difficult 10 11<1cJ.: how many of these <1nimals exin in the wild (Karanlh and Nichols. 2010). but also it limirs what is lmo\vn about 1heir behaviour. the availabli tnowledge. hov.-ever. indk <ttes tha1 the captive en..,ironment differs drama1k ally from '"hat these animals ar? dc!;igncd 10 experience..lnd i1 b not :;urpri:;ing th.l.1 tigcrn can fare poorly in capri\'hy. \Vild adult tigers typically live a solitary existence. occupying large home ranges (Sunquist ). 'l'hey spend mos1 of the day either i.n.active or pa1101jing their 1erritory (Zhen-sheng et <1L. 2002!. <.nd hun.t nocturnally (although some hunting c<1n occur during the daytime) (Sunquist ). In captivity, howtver.the environment for tigers is vastly different. They ar? often housed '"ith. or in close proximity 10. other tigers(as well as other species). and spa1ial <1nd financial cons1rainis testrk t encjosure size. resulting in s ignificant}~ smal!cr :erritories. f'un her. the absen«of a live prey base and. inste<td. the provision of pre-prepared die1sdoes not allow for the performance of their n.atural hunling behaviours <.nd energyexpenditure. 1.asrty. tigers are often expected n be active <tnd \'isible during zoo opening hours \\'hen visttors are present. \\'hich confl icts \\'ilh their na1ural activity pattern. Ci\'en these differences. it is not surprising 1h.;t 297

316 this speocits displays high lt'vt-ls or '1C":r'COtypic behaviour ln captivity (for a more detailed di.scusslon or tl\t impact of sud!. factors on captive animals. ~e ~torg11n and Tromborg. 2007). However. it is also recognised I hill 01her factors - such as an Individual animal's pcr$0nahty.tnd history - c.tn also contribute to thcsc.' bch.ivlours. 2. Environ.mental enrlchjmnt for c.apttw tlgtn Sttttotypic be'ha'vioc:r 1n upt:ivt t11-trs h.ls not bttn tpottd. and numtrous dfocts h.lvt bttn eomp1oytd to ttdutt IL Tht prima:l'y mt.lns for do&nl io hh bttn lhrough UM" or mvironmental enrichmm t t«hnlqoh anc! mere is.a rc-bm.~ly ~ bocy of hte:r lft prnalnl~ to enrich mtnt for aptiw :ends. In rttmt deadn. a number of lnphudons have implemented p roctuc':s to en.ha.nee tht husb!nd:ry and management or tht'sc': an lm.ils, adopdng an integrated, holistic approach (Hutchins and Smith. 2003; Uw et al : Mellen and Scvcnlch M.>cf'he1\ 2001). Uowl"Y'Cr, fl'\\' environment.:il C'1'1richmcn1 s-1ud!es have focused specifically on dgcrs alone. Much or this research orlgina1cs from the USA and foc:usc':l prim.irtlyon only rwo ctitr sub-species - Amur and S1;1m.uran. funher. papers in the pttr revie1.-.red literature.itt llmlccd in numbers. but a~ supponed by various confcrmcir procttd!np and.mec dow aaocnu (Stt T.lblt 1 for dtt.ai!s or pttr-~-1'.'d t11tr Mrichment srudies to d.atc). l br ~ cl these' C'ffotu ducuswd h.n'e focuwd on c:nl'knnrn1s th.at wp>t Neural fhding and huno.ng bdt.a\lloul"l. Wlll'I othtts also Jddmsing: olfactory scim.uuition nd mdowtt sltt. Two mas of titsser exploration.are $0Cio\I Mm;lttg!'Id human con111n (in the form of tr ining.and as a husbandry st)'lt). J, En.rlch.ment ttth.njquc':s ta.rgt:tln3 n.trur.il feeding.ind hunting behaviours Given that \\'ild tigers devote ;i l.irge proponion of their.ictivt' time lo hunting and consuming prty, it is not surpruing that the majority or documented eonvironrnerual c:nrlchmttt effons bavir add:r~ tl'ic'jc tw,tur.tl hunting.and fttd!ng behaviours. Wild ci,gffs. COftJUtnr a tangt" oi P"Y p.lflicu!arty -...itd bole.ind dttr:.and. in ll't.uaras \\'hrn bfp' species.n urwv.ail.tllt. sm.o\1"1' ptty, SIJdl.u Cita.l dttr.art tafll.'lt'd (Sgnqi,11R. 2010). In a.:!dkion CO l~ rimt spetll uptofinl pcul'llial hurll.lftc MUS and P.alkm;g: potential prey (Sunquist. 2010). 11ctn ttquir"f' a varltty ofbehaviours(suc:h as seizin.a. puu1ns. gnping:, and the "killing bite') and signifi can1 eo:ncrsy cxpr-nditurc to suettssfully bring do.,.,.n these an lm.alj (Scldenstld:er and Md>ougaJ, 1993). l) pically. wildtlgcn hun1 bctweenda\\'n.tnd dusk. but display some behavioural nexlblllcy and w ill hunt opponunisrically at any time Ir the shuatlon arises (Seldenstldcer a.ndp.1c0oug.tl, 199J: Sunqulst.2010). Hunts.trt not a~ways succt'ssful. and ti1cn Wlll ieflt!ra!ly take Lu1eo prey once a v.ttk (Sunqui.sc }. rctuml~to their kfd for up to six days until lutk 1Tm,uns (St-i~nstlcUr ~nd ~td>oupl. 1993). Despite this readily ~.ail.ableo knl1\\1rdgir lboct tht ltnpon~na of hu..'lring Mid c:omumlfll- ptty, aipcivt' fetd in& rourims for tigers ha\'t" t)'jlicaly 1nvolwd daily feeds of pre-prtp.ared 6eu. with gre: tt tml)h.tsis placed on nutritional v.alut. r.athtt than natural food (Law t'i ll 199?: ~1 t'll~n and Shtpherdson, t997i Captive rcr<1.in.a guide.lint's d.,.oc.i1c such dk!u 1 (Oierenfeld C"l al ). Since processed diets do not require natural roragins l.lc tics. energy cxpc-nd ~1u rc. or much use of.:ippcndagcs. c hlj has various physlologk al and psychological implicatloru for tigers (Law cc.ti : Skibiel et al_ 2007). lnclud ing greater pc-rlorm.:ince of stereotypic behaviour (Lyoiu et.il l997). fnrichtnc'nt effon s have therefore focutcd on stimuw.t1ns cnnc n.ttw'a! beh.a:\iocts and attcmptlq to repbc.att:. to lhc t'xttr.t possi>le. fttdlng in thir wild. This has bttn.actlltwd 1hrough the provision of "'1W"I bod items.. ~ co fttdltlc routines and food pnms1on..as \\'f:u.s novel 10)"$(ol!jttu to stimulatt hllf'.dlg bcll.avioun. 3.J. Novdfood lltt!u o:s mrldimenc Some rc1t.trchc:rs (Rashaw ct at. 2000: Stark. 2005: Skibiel cl aj., 2007) h.tw! explored novel food items.u e nrichn1cnt ror c.11ptlvc dgcrs. Results have shown th.it items such J.S anlmai bones, care.asses (whole and pa.rd.ii). and fish (fflotcn and llvt) un coniribt.tte to an incrcast' in narur.il hunrina bchl viours (sud'i. JS c.ipturing. mock killing and fut ptucjdng..u well as consumption). Such imns hi'lf also bt:c:n shown to lead to Ceae.ases in ste~ typic: bth.aviour. whh ntts v;uying for eacb study. &ar:sh.lw et.al (2003) s.aw Pldtll ~ wi1h tht prov'lsioa cl ft.ah. from 60S a.t t>.&wknr down to llll. \<o.'tich w.ts. susuantd l\\'o days.alttt provklon. Despite lhis noublt ckc:rust'. W result w.is Mft-SIJRll"tunt: ~'l'f. the.authors.ittril'altc this to the s.m.au s.11mple sitt (rt.. 2). Similarly. in.a conritr en«proceed.in.a, Sc.ark (2005) discusses how padng wu compk tt'ly.abolished w!1h the provision of calr care.asses to a tigress. or nott'4 howctver, thi.s individual pa«d al.t r.ii c of9% orbascjine scans,dtmonsira.ting a lessdranic rcduc tion in this behaviour 1han those in Bashaw ct l (2003) study. Rcg.trdlC'ss. both resulu indica1e the SOCCC$.S ofthlj enrichment te<hnlquc Atrmd /ttdbti roi.:rt.'lll's Ol'ld methods of food dtsplgy OS mndlmtflt lttsc.arcb bu~ txp1ottd a.."t.ercrions co food Pf'OVI" "' ln m atumpc co rc-pllat.e tht... ;Id. wheft' ogr:rs do not hunt daily.somt a:pcl\~ lnstiturions do nck fttd thf'trugit:rs e"\'n)' da.y..-. liugcstt'd 1n guidelines fot housing c.apt we feuds (DicRt":nfC":ld cc al ). Howevrr. this may.accu ally Jc.ad to.in 'lnac1se In pacing on the non-feeding d.a.ys. lyons et al,. (1997) found that some felid species (UCH'IJ. chtttah. jungle c.i1. Stbtrian lynx. jaguar, leopard. Pcrslan leopard..lnd Slbcrllln leopard) pa«d significantly more on non-focdlngd.iys rhanon fhding days. "'~osr notjbly, l l SS decrease in IJ)adnt(from 18" on non feeding days. ton on fttdingd.iys) w~sobs.erwd In a fem.a~ jaguar. ttowe.. -c:r. In this s..me study. chrtt Amur rigers fed on.i thttt4.ay cydt e..,.flmf'!i Nia...,co.tnn. llullt.l ~tkfl..,.. fl ~ """'& ~ ""'.. ~_..~ (( TM.-, ,.-.~Wt'~--~~ ~ c<... olllb,., fl~-0~.lciol.p l}. 298

317 T;Jl>lt 1 OK.:..!s of ugtr cr.ncllr.:'im: p.:ipers :i :he Pl"C'f tl"\'lev-'ffi!itf!r.:ir.1~. Entimmcnt c.:.:~ Enrlmmtn1 type "'".1mll<'rfa:1>4?K!cs of 11.3tn ""'~ /iutllor(to l'«dlfl:$, Uve!hb.:ind bonoc$ l. 1 S-.im;it r.in USA Suhaw ti.:il f200l) rcmjns 110XtS l.2nnut ():rope jer.tl)'.lml SCbrn!d 12002) Alt('mS~i J'lt ()'dt l,2 M u l.11-0ns f1.t.{1991) tls'!'l.j :d bones 1,2 OSA Skil;i('I te 4{2001) Huntir'lt l'~iic.;,nil'! ;1nd.;mi M01 so..tru.is. V" OSA M.i.it.ov-'kl.;,m,I Wkl1K ( 1961) M.:1,,1 kt.rs.'in" frort( n?0000 bas l,) Sum;il <tn OSA \,1;n \ ltllt ~n,,1, (2~) O!faUOI)' Spk t s ).2 ' USA Sk!lllc! t< -". (2007) ExMb!t flx.:.:iom l.o S-.1m.:ic~.in: l.oamllf USA \\ 11!1e ei.:11. (2003) Ztbf".1 d ur..;.:.00 s n-nttd sq-.1~1'1 l.l S-.1m.11~.1n USA \!.:.n Mer:er et ~I. (2008) Endos1nt Sil:t- (95 1 tto~ ~s 1 ~ 1TY') 1,1 Sum;il <tn OSA t\lsnw N.aL(2001) Sil:t () Ul 111~ - (Oi"li'~d lo ~S in v.-il).is 1,2 M ou l,y(lnsf1.t.{1991) ff'ch)w:\'s) "" Sil:t (tw<i ('1JCIOS0> '(') - $i(('s 1msprtill('(I) t,c Sum.;,l <tn OSA \\!hilt ('I,al, (2003) SXl.t! Sod.:.1 to-.1s!ng:(poiw W!f'S'".1s Ul!!t:ity 0.15 ( muld ple sub-sl)c'a.es J OI' Koc.w:ket.ll.(:.10001!IOUS1t1$,) ():rope Sod.:.1 to-.1s!ng:(two p-.1ps - \,ity!ng: O.li USA M!Uer ~r.d kuh.lr (: <Ot~~in.-tions) NII: Son:.- sr.kllcs ~P?Nr :wic~ due lo tl!em docun:.-r.dr..g: nxtt fh;lr. or.c (otr.:'i ol er.nclln:.-r.c. Sub-Silt~ oftit:ff' s unlder.c!f~tllyllrld. ~ Gender of 11.3tr/s unxle1111.fll'd. "" "'""' (whereby they \vere fed r:very third day, rather 1han every consecuti\'e day) did not pace at all during 1he experiment <1nd so 1he effect on sterec cypy cannct be determined. No other behaviours associated \Vilh feeding \vere ana1> sed. r unherexploration in10 the roleof1;his form of enrichment is requirod. t he effect of feeding boxes has also been explored. i'o examine whether pacing in tigers.is caused by ftuscra1ed <1ppetiti\'e foraging behaviours.. jenny <1nd Schmid (2002) explored the impact of feeding boxes (\vhereb>' food \vas placed in <1 box <1ndcouldonlybe<1c:«ssed if opened by the <1nimal)on one male and one female Amur tiger. Both tigers demonstrated a decrease in pi!d tl$ with the addition of feeding boxes. but there \\'ere diffettnces for e<ich individual \Vhen housed solitarity and paired.although the female demonstrated a significan1 reduction in pacing under both conditions. the male only demonst ra1ed a significan1 red uc- 1ion in pacing when housed with rhe female. Both tigers displayed <1n increase in species-1ypk aj behaviours during 1he experimental conditions.. \Vith th e female exhibiting an increase in sleeping. and the ma.le sho\ving an increase in <1ctive behaviours at the feeding b::>xes. While promising. being the only study of its kind whh cap1iw- dgers, 11ddi- 1iona1 research is needed to substantiate these findings Na'Vtl roysjab}ec.rs as tnrichrneru i o compensate for the lack of li \."e prey in captivity <1nd 1he sobsequen1 lack of hunting-related energy expendi- 1ure. zoos have been creative in developing l\ov\"i 'play devices to encourage these natural behaviours in their 1igers. 'l'hese include an ifidal prey in the form of plastic rabbit replicas and bird sounds: cardboard box animali: burlap sacks hung from trees 1hat move \Vhen auacked : ropes for tugging (a 'tiger tug'): barrels cor11aining s1ones for sound: watermelons that roll av. ay \Vhen chased; met<1i kegs and frozen blood balls (Hare and jarrand. 1998: MarJ.:owitz and Laf orse. 1987: f oulsen and f\>1il!er, 1996: Koynon. 2000: Star k. 2005: Van ~1 ener et al_ 2008). These papers - including <1necdotaJ accoun1s. conference proceeding.s, and joum a1 <1rtk les - <111 indicate the success of such items in increasing sped es specific behaviours a.nd d~crtas:i ng pacing. for &ample. Hare and Jarrand (1998) a.nec:dotally discuss hol.v burlap sacks encouraged natural hunting behaviours. including stalking. approaching. a.nd biting. 4. Olfactory stlm u.lation as enrichment Wild 1igers devcte signific<1nt time to patrolling 1hd r territories.. and therefore some focus has a.lso been placed on s1imulating natural territorial behaviours th rough olfactory (scent) enrichments in c<1ptivit>" Many felid species. including tigers.communicate \vi th ether animals through sprayingands«n1 rnarlcingand, in the \vild, they will spend much lime patrolling <1nd marking their territories (CJark a.nd King. 2008: Ki.tchener, 2000). In many c<1pfive environments \Vhere 1igers constantly occupy the same space. there is a limited need for scent-marking and 1erritoriaJ behaviours.. since rhere is little olfactory change in che exhibits. 7.oos have addressed this through the provision of various scent enr ichments Addfrion oj seen rs and objeas ro riger enc.losures '!'he addition of scents. such as spices, faeces fro m prey animals.. and woodchipsfsubscra1es from other enclosures. can encourage these natural communic<1tory and 1erritoriaJ behaviours (such <IS patrolling and spraying) and increase stimulation (f\>1ellen <1nd Shepherd.son. 199?: Skibiel et al ). Skibiel et a l (2007) compared the presentation of spices (cinnamon. chilli po...'der. and cum.in) on the behaviour of six fetid species (including three tigers). ror all animals combined. it \vas found tha1 the spi«s 299

318 s significandy ina e.ased.activity level from baseline and c.aused <t signific.ant decre.ase in proportion of sc.ans spent p.acing. Not.ably, tigers exhibited the highest incre.ase in active behaviours ( :1' 8.03%) and <t decrease in pacing of (:l2.3.t,9%) with the implemen1a1ion of the spices. Similarly, van ~1euer et.al (2008) foond that tigers c.an be enriched by zebra dung and scented squ.ash, both of which incre.ased behavioural diversil>' f:leing the only l\vo studies to explore scen1 enrichment, such positive results need to be sopponed by addilionaj research ndo.rure rotations A more comp Jex form of olf.actory enrichmen1 has been demonstra1ed by zoos tha1 rotate 1hdr animajs betv.<een exhibits. Enclosure rotation is a process where severaj ani mals may occupy <t number of exhibits simullaneousjy or serially, and are moved from one to 1he next \'lithin <t sin gle day or on differen1 days. Kot.ating animals in such <t wa>' is thought to more.accurately resemble the natura.i environment.. \Vhere ether species a re presen1 and ani mals haw interconnected home ranges and path\vays(coe, 200'1). This can allow for various environmen1aj stimuli to be encountered, such <is ph)'sk<tl variation and the scents of other animals (White et <tl 2003). While studies exploring the effects of this method have tended to focus on primates (Coe. 200t,), one study com pared the behaviour of <t tigerrota1ed through four exhibits (simultaneously housing orangutans, f\~ajaya n tapirs. sia mangs and babirusa) to a tiger housed in <t solitary exhibit (White et al., 2003) f indings revealed no signifk<in1differ ences in most behaviour c<itegories. ind uding stereotypic pacing, but 1he l\vo did differ in time spen1 resting and engaged in 'othef behaviours (including species-t> pical behaviours, such <is spraying and grooming). '!'he rotated tiger res1ed significantly less (37.31% of scans versus 52.20%) and eng<iged in 'ether" behaviours signific<intly more (S.67% versus t.69 ) than the trad[tionally housed tiger, indicating the peten1ial for encjosurt rotation to increase.active. species-typk aj behaviours. 5. Enclosures as enr ichment Aspeasof an enclosure, such as its size and complexity, can <tlso inftuence the occurrence of s1errotypk behaviours (Lyons et al., 1991) and therefore could be altered as <t form of enrichment. WhBst differing depending on the region, \Vi!d tigers occupy large, individual home ranges of 15-10,00 kml (Karanth and Nichols, 2010). However, home ranges also vary in size depending on prey abun dance (Sunquist <ind Sunquist, 2002). 'l'igers are found to be dispersed over long distances (up to hund.reds of ldlo meters) <ind can navel benveen Sand 30 kmjday(k.aran1h and Nk hols.20 10).lncaptivit>'. however, thisrange isdras rically reduced, especial!)' in urban zoos (<ts opposed to open-range settings) Restricted space reduces the.ability for these animals to perform na1ural territorial behaviours to 1heex1ent 1ha1 they would in the wild and the proponion of rime devoted to 1his (assuming 1hat <t smaller 'territory' in capti\'ity requires less time to patrol and scent mark). Although limited. some research has explored 1he impact of enclosure size on the beha\'iour of c.aptive tigers l:'hd osure.size Lyons et al. (1991). \Vhite ct <ti. (2003). and Bashaw et <tl (2007) have all sho\vn that tigers housed in larger enclosures.are more active. In a study of nine felid species. including three tigers. Lyons ct <ti. (199?) found that cats housed in larger enclosures shov.<ed higher average move men1 than those in smaller ones. HO\vever. 1he.authors note that signific<int differences in pacing v.<ere not found, despite.animals in smaller encjosures pacing more than those in larger ones. Similar 10 Lyons et aj. (1991). Whi1e et al (2003) found that \'/hen housed in a smajl unna1- uralisric exhibit <t male Sumattan tiger made very Uttle use of the enclosure space <ind spent a large proportion of time resting. compared to its behaviour in a larger, na1- uralisric exhibit. f\tore recentl)'. aasha\v et al. (2007) found an.average incre<ise in activity of around 6% (from 5% to 11%) and <t decre<ise in s1ereot>'pk pacing of around 15% (from '12% 10 2/%) of sc<ins in tv.-o tigers when housed in a large exhibit (9Si ml) compared to a smaller one (1 19 m2). However, these differences \vere non-signific<int and it ' ' 'il.s net det<iiled how freq.uently each.animal was moved bet\\'een 1hese enclosures, making it difficult to draw reliable conclusions benveen exhibit size and behaviour. further research in this are.a is required. 6. Sodal stim ulation as enr ichment Civen that feeding <ind territorial behaviours are dom inant in wild tigers, it is not surprising 1ha1 much of the litera1ure has focused on enrichment efforts 10 encour<ige these. Of lesser recognition is social enrichmen1 for this species.. S\\ aisgood and Shepherdson (2005) highlight that soli1ary.animals typically fare worse in captivity than more social species due to Jack of.access 10, or Jack of bene fit from interacting with, conspedfics.. Appropria1e social stimulation, beth within and between species, c<in be con side red enriching \'ia incre.ased arousal and the opponunity for interactions (Shepherdson, 2003). but 1his has no1 been explored in sufficient depth for tigers.. Wild tigers are primaril)' sol[tary in nature, usuaj)y only interacting for mating purposes and \'/hen females re<ir cubs (Kitchener. 2000). 1-'urther,.adult tigers main1ain exclusive home ranges, with female ranges ta.rely over lapping and male ranges usually overlapping one to three females (Sonquis1 <ind Sunquist, 2002). He Never, despile this solitary nature, tigers.are not <1n1isodal and there have been some repons of.additional conspecific interaa ions in \Vi!d tigers, such as group hunting (Kitchener. 2000:Tilson et al ). Shoemaker ct <tl(l99?) indicate tha1 tigers do no1 necessarily have to be housed sing})'. and in some cap tive settings they are housed in pairs. trios and/or groups (Law et <ti., 1991: Mellen and Shepherdson, 1997: ~t iller and Kuhar, 2007). While the use of social housing is prob ably aimed at saving space and the expense of building separate enclosures (Law et al., 1991). it is not cle<ir <it presen1 \Vhether or not this could be enriching. V<trious tiger enrichment rela1ed studies indicate the occurrence 300

319 of social housing when provid!n& dt'lails ol study animals (ror example. see Lyons ct.i.i ) bu1, lnteresringly, few studies haw focustd on this capn.. e aspen as a form of enrichmenl Social housing 1\1.'0 studjei (De Rouc:k ~ : ri.~ine r and Kuh.ar. 2007) explored d1ffe1cnt socw l'm>uslng combinations of tigers in various Europc.&n and Amfric:an instirctions. rntaling miud find.mp. Dt Rouck tt: al (2005) found that. NWYl(U,,.-I m c,illf,11roy MlrU'ft fivt' iuir-hntkl'd riv«pnbnitd a WI.icier v.uitty ol~ts.. ~ch wi.~ primuily interaaiw m rwtutt. In corit:r~ by exploring 1.bt imp:tet of socw housing on FOUP tl SD: fem.alt tigrs. Jo.tiller and Kuhar (2007) found ttut owr a six.-)oeat pl"riod non<ontaa aggmsion and voc.alis.ulons increased. while spraying and soda.i proximity dccrta$t'd. 1-'unher. spra) ing correlated positivtly with con111e1 a,a.rwion. while social proximity correlated Mgatlvcly with non-contact aggression. 1'hes.e results indk ace lht' pot enrlal Implications thal can atis.e from social houslftji. 'I hb Is lmponant sin«. to date. there is litdr information :.bout the fe.uibilicyof largc social groups of ti;grrs and nont on s:i.me-sexgrou ping. fw ther. 8.!shaw e1 al. (2007).tind ~ kolxk e1 al (2005) have both found th.at agas duplay mott p.tdng ~en houstd ad_jacttt 10 Olhtt t1gtrs. Tl'lis lnlonn.xion. combined with lht mixed findings of soo.al ~ lnd.ic.tlts that if ligrn arr to c:oodnue to bt' hourd tn socuj conditions lhttt is a dear nttd for.addn:~ rcswd'i to substantiale such practices. k.c:sl'meh with m\l.llff!clids has ~\oeall'd oogariw imparu of sodl1 hou'lna. ~has reducl'd repro c!uctive success i.npair-houj.td.1.n!mals and higher le\ els of pacing (Law et al.. 199? ; MC':llen.uwi Shepherdson. 1997). Given the dracasing number or tigers remaining in the wild. ii is imponant to address 001 only the enrichmen1 value of social housing. but.tljo 1he lmp;ic1 that this mighl have on 1heit ttprodocuvt ~h.ivlours. Funher. futurr resejjdl wlu nttd to consider the age or tigers in social conditions. Accordlni 10 the.soc:ui behaviour of 'o\ijd ligen.. it would be txp«ttd that ~unger tigers would Cispl.ay men posa1rw bcnrfiu of social housing comp.ated 10 older/adull 1i:pn (shier abs mnait1 in thrir liners with lhl'ir mothrn.. whm-.n.u!ulu tiw indtpen Undy). and dlls mlgb1 s:trw metes. For example. m unpublished honours 1hcs:is (~ tht pnnury author} found that socially boustd cubs (6 moru hs of age) and sub-adults (1 year of age) d1.spl a)~ hi;htr propon lons of interac fi ve behaviours wnh each Olhl'f (panicular1y play-related bthaviours)comparcd to.ldult t1grrs(averag:e 16%of scans versus 1sor scaru.. respectivtly) Hum.an Jnteraalon cu t11rl<h me111 ltg:ers are typkally J)l'"l &:tlvc"d u d.mgerous. with a reputation.!s man-eattts. np«wly In tndl.i (Azad et.!l : Nyhus et al~ 2003: Tiitson tt..._ 1997). Thus. possibly Cue to thti.r sitt. strtfl#h. mt poem W dj.!'lg'tts of haying physic.aj conua W1lb Uw.m. nd 1ht resoun:u involvtd. t.iman intttaenon tw.s rattly. until more recendy. bttn. prac:tired... ith upm.t ugm. Whilt hunwi.s ana rigrs h.w bttn lmtrat'ting outside oi official zooloijui lnsritutionj throol}lout his.lory (e.g.. in d 1cusl':s), it ts h.a1 only been In rt'ctn1 decades that these intcractlon1 In zoos h vt lncrc.i1td. t'or the purpose of this djscusslon. hun,.in inter.-ctton will refer to th.'tt bet... 1ecn keepers.tnd ti a er" 1 Allhough keepers are a cons1ant ractor in captive tfget's cnvi ronn,tnt. human interaction h.1$ no4: lrcn sden1ifk.illy rxplortd as.tn enrichment trchnique.d1?spilr it h.ivtng tm potentl I to be tnrichlng. both in the form J(tr.iinlns.ind as a l'lusbiandry Stylr. Beyond the ""lf re of ind.jvidu.ll users. ii ts.also rttocj'litt'd that f.mors in 1he capm tn'v1ronmrn1 ~ fw".vriw>ty Ah# M"tiftul ~Ir,Hli f'l"'l.vtlty.. -. n thus it l:ic'vt't'i more pminim.1 dw.t lhrmp.kt oltlul't'wi in.c':t'.tittaon on 1i:gen is o:plortd in order o nwnt.mn NCU ral bd\lvdurs.ind minimise gencci&: cb.arge 6.2. J. tti1tf~ mnforceme1u mrlni.ng 1'r.a.lnlns, defined here as any plann!d.tind t.irtct e-d procedure whereby a zoo keepe.r cncrur~c':s the performance or specific behaviours in tiger>. h.u long been performed wllh many species. particuh rty non hun,.in prim.utt;. 6.lsed on principles of opcran1 m nd 1lonln;. It u1u.illy t.lkcs the form of positive reinforcmten1 c r;i ln~ ing. whr'f'c'.tnl.ma~ are rewarded (rather th.a.n punishtd) for perfom'l.lng a dt:s:i!ed behavioural t'l'spon:jf (Lluk.ind Oamond. 1998). This process thttt:fort IC':lJH on thr vol Wll.lfY cooptruoft of dlt subject. and pr<vidts uw.anlnw w.u..ap.ltttdtpffofconudo tttbteimt (~t.tnd Budw\ln~Smit.h. 2007). Tigers haw panop.m.td In min ire in drcu1e1.ind other unofficial insritmons tmnshoui hu.tory. \'l/hds1 il does not appear tn:t any rd~titk me.arth his Mq)lored this. ft would seemth.lt the pl"lmuy pu IJI05t' or 1 his practice in such environm:-:ncs h.as bttn ror the encrrt.ilnmitn1 of the public. krgardlcss of the motl\ a tion. this does h(ghlight the potl'ntial tr.i n.lbility or tl,c'sc anlm.ils. It U only more recently that zoosjnd othrr ofrd11l wildlife lnilltutlons have adopted such p1o1etkes. Al prrstnt. any Information about training pr.lctlces with tl.&c't's b putt!)'.meodotal. Sf?atch~s of %00 WC"bslltS revt.al tl\.lt v.11rlous Institutions worldwik.tre prxtldng u;uruns 'o\'ilh t~r animals. funhtt. bocb. Mtdi.n.i (2005} nd Thom.I. (2004) discuss tbt use ol pu>lic: tnlr:urta ~ ~.11 tl'lr arcnx too. where visitors._~.tbijt 10 w.:uch kttptn u.dnins tigers lhrough mtsb fl.'flclnl. Countlru video foot1sr of such praaias is.tso frttly.iv.a.11lblt on wd>s1tn such ~ YOU1"ube. Addition.ally tw0 Au.flri!W\ in~t1 1udons Austra!U Zoo and Ott.a.mNOrid-lu.nd"fW tht'ir tiscrs.11nd practice hands-on contact. both on ;tnd off-c':xh\l)!l, wl1h k~pers spen dingthc most pan oft hr day in phy,le.ll con1.tet with theanimals(pcrsonal observatk>n: for add! 1 IOMI decal Is orkeeper-dgrr intrr;_c1ioru '11 Orellm worki.acc 8road and Weiler.1998: l'hillij»and P~ck. 2007). Despite t he! 1.tckof scientific support for such pr.ti:1jcts with tigers. common on all of me \vebsitrs for the.sc": zoos: 11 thtrffon cohl1t'lllght 1hat such practicei<lre occvrrinaout or a 1 Ill,_ ,,..~-~...,,..._.. ~... --~""""...,,.."" ~ b. ~If.. MffOr',...,...,...., f'llftl...,. nww..oe..-.~ bt...,...i:.y... tw~., l"-t~ ""'-~«iorl... ~ 301

320 need to enrich and improve 1he capti\'e exis1ence of 1hese animals. the li1era 1ure hea..,ily supports such practices \\'i 1h non-human prima1es(b.asset1 et al : 81oomsmith and Else. 2005: Bloom.smith et at. 1998: Calahan and 8reder. 2003: l..lule et al : Reinhardt, 2003: Savastano et aj : Schapiro et al ): but. aside from one \'erybrief anecdotal mention of a training program for a tiger a1 1>ort I.and Zoo (f\tellen <ind Shepherdson. 1991), whether the same benefits apply to this species remain unlcno\\'n. It therefore appears timely to explore the impact of these pracuces on ttgers ana to 1nwsugate the claims tnat uiey are benefid aj Husbandry styte In addition 10 1he role of ttaining <is enrichment. e ther keeper-tiger in1eractionsshouldbeconsidered. f\tellen and Shepherd.son ( 1991) highligh1 that keepers play a signifi cant role in the welfare of captive felids. aenefits include increased reproductive success. reduced pacing. and stim uja1ed cogniti\'e abili1ies as a result of increased keeper interaction (IV1el!en a nd Shepherdson. 1997: S\vaisgood and Shepherdson. 2005} indicating 1hat it may be enrich ing. Such benefits ha\'e not been specifically explored '"i1h captive tigers. although 1here are aspens of keeper-tiger interactions discussed in the anecdotal literature. Oocumented in1eractions ha\'e occurred at the Calgary Zoo. where a 'tiger tug' \vas used as a play device betv.-een tigers and keepers through protective fend ng(t:.oulsen and IV1il!er. 1996). The authors expjain tha1 this was a souru of increased interest for 1he tigers. with individuals that did net previously show interest in J.:eepers actively soliciting pjay '"hen J.:eepers v.-ere present. Despite being anecdotaj. this evidence of tigers voluntarily initiating contact high lights a positive benefit of tiger-keeper interactions. \Vi1h an increase in the le\'els of interaction occurring betv.-een tigers and keepers. for ex.ample a1 Austt alia Zoo and Ore am v.'ortd. it <ippears important to explore 1he enrichment \'alue of these practices. ' successful a particular enrichmen1 technique is \'iev.-ed as and minor changes in pacing may be over!ocj.:ed. despile the pcten1ial welfare gain for an individual animal One po1entiaj me1hod to <illo''' for a more complete measurement of enrichment techniques could be to couple behavioural data \vith a biochemical marker of stress. such as a cortisol <inalysis (for an example of a combined behavioural and cortisol analysis of cattle. see 8ris1ow and Holmes. 2001). In addition to adding complemen1ary data to 1>ehav1oura1 studies 01 a ger enncnment. there al.so needs to be more consis1ency in the behaviouraj designs emplo}'ed. t he majority of 1he aforementioned studies involved the instantaneous sc<in sampling technique ( f\~artin a nd aa1eson. 2007). whereby the behaviour of the animals \vas recorded <tt particular intervals. HO\\'ever. the length of in1ervals and amount of da1a collected is inconsisten1 across studies. also differing in the timing and duration of any post-enrichment data. WhiJe it is net al\vays possible to be consisten1 due 10 other fa ctors. such as sample and enclosure size.. some general guidelines for future research could allow results to be more compar<ib!e. for example. with regard to time intervals. it is suggested tha1 30 s to l min intervals \vould be appropria1e where one <inimal is being observed. 1 min for 2 -'1 <inimals are being observed at the same time. and 2-3 min for more than 5 animals (based on 1he authors - M.S. and C.L - experience with tiger observations} finally. sintt there is no uniform e1hogram for documenting tiger beha..,iour. studies vary in 1he behaviour categories they use.. making comparisons difficult. In order 10 be able to accuratety compare data across <inimals and institutions. for different enrichment techniques. there is a sirong need for a more defined and consistent <ipproach to data collection. f Ablt l riw:1 b rt.;:wiwr t.:ilt1tu1 i\'s, 7. ~!ethod ologlcal considerations ror future research Not only is it impon an1 to consider 1he findings of tiger enrichment studies. but <ilso it is vital to recognise the mediodolcgies used. All of the aforementioned studies have employed behaviour al observations of 1he ani mals in order to measure the impact of the enrichments. This is fitting. sin«enrichment seeks to increase animal activity, increase active behaviours. and redu«stereo typic beha..,iour (Sv.:aisgood and Shepherdson. 2005). It al.so seeks to improve animal PS>'Chologic<il v.<ell being (Swaisgood and She pherdson. 2005). and Laule and Oesmond ( t 998) suggest that psycholcgk aj \vell being should be measured by a1 least rwoofthe fo!io\\'ingcriteria: behaviour. health, reproduction. and longevity. 'l'herefore. it is apparent that future research mus1 taj.:e into account more than jus1 the behaviour of the <inimals. ror example. many of the studies discussed here employed pacing as a measure of enrichment success: ho\\'ever. indi\'idual animals. both \\'ithin and betv.-een zoos. varied in 1heir levels of this behaviour. 'l'his variation will infl uence hol.v Out olsiil!t \\'.:..le. rnn. Jur..Jl. ~.mb. fw!m. p.:.uol :crr!!of)' (imluc!lll8 Jn!lf o')jttis. tlellmcn. ipn yj. grooo Sl'lf. YOC;i!bt...lk'rr. n tt1m. il!.upen d.:iwi. urln;i1c(cltttui:e SU;. s:;ind or 51.y down (W!l?I or witllouc eyt'~opl'!i'i) Any OOl'ISunUJtion btli.;1vi(l11 (f(l1 f?l.antj4t; d1.;1s. pluc-l.1kk. bilo:. chc-...,,...,.,, lu'ty bth.a'\ AA11 dlirttt,,, " (':'lfi(h tn('!tl ~ (fo< f?l<m'!plt: snillpull sw.:il l\lt, rolon[oj'i. d14 <l:o:w) lu'ty ll('h.-.iou1 din:c-tt<l.:it,am)l h('r lilh' or ~':lm;ii'i (for cx.:.ml)k: s:;i!k, l)lly, Sl': w.lh.t:room.~.u!: 00'".1kl be subdivided!r.i:o #llll.il~( po~!l~).ln:l.lst:t\'$$/lot' l'!1t3'll;\\') ~i'llt'r..ci:loi'1$ Any.ipp.:irm:ly fu Mtlon!cu, ro:pela1~ ~vi-. t nyo!ylll8.ll SNn two ttl)l'd U01'1s of me btll.f\'lo'-.1r (i octl.'5 ~ns or cxtcuive 31000!1"15) lu'ty bth.a-vi<lu1 11:.~l dr:lo 11)( lit into onr ot the 4bo \' <''1\'t:(l!'WS lu'ty i115t,am,''t' wh('li'l hi' btt.;:..,ioor (If.a lijtc1 is n(lt vh itik to obsi:l'vft 302

321 303

322 304

323 Appendix C Key methodological details of some past zoo animal studies (impact of visitors and keepers) To complement the information in Chapter Three, this section contains key methodological details of 32 studies investigating the influence of humans (zoo visitors and/or zoo keepers) published between 1984 and

324 Table C. Summary of key methodological details from peer-reviewed studies, published between , of the influence of people (visitors and/or keepers) on zoo animals (excluding studies of felids) Study focus (human element) Visitor density Few (<5 visitors per cluster) versus large (>8) Visitor density Density levels (not specified) Visitor density None, low (1-999 visitors), medium ( ), high (>7000) Visitor density High versus low (not specified) Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Orangutans (n = 12)/Chester Zoo (UK) Mandrills (n = 3)/Zoological Garden of Vienna Schoenbrunn (Vienna) Colombian spider monkeys (n = 6)/Chester Zoo (UK) Golden-bellied mangabeys (n = 10)/Sacramento Zoo (USA) Within-groups, repeated conditions Within-groups, repeated conditions Within-subjects, repeated conditions Within-subjects, repeated conditions Behaviour: Categories: adult sack use, infant holding, approach, sitting, foraging, object manipulation) Behaviour: Categories: behaviour directed towards visitors, abnormal behaviour, general behaviour, and social behaviour rules*) Direct behaviour obs^: - Scan sampling using instantaneous sampling at 30s intervals (60min period for solitary, 90min for group housed animals) Direct behaviour obs: - Continuous recording of each animal for 10 hrs/day Hormone (cortisol) levels Urinary cortisol sampling: - validated by enzymeimmunoassay - collected during zoo closure (no visitor condition) and then the morning following the particular day of zoo visitation Behaviour: - Aggressive behaviour: facial displays indicating threat (openmouth or brow threat) Direct behaviour obs: - Ad libitum, continuous recording of aggressive behaviour displays throughout 15min observation sessions at each cage (n = 3) (time/days of obs unspecified) Study duration/ data collection 6mth study period: - various times of day and days of week for different animals - 55 hours obs total. Study period = NS: - 20x days with visitors present, 20x days with visitors absent (unspecified period) Study period = NS - approx. 25 samples collected per animal - always collected between 0700 and 0800hr, 3-4x per wk Study period = NS: - total 453 sessions/113.25h of observation Study details Birke (2002)**^^ Chamove et al. (1988)** Davis et al. (2005) Mitchell et al. (1992b) 306

325 Table C. Continued Study focus (human element) Visitor density Busy (weekend/public holidays) versus quiet (weekdays) Visitor density Continuous Visitor density None, small (1-5 visitors), large (>5 visitors) Visitor density Low v high (based on zoo attendance numbers) Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Siamang (n = 6) and Corella (n = 1)/Adelaide Zoo (AUS) Male pileated gibbon (n = 1)/Blackpool Zoo (UK) Diana monkeys (n = 6)/Edinburgh Zoo (UK) Gorilla (n = 6)/ Belfast Zoological Gardens (Ireland) Within subjects, repeated conditions Within groups, repeated condition Within-groups, repeated conditions Within-groups, repeated conditions Behaviour: - categories not defined - location in exhibit Behaviour: - Self-injurious behaviour (selfbiting) and abnormal behaviour (chewing orally) Behaviour: - categories: observing, feeding and chewing, playing, grooming, resting and sleeping Behaviour: - State categories: stand, sit, rest, autogroom, aggression, move, socialize, bang barrier, abnormal - position (in front of visitors' viewing area) rules*) Direct and video recorded behaviour obs: - 20min obs sessions at each enclosure (n = 2) using two methods: 1) Scan sampling using instantaneous sampling at 30s intervals during noninteractive periods; 2) Alloccurrence sampling during interactive periods - 20min sessions on four days per visitor condition Direct behaviour obs: - Scan sampling using instantaneous sampling at 30s intervals during 2hr blocks at different times of the day ( h, h, h) Direct behaviour obs: - Scan sampling using instantaneous sampling at 5 minute intervals during two time sessions ( h; h) per day Direct behaviour obs: - Scan sampling using instantaneous sampling at 5min intervals Study duration/ data collection Study period = NS: - 4x 20min sessions at each enclosure ( h, h, h, h) 5mth study period: - 18 non-consecutive days of data collection 10 day study period: - data collected daily 5mth study period: - 4hrs per day ( ) for 20 days of high density and 20 days of low - total 80hrs/animal/condition Study details Nimon & Dalziel (1992) Skyner et al. (2004) Todd et al. (2007) Wells (2005) 307

326 Table C. Continued Study focus (human element) Visitor intensity % of enclosure perimeter exposed to public Visitor intensity No visitors, passive visitors (no visitor tries to interact with animals), active visitors (at least one visitor tries to interact with animals) Visitors - intensity None, small (1-5 visitors) active ( 1 visitor attempting to interact with animal/s) groups, large ( 6 visitors) active groups, small passive (no visitors attempting interaction) groups, large passive groups Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording rules*) Black rhinos (n = NS; 15 enclosures)/8 zoos (NS) Lemurs (n = 3)/Chester Zoo (UK) Ring-tailed lemur (n = 3), Mayotte lemur (n = 3), black spider monkey (n = 6), white-fronted capuchin (n = 4), patas monkey (n = 5), de Brazza monkey (n = 3), Sykes monkey (n = 5), talapoin (n = 6), Barbary macaque (n = 7), lion-tailed macaque (n = 4), celebes black macaque (n = 2), hamadryas baboon (n = 8)/Chester Zoo (UK) Correlational Corticoid metabolite values Fecal cortisol sampling Within-groups, repeated conditions Within-groups, repeated conditions Behaviour: - Categories: interactions with autdience, interactions with cage mates, locomotory activity Behaviour: - interactions with audience - interactions with cage-mates - locomotory activity - spatial dispersion - collected in morning from enclosures/overnight housing Behaviour obs (method = either direct or video recorded; NS): - 1min samples (method published elsewhere***) Direct behaviour obs: - 1min samples - all occurances of interactions (audience and cage-mates) - scan sampling of locomotory activity and spatial dispersion using instantaneous sampling at 10s intervals Study duration/ data collection 1yr study period: - wkly sampling - total number of samples = NS 6mth study period: - 76 samples in total 8mth study period: - times/days of obs not specified x1min obs audience directed behaviour, 539 x1min obs cage-mate directed behaviour, 459 x1min obs locomotory activity, 465 x1min obs spatial dispertion (but not specified how many obs per exhibit/species, or per condition) Study details Carlstead & Brown (2005)**^^ Hosey (1989)**^^ Hosey & Druck (1987) 308

327 Table C. Continued Study focus (human element) Visitors - intensity None, small (<5) active groups, large ( 6) active groups, small passive groups, large passive groups Visitor intensity Enclosures (n = 3) of different styles, with increasing visitor access around perimeters Visitor noise Silent v. noisy - (talking/singing loudly) (manipulated by researcher) Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Ring-tailed lemur (n= 4), mongoose lemur (n= 2), redruffed lemur (n= 5), squirrel monkey (n= 2), francois langur (n= 2), spot-nosed monkey (n= 2), De Brazza's monkey (n= 2), goldenbellied mangabey (n= 8), gibbon (n= 3), orangutan (n= 5), common chimpanzee (n= 4)/Sacramento Zoo (USA) Soemmerring's gazelle (n = 17)/Giza Zoo (Egypt) Orangutans (n = 10)/Chester Zoo (UK) Within-groups, repeated conditions Behaviour: - interactions with audience - interactions with cage-mates - locomotory activity - spatial dispersion in cage Between-groups Behaviour: - Reactivity (categorized as: attention to observer/visitors/sounds/noise by freezing, moving heads or body, or running; vocalization; agnostic behaviour (fight, flight, threat, submission)) Within-groups Repeated conditions Behaviour: Time budget (categories: adult sack use, looking, holding, approaching, sitting, object manipulation) rules*) Direct behaviour obs: - Scan sampling using one-zero sampling of group at approx. 15s intervals (session time length unclear) Video recorded behaviour obs: - Two methods during 60min observation periods of each exhibit (n = 3) at various time points ( , , ): 1) focal sampling, continuous recording of males; 2) ad libitum, continuous recording of females Direct behaviour obs: - Scan sampling using instantaneous scan sampling at 30s intervals for 3mins prior to visitor arrival and 3mins during visitor presence - Rpeated after 20 minutes but with opposite group type Study duration/ data collection Study period = NS: - total 2033 scans 5mth study period: - total observation hrs unspecified 6wk study period: - Repeated for 10 nonconsecutive days Study details Mitchel et al. (1992c) Mansour et al. (2000) Birke (2002)** 309

328 Table C. Continued Study focus (human element) Visitor presence/absence Presence = 6 visitors Visitor presence/absence On- versus off-exhibit housing (Part 1-1 group on- exhibit and 3 groups off; Part 2-2 groups switched between on- and offexhibit, and 2 groups off only) Visitor density Impact of number of visitors at exhibit Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Cotton-top tamarins (n = 3), Diana monkeys (n = 5), and ring-tailed lemurs (n = 4)/Edinburgh Zoo (UK) Four groups of cotton-topped tamarin (n = # )/zoo = NS Four groups of cotton-topped tamarin (n = 3)/zoo = NS Within-groups, repeated conditions Between-and within-groups Behaviour: Categories: agnostic, grooming, affiliation, inactivity. Behaviour: - Categories: amicable (nestbox share, contact, proximity, allogroom, play, food-beg, mount), agnostic (attack, fight, chase, threat, avoid) Correlational Behaviour: - Categories: amicable (nestbox share, contact, proximity, allogroom, play, food-beg, mount), agnostic (attack, fight, chase, threat, avoid) rules*) Direct behaviour obs: - Scan sampling using instantaneous sampling at 10s intervals during 10min periods at one of three times of day (1000, 1200, or 1400hr) Direct behaviour obs: 2-part study (but same for both) - 1 hr sessions (during one of the following periods per day: hr, hr, hr); - sampling and recording = NS (but assume recorded behaviour duration and frequency) - 1 hr sessions (during one of the following periods per day: hr, hr, hr); - sampling and recording = NS (but assume recorded behaviour duration and frequency) Study duration/ data collection 2mth study period: - Each animal observed for total of 40mins during two separate months Part 1 6mth study period: - each group observed 24 times (3 or 4 obs days per week) Part 2-6mth study period - experimental groups (n = 2) observed 36 times each; control groups (n = 2) observed 24 times - 6mth study period: - 24 obs (3 or 4 obs days per week) Study details Chamove et al. (1988)** Glatston et al. (1984)** Glatston et al. (1984)** 310

329 Table C. Continued Study focus (human element) Visitor presence/absence As implied Visitor presence/absence Presence (zoo open to public), absence (zoo closed to public) Visitor presence/absence Presence (on-exhibit), absence (off-exhibit) (long term effects) Visitor presence/absence As implied Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording White-cheeked gibbons (n = 2)/Duisburg Zoo (Germany) Lion-tailed macaques (n = 30)/8 zoos (India) Lion-tailed macaques (n =7)/ Thiruvananthapuram Zoo (India) Pied tamarins (n = 11), black lion tamarins (n = 7)/Jersey Wildlife Preservation Trust (Jersey) Within-groups, repeated conditions Within-groups, repeated conditions Within-groups, repeated conditions Within- and between-groups (species), repeated conditions Behaviour: - Social proximity (categories: contact, proximate, and out of reach) - Behaviour categories: feed, locomote, self groom, social groom, visitor directed behaviour, rest, other, not visible Behaviour: - States (major categories: abnormal and normal) and events (major categories: abnormal and normal)* see study for full list - Enclosure use (four locations) Behaviour: - States (major categories: abnormal and normal) and events (major categories: abnormal and normal)* see study for full list - Enclosure use (four locations) Behaviour: - Categories: approach cage front, threat, interactions with visitors, piloerection rules*) Direct behaviour obs: - Scan sampling using instantaneous sampling at 30s intervals for 3x 10 min sessions (1100, 1300, 1500h) Direct behaviour obs: - Two sampling methods during sampling period (at least 1hr sessions - maximum unspecified): 1) 1x instantaneous scan (behaviour and location); 2) focal sampling with continuous recording for 15 minutes (behaviour only) - all days of week between h Direct behaviour obs: - Continuous focal animal recording for 15 minutes during each sampling period (one or multiples of 1hr) - States recorded as % of time and events as frequencies/hr Direct behaviour obs: - Continuous sampling of groups (n = 4) during 2hr observation periods (one period per group) Study duration/ data collection - 13 visitor present and 15 visitor absent days over approx. 6mths - 6mth study period: - 2x days of data per animal at each zoo) - Approx. 5.6hr visitor presence and 1.6hr visitor absence per animal in total 10mth study period: - observed on-exhibit, then observed 3mths later offexhibit - Approx. 5.1hr on-exhibit and 5.4hr off-exhibit per animal Study period = NS: - total observation time approx. 9hrs with and 7hrs without visitors Study details Lukas et al. (2002) Mallapur et al. (2005)** Mallapur et al. (2005)** Wormell et al. (1996) 311

330 Table C. Continued Study focus (human element) Visitors - multiple factors Combined impact of retreat space (no retreat, semi-retreat, or full retreat), visitor density (low = 1-6 visitors, medium = 7-12, and high = 13-18), number of visitor touches per hour, and species Visitors - multiple factors Interaction of mean number of visitors and presence/absence of feeding enrichment Visitor presence/absence Presence = 6 visitors Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording African pygmy goat (n = 5) and Romanov sheep (n =2)/ Zoo Atlanta (USA) Western lowland gorilla (n = 20)/ Port Lympne and Chessignton zoos (UK) Cotton-top tamarins (n = 3), Diana monkeys (n = 5), and ring-tailed lemurs (n = 4)/Edinburgh Zoo (UK) Predictive correlational Within-groups, repeated conditions Within-groups, repeated conditions Behaviour: - Undesirable behaviour: head tossing, head butting, foot stamping, rearing, nose-blowing, and leaving Behaviour: - Anxiety behaviours (selfscratching and visitor watching) Behaviour: Categories: agnostic, grooming, affiliation, inactivity. rules*) Direct behaviour obs: - Focal sampling using one-zero recording of each animal for 8min period - data collected once/day/animal during first hour yard opened (alternating daily between 1000 and 1100) - Ad libitum recording of number of touches per animal/session Direct behaviour obs: - Continuous focal animal recording during 15min periods (times unspecified) Direct behaviour obs: - Scan sampling using instantaneous sampling at 10s intervals during 10min periods at one of three times of day (1000, 1200, or 1400hr) Study duration/ data collection Study period = NS: - Approx. 27hr behavioural data total (14.67hr no-retreat, 9.52hr semi-retreat; 3.60hr full-retreat) Approx. 2mth study period: - total 151hrs obs data 2mth study period: - Each animal observed for total of 40mins during two separate months Study details Anderson et al. (2002) Carder & Semple (2008) Chamove et al. (1988)** 312

331 Table C. Continued Study focus (human element) Visitors - multiple factors Absence (no visitors), active presence ( 6 visitors, 1 trying to gain animals' attention/interact), passive presence ( 6 visitors, 0 trying to gain attention/interact); Differences between species, as well as for the following variables: body weight, aboreality, group size, and mean length of residence at zoo Visitors - multiple factors Interaction of visitor number (11-20, 21-30, 31-40, >40 visitors), activity (standing, looking, taking photos, holding food) and proximity to animals (>/< 10m) Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Talapoin monkeys (n = 6), brown lemurs (n = 3), ringtailed lemurs (n = 4), white-fronted capuchins (n = 4), black spider monkeys (n = 6), lion-tailed macaques (n = 4), de Brazza monkeys (n =3), Sykes monkeys (n = 5), patas monkeys (n = 5), black macaques (n = 2), barbary macaques (n = 11), hamadryas baboons (n = 8)/Chester Zoo (UK) Orangutan (n = 11)/ Singapore Zoo (Singapore) Between-groups, repeated conditions Within-groups, repeated conditions Behaviour - Behaviour directed by the animals towards audience (e.g. threat, submission, begging) - Locomotor activity (movement) Behaviour: - Categories: idle (incl. motionless, autogrooming, expelling waste, or looking around), look at visitors, move, feed, play/social, regurgitate, beg rules*) Direct behaviour obs: - Two methods during 1min period: 1) Continous recording of behaviour directed towards audience; one-zero sampling at 10 s intervals of locomotor activity Direct behaviour obs: - Scan sampling using instantaneous sampling at 10min intervals Study duration/ data collection Study period = NS: - Total 483x 1min observation periods (unspecified data collection period) 5mth study period: - weekdays and weekends between 0930 and 1700hrs -192hrs of obs total Study details Chamove et al. (1988)** Choo et al. (2011) 313

332 Table C. Continued Study focus (human element) Visitors - multiple factors Cage changes with different numbers of visitors to each (low, medium, high - undefined) Visitors - multiple factors Interaction of intrinsic (e.g. animal personality, sex) and extrinsic (e.g. animal group) factors in response to crowd size (low v high, based on zoo attendance numbers) Visitors - multiple factors Interaction of enrichment (new and 1day old) and visitor numbers (high/weekend and low/weekday) Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Golden-bellied mangabeys (n = 10)/Sacramento Zoo (USA) Western lowland gorilla (n = 18)/Zoo Atlanta (USA) Chimpanzees (n = 11)/Los Angeles Zoo (USA) Within-groups, repeated conditions Within- and between- groups, repeated conditions Withingroups,repeated conditions Behaviour: - Categories: aggressive displays (towards people, towards primates in other cages, or within group), grooming and sexual behaviour, play Behaviour: - Activity budget categories: object examination, self-directed behaviour, feed forage, and undesirable behaviour - Social behaviour categories: affiliative, noncontact aggression, contact aggression, and displace - Rates of wounds (from other gorillas) Behaviour: - 12 time-use categories: aberrant behaviours, caregiving/receiving between mothers/young, fighting and displaying, exploring, foraging, feeding, grooming, playing, using objects, sex, traveling, watch/idle - 6 social contexts: friendly, agnostic, mother and infant, public oriented, solitary and nonsocial, submissive rules*) Direct behaviour obs: - Field notes during 15min observation sessions at each cage (n = 3) (no standardised sampling method during field note collection; times/days unspecified) Direct behaviour obs: - two methods during 1hr sessions (balanced across the day, on different days of week): 1) scan sampling using instantaneous sampling at 5min intervals for activity budget; 2) ad libitum one-zero sampling for social behaviour - Wound rates obtained from keeper records Video recorded behaviour obs: - Scan sampling using instantaneous sampling at 60s intervals Study duration/ data collection Study period = NS - total 453 sessions 4yr study period: - data collected approx. 2x/wk per group (n =4) - total of 66hrs low and 53hrs high selected for analysis 3mth study period: - videotaped on 4 consecutive days (Sat-Tues)/wk for 5wks - 15hrs of recordings randomly selected from total for analysis Study details Mitchell et al. (1991a) Stoinski et al. (2012) Wood (1998) 314

333 Table C. Continued Study focus (human element) Visitors - other Reduced visibility - provision of camouflage net barrier to viewing area of exhibit Visitors - other Visitor 'size' (with groups of >5 asked to approach the viewing window and watch for at least 5mins, either a) standing as tall as possible, or b) crouched so only their heads were above the base of the window) Visitors - other Interaction sequences between animals and visitors Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Western lowland gorilla (n = 6)/Belfast Zoo (UK) Cotton-top tamarins (n = 3), Diana monkeys (n = 5), and ring-tailed lemurs (n = 4)/Edinburgh Zoo (UK) Chimpanzees (n = 24)/Chester Zoo (UK) Within-groups, pre/post Within-groups, repeated conditions Behavioural: - Activity budget (resting, standing, sitting, moving, looking, position, aggression, social, abnormal, autogroom, bang glass) Behaviour: - Behaviour categories: agnostic, grooming, affiliation, inactivity - Glances toward viewing window Between-groups Behaviour: - Interactive categories: eye contact, vocal, nonvocal sounds, locational change, gestural, begging rules*) Direct behaviour obs: - Scan sampling using instantaneous sampling at 5min intervals for 2hrs ( h) Direct behaviour obs: - Two methods during 1min sessions: 1) scan sampling using instantaneous sampling at 5s intervals; 2) ad libitum recording of glances Direct behaviour obs: - Observation of interaction sequences (that is, any of the interactive behaviours in which the recipient of the interactions responded to it) between animal and visitor; determined by random selection of human visitor approaching enclosure - Sampling and recording = NS (but assume continuous recording of sequence) Study duration/ data collection 2mth study period - 5days/wk for 1mth baseline and 1mth with barrier Study period = NS - Total data collection unclear (but 8 groups of visitors involved, so may = 8x sessions) 4mth study period: - times/days data collection = NS Study details Blaney & Wells (2004)^^ Chamove et al. (1988)** Cook & Hosey (1995)^^ 315

334 Table C. Continued Study focus (human element) Visitors - other Gender differences in aggression between humans and animals Keepers - training Positive reinforcement of shifting behaviour Keepers - training Long-term monitoring of voluntary husbandry behaviour (presenting tail fluke for blood sampling) Keepers - training Impact of training (outside of training hours) Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Golden-bellied mangabeys (n = 11)/Sacramento Zoo (USA) Giant pandas (n = 2)/Zoo Atlanta (USA) Harbor porpoises (n = 4)/ Fjord and Baelt research and visitor experience facility (Denmark) Abyssinian colobus monkey (n =8)/ Paignton Zoo (UK) Withingroups,repeated conditions Within-groups, pre/post Within-groups, AB longitudinal Within-groups, repeated conditions Behaviour: - Aggressive behaviour: facial displays indicating threat (openmouth or brow threat) Behaviour: - Compliance with training command (i.e. performance of conditioned response) rules*) Direct behaviour obs: - Ad libitum, continuous recording of aggressive behaviour displays throughout 15min observation sessions per enclosure (n = 3) (times/days unspecified) Direct behaviour obs: - Recording of whether or not animal complied with training commands before and after training period Hormone (cortisol) levels Blood cortisol sampling: - Two sampling methods for comparison: (1) collected after removing animal from water (taken between 10-40mins of removal from water); (2) collected during voluntary husbandry behaviour in water (procedure lasting 1-3 mins) - blood taken from dorsal side of fluke (in both conditions) Behaviour: - State behaviours (categories: social, allogroom, play, threat, proximity, rest social, rest, feed, move, fight) - social behaviours (proximity to other colobous, and colobusinitiated interactions with humans - including keepers, staff and visitors) Direct behaviour obs: - Two methods: 1) Scan sampling using instantaneous sampling at 30min intervals (between hrs) for state behaviours; 2) 2x 10min instantaneous focal follows (10s intervals) per individual per day (social behaviours) Study duration/ data collection Study period = NS - total 453 sessions 8mth study period: - 45 sessions over 1mth pretraining; 3-4 sessions/wk over 7mth training period Study period varied per animal (from 1-7yrs): - on average, at least one sample/mth per animal (number of samples for each method = NS) >3mth study period: - 4x 12 day periods (1 prior to training, and once per mth for 3mths after training started) Study details Mitchell (1992a) Bloomsmith et al. (2003) Desportes et al. (2007) Melfi & Thomas (2005) 316

335 Table C. Continued Study focus (human element) Keepers - multiple factors Combined impact of positive reinforcement training and play therapy Keepers - other Animals' friendliness to keeper (keeperrated) Visitors and keepers Presence of keepers within and in-front of exhibit, with and without visitors present Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Lowland gorillas (n = 7)/ Barcelona Zoo (Spain) Black rhinos (n = 26)/10 zoos; white rhinos (19)/6 unspecified zoos Zulu suni (n = 2), slenderhorn gazelle (n = 2-3), Dorcas gazelle (n = 6-7), impala, yellow-black duiker (n = 2-3), Mhorr's gazelle (n = 5), lowland nyala (n = 3-4), Nile lechwe (n = 2-3), Arabian oryx (n = 3), bongo antelope (n = 7-8), sable antelope (n = 2-3), greater kudu (n =2)/zoo = NS Within-group, pre/post Behaviour: - Individual behaviours (neutral - inactivity, locomotion, feeding); positive/desirable - acrobatic play, instrumental play; negative/undesirable - deficit behaviours (multiple categories), interaction with the public) - Social behaviours (positive/desirable - locomotion, feeding, social play, social grooming, affiliative behaviours; negative/undesirable - threatening, agnostic)frequency and duration Within-groups Corticoid metabolite values; Keeper-ratings of animal friendliness Within- and between-groups, repeated conditions - Behaviour: - Categories: interactive (visual orientation, approach, physical contact, flee, social play, sexual behaviour, threat, aggression); non-interactive (eat, drink, eliminate, sniff, vocalize, pace, nonsocial play) rules*) Direct behaviour obs: - Continuous focal animal recording for 10mins per animal between hr baseline (times NS for experimental) until 10hr recorded per individual Fecal sampling (collected in the mornings) Zoo keeper survey Direct behaviour obs - Focal animal sampling of one male and one female of each species for 10s/min each, using all-occurrence recording, per 30 minute observation period - four conditions: (1) keeper within exhibit, with visitors present; (2) keeper in-front of exhibit, with visitors present; (3) keeper within exhibit, with visitors absent; (4) keeper infront of exhibit, with visitors absent. Study duration/ data collection Two repetitions of the study, 1yr apart: - 2x baseline and 2x experimental conditions, with 10hrs/animal per condition - total number of days = NS 1yr study period: - wkly fecal sampling Approx. 1.5mth study period : - all obs conducted 3hrs prior to sunset - 48hrs total observations (number of obs per condition = NS) Study details Carrasco et al. (2009) Carlstead & Brown (2005) Thompson (1989) 317

336 Table C. Continued Study focus (human element) Visitors and keepers Threat responses of animals to different humans (observers, keepers, visitors) as well as conspecifics Sample/Zoo Design Outcome measure/s Sampling method (for behaviour: recording medium, sampling rules, and recording Golden-bellied mangabeys (n = 10)/Sacramento Zoo (USA) Within-groups, repeated conditions Behaviour: - Aggressive behaviour: facial displays indicating threat (openmouth or brow threat) rules*) Direct behaviour obs: - Ad libitum, continuous recording of aggressive behaviour displays throughout 15min observation sessions at unspecified times on weekends Study duration/ data collection 8mth study period: - total 453 sessions (unspecified number of days) and heterospecifics * Recording medium = direct observation, video recording, etc.; sampling rules = ad libitum, focal, scan, or behaviour sampling; recording rules = continuous recording, or time sampling (including Study details Mitchell et al. (1991b) instantaneous sampling or one-zero sampling) (Martin & Bateson, 2007); ^ obs = observation; ** study included multiple human impact categories and is therefore presented multiple times within the table; ^^ study included other, non-animal elements (e.g., visitor surveys) which are not reported here; NS = not stated; # range in sample size due to fluctuations in included animals during the study (e.g., due to deaths or removal from exhibits). 318

337 References Anderson, U.S., Benne, M., Bloomsmith, M.A., & Maple, T. (2002). Retreat space and human visitor density moderate undesirable behavior in petting zoo animals. Journal of Applied Animal Welfare Science, 5, Birke, L. (2002). Effects of browse, human visitors and noise on the behaviour of captive orangutans. Animal Welfare, 11, Blaney, E.C., & Wells, D.L. (2004). The influence of a camouflage net barrier on the behaviour, welfare and public perceptions of zoo-housed gorillas. Animal Welfare, 13, Bloomsmith, M.A., Jones, M.L., Snyder, R.J., Singer, R.A., Gardner, W.A., Liu, S.C., & Maple, T.L. (2003). Positive reinforcement training to elicit voluntary movement of two giant pandas throughout their enclosure. Zoo Biology, 22, Carder, G, & Semple, S. (2008). Visitor effects on anxiety in two captive groups of western lowland gorillas. Applied Animal Behaviour Science, 115, Carlstead, K., & Brown, J.L. (2005). Relationships between patterns of fecal corticoid excretion and behaviour, reproduction, and environmental factors in captive black (Diceros bicornis) and white (Ceratotherium simum) rhinoceros. Zoo Biology, 24, Carrasco, L., Colell, M., Calvo, M., Abello, M.T., Velasco, M., & Posada, S. (2009). Benefits of training/playing therapy in a group of captive lowland gorillas (Gorilla gorilla gorilla). Animal Welfare, 18, Chamove, A.S., Hosey, G.R., & Schaetzel, P. (1988). Visitors excite primates in zoos. Zoo Biology, 7, Choo, Y, Todd, P A, & Li, D. (2011). Visitor effects on zoo orangutans in two novel, naturalistic enclosures. Applied Animal Behaviour Science, 133, Cook, S, & Hosey, Geoffrey R. (1995). Interaction sequences between chimpanzees and human visitors at the zoo. Zoo Biology, 14, Davis, N., Schaffner, C.M., & Smith, T.E. (2005). Evidence that zoo visitors influence HPA activity in spider monkeys (Ateles geoffroyii rufiventris). Applied Animal Behaviour Science, 90, Desportes, G., Buholzer, L., Anderson-Hansen, K., Blanchet, M-A., Acquarone, M., Shephard, G.,... Siebert, U. (2007). Decrease stress; train your animals: The effect of handling methods on cortisol levels in harbour porpoises (Phocoena phocoena) under human care. Aquatic Mammals, 33, Glatston, A.R., Geilvoet-Soeteman, E., Hora-Pecek, E., & van Hooff, J.A.R.A.M. (1984). The influence of the zoo environment on social behavior of groups of cotton-topped tamarins, Saguinus oedipus oedipus. Zoo Biology, 3, Hosey, G.R. (1989). Behavior of the mayotte lemur, Lemur fulvus mayottensis, in captivity. Zoo Biology, 8, Hosey, G.R., & Druck, P.L. (1987). The influence of zoo visitors on the behavior of captive primates. Applied Animal Behaviour Science, 18, Lukas, K.E., Barkauskas, R.T., Maher, S.A., Jacobs, B.A., Bauman, J.E., Henderson, A.J., & Calcagno, J.M. (2002). Longitudinal study of delayed reproductive success in a pair of white-cheeked gibbons (Hylobates leucogenys). Zoo Biology, 21, Mallapur, A., Sinha, A., & Waran, N. (2005). Influence of visitor presence on the behaviour of captive lion-tailed macaques (Macaca silenus) housed in Indian zoos. Applied Animal Behaviour Science, 94,

338 Mansour, A.A.H, Zakaria, A-H., & Fraser, A.F. (2000). Effect of enclosure quality on reactivity and welfare of captive soemmerring's gazelle (Gazella soemmerringii). Journal of Applied Animal Welfare Science, 3, Martin, P., & Bateson, P. (2007). Measuring behaviour: An introductory guide (3rd ed.). New York: Cambridge University Press. Melfi, V.A., & Thomas, S. (2005). Can training zoo-housed primates compromise their conservation? A case study using Abyssinian colobus monkeys (Colobus guereza). Anthrozoos, 18, Mitchell, G, Tromborg, C T, Kaufman, Jennifer, Bargabus, Shawna, Simoni, Rosileen, & Geissler, Victor. (1992c). More on the 'influence' of zoo visitors on the behvaiour of captive primates. Applied Animal Behaviour Science, 35, Mitchell, G., Herring, F., & Obradovich, S. (1992a). Like threaten like in mangabeys and people? Anthrozoos, V, Mitchell, G., Herring, F., Obradovich, S., Toromborg, C., Dowd, B., Neville, L.E., & Field, L. (1991a). Effects of visitors and cage changes on the behaviors of mangabeys. Zoo Biology, 10, Mitchell, G., Herring, F., Tromborg, C., Dowd, B., Steiner, S., & Obradovich, S. (1992b). Targets of aggressive facial displays by golden-bellied mangabeys (Cercocebus galeritus chrysogaster) at the Sacramento Zoo. Applied Animal Behaviour Science, 33, Mitchell, G., Obradovich, S., Herring, F., Dowd, B., & Tromborg, C. (1991b). Threats to observers, keepers, visitors, and others by zoo mangabeys (Cercocebus galeritus chrysogaster). Primates, 32, Nimon, A.J., & Dalziel, F.R. (1992). Cross-species interaction and communication: a study method applied to captive siamang (Hylobates syndactylus) and longbilled corella (Cacatua tenuirostris) contacts with humans. Applied Animal Behaviour Science, 33, Skyner, L.J., Amory, J.R., & Hosey, G.R. (2004). The effect of visitors on the selfinjurious behviour of a male pileated gibbon (Hylobates pileatus). Der Zoologische Garten, 74, Stoinski, T.S., Jaicks, H.F., & Drayton, L.A. (2012). Visitor effects on the behavior of captive western lowland gorillas: The importance of individual differences in examining welfare. Zoo Biology, 31, Thompson, V.D. (1989). Behavioral response of 12 ungulate species in captivity to the presence of humans. Zoo Biology, 8, Todd, P.A., Macdonald, C., & Coleman, D. (2007). Visitor-associated variation in captive Diana monkey (Cercopithecus diana diana) behaviour. Applied Animal Behaviour Science, 107, Wells, D.L. (2005). A note on the influence of visitors on the behaviour and welfare of zoo-housed gorillas. Applied Animal Behaviour Science, 93, Woods, B. (1998). Animals on display: Principles for interpreting captive wildlife. Journal of Tourism Studies, 9, Wormell, D., Brayshaw, M., Price, E., & Herron, S. (1996). Pied tamarins (Saguinus bicolor bicolor) at the Jersey Wildlife Preservation Trust: Management, behaviour and reproduction. The Dodo: Journal of the Jersey Wildlife Preservation Trust, 32,

339 Appendix D Key methodological details of some past zoo visitor studies (impact of a general visit) To complement the information in Chapter Three, this section contains key methodological details of 15 general zoo visitor studies, published between 1992 and

340 Table D. Summary of key methodological details from peer-reviewed studies, published between , of the impact of a general zoo visit on the visiting public Study focus (Zoo element) Impact of aquarium visit on visitors' shortand long-term knowledge, attitude, behaviour change^ Interaction of variables (education, gender, leisure reading and knowledge about subject matter) on attitudes toward antelopes Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/days Adult visitors (n = Within- and 3mth period at 306)/National betweengroups, aquarium (plus Aquarium in follow-up): Baltimore (USA) longitudinal - mostly weekdays Visitors, age = NS (n = 128) /Birmingham Zoo (USA) Three methods: Part 1: Medium: Semi-structured interview Time point/s: pre (n = 100) and/or post visit (n = 203) Location: Zoo entry and/or exit Participation time: 15-20mins Sampling method: random Part 2: Medium: Personal meaning mapping Time point/s: pre and post visit (n = 103) Location: At zoo entry and exit Participation time: NS Sampling method: random Part 3: Medium: Telephone interview Time point/s: 6-8wks follow-up (n = 48) Location: N/A Participation time: 10-15mins Sampling method: N/A (taken from samples in Parts 1 and 2) Correlational Medium: Survey (NS) Time point/s: Once only Location: zoo entry and exit Participation time: NS Sampling method: NS Part 1: Questionnaire: Pre-visit: - Awareness and understanding of the concept of conservation; conservation-related knowledge, concerns and behaviours; perceptions of relationship to conservation issues; Chesapeake Bay-related knowledge (types of questions = NS) - Based on past research: Partially Post-visit: - Sense of aquarium's message, conservation at the aquarium, and motivation to be involved in conservation (types of questions = NS) - Based on past research: No Part 2: Personal meaning mapping: - extent, breadth, depth, and emotional intensity associated with understanding of conservation - Based on past research: Yes Part 3: Telephone interview: - Same content as post-visit questionnaire and personal meaning mapping Questionnaire: - Attitudes toward antelopes (rating scale questions) rating scale on three items for antelopes - Based on past research: No Study details Adelman et al. (2000) NS Bitgood (1992) 322

341 Table D. Continued Study focus (Zoo element) Relationship between zoo experience and attitudes toward animals/species^ Impact of visit to Conservation Station on long-term intended conservation action Impact of prior knowledge and interest on visitor attitudes to learning Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/days 3mth period - Days = NS Adult visitors (n = 206)/ Cleveland Metroparks Zoo (USA) Adult visitors (n = 702)/Disney's Animal Kingdom (USA) Adult visitors (n = 100)/National Aquarium in Baltimore (USA) Correlational Medium: Interview Time point/s: Once only Location: 4 sites in zoo Participation time: NS Sampling method: NS Within- and betweengroups longitudinal Within- and betweengroups Medium: Interview Time point/s: pre and/or post visit at zoo; 2-3mths follow-up (n = 72) Location: At exhibit and telephone interview Participation time: NS # Sampling method: random Medium: Interview Time point/s: pre and post visit Location: At zoo entry/exit Participation time: mins total Sampling method: random Questionnaire: - Attitudes toward individual animal and species (rating scale questions) - Emotions (rating scale questions) - Perception of animal (rating scale questions) - Motivation for visiting (multiple choice question) - Based on past research: No Questionnaire: Pre/post/follow-up: - Prior/future/current conservation behaviour actions (rating scale questions) Based on past research: Yes Questionnaire: Pre-visit: - Awareness, knowledge, attitudes, and perceptions related to conservation (multiple-choice, rating scale and open-ended questions) Post-visit: - As above, plus understanding of aquariums message, conservation at the aquarium, and motivation to be involved in conservation (closed and open-ended questions) - Based on past research: No 6mth study period for initial survey at zoo - Days = NS 3mth period: - Days = NS Study details Clayton et al. (2009) Dierking et al. (2004) Falk & Adelman (2003) 323

342 Table D. Continued Study focus (Zoo element) Impact of aquarium visit on knowledge, attitudes and behaviour toward seafood Relationship between visitor predispositions and cognitive and affective experiences and reactions Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/days 2yr study period: - Days = NS Visitors, age = NS (n = 726)/Monterey Bay Aquarium (USA) Adult visitors (n = 825)/Brookfield Zoo (USA) Withingroups, longitudinal Predictive correlational Two-part study: Medium: Survey (NS) Time point/s: Post-visit and 4mths follow-up (n = 400 visitors) Location: Zoo exit and telephone interview Participation time: NS Sampling method: Convenience (those who took a Seafood Watch pocket guide upon exiting) Part 2: Medium: Focus-group (n = 64 visitors; n = 6 groups) Time point/s: NS Location: NS Participation time: NS Sampling method: NS (but stated that participants volunteered; subset of original sample) Medium: Self-report questionnaire Time point/s: post-visit Location: Four exhibits Participation time: 5-10mins Sampling method: random Part 1: Questionnaire: Post-visit: - Knowledge about sustainable food issues, attitudes regarding environmental and sustainable food issues, interest in sustainable seafood tools/information, previous use of Seafood Watch pocket guide, current seafood purchasing behaviours (types of questions = NS) Follow-up: - As above, plus perception of impact of pocket guide on awareness and behaviour (type of questions = NS), and use of pocket guide (openended questions) - Based on past research: No Part 2: Focus groups: - Experience using Seafood Watch pocket guide (questions = NS) - Based on past research: No Questionnaire: - Exhibit experiences - personal thoughts, feelings, and activities at the exhibit (rating scale questions) - Overall reactions - fun and enjoyment, concern and empathy for animals and nature, interests, understanding, and beliefs about animals/conservation (rating scale questions) - Based on past research: Yes 10wk study period: - 1hr per day at various times (between 1100 and 1400hr) - days = NS Study details Kemmerly & Macfarlane (2009) Luebke & Matiasek (2013) 324

343 Table D. Continued Study focus (Zoo element) Entry versus exit knowledge and attitudes towards African apes Impact of zoo visit on knowledge regarding Lion-tailed macaques (compared to general public) Relationship between feelings and cognitions after visiting polar bear exhibit, and conservation attitudes Adult visitors and public (n = 900)/Shri Chamarajendra Zoological Gardens, Aringnar Anna Zoological Park, Thiruvananthapura m Zoo and accompanying city centres at locations of each zoo (India) Adult visitors (n = 30)/Ouwehands Zoo and Rotterdam Zoo (Netherlands) Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/days Adult visitors (n = Betweengroups 15wk study 1000)/Lincoln Park period: Zoo (USA) - various times on weekdays Betweengroups Exploratory case study/correla tional Medium: Interview Time point/s: Before or after viewing Great ape exhibit Location: Entry/exit of Great ape exhibit Participation time: 10mins Sampling method: Random Medium: Interview Time point/s: Once only Location: Zoo exit or city centre Participation time: NS Sampling method: Random Medium: Interview Time point/s: post-visit Location: NS Participation time: 30mins Sampling method: random Questionnaire: - Knowledge about chimpanzees or gorillas (true/false questions) - Attitudes to chimpanzees or gorillas (rating scale questions) and conservation attitudes towards gorillas and chimpanzees - Based on past research: Partially Questionnaire: - Knowledge about Lion-tailed macaques (closed and open-ended questions) - Based on past research: No Semi-structured interview: - Questions assessed general response to viewing animal, feelings associated with animal, knowledge about animal (open-ended questions) - Value orientations toward animal and attitude to conservation (rating scale questions) - Based on past research: Partially Study details Lukas & Ross (2005) NS Mallapur et al. (2008) 1mth study period: - 3 days/zoo (weekend and weekday) Marseille et al. (2012) 325

344 Table D. Continued Study focus (Zoo element) Emotional impact of viewing different species (okapi, snakes, gorilla) Impact of visits to orangutan exhibits on visitor knowledge, attitudes, and intentions for future behaviour Connection to wildlife and influence of viewing charismatic mega fauna on conservation outcomes; zoo versus wild experience Adult visitors (n = 279)/ Brookfield Zoo (USA) Adult visitors (n = 240)/Adelaide Zoo, Melbourne Zoo, Taronga Zoo (AUS) Visitors, age = NS (n = 868) /Brookfield Zoo, Shedd Aquarium, Zoo Atlanta (USA), and various wildlife parks (Tanzania) Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/days Betweengroups 3mth study period: - Days = NS Betweengroups and predictive correlational Predictive correlational; betweengroups Medium: Self-report questionnaire Time point/s: Once only Location: At okapi, snake, or gorilla area Participation time: NS Sampling method: Random Medium: Self-report questionnaire Time point/s: Once only (per zoo) Location: Orangutan exhibit exit Participation time: 5-10mins Sampling method: Random Medium: Survey (NS) Time point/s: zoo = during visit (n = 452); wild = post-visit (n = 416) Location: zoo = within zoo (not entry/exit); wild = Tanzania airport Participation time: NS Sampling method: zoo = systematic; wild = census approach Questionnaire: - Emotions toward the animals, as well as evaluative and arousal emotions, when viewing exhibit (rating scale items) - Based on past research: Yes Questionnaire: - Knowledge about orang-utans (multiple choice quiz, as well as yes/no and open-ended questions) - Attitudes toward orang-utans (rating scale; Animal Attitudes Scale) - Intentions for future behaviour (yes/no responses) - Subjective norm regarding orang-utan conservation (rating scale) - Experience satisfaction (rating scale questions) - Based on past research: Yes Questionnaire: All rating scale questions - Connection to wildlife - Species characteristics (e.g., physical and emotional attributes) - Trip characteristics (e.g., authenticity and interspecies interaction) - Conservation caring (i.e., connection to species) - Conservation behaviours (species-specific and general) - Based on past research: Yes Study period = NS - 6 days (one weekday and one weekend per zoo) 3mth study period at zoo 1mth study period in Tanzania Study details Myers et al. (2004) Pearson et al. (2013) Skibins et al. (2013) 326

345 Table D. Continued Study focus (Zoo element) Relationship between conservation caring and conservation behaviour^ Impact of zoo visit on conservation outcomes Sample/Zoo Design Method Instrument/Outcome measure/s Study duration/days Predictive 3mth study correlational period: - Days = NS Visitors, age = NS (n = 862)/Brookfield Zoo, Shedd Aquarium, Zoo Atlanta (USA) Age unspecified visitors (n = 786)/Philadelphia Zoo (USA) Medium: Survey (NS) Time point/s: pre (n = 411) and during visit (n = 452) Location: zoo entry points (pre) and within zoo (during) Participation time: NS Sampling method: random Correlational Medium: Self-report questionnaire Time/points: Once only Location: Zoo exit Participation time: 10mins Sampling method: NS (relied on voluntary approaches from visitors, and also approached visitors) Questionnaire: All rating scale questions - Connection to wildlife - Conservation caring (i.e. connection to species) - Species and biodiversity oriented behaviours - Based on past research: Yes Questionnaire: - Conservation outcomes - conservation motivation, knowledge, and attitudes/values; pro-conservation consumer skills; readiness to take conservation action (all rating scale questions and asked retrospectively as well as current) - Experience at zoo - interactions with staff, reactions to and experience with exhibits, and engagement with conservation education (frequency and rating scale questions) - Based on past research: No, but developed in consultation within zoo 2x 1mth blocks (1yr apart) - days = NS Study details Skibins & Powell (2013) Wagner et al. (2009) ^ = study involved other elements (such as evaluations of other experiences or behaviour observations) but since these are not related to the impact of interactive animal experience on visitor outcomes (i.e., knowledge, attitudes, emotions, or behaviour) these elements have been excluded; NS = information not stated. 327

346 References Adelman, L.M., Falk, J.H., & James, S. (2000). Impact of National Aquarium in Baltimore on visitors' conservation attitudes, behavior, and knowledge. Curator: The Museum Journal, 43, Bitgood, S. (1992). The impact of a zoo visit on attitudes: A preliminary report on interaction effects. Visitor Behavior, 7, Clayton, S, & Brook, A. (2005). Can psychology help save the world? A model for Conservation Psychology. Analyses of Social Issues and Public Policy, 5, Dierking, L.D., Adelman, L.M., Ogden, J., Lehnhardt, K., Miller, L., & Mellen, J.D. (2004). Using a behavior change model to document the impact of visits to Disney's Animal Kingdom: A study investigating intended conservation action. Curator: The Museum Journal, 47, Falk, J.H., & Adelman, L.M. (2003). Investigating the impact of prior knowledge and interest on aquarium visitor learning. Journal of Research in Science Teaching, 40, Kemmerly, J.D., & MacFarlane, V. (2009). The elements of a consumer-based initiative in contributing to positive evironmental change: Monterey Bay Aquarium's Seafood Watch program. Zoo Biology, 28, Luebke, J.F., & Matiasek, J. (2013). An exploratory study of zoo visitors' exhibit experiences and reactions. Zoo Biology, 32, Lukas, K.E., & Ross, S.R. (2005). Zoo visitor knowledge and attitudes toward gorillas and chimpanzees. The Journal of Environmental Education, 36, Mallapur, A, Waran, N, & Sinha, A. (2008). The captive audience: The educative influence of zoos on their visitors in India. International Zoo Yearbook, 42, Marseille, M.M., Elands, B.H.M., & van den Brink, M.L. (2012). Experiencing polar bears in the zoo: feelings and cognitions in relation to a visitor's conservation attitude. Human Dimensions of Wildlife, 17, Myers, O E, Saunders, C D, & Birjulin, A A. (2004). Emotional dimensions of watching zoo animals: An experience sampling study building on insights from psychology. Curator, 47, Pearson, E.L., Dorrian, J., & Litchfield, C.A. (2013). Measuring zoo visitor learning and understanding about orangutans: evaluation to enhance learning outcomes and to foster conservation action. Environmental Education Research, 19, Skibins, J.C., & Powell, R.B. (2013). Conservation caring: Measuring the influence of zoo visitors' connection to wildlife on pro-conservation behaviors. Zoo Biology, 32, Skibins, J.C., Powell, R.B., & Hallo, J.C. (2013). Charisma and conservation: charismatic megafauna's influence on safari and zoo tourists' pro-conservation behaviors. Biodiversity and Conservation, 22, Wagner, K., Chessler, M., York, P., & Raynor, J. (2009). Development and implementation of an evaluation strategy for measuring conservation outcomes. Zoo Biology, 28,

347 Appendix E Maps of Adelaide Zoo and Monarto Zoo 329

348 330

349 331

350 Appendix F Published version of Chapter Four 332

351 333

352 334

353 144 Szokalski et al. emajj via sev<.-n const.-'lltiog wos/wildlife parks io.l\us. tr<llia and NC\\' Zealand. Respondents ranged io <lh'e. v,:ith 31%aged bctwceo 20 and 29 yea.rs; 31% ag years; 21% aged years; 4% aged years; and 1% aged 60+. The majority of pa rticipa ots \\'t.-"tc employed io public facilitjes (11 = 63), v,:ith a further 19 in priv-.tte facilitjes (four o thet/uojdentified). Approxima tely ha lf v,:o rked in the United Stales (51%), 16% in Australia, 1 lo/o in NC\\' Zealand, a nd 11% io the United Kjngdom (11% from v-j.r ious o ther countr ies; one uospccified). ~1os t pa rticipa nts were working io keeper roles (11 = SO), v,:ith a further 26 holding senior keeper positjons, and seven holdjng hjgher positioos (such as zoo manaj,>e-r or director; three uospooified). ~1ean yean> of employment in current role \\'t.-"tc 6.2,9 (SD = 6.54), v,:ith a miojmum of 2 months and a maximum of 33 y<.-ars. A large proportjon of pa rticipa nts (69'l/o) had pre\'ious.ly \\'Orked in the zoo ind ustry, with a range of 8 months Lo 35 yean' e:tpt.-tiencc (r\1 = 7.65years, SD= 1.66). Questionnaire.l\ self report q ucstioooa irc w4s generated in Tel1U s2. - a U ni.,1:rs.ity of South Australia ooline sur\'ey soflw4rc applic.ation. ln addition to demographic details, the fojjo..,,ing ioformatioo wjs sought: IJetermining currtn.t liandlinx practices and us.e of training P.artjcipaols were asked which species of big cat/ s they currently work \\'ilh, what lt..,,"ci of ha odling they practjoe with them, a nd \\'hether or not they practioc: Lr4ining. P..ir ticipan ts were also asked to indicate, on average, ho\\' much daily tjme they spend in coo tact..,,ith, a od/or tr<tioing, the anima ls. hasic opinions of the different li11ndling pr11dices and training. P.<trtjcipaols v.'cre asked how strongly they agreed that the t.'llrrcnl handling and tr<tiojng pr<tctices t."mployed at their instjtutioo a llow the needs of big cal~ keepers, and \'isiton> to be s.imultancous.ly met. T hey were asked \\'hich of the Lhr<.'C levels of contact they belje\>e to be be:st for each of the three groups, and v. hether or not Lr4jning is bt.-neficia l to each of the throe groups (ycs/ oo / uosurc response). lnfomiadon specific to the different lei-els ofli11ndling and training. To allo\\' a more in-depth understandiog of pa r tjcipanl opinions, respondents v.'cre asked to r<tte ho\\' strongly they agn:c (wht."tc I = strongly di.sagr<.'c and S = strongly agroe) that each Je\>el of ha odling and Lr4jning contributes Lo five difft."tcnl benefi ts to big cats, k<.-cpers, and visitors (these bc.nefils ""'C-re deri\'cd from the Jiter4 ture about these practices \\'ith o thet species; limitatjons were not asse:sscd due to a lack of ioformation about these in the ljter<tture): I. Benefits Lo big cats: imprm-ed physical \\'ell being; impro.,-ed psychological \\'<:II-being; impro\ ed cognili\'e stimulatjon; improved social stjmulation; impro.,"cd behavioral stimulatjon. 2. Benefits to keepcn : impro\ ed relatiooshjp v. ith big cats; impro.,"cd mo\"cment of big cats (e.g., betwceo e n closures); impro\'cd enclosure maintenance; improved job s.atisfactioo; impro.,"cd ca re of big ca ts (e.g., vcteri na ry clt.ects). 3. Benefits to \'isiton>: impro\ ed enlerta inmeot; improved knowledge of big cats; impro\ ed e njoyment; improved "oonot."ction" with big cats; imprm-ed information about big cat coosefv'4tion methods. lo o rder to ideotify additjonal perocived cooseq uenccs of these practioc:~ participan ts ""'C-re. asked Lo describe any other benefits or limitations associated with each handliog type and tr<tin ing in open-ended questioos. Statistical Analysis Basic percentages ""'C-re. examined to understaod the use of the diffe~-n l handling methods and tra ining, as well as respondents' outright opioions of these. To determine whic h practjce (hands-on, prott.'cted, and hands-off contact, and training) wjs viewed as most bc.neficial for each of the three groups (big cat~ keepers, and visitors), rat ingsforeach group\\'t:re pooled to gent.t<tlea mean benefi t rating, r4nging from I to S (wht."tc 1 = strongly d isagree and S = strongly agree with the beoefits Lo each group). T he responses to these statements v.'cre no t normally d istributed, and both Jog and square root transformations fajjed to normalize the data. As such, the~ scores V.'Ctl! subjected to Friedma o's..\nova analyses, v. ith \Vikoxon signed-rankoo post hoes (Siegal and C-J.SteUa o, 1998] in SPSS \"Crsion 17 (all post hoc oomparisoos \\'t.-"tc cond ucted v. ith a Dooferroni correction applied, corrected a = 0.008). G iven the explor<ttory nature of the opcnended questions regarding parlicipaol pen:ci\'cd benefi ts and limitations of each handliog pr.ictice a od of Lrainiog for the thn:c group~ the~ results were not subjected to statistical analysis. l nstead, the~ qualitati""c data v.-ere o r ganized into categories [similar Lo Hosey and ~1elfi, 2010; Wilson e l al., 2003]. RESULTS How Commonly Performed Arc the Three Different Handling Methods and Trai ning with Big Cats? Results r<.,-vca JOO tha t 19.S /o (n = 11) of partic ipan ts pr<tclicc hands-on contact, 87.2 /o (n = 15) practjce protected contact, and 20.9'>/o (n = 18) practice hands-off contact. A further 11.9'>/o (11 = 67) pr<tt.1.ice Lrainiog. Rates 335

354 Zoobepef..Sig Cat lnterectiont 145 TARI. I. Rates or budllna aod trtlaina n a perceotttt et each o r lhc three i:roups (big cats, koc:pers. lnd visitor'))...,.pm; worldq wltli tada sptclctl Re5ulu "-tc individually discussed for each group. Spcries (11 = nunlbc:r of keepers... orking \loith species) Hands-on ProtcOlcd H1utds-01T Training Tlgct (n = 65) 9.23% 89.23% lll.460m 69.23% Lion (11 = 61) 6.56% 81.9'1% 18.03% 61.JOtl. Chtttah (n = 60) 42.31% 57.69'.4 I l.5-l"'.4 S0Jl0% Uop.ud (1t = 27) 7.41% 70.)7% 29.63% 51.85% Jagu;ir (II = 12) 0.00% IJ.m. 31..l.!% SO.DO% Co1:1pr (n = 24) 16.67% IJ.m % SO.DO% Snowkopard ( =21) 9.52'< 90.48% t.:!9% 11.43% TARI. 2. Pf:rttotace of kttpm,..,bo ~ that each c,,.. el hancllitle; and cnloldg" most bnicftclal for th~ arrlmak. kttpm, and \"kltors Hands-on Protected l l!uldfi OfT contact contlci contact Training - Anim.-:l!s 12'A 8% Keepers 7% 114% 9% 99% Va:itots 7% 43% 88% - of t-.andling-styk and tra.iniq ""U1Cld by species. as stiov."d in Table 1. Keepers reported,.pcndanc a mean maximum of 1.94 hr(sd = l.46) (minimum 0.33 and maximum 6.0 hr) engaged in bands-on oontac..'t per day (with aji animals), I, 70 hr (SD= I.47) in prott.:ccc:d contact (mjnimum 0.17 and maximum 8.0 hr), and 1.09 (SD.99) trajning(mjn. imum 0.17 and maximum 4.0 hr). Some rcsponso \lo'cte excluded d ue to nol being llblt to confidentlycajculate the number of hours (four for protected conlru:t and fo ur for training). What Ale KIJepets' Buie: Opinions ol lhe Thlllllenont Levels ol Han<lllng and Training? Respondents v.a-t: ubd to rate how stmngty they agreed thal the current hand.hn& and training pr.!c:ticcs used al their institution alkwo for the combined needs of the animals., keepers. and vi11o1ton to be met. The majority of participants (74.S /o) were in u.arc:cment that their in stitution d id meet th~ ncedi (S4. 7% ttgrcoo and 1 9. ~/o strongly agreed). They we n.: also asked to indicate >,a, hich le\'d of handling and trainin{l they believed is best for the three aforementioned groups. Results are sho\\-n in Table 2. How Strongly Do Keepora Agree w1th the Potential S..neflts ol Each Level of Handling and Training? To pt'o\iid.e a more thorouah uriderstanding of koepa opinions of the cbltcn:nt hllndling methods. pa.r tic::ipants \\'CfC asked to JCllt their aarocmrnt \\ith \'arious statements about the benefits or each kvcl of hand.jing for Koopor Oplntona 01 Handling and Training 1or Big Cats There WI$ a statisticauy significant difference in kccpc...-r pcr<:ci,~d benefits of each leo.~ I of handlin& and tr.aini.na (or bis cats,. x 1 (3) = P < Median {IQR) pcroeh'<d bmdils We hajtds.oa OJOUICt. pn>l<eled coot.act. hand.m>fl" contact. at:d training "~ ).0 ( )..O ().4-4.2). 2.6 ( ). and 4.2 (4»-4.8). re spccth~i)'. Poit hoe comparisons muled.w.pufan1 dif~ facno:. be'lv."ccn the pcroci\'cd benefits or ltlininc and han<b-<>n (T 10, P < 0.001, r= -0.53). pro1oeted (T= ll. P < 0.001, r= ), and hands-off contact (T 4, P < I' - 0.S?), such that training " 'all pen:eh'cd as mon; beneficial than any of the5e praetioci. f urther, protoch.'d contact " 'tts pcm.-ivcd as significantly more bcn efic:ial thun both hnnds-on contact (T = 14, P < r = ) and hands-off c:or.tact (T = 1. P < 0.001, rs ): the diffe ~N.'t' bctwttn the pcroci''cd ~<:fil rot itlp of hands-oa and bar:ds-olt contacl "-.u not imjnificant (P=0.032). A"'6t#IMI #-Nfirs a.ii limi1"1i011j A au.mbcr of participants provided de&.aib of addi tional bcncfili and limitations of the three b'd;of conlllcl and of tntininc for big cats. As seen in Table 3. p;trl.icipants reported u. number of limitatjons associated " 'ith bands on and hand11-oll' contat..1, and more bc:ncfilll llllliociutcd with protcctt.'d contact and tr<tinjng. For each mcthod, participanu al:w noted that the benefitl/limitutions \\'ill depend on other factors, such as the indi, ldu.al animal (for u.amplc. pccies. size. and history). the ltoct UP of the f<tcility and housing style. the stajj' im-ohw. the t)'l)c or prognun and n:m.ron:cmcnt used. as v.-dl u other tnrieh axol/.umulatioc ofl"ercd to the animal Keepor Opinion ol Handling and Ti'olnlng!0< KIJepets There Wll.li a sta:tisticaljy s.igoiflcanl difference in per ccivcd bencot11 or each level of handling and truinina for kecpcn. x 2 ()) 14S.6S, P < O.OS. Median (IQR) per ccivcd bcnc.111$ for hands o n contact, protected contact, handllo(llt cc:mtttl..1, nnd trjining \lo'cre 3.0 ( ). 4.0 ().µ,6), 2.2 ( ), and 4.6 (4.0-S.0). rc>pccliv<ly. Post hoc comparisons rei.'cajed s.igoi cant differences be twllcl'l lhe pcroci\'cd beceftts of tr.aining a.cd barw.t.-on ( T = 5.P <0.001.r -0.53),protcch:d(T= 14.P <0.001,r= -0.37).aod lwlds-olf(t= I. P <0.001.r ) och th.at ll'lllwll -...u pc-ta:n..:d as more bmcficial lhan acyo( tbck practio:s. further, proleck.d contac:t "''*' pcrccl'\'ed u WgniflcantJ)' mon: bcr.c&ial than both hand"" "" con taot(t 18. P < 0.001, r= -0.39)and hanch-oftcont.11.c:t (T= 3. P < 0.001, r = - 0.S?); and hands.on contact w11.s 336

355 146 Szokalski et al. TABLE 3. Addltlo1iaJ beodlu and llniltarion..; of handling and training for big cau Benefits Hands-on Better health/ medical care (n = 3) Greater trust/bond with keeper (n = 2) Higltcr breeding success (n = I) lntpt'o\'cd welfare (n = I) Protected Ssfer(n= 12) Better training and 'l.'cterinaf)' care (n = 7) Reduces stress and aggression (n = 7) tnereascs choice. and control (n = 1) Provides cognitive. stimulation (n = S) En-courages natural behavior (n = 2) Enjoyable.foranimal(n= I) Oeercascs boredom (n = I) Allows for conspecific interaction (n = I) Gh-esanimal morc.space (rt= I) lntpt'o\'cs O\'Crall animal welfare (n = I) Hands-off Gh-es animal ehoioc (n = 2) Allo"''S natural animal behavior (n = I) Beneficial method for rehabilitation animals (n = I) Training Better health/ medical care (n = 8) Stin1u)ating and enriching (rt= 6) Allo"''Sstronger bond between animal and kttpcr (rt= 4) Can reduce negative behaviors (e.g. strcs.i; rt= 3) lntpt'o\'c physical, mental. emotional well being (rt= 3) Limitations Unsafe and leads to animal euthanasia (n = 28) Unnecessary (rt= 9) Influences natural behavior (n = 7) RemO\csehoicc. and control (n = 3) Has negalh'c consoquenocs (e.g.. punishment; rt= 3) Reduces breeding succcs..; (n = 2) RemO\cs mother rearing (n = I) Gives animal wrong impression of humans (n = I) Los.sof dignity for animal (n = I) Limits anin1al's exhibit access (n = I) Potentially reinforces unnatural behavior (n = I) f\ 1ore beneficial for keeper than animal (rt= I) Can lead to negative. behaviors (e.g. stereotym ; n = 8) Unnecessary (rt= 6) Lack of bond bctw«n animal and keeper (n = 4) Lack of health care and behavioral monitoring (n = S) Lack of enrichnlcnt (n = 3) Can be a source ofsuess for anin1al (n = I) Could be.considered unnatural (rt= I) Potentially reinforces unnatural behavior (n = I) peroeivcd a.s significantjy more bcneficiaj than hands-o ff conlacl (1'= 20, P < 0.001, '= - 0.lS). Additional btn~fit1 11nd limitations \Vhilst ft.."\.1.-er comments were made about the benefits/ limitations of eaeh practice for kcc.-pers compared lo the animals, participants s.imilarly noted more limitatjons for hands-on tha n the olhc.-r methods (see Table 4). P.4rtieipanls noted that the benefits/limitations of any pr<tclicc for keepers v. ilj depend on the q uajificalion/experiencc of the individ ua l keeper, the.species of cat, compliance of the a nimal, the purpose of the parlicujar prat..-i.icc in any situatio n, the type of rein foroemenl used, a nd the q uajity of the program. Keeper Opinions of Hand ling and Training for Vis ito rs There v. as a stalisticajly significa nt differ<.-"llce in per oei''cd benefi ts of each level of handling and training for big cats, x 2(3) = 98.06, P < ~1edian (lqr) percei.,-ed bc.nefils for ha nd.s-on contact, protected contact, ha nds. off contact, and training v.-ere 3.0 ( ), 3.1 ( ). 2.6 ( ). and4.0( ), respt.-ctivcly. Post hoc com parisons revealed significa nt d i.lterenoe.s betv.-een the per o:i\-ed benefits o f tra ining and hands-o n er = 11, p < 0.001, r = ), protected (1' = 14, P < 0.001, r = ). and hands-off contact ( r = 4, P < 0.001, r = ) such that lr4jning was perocived a.s more benefi ciaj than any o f these practices. Further, protected contact wa.s pen:cl\ ed as significan tly more bc.neficiaj than both hands--0n oon tat..1. (1' = 23, P < 0.001, r = ) and hands-o ff oonta t..1. (1' = 10, P < 0.001, r = ); and hands-on contact wa.s perocived as signjficantjy more beneficial than hands--00' oon tat..1. ( r = 27, P = 0.003, r = ). Additio1111I bt.nefits and lin1itations As shown in Table 5, a number of comments were ma de by par tic ipan ts about the benefits a nd limitations of the three levels of contact a nd training for visito~ with partjcipanls no ting that benefits a nd limitations depend on the m-erajj educationa l program of a facijity a nd the way informatjon is communica ted lo vis.itors, the ind i vid ua l ha ndler and the training used, whethc.-r the handjing/ tra in ing is conducted on - or off-exhibit, hov. safe the program is, a nd tha t no one type: has an impact on a \ isitor and his/ her learning. lt v. as also no ted tha t any 337

356 Zookeeper-Sig Cat Interactions 147 TABLE-'1. AddJtiooal btdeflts a1..:i U1nitatloos of ha1..:lllng and trahliog for ketpers Hands-on Protected Hands-off Training Benefits lntpt'o\'cs keeper's knowledge of aninlal behavior (n = I) Stin1u)ating fo r kttpcr (n = I) Allo"''S strong connection between kttpcr and animal (n = I) Ssfer(n = 12) Allo"''S keeper to bond with animal (n = 6) Allo"''S keeper to provide. better health/mcdieal care.(n = 4) Challenging/stin1ulating fo r keeper (n = 2) Interesting/engaging for keeper (n = I) lntpt'o\'cs keeper well-being (n = I) Teachei> keeper to communicste with anin1al ( = I) Lesssuessful for keeper (n = I) Can be practiced by most keepers (n = I) Ssfe(n = 4) Still able to achieve things with this approach (n = 2) Does not impact job satisfaction (n = I) Strengthened relationship with animal and sub~quent positive. impact on work (n = 5) Rewarding and ghcs kttper pride in work ( = 3) Allo"''S keeper to provide. better health/n-.cdical care.(n = 3) Ssfe (n = I) Results in happier/better keepers (n = I) U nncoessary (n = 5) Limitation.i Dangerous (n = 3) Potentially puti pressure on kttpers (n = I) Potentially dangerous (n = 2) ~iay lead to job dissatisfaction/ frustration ( = I) Poorer pl'0\ 1 ision of heath/medical care(n = I) Unsafe (n = 2) Some keepers may be unsuitable to practice it (n= I) ~iore about meeting keeper needs than animal needs (n = I) TABLE- S. AddJrlooal bt1wflts a1..:i U1nitatloos of ha1..:lllng and trahilng for vliltors Hands-on Protected Hands-off Training Benefits ~iore. enjoyab!e. cx perienoc (n = 4) lmpro,cs opinion or animals (n = 2) Potential to increasecducstion and conservation suppon (rt = 2) Sends more accurate n-.cssagc about animal (n = 1) High education value (n = 6) Allows visitor to connect with animal (rt = ) Safer (n = 3) Interesting for visitors (n = 3) ~iore. natural (n = 2) ~iaycnoourage financial suppon (n = I) Sends more accurate n-.cssagc about animal (n = 6) Stting animals in this fo rn1 should be. sufficient for visitors (n = I) Sends accurate message (n = 3) Educational (n = 3) ~iaycnoourage financial suppon (rt= I) Encourages connection between visitor and animal (n = I) Limitations Sends wrong mei>sage. about dangerousness of animal and encourages pct ownership (rt= 39) Unnecessary (n = 6) Dangerous (n = 2) Could result in fest if a visitor has a negative experience (n = I) Sends inaccurate. message. about animals (n = 4) ~ot as beneficial as hands-on (n = I) Possible discase trans.rnis.iion from animal to visitor (n- 1) Often no opponunit)' for keeper talks (n = 2) Animal not as interesting (n = 2) ~o opportunity fo r connection berwttn visitor and animal (rt = I) Sends inaccurate. mcssage.(n = 2) 7.tJO mot' ' 338

357 148 Szokalski et al. k.-vcl of contact sho uld oojy be practiced by )a:epe~ not \1isitors. DISCUSSION A.lthough protected contact is the most popular handling method used with big cats a mong this sam pie, some keepers employ a combinatjon of two or more methods, and the majority use training. Partjcipants v.urk with a range of big c.at tax.a, ofleo more tha n one, most eommoojy \\'ith ti.gers and lions. Cheetahs, leopards, eougars, and soov. Jcopards a re v.urked...,ith a t a moder ate k.-vcl, and fc\\'t.-"f kt.'cpers work \\'ith jaguars. Prott.'Cled eontact is used most frequently with all species, \\'hereas the use of ha nds-on and hands off eontact \.<lries bet wcicn spt.-cics. No tably,over 40% of keepers who work v. ith ehceta hs practjce hands on contact v. ith them, compared to a maximum of 11% for kcicpers \\'-Orking with the other spi.-cics; and hands off coo tact is u.sed more...,ith leopards and jagua rs. T his may be attributed to the nature of each indjvidual species, such as the cheeta h generally being Jess eonfrontatiooa l tha n the lion, for ex.ample. Simila rly, ha lf of ajj keepers practice tr4ining v. ith cheeta hs, leopards, jaguars, and cougars, and more (60-ifP/o) with tigers, Ii oos, and snow leopards. Further, respondents spend more tjme eogagoo in hands<-0n and protected contact com pared to trjjning. This is not unexpected, since trjjning is a more forma l proocdun; ljkely to ooeur for set periods of time, v. hert.-as coo tact oojy may occur more frequentjy. \Vhile specific to the sample studied, these findings pro \1ide a starting point for understanding the current global use of these prjctjces. Reflot.."'ling its high u.sc among this sample, prott.'cled eontact wjs percci\'cd as the most beneficia l ha ndling method for all three groups (big c.ats, kt.'cpers, and visi tors), both \\'hen participants \\'t.-'1'c asked outright a nd in our benefit r<ttiog scajc. Fe\\. pa rticipa nts (around 1<Y>/o) rt.-ported hands<-00 to be most beneficial for ajj three groups wht.-'ll asked outright; and few belit.--..."cd hands o ff to be most beneficial for big cats (S /o) and keepers ('Pio), whereas around 45 /o belit.--..."cd it to be best for visitors. Altho ugh ha nd,s..oo contact rt.'ceived a higher rating than hands off on our r4tiog sc.ale for a lj three groups, this is sotm"\\'hat misleading, since median scores for ha nds,. on \\'t.-'1'c 3 (of S) for each, rt.-preseotiog "neutral/ unsure." T hus, while hands off is not vie\\.'cd as being beneficial to any group based on the scaje responses, it does not neccs- sarily mean that hands o n is pt.-"toei""cd any more fa\'orably (based on this informa tion alone). The open ended questions hjghjighted potentjaj ad ditional benefits and limitatio ns of each tt.'chnique. Pro tccted contact received high bt.-"llcfit ratjngs and primarily positi\ ecomments for eaeh of the three groups. Similar to previous research (Baker, 2004; Hosey and Melfi, 2010), it wjs considered safe for big cat~ pro.,id ing them v. ith better health care. as well as reduc ing ncgati""c behavior, incrt.-asiog choice, ajjowing animajs to bond with keepe~ and improving cogniti\'c stjmujation. Like\\. isc, for kocp e~ protected contact \\'as considered safe, a llowing them to develop a bond...,i th their an imals, a nd improving the provision of heajth/ medic.al care; again, consistent with prior research [Bayne, 2002; Hose.y and Melfi, 2010). Pro tectcd eontact w4s belie\'00 to send accu.r<tle messages about animals to zoo visitors a nd to have a high cduca tional \'aluc, a lso reflectjve of the q uite c:xtensi"-c litera lure suggesting that visitors enjcy seeing acti""c animals [f\1argulis et a l., 2003) and that interactj\'e programs successfully increase visitor attitudes, knowledge, and eon sen 111ioo beha\'ior [UalJa ntyne and Packer, 200S; Uroad and \Viek.-"T, 1998; K reger and ~1eoeh, I 99S; L..indt..-"tna on ~1a uhies and Kame r, 2006; J>o\'C.Y and R ios, 2005; Ross and G illespie, 2009; S\\ anagan, 2000; Visscher et aj., 2009; Zcppel, 2008; Zeppel and 1'.1ulo in, 2008}. Fe\\, oomments were made about limitations of protected coo tact for any group. These findings strengthen the results from the ben efit rating scale, while providjng addjtionaj insight into the potentjaj positive outcomes of this approach. Responses to our rating scale showed tha t kt.'cpers were somev. hat ambiv.denl about the benefits of hands on contact. The majority of open ended comments suggest that participants typica ljy belie\ e thjs practioe is no t a pus iti\ eone, with comments reflecting how dangerous handson coo tact can be, both for anima ls a nd kt.'cpt.-"ts; one re spondeotexpla ined,"... if aatt docs a ttack a keeper, that anima l \\'ill often end up shot. This is not fa ir to the cats as they arc acting on naturaj instincts a nd should no t ha\'e been put in the position to be able to attack anyo ne." De spite many participants raising this issue and the strong sentiment with whic h it v. as oommuniatted, none men tioned spt.'cific incidents where this oocurred. Regardles~ big cat a ttacks do oocur in captjvity, often with serious consequences (Chapenoire et aj., 2001; Cohle et a L, 1990; l lejna, 20 IO; Lazarus e I aj., 200 I; Loe and Ros.kaf t, 2004; ~1urphy et a L, 2001; Nyhus et al., 2003; SchiJJer e t al., 2007J. Respondents also belie\ ed ha od,s..oo eontact negati\>ely influences the naturaj behavior of big cats, and is unnecessary since protected oontat..-"1 can yield similar ben efits (e.g., a close keept.-"t animal bond) without the added risk. This supports Carlstead's (2009), finding that rela tionships betv.-een animals and kt.'cpers are more favor able when a keeper inter<tcts through a bar rier as opposed to entering the an imal's e nclosure. Partjcipants most fre q uently noted tha t ha od,s..oo coo tact sends the wrong message to visitors, implying that it is safe to ioterat..-"1 closely with big c.ats, tht..-'1'cby polt..-"lltia lly eneouraging pet own en.hip; as one participant explained: "\Vhat the public walks aw<ty from is a s.kt.."\\'cd image of what these cats are and do. T hey sec traint..-"ts touch them and then they think tha t all cats can be touched. They then "\\ ant one" with out (sic) thought to the cost~ care and beuerment of the spt.-cics... Of note, most of our respondents work in the USA, v. here it is JegaJ in many states to kt.'cp exotic 339

358 Zookeeper-Sig Cat Interactions 149 felines a.s pc-ls, au:ouoliog for Lhe lar,i,>e numbet of parlic i paols expressiog this \'it.."\\'. These consequences have been prc\'iously discussed in the liler<t.ture (Kreger and r-.1eoch, i 99S; Nyhus et aj., 2003i Despite thcsecona:rns, a sma ll oumber of respondt..'1lls expressed tha L hands-on contact allo-...-s a strong keeper-big cat bond, assists -... ith medical care, and pro\'ides an enjoyable cxpe:rienoe for \'isitors. ~1ore research i.s nt.-cdcd to explore these benefi Ls. \Vhile rc:spondeots rated hands-off contact a.s no t beneficial for a u three groups, the opeo-ended responses ""'ere mixed. For big cat~ this approach v. as considered limiting, leading to negati\'e beha\'iors (e.g., stercotypy), v.ith k<.-cpcrs being a source of stress for animals under such cirewnstanccs. O ne respondent explained, "l fa\' ing an animaj which is ma naged 'hands off in c.apti\'ity and in constant close proximity o f kec..-pers/visitors can ca use aggressive, defc.. "tlsive and stressed animals." Further, participants highlighted tha L hands-off contacl does oo t.stre ogtheo bond.s or improve health/medical care. Jn cont.rjsi, despite hands-off being reported as not bc:oeficial for k<.-cpc..ts a nd \'isiton> on the rating scales., the open-ended responses actually raised mo re benefits than limita tions (albeit the O\'t:r<t.U numbc..-r of comments wjs \'t:fy low compared to the othe r practioe.s). For keepers, participaols discussed it as safe and sending an au:urate message to visitors about \\'ha L big ca Ls are reajly like. T hese finding$ suggest that v. hat may be bc.. '1lefic iaj for one group is 001 nt.'cessarily so for a oo ther, and zoos must be aware of this. Almosl all participants agreed that training is beneficia l for big cars and keepers, \\'ith 88 /o agreeiog that il is for visitors. T his agm.."ment wjs higher than that for any of the thr<.'c handliog practioes. Rating scale and open-ended responses.substantiated these findings. For big c.ats, respondeols sa\1: trainiog as beoeficia l to \'Cl erinary/ medic.al proocd ur~ and a.s.stimulating and e n riching, simila r to discussions aboul.such practices with othe r species (Hose.y a nd ~1elfi, 2010; Lambeth et al., 2006; l a ule and Desmond, 1998; LauJe et al., 2003; ~1artin, 2004; McKinley et al, 2003; Phillips et a l., 1998; Prcsootl a nd Bucha nan-smith, 2003). Consistent with findings in primate research [fiayne, 2002; Sa\'astano et al, 2003}. participants highlighled that traioing sigoific.antjy strengthens the keeper-a nimaj bond, positi\'ely impacting a keeper's work. O ne participant highlighted this ""'CU: "O noe you form that bond \\'ith the anima l you are lr4ining, you arc inspired to do more for the anima l bec.ause it is importan t to you therefore you clean belier, yo u give more enrichmeol, you get in\'oh-ed in coosct, ation, you learn to enjoy talking to the public about yo ur a oimajs..." Finally, like \\'ith protcx.1.cd contact, participaols primarily discussed how tr<i.ioing sends an accurate message to w o \'isitors aboul the animals and the importance o f caring for them, and that il is therefore also educational. Minimal negatjve commeols \\"etc made about 1hjs practice. These findings add support to the rat- ing scales and pro\'ide additional informa1jon aboul the lx..-oefits of training. lo this.study, handling and trjjning were treated a.s separjte entities whereas, io rea lity, trainiog musl be accompanied by some level of handling (no t a lw<i.ys \'ioe versa). Our fioding$ a llowed us to tease a par L these eff(.'cts, re\ <.~ling that for all thr~ groups (animals, kec..i:iers, a nd visitors). traioing is viewed as most beoeficia L R>r big cats, this is not surprising, given the potentiaj eor ichmeol value of tr4ining (Laule and Desmond, 1998; Laule et aj., 2003; Prcscoll and Buchanan-Smith, 2003), a od its.struc tured nature a ljowing aspects of anima l well-being,.such as cogniti\ e stimujation, to be addressed. l o contrast, coo tact only ca o be less structured and may not be targeted.specific.ally to impro''c a oimal well being. Respondents highlighted tha L the benefits a nd limitations of each practice depend on \ arious othe r factors. Fo r big ca Ls, individ ua l factors (gender, aj,>e, spt.'cies, histor)', and personality) inouencc ho\\ much each animaj lx..-oefits from these practices. T his is consistent v. ith Carl.stead (2009] who highjighled tha t species differe ooc:sooeur in thecomponeots that contribute to the oa ture of interactions belwceo animals and their kec..-pers, and v.ith an animal's affinity to keeper. Similar ly, partic ipan ls discussed how indi\'idual keepc..-r factors \\'ill de.termine how much they benefil, v. ith some suggesting that not all keepers are.suited to practice each ha odjing method a nd/or traioing. l o a study of a hands-on facility, Phillips and P<.'Ck (2007] fo und that ki.."(..i:ier persooa lity influences inter<i.ctions be Lween keepc:n and tigers. f inally, participants noted that the over.di educatiooa l a pproach of a facility impacts the expc..-rience of zoo visitors. Lik<.."\\'ise, prior research ha.s indica ted that individ ua l factors ca o ajso io.flueoce how visitors experieoce and respond to zoos (e.g., see Dallan Lyne and P.<t.ckcr, 200S; Falk and.l\delma o, 2003; Falk et a l., 2007; r-.i organ a nd Hodgkinson, 1999]. WhiJe direct conclusioo.s c.anool a lways be dr<i.v. o belweeo a particular zoo pr<t.t..1.icc a od its consequences, it is important to coosider these vjriable.s when inter-preting o ur results a nd \\'hen addressing the experieoce of a nima ls, keepers a od visitors. T hjs study ha.s plaa:d a hea\'ier focus on identifying the benefi Ls of different ha nd ling and tr<i.ioing prat..1.ices. f.i0\.\'<..""1:t, a number of limita tions have also bocn raised by keepers and, in the interest of promo tiog optima l wellbeiog for c.aptive animal~ these are of concern. For ex.ample, a hands-on a pproach is tho ught to influence naturaj cat behavior and ri:mo''c choioe and control from the animal. Similarly, a hands-off a pproach has bceo tho ught to lead to negati\ e beha\'iors and reduced health care. Each of these factors is associated \\'ith reduced welfare. It is \\'idely recognized that \'arious fat..1.ors in the ca ptive environment cao lead to diminished ""'C-IJ being, and huma o interaction is no exception. Little is understood about how huma o.s might negatjvely intlut.. '1lce big cat welfare, \\'ith only a smaji oumber of studies offeriog coooicting 7.JJO Biofog)' 340

359 150 Szokalski et al. fiodjng;s aboul lhe influence of unfamiliar humao.s (zoo \1isiton) [~1a1Ja p ur and Chellam, 2002; ~1a rgulis e l al, 2003; O'Donovan et al., 1993; Sellinger and Ha, 2005), and nolhing documenled aboul the influenoc: of famil iar humao.s (zook<.-cpen). O ur fiodjng;s have pro.,ided a sta rljng poin l for future research in this regard. The partic ipants in o ur study ha"-e provided a s.ignifican l nt.."'a' understanding about the possible consequenocs of different handling practjces a od traioing v,:ith big alls. A.s a research area in its iofancy, there is a clear need for the benefits and ljmita tioo.s raised here to be substan tiated. As ha.s been done for prima tes, behavioral, and physiologica l data are required to k nd support to the notioo that protected contat..1- and training are beneficial for big cats. And, while our sample size \\ as sufficient for this initial investigation, a large r number of re.spondents, with a more e\.-en.spread using each practice, could pro \1ide a better uodersta odjng. G iven that this sun'<:.y relied oo self report measure~ some obj<.'ctivc methods to in\ es. tjgate the benefils and limitations to keepers v.uuld add W'Cight to these findings. F'inaJJy, independent measuremenl of the impat..1- of these different methods o n zoo vi.sit ors is also required, and could be achje\'cd through ol> sen ational and survey research of vis.it ors who encounter these differeol pr4ctice:s. 0\'t:r.tU, zookeepers ha\>e pro \1ided \1aluable informatjon that v.ill help direct and compliment future research on thj.s sotm"'a hat cootroversial topic. CONCLUSIONS I. Based on this sample, protected contact i:s the mosl commonly used handling method \\ ith big cats and tr4jning i:s used more ofk-"ll than not. 2. Partic ipants rated protected contact a.s more benefic ial tha o hands --0n and ha od.s-off coo tact for big cats, kcc..-pers, a nd visitors; perceived beoefit.s include tha l il allows a bond to form between keepc.-"t a od a ojmal, and has hjgher educatiooa l bc:oefit.s for zoo \ i:sitors. 3. Partic ipants were quite ambiv41ent when rating ho\\' beneficial ha ods--00 contat..1- is, bul provided numerous comment.s aboul ils pcrocivi:d limitatjons, particularly that itca o bedaogc.-"tous for cats and k<.'cpt.-'ts, and sends the \\ rong mcssaj,>-e to zoo \1isiton. 4. While partjcipants rated hands--0ff contact as ool beneficial for any of the three group~ additional commeols suggested il could lead to negali\ e behaviors in big ca ls, but is safer for keepers a od may send accurate messages to zoo visitors. S. Trainiog was r4tedas being more beneficial for all three partjes than any of the handling method~ \\'ith respondents believing il to be of par lk.--ular benefi t to the health ca re of big utls, a ljowing a strong bond bet wcicn animal a nd k<.-cper, a od sending a positi\'c message to zoo \ i:sitors. 6. Zookeepers a re \\'<:JI positioned to provide information a bout the use of these d ijtere.nl handling practioc:.s and trainiog, and ha""c provided a base for future, objective e:tplorations of these pr<tcticcs. ACKNOWLEDGMENTS \Ve acknowled.!,>-e the assista oce of the \1arious zoo kocper orgaojzatjons (.l\ustr41asia o Society of Zoo Koc p ing, Ame rican Associatjon of Zoo Keepc~.l\ssociation of British a od Irish Wild A ojmal Keepers, Zoo Dio l ogy, Auslr41asian Zoo Keeping, a od Asociacion l berica de Cuidadores de Animalcs Sal\'ajes), and w os/y.ildlife parks (Zoos South Austr<tlja, Zoos Ytctoria, Ta ronga Zoo, Cairns \Vildlife Safari Reserve, \Vetlington Zoo, Aucklaod Zoo, a od llamiltoo Zoo) iovoh cd io distribut ing our questionnajre to their membc.-'1"1> and staff. \\~also v.ish to tha nk a ll of the respoodent.s \\'ho dcdjcated their time to answeriog o ur q uestionnaire. REFERENCES IJaln KC lkr.. d io C1f pusiti''i: ht.:n:cn intc:u :t3on fm w 1..ially ho ur-d cbirnpllr.::t.«$. Acir.i \\l:lfiw.: 13:ll IJalfar.tyne R. r 11eker I t':'omi>t:.ng er.,i roomenta lly )~~11inabk 11ttitudr:l and bd111viou.r tl-.rough lh.-e o:l-.c1:.x k :u-ning i.:1qio:f.cno:ec:.,,.tud is the Slllk of llu: p.mc? Environ EdllC Ri:s 11: IJa~tt L. Uuo:b11r..:tn Sr.1ith llm Efrccts t>f p~dio:lllh~ily CICl the v.~lfaro: u( o:apti\~ ranim11b. Appl Anim lkltav Sd 102:21J 245. IJaynr K Oo:~li>pmcnt ul the hu.man n:)l.'.11n:h 11nir.1:tl bood 11nd its :.tltpuct un ralim.:11 wcll briiq;. IL:\ R I 43 :A 9.!kin r Cuug11: att11o:ki1 on hum.:mi1 ~n tlte Un!tcd Stllll:ll 11nd Car.11d11. \Vildlifc Sao: n 19:<I IJ:u11d S. \\'ickr n Capti\~ anir.iab and intcrprct:j.t3on: 11 tak uf two tiger cxhib:t11. I Tot..'Ti~ Sttld:Ci19: Clrli1tcad K A cor.1pua1ivc approach lu lite study Clf lc.:pn 11r:ir.1aJ rcblioo,;hips ~ ti-~ :WU. Zoo UiuJ 28: Clrr.1.."lt\'.I L,.Colcll M. C.'ah u M, Abclto?\{T. Velasco M. Poiada S lkndits ul trair:ir..g/pfa)i iq; llu:n r-r in a group of capti~ lowland gurillu (Gori!laguril{a pill:!). Anitn \\'cl fax 18!9 19. Chapcno:x S. C111:1iadc B. Lcg.."l:!s M IJa.'Cio: in.'ltinct in 11 fdir...:, Am J l'urcn ~1cd l"j.th 22:46-SO. 0 11xtun A~t The potential oft!-...: hum11n-anin:cl ~b.t3onitlip all 11n en,i roor.iental cnricltmer.t for tlte... 1:1rax Clf wt>-huu.'ll:d :mim.:1111. AppJ Anim lklta\ S1..i l l3:1 10. Cohk SD. H11rb.n <..-V.'. li:u-l11n G n.tal big 0:111 ratta..11.s. Am I Fu~n Med r 11tlt 11 :20S 212. Cob.han Ii, IJ:i:W C l'dr.1a1c t..-:j.ir:ir_,g 111 Di,;r)C') ii Anirn:tl Kiiq; dc1m. J Appl Anim \\'clfa.-i: Sd 6: n.llt H I. Adi:tman L~ fr.,'l:sti.g11tiiq; llu: impuct Clf prior knuv. I ~gi.: rand inti:xsl on aq1.::ubr.1,;~tor k:ur:ir_,g. J RCll Sc.i Teach 40; n.llt Jli. Ri:inhard E?\{. \'croon CL. Uroonenkra.nt K. Del.me NL. Heim!:ch IE \Vhy ioos & aqi:11riun1~ matter: U)l.'.i;iciiq; lite :n1pw:t uf 11 xoo ' i sit. S~vn Spri~?\{0: A~iation Clf ZO<H: & Aqua7iumll. Ft.:lu-m11n NE. Ladewig Ii Clt:u-11eti:ristia uf ralim:lls 'J~d in :wu ir.ti::pxt11lioo: ra ~)Tilhc:cis uf m1e1rd1. J l.nccrpret Ri:~ 13: l-ltjna P. 2GIO. A fatal k upwcl attack.. I Foo:nsio: Sc.i SS: !-loscy G. Mel.Ii V Hu.m.:ln animal bonds bctw1ien ioo profa,;san 11l:111r..cl the 11.nim11l:c in tbcir c::t:1:, Zot> O:ol 29:1 14. l-loy JM. Murr"y PJ. 1'r.llo: A Thiny yc11n b tcr. er.ddt.nu:r.t prn1--ic.1:s for ca.p(i''i: mammab. Zuu o:ol 29: Kirk wooo JK \~!fare. h~tiband!y and vcti:rinra.ry C'.l.;"l: Cl(.,..:Id 11r:ir.1ab in C'.tplh~ ty: chaft8.:ll in rattit00..:11. prog.'l.'.ss in lncw. k dt.oe and tto:bnxjuc$. Int Zoo \l::ubl: 28: Kfl:gi.:: MD. Mcr.. d 1 IA \'i:citm 11.nim11I ~nler11tlimts at ll-...: xoo. Anllu-ui0058:14J

360 342

361 Appendix G Online questionnaire used in Chapter Four 343

362 Dear participant. Thank you for your interest in this study. It involves completing this online questionnaire, which is both voluntary and anonymous. Returning the questionnaire indicates your consent. Once analysed, the data will be written up as part of my PhD thesis and will be submitted for publication in a relevant academic journal (e.g., Zoo Biology). The data will only be accessed by myself and my supervisors. The information collected as part of the current study will be retained for five years, stored in a secure place in the School of Psychology office, Magill Campus, at the University of South Australia. The researcher will take every care to remove responses from any identifying material as early as possible. Likewise individuals' responses will be kept confidential by the researcher and not be identified in the reporting of the research. However the researcher cannot guarantee the confidentiality or anonymity of material transferred by or the internet. SECTION 1. THIS SECTION ASKS FOR SOME DETAILS ABOUT YOURSELF AND YOUR PAST/CURRENT WORK EXPERIENCE. 1. Gender - Please select one - 2. Age - Please select one - 3. Which country do you currently work in? (Enter text into this box, maximum 2000 characters) 4. What type of institution do you currently work in? (E.g., Public zoo/private zoo/wildlife sanctuary/etc) (Enter text into this box, maximum 2000 characters) 5. Which species of big cat/s do you currently work with (you can select more than one): Tiger Lion Cheetah Leopard 344

363 Jaguar Cougar Snow leopard 6. How long have you been working with big cats at your current institution? (Enter text into this box, maximum 2000 characters) 7. If you have worked with big cats prior to your current position, approximately how long did you spend doing so? (Enter text into this box, maximum 2000 characters) 8. What is your current job title (E.g., Manager/Curatory/Keeper/etc)? (Enter text into this box, maximum 2000 characters) 9. Please rate how strongly you like working in your current job: - Please select one Zoos and wildlife parks differ in the level of contact that occurs between keepers and big cats. Please read the following definitions and select which one best describes the nature of the interactions between yourself and the big cats with which you currently work (you can select more than one, since different practices might be used with different species within the same institution): Hands-on contact: Physical contact between keepers and big cats, with no protective barrier (for example, entering big cat enclosure with the animals present or taking big cats for walks) 345

364 Protected contact: Contact between keepers and big cats with a protective barrier between the two (for example, interacting with big cats through a mesh fence) Hands-off contact: No contact (physical or protected) between keepers and big cats 11. Based on the definition above, please select which big cat species you currently practice hands-on contact with (if any): Tiger Lion Cheetah Leopard Jaguar Cougar Snow leopard 12. Based on the definition above, please select which big cat species you currently practice protected contact with (if any): Tiger Lion Cheetah Leopard Jaguar Cougar Snow leopard 13. Based on the definition above, please select which big cat species you currently practice hands-off contact with (if any): Tiger Lion Cheetah Leopard Jaguar 346

365 Cougar Snow leopard 14. If you engage in hands-on contact with the big cats, approximately how many hours in a working day do you spend interacting with them? (Enter text into this box, maximum 2000 characters) 15. If you engage in protected contact with the big cats, approximately how many hours in a working day do you spend interacting with them? (Enter text into this box, maximum 2000 characters) 16. Individual institutions also differ in the level of training that occurs between keepers and big cats. For the purpose of this study, training refers to any planned and targeted procedure whereby a zoo keeper encourages the performance of specific behaviours in big cats (for example, positive reinforcement training). Based on this definition, do you participate in training with big cats in your current role? Yes (please proceed to Question 17) No (please proceed to Question 19) 17. Based on the definition above, please select which big cat species you currently practice training with: Tiger Lion Cheetah Leopard Jaguar Cougar Snow leopard 347

366 18. If you engage in training with the big cats, approximately how many hours in a working day do you spend training them? (Enter text into this box, maximum 2000 characters) 19. If you have worked in the zoo/animal industry prior to your current position, approximately how many years did you spend doing so? (Enter text into this box, maximum 2000 characters) 20. If you have worked in the zoo/animal industry prior to your current position, please list which species you have worked with: (Enter text into this box, maximum 2000 characters) 21. If you have worked in the zoo/animal industry prior to your current position, please select which of the following practices you have engaged in with animals in the past(can be more than one): Hands-on contact Protected contact Hands-off contact Training SECTION 2. THIS SECTION ASKS FOR SOME DETAILS ABOUT THE BIG CATS THAT YOU CURRENTLY WORK WITH. 22. If you currently practice hands-on contact with any of your big cats, on average how would you describe the behaviour of the animals involved (you can select more than one if appropriate): The animals seek interaction 348

367 The animals are cooperative The animals don't seek interaction The animals are hostile toward keepers/zoo staff 23. If you currently practice protected contact with any of your big cats, on average how would you describe the behaviour of the animals involved (you can select more than one if appropriate): The animals seek interaction The animals are cooperative The animals don't seek interaction The animals are hostile towards keepers/zoo staff 24. If you currently practice hands-off contact with any of your big cats, on average how would you describe the behaviour of the animals involved (you can select more than one if appropriate): The animals seek interaction The animals are cooperative The animals don't seek interaction The animals are hostile toward keepers/zoo staff 25. If you currently practice training with any of your big cats, on average how would you describe the behaviour of the animals involved (you can select more than one if appropriate): The animals seek interaction The animals are cooperative The animals don't seek interaction The animals are hostile toward keepers/zoos staff 26. Do the big cats at your institution display any 'abnormal' behaviours, such as pacing, furplucking, excessive grooming, etc? Yes (please proceed to Question 27) No (please proceed to Section 3) 349

368 27. If you answered 'yes' to the previous question, approximately how many hours per day do the big cats at your institution spend engaging in such behaviours? (Enter text into this box, maximum 2000 characters) SECTION 3. THIS SECTION REFERS TO YOUR OPINIONS OF DIFFERENT BIG CAT HANDLING AND TRAINING PRACTICES. 28. Zoos and wildlife parks are faced with the triple challenge of simultaneously meeting the needs of their animals, keepers, and visitors. How strongly do you agree that the current handling and training practices used with big cats at your institution allow the needs of these three groups to be met? - Please select one Please rank the order in which the three groups below are important to you: Big cats 1 Keepers 1 Visitors Please select which type of handling practice you believe is best for big cats Hands-on contact Protected contact Hands-off contact 31. Please select which type of handling practice you believe is best for keepers Hands-on Protected contact Hands-off 32. Please select which type of handling practice you believe is best for visitors 350

369 Hands-on contact Protected contact Hands-off contact 33. Do you think training is beneficial to big cats? Yes No 34. Do you think training is beneficial to keepers? Yes No 35. Do you think training is beneficial to visitors? Yes No SECTION 4. THIS SECTION RELATES TO YOUR OPINIONS OF HANDS-ON CONTACT WITH BIG CATS. Sub-section 4A - For each of the following questions, please rate how strongly you agree that HANDS-ON CONTACT contributes to the following benefits for BIG CATS: 36. Improved physical well-being - Please select one Improved psychological well-being - Please select one Improved cognitive stimulation - Please select one - 351

370 39. Improved social stimulation - Please select one Improved behavioural stimulation - Please select one If you believe there are other positive or negative consequences for big cats as a result of hands-on contact, please describe them below: (Enter text into this box, maximum 2000 characters) Sub-section 4B - For each of the following questions, please rate how strongly you agree that HANDS-ON CONTACT contributes to the following benefits for KEEPERS: 42. Improved relationship with big cats - Please select one Improved movement of big cats (e.g. between enclosures) - Please select one Improved enclosure maintenance - Please select one Improved job satisfaction - Please select one Improved care of big cats (e.g., veterinary checks) - Please select one - 352

371 47. If you believe there are other positive or negative consequences for keepers as a result of hands-on contact, please describe them below: (Enter text into this box, maximum 2000 characters) Sub-section 4C - For each of the following questions, please rate how strongly you agree that HANDS-ON CONTACT contributes to the following benefits for VISITORS: 48. Improved entertainment - Please select one Improved general knowledge of big cats - Please select one Improved enjoyment - Please select one Improved 'connection' with big cats - Please select one Improved information about big cat conservation methods - Please select one If you believe there are other positive or negative consequences for visitors as a result of hands-on contact, please describe them below: (Enter text into this box, maximum 2000 characters) 353

372 SECTION 5. THISE SECTION RELATES TO YOUR OPINIONS OF PROTECTED CONTACT WITH BIG CATS. Sub-section 5A - For each of the following questions, please rate how strongly you agree that PROTECTED CONTACT contributes to the following benefits for BIG CATS: 54. Improved physical well-being - Please select one Improved psychological well-being - Please select one Improved cognitive stimulation - Please select one Improved social stimulation - Please select one Improved behavioural stimulation - Please select one If you believe there are other positive or negative consequences for big cats as a result of protected-contact, please describe them below: (Enter text into this box, maximum 2000 characters) Sub-section 5B - For each of the following questions, please rate how strongly you agree that PROTECTED CONTACT contributes to the following benefits for KEEPERS: 60. Improved relationship with big cats - Please select one - 354

373 61. Improved movement of big cats (e.g. between enclosures) - Please select one Improved enclosure maintenance - Please select one Improved job satisfaction - Please select one Improved care of big cats (e.g., veterinary checks) - Please select one If you believe there are other positive or negative consequences for keepers as a result of protected-contact, please describe them below: (Enter text into this box, maximum 2000 characters) Sub-section 5C - For each of the following questions, please rate how strongly you agree that PROTECTED CONTACT contributes to the following benefits for VISITORS: 66. Improved entertainment - Please select one Improved general knowledge of big cats - Please select one Improved enjoyment 355

374 - Please select one Improved 'connection' with big cats - Please select one Improved information about big cat conservation methods - Please select one If you believe there are other positive or negative consequences for visitors as a result of protected-contact, please describe them below: (Enter text into this box, maximum 2000 characters) SECTION 6. THIS SECTION RELATES TO YOUR OPINIONS OF HANDS-OFF CONTACT WITH BIG CATS. Sub-section 6A - For each of the following questions, please rate how strongly you agree that HANDS-OFF CONTACT contributes to the following benefits for BIG CATS: 72. Improved physical well-being - Please select one Improved psychological well-being - Please select one Improved cognitive stimulation - Please select one Improved social stimulation - Please select one - 356

375 76. Improved behavioural stimulation - Please select one If you believe there are other positive or negative consequences for big cats as a result of hands-off contact, please describe them below: (Enter text into this box, maximum 2000 characters) Sub-section 6B - For each of the following questions, please rate how strongly you agree that HANDS-OFF CONTACT contributes to the following benefits for KEEPERS: 78. Improved relationship with big cats - Please select one Improved movement of big cats (e.g. between enclosures) - Please select one Improved enclosure maintenance - Please select one Improved job satisfaction - Please select one Improved care of big cats (e.g., veterinary checks) - Please select one If you believe there are other positive or negative consequences for keepers as a result of hands-off contact, please describe them below: 357

376 (Enter text into this box, maximum 2000 characters) Sub-section 6C - For each of the following questions, please rate how strongly you agree that HANDS-OFF CONTACT contributes to the following benefits for VISITORS: 84. Improved entertainment - Please select one Improved general knowledge of big cats - Please select one Improved enjoyment - Please select one Improved 'connection' with big cats - Please select one Improved information about big cat conservation methods - Please select one If you believe there are other positive or negative consequences for visitors as a result of hands-off contact, please describe them below: (Enter text into this box, maximum 2000 characters) SECTION 7. THIS SECTION RELATES TO YOUR OPINIONS OF TRAINING WITH BIG CATS. 358

377 Sub-section 7A - For each of the following questions, please rate how strongly you agree that TRAINING contributes to the following benefits for BIG CATS: 90. Improved physical well-being - Please select one Improved psychological well-being - Please select one Improved cognitive stimulation - Please select one Improved social stimulation - Please select one Improved behavioural stimulation - Please select one If you believe there are other positive or negative consequences for big cats as a result of training, please describe them below: (Enter text into this box, maximum 2000 characters) Sub-section 7B - For each of the following questions, please rate how strongly you agree that TRAINING contributes to the following benefits for KEEPERS: 96. Improved relationship with big cats - Please select one Improved movement of big cats (e.g. between enclosures) 359

378 - Please select one Improved enclosure maintenance - Please select one Improved job satisfaction - Please select one Improved care of big cats (e.g., veterinary checks) - Please select one If you believe there are other positive or negative consequences for keepers as a result of training, please describe them below: (Enter text into this box, maximum 2000 characters) Sub-section 7C - For each of the following questions, please rate how strongly you agree that TRAINING contributes to the following benefits for VISITORS: 102. Improved entertainment - Please select one Improved general knowledge of big cats - Please select one Improved enjoyment - Please select one - 360

379 105. Improved 'connection' with big cats - Please select one Improved information about big cat conservation methods - Please select one If you believe there are other positive or negative consequences for visitors as a result of training, please describe them below: (Enter text into this box, maximum 2000 characters) SECTION 8. THIS SECTION REFERS TO YOUR OPINIONS OF BIG CATS How strongly do you feel about big cats? - Please select one How important do you think big cats are? - Please select one How concerned are you about wild big cats becoming extinct? - Please select one How important do you think it is that big cats are saved from extinction? - Please select one How strongly do you like working with big cats? 361

380 - Please select one How strongly do you agree that the work you do is contributing to big cat conservation? - Please select one - Thank you for your time. Your participation is greatly appreciated. If you would like a copy or summary of the research findings, please feel free to contact me (Monika.Szokalski@unisa.edu.au; XXXX) and I will provide this to you within six months. This project has been approved by the University of South Australia's Human Research Ethics Committee. If you have any ethical concerns about the project or questions about your rights as a participant, please contact the Executive Officer of this Committee, Tel: XXXX; Vicki.XXX@unisa.edu.au. Thanks again, Monika and supervisors [Dr Carla Litchfield (UniSA) and Dr Wendy Foster (ZoosSA)]. 362

381 Appendix H Sample questionnaire from Chapter 4 Developed by incorporating findings from Chapter Four, this section includes a sample questionnaire which could be used to understand the benefits and limitations of humananimal interactions in zoos. The questionnaire has been generated by combining the suggested benefits and limitations for each handling and training practice as stated by the participants, and writing these as survey items, using both positively and negatively worded statements. In many instances, there was overlap across the types of practices regarding benefits and limitations, and so these have been merged into one item. For example, with regard to these practices for large felids, 12 comments were made regarding protected contact being a safe practice, while 28 comments were made regarding hands-on contact being a dangerous practice; as such, these have been combined in the following statement: I believe that interacting with the animal/s in this way is a safe practice for the animal/s. The merging of comments has resulted in 19 items for animals, 12 items for keepers, and 12 items for visitors. Responses could be tallied, keeping in mind that the negatively worded questions (such as items 3, 6, 8, 11, 13, 15, 18, 19 in the animal section), would need to be reverse scored first. This would give a possible rating range of for animals, for keepers, and for visitors, with higher scores indicating agreement that the particular practice is beneficial, and lower scores indicating that the practice is not beneficial. The sample inventory is to follow and is written for use with animals in general, not just large felids. 363

382 Instruction: Read the statements below and place a tick in the box to indicate how strongly you agree with each of the statements (where 1 = strongly disagree and 5 = strongly agree) Animal perspective I believe that interacting with the animal/s (replace with species) in this way (replace with handling style): 1. Is a safe practice for the animal/s 2. Is stimulating/enriching for the animal/s 3. Negatively influence the animal/s natural behaviour 4. Encourages bond/s between animal/s and their keeper/s 5. Allows for better veterinary care for the animal/s 6. Reduces negative behaviours/states (e.g. stress, aggression, boredom) 7. Provides choice and control for the animal/s 8. Provides the animal/s with cognitive stimulation 9. Reduces the animal's dignity 10. Allows for better medical/health care/behavioural monitoring 11. Has negative consequences for the animal/s (such as punishment) 12. Improves the animal/s' breeding success 13. Limits opportunities for mother-rearing 14. Is necessary for the animal/s 15. Limits the animal/s access to their exhibit/exhibit space 16. Allows for social enrichment with conspecifics 17. Assists with the husbandry of rehabilitation animals 18. Reinforces unnatural behaviour/s 19. Is unnatural for animal/s Keeper perspective I believe that interacting with the animal/s in this way: 1. Is safe for keeper/s 2. Puts unnecessary pressure on keeper/s 3. Allows keeper/s to provide better health/medical care 4. Is rewarding/allows keeper/s pride in their work 5. Is stimulating/challenging for keeper/s 6. Is necessary for keepers 7. Contributes to positive keeper well-being (e.g., happier, less stressed) 8. Encourages bond/connections/relationships between keeper/s and their animal/s 9. Increases keeper/s knowledge of the animal/s and/or their behaviour 10. Is engaging for keeper/s 11. Results in poor job satisfaction 12. Teaches keepers how to communicate with their animals Visitor perspective I believe that interacting with the animal/s in this way: 1. Is enjoyable for visitor/s 2. Prohibit visitor experiences (e.g. keeper talks) 3. Positively impacts visitor/s' opinions of the animal/s 4. Sends 'wrong' messages about the animal/s to visitor/s 5. Is educational for visitor/s 6. Has the potential for disease transmission from animal/s to visitor/s 7. Is necessary for visitors 8. Is interesting for visitors 9. Allows for a more 'natural' experience 10. Can increase visitor/s' conservation/financial support 11. Has the potential to contribute to visitor/s fear of the animal/s 12. Allows a connection between visitor/s and animal/s 364

383 Appendix I Published version of Chapter Five 365

384 lr..rernu.riw:ul Jo :..n:al of Compu.ra!i~"e P5) t:liology ~. Copyriglt by the ltttematimal Society fot Cornp:uative Psychology Behavioral.l\>lonitoring of Big Cats Involved in 'Behind-the-Scenes' Zoo Visitor Tours Monika S. Szokalski Uni1:ersiiy of South i\ustralia, i\ustralia Wendy K. Foster Zoos South i\ustralia, i\ustralia Carla A. Litchfield Uni1:ersiiy of South i\ustralia, i\ustralia W'hik. intt l'\lcth'c :auts have been argued to hold great OOClf.Cl"\'ation poietttial fot zoo vis:itoni. Llie influence on the. panicipating 2!'li1nil's behaviot is often ignol\.'d. To in\'cs?igate th.is. 11>e obsetvtd the bchaviot of one. Suniattan tigct (1>"1ntl:tf(I tlgfij $Tr.1U.frlit') and thiee African lions (l >antj:.eru /et> /eo) involved in a protec(cd C01':t1Cl tol.lt, as 11>elJ 3." lhat of thtce chocuhs (Achfooyx j 1tlxitu1) irt\'oh'cd in a hrut<!s-oo 1oor. a1 Zoos Soolh,\ usua1i3. lns1antai~1:ous scan sa1nplirtg (30-s intcf\ a!s) \l.>11.s used to tccofd a 1irnal bcha\ i()( before., durirtg. and aftct bch ind the~cenes iouts, 3." ~ell as for equivalent li)ics on non-tout days. ovet a thtce-1no11th period. Es1iomed proxirnity (close., <1ti~ 1nodciate., 2-5 in: and dista111. > 5 m) to humans ~-as also teeonk d as ~ indirect tl)c3.<;ute of intet.lc'lio1l. The ani1:ia1s in lite. proi:ea td 00111ac1 tout displllt'cd decr\'3.sed inactivi1y atid incre.ased feeding al)(! pacing <!.utirtg the touts, 001rpaitd io bcfofe a!ld after. We suggest 1h31 the increa.<;ed pacing is!flote associated ~ ith the animals being fed d11:ting!he touts. talhet than lhe touts bci.ng a S'.ressfol c,xpctiencc. Those. in lhe hrutds on tout shov.'cd variatioo in proportions of t1rultipk. be-h3viot categories aoo pri1:iarily these v.'ete shifts in species-typical bcha\ ioo. In c«nta.<;t to thor.c. in the. J)(ilteetcd OOC1tac toll:t. lhey s0011,--cd d1:ctcas1:d pacing durirtg the tou:t sessioos. No agg.ressi\'c. or o!het~ isc. antagonistic lt"havi«s ditcc(td at ltutl'isns ~'Cfe obsq"ved by arti1tla1s in eithe1 toll:t, ~ ith!hese ani1na!s typically spending l'll(lfe thin half of the-ir tour tintes in dista111 proxi1ni1y to kecpcts and \ isitoo. Co1tlbin1:d. thtse fiooings indicate. th:a1 luge fclid btha\ iot itlay be. altered by pa!licipitioo in intet3cti\ e touts, but lha1 l.ltese changes ate no~ necessarily ittdi-cative of co1npromised 11>ell being. Additionsl.~r.caroh is needed to detet1tline lhe intpacl tha~ thef.c. experiences ate. having oo the v. elfate of the 1111it1'1Sls. This study reirtfofces!he poienti31 fot be-jiaviota! t1~nitoting to be used a." a inelho:i fot ass1:ssi11g the inouelx"e of visitocs on UlO a 1io'l3ls. The prilnary goaj of moden1 200S is to protect anijnajs fro1n extinction. This is addressed nor t)n l> thmueh t'.. 11,1ivt: hrep.tline, 1'me,t :-un'-. n".(;f".:ur.h, :lntl i n.~itu t'..tw\~l"vl'llinn \\ ode hut l'llm thtough the ed"ucation and engage1nent of the visiting public. ln otder to comb.ne the neecl11 of animals (opti1nal..-.. ell-being) and visitors (ptimarily, recteation and education), as well as zoo staff. recent decades ha e seen a shift in lhe way animals and visitors intetacl and encountet each other. Today's exhibits are becoming incteasing.ly naturalistic (Fetnande2. Tamtx rski, Pickens, & Tiinbetlake. 20J9), ai1ning to addre..11s the psychological and physical needs of the ani1nals. whilst providing visitots with a 1nore realistic experience. F'utlhet breaking. the barri!r to traditionaj exjlibit viewing. 200 visitors are being. offered various intetacti" e experiences - including behindthe-...11cenes toun of off-exjlibit 200 facijjties, entering ani1nal enclosutes. touching aniinals, and This study OO(Jlj}ltfS ~'ith the cutren1 Jaws of Austtalia. a1'1d elhic.s apptoval wa." gtartled fto1n the lnst.ilute of fl.1edical rutd \'1:terill31)' Sci!nces (Sooth AUSlialia: ptojl'"ci 1tumbct 29/11). We v. ould like to lhartk!he catnivote ket pets a1 ~ S,\ fot supponirtg this s1udy. p1ttticulatly JSS()() Hllkof. Anna 8ert 1('!1 atld t.iichc!le Lloyd. V.'e also!hank two ltltoil)'rtl(ius fe'\'i e11.-ets fot!heir substatitial a1td w.luab!e feedback on elu1ier \'ttsioos cl this 1tlartusctip1t Coftespondence C6nocmi.ng I.his article shou?d be addmscd 10,\tooika S Szokalski. School of Psycho?ogy. Social W()(l: and Socia! Policy, U1i\-etsi1y of South AUS1talia. GPO Box 2471.,\delaide SA :\ US<ialis. (11.1onika.SU'lka1sl:i un isa. cdu..au). 366

385 having photographs taken with them. Although there is currently a paucity of infonnation about Lhe influence of dlese interactjve animal experiences on \'isitors, research tends to suggest that the educatjon and conservation value or experie nce..~ that allow visitors to do 1nore than simply view an animal's exhibit (such as witoes:sing keeper talks) is high (Ballaruyne & Packer, 2005: Broad & Weiler, 1998; Lindemann-Mauhies & Karner. 2006: Povey & Rios, 2005; Visscher, Snider, & Van<llr Stoep. 2009). Moreover, die infiuence on the animals is rarelyconsidere<i in the literature and.!jl order to support the continued use of animals in th-!se prograrns, it is vital to understand how participation might alter their v1ell-being. () f particular interest is the use or larger. potentially dangerous i.lrliinals in dlese encounters: namely, big cal< (or large felids. Fe/is ssp.). A1 present, 1.0os and wildlife park.1 world-wide offer various vc1itor-big cat encounters and the opportunity for people 10 engage in these appears 10 be increasing. What makes diis panieularly controversial is that these animals rarely have direct physical contact with hurnans in d'e wild, nnd man)' of the recorded internction..~ (portrayed in both the scientific literature and the mocia) docu1nent injuries and fatalitj e.~ sustained by hu1nans as a result of 'attacks' from these ani1nals (Packet, lkanda, Kissui, & Kushnir, 2005). Further. these species are typically solitary, with the exception of lions and, to a lesse1 extent, some male chootal> (Macdonald, Loveridge, & Nowell, 2010). Therefore. regular con t~t and interaction w i ~1 zoo visitors oould resuh in negr.tive repercus..~ ions ror these :mi1nals. lnfonnation surrounding the influence of visitors on 'l.00-hou.sed aninmls has focused ahnon exclusively on visitors at exhibits. a..c; opposed to i11tera.ctive tours. It is generally agreed thm '" visi1ors crui ac1 as a source of stress for animals (Clax1on, 2011; Davey , 2007: Fernandez, e1 al : llosey, 2005, 2008; Morgan & Tromborg. 2007), however it has been lllgued thm visitors may also be enricliing. panicularly due 10 the variability d1ey provide (Davey, 2007). Of note, tl1e majority or tl1e visitor-impact resell!ch has focused on primates and Davey (2007) argues that "''e carlrlot generali1.e this inforn1atio11 to othet species. In reviewing resea."ch on the effects of unfa1niliat humans on different taxono1nic gtoups, l losey (2008) concljded that. compared to primates, felids are much less susceptible (behaviorally) to the pre..c:cnce of zoo visitors. Only a small handful of studies have inves1jga1ed ~1e innuence of visiton in relatjon to large felids and the resull< are conflicting (Claxton, 20l 1; Hosey, 2008). On the one-hand, C)'Dcoovan. lii 1dle, Mckeown. ru1d ()'Doi1ovan {1993) found 10 sigoificant change.c: i11 the beha,.ior of fe1nale cheetahs ru1d their cubs in resporlc:e to visitor presence. Similarly, Margulis, Hoyos, and Anderson (2003) inve.<tigated the behavior of six 1.00-houscd felid specie.< (including lions, leopru'<ls, tigers and snow leopards) in response to visitors and found no vll!iation in activity ba.c:el on the presence or absence of visitors. The authors concluded that d1e..c:e animals do not appear to respond to visitor distutbru1ce or auen1pts to engage. h1 contra.cit, ~1 allapu r and Chellam (2002) observed tha1 leopards (n = 14) a1 four 1.oos in southern India displayed higller levels of activry and lower levels of resting on visitor absence days compared to visitor presence days. Likelllc<e. Sellinger ll!ld Ila (2005) found that both visitor density (number of visitors) ll!ld intensity (behavior of visitors. ranging fron1 qujet - no loud talking, to extrerne - loud talking and/er shouting) innuenoed the behavior of l \,l,'o jaguars at Woodland Park Zoo, Seaule. ~t o... ever, their ti11di11gs were so1newhat differeru to Mallapur and Chellam's (2002) si 1ce tj1ey found that lo\\ er visitor density wns a.c:sociated with increased tirne spent out-of...,c:ig.ht by both anilnals compared to higher de1l1ities. and d>at the female displayed higher pacing at low visitor intensity (characterized by relmively quiet talking among vc<itcrs) compared to higlier imensitie.<. ~1oroover, although non-significant, aggressive behavior iu the 1nale sho\\ ed a trend to increase

386 with higher visitor densitje.11. Based on onjy t\vo animajs, 1nore research is needed to substantiate these finding. Regardless. this inform~uj on indicate..11 that the visiting public can influence big cat behavior in "'arious ways. ln addition to this lack of concl"usi e knowledge regru'ding the iinpact that visitors ha e on big cats at their exjlibits. research to date doe..11 not aw ear to ha"'e explored the influence of interactive tours. Indeed. "'ery liuje is ki1o"' n about the visitor effect during close encou.nters involving any specie..11, with aquarium-housed dolphins participating in 'swim-\\ ith-dolphin' prograjus being the primary focu..11 of such research (Brensing & Linke. 2003; Kyngdon. ~1inot, & Stafford, 2003; Samuels & Spmdlin. 1995; Trone, Kuczaj, & Solangi. 2005). Overall. tliese studie..11 ha"'e concluded that the animals are not being adversely influenced by their participatjon in such experience..11, both in lhe short- and long-tenn. Since dolphins are social animals, these findings cannot be generalized to other. less social specie.11 like big cats. What these srudj e.11 do highlight, ho"'re" er, is that behavioraj analysis of anfrnals in such prograins can be an efficient and effective 1nethod for 1nea.11uring animal y, elfare. j u..11t as it has been for big cats in both enrichment-related and visitor-impact studie..11 (for example, Bashaw. Bloorn..11olith. Marr, & Maple. 2003; Sellinger & Ha, 2005). Aims of the Cur rent Study At present, big cat tours in zoological organiui.tions aw ear to be taking the fonn of either protected or hands-on contact. Protected contact refers to situatjons where people indirectly interact with big cats through so1.1te form of barrier, such as mesh fencing, whilst hands-on contact refers to situatjons \\-here there is no barrier bet\\ een the t\\ O and direct physical interaction is possible. Zoos South Au..11tralia (Zoo..11 SA; including Adelaide 'loo and Monarto 'loo) is one Au..11tralian organization currently offering big cat encou.nters to lheir visitors, including both a ptotected contact tour of the tiger and lion facilities :it a city :wo, a nd a hands.on cheetah tour at an open-rai1ge zoo. It therefore serves a.11 an ideal place in whjch to explore the impact of such practices on the anfrnals in"'olved. ln this case study. ""'e sought to empirically exrunine the behavior of three group.11 of big cats (tigers, lions, and cheetah) doring the.11e behind-lhe-scenes zoo visitor tours, in order to detennine "'-hether any behavioral changes y, ere exjlibited that 1nay indicate ajtered welfru'e. Species-typical behaviors were used a.11 indjcators of positive well-being, while stereotypic behavior (pacing) indicated the po..11sibility of stress or compromised well-being. ln addition. y, e y, ere interested in doctunenting the nature of interactions between these aniinajs and hu1nans (including both keepers and visitors). since little is understood about ho\\ big cats respond to people ru1d the behaviors they engage in during intetacti., e tours. As ru1 indjrect 1nea.11ure of interaction. y, e investigated physical proxilnitie..11 of big ca l! to hu1nans during tours. Subjeccs Method The subjoois inw>l\'td \4.'efC!litre separate groups of llk!ult big cats cuttcrttly parlicip:iting in intcroctive touts.: ooe St1t'l'l31tSn ti,s.ct (Pu.nr!i~ro rigris.\'(.lmurru.~) atld thfte Africat1 lio 1s (f'w.tl:t:f(j /eo le<>) involved in Ute. protected contaci tout at Adelaide Zoo, at1d llitoe Africat1 cl'tcttah (:\cloonya J~1'af{4S) invoh"cd in Ute h.aoos-0n tout a1,\fot1st10 Zoo (sec Table 1 fot indi\ i<!.ua! details). The tiger ai)d lions have been inw>l\'td in the protcc(cd co111ac1 tour for apptoxirnatcly the ls s.1 fout years, wbils~ the cttcetah have been par~ici pating in hatlds-oo "'isi!ot exfx'tienct:s sinoe.!hey wete cubs (eight years)

387 Table I lr.db1dirll! s!lb'ecr d~rails Speci:1 Narnes (Gendet) D.O.B ll fl.ace of birth Apptoxirnate age of arri, 'SI a1200\'l SA SLSyr"ll Rearing histoiy,\.fotho:t African lioo Mujatnbi (t.f) t.~ogo Zoo, Aus.ttalia Yi.i i (F) & 06.0S.2001.\i.tekland Zoo, Ainarti (FL New2'als 11d Skulrul3 ("-f), l ~ ~foosno Zoo. Tsots.ie (fl.1) & Aus.ttalia 11 Askari a = lionesses bom in same liucr; b = cheeta."i bom in ~nte liner JVUr~. Subject Ho usin~ and Feeding Routlnes 2ytS I )'f NIA ~fothet,\ fothet Hunian Adelaide Zoo.,\!though,\delside Zoo houses OllC rna?c. 111)(! J\l.'O female Sun'ISlt'Sn tigers. oojy the l'll3!e wa." obsef\ ed crue to lhe two fcnia!es alletnsting bcl.,.,ttn bci.n.i housed on-oisplsy 111'1d off.display f()( lhe toojti, I inti ting d:aia consstcncy. All ti gets ere ltoor.cd.soliterity and the ind,vidual of intefes1 is typically housed in the ssrne ot1-ex.hibi1 cnclosurooo a dsijy ba.~is '. 0 \'Cmigl'n. lhis tiget is gi\ en access 10 an off.exhibit ct1e-los.ure and is then pu1 on.cxh.ibi1 in the 1notnings - the 1iming of th.is vaties dejx'nding on Cl)(!asUie itlainte-nsnoc. bu! is ususljy aroul)d 8.1S31!l oo tour d~ys at!l.l bcfu1c u 10 oyt ning ll!.iuj) (9.30;..;iii) un t11.-x:1 \uur 1.-;..;ys. Tiu:. un<.d tibit <:.~m:los11:tt i~ 1:1ppto>.iirnuc:ly 2:?.3n1'-' atlll oontairts l 1no:n, 8 pool, atid "'arious usro ro.lis.tic plenl" and subs1l'srt. Thero ate a ntunbet of shaded ind tcueat locations, a." v. e.ji a." log_i and rocks for e!evadoo.,. The l ~n facility a1 Adc!aide Zoo is 001npiised of foor adjacct1t cnclosufes (lhn:e afe-3s rn C3Ch in s.i:.:e and the ()(het-170 in!). Fences COrll)CCting Ute. enclosures can Ile opened to pro\ i<le the aninials with acocss 10 tnofe than ooe enc1c9jre at 8 ti!f1e~ usual.i)'. the fences be1wl'ttl the tliffc. sti'isuct enclos.ures rct1'1sin open (ro1alli11g nl:). Esch of these,!nclos.ures contain ti~tly diit based flooring. v.ilh various uees andlot sn!all phuns irt C3Ch., ood raised pla!fonns or tocks f()( e!evaiioo. The latt;et ejlhibit also couairts 8 taised, grassy 1nou11d and a large tree fot clinlbing. Sinoc. the n!ale lioo e-xpetienres 8 nltdical oot1dili0il, ""'llich ha." nlade h.inl J}tone io r.ci.i:ufes in the pa.~t end therefore. ~\llnctabk around olher aniti'isl.s, he is housed separately fro1n the tv. o fcn'is!ts. The anin'is!s fife 1tlai.n1ained in the Ort cx.hibi1 etelosurcs tlt:mughou~ lite day at1d night. During cx.hibi1 itlaintenatice on 1()(1-(0lli day nwmi.ng'i (usually lx-1v. een 9.QOaitl 9.30am).!hey are switched lx-1v.--een lite. st:iallet and lstger enclosures (usually gi\'ing dx-1n t""'o days in C3Ch e::hibi1 111 a ti.ifie). This typically t~uires Ute. ti'is!e lion 10 be l:e!)( in a Jock-away area whilst the fcinales ate shif1e<i. \Oohkh. mull" irt hi1n being off.exhibit during ther.c C.!t'ICS. rn order io tej}licatc. 111ural ftt.ding in lite. ""'ikl. lllese aninials ete no~ fed C\'Ci)' day. The tiger at1c lio 1s arc. typically :Cd oo Tuesdays, Thursdays. Sa!Uidays. Sundays tnd public holidays. Wh.ilst the propo1lion of food provided 10 C3Ch. etin'isi vaties aoootding io W it individual \l.'ci.ghts. hotse itlea1fom'ts a 1najor pat1 of the-ir dicl They are also fed chicktn quite fegulsfl y. ""'ith kangaroo. tebbi~. wtkey :md goat ptovidcd ""'hen availab?c.,\fca~ is ususljy gi\'cn a." p<11dal i::..u:a.'>k'.» ' d.\b Big Cai suppltt1~~u 1 :.WUtl.l. Th t st: ai1iluals me U) U11Uy f1.-"'l'.l in t11cit ou c>.blb ii t:i:a:!j)ults, ai: atound 230p1n for the tiger atid 2.00pin for the lioos. Dunng the touts, these a 1in'IS!s arc. also fed a Sti'ISll pofti01) of 1tlea1(e.g.. a 1ut l:ey neck). 8 S pall of the e-xpetience involves visitofs ha\'ing the oppon:unity 10 feed an animal. f\lonarto Zoo. Thi.s 1.00 cuttcnlly houses two g.to"llj}s of 1nale disj}jay cheetahs (as ""-ell a." bfecding fe1nalcs off-exh.ibi1). ftov. e\ er 01lly ooe of these is involved in '-'i.s.klt touts. The oo-ejlhibit facil ities fife contpri~ of lhtoc adjaccn1 tjlhibits. The ti'isin e.xhibi1 (w-tlich has a '-'isitot \'iewing pletfonn attaehed) is ""'lletc;. lhe thfee s.u!ijccu arc. hoof.cd eich d3y. as ell as fflost nishi.s of lhe v. eelc. Ths e.xhibit is epptoxirnate.jy 4500 itl. 111)(! oontair.s ''atioos nsroro.lisl!c plat1ts and subs1ta1e. J)ro\ iding '-'atious s.hadcd )Cations fot lhe aninials. It also inctudes 8 Sti'ISU ' -1ut' filled ""'ith h3y, a lo""' platfoml. a latge tree kig and 1'3.ised J}latf«m f()( clinlbing. a." \l.'cll 8 S a srnall d!inking pond. Sc\'Cial 1ighL" pct ""eek. the animals ate housed in ooe of lhe ()(bet on-e.xhibit enclosures, ""'llich oot1t11~ns siitlilat foenucs a." 1 0.."Ca.~iotally. ooe of the 1v. o fe1na1e 1igcts is si~-en access» thi.s e!x"losufe and lite n'is!e is e-itht-r housed off.cxhibi1 or in a 1()(heroo ejlhibit ctle-los.ure

388 !heir ftgulat CJlhibit. bu1 is roogbjy f.c'vcn times l31sct. Mien!his OCCJl'S, the!.lttce cheernh arc. then sh.iftcd 10 thc-ir ftgulat c.xhibit in the moming. ususljy before zoo oix- 1ing houts (9.30ain). Doe 10 thcr.e. a 1ion l.s being hai)d l'3.ised.!hey hsve reguh'1r physical ('()(ltflc( with W it kccpcts, u'lto ct1tcr!he ct1clos.ute for routil)c. 1tlai.ntcttance on a daily bas.is. U~u.,uy thb o.::,;ur,. bc-fo~ l O.<>OM' on b.:>th!<)ur M d non tour d4y.. A na11.11'll1istic feeding paucm i.s ajso adopted a1 this zoo. wlietcby the cheetah ate usually 01lly fed five days per \4,'0Ck (with!\1o003y and Thursday being noo focd <lays). Thci.r diet includes jxlrlions of bttf, hon;e, kangaroo and rabbit. Typically. e!'ic'b individual is fed kg of nlt'31 each (inctudirtg bone). bu1 ooca.~ional!y they will be gi\'ctl a kg kangaroo carcass to oonsume togctltcr. Kccpcts like. 10 avoid a f.c't feeding time for these aninn ls al)(!. 1111lil.s1!hey ate often fed a1 around 3.00p'n during a kccpct talk. thi.s C3il occur at any time of day. When toots ate conduacd Oil feeding dsys. lhe.se arti1tlals are us.ually gi\'cf'l the it daily feed during!he tour. Tour Details Prote-cced contact. lour at :\dejakle Zoo. This tour is conducted up 10 tlttce tl'k'lc'tlirtg.i (Tuer.day, Thursday and Sunday) pet 11>eek. rt Sla!t.s ptioc' to the 7.00 oix- 1irtg and fun.s for approximate?)' Ol)C. ft()ur (8.-1Sain-9. ~Sam). during \4,tlich up 10.six \ isitot's (mi 1inruttl at;e 12 )'t'.ars) ate 1outcd fit'st!.lt:mugb the tiget on and off-cxh.ibi1 facilities. and ltlct1 lhroogh the lion OO and off-c.xhibit facilities. Whctt \ i.siling!he tiget facilities, 'lisi1ots view the aninials ftotll outside. of l.ltc enclosures. Visitol"S view!he tigct of itttcres1 bou.scd in h.is regular on-cxh.ibi1 cnck>sure. 1111lcre he. is al.so loca1cd before. and aftct the tout. \\'hct1 \ is.iting Ute. Jioo facililics. ooe of the :>malk r Jioo.exhibits is vacated of aninials and 'lisitofs.cntet lhis enclosure,.so lha1 they can vicu the tl)ale lion Oil Ol)C. :>ide and the female lions on tlte othct in thc-ir tcgulat oo-cxhibi1 ct1d os.urcs, whetc. they arc. also housed befotc. and after the tour. Tlte toot group gooctall)' fcf'l'isins with e!'ic'b individual tiger for IS ntin. at1d with the lio 1s cot1lbincd fot a1!()(hcr IS itlin. Visitofs arc. intfoduced to each of Ute. individual arti1tlal:s. learn abou1 their life 31 Adc.Jaide 7..oo. h3ve a chat1ce to feed them th.mush nte.sb fencing. and Ltl:e pool:ograph.s of lheitl. i o nn intain visitor safe!)'. visitot's ate as.l:cd by lite. keeper to tcmain approxion rely I rn away frotn the.cxhibi1 fencing at all tinte.s. The oojy.exception to this is \4,tlen feeding an anin!al. durirtg which tiiflc!he k('('pct apptooches the fct1ce with the vis:itor and the. vis:itot plsocs a piece. of rnca1 through!he.cxh.ibi1 fencing. Visitot'S ate. insliuctcd 1()( to pbce thc-ir hand thfough the. fencing. 10 le( go of the food ooce the animal ha.i taken ii. and 10 step bact ooce they h.ave done I.his. One kccpct al)(! one 2.00 \ o!un1ocr ate always presct1t tlttougho11.~ the 1out. Rand~ on tour at ~tonarco Zoo. Th.is tour is conducted up to three momings (Mooday. Thutsday at1d Saturday) per \4,'0Ck, for approxit1n 1d y Ol)C. OOur ( JOOOam-11.00anl). During th.is.cnco11.ntct, up to four \ isi100 (rni 1ioruitl age l6 years) ate takctl into the ch~lah cxh.ibi1 with the a 1ion l.s ptcsen1 (with no piotec1.h'c batriet bciwccn lhool). whetc. they ff);l('t Ute. indi\'idual cheetahs. le2tn abou1!.ltcit band-rearing situation. at1d wi1t1ess ~fflc po:>itive fci. 1forocnient ttai 1ing and aninial feedirtg ( Oil feedirtg days oojy). They afe also able to pal the. arti1tlal:s and h.ave pliotogtaplis taken wilh thctn. The cheetah reitlain in lhis.cxh.ibi1 bef otc and af1et the 1our. Bc-irtg a h3nd.s-0n tou r. a rtuitlbcr of t1)!'.as1u-es have been established to itla~ntain visitor safety ai)d these a.-e.explained to visitofs bcfofe they Ctltct the cheetah etid os.ure. They ate insuucted to remain s.1anding a1 all titt)('s. and 1()( 10 Cfouch down to the anin!al's level. They ate also irts1fllcted fl()( 10 approach lite a 1ion ls unless asked to do.so by a kccpct. at1d arc. told 10 reitlai.n s.tiu and 10 refrain from tot.tehins a chcc1ah if ol)c. approaclics thctn. When a l:ttix-r does allow 'lisi1ots 10 p.31 the animals.!hey ate irts~fucted not to 1oucb the cheetahs' faces. oot 10 look the aninials dittt1ly in the. eyes. To promote. safe itttcl'llc'lions, the cheetah arc. positi\'c.jy tci. 1forocd (with a oornbirtation of \".cttlal ptai.se. patting. aoo \ arioo.s food pastes) 1111lct1!hey arc. in close proximi1y 10 visitofs.!\1oreo,'cr, two l:eepefs ate always ptcsen1 Oil tltis!olli: one to ('()(ldll'cl the touj and the. othct to n10 1i1ot acti\ i1ics of the a 1ion ls and \ isitol"s (e.g.. 10.ciu.ul'C tha~ visi1or"s ai-e no1 lea\'ing Ute. toot group). As a precau1ions1y!flca.~uic, l:ttptfs also catty sptay bou!es filled with ws1er. 1111licb niay be used to deter any potentially d a11ge1ous anin!al behavior (at1ccdotal accounts fi'ool kecpcts indicate 1.h.a1 th.is ha.i ne\'ct needed to be uf.c'd duritlg a visitor tout). Bcha\'iot wa.i cadcd according 10 an ctttos:l'aitl containing cate~ics of fejtd beha, ioc': <lefini1io 1s wcte adapted fto1tl pas.1 folk! ft.search (Skibicl. Tfevino, & Knsughcr, "2007: Uon p:alhy et al., "2007: Van MeltCt, Hatti.ngct, & Bolen. "2008: Wideboo...,'Ski & Brown, 1998). These definitions u-ere ftirl.ltcr ftfincd ba.~cd on beh.avior obf.c'l"\".cd durirtg pre!io)irtary obsenation.s, ptior to fo!rnal dma oollcctioo ('()(J~1tlcncing (these obser.. atio1m 11>ete not inc.ludcd in!he anal)'sis). looi\ idual bch3viol"s 11,--et c. thct1 condcttf.c'd irtto beh3viot catct:oties (Tab?c. "2). Es~i 1tlated proxi 1tli~y be1wcen anion l.s and i>oopk (l:cenets a1'1d zoo visitots). was teeo!ded b)' visual ins!.\cctioo and divided into thtee. atcgotie.s: close (< 2 in). n!odctstc (2 5 m). and distant (> S in). Any o!.lt.cr occjn-enocs al)(! pos:>ib?e. CXll'Sneous

389 varisblts, s:ucll as the pfesct1ce of neighboe'ing ('()(lspttific.s, or ktepets al c.xhibia (oul..;i<le of tour titl)?s). 'ol.'c'ie also 1ia.:cd irtfonnally oo!he da1.a tee0td sheets. i able 2 Catego1y li1active Fteding Co 1spceific inter.action Pacing Aggressioo Out Of si.ght Othct In.eluded bcha\'ioo Si1. S'land. ot tay u'hils1 11()1! eng~cd in any other bcha, iot (animal itlay be vigilattt or not ''i&ita1u dwing any of thtse bellsviots) Walk.cliitlb. 1tlartipuh'1te object, ~iatcb grou1td/gra...;s. ctoucli. sniff self. scta!ch self, groom self, \'OCali.1'!, ot atty olfac(01)' bcha, ior (t ~.. spfay. Ocluncn) Aoy corts'lutlpiion bch.avio( (c..s,.. C3lirtg. d~gi ns food) AtJy direct aoiliative intttactive btha"ioc' u'ith a conspecific, including llt3d 1ub. sniff. paw a1. plsy. 'pm enn OC' olhct sexual bellaviofs, groom (bch3viots tnay be either dircci OC' th1 ou.gt1 fenoc, al)d niay ot l'l3y 1()! il)chk!e affuiative \'OCaliuuions) NJ)' dittc~ a.fliliinivt itlitl'\l('lh't bcha\'ioc' wi1h a h1unan ():cej)tf atidlot \'isitoc'), i~l udi.ng tas:ing mca~ from (inclodirtg olhet bcha, ioo. S'llcli M sttetcllirtg u-p ro take food. or ''~ali.1.i ng). bci.ng palttd (w-tlilst ensat:ed irt any olhet bcha\'iot. soch a.<i si1. stsnd ot cat~ tiu1y include being rewarded), cng~e in uai.nirtg (so:ili as si1 oo ooitlmand). or routine he-altb cheeks (s Jch as being wc.igltcd on sca!ei) Repetitive walk or trot at0u11d elx"losure witlioot att apparent goo.i (displaying at k-as1 tv.."o tepelitions of the Satl'le pauctn of it"lovemcnt: may OC' may not include vocali1.ation) 8a11gi11g on ct1closure dooc's/fcncing. suiking a1 amthtt artiitlal (tiny ot uny 1101 include aggfessive \'0Cali2.a1ions. St.teh a.<i grou,lirtg or llis5'ing) Aoy ius~attce u'lteie the a 1ion 1Jii.s beh.3vior is 001 visib?c to obsetvet Aoy bcba, ioc' Ml listed above (c..g,.. defecate, u-rina1c., drink. regutt:itatc., or enwlt:e with Sptcific enticho)?nt -device) Prooedure The!>!udy used at1 obstfvali0"9l dcs.ig" 11.'ith d9la (atiirnal beha\ it}r atid pruxirnity) be.ing coul'cioo iu 30-s ittter'11als acootdirtg to ius~attl.sllcous scan siunpling n1 '1hods (11iutin & BatcSOil. 2001). All obsctvations occutrcd w-tlcn iu1itin ls wete in thc-ir on-c.xhi!mt enclosures. Obsetvations ete ust.tal.ly ool?ctted fo( ooe hour before c-~h tour (with the exotplion lha1 01lly half an hour of J)te tour da!a wss oolkc(td for!he tiigcr. due to husbandty needs and tile tour commencing close to the st3.:'1 of the wort day). the durodoo of the tout (approxiitlately IS min for tile ti get. 15 min for lhe lio1ts. and ooe hour fo( lhe cbtciahs). al)d the hout aftet the tout. As such. obsc1vslions typically spatt1'1cd ftoitl 8.15aitl - loam for the tiger. :8:30am aitl for the lions. am 9:00am - 12:00pm fot!he clttttsh. Behavior was lhct1 obsetvcd foe' the same ti1:ne petiods oo Mt1 1out days (Moodays. Wed11esdays al)d F'ri-dsys fo( the Lige( attd l io11~ Tuesdays. Wednesdays attd F.ridays foe'!he cheetah). Tile data oollcaioo period spattl)cd frocn Qc(ober to Dcceitlber 2011: typi<:ally. 01'1C 1our and ooe 1!()() tou! day wcte obsetvcd pct iu1it1n l gfou-p per 11>tek O\'Cr this tlll\.'e n10nth period:. This allo"'ed fot 12 tour atld 12 Mn-1ourd:a)'S of d31a foe' C3Cb individt.tal a nimal, with the exception Iha~ ooly 10 toor -dsfs u'tte collected for the tiget due 10 Ml enoug.ll tours bc-ing oondoctcd durirtg lhe srudy period and chattgcs to the oo -clth ibi~ tin nsgcmen! of lh isl artiitlal toward tile end of the srudy. One rese.atchct (11.S.) oonduc(td au beh3vior obscr'\ ations. i C'> 1es1!he ft liabili1y of the t:lhognun and ptoximi1y estiitlates. an i111er-ta:ct as...;cssmcn~ was ootlduci:cd by a SCOOil.d. indt f)cndc111 obsel"\'c(.!\1can agftel"tlcnt between t.t.. ~. and this obse1w r was 95~ for bcha,!oc' and 91% for proxlntity. achieved in ol'ic hour or sinwlianeaus -dsia colleclion for eacb spccic:i. i o detennine propo.1tions of bcha, iots petfotnw:d by ca-ch animaj. counts wcte to1alk d fo( each individ.uaj iu1itin l across each ('()(lditiotl ( pte. durirtg and pos!-(ou! ses...;ioos on rour days. a.<i well as equi, akrtt pro. during and post rour time blocks Oil nort tour-dsfs). A tl~n \'Slue fot the ptl'ttn1age of sciu1s lha1 C3Cb indi\'iduaj iu1it1n l spent engaged in C3Ch bcha, iot c-a1egoty 'ol.'ss tlten gcnctated. \\'here an individual was off-cxh.ibi1 for arty pan of att la!thoug.ll this wa.<i the aim. it was 11()( always possible to conduc1 one touj' iutd ooe t1on toor day eacli 11>tel: foe' ea.zb group, due to fac(ofs such a.<i heavy rain iutd touts noi being on du-c to M \ is.i10c' bookings

390 obsct1. aiit>n period, these. scans v.'t te c,xc-ludtd fro1n the taaj n.urnbet of scatis fot that petiod. such ths! bchaviots 11>et e. only caj:u!ated a." propo1lions of on-cxhibi1 sciu1s. Sinoe. t'olo'o types of touts 11,--ete. obsened in th.is itudy, anslyscs v.-cte oonductcd separnte!y f()( a 1tln ls in the. ptoioote:i cot11aci tour and those in!he hands-on toot. Data f()( each toot 11,--ete. analysed by combi 1in; die daily ptopor1ioos of behaviots f()( esc"b ani1tlal in thsl tour. l)ue 10 lhe data no1 tj)c(:ling lhe assu1np1ions fot pats1ncuic 1111al ysc~ (panicu!arly Ois1!he data 11,""tte not noimall y distributed), noo patafflcttic analyses v.'t te t itlployed. F'riedrnati s A:NO VA was used to tcs1 tor bch3vioiaj <htte.re1icc.s a:tos~ the tllttt tour day con<htions (pre. dunng atld posl) 1or each IO.ut type: v.'ltci-e lhe tests 11,""tte significant. pos1 OOc tes.t$ 11,""tte ('()(lducted us:ing \\'ilooxo1) sigtlcd tsnl: «itlp:arisons. These ~s.ts 11, (te also used to tcs1 for diffc-ftt1ces in ptopottiatls of scans spcn1 in c-lose. n!odtl'ate 3tld dis.ta.."t proxi1tli1y IO both.<ecpets al)(! vis:itots f()( bolh tour groups. All analyses 11,""tte co1s:idered significa111 at lhe Ji< O.OS bvc.j. These. 1111alysc~ inc!'uded a toi.al of 46 d3ys of data for Oie JHOltctcd cot1tac( tout atid 36 days for the hsnds-oo!our. Wilcoxoo signed-tank cornpari.s()(is 11,--ere also 1scd IO test for bcbf!\ iotal diffc-fcncc.s bc1ween tte.!olli and noo tour day ('()(lditions (pit, dutins and post) f()( each toor 1ype (,p < 0.05 le\'tl of s:isnificanoe). When 001i:paring!Our d3ys to tlort tour days f()( the protected COiltaci tour, data frocn the ti get 11>ete. 1<>1 i1icludcd since this flrtiitltll 11,-ss often off-cxh.i)i1 (due to husbiutdty and ct1clos.ute 1tu1in1ooat1ce) until l111c-r in lite. 1noming on lflost Mil tour days atid ii v. a." lhought that th.is t1\igh1 influence lhe bcbf!\ ioo displayed by th.is a 1ion l in lhe diffc-fttlt ti iflc blocks.,\ s such. these. cornparisatls only inch:ded lion d3w.. involving 36 iout days and 36 no n-tour days. For!he cheei.ah. lhesc c-)rtiparisoos s:imilatly inc-luded 36 tout deys and 36 1'1Cln-(Olli days.,\u ~tfltistical atiajyses 11, ct e conduacd in SPSS \'ttsio Result.~ AnaJyses {behavioraj differences bet\vten the three tour day sessions, bet\'\ een tour day and non-tour day se.cisions, and proximity to humans) are desctibed sepatately for both tour groups here.in. UbSetvauons 01 bettav1or categories lhat wete considered vety tow (tes..ci lhan :>% of sca11s for au conditions in a given compatison) have not been included in the analyses. l'rotected Contact Tour Changes in be-havior on tour days. D1fferences,... ere seen in inactive, x~{2) = : p < 0.001,feeding. 1'(2) = 58.63, p < 0.001, and paoing bellavior, x'(2) = p < tcro.<s the thtee tour day conditions for lhe protected contact group {Figure I). Inactive behavior \l. as er doring tours compated to both pre-tour. T = 2S, p < , r = -0.58, and post-tour. T = 23, p < 0.001,r = as \'\ ell as highet post-tout ca1npared to pre-tour. T = 329, p < 0.05, r = Feeding was higher doring tours compated to b:>th pre-tour, T = 0. p < , r = -0.55, and pos.ttour. T = 50.50, p < 0.001, r = , and higher post-tout co1npared to pre-tout, T = 6, f. < 0.001, r = Pacing was higher doring tour comptred to both pre-tour, T = , p < r = , and post-tour. T = p < 0.001, r =..0.52, and higher pre-tour compared to pcgt-tour, T = 29, p <0.02. r = Thete wa.ci no difference in proportions of acti e species-typical behavior aero~ the thtee conditions (p = 0.523)

391 100' 90'. ' il ~ ' 0 "!!. = ~ SO' '. ;oo 40' JO' 20' I I 91 Duru~ taur Pottl l(iw" 100. oo. Pigur~ I. The ptopon:ioo of scans obsened a." inact.h't, active spcci:es-1ypica1. fcedirtg and pacing actoss lhe tlttre tour d3y conditions for the protected co 1L&C1 toot. Bats iooicate ~2SB a nd av.ctisks indicate 1h:s1 a s1.a1is1ica1ty sig 1ifican! diffett1'1ce.,.,'ss fo11.nd for a beh:sviot categ()(y. Tour day versus non.tour day behavior. Compatisons of lion beha\'ior during the pre-, doring and post-tour ti1ue blocks on tour and non-tout days can be seen in Figllre 2. Compatisons o f the pre-tour block tevealed that lhe only difference in behavior \'\ as that pacing was higher on tour days compared to non-tour days, T = 84. p < r = (inactive, p = 025; active species-typical, p = 10.05). Co1nparison..11 of the during-tour block revealed et p!'opojtions of inacti" e behavior. T = 6. p < r = -0.61, on tour days co1npared to non-tour days, but higher feeding. T= 0.p < r = -0.53, and pacing, T = 37,p < 0.05.r = -0.53, on tour days co 1npared to non-tour days. Proportions of acti e species-typical behavior in the doring tour ti1ne block did not differ between the two days (p = 0.13). Co1n.parisoo.s bet\'\ een the pos.t-tour block on toor and non-tour days re\feitled no differences in proportion.'l of behavior (inactive, p = 0.46 ~ active species-typical, p = 0.77, and pacing, p = 0.09). InteractiYt-!behavior and proxintity to bunulns. A mini1nal proportion of scans were spent interacting \'\ ith hujnans for any of the aniinal~ during the protected contact tou.r (M = 0.98%; SB = 027%). A 1nini1nal a1nount of aggressi\fe ibeha\'ior was obsef\:ed (M = 1.94%; SE = 0.66%) and none of this was directed at the tour group. As an indirect 1neasure of inte.raction, proximities to humans \'\ ere also recorded and can be seen in Figure 3. Animals in the protected contact tour varied i11 their observed proii1nities to keeper/s, x. 2 (2) = 26.18, p < 0.00 I, spending 1nore scans in distant co1npared to both close. T = 132., p < 0.001, r = -0.47, and 1nodera1e. T = 128, p < 0.001, r = proximity. A siinilar pauem was observed for proximity to \risitor/s, with thi.1; varying in distance, x:(2) = 20.34, p < ~ and the aniinals spending more scans in dis.tant co1npared to both close, T = 14 8, p < , r = -0.43, and 1noderate, T = 215.:50, p < 0.00 I, r = -0.37, proximily

392 100'0 90'~ SO'> 70L ~ 50"'o O"~ oa~ IC" O" 100'. < JO'O.E 20' ' 1000 OOo -- M,-.~.,...,... Prt btilui\"'ior 90'. ~ 80'.o I ~ ~ '0 " 50'o E 40' T""" ~ Non-I.QUI - lll.lc1wt Acti\'t~- P'.ciq:.,..., D11rin2 b,b:i,'1or l OO'o 90'o 80'0 'IO' 60'. 500, ' '' 300, :?O'o 100. oo.... ~ Actw~~ -~ """" P~" btl1;t\lor Figw~ 1. Co1:nparisoos of obsti''td behavior <!:uting!he. three thne periods (pte. ctudng :uxl pol>t} on t\)ur al)d no11-1,1ur days fof the. prooectcd coiu~t!(!ur. Bars indica1e :1:2SE and as.1ttisks indica1c Uiat a s.tati.sti-cauy slgnifi~111 diffcrenoc Y.':li fov.nd fot a bch3vior Clltc!<lC')'

393 100' 90' ' ''., ~ 111' ~ 0 (H', E 50' t.,,., =-,... " :;:... ""... ~s. \~~ Pigur~ 3. Eslin'ISted prox_ntities to keepetls eoo ''isitotls durirtg Ute pro.:ea ed contaci tout. Bets indicate :t2se Md as.tetisks indieatc. that a s.1.s.tically sig 1ificant difference.. ~1a.o; fool)(! fot a cotnparison. Hands-on Tour Changes in behavior on tour days. Differences "'ere seen in inactive. x~2) = p < ac1i'-e species-1ypical, -/(2) = 15.17, p < 0.001, feeding. -/(2) = p < 0.001, human in1erac1ion, t (2) = p < 0.001, pacing, i (2) = 60.Cl?, p < 0.001, and ou1-of-sigh1, x?(2) = 9.21,p < 0.05, behavior acto.11s the three tour day conditions for the animals in the hanclson tour (Figure 4). lnlctive behavior "va.1; hjgher post-tour compared to both pre-tour, T = 58, p < 0.00 I, r = and doring tour, T = 115, p < 0.001, r = Acti\fe specie.11-typical \'\ a.11 higher pre-tour compared to post-tour. T = 98.p < 0.001, r = and higher doring tour co1npared to post-tour, T = 121, p < r = Feeding was higher doring tours compared to both pre- 1our. T = 0, p < r = -0.57, and post-lour. T = 60, p < r = as well a.< higher post-tour co1npared to pre-tour, T = 0, p < r = flo1nan interaction was higher doring tours CO!llpared to bo1.h pre-tour, T = O, p < 0.CX>l, r = -0.62, and post-tour, T = O.p < r = Pacing "va.1; higher pre-tour compared to both during tour, T = 0, p < 0.00 I, r = -0.62, and post-tour, T = 4, p < 0.001, r = , and higher post-tour compared to doring tour. T = 2, p < 0.00 I, r = Finally, the proportion of scans observed as out-of-sight for animajs on thjs tour was higher during tour compared to pre-tour, T = , p < 0.02, r =

394 loo". 90' ' 80' ' 10'. ~ W " f SO' ~ 40'. c. JO' ~ 20-. lo'. O' in~ h:ln-e~- C)J'l<ill' f~ulf,' Bl'11:.\'i~1r c-:il ti,or~ Pril!' lour '-'l'u\ IOllr Po..c-cow Pig1u~ 4. The ~oition of scans obsel'\'td a.; inacti\'t, acli\"c s.pecies typical. fo.'dirtg, hu1tu11i int1:tae1ion. pacing at1d out-of sighl across the three!ow' day comitioos f()( Ute. hands-on tout. 8 3.!'S indicate :t:2sb atld as. 1:tisl:s indicate. lha1 a '-t::..ti$tk-::illy,;e,11.ir..:~l difl'cfc-ucc Wllll f<l~u'ld f (lt ll. bch11.,io0t ClltCeotY Tour day versus non-tour day behavior, Compatisons of cheetah behavior in lhe pre., during ai:d post-tout time block.11 on tout and non-tour dj.ys can be seen in Figure 5. Co1nparison..11 oi the pre-tour block revealed that lhete was 1nore inacti e behavior on tout days compared to non-tour days, T = 194, p < 0.05, r = -0.26, and niore feedi ng on non-tour days co1npared to tour days, T = 0, p < 01:10 1, r = There..-., ere no other beha\'ioral differences in this time block (actj\fe specie.11-typical. p = 0.46; pacing, p = 0.94; out-of-...11ight, p = 0.3 1).,\ nujnber of beht:\'ioral differences were observed in the during tour tjjne block between tour and non-tour days. Jnacti" e. T= 112, p < 0.001, r = -0.41, and pacing, T = 9,p <01)5, r = -0.36, were lo\'\ er on tour days coin pared to non.tour days, whilst acti oe specie.11-typical, T = 7 1, p < 0.00 I, r = -0.48, feeding, T = 0, p < 0.00 I, r = -0.57, and hujnan interaction, T = 0, p < 0.001, r = -0.62,.-.;ere all hjgher on tour days compared to non-tour days. There was no difference in scans spent out-of-sight doring this ti1ne block on tour and non.tour days (p = 0.05). Co1nparisons of behavior in the post-tour b lock ind ico.tc th tlt inoctivc bcho.vior wos IO""'Cr (1) tour doys oornpo.rcd to non tour days, T = 17650, p < 0.00 I, r = -029, whilst acti oe species-typical, T = 73. p < 0.00 I, r = , 3rl.d 0Ut-of-...11ight, T = 113, p < ons. T = were hjg.her on tour days C01llp3ted to nontour days

395 loo'. 90'. SO' 70'. 60'. SO'.... ' "' l.o' 10-. I O" '5«11-. Al.,._,_.... P1 lwh1,l01 100' i PO'o 10' '; 70' " ' '' ~ 50' i. " ''.IO'o 1 - :?~. 10'> I O'o 1~. 90' so.. 70'... ;oo, ' " 30' io <>',,,_.....,._ -,,,., O.riuc ~a..,ior Fig1u" 5. Cornp:riS(l(lll of obstf'. cd OC-J13\'ior ting ll'lt lh!ec. lil.'lle periods (pre. duritlg end pos1) on tout 1100 no11 touf <!3ys fo:t the hati~-on!{!ur. Bus indicate :t2se atld astcti.s!r::s indica:c. lhat a Sl.S'!istical!y s:igrtificantdiffcten«"'as fou11d tot a beh3vio( caicgo-ry. 377

396 Interactil'e beha,ior and proximities to humans. Tile prior analyse..11 re"ealed that the animals in lhe hands.on tour spent a noticeable proportion of scans interacting with hu1nans during the tours. All of the.11e direct interactions tvere affiliative in nature and no ag.gressi e behaviors towards hu1nans were observed in this study. 1\ nalyses of proximity data (Figure 6) revealed that lhe animals on this tour varied in their obser\'ed proxilnitie.11 to keeper/s, x 1 (2) = 51.72,p < spending 1nore scans in distant compared to both close, T = 110.p < 0.001, r = , and moderate proximity, T = 0, p < r = -0.62, as well as more scans in close co1npared to 1noderate proximity, T = 1, p < 0.001, r =.o.ro. A si1nilar patten1 was observed for proximity to visitor/s, with this varying in distance, :((2) = p < O.CX>I. and t))e animals spending. more scans in distant co1npared to both close, T = 148, p < O.CX>I, r = -0.49, and 1noderate proxilnity, T = p < 01)0 I, r = o. 90', so... = ~, 70'".. ~ Q 6()0, ~!l so- ~ t ~ "' 4()0, = 30', ~, 20-o Close (<lm) Modtta,. (2-Sm) Distant (>5m) 100. oo. x Kttperts Vi.sitor/s fig1u~ 6. E.niroatcd pfoxio1iticl to kctpcrli 111'1d \'isilot/s during lite h3!lds' 1'!!Our. Bars indic!ttc 1'1SE!t1'1d a.~tctiskl it1dica1c I.hat a statistically siyiificartl diffcn.'tlre -...-ss found for a compari~n. Disc-ussion To date, little research has been conducted to understand ho"'' big cats 1night respond to interactive zoo visitor tours. Tile present study sought to in\'estigate whether short-term behavior differs in tigers and lions involved in a protected contact tour, and in cheetah in a hands-on tour, as y, eu as how the behavior of both of the.11e group.11 differs on tour days compared t.o non-tour days. Differences in behavioral profile..11 pri1narily in\'o)\'ed observations of inactive, actj\'e, pacing and feeding beha\'ior for both groups. with the addition of hu1nan interaction and out-ofsight for the hands-on group. 11ljs study \\ as also interested in determining the nature of interactive behaviors perfonned by the..11e aniinals in relation to hu1nans during. the tours, with

397 estimated proxi1nity to keepers and visitors additjonally used as an indirect 1neasure of interaction. Changes in Beba\ior with the Protected Contact Tour \\'hen co:1nbined. inactive and active species-typical behaviors accounted for a large proportion of behavior in any of the conditjons exrunined for the tiger and lions in the protected contact tour. Gj\ren that these categories co1nprised of natural fetid behaviors, this could be indicati" e of po.(oitive well-being. In particul~r. the hjgh proportion of inactive behavior i..11 consistent with research of these sajne species i.n other 200..'I (for exajnple. see Jenny & Sch1nid, 2002; Margulis et al ; Quirke & 0 1 Riordan ), as well as with th.eir natural activity patterns (Gugg.isberg, 1975). l lowever. so1ne changes y, ere observed in the ptoportions of behavior of these animal.'! across condition..'!, partjcujatly regarding inactivity, feeding and pacing. On tour days, the re wa.11 a decrease in inacti... ity and an increase in feedictg d"uring the tour compared to befcm"e and after. When co1npared to the equivalent time periods on non-tout days, proportions of inactivity only differed duting the tour block (being Jo\'\ er on toui' days co1npared to non-tour days), indicating that participating: in tours may decrease the 3JUount of inacti oe behavior engaged. in by these animals. This can largely be accounted for by feeding. and pacing behaviors being significantly higher on toui' days compared to non-tour days during the tour time block. However. since the aniinal..'1 y, ere not usuajjy fed on non-tour days, it is not pos..11ible to conclude \'\1lether participating in tours innuences the performance of this 'behavior. nor ho\'\ feeding. might have altered other behaviors. lnacti oe behavior increased again after the tours, and feeding. and pacing. reduced. ~1oroover. there were no differences in inactjve,. ac1jve and pacing behavior in the pc~st-tour conditjon on tour and non-tour days, indicating. that their ifl.\'olvement in the tours 1nay not be impacting their behavior pattern at these times. The changes in pacing behavior runong: these aniinal..'i are of some concen1. Pacing. is a common fonn of :Stereotypic behaviol' in big cats (Mason, Clubb, Latham. & Vickery. 2007) and, depending on circu1n..'ltance..11, its presence can be suggestive of compro1ni..'led anfrnal well-being (Carl..'!tead. 1998).. In this study, pacing le oels were highest doring the tour compared to before and after, a.11 y, ell as higher before than after. Moreover. this behavior wa.11 hjgl)er in both the preand during tour time blocks on tour days cor.npared to the.11e tjjne blocks on non-tour days, indicating that the touts aw ear to be innuencing: this behavior. We propo..11e sor.ne reasons for thi11. For the lions. \'\ e suggest that this behaviol' 1nay be a~ociated \'\ ith the1n being fed during the tours, \'\-hen food was present. This behaviol' ha.11 been obser\ ed ptiol' to other feeding. ti1nes at thj..'1 200 (persontlf obsen«llion S'.), as y, eu as in lion..'! at Monarto 'Z.co \'\1lO are also fed on behindthe,scenes toulli (these lions were originally included in this study, but data collection wa.s cea.sed doe changes in group co1nposition both on- and off.exhibit). lndeed. research indicat~ that animal..'! may pace in association with feeding. and cartstead ( 1998) explains that te1nporally predjctable feeding schedules ha\'e been linked to stereotypy in so1ne carni \'Ores. Ba.11sett and Buchanan-S1nith (2007) al110 argue that food a:nticipa1ory activity - characte rized by increa.11ed aro"usal and activity - is typically seen in animals which are fed on a regular sclledule. Since these animal..'! are regularly fed (albeit a small arnou.nt of food) d"uring the tours, thjs pacing could be a fonn of food anticipatory activity. 11lat these animals paced Jess in the pte- and d"uring tour blocks on non-tout days co1npared to tour days, and 1ninimally in the post-tour block on both days, adds some support to this atg"urnent, since feeding only occurred dul'ing. the tour condition. Before dtawing a concrete conclusion about the cause..11 of their pacing.. it y, ould be iinportru1t to

398 obsetve tht it behavio r o n tours without feeding, as \Ve.JI as theit feeding behaviors at other tinte..11 of day. EfiorL11 to 1ninimise this behaviot could include ijteting the otder in \'\<hich these tours are conducted. to redllce lhe ptedictability a.11socia1ed with feeding at this time (fot a more detailed discussion of the negative effecl11 of predictable feeding, see Bassett & Bllchanan-Smith, 2007). 1\ nothet altetnati e could be to introdllce a teliable feeding signal (such as a buzzer) to indicate to the aniinab when they will be fed (Bassett & Bllchanan-S1nith, 2007). The rationale behind this is that by knc"'"ing exactly \'\<hen lhey \'\ ill be fed, the lions could \'\ ait for this cue rather than rely o n o ther cues (in this instance. the visual, auditoty and o lfactoty cues that a tour gtoup is present, which pre<ede feeding). Since pacing levels \'\ ere high during the tour (42% of scans for lions alone), and that this behavio r has the potential to neg.ati\'e.jy impact the visitor experience (Millet, 2012), the iinportance of addressing. it exlends beyond individllal animal well-being. For the tiger. \'\ e belie"'e that the pacing obser ed in all thtee tour day conditio ns \'\ as 1nore telated to the uncontrolled sti1nulatory access of neigh boring conspecifics, since this individual's pacing predo1ninantly occurred along the peri1neters of his enclosures which ate shared wilh female tigers. Although the tiger's behavior was not included in the comparison of tour and non-tout days in this study. pacing in the same locations was obsen:ed in this individual thtoughout the day in an earliet, un published study by dle pri1naty author. In addition. although he \'\ ould someti1nes engage in a few bouts of pacing alcng. lhe fence whete touts Y<ete conducted as the tour gtoup was arriving, he would usually lhen lay by the fence (often chuffi ng) until he was fed. Pacing in response to conspecifics has been witnessed in tigers by other researchers and the use of visual barriers to block the view of neighboring conspecific..11 o'light reduce this behavior (3ashaw, Kelling.. Bloo1nsmilh, & ~1aple. 2007; Miller. Beninger. & Mellen, 2008). Indeed, a "isual battier ha.11 now been put in place at this exhibit since this study was conducted, but research has not yet explo red ho\'\ inflllential this has been in teducing pacing behavior. InteractiYe-be-havlor and proximity to humans during th e-protected contact tour. The ani1najs in the ptotected contact tout were observed to spend negligible proportions of scans interacting. wilh humans dllring the tours ( < I% of scans). By definition, these intetactions were affiliati e in natute, OCCllrri.ng when the animals Y.'ere fc:d by keepets o r visitors through fencing. Si1nilarly, ~ ety little (< 2% of scans) aggtessio n w a.11 obser"'ed d oring the tour and is pri1natily accounted for by the lions banging on the doors at the back of their enclosure..11 (with ot without \'OCaliz.atioo) \'\ hen the tour group was attiving (a fe.,v scans also in"'ol ed the fe1nale lions growling at each other over food). This behavio r was also obsetved at other ti1nes when keepers walked through these area.11 alone. As with pacing, fotmal docu1nentation of this behavior at other ti1nes o f d t:y \'\ ould be needed to detero'line \'\<hether this is a tesponse to the ptesence of the tour group, or :J.n anticipatory behavior related to keepers being in the area and the _po..11sibility of feeding.. Other than this, no direct agg.ressi e behaviors '"'ere obsen ed. Being a protected contact tour. thete is obviously limited opportunity for direct physical interactions. either positive or negati e, to take place. As such. proxi1nity data \'\ as al110 used a.11 an inditect interactive 1nea.11u re. These animals spent lhe majo rity of tout scans in distant proximity (> 5 1n) fro1n the tour gtoup. \Vllen the 1nimals \'\ ere in clo..11e proxiinity. this \'\ a.11 usually until and/or dllring feeding. tijne. F'ot ex:un ple, the tis e r would often Jay tisht by the fe ncii~ u ntil he \ Y:'l.S fed, aftet v.'hich ti1ne he y, ould move a\'\ ay and co nsu.nte hi.11 food. Obviou~y, this data can.not be used to draw conclusions about whethet o r not the animals preferred to be clo..11e to people, but it appears that they o nly :emained close for the purpose of obtaining food. Condllcting. tours ""'ithout feeding y, ould ag.ai 1 be u..11eful in determining. ho w this 1night in fluence proximity to keepers and visitors

399 Chllllg<-~ in Behn>ior with the llllllds-on Tour Si1nilarly to the protected contact tour, high le\'el11 of inactivity were obsm ed in all conditions by the cheetah in the hands-on tour. Again, this is consistent \'\ ith the nattiraj activity pattern of thi.11 species (Gugg.isberg, 1975) and could be indicative of positive y, ell-being. 1\ nujnber of changes in proportions of behaviors \Vere witnessed, both on tour days and \Vhen co1nparing tour days to non-tour days, but these ""'ere largely shifts in types of naturaj beh aviors; for example, incteased ad ivity and decteased inactivity. 1\g.ain, similar to the protected contact group, there \'\ a.1; an inctease in feeding during the tour session, and this \'\ as higher in lhe doring tour time block on tour days coin.pared to non-tour days. since the animals ate often fed doring the tour. l lowe er, there wa.11 1nore feeding behavior during the pre-tour ti1ne block on non-tour days co1npared to tour days, as the ru~iinals y, ere sometinte..'i fed at this time on non-tour days. These differences in feeding routines :inake conclt1sion..'i about behavior change associated with the tours quite difficult. Ani1nals in this group al110 spent approxiinately 10% of scan.'! out-of-sight doring each condition and thi..'1 means that some values for the other behaviors are probably underreported (e.g., \Vhen aniinal..'i ""'ere out-of-sight for longer periods of ti1ne, it is thought that they y, ere probably laying down in.'!haded locations). These animals displayed varyi11g le,.el..'1 of pacing behavior across conditions ( < 26% of.'!can.'! in any condition), "'"hich again are \VOrthy of discussion. Like this behavior in the tiger, y, e propose that thi..'1 is related to the presence of neighboring cheetah, since pacing al'\'\ ays occurred along. the boundary of lheir exhibit shared ""'ith conspecilics. either when the other cheetah y, ere being. shifted in or out of lhe adjacent enclosure. or when keepers \'\ ere performing routine 1naintenance prior to the other cheetah being given access to the exjlibit. 11lese activities usually occurred in the mornings, roughly half an hour before the visitor tour.'! started, which is consistent with the finding of higher pacing in tile pre-tour condition co1npared to tour condition on tour day.'!, as y, ell as the lack of difference in the proportion of pacing in the pre-tour ti1ne block on tour ru1d non-tour days. Moreover, these animals displayed no pacing during. the tours. and..'lignificantly 1nore doring thi.11 ti1ne block on non-tour days compared to tour day.'!. Being territorial ani1nals. this may be an expression of their naru.ral patrolling behavior. Visual barriers may again be u.11eful here but, gi.. en the size of the enclosures at this zoo (being. open-range), this is probably not fea.11ible. InteractiYt-behavior and proximity to humans during the-hands-on tour. Animals in this tour demonstrated a noticeable (> 10%) portion of scans engaged in human interaction behavior during the tours, which was J)ighest doring this time block compared to pre. and post tour. These behaviors typically in"'olved the ani1nals.'lilting with, and being patted by, keepers and visitors, as y, ell as engaging in training wilh keepers. Gi., en lhe nature of this tour - being hands-on - there are opporttinities for direct interactions, and so we \'\ ould expect this to account for some of the activity budget of these ru1i1nals. No ag.gre.11si., e or otherwise antagonistic behaviors were displayed by the cheetah toward hu1nru1s d"uring the tour (nor during any other condition). indicating no obvious threats of dang.er. de.11pite the physical contact in"'ol"'ed. In addition to these direct interacti., e behaviors, the proxi1nity data cru1 also contrib"ute to our understru1ding. of indirect interactions. De.11pite the proportion of scans spent engaged in direct interactions, these aniinals y, ere in close proximity (< 2 m) to keepers for approxiinately 35% of.'!cans, and to visitor.'! for 25%. A.11 su ch ~ they re1nained close for notable proportions of tour scans, despite not interacting physically the whole time. flowever, they also spent o"'er 50% of.'!cans in '

400 distant proxilnity to hu1nans (keepers or visitors) indicating that, despite being in a situation where direct physicaj interaction \'\ as po..~sible, the animals sllo\'\ ed a preferec1ce for being. a\'\ ay fro1n the tour grou p. Positive A\pects of the Zoos SA. Interactive Tours AJthough these findings do sugge.~t that interactive tours can alter the type..~ and proportions of big cat behavior, both g.roups in this study often engaged in large ajnounts of species-typical behaviors. which is a positi e lioding. \Vhile so1ne pacing was obsen ed, this 1nay not be the result o f the tours being a stressful ex.perience. 1\ s such is important to discllss some of the potentially p0;.~ itive factors of the tours. Firstly. predictability has been said to be an important factor in the capti., e environment, depending oct the cil'cumstance (~1organ & Tromborg. 2007). Both groups of animals ha\'e been participating in these tours for a nujnber of years and the nature of the tours have remained relatively consistent, resulting in a high degree of predictability. Although different visitors participate in the tou:rs, lhe predictability that tours do not pose threats to the anilnajs ma.y be important here. Another factor is that none of the ani1nruls 31'e forced to interact with.. risitors or to engage in specific behaviors. but are po.~ itively reinforced for their participation. This reinforce1nent can be direct (such rus cheetahs receiving verbal praise for sitting on co1njnand whist visitors pat them), or si1nply feeding. all animals at the end of a tour. Positi\fe reinforce1nectt training has been ell-argued as the most successful and appropriate training technique, and has been suggested to be beneficial to animal elf are in the zoo setting. since it increases control and predictability o.. er aversive events (Bassett & Buchanan-Sn1tth. 2007: Laule & D<.<mond, 1998). Funher. these animajs ha\'e the opportunity to retreat if d.esired. Forced proxiinity to hu1nans has been highlighted as potentiajjy stressful for zoo-housed animals (Morgan & Tromborg, 2007). and there are many fe atures in each of the enc losure.~ \'\ hich could be used by the animals to re1nove themselves from v iew and potentially reduce any stress associated with visitor presence. Indeed, e saw that anio:iajs in both tours spend 1nos.t of their tour ti1ne in distant proximity to keepers and visitors. This is more so important in the cheetah tour, where visitors actually enter the exjlibit. ln many instance..~. lhe anijnals were not in clo..~e or moderate proximity to humans and ould often interact for 5-10 lllin before walkin.g a\'\ ay and laying down in the sllade, a\'\ ay fro1n the group. \\'hen the aniinals did this, lhey \'\'ere ne\fer followed by keepers or visitors. Like predictability, having retreat space..~ has been found to be iinportant in other interacti., e program..~. and in 1noderating. anilnal behavior and improving animaj welfare (Anderson. Benne, Bloomsmitll; & Maple. 2002, Samuels & Spradlin; 1995). One fi naj factor is that these tours are always cond"ucted by zoo keepers. Whilst it is unlikely that tours like this would e\fet occur without an experienced zookeeper present. a stlldy of dolphins at various facilities in the United State.~ found that potentially risky behavioral interaction..~ were lo\'\ er in ses..~ ions where trainers had direct control over the ii1teractions bet een dolphin and \fisitor. compared to those \'\ ithout direct trainer control (SaJnuels & Spradlin, 1995). The presence of a familiar keeper will undoubtedly influence how the aniinal responds since lhere is an existing rel ationsllip. It 1nay be that haying a keeper present mediates the interactions, 1naking it a more po..~ itive experience. Indeed, Claxton (20 11) sug.g.e.~ts that it r.nay be po.~sible for an animal's interactions/relationship with unfamiliar humans (e.g.. zoo visitors) to be influenced by the relation..~h i p that they have with fajnili31' hu1nans (e.g.. zookeepers). Alternatively, the animajs in this study 1nay not ha\'e been respooding to \fisitor presence at all, since the aniinals - ~9. 382

401 appeated to respond to the tour group in the same way as they \VOUld when keepers y, ere present at other ti1nes of the day (for example, becoming alert upon keeper atrival, or pacing before a feed). Confounding Variables and Directions for f'uture Research \\'hen combining re..111ult.111 fro1n each of the tour types, it is evident that these tours y, ere ha"'ing so1ne inflllence on animal behavior. This is consistent with the small ainount of previou..111 rcsco.reh that ho..~ shown big cat bcha..' i<>r to vo.ty in associatio n w ith regular 200 vi..«;itotion (Mallapur & Chellam, 2002; Sellinger & fla, 2005). Tiie find ings from this siudy add to tliis post research. by providing an understanding of the pote1uial influence of clo..111er visi10r interactions on animal behavi>r. HO"'e"'er. drawing conclllsions about which aspects of the tours contributed to alterations in behavior i..111 difficult. 11lere are al "' ays likely to be a number of confounding "'ariables in zoological stlldies, many of which are una.. oidable if the purpose of the research is to observe ani1nals behaving according to their regulat routines. This often makes dra\\. ing conclusions about behavior change quite challenging. and so we ha"'e discu..111sed so1ne of these issues here ar.d provided suggestions for how they 1night influence our results. as y, eu as how they 1nig.ht beaddre..111sed in furureresearch. Firstly, since there was a difference in the feeding routines of these animals on tour and non-tour day!, it would be important for future studies to consider keeping this potentially confounding " ariable consistent acros.111 conditions. in order to 1nore fully undersw.nd the influence of tours. Tilis was not possible in our st\jdy, since y, e did not "'aj1t to make any changes to the regtjlar routines of the anijnals in ed. ~1oreo"'er. it is suggested that it might be possible that the aniinal111 here responded to the tour group in the sa1ne \\. ay that they do to keepers alone. at other ti1nes of day. It y, ould be iinperati"'e for future reseatch to docu1nent aniinal responses to keepers at "'arious times throughout the day, to detennine whether S\JCh a siinilruity doe.111 exist and to draw 1nore infor1ned conclllsions about ani1nal responses to participating in tours. fl o..111ey (2008) argu e.111 that both the quality and quantity of ani1nal interactions with keepers differ fro1n those with "'isitors, therefore separating these intluences would also be iinp()rtant. Moreover, Phillips ai1d Peck (2007) found that tiger keepers differ in their individual interaction style with their ani1nals. so understanding if ai1d ho"' big cats respond to different keepers/keeping style.111 should be add:essed. AlthO'Jgh two different types of tours (protected contact and hancl111-on) were explored in trus st\jdy, lhe be.lla"'1or 01 both groups 01 anunajs wa.111 not compared and so conc1us1ons about which tour might have a greater influence of behavior. and welfate, cannot be 1nade. Such a co1nparison was deemed inappropriate here. since there ate variou..111 other l ifferences in the histori e.111 and ctutent husbai1dty of these l\\'o groups of aniinals \\.-hich could account for behavioral differences. Addres.111ing the influence of varying type.111 of tours could be made po..111sible at organizations "'-here the same animals participate in different type.111 of tours..!\long these same lines, whethet animals ha"'e been hand-reared and the amount of prior intetaetb n that they ha.. e had with humans could aocou.nt for beha"'ioral difference.111 d"uring tours, so it " 'ould be u..111eful to co1npare the sajne species \\. ith different le"'els of interaction, to deterrnine whether any interspecies differences exist. Tilis could ajjow 2oolog.ical organizations to understand which big cat species are m >St suited to the different types of tours. It has been proposed that vatious factors will account f+)f a particular aniinal' s response to hu1nans - such as specie.111, extt nt of handling in early life, ani1nal personality and te1nperainent (Da"'ey, 2007; fl o..111ey, 2003). As such. the l OO 383

402 inve.~tig.ation of ho\'\ individuaj factors relate to interacti e tours \'\ ould ajso be useful in being able to predict which animals \'\ ould be suitable for invol e1nent in such programs. As \'\ ith 1nany studjes of zoo/aquatium-housed aniinajs, this octe \'\ as limited to a s1nall nujnber of subjects, at one organization. In order to generalize findings beyond these individoals, future research should consider a greater number of aniinals, at multiple institutions. a.~ well a.s other big cat species. ~1oteover, the g.eneraliubility of these findin,g.s is li1nited in that the individual animals invol ed. here ha e participated in the tours for a number of years (four years for the tiger and lions, and 1nost of their li"'e.~ - eight years - for the cheetah) and ha e likely adapted to interacti e tours to so1nedegree. Moreo"'er, being hand-raised, the cheetah have daily physicaj interaction with their keepers, providing. the1n with further opportuniti e.~ to adjust to interacting. \'\ ith people. In this way. it wa.~ only possible to exajnine iinjnediate, short-tenn behaviors associated with tours. It would be iinportant for furure resea:rch to examine long-tenn change.~ in animals \'\<ho ate not yet in"'olved in such a prograin. In tenns of gauging the influence of tours, although the use o f multiple measures of... ell-being have been ad"'ocated (Szokalski, Litchfield. & Fo..~ter. 2012), behavior was deemed the 1nost efficient method here. Since,...e... ere interested in imntediate effects. and bec.au..~e 1nany other factors in tjie zoo environment may contribute to stress. a corticosteroid analysis (or siinilat) did not appear suitable here, but may be of valoe to a long.er-term i nve.~tig.atio n which focu..~es on the establishment of tours with ne\'\ animals. A.~ a re..~ult of focusing solely on beha"'ior, our findings do not provide a complete picture of animal responses ilo tours, and this limits the conclu..~ ions drawn. Beyond the behavior and welfare of aniinajs invol ed in interacti" e tours, future research should also addre..~s the safety of zoo visitors during. such experiec:ice..~. Although no direct a;gre..~sive or other\'\ ise antagonistic behaviors to\'\ ards hu1nans \'\ ere observed in this study, there is a wealth of literature to ij:idicate that big cats can cause both injuries and fatalities to people, in the \'\ ild and in various forms of captivity (Beier, 1991; Olapenoire. CaJniade, & Legros, 200 1; Coble. Harlan. & Harlan. 1990; Hejna, 2010; Lazarus. Price. & Sorensen. 2001; Loe & Roskaf1, 2004; Mu.tphy, De1npsey. & Kneafsey. 2007; Nyhus, Tilson, & Tomli 11son. 2003; Packer et al, 2005; Schiller, Cullinane, Sawyer. & Zietlow, 2007). lndeed. when sul"\ eying zoo keepers about their opinions of interacting with big cats. we found that safety \'\ as a 1najor concern a1nong participants (Szokalski. Litc hfield, & Fo..~ter. in pre..~s). A nu1nber of these keepers discussed hujnan safety as an is..~ue \'\ hen a hands-on style is adopted; lhey also expressed the concern that if an "attack" does take place. the animal invol'ved is usually euthanized. highlighting the neg.ati e consequences for both people and animals \'\<hen interacting. Typically, participants explained that a protected contact style is s:afer and more appropriate than hand..~-on. For the..~e reason..~. it \'\ ill be important for re..~earch to continue to in"'estig.ate the behaviors displayed by these aniinals d"uring. tours, and how the safety of both hujnans and animals 1nig.ht be preserved or co1npromised. Conc-lusions Fen1dandez et al. (2009) argue that those interactions which meet the entertainjnent and ed"ucation aim.~ of visitors, \'\<h il~t also being. helpful for aniinal \'\ elf are, should be encouraged. 1\ lthough more research is 1:ieeded to detennine whether visitors benefit from interacting wilh big cats, lhe results of lhis stu.dy provide a starting point for understanding how participation in interactive zoo visitor to"urs 1night influence ani1nal behavior. The paper also supports the continued use of behavioral monitoring. as an effective 1neans to un.derstand animal welfare. Difference.~ in proportions of behaviors - both species-typical and stereotypic. \'\ ere obser\ ed l OI 384

403 bet\'\ een the conditjons in this study; ho\'\ e\fer. due to the multjple confounding variables inherent in this type of research, it is not po.1;sible to draw conclusion.s about how animal..-.. elf are might be influenced by these tours. L1 particular. it is important to ackno\'\ ledge that it cannot be detennined, fro1n this in\'estigation alone. which a.1;pect.1; of the tours ate responsible for the changes observed. This \'\ ill need to be explored in future research. ~1oroo\fer. since the presence of stereotypic behavior can be indicative of co1npromised..-.. ejj-being, it will be of iinportance for future reseatch to add.re~ thjs in order to support the continued in., olvement of big cats in interactive progra1n..1;. flowever, \'\ e do suggest that the o.. erall conditions under whjch these tours oocur - such as predjctability. allowing the aniinals choice to participate, the u..1;e of positi., e reinforcement, and the provision of retreat spaces - appeat quite po..1;itive and should be encouraged el..1;ewhere. There is clearly a need for 1nore research in this area and it i..1; hoped lhat this st\ldy will pro1note the in., es.tigation of big cats in\'ol\'ed in interactive tours at other 2oolog.ical facilities. Reference.~ Anderson. U. S. Benne. ~1. Bloomsmith. ~1. A.. & ~1 aple. r. L. (2002). Retreat space and human visitor density modc:-rate undesirable bc:-havior in petting zoo animals. Journal of Applied Anin1a{ n:elfare Science BalJantyne, R. & Packer, J. (2005). Promoting c:.nvironmentally sustainable ailiiudes and behaviour through free-choice learning experiences: \Vhat is the state of the game'! E.'nviron,nenral Educa1ion Research. 1I Ba.shaw. ~1. J. Bloomsmi!h, r..t. A. r..tm. r..t. J., & ~1aple. r. L. (2()03). ro hunt or not to hunt:> A feeding e nrichment experimoot with c-.apti\'c large (elids. Zoo Biology, 22, Ba.shaw, r..t. J.. Kelling. A. S.. Bloomsmitlt, ~1. A. & ~1aple. 'I'. L. (2007). Environmental effect-. on the bc:-havior of wo-hooscd lions and tigers, '>"ith a case study or the c:ffec.ts of a \'isual barric-r on pac.ing. Applied Anin1a{ Behaviour Science. /(), Basscu, L.. & Buchanan-Smitlt, H. ~1. (2007). Effects of predictability on the "''Clfare of captive animajs. Applied Anintal Behaviour Science Beier, P. ( 1991 ). Cougar auac.ks on humans in the United Sta1es and Canada. n:ildlife Society Bulletin, 19, 40J-412. Brensing. K. & Linke. K. (2003). Bc-ha\'ior of dolphins towards adults and children during svd m witlt dolphin programs and to..., ards children witlt disabilities during therapy sessions. NuJ1roz-00.f, 16, J 15-JJI. Broad. S. & \Veiler. B. ( 1998). Capti, e IUlimals and intcrprctation A tale or t"''o tiger exhibits. Journal of '/'ouri.wn Studie.f, Carlste.ad, K. ( 1998). Determining the causes or stereotypic behavior in zoo carnivores: rowards appropriate e nrichment stra1egies. In D. J. Shc:pllerdson. J. 0. ~1ellen. & ~1. Hutc hins (Eds.), Second 11a1ure: f:nvironn1e111a { enrich,nen1 f or cr.p1h e anintals (pp ). \Vashington: Smithsonian Institution Press. Chapenoire, S. Camiade, B. & Legros, ~1. (2001). Ba.-.ie instinct in a feline. 1'he Anierican Jounial of / 1 oren.sic,'.fedicine and Pathology Claxton. A. ~1. (20 11 ). '('Ile potential or the human-animal relationship as an environmental enrichment (or the v.-el(are of zoo-housed IUlimals. Applied.4ni,nal Behaviour Sc ience. I 33. I JO. CohJe. S. 0. Harlan, C. W., & Harlan, C. ( 19'>0). Fatal big cat a!tacks. 'The A,nerican Journal of f "oren.f ic i"ledic ine and PatJ1ology, I Davey, G. (2005). '('he 0 \ isitor effect." Zoo.f ' Print, Davey, G. (2006). Visitor behavior in zoos: A review. Anrhrozoo.f, 19, Davey, C. (2007). Visitors' e ffects on the v.-elfarc of animals in the zoo: A review. Journal of Applied Anin1a{ \Ve/fare Science. JO, 16'J- J83.. l

404 386

405 387

406 Appendix J Pooled data for analysis in Chapter Five This section contains graphical representations of mean and standard deviation proportions of active species-typical, inactive species-typical, and pacing behaviour, as well as estimated proximities to humans, for individual lions (n = 3) and cheetahs (n = 3) to confirm pooling of data for analysis in Chapter Five. 388

407 Mujambi Yizi Amani Mujambi Yizi Amani Mujambi Yizi Amani Mean % of scans Mujambi Yizi Amani Mujambi Yizi Amani Mujambi Yizi Amani Mean % of scans Mujambi Yizi Amani Mujambi Yizi Amani Mujambi Yizi Amani Mean % of scans Non-tour Tour Pre During Post Active species-typical behaviour Non-tour Tour Pre During Post Inactive species-typical behaviour Non-tour Tour Pre During Post Pacing behaviour Figure J.1. Mean percentage of scans in which each individual lion (Mujambi, Yizi, Amani) engaged in active species-typical (top), inactive species-typical (middle), and pacing (bottom) behaviour in each condition (pre, during, post) on non-tour and tour days [Note: Bars indicate standard deviation] 389

408 Skukuza Tsostie Askari Skukuza Tsostie Askari Skukuza Tsostie Askari Mean % of scans Skukuza Tsotsie Askari Skukuza Tsotsie Askari Skukuza Tsotsie Askari Mean % of scans Skukuza Tsotsie Askari Skukuza Tsotsie Askari Skukuza Tsotsie Askari Mean % of scans Non-tour Tour Pre During Post Active species-typical behaviour Non-tour Tour Pre During Post Inactive species-typical behaviour Non-tour Tour Pre During Post Pacing behaviour Figure J.2. Mean percentage of scans in which each individual cheetah (Skukuza, Tsotsie, Askari) engaged in active species-typical (top), inactive species-typical (middle), and pacing (bottom) behaviour in each condition (pre, during, post) on non-tour and tour days [Note: Bars indicate standard deviation] 390

409 Skukuza Tsotsie Askari Skukuza Tsotsie Askari Skukuza Tsotsie Askari Mean % of scans Mujambi Yizi Amani Mujambi Yizi Amani Mujambi Yizi Amani Mean % of scans Keepers Visitors 0 Close Moderate Distant Figure J.3. Mean percentage of scans in which each individual lion (Mujambi, Yizi, Amani) was observed in each proximity category (close, moderate, distant) to humans (keepers, visitors) [Note: Bars indicate standard deviation] Keepers Visitors 20 0 Close Moderate Distant Figure J.4. Mean percentage of scans in which each individual cheetah (Skukuza, Tsotsie, Askari) was observed in each proximity category (close, moderate, distant) to humans (keepers, visitors) [Note: Bars indicate standard deviation] 391

410 Appendix K Survey materials used in Chapter Six This section includes the information sheet, consent form, and pre-experience and postexperience questionnaires used in Chapter Six. Note that this study involved different species and visitor experiences, however only an example of one of these is included in this section (to avoid unnecessary repetition of materials). Questionnaires for all experiences were identical, except where the name of species (i.e., tiger, lion or cheetah) and experience type (i.e., exhibit or behind-the-scenes tour) needed to be changed. 392

411 PARTICIPANT INFORMATION SHEET Project title: The impact of different big cat experiences on zoo visitors Researcher s name: Monika Szokalski Supervisors names: Dr Carla Litchfield and Dr Wendy Foster Contact telephone number: Monika: (08) 8302 XXXX; Carla: (08) 8302 XXXX Monika.Szokalski@unisa.edu.au School of Psychology, Social Work, and Social Policy Division of Education, Arts, and Social Sciences University of South Australia Dear participant, This study is designed to explore the impact of different captive big cat experiences on zoo visitors. I am inviting you participate since I feel there is a great need to evaluate the experiences that visitors like yourself are having. I am particularly interested in the experience of visitors before and after different types of captive big experiences, namely visiting a traditional exhibit or participating in a more interactive experience, such as going on a behind-the-scenes tour. Participation in the current study will involve completion of a brief questionnaire before/after your visit to the exhibit/behind-the-scenes tour today. Participation in the current study: will be anonymous is completely voluntary and you are free to withdraw at any time will not involve any changes to the experiences you engage in at the zoo today, and not giving consent to participate will not in any way affect your experience at the zoo today. The information collected as part of the current study will be retained for five years, stored in a secure place in the School of Psychology office, Magill Campus, at the University of South Australia. This data will be non-identifiable, as you will remain anonymous in the data collection. For any further questions about the present study, please contact myself or my supervisors (see contact details above). This project has been approved by the University of South Australia s Human Research Ethics Committee. If you have any ethical concerns about the project or questions about your rights as a participant, please contact the Executive Officer of this Committee, Tel: XXXX; Vicki.XXXX@unisa.edu.au. If you would like a copy or a summary of the research findings feel free to contact me. Thank you for your participation in this study. Regards, Monika Szokalski 393

412 PARTICIPANT CONSENT FORM Project title: A study of tiger behaviour at Australia Zoo. Researcher s name & contact details: Monika Szokalski Monika.Szokalski@unisa.edu.au or (08) 8302 XXXX Supervisors names: Dr Carla Litchfield (UniSA) and Dr Wendy Foster (Zoos SA) I have read the Participant Information Sheet and the nature and purpose of the research project has been explained to me. I understand and agree to take part. I understand the purpose of the research and my involvement in it. I understand that I may withdraw from the research project at any stage and that this will not affect my status now or in the future. I understand that while information gained during the study may be published, I will not be identified and my personal results will remain confidential. Name of participant: Signed: Date: I have provided information about the research to the participant and believe that he/she understands what is involved. Researcher s signature and date: This project has been approved by the University of South Australia s Human Research Ethics Committee. If you have any ethical concerns about the project or questions about your rights as a participant please contact the Executive Officer of this Committee, Tel: XXXX; Vicki.XXXX@unisa.edu.au. 394

Stereotypic pacing in Siberian tiger (Panthera tigris altaica) and the effect of interspecies presence: A case study

Stereotypic pacing in Siberian tiger (Panthera tigris altaica) and the effect of interspecies presence: A case study Stereotypic pacing in Siberian tiger (Panthera tigris altaica) and the effect of interspecies presence: A case study Table of Contents List of Figures... 2 List of Tables... 3 Abstract... 4 1.0 Introduction...

More information

Kathleen Krafte, Lincoln Larson, Robert Powell Clemson University ISSRM: June 14, 2015

Kathleen Krafte, Lincoln Larson, Robert Powell Clemson University ISSRM: June 14, 2015 Kathleen Krafte, Lincoln Larson, Robert Powell Clemson University ISSRM: June 14, 2015 Big cats are keystone carnivore species that drive the structure and function of biological communities in diverse

More information

Welcome to the Animal Ambassador Program from IFAW!

Welcome to the Animal Ambassador Program from IFAW! Welcome to the Animal Ambassador Program from IFAW! This month s featured animals: big cats A tiger s stripes are like a human s fingerprint; the pattern is unique to every tiger. These distinctive stripes

More information

What is a tiger? Tigers are felids (members of the cat family). They are in the genus Panthera.

What is a tiger? Tigers are felids (members of the cat family). They are in the genus Panthera. CONTENTS Page What is a tiger?... 3 Tiger Distribution... 4 Tiger Habitat... 5 Key Tiger Facts... 6 Threats to Tigers... 7 Tiger Conservation... 8 Why Conserve Tigers?... 9 Tiger Activity Sheets... 10-13

More information

Introduction to the Cheetah

Introduction to the Cheetah Lesson Plan 1 Introduction to the Cheetah CRITICAL OUTCOMES CO #1: Identify and solve problems and make decisions using critical and creative thinking. CO #2: Work effectively with others as members of

More information

ORIGINAL RESEARCH PAPER

ORIGINAL RESEARCH PAPER 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 ORIGINAL RESEARCH PAPER TIGERS STEREOTYPIC PACING AND ENRICHMENT Ivana

More information

Role of Temperature and Shade Coverage on Behavior and Habitat Use of Captive African Lions, Snow Leopards, and Cougars

Role of Temperature and Shade Coverage on Behavior and Habitat Use of Captive African Lions, Snow Leopards, and Cougars Xavier Journal of Undergraduate Research Volume 4 Article 7 2016 Role of Temperature and Shade Coverage on Behavior and Habitat Use of Captive African Lions, Snow Leopards, and Cougars Caitlin Mack Follow

More information

Effects of interactive visitor encounters on the behaviour and welfare of animals commonly housed in Australian zoos

Effects of interactive visitor encounters on the behaviour and welfare of animals commonly housed in Australian zoos Effects of interactive visitor encounters on the behaviour and welfare of animals commonly housed in Australian zoos www.animalwelfare.net.au Lydia Acaralp-Rehnberg PhD candidate Supervisors: Dr Ian Bland,

More information

Your web browser (Safari 7) is out of date. For more security, comfort and the best experience on this site: Update your browser Ignore

Your web browser (Safari 7) is out of date. For more security, comfort and the best experience on this site: Update your browser Ignore Your web browser (Safari 7) is out of date. For more security, comfort and the best experience on this site: Update your browser Ignore Activityengage CO NSERVATIO N AND BIG CATS What problems threaten

More information

The Forgotten Taxa : Past, present and future directions in zoo research. Dr Sonya P. Hill

The Forgotten Taxa : Past, present and future directions in zoo research. Dr Sonya P. Hill The Forgotten Taxa : Past, present and future directions in zoo research Dr Sonya P. Hill Department of Biological Sciences, University of Chester hill.s@chester.ac.uk @DrSonyaPHill Happy 20 th Anniversary,

More information

Social and Reproductive Behaviours in the Cheetah (Acinonyx jubatus) in A Captive Population

Social and Reproductive Behaviours in the Cheetah (Acinonyx jubatus) in A Captive Population Social and Reproductive Behaviours in the Cheetah (Acinonyx jubatus) in A Captive Population Rebecca Bradford-Wright B.A. Hons A THESIS SUBMITTED FOR THE DEGREE DOCTOR OF PHILOSOPHY School of Biological

More information

110th CONGRESS 1st Session H. R. 1464

110th CONGRESS 1st Session H. R. 1464 HR 1464 IH 110th CONGRESS 1st Session H. R. 1464 To assist in the conservation of rare felids and rare canids by supporting and providing financial resources for the conservation programs of nations within

More information

Tigers Stereotypic Pacing and Enrichment

Tigers Stereotypic Pacing and Enrichment Annual Research & Review in Biology 4(10): 1544-1550, 2014 SCIENCEDOMAIN international www.sciencedomain.org Tigers Stereotypic Pacing and Enrichment Ivana Gardiánová 1* and Petra Bolechová 2,3 1 Demonstrational

More information

POSITION DESCRIPTION. Organisational Context: Important Functional Relationships: Page 1. Job Title: Reports To: Direct Reports: Position Purpose:

POSITION DESCRIPTION. Organisational Context: Important Functional Relationships: Page 1. Job Title: Reports To: Direct Reports: Position Purpose: Page 1 POSITION DESCRIPTION Job Title: Reports To: Direct Reports: Position Purpose: Keeper Level One, Two and /Animal Care Manager Nil A Zoo Keeper is responsible for providing quality animal husbandry

More information

Your web browser (Safari 7) is out of date. For more security, comfort and the best experience on this site: Update your browser Ignore

Your web browser (Safari 7) is out of date. For more security, comfort and the best experience on this site: Update your browser Ignore Your web browser (Safari 7) is out of date. For more security, comfort and the best experience on this site: Update your browser Ignore Activityengage BIG CATS IN CAPTIVITY What are the negative consequences

More information

BEHAVIOUR OF THE DOMESTIC DOG (Canis familiaris)

BEHAVIOUR OF THE DOMESTIC DOG (Canis familiaris) THE INFLUENCE OF CEREBRAL LATERALISATION ON THE BEHAVIOUR OF THE DOMESTIC DOG (Canis familiaris) A thesis submitted for the Degree of DOCTOR OF PHILOSOPHY by Luke Aaron Schneider B. A. (Hons) School of

More information

Required and Recommended Supporting Information for IUCN Red List Assessments

Required and Recommended Supporting Information for IUCN Red List Assessments Required and Recommended Supporting Information for IUCN Red List Assessments This is Annex 1 of the Rules of Procedure for IUCN Red List Assessments 2017 2020 as approved by the IUCN SSC Steering Committee

More information

Ruppell s Griffon Vulture

Ruppell s Griffon Vulture Species Status IUCN: Critically Endangered ESA Status: Not Listed CITES: Appendix II TAG: Raptor TAG AZA SSP DESIGNATION: Yellow GEOGRAPHIC REGION: Africa BIOME: Savanna EXHIBIT DESIGN AND MANAGEMENT HUSBANDRY

More information

ISSN CAT news. N 63 Spring 2016

ISSN CAT news. N 63 Spring 2016 ISSN 1027-2992 CAT news N 63 Spring 2016 02 CATnews is the newsletter of the Cat Specialist Group, a component of the Species Survival Commission SSC of the International Union for Conservation of Nature

More information

Key considerations in the breeding of macaques and marmosets for scientific purposes

Key considerations in the breeding of macaques and marmosets for scientific purposes Key considerations in the breeding of macaques and marmosets for scientific purposes Key considerations in the breeding of macaques and marmosets for scientific purposes Laboratory Animal Science Association

More information

Half Yearly Examination for Primary Schools Year 5 ENGLISH (Listening Comprehension) Time: 30 minutes. Teacher s copy

Half Yearly Examination for Primary Schools Year 5 ENGLISH (Listening Comprehension) Time: 30 minutes. Teacher s copy Half Yearly Examination for Primary Schools 2017 Year 5 ENGLISH (Listening Comprehension) Time: 30 minutes Teacher s copy Guidelines for the conduct of the Listening Comprehension Examination ALL INSTRUCTIONS

More information

Public consultation on Proposed Revision of the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes 2004

Public consultation on Proposed Revision of the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes 2004 RESEARCH INTEGRITY Animal Ethics Committee Web: http://sydney.edu.au/research_support/ethics Project Officer Australian code of practice for the care and use of animals for scientific purposes Health and

More information

Social media kit for World Wildlife Day 2018

Social media kit for World Wildlife Day 2018 Social media kit for World Wildlife Day 2018 Date: 3 March 2018 Theme: Big Cats predators under threat Main hashtags: #WorldWildlifeDay, #PredatorsUnderThreat, #WWD2018 Supplementary hashtags: #DoOneThingToday

More information

Supporting Information

Supporting Information Supporting Information Table S1. Sources of the historic range maps used in our analysis. Elevation limits (lower and upper) are in meters. Modifications to the source maps are listed in the footnotes.

More information

Københavns Universitet. Companion animal ethics Sandøe, Peter; Corr, Sandra; Palmer, Clare. Published in: Luentokokoelma Publication date: 2013

Københavns Universitet. Companion animal ethics Sandøe, Peter; Corr, Sandra; Palmer, Clare. Published in: Luentokokoelma Publication date: 2013 university of copenhagen Københavns Universitet Companion animal ethics Sandøe, Peter; Corr, Sandra; Palmer, Clare Published in: Luentokokoelma 2013 Publication date: 2013 Citation for published version

More information

Endangered Species: The cheetah

Endangered Species: The cheetah Endangered Species: The cheetah By Gale, Cengage Learning, adapted by Newsela staff on 01.05.18 Word Count 626 Level MAX Image 1: Cheetahs are famous for their round, black spots, which help them to hide

More information

Conflict-Related Aggression

Conflict-Related Aggression Conflict-Related Aggression and other problems In the past many cases of aggression towards owners and also a variety of other problem behaviours, such as lack of responsiveness to commands, excessive

More information

Malayan Tiger Updated: April 8, 2018

Malayan Tiger Updated: April 8, 2018 Malayan Tiger Updated: April 8, 2018 Interpretation Guide Status Danger Threats SSP Yellow Critically Endangered (IUCN Red List) Their main threat to habitat loss is deforestation due to palm oil plantation

More information

Animal Welfare Assessment Transfers Checklist

Animal Welfare Assessment Transfers Checklist Animal Welfare Assessment Transfers Checklist Our Animal Welfare Commitment The believes that consideration of an animal s welfare must include its physical, physiological and mental state and that good

More information

The Kennel Club has long campaigned for a ban on the use and sale of electric shock collars in Scotland.

The Kennel Club has long campaigned for a ban on the use and sale of electric shock collars in Scotland. PE1555/E The Kennel Club and Scottish Kennel Club response to the Public Petitions Committee - PE1555 (electric shock and vibration collars for animals) call for evidence Summary PLEASE NOTE THAT FOR THE

More information

MSc in Veterinary Education

MSc in Veterinary Education MSc in Veterinary Education The LIVE Centre is a globally unique powerhouse for research and development in veterinary education. As its name suggests, its vision is a fundamental transformation of the

More information

Zurich Open Repository and Archive. Feeding live prey to zoo animals: response of zoo visitors in Switzerland

Zurich Open Repository and Archive. Feeding live prey to zoo animals: response of zoo visitors in Switzerland University of Zurich Zurich Open Repository and Archive Winterthurerstr. 190 CH-8057 Zurich http://www.zora.uzh.ch Year: 2010 Feeding live prey to zoo animals: response of zoo visitors in Switzerland Cottle,

More information

Meredith J. Bashaw a, Angela S. Kelling b, Mollie A. Bloomsmith b & Terry L. Maple b a TECHlab, Zoo Atlanta, and Center for

Meredith J. Bashaw a, Angela S. Kelling b, Mollie A. Bloomsmith b & Terry L. Maple b a TECHlab, Zoo Atlanta, and Center for This article was downloaded by: [Dr Kenneth Shapiro] On: 09 June 2015, At: 07:05 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer

More information

Development of Council of Europe Conventions for Protection of Animals - ethics, democratic processes, and monitoring

Development of Council of Europe Conventions for Protection of Animals - ethics, democratic processes, and monitoring Joint COE EU TAIEX OIE WORKSHOP Development of Council of Europe Conventions for Protection of Animals - ethics, democratic processes, and monitoring David G. PRITCHARD Chairman T-AP Standing Committee

More information

international news RECOMMENDATIONS

international news RECOMMENDATIONS The Third OIE Global Conference on Veterinary Education and the Role of the Veterinary Statutory Body was held in Foz do Iguaçu (Brazil) from 4 to 6 December 2013. The Conference addressed the need for

More information

Surveys of the Street and Private Dog Population: Kalhaar Bungalows, Gujarat India

Surveys of the Street and Private Dog Population: Kalhaar Bungalows, Gujarat India The Humane Society Institute for Science and Policy Animal Studies Repository 11-2017 Surveys of the Street and Private Dog Population: Kalhaar Bungalows, Gujarat India Tamara Kartal Humane Society International

More information

February 1, 2018 Robert Gibbens, D.V.M. Director, Animal Welfare Operations USDA/APHIS/Animal Care

February 1, 2018 Robert Gibbens, D.V.M. Director, Animal Welfare Operations USDA/APHIS/Animal Care February 1, 2018 Robert Gibbens, D.V.M. Director, Animal Welfare Operations USDA/APHIS/Animal Care Via e-mail: ; acwest@aphis.usda.gov Dear Dr. Gibbens: I am writing on behalf of PETA to request that the

More information

Unit 3 Sustainability and interdependence Sub Topic 3.4: Animal welfare

Unit 3 Sustainability and interdependence Sub Topic 3.4: Animal welfare Unit 3 Sustainability and interdependence Sub Topic 3.4: Animal welfare Page 1 of 12 On completion of this topic I will be able to: Describe the costs, benefits and ethics of providing different levels

More information

EXOTIC PETS The landscape has changed

EXOTIC PETS The landscape has changed EXOTIC PETS The landscape has changed Today smaller animals predominate Problems with exotic pets Animal welfare Species reduction and/or extinction, disruption of ecosystems Introduction of foreign diseases

More information

International Shelter Programme for Applied Canine Studies (ISPACS)

International Shelter Programme for Applied Canine Studies (ISPACS) Page 1 International Shelter Programme for Applied Canine Studies (ISPACS) Understanding Real-Life Skills Improving Staff Skills Positive practices Meeting the needs of Rescue Centre staff and dogs Improving

More information

King Fahd University of Petroleum & Minerals College of Industrial Management

King Fahd University of Petroleum & Minerals College of Industrial Management King Fahd University of Petroleum & Minerals College of Industrial Management CIM COOP PROGRAM POLICIES AND DELIVERABLES The CIM Cooperative Program (COOP) period is an essential and critical part of your

More information

Effective Vaccine Management Initiative

Effective Vaccine Management Initiative Effective Vaccine Management Initiative Background Version v1.7 Sep.2010 Effective Vaccine Management Initiative EVM setting a standard for the vaccine supply chain Contents 1. Background...3 2. VMA and

More information

UNIT Animal Care: Reptile and Amphibian Care (SCQF level 5)

UNIT Animal Care: Reptile and Amphibian Care (SCQF level 5) National Unit Specification: general information CODE F6SR 11 SUMMARY This Unit may be suitable for candidates who seek a basic knowledge of caring for reptiles and amphibians. OUTCOMES 1 Identify suitable

More information

Keywords: Acinonyx jubatus/breeding/captivity/cheetah/management/off-exhibit

Keywords: Acinonyx jubatus/breeding/captivity/cheetah/management/off-exhibit Frank, J. and Saffoe, C. (2005). Breeding management strategy for cheetahs (Acinonyx jubatus) at the Smithsonian's National Zoological Park. Animal Keeper's Forum 7/8: 393-397. Keywords: Acinonyx jubatus/breeding/captivity/cheetah/management/off-exhibit

More information

Distressed Animal Behaviors and Some Recommendations for Improvements at the Kuala Lumpur Zoo, Malaysia Amber Haque Published online: 04 Jun 2010.

Distressed Animal Behaviors and Some Recommendations for Improvements at the Kuala Lumpur Zoo, Malaysia Amber Haque Published online: 04 Jun 2010. This article was downloaded by: [Dr Kenneth Shapiro] On: 09 June 2015, At: 06:55 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer

More information

EFFECTIVE LEADERSHIP. Primary Qualities of An Effective Leader

EFFECTIVE LEADERSHIP. Primary Qualities of An Effective Leader EFFECTIVE LEADERSHIP A leader s responsibilities are tremendously diverse. CFRs and other puppy raising program staff are always happy to provide additional information and tools as needed. There are many

More information

Turtle Watch: Enhancing Science Engagement

Turtle Watch: Enhancing Science Engagement Turtle Watch: Enhancing Science Engagement Elaine Lewis, Louise Nielsen, Jennifer Pearson and Catherine Baudains Introduction Engaging students in authentic science settings and real-life contexts is viewed

More information

MODULE 3. What is conflict?

MODULE 3. What is conflict? This module incorporates the Human Wildlife Conflict Toolkit developed by BioHub with sponsorship from the FAO SADC Subregional office. The module focuses on conflict between humans and cheetah and wild

More information

Chapter 13 First Year Student Recruitment Survey

Chapter 13 First Year Student Recruitment Survey Chapter 13 First Year Student Recruitment Survey Table of Contents Introduction...... 3 Methodology.........4 Overall Findings from First Year Student Recruitment Survey.. 7 Respondent Profile......11

More information

STUDENT MANUAL CANINE SEARCH SPECIALIST TRAINING UNIT 8: ADVANCED RUBBLE SEARCH

STUDENT MANUAL CANINE SEARCH SPECIALIST TRAINING UNIT 8: ADVANCED RUBBLE SEARCH STUDENT MANUAL CANINE SEARCH SPECIALIST TRAINING UNIT 8: ADVANCED RUBBLE SEARCH Unit Objective Enabling Objectives Upon completion of this unit, you will be able to explain the rationale for canine foundation

More information

Overview LANCTB1. Observe, assess and respond to the behaviour of dogs. Observe, assess and respond to the behaviour of dogs

Overview LANCTB1. Observe, assess and respond to the behaviour of dogs. Observe, assess and respond to the behaviour of dogs Overview This standard covers observing, assessing and responding to the behaviour of dogs. The standard involves understanding the normal behaviour of the type of dog that you are working with and observing

More information

Comparison of Several Types of Enrichment for Captive Felids

Comparison of Several Types of Enrichment for Captive Felids Zoo Biology 26:371 381 (2007) RESEARCH ARTICLE Comparison of Several Types of Enrichment for Captive Felids Amy L. Skibiel, 1 Heather S. Trevino, 1 and Ken Naugher 2 1 Department of Biological Sciences,

More information

Assessment and Opinion of Health and Welfare of Animals at Spring River Park and Zoo Christine Capaldo, DVM March 9, 2017

Assessment and Opinion of Health and Welfare of Animals at Spring River Park and Zoo Christine Capaldo, DVM March 9, 2017 Assessment and Opinion of Health and Welfare of Animals at Spring River Park and Zoo Christine Capaldo, DVM March 9, 2017 Introduction: My name is Christine Capaldo, DVM and I am a veterinarian who recently

More information

About Animals Asia. Who we are. Our work

About Animals Asia. Who we are. Our work About Animals Asia Who we are Animals Asia is devoted to ending the barbaric practice of bear bile farming and improving the welfare of animals in China and Vietnam. We promote compassion and respect for

More information

Management of bold wolves

Management of bold wolves Policy Support Statements of the Large Carnivore Initiative for Europe (LCIE). Policy support statements are intended to provide a short indication of what the LCIE regards as being good management practice

More information

Promoting One Health : the international perspective OIE

Promoting One Health : the international perspective OIE Promoting One Health : the international perspective OIE Integrating Animal Health & Public Health: Antimicrobial Resistance SADC SPS Training Workshop (Animal Health) 29-31 January 2014 Gaborone, Botwana

More information

The IUCN Red List of Threatened Species

The IUCN Red List of Threatened Species The IUCN Red List of Threatened Species: Celebrating 50 years Background, lessons learned, and challenges David Allen Regional Biodiversity Assessment Officer, Global Species Programme, Cambridge The IUCN

More information

BIAZA Animal Transfer Policy (ATP)

BIAZA Animal Transfer Policy (ATP) vember 2014 British and Irish Association of Zoos and Aquariums BIAZA Animal Transfer Policy (ATP) Preamble: This document is to assist members to carry out due diligence in respect to sourcing or disposing

More information

Investing in Human Resources in Veterinary Services

Investing in Human Resources in Veterinary Services Investing in Human Resources in Veterinary Services 9 th Conference of Ministers responsible for Animal Resources in Africa Meeting of Experts Abidjan, Côte d Ivoire, 16-17 April 2013 Dr. Etienne Bonbon

More information

Surveillance. Mariano Ramos Chargé de Mission OIE Programmes Department

Surveillance. Mariano Ramos Chargé de Mission OIE Programmes Department Mariano Ramos Chargé de Mission OIE Programmes Department Surveillance Regional Table Top Exercise for Countries of Middle East and North Africa Tunisia; 11 13 July 2017 Agenda Key definitions and criteria

More information

Rare Asiatic Golden Cats are World-First Test Tube Babies

Rare Asiatic Golden Cats are World-First Test Tube Babies Rare Asiatic Golden Cats are World-First Test Tube Babies One of the Asiatic golden cat cubs at around 40 days old. Credit: Imke Lüders A pair of Asiatic golden cats have been bred using artificial insemination

More information

Keywords: 1ZA/Acinonyx jubatus/cheetah/hair sample/method/monitoring/scent

Keywords: 1ZA/Acinonyx jubatus/cheetah/hair sample/method/monitoring/scent Thomas, P., Balme, G., Hunter, L., and McCabe-Parodi, J. (2005). Using scent attractants to noninvasively collect hair samples from cheetahs, leopards and lions. Animal Keeper's Forum 7/8: 342-384. Keywords:

More information

Care For Us Arc$c Wolf (Canis lupus arctos)

Care For Us Arc$c Wolf (Canis lupus arctos) Care For Us Arc$c Wolf (Canis lupus arctos) Animal Welfare Animal welfare refers to an animal s state or feelings. An animal s welfare state can be positive, neutral or negative. An animal s welfare has

More information

Supplementary Materials for

Supplementary Materials for www.sciencemag.org/content/343/6167/1241484/suppl/dc1 Supplementary Materials for Status and Ecological Effects of the World s Largest Carnivores William J. Ripple,* James A. Estes, Robert L. Beschta,

More information

March 16, Guide's space recommendations as a minimum while always recognizing that performance standards also must be met.

March 16, Guide's space recommendations as a minimum while always recognizing that performance standards also must be met. Comments of The American Association of Immunologists (AAI) to the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC) Regarding the 8 th Edition of the Guide

More information

Mexican Gray Wolf Reintroduction

Mexican Gray Wolf Reintroduction Mexican Gray Wolf Reintroduction New Mexico Supercomputing Challenge Final Report April 2, 2014 Team Number 24 Centennial High School Team Members: Andrew Phillips Teacher: Ms. Hagaman Project Mentor:

More information

DHOLE PROTECTION GUIDE CREATED BY

DHOLE PROTECTION GUIDE CREATED BY DHOLE PROTECTION GUIDE CREATED BY INTRO In this presentation we are talking about the endangered species name Dhole which is a red dog that lives in the Middle East and India which there are only 2,500

More information

A-FAVP.1 Foundations of Advanced Veterinary Practice

A-FAVP.1 Foundations of Advanced Veterinary Practice A-FAVP.1 Foundations of Advanced Veterinary Practice Credits: Provider: 10 (100 hours) Veterinary Postgraduate Unit School of Veterinary Science RCVS Content Covered Conceptual Outline The scope of the

More information

Bobcat Interpretive Guide

Bobcat Interpretive Guide Interpretive Guide Exhibit Talking Point: Our job as interpreters is to link what the visitors are seeing to The Zoo's conservation education messages. Our goal is to spark curiosity, create emotional

More information

and suitability aspects of food control. CAC and the OIE have Food safety is an issue of increasing concern world wide and

and suitability aspects of food control. CAC and the OIE have Food safety is an issue of increasing concern world wide and forum Cooperation between the Codex Alimentarius Commission and the OIE on food safety throughout the food chain Information Document prepared by the OIE Working Group on Animal Production Food Safety

More information

The S Files Success with Maria: Sunshine: Biting Reported by S.G. Friedman, PhD and L. McGuire

The S Files Success with Maria: Sunshine: Biting Reported by S.G. Friedman, PhD and L. McGuire The S Files Success with Maria: Sunshine: Biting Reported by S.G. Friedman, PhD and L. McGuire In Press, Good Bird Magazine Volume x(x), pp-pp The S Files are real case studies of behavior challenges faced

More information

Presentation by Major General Peter Davies, Director General of WSPA, to the second OIE Global Conference on Animal Welfare. 21 st October 2008

Presentation by Major General Peter Davies, Director General of WSPA, to the second OIE Global Conference on Animal Welfare. 21 st October 2008 Presentation by Major General Peter Davies, Director General of WSPA, to the second OIE Global Conference on Animal Welfare. 21 st October 2008 Work of Non-Governmental Organisations supporting the implementation

More information

A1 Control of dangerous and menacing dogs (reviewed 04/01/15)

A1 Control of dangerous and menacing dogs (reviewed 04/01/15) A1 Control of dangerous and menacing dogs (reviewed 04/01/15) 1 Introduction 1.1 For as long as human beings continue to interact with dogs, there will be incidents of dog bites. However, the frequency

More information

The Dog and Cat Management Board. Policy and Procedure for the training of dogs subject to a dangerous dog order

The Dog and Cat Management Board. Policy and Procedure for the training of dogs subject to a dangerous dog order The Dog and Cat Management Board Policy and Procedure for the training of dogs subject to a dangerous dog order Description: A policy and procedure for the training of dogs subject to a dangerous dog order

More information

WORLD ANTIBIOTIC AWARENESS WEEK

WORLD ANTIBIOTIC AWARENESS WEEK # AntibioticResistance WORLD ANTIBIOTIC AWARENESS WEEK 14-20 NOVEMBER 2016 2016 CAMPAIGN TOOLKIT TABLE OF CONTENTS Why we need a global campagin... Campagin objectives... Key messages... Calls to action

More information

14th Conference of the OIE Regional Commission for Africa. Arusha (Tanzania), January 2001

14th Conference of the OIE Regional Commission for Africa. Arusha (Tanzania), January 2001 14th Conference of the OIE Regional Commission for Africa Arusha (Tanzania), 23-26 January 2001 Recommendation No. 1: The role of para-veterinarians and community based animal health workers in the delivery

More information

Primate Welfare Meeting

Primate Welfare Meeting Primate Welfare Meeting 15 November 2007 Advances in Primate Housing Agenda - NC3Rs Primate Welfare Meeting 2007 09.30 REGISTRATION and COFFEE 10.00 10.10 Welcome and introduction 10.10 10.30 New housing

More information

GUIDELINES FOR APPROPRIATE USES OF RED LIST DATA

GUIDELINES FOR APPROPRIATE USES OF RED LIST DATA GUIDELINES FOR APPROPRIATE USES OF RED LIST DATA The IUCN Red List of Threatened Species is the world s most comprehensive data resource on the status of species, containing information and status assessments

More information

Mental stim ulation it s not just for dogs!! By Danielle Middleton- Beck BSc hons, PGDip CABC

Mental stim ulation it s not just for dogs!! By Danielle Middleton- Beck BSc hons, PGDip CABC Milo, Congo African Grey by Elaine Henley Mental stim ulation it s not just for dogs!! By Danielle Middleton- Beck BSc hons, PGDip CABC Dexter, Green Iguana by Danielle Middleton-Beck Exotic pets include

More information

Dog Off Leash Strategy

Dog Off Leash Strategy STRATHCONA COUNTY Dog Off Leash Strategy Phase 2 Report: Consultation Summary December 03, 2014 ENCLOSURE 4 STRATHCONA COUNTY Dog Off Leash Strategy Phase 2 Report: Consultation Summary ENCLOSURE 4 Table

More information

Dog Behavior and Training - Moving with Your Dog

Dog Behavior and Training - Moving with Your Dog Kingsbrook Animal Hospital 5322 New Design Road, Frederick, MD, 21703 Phone: (301) 631-6900 Website: KingsbrookVet.com Dog Behavior and Training - Moving with Your Dog Our family is moving. Should I be

More information

Is It a Cheetah? By Stephanie S. Tolan Stephanie S. Tolan

Is It a Cheetah? By Stephanie S. Tolan Stephanie S. Tolan Is It a Cheetah? By Stephanie S. Tolan 1996 Stephanie S. Tolan It's a tough time to raise, teach or be a highly gifted child. As the term "gifted" and the unusual intellectual capacity to which that term

More information

Global Communication on AMR in Animal Health: Tripartite and OIE Efforts

Global Communication on AMR in Animal Health: Tripartite and OIE Efforts Catherine Bertrand-Ferrandis Head of the Communication Unit Taylor Gabourie AMR Communications Officer Global Communication on AMR in Animal Health: Tripartite and OIE Efforts Marrakech, Morocco 29 31

More information

Table of Contents. A Pet Parenting Guide Page 5

Table of Contents. A Pet Parenting Guide Page 5 Table of Contents Chapter I Wellness Care for Your Puppy s Behavioral Health 11 What is Behavior Wellness? Characteristics of Behaviorally Healthy Dogs Breed and Individual Differences : A Blueprint for

More information

lasting compassion and

lasting compassion and Approved by the Board 26 June 2015 Po lasting compassion and DATE UPDATED POLICY HOLDER NEXT REVIEW DATE JUNE 2015 SENIOR WELFARE ADVISOR JUNE 2017 PURPOSE The Animal Welfare Policy describes the standard

More information

Mark Scheme. November Functional Skills English. Reading Level 2 E202

Mark Scheme. November Functional Skills English. Reading Level 2 E202 Mark Scheme November 2015 Functional Skills English Reading Level 2 E202 Edexcel and BTEC Qualifications Edexcel and BTEC qualifications are awarded by Pearson, the UK s largest awarding body. We provide

More information

"Education is not the filling of pail, but the lighting of a fire." WILLIAM BUTLER YEATS

Education is not the filling of pail, but the lighting of a fire. WILLIAM BUTLER YEATS TEACHERSGUIDE WINNIPEG HUMANE SOCIETY The Winnipeg Humane Society is a charitable organization dedicated to the protection, welfare and dignity of all animals. We offer educational programs from preschool

More information

DECLARATION of the First Conference on Animal Welfare in the Baltic Region RESPONSIBLE OWNERSHIP 5 to 6 May, 2011, Vilnius, Lithuania

DECLARATION of the First Conference on Animal Welfare in the Baltic Region RESPONSIBLE OWNERSHIP 5 to 6 May, 2011, Vilnius, Lithuania DECLARATION of the First Conference on Animal Welfare in the Baltic Region RESPONSIBLE OWNERSHIP 5 to 6 May, 2011, Vilnius, Lithuania Animal welfare is a complex and multi-faceted issue with an impact

More information

STUDENT MANUAL CANINE SEARCH SPECIALIST TRAINING UNIT 3: ROLE OF THE HELPER

STUDENT MANUAL CANINE SEARCH SPECIALIST TRAINING UNIT 3: ROLE OF THE HELPER STUDENT MANUAL CANINE SEARCH SPECIALIST TRAINING UNIT 3: ROLE OF THE HELPER Unit Objective Enabling Objectives Upon completion of this unit, you will be able to describe the function of the helper. You

More information

3 rd International Conference of Ecosystems (ICE2013) Tirana, Albania, May 31 - June 5, 2013

3 rd International Conference of Ecosystems (ICE2013) Tirana, Albania, May 31 - June 5, 2013 3 rd International Conference of Ecosystems (ICE2013) Tirana, Albania, May 31 - June 5, 2013 ANIMAL WELFARE IN ALBANIA Prof. As. Dr. Ylli Biçoku* * Agricultural University of Tirana, Tirane, Albania Corresponding

More information

Policy on Animals on Campus

Policy on Animals on Campus Policy on Animals on Campus Owner: Reviewed by: Deputy University Secretary Health and Safety Committee (October 2018) Appendix A Health and Safety Committee (May 2018) Approved by: Council (June 2018)

More information

KCAI Scheme Online Assessments: Criteria

KCAI Scheme Online Assessments: Criteria 1 Welfare, care and needs of the dog This assessment tests your knowledge and understanding of canine welfare, care and needs of dogs in relation to their life stages from puppy to adult. the physical

More information

AFRICAN CATS AT DISNEY S ANIMAL KINGDOM THEME PARK. Educator s Background Information

AFRICAN CATS AT DISNEY S ANIMAL KINGDOM THEME PARK. Educator s Background Information EXTRA FEATURE: CARING FOR AFRICAN CATS AT DISNEY S ANIMAL KINGDOM THEME PARK Educator s Background Information Lions and cheetahs are among the hundreds of animals that call Disney s Animal Kingdom Theme

More information

Professor David J Mellor Professor Kevin J Stafford Co-Directors

Professor David J Mellor Professor Kevin J Stafford Co-Directors Professor David J Mellor Professor Kevin J Stafford Co-Directors Collaborating Centre for Animal Welfare Science and Bioethical Analysis: Founding Partner http://animalwelfare.massey.ac.nz Evolving Veterinary

More information

of Conferences of OIE Regional Commissions organised since 1 June 2013 endorsed by the Assembly of the OIE on 29 May 2014

of Conferences of OIE Regional Commissions organised since 1 June 2013 endorsed by the Assembly of the OIE on 29 May 2014 of Conferences of OIE Regional Commissions organised since 1 June 2013 endorsed by the Assembly of the OIE on 29 May 2014 2 12 th Conference of the OIE Regional Commission for the Middle East Amman (Jordan),

More information

2017 ANIMAL SHELTER STATISTICS

2017 ANIMAL SHELTER STATISTICS 2017 ANIMAL SHELTER STATISTICS INTRODUCTION Dogs and cats are by far Canada s most popular companion animals. In 2017, there were an estimated 7.4 million owned dogs and 9.3 million owned cats living in

More information

NATIONAL CODE OF PRACTICE

NATIONAL CODE OF PRACTICE NATIONAL CODE OF PRACTICE Version 3 February 2017 Table of Contents PREFACE... 3 INTRODUCTION... 3 VISION... 4 MEMBER CODE OF ETHICS... 5 WHAT DO PET INDUSTRY ASSOCIATION OF AUSTRALIA MEMBERS BELIEVE?...

More information

Aquarist. Jobs at an Aquarium

Aquarist. Jobs at an Aquarium Aquarist The primary responsibility of an Aquarist is to care for the fish and invertebrates living in the many exhibits throughout the Aquarium. This includes feeding the animals and maintaining their

More information

Mexican Gray Wolf Endangered Population Modeling in the Blue Range Wolf Recovery Area

Mexican Gray Wolf Endangered Population Modeling in the Blue Range Wolf Recovery Area Mexican Gray Wolf Endangered Population Modeling in the Blue Range Wolf Recovery Area New Mexico Super Computing Challenge Final Report April 3, 2012 Team 61 Little Earth School Team Members: Busayo Bird

More information

The Big Cats At The Sharjah Breeding Centre Answers Key

The Big Cats At The Sharjah Breeding Centre Answers Key The Big Cats At The Sharjah Breeding Centre Answers Key We have made it easy for you to find a PDF Ebooks without any digging. And by having access to our ebooks online or by storing it on your computer,

More information