Social dispersal by domestic chicks in a novel environment: reassuring properties of a familiar odourant

Transcription

1 ANIMAL BEHAVIOUR, 2002, 63, doi: /anbe , available online at on Social dispersal by domestic chicks in a novel environment: reassuring properties of a familiar odourant R. BRYAN JONES*, LUCILLA FACCHIN & CAROLINE McCORQUODALE* *Roslin Institute, Edinburgh Dipartimento di Psicologia Generale, Università di Padova (Received 4 April 2001; initial acceptance 3 May 2001; final acceptance 9 October 2001; MS. number: 6890R) It may be possible to exploit olfactory attachments shown by chickens, Gallus gallus domesticus, to improve their welfare. In the present study, chicks were housed in groups of 15 in wooden boxes from 1 day of age. Experiment 1 compared fear responses in pairs of 8 10-day-old chicks with no previous experience of vanillin when they were exposed to an unfamiliar open field containing a dish of food in either the presence or absence of this odourant. The vanillin had no detectable effects. Chicks were housed similarly in experiment 2 but this time dishes containing vanillin were placed underneath the wire floor. We then asked if its presence would increase social dispersal and reduce fear when pairs of chicks were tested in the open field. At 8 10 days of age, two cagemates were placed close together in an open field in the presence of either the familiar odourant or a colour-matched solution of odourless food dyes (control). When the open field contained vanillin the chicks moved apart significantly sooner (minimum criterion=20 cm) and also tended to pace, preen and peck at the environment more often, although not significantly so; the proportions of pairs that moved apart and that fed were significantly greater. Novelty elicits fear and frightened chicks would be expected to move apart and to feed only when their fear levels had dissipated sufficiently. Our results confirm the existence of olfactory memories in domestic chicks, they suggest that behavioural modification reflected the familiarity of vanillin rather than any anxiolytic properties of this odourant per se, and they support our hypothesis that familiar odourants can act as reassuring agents in otherwise unfamiliar situations. These findings may have important implications for poultry welfare and productivity. Despite their possession of a moderately well-developed olfactory system (Bang 1971; Rogers 1995), the likelihood that the domestic fowl, Gallus gallus domesticus, could regulate its behaviour in response to olfactory cues was generally dismissed or neglected for many years. Indeed, their sense of smell was either thought to be totally lacking or very poor (Wood-Gush 1970; Fischer 1975; Kare & Rogers 1976). However, there is now compelling behavioural and neurobiological evidence that chickens can respond to odours in a variety of functional contexts (Rogers 1995; Jones & Roper 1997). For the sake of brevity we focus only on behavioural findings here. First, odours associated with predators, blood or specific aversants, such as methyl anthranilate, elicit marked alarm, avoidance and/or disgust responses (Jones & Black 1979; Fluck et al. 1996; Marples & Roper 1997). Second, treatment of Correspondence: R. B. Jones, Welfare Biology Group, Roslin Institute, Edinburgh, Roslin, Midlothian EH25 9PS, U.K. ( bryan.jones@bbsrc.ac.uk). L. Facchin is at the Dipartimento di Psicologia Generale, Università di Padova, Via Venezia 8, Padova, Italy The Association for the Study of Animal Behaviour. Published by Elsevier Science Ltd. All rights reserved. familiar food with an unfamiliar odourant induces temporary neophobia in young chicks (Jones 1987a; Turro et al. 1994; Marples & Roper 1996). Third, chickens can use odours as discriminative stimuli in a number of experimental paradigms, including operant conditioning (Stattelman et al. 1975; Porter et al. 1995; Roper & Marples 1997). Fourth, high concentrations of otherwise attractive odourants act as deterrents (Burne & Rogers 1996). Finally, odourants with which chicks have been reared are attractive to them when they are subsequently presented in otherwise unfamiliar environments (Jones & Faure 1982; Jones & Gentle 1985; Turro-Vincent 1994; Vallortigara & Andrew 1994; Burne & Rogers 1995; Jones & Carmichael 1999). The catalogue of familiar odourants used in these studies includes clove oil, geraniol, limonene and vanillin as well as those associated with soiled substrate taken from the chicks home cage or with nest litter. Such attraction to familiar odours generalizes to include chicks from four breeding strains as well as diverse methods of odour presentation, for example, on wood shavings, in /02/$35.00/ The Association for the Study of Animal Behaviour. Published by Elsevier Science Ltd. All rights reserved.

2 660 ANIMAL BEHAVIOUR, 63, 4 dishes or in perforated tubes. Collectively, these findings strongly suggest that domestic chicks are capable of forming olfactory memories and attachments. Furthermore, chicken embryos can respond to odourants once the beak has penetrated the air space (Tolhurst & Vince 1976) and chemosensory learning in the chicken embryo was demonstrated when chicks that had been exposed to strawberry before hatching subsequently spent more time in a strawberry-scented area than those with no such experience (Sneddon et al. 1998). A previous report (Jones & Gentle 1985) that the presence of a familiar odourant decreased fear-induced behavioural inhibition when chicks were tested individually in an open field (novel environment) suggests that familiar odours might also provide reassurance in unfamiliar situations. The latter hypothesis could have practical implications for animal husbandry and welfare. We therefore re-examined it in the present study using a new and more rigorous index of fear reduction (see below). Open-field behaviour, at least in birds, is thought to represent a compromise between two opposing tendencies, namely social reinstatement and predator evasion (Gallup & Suarez 1980; Faure et al. 1983; Jones 1989, 1996). Sociality (the motivation to be near to conspecifics) in a vast number of species must represent, at least in part, a response that confers some protection against predators (Pulliam & Caraco 1984; Yaber & Herrera 1994; Lima 1995). This hypothesis is supported by reports that chickens and other birds are strongly attracted to familiar visual and auditory stimuli in otherwise novel surroundings (Salzen 1979; Jones 1987b; Bolhuis 1991), that they work hard to reinstate and/or maintain close social contact in unfamiliar test situations, such as runways and T mazes (Suarez & Gallup 1983; Vallortigara et al. 1990; Jones et al. 1999), that social reinstatement is increased by prior exposure to a frightening event (Marin et al. 2001), that domestic chicks and quail, Coturnix japonica, tend to stay close together in novel areas (Launay et al. 1991; personal observations), that the presence of familiar companions reduces fear and distress (Hogan & Abel 1971; Jones 1983; Jones & Merry 1988), and that fear is inversely related to the likelihood that adult hens will disperse in a novel environment (Grigor et al. 1995). Fear-induced predator evasion in the open field is characterized by silence and immobility (Suarez & Gallup 1982; Jones 1987b, 1996) and it has been proposed that contact with the experimenter immediately prior to open-field testing mimics a predatory encounter (Suarez & Gallup 1981, 1982). Collectively, these findings suggest that the presence of a companion would be likely to attenuate social reinstatement motivation and that a predator evasion strategy (immobility) would predominate when pairs of chicks were placed close together in a novel test arena by the experimenter. A logical extension of our argument is that chicks would be expected to move apart from one another only when their fear levels had waned sufficiently to allow the expression of escape and/or exploratory responses. We used this premise to test our hypothesis that a familiar odourant (vanillin) could serve as a reassuring agent when chicks are placed in otherwise unfamiliar surroundings. To establish whether vanillin possessed any anxiolytic properties per se, we reared chicks without any exposure to it in experiment 1 and then tested them in pairs of cagemates in an unfamiliar open field in the presence or absence of this odourant. In experiment 2, we reared chicks with vanillin and then asked if its presence would increase social dispersal when familiar chicks were tested in pairs in the open field. Vanillin was chosen as the test odourant because it is not associated with toxicity in plants, it has no discernible irritating properties, it is easily available, it is relatively inexpensive, and chicks are known to form attachments to it (Jones & Carmichael 1999). Since sex differences have been reported in the open field and social affiliation responses of domestic chicks (Jones 1987b; Vallortigara et al. 1990; Vallortigara 1992), we used only female chicks in the present experiment. This choice was made because of its greater practical relevance, that is, the number of hens used in industry greatly exceeds that of cockerels. Experiment 1 METHODS Animals and husbandry We obtained 90 female ISA Brown chicks at 1 day of age from a commercial supplier. Immediately upon receipt they were randomly allocated to six groups of 12 chicks that were housed in each of six wooden boxes. We then subdivided each group of 12 chicks into six pairs and marked each member of each pair in a similar and distinctive fashion with indelible ink. Thus, although the same marking system was used in each box, both members of a pair bore a mark that was exclusive to them in that box. The latter procedure facilitated the subsequent identification of matched pairs and the random allocation of these pairs to the two test conditions described below. We then added three unmarked chicks from the remaining pool of 18 to each of the six boxes, bringing the number of chicks per box to 15; these would have been used as replacements if any of the marked chicks had died. The home boxes measured cm and 40 cm high and they rested on 1-m-high shelving covered with black builders paper. Wire lids prevented escape and the 1-cm wire-mesh floor was raised 2 cm off the floor to allow passage of excreta. Food (chick starter mash) and water were provided ad libitum in plastic hoppers suspended on wire grids from the tops of the side walls of each cage; these could be removed remotely for maintenance purposes. A combination of background heating and a dull emitter heater suspended above each box maintained an ambient temperature of ca. 27 C and the photoperiod was hours. Two plastic petri dishes, each containing 10 drops (measured using a teat pipette) of distilled water, were placed under the grid, one in each half of each box ca. 8 cm from the end walls. These were replenished twice daily but the chicks remained otherwise undisturbed until testing began.

3 JONES ET AL.: FAMILIAR ODOUR AND FEAR IN CHICKS 661 Test procedures We tested the chicks once only in pairs at 9 or 10 days of age in one of two circular, open fields (novel environment) 75 cm in diameter. Both members of a pair were taken from the same home box and both bore the same identifying mark (see above). We randomly chose one chick from each pair and added another mark on the head immediately before testing to aid identification. We recorded only the behaviour of the marked chick in each pair because data from both members of a pair could not have been regarded as independent. An overhead camera was switched on (see below) and we placed the chicks simultaneously against the wall of the open field so that they were touching and facing each other. The experimenter then retreated from the test room. The 45-cm-high metal wall of the open field was painted matt black and it rested on a 1-cm wire-mesh floor raised 2 cm off the grey concrete floor of the test room. We placed a petri dish containing either 10 drops of vanillin (odourant group) or a solution of food dyes (control group) under the centre of the wire-grid floor immediately before testing began. For the control stimulus, the odourless food dyes Hexacol Indigo Carmine and Hexacol Ponceau (Pointing Ltd, Prudhoe, U.K.) were added to water until the colour of the solution approximated that of the brown vanillin. We renewed the vanillin (4-hydroxy-3-methoxybenzaldehyde; Supercook, Leeds, U.K.) and control solutions at regular intervals. We placed another petri dish containing familiar food (starter mash) on top of the grid floor in the centre of each open field. The open fields were situated in separate rooms to eliminate the confounding effects of residual odours and we rotated the control and experimental treatments across rooms across test days. Apart from the presence of the odourant all other environmental conditions were similar in the two rooms. The ambient temperature in each test room was similar to that at which the chicks were reared (27 C) and illumination was provided by a 15-W lamp suspended 116 cm directly above the centre of the open field. A Panasonic Industrial Colour CCD microcamera (WV-KSi51, 8 2 cm) was suspended 80 cm above the centre of each open field and the responses of the marked chick were recorded on videotape. Upon analysis of the videotapes we recorded the latencies until the marked chick walked and until the distance in the open field between the centres of each chick s head reached 20 cm. This distance was chosen because it could be attained only if the chicks moved out of physical contact; it could not be reached if they simply turned away from each other but remained touching. We calculated distances by measuring the distance between the chicks heads on the cm monitor screen and then multiplying this value by 4 so that it translated to real-life distance in the 75-cm-diameter open field. We also measured the numbers of paces, pecks at the environment, pecks at the food and preening bouts shown by the marked chick. The distance between the centres of each chick s head was also recorded at intervals of 30 s throughout the 30-min observation period; the 60 values were then summed and averaged to provide a mean social dispersal score for each test. We tested six pairs of chicks from each home box in this way. Thus, 36 pairs were tested in total, 18 in each condition. We tested 18 pairs of chicks on each of the 2 test days. All chicks were immediately returned to their home box after testing, we cleaned the floor of the open field with a damp cloth, and replaced the food, vanillin and control stimuli before the next test began. Statistical analysis The latencies to walk and to move 20 cm apart were transformed to logarithms while counts of behaviours, for example paces, pecks, etc, were transformed to square roots to ensure that variance was approximately constant with respect to the mean. The measures of social dispersal (distance between chicks) did not require transformation. We then carried out analysis of variance between and within boxes to estimate the effects of test condition (odourant, no odourant), day and their interactions. Several pairs of chicks remained in close proximity throughout the observation period and many birds failed to feed, preen or peck. Therefore, because counts or latencies of these behaviours contained many zeros or maximum values of 1800 s, respectively, the proportion of birds feeding, moving apart, pecking or preening were also analysed with a generalized linear mixed model assuming binomial variation between birds within boxes and extra variation between boxes with a normal distribution and constant variance in the logistic scale. The statistical significance of the main effects of treatment, day and their interaction was established with Wald tests (Genstat 5.3 Committee 1993). Experiment 2 Animals Two batches of 75 and 135 female ISA Brown chicks, respectively, were obtained at 1 day of age from a commercial supplier and at intervals of 4 weeks. Immediately upon receipt they were randomly allocated to five and nine groups of 12 chicks that were housed in five and nine, respectively, of the wooden boxes described above. We used the same marking procedure as in experiment 1 to facilitate identification of matched pairs of chicks for subsequent testing. We added three unmarked chicks to each box to cover for any deaths of marked chicks. Thus each home box contained 15 chicks. Husbandry procedures also followed the same protocol but this time we placed two plastic petri dishes, each containing 10 drops of vanillin, under the grid floor, one in each half of each box ca. 8 cm from the end walls. We cleaned the petri dishes and renewed the odourant twice daily throughout the experiment; the odour of vanillin was detectable to the human nose at all times. The chicks remained otherwise undisturbed until testing began. Test procedures We tested the chicks once only in pairs of chicks at 8, 9 or 10 days of age in one of the two circular, 75-cm-diameter open fields described above. Both members of a pair were taken from the same home box and both bore the same identifying mark (see above). We

4 662 ANIMAL BEHAVIOUR, 63, 4 Table 1. Effects of the presence of a novel odourant (vanillin) on the open-field responses of domestic chicks tested in pairs of familiar cagemates Test situation Measures Odourant No odourant Sedtrans Variance ratio P Latency to separate (s) (5.41) (5.21) Latency to walk (s) 96.5 (4.57) 72.2 (4.28) No. of paces 52.4 (7.24) 73.6 (8.58) No. of environmental pecks 18.9 (4.35) 36.6 (6.05) <0.05 No. of pecks at food 1.3 (1.15) 1.7 (1.31) No. of preening bouts 0.4 (0.65) 0.2 (0.46) Distance apart (cm) 3.03± ± The table shows back-transformed means, means (in parentheses) and standard errors of differences (sedtrans) in the transformed (logarithmic or square root) scale. Latencies were transformed to logarithms and counts of behaviours to square roots. Untransformed means and their standard errors are shown for distances apart. Degrees of freedom=1,60. randomly chose one chick from each pair and added another mark on the head immediately before testing to aid identification. We placed a petri dish containing either 10 drops of vanillin (odourant group) or a solution of food dyes (control group) under the centre of the wire grid floor and beneath a similar dish containing food immediately before testing began. The control stimulus again consisted of a solution of the odourless food dyes described in experiment 1. We renewed the vanillin and control solutions at regular intervals. We placed the chicks simultaneously against the wall of the open field so that they were facing each other and in close physical contact. The chicks behaviour was again recorded on videotape with an overhead microcamera and we noted the same behavioural measures upon subsequent analysis. Six pairs of chicks from each home box were tested in this way. Thus, 84 pairs were tested in total, 42 in each condition. We tested 28 pairs of chicks (14 pairs in each condition) on each of the 3 test days. We then returned them to their home box, cleaned the open field with a damp cloth, and replaced the food, vanillin and control stimuli before the next test began. The latencies to walk and to move 20 cm apart and the measures of social dispersal (distance between chicks) were transformed to logarithms while counts of behaviours, for example paces, pecks, etc, were transformed to square roots to ensure that variance was approximately constant with respect to the mean. We then carried out analysis of variance between and within boxes to estimate the effects of treatment (odourant, no odourant), day, batch and their interactions. Many birds again failed to feed, to move 20 cm or more away from the companion chick, or to preen during the observation period (44/84, 19/84 and 34/84, respectively). For that reason, we again analysed the proportion of birds feeding, moving apart, or preening, with a generalized linear mixed model assuming binomial variation between birds within boxes and extra variation between boxes with a normal distribution and constant variance in the logistic scale. We used Wald tests to establish the statistical significance of the main effects of treatment, day and their interaction. RESULTS Table 1 shows the open-field responses of chicks that had received no prior exposure to vanillin when they were tested in the odourant and control situations in experiment 1. They are presented as back-transformed means, which can be interpreted approximately as medians of the original observations. Control chicks pecked significantly more at the environment. Although the controls also tended to walk sooner and more and to move further apart there were no other significant differences. Chicks that had been reared with vanillin in experiment 2 and that were then tested in pairs in the open field moved apart significantly sooner and showed greater social dispersal scores in the presence of the familiar odourant than those observed in the control test situation that did not contain vanillin (Table 2). The chicks also tended to walk sooner and to show more pacing (+28.6%), environmental pecking (+19.8%), preening (+42.9%) and pecks at the food (+33.9%) when vanillin was present, although none of these differences reached significance. There were no detectable effects of batch. All measurements showed some evidence of an interaction between batch and day but this was not considered further since there was little evidence of an overall effect of day (P always>0.05). The proportions of chicks in the control group that failed to feed, to move 20 cm or more apart, or to preen, respectively, were 30, 16 and 19 whereas those in the odourant group were 14, 3 and 15. Table 3 presents the back-transformed mean proportions of vanillin-reared chicks that showed selected behaviours in the odourant and the control test situations. The proportions of chicks feeding and moving 20 cm or more apart in the open field were significantly greater when the familiar odourant was present rather than absent. There were no detectable effects of day or of any interactions between day and treatment. DISCUSSION Certain fragrances have been attributed with moodenhancing properties and the psychological effects of

5 JONES ET AL.: FAMILIAR ODOUR AND FEAR IN CHICKS 663 Table 2. Effects of the presence of a familiar odourant on the open-field responses of domestic chicks tested in pairs of familiar cagemates Test situation Measures Odourant No odourant Sedtrans Variance ratio P Latency to separate (s) (4.91) 502.7(6.22) <0.001 Latency to walk (s) 36.6 (3.60) 42.5(3.75) No. of paces (11.47) 94.3(9.71) No. of environmental pecks 12.1 (3.48) 9.7(3.11) No. of pecks at food 24.8 (4.98) 16.4(4.06) No. of preens 2.1 (1.44) 1.2(1.11) Distance apart (cm) 15.7 (2.75) 9.1(2.21) <0.001 The table shows back-transformed means, means (in parentheses) and standard errors of differences (sedtrans) in the transformed (logarithmic) scale. Latencies and distances apart were transformed to logarithms and counts of behaviours to square roots. Degrees of freedom=1,60. Table 3. Back-transformed mean proportions of chicks exhibiting selected behaviour patterns when tested in pairs of cagemates in an open field in the presence or absence of a familiar odourant, with logit means (in parentheses) and standard errors of the differences on the transformed scale (sedtrans) Test situation Behaviour Odourant No odourant Sedtrans W 1 P Preening (0.822) (0.565) Feeding (0.492) ( 0.716) <0.01 Social dispersal (2.986) (0.586) <0.001 W=Wald statistic. The criterion for social dispersal was that the chicks had to move more than 20 cm apart. essential oils have long been acknowledged in aromatherapy and folk medicine (Tisserand 1988; Lehrner et al. 2000), although it is not clear if these phenomena reflected some intrinsic anxiolytic effects of the odourants or simply the evocation of pleasant and comforting memories. In view of these reports it was important to establish whether vanillin possessed calming properties per se. Therefore, in experiment 1 chicks that had no previous experience of vanillin were tested in pairs in a novel environment (open field) in the presence or absence of this odourant. In this case, the presence of vanillin was associated with a significant reduction in environmental pecking as well as nonsignificant tendencies towards delayed ambulation and social dispersal. These findings are indicative of heightened fear/ neophobia (Jones 1996) and thus they clearly show that vanillin possesses no intrinsic anxiolytic properties for domestic chicks, or at least those that have received no previous exposure to it. Unlike experiment 1, vanillin was incorporated in the home cages of all the chicks used in experiment 2 from 1 day of age. In this case, when pairs of domestic chicks from the same home cage (of 15 chicks) were placed close together in the open field, they moved apart significantly sooner and further and more of them fed when the familiar odourant was present rather than an odourless, colour-matched solution of food dyes. Chicks tested in the presence of the familiar odourant rather than in the control condition also tended to walk sooner and more, to peck at the environment and at food, and to preen more often, although not significantly so. The observed differences were all consistent and we therefore consider them likely to have biological relevance. Novelty is a potent fear elicitor and fear takes precedence over and exerts a progressively inhibitory effect on behaviours generated by all other motivational systems (Hogan 1965; Jones 1987b, 1996; Boissy 1995). Like other animals, chickens perceive enforced exposure to an unfamiliar environment as frightening (Faure 1981; Faure et al. 1983; Jones 1987b, 1996). Furthermore, contact with the experimenter immediately prior to open-field testing is thought to mimic a predatory encounter and thereby to promote the adoption of immobility (Suarez & Gallup 1982; Jones 1987b). Simplistically therefore, chicks that remain inactive in a novel environment are considered to be more frightened than those that ambulate, eat, preen or explore the environment (Faure 1981; Faure et al. 1983; Jones 1987b, 1996). Frightened chicks would also be expected to remain close together when tested in pairs in a novel environment and to move apart from one another only when their underlying levels of fear have dissipated sufficiently (Suarez & Gallup 1983; Jones 1987b; Vallortigara et al. 1990). Similarly, the readiness of socially tested adult laying hens to disperse in a novel arena is thought to be inversely related to fear (Grigor et al. 1995). In view of the above reports, our results are

6 664 ANIMAL BEHAVIOUR, 63, 4 all indicative of reduced fear in the presence of the familiar odourant (Faure et al. 1983; Vallortigara et al. 1990; Jones 1996). We cannot rule out a role for trigeminal chemoreception in the present study. However, the thresholds of sensitivity are considerably higher for the trigeminal than the olfactory system (Tucker 1965; Tolhurst & Vince 1976) and the odour of vanillin was not particularly strong, at least to the human nose. Therefore, we consider our findings more likely to reflect an olfactory rather than a trigeminal phenomenon. Unlike our finding that more chicks fed when the open field contained the familiar odourant, chicks that had been exposed to a moist-food odourant from embryonic day 20 to 18 h after hatching pecked less than controls at a small coloured bead attached to a delivery system releasing the odour of moist or dry food (Burne & Rogers 1999). This apparent inconsistency may simply reflect methodological differences, because the chicks used by Burne & Rogers (1999) were housed individually, they were tested in their home cages at 1 2 days of age, and they were not given the opportunity to feed at test. Our results have both fundamental and strategic relevance. First for example, the differences observed here between the chicks reactions in the presence or absence of the familiar odourant reinforce earlier suggestions that olfaction can play an important role in regulating the behaviour of the domestic fowl (Jones & Gentle 1985; Vallortigara & Andrew 1994; Porter et al. 1995; Burne & Rogers 1996; Jones & Roper 1997; Jones & Carmichael 1999). Our findings also further confirm the existence of olfactory memories in the domestic chick. Second, the enhanced readiness to disperse and to feed in a novel environment found here when pairs of chicks were tested in the presence of vanillin is consistent with an earlier report that the presence of a familiar odourant (geraniol) decreased fear-induced behavioural inhibition when chicks were tested individually in an open field (Jones & Gentle 1985). Collectively, these findings strongly support our suggestion that familiar odourants can act as reassuring agents in otherwise unfamiliar situations, at least for the domestic fowl. This phenomenon might also readily generalize to include other species. The use of familiar and reassuring fragrances is becoming an accepted method of reducing anxiety in human beings (King 1988; Lehrner et al. 2000) and a potential antidepressant effect has been claimed in rats, Rattus norvegicus (Komori et al. 1995). Similar olfactory therapies could have important implications for issues relating to the management, welfare and productivity of poultry. For example, because novelty is frightening (Hogan 1965; Jones 1996), chickens are likely to experience fear and distress when they are translocated from one environment to another, for example, from the hatcher to unfamiliar brooding cages and from there to the rearing environment (Jones 1996); they are often reluctant to venture out from the poultry shed on to an exposed and unfamiliar free range area (Grigor et al. 1995); and changes in the colour or smell of their food (Jones 1987a; Marples & Roper 1996) or in the appearance of the stockperson (Jones 1984) can elicit neophobia and fear. The sudden, intense or prolonged elicitation of behavioural and physiological fear reactions, such as panic or chronic adrenocortical activation, can cause injury, pain or even the death of the animal as well as marked decreases in its reproductive performance, growth and food conversion efficiency (Mills & Faure 1990; Boissy 1995; Jones 1996, 1997; Hemsworth & Coleman 1998). The induction of stress, such as might be caused by translocation of birds to unfamiliar housing systems or by environmental impoverishment, has also been implicated in the aetiology of feather pecking (Vestergaard et al. 1997; El-Lethey et al. 2000); this potentially catastrophic behavioural vice can cause energy wastage, injury and, if it escalates into cannibalism, the painful death of numerous target birds (Blockhuis 1986; Jones & Hocking 1999). Our findings suggest that by incorporating a familiar, perhaps imprinted, odourant into a new environment or perhaps applying it to novel diets or to unfamiliar stockpersons we might reduce chickens fear of novelty and its potentially harmful consequences. There is now mounting evidence that olfactory memories can be formed in the chick embryo as well as the neonate (Rogers 1995; Sneddon et al. 1998). Therefore, identifying the developmental stages at which attachments to odourants are most strongly established presents an exciting and worthwhile research opportunity. Acknowledgments This study was commissioned and supported by the U.K. Ministry for Agriculture, Fisheries and Food. The Roslin Institute is supported by the Biotechnology and Biological Sciences Research Council. We thank Christine Ruschak and Anna Tassi for their invaluable technical assistance. We are also grateful to the University of Padua, Italy for providing Lucilla Facchin s scholarship. References Bang, B. G Functional anatomy of the olfactory system in 23 orders of birds. Acta Anatomica (Suppl), 58, Blockhuis, H. T Feather pecking in poultry: its relations with ground pecking. Applied Animal Behaviour Science, 16, Boissy, A Fear and fearfulness in animals. Quarterly Review of Biology, 70, Bolhuis, J. J Mechanisms of avian imprinting: a review. Biological Reviews, 66, Burne, T. H. J. & Rogers, L. J Odors, volatiles and approachavoidance behavior of the domestic chick (Gallus gallus domesticus). International Journal of Comparative Psychology, 8, Burne, T. H. J. & Rogers, L. J Responses to odorants by the domestic chick. Physiology and Behavior, 60, Burne, T. H. J. & Rogers, L. J Changes in olfactory responsiveness by the domestic chick after early exposure to odorants. Animal Behaviour, 58, El-Lethey, H., Aerni, V., Jungi, T. W. & Wechsler, B Stress and feather pecking in laying hens in relation to housing conditions. British Poultry Science, 41, Faure, J. M Bi-directional selection for open-field activity in young chicks. Behavior Genetics, 11,

7 JONES ET AL.: FAMILIAR ODOUR AND FEAR IN CHICKS 665 Faure, J. M., Jones, R. B. & Bessei, W Fear and social motivation as factors in open-field behaviour of the domestic chick. Biology of Behaviour, 8, Fischer, G. J The behaviour of chickens. In: The Behaviour of Domestic Animals (Ed. by E. S. E. Hafez), pp London: Bailliere Tindall. Fluck, E., Hogg, S., Mabbutt, P. S. & File, S. E Behavioural and neurochemical responses of male and female chicks to cat odour. Pharmacology, Biochemistry and Behavior, 54, Gallup, G. G. & Suarez, S. D An ethological analysis of open-field behaviour in chickens. Animal Behaviour, 28, Genstat 5 Committee Genstat 5, Release 3 Reference Manual. Oxford: Clarendon Press. Grigor, P. N., Hughes, B. O. & Appleby, M. C Emergence and dispersal behaviour in domestic hens: effects of social rank and novelty of an outdoor area. Applied Animal Behaviour Science, 45, Hemsworth, P. H. & Coleman, G. J Human-Livestock Interactions. The Stockperson and the Productivity and Welfare of Intensively Farmed Animals. Wallingford: CAB International. Hogan, J. A An experimental study of conflict and fear: an analysis of behaviour of young chicks towards a mealworm: Part 1. The behaviour of chicks which do not eat the mealworm. Behaviour, 25-26, Hogan, J. A. & Abel, E. L Effect of social factors on response to unfamiliar environments in Gallus gallus spadiceus. Animal Behaviour, 19, Jones, R. B Fear responses in domestic chicks as a function of the social environment. Behavioural Processes, 8, Jones, R. B Experimental novelty and tonic immobility in chickens (Gallus domesticus). Behavioural Processes, 9, Jones, R. B. 1987a. Food neophobia and olfaction in domestic chicks. Bird Behaviour, 7, Jones, R. B. 1987b. The assessment of fear in the domestic fowl. In: Cognitive Aspects of Social Behaviour in the Domestic Fowl (Ed. by R. Zayan & I. J. H. Duncan), pp Amsterdam: Elsevier. Jones, R. B Avian open-field research and related effects of environmental novelty: an annotated bibliography, Psychological Record, 39, Jones, R. B Fear and adaptability in poultry: insights, implications and imperatives. World s Poultry Science Journal, 52, Jones, R. B Fear and distress. In: Animal Welfare (Ed.byM.C. Appleby & B. O. Hughes), pp Wallingford: CAB International. Jones, R. B. & Black, A. J Behavioral responses of the domestic chick to blood. Behavioral and Neural Biology, 27, Jones, R. B. & Carmichael, N. L Domestic chicks are attracted to a familiar odorant in a novel test situation: a brief report. Applied Animal Behaviour Science, 61, Jones, R. B. & Faure, J. M Domestic chicks prefer familiar soiled substrate in an otherwise novel environment. IRCS Medical Science, 10, 847. Jones, R. B. & Gentle, M. J Olfaction and behavioral modification in domestic chicks (Gallus domesticus). Physiology and Behavior, 34, Jones, R. B. & Hocking, P. M Genetic selection for poultry behaviour: big bad wolf or friend in need? Animal Welfare, 8, Jones, R. B. & Merry, B. J Individual or paired exposure of domestic chicks to an open field: some behavioural and adrenocortical consequences. Behavioural Processes, 16, Jones, R. B. & Roper, T. J Olfaction in the domestic fowl: a critical review. Physiology and Behavior, 62, Jones, R. B., Marin, R. H., Garcia, D. A. & Arce, A T-maze behaviour in domestic chicks: a search for underlying variables. Animal Behaviour, 58, Kare, M. R. & Rogers, J. G. Jr Sense organs. In: Avian Physiology (Ed. by P. D. Sturkie), pp New York: Springer- Verlag. King, J. R Anxiety reduction using fragrances. In: Perfumery: The Psychology and Biology of Fragrance (Ed. by S. van Toller), pp New York: Chapman & Hall. Komori, T., Fujiwara, R., Tanida, M. & Nomura, J Potential antidepressant effects of lemon odor in rats. European Neuropsychopharmacology, 5, Launay, F., Mills, A. D. & Faure, J. M Social motivation in Japanese quail Coturnix coturnix japonica chicks selected for high or low levels of treadmill behaviour. Behavioural Processes, 24, Lehrner, J., Eckersberger, C., Walla, P., Potsch, G. & Deecke, L Ambient odor of orange in a dental office reduces anxiety and improves mood in female patients. Physiology and Behavior, 71, Lima, S. L Back to the basics of anti-predatory vigilance: the group-size effect. Animal Behaviour, 49, Marin, R. H., Freytes, P., Guzman, D. & Jones, R. B Effects of an acute stressor on fear and on the social reinstatement responses of domestic chicks to cagemates and strangers. Applied Animal Behaviour Science, 71, Marples, N. M. & Roper, T. J Effects of novel colour and smell on the response of naive chicks towards food and water. Animal Behaviour, 51, Marples, N. M. & Roper, T. J Response of domestic chicks to methyl anthranilate odour. Animal Behaviour, 53, Mills, A. D. & Faure, J. M Panic and hysteria in domestic fowl: a review. In: Social Stress in Domestic Animals (Ed. by R. Zayan & R. Dantzer), pp Dordrecht: Kluwer Academic. Porter, R. H., Turro-Vincent, I. & Picard, M Role of chemical cues in food choice by 3-day-old chicks. Chemical Senses, 20, 152. Pulliam, H. R. & Caraco, T Living in groups: is there an optimal group size? In: Behavioural Ecology. An Evolutionary Approach. 2nd edn. (Ed. by J. R. Krebs & N. B. Davies), pp London: Blackwell Scientific. Rogers, L. J The Development of Brain and Behaviour in the Chicken. Wallingford: CAB International. Roper, T. J. & Marples, N. M Colour preferences of domestic chicks in relation to food and water presentation. Applied Animal Behaviour Science, 54, Salzen, E. A The ontogeny of fear in animals. In: Fear in Animals and Man (Ed. by W. Sluckin), pp New York: Van Nostrand Reinhold. Sneddon, H., Hadden, R. & Hepper, P. G Chemosensory learning in the chicken embryo. Physiology and Behavior, 64, Stattelman, A. J., Talbot, R. T. & Coulter, D. B Olfactory thresholds of pigeons (Columbia livia), quail (Colinus virginianus) and chickens (Gallus gallus). Comparative Biochemistry and Physiology, 50A, Suarez, S. D. & Gallup, G. G Predatory overtones of open-field testing in chickens. Animal Learning and Behavior, 9, Suarez, S. D. & Gallup, G. G Open-field behaviour in chickens: the experimenter is a predator. Journal of Comparative and Physiological Psychology, 96, Suarez, S. D. & Gallup, G. G Social reinstatement and open-field testing in chickens. Animal Learning and Behavior, 11, Tisserand, R Essential oils as psychotherapeutic agents. In: Perfumery: the Biology and Psychology of Fragrance (Ed. by S. van Toller & G. H. Dodd), pp London: Chapman & Hall.

8 666 ANIMAL BEHAVIOUR, 63, 4 Tolhurst, B. E. & Vince, M. A Sensitivity to odours in the embryo of the domestic fowl. Animal Behaviour, 24, Tucker, D Electrophysiological evidence for olfactory function in birds. Nature, 207, Turro, I., Porter, R. H., Picard, M Olfactory cues mediate food selection by young chicks. Physiology and Behavior, 55, Turro-Vincent, I The ontogeny of feeding behaviour in young chicks kept under intensive husbandry conditions. Ph.D. thesis, University of Tours. Vallortigara, G Affiliation and aggression as related to gender in domestic chicks (Gallus gallus). Journal of Comparative Psychology, 106, Vallortigara, G. & Andrew, R. J Olfactory lateralization in the chick. Neuropsychologia, 32, Vallortigara, G., Cailotto, M. & Zanforlin, M Sex differences in social reinstatement motivation of the domestic chick (Gallus gallus) as revealed by runway tests with social and nonsocial reinforcement. Journal of Comparative Psychology, 104, Vestergaard, K. S., Skadhauge, E. & Lawson, L. G The stress of not being able to perform dustbathing in laying hens. Physiology and Behavior, 62, Wood-Gush, D. G. M The Behaviour of the Domestic Fowl. London: Heinemann. Yaber, M. C. & Herrera, E. A Vigilance, group size and social status in capybaras. Animal Behaviour, 48,