The effect of early environment on neophobia in orange-winged Amazon parrots (Amazona amazonica)

Transcription

1 Applied Animal Behaviour Science 89 (2004) The effect of early environment on neophobia in orange-winged Amazon parrots (Amazona amazonica) Rebecca A. Fox, James R. Millam Department of Animal Science, University of California, One Shields Avenue, Davis, CA 95616, USA Accepted 7 May 2004 Abstract Early experience is often a significant factor in shaping animals later behavior. Early maternal separation is associated with negative behavioral outcomes, such as increased fearfulness in rats, while higher levels of maternal grooming during the neonatal period are associated with decreased fearfulness and increased exploratory behavior. This finding may have implications for the welfare of captive parrots, many of which are hand-reared for the pet trade. We investigated the effects of three different rearing conditions on the neophobia of juvenile orange-winged Amazon parrots (Amazona amazonica). Hand-reared (H, N = 6), parent-reared/human-handled (birds which were handled five times/week for 20 min/session between 2 and 8 weeks of age, PH, N = 6), and parent-reared without handling (P, N = 7) parrots were tested for neophobia between the ages of 4.5 and 6 months of age by measuring their latency to feed in the presence of five different novel objects. The parrots neophobia was assessed again at 12 months of age by measuring their response to a novel object hung in their home cage. Although PH birds were groomed by their parents significantly more than P birds (F 2,16 = 6.21, P = 0.01), there was no significant difference in neophobia between the two groups (F 1,11 = 0.41, P = 0.53, 1 d.f.). H birds were significantly less neophobic than P and PH birds until 6 months of age (F 1,17 = 9.25, P = 0.007, 1 d.f.). At 1 year of age, P, PH, and H birds exhibited comparable levels of neophobia. Our results suggest that the development of neophobia in orange-winged Amazons is not related to parental care, but may be related to the level of novelty that the chicks experience during early life Published by Elsevier B.V. Keywords: Parrot; Psittacine; Behavior; Enrichment; Neophobia; Behavioral development; Hand-rearing; Amazona amazonica Corresponding author. Tel.: ; fax: address: rafox@ucdavis.edu (R.A. Fox) /$ see front matter 2004 Published by Elsevier B.V. doi: /j.applanim

2 118 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) Introduction Within populations, individuals often exhibit consistent differences in behavior, both in the laboratory and in the wild (Benus et al., 1991; Verbeek et al., 1994,1996; Gosling and John, 1999). For example, some great tits (Parus major) are highly sensitive to novelty, while others are not (Verbeek et al., 1994). This difference in behavioral strategy, also referred to as coping style or personality (Benus et al., 1991; Capitanio, 1999; Gosling and John, 1999), influences behavior across a variety of contexts. Great tits, which are highly sensitive to novelty spend more time exploring novel objects, are quicker to respond to changes in the environment and tend to be less aggressive in an aviary setting (Verbeek et al., 1994,1996,1999). Within populations, individual personality types often have approximately equal fitness, and may serve to increase the likelihood that populations can adapt to environmental change (Wilson, 1998). Commonly measured personality traits include reactivity to novelty (Benus et al., 1991; Verbeek et al., 1994; Capitanio, 1999; Francis et al., 1999; Levine, 2001; Meaney, 2001), aggressiveness (Benus et al., 1991; Verbeek et al., 1996; Capitanio, 1999), and sociability (Capitanio, 1999). Personality factors tend to be stable throughout an animal s lifetime, and can be used to predict an animal s behavior across a wide variety of situations (Benus et al., 1991; Verbeek et al., 1994; Capitanio, 1999; Meaney, 2001). Although personality has been shown to have a strong genetic component in some species (e.g., Carere et al., 2001), early experience is probably at least as important as genetics in shaping personality, especially in species with long periods of immaturity. Mason (1979), in fact, suggested that early experience strongly influences the phenotypic outcome of development, and that variation in early experience may maintain phenotypic flexibility within populations. Evidence for nongenomic inheritance of hypothalamic pituitary adrenal axis reactivity to stressors in rats (Francis et al., 1999) provides empirical support for this view. In rats, the transmission of stress reactivity across generations is mediated by maternal grooming behavior (Francis et al., 1999; Levine, 2001; Meaney, 2001), but any aspect of animals early experience can potentially influence personality development. Francis et al. (2002) also showed that the level of environmental enrichment, to which rats are exposed before puberty, strongly influences their reactivity to novelty as adults. An understanding of the influence of early experience on personality development is relevant to the welfare of captive reared animals, as well as to more general evolutionary questions. Many parrots raised for the pet trade in the United States are hand-reared (removed from the nest at a young age and reared by humans). Although hand-reared parrots tend to be quite tame and are highly valued as companion animals (cf. Millam, 2000), hand-rearing typically involves permanent maternal separation. In cockatiels (Nymphicus hollandicus), hand-rearing has been shown to produce reproductive disadvantages, which are probably related to abnormal habitat imprinting and sexual imprinting (Myers et al., 1988). Furthermore, at least in mammalian species, even short periods of maternal separation (5 24 h) may lead to other changes, such as increased stress reactivity, which may persist throughout an animal s lifetime (Vazquez, 1997; Levine, 2001; Meaney, 2001). Studies of orange-winged Amazon parrots (Amazona amazonica) have shown that handling parrots regularly for brief periods as nestlings can also be quite effective in producing tame parrots (Aengus and Millam, 2001). Although it has been suggested that taming parrot chicks by

3 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) neonatal handling prevents the development of the behavioral abnormalities associated with hand-rearing, such as abnormal sexual and social imprinting and inappropriate reproductive behavior (cf. Millam, 2000), this has not been established empirically. The behavioral phenotype of hand-reared and neonatally handled orange-winged Amazon parrots is quite similar (with respect to tameness) and differs substantially from that of birds, which have been parent-reared without handling (Aengus and Millam, 2001). However, it is not known how hand-reared, neonatally handled, and parent-reared birds differ with respect to other behaviors. Thus, the present experiments were designed to address the question of whether hand-rearing, neonatal handling, and parent-rearing alone result in different outcomes with respect to more species-typical behavior, specifically neophobia. Neophobia (operationally defined as reluctance to feed in the presence of a novel object) was used as an index of behavioral differences associated with three different rearing methods (i.e., parent rearing, parent rearing with human handling, and hand-rearing) because differences in rearing conditions are typically associated with changes in stress reactivity and fearfulness in mammals (Vazquez, 1997; Levine, 2001; Meaney, 2001). Chicks feather quality was also measured after removal from their parents cages or the nursery, since neonatal handling is also associated with differences in parental grooming behavior in mammals (cf. Meaney, 2001). 2. Methods 2.1. Subjects Nineteen orange-winged Amazon chicks from eight unrelated pairs of parents were used for this study. The chicks were reared at the University of California, Davis, Avian Science Research Facility in the spring of Six of the chicks were removed from their parents nestboxes at 2 weeks of age for hand-rearing. The remainder of the chicks remained with their parents until 3 4 months of age. Six of these parent-reared chicks were handled by experimenters between 2 and 8 weeks of age. Each of the handled chicks was handled by an experimenter for 20 min per day, 4 5 days/week. The remaining seven chicks were left undisturbed. Hand-reared chicks were maintained on Roudybush Optimum Handfeeding Formula (Roudybush, Inc., Woodland, CA) until weaning at approximately 3 months of age, after which time they were maintained on Roudybush Low Fat Maintenance Diet. Parent-reared birds and parent-reared/human-handled birds were fed Roudybush Breeder Diet by their parents. After removal from their natal cages, parent-reared birds were also maintained on Roudybush Low-Fat Maintenance Diet. All birds in the study were also provided with fruits and vegetables approximately five times per week as well as a variety of enrichments in their home cages Housing Hand-reared chicks At 2 weeks of age, hand-reared chicks were separated from their parents and removed to a nursery. They were housed in clear acrylic brooders maintained at 37 C until approximately

4 120 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) days of age, at which time they were partially feathered and able to thermoregulate. The hand-reared group consisted of one pair of siblings and four unrelated birds. To minimize chances of disease transmission, only chicks from the same nest were brooded together, so that the pair of siblings was brooded together and the four unrelated birds were brooded singly. After 35 days of age, chicks were housed in open plastic bins. At this point, birds were housed two per bin, so that the siblings remained together and unrelated birds of similar age were paired. Terrycloth towels and paper towels were used as bedding substrates throughout. Each brooder or bin was equipped with a stuffed terrycloth doll to serve as a surrogate nest-mate for huddling behavior. Chicks were allowed to fly freely about the nursery after fledging at approximately 56 days of age. Once able to perch on the side of the bins (at approximately days of age), hand-reared chicks were able to interact with one another during the day. The photoperiod in the nursery was 12 h of light and 12 h of darkness (12L:12D). Food and water were provided ad libitum after the chicks were 35 days of age Parent-reared chicks Parent-reared (P) and parent-reared/human handled (PH) chicks were housed with their parents until weaning in suspended welded-wire cages (1 m 1m 2 m) provided with grandfather-clock-style galvanized metal nestboxes (40 cm 38 cm 76 cm) partially filled with pine shavings. Nestboxes were equipped with partitions, which could be inserted to separate chicks/eggs from parents to facilitate removal from the nestbox. The photoperiod was 14L:10D. The longer photoperiod was used to encourage breeding and was maintained until chicks were removed from their parents home cages. Food and water were available ad libitum. Chicks remained in the nestboxes until fledging at approximately 8 weeks of age. They remained with their parents for another 5 weeks after fledging Housing of juveniles When the chicks were of 3 4 months of age, they were removed from the nursery or their parents cages and transferred to another room in the facility. Chicks were housed in parrot cages (approximately 60 cm 90 cm 180 cm) outfitted with natural manzanita branch perches and various enrichment devices (wooden toys, rope perches etc.). Chicks were socially housed with either a sibling (parent-reared birds) or in the case of hand-reared birds, an unrelated bird of similar age if no sibling was available. Two birds (one PH bird and one P bird) were housed singly because of uneven numbers in the groups. The photoperiod was 12L:12D, and food and water were available ad libitum Treatments At hatching, chicks were randomly assigned to one of three treatment groups. The three groups were parent-reared (P; N = 7, from four clutches), parent-reared with human handling (PH; N = 6, from three clutches), and hand-reared (H; N = 6, from five clutches). To control for potential effects on parental care due to the exclusion of parents from the nestbox during the handling procedure, clutches did not contain both handled and nonhandled chicks.

5 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) Parent-rearing alone Parent-reared chicks were housed with their parents as described above. The nests were left undisturbed and no handling of chicks occurred Parent-rearing with human handling Chicks were housed with their parents as described above. Chicks were removed from the nest for handling four to five times weekly from 14 days until fledging (approximately 56 days of age). Each chick was assigned one or two handlers and was handled by the assigned handlers throughout the handling period. Handling consisted of weighing the chicks followed by min of gentle handling (stroking allowing the bird to perch on a hand or arm, etc.) and talking softly to the chick Hand-rearing Chicks were removed from nestboxes at 14 days of age and housed as described above. They were fed orally via syringe at regular intervals over a 16-h period each day (they were fed once during the dark period until feedings were reduced to two per day). Chicks were hand-fed by the investigator and three undergraduate research assistants. Two-week old chicks were fed every 4 h, with between-feeding intervals increasing to approximately 5 h at 3 weeks of age and 6 h by 5 weeks of age. After 5 weeks of age, feeding intervals were increased by 1 h every days. Chicks were fed as much as they would consume in a feeding (amounts varied from 20 ml/feeding for a 2-week-old chick to approximately 120 ml/feeding for an 8-week-old fledgling). A variety of solid food, including Roudybush Low-fat Maintenance diet, Cheerios breakfast cereal (General Mills, Minneapolis, MN), and assorted fruits and vegetables were offered ad libitum after the chicks were of 5 weeks of age. Chicks were eating independently by 90 days of age Feather scoring Composite feather scores were determined to characterize the impact of neonatal handling on parental grooming behavior. Chicks feather condition was scored approximately 1 week after removal from their parents home cages using the method developed by Meehan et al. (2003). Five areas of each bird s body were assessed: chest and flank, wings, back, tail, and legs. Each area was scored on a scale of 0 2, with 0 representing complete defeatheration and skin damage and 2 representing excellent feather condition. Scores from the five areas were added up, for a possible composite feather score of 10 points. Data were analyzed using one-way ANOVA for treatment group (SAS Systems, Cary, NC) Tests of neophobia Novel objects tests After the chicks had been separated from their parents for approximately 14 days (approximately 3.5 months of age), they were habituated to a dish containing unsalted peanuts in the shell hung on the back wall of the cage once a day by the experimenter. The birds had no prior experience with either the dish used in the testing or with peanuts. Peanuts are

6 122 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) a favorite food of many parrots including the experimental birds, which, once habituated, generally began consuming peanuts within 10 s of the dish being placed in their cage. After a habituation period of approximately 4 weeks (which ended when all birds were reliably eating peanuts from the dish), a series of five novel-objects tests were performed 8 14 days apart beginning when the chicks were approximately 4.5 months old. The test consisted of hanging a novel object (approximately 16 cm 10 cm 10 cm) beside the peanut dish and observing the birds behavior for 30 min via a video camera. Novel objects were a plastic cooking spoon, a green plastic mug, a child s toy truck, a ball of cloth, and a roll of adding-machine tape. Each pair of birds received each novel object in the order presented above. Each bird s latency to approach the dish and eat a peanut was recorded. Longer latencies of approach were indicative of higher levels of neophobia. Latency data were log-transformed to homogenize variance. Although each novel object was presented only once to each pair of birds, the same birds were used for all novel objects tests. Because the data consisted of several measurements for each bird, the data points for each novel object test cannot be assumed to be independent, therefore the data were analyzed using repeated-measures ANOVA. One chick from the PH group died of a respiratory infection after feather scoring and the completion of three of five novel objects tests. This bird s latency to feed in the first three novel objects tests was included in the statistical analysis and was treated as a missing value in the two tests administered after its death Toy presentation and neophobia rating scale When the birds were approximately of 1 year of age, we again assessed their level of neophobia by pairing the presentation of an unfamiliar commercial parrot toy (Avitech Large Chili Strand, Avitech International, Frasier Park, CA) with a simple behavioral rating scale for fearfulness. The toy was hung from the highest point of the cage ceiling, the birds were allowed to settle for about 30 s, and their behavior was observed for approximately 1 min. The rating scale consisted of five possible scores ranging from 0 (approach toy) to 5 (panic flapping, falling from perch, etc.). The toy was left in the birds cages for 30 min following the end of the 1-min observation period. At the end of this 30-min time period, the birds position in the cage was recorded, and it was noted whether or not the toy appeared to have been played with, based upon evidence of chewing (bite marks, scratches etc.). 3. Results 3.1. Feather scoring Although parental grooming behavior was not directly observed, the impact of neonatal handling on parental grooming behavior was inferred through a comparison of the chicks feather scores at approximately 15 weeks of age (1 week after the HR birds had been moved from the nursery and the P and PH birds had been removed from their parents home cages). One-way ANOVA revealed a significant effect of treatment on feather score at 15 weeks of age (F 2,16 = 6.21, P = 0.01)(Table 1). Comparisons between the groups revealed that

7 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) Table 1 Mean ± S.E. feather scores of juvenile Amazons by rearing treatment, 1 week after removal from parents or nursery (approximately 15 weeks of age) Treatment group N Feather score Parent-reared ± 0.37 a Hand-reared ± 0a Parent-reared/human-handled ± 0.56 b Lower scores represent poorer feather quality. Means with unlike letters (a, b) are significantly different (P < 0.05, ANOVA). differences between the H and P groups were not significant (t = 1.96, P = 0.07). However, feather score differences between the PH and P groups were significant at the 0.01 level (t = 4.62, P = ), indicating that PH chicks had significantly poorer feather quality than both P and H chicks Novel objects tests There was no significant effect of treatment group on neophobia when P and PH birds were compared (F 1,11 = 0.41, P = 0.53, Fig. 1). Neonatal handling thus had no effect on Fig. 1. Latencies (mean ± S.E.) to feed from a dish of favored food in the presence of a novel object for parent-reared (P) and parent-reared with human handling (PH) birds in five tests administered between 4.5 and 6 months of age. Repeated measures ANOVA showed that the behavior of the two groups was not significantly different (F 1,11 = 0.41, P = 0.54).

8 124 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) Fig. 2. Latencies (mean ± S.E.) to feed from a dish of favored food in the presence of a novel object for parent-reared (P/PH) and hand-reared (H) birds in five tests administered between 4.5 and 6 months of age. Data from P and PH groups were combined for analysis. Repeated measures ANOVA showed a significant difference between P/PH and H birds (F 1,17 = 9.25, P = 0.007). the latency to feed in the presence of a novel object. The lack of significant differences between these two treatment groups allowed for the data to be combined for further data analysis. This combined group of P and PH birds is referred to hereafter as all parent-reared birds (P/PH). Latencies to feed in the presence of a novel object differed between H and P/PH birds (Fig. 2). In contrast to the P/PH birds, which were already somewhat neophobic (exhibited long latencies to approach the dish + novel object combination) at the onset of testing, H birds showed lower levels of neophobia until the last test. Higher latencies to feed in the presence of a novel object were observed for the P/PH birds than for H birds in all except the final test, when the birds were approximately 6 months of age. Repeated measures ANOVA revealed a significant effect of treatment group (F 1,17 = 9.25, P = 0.007) on latency to approach a novel object when the P/PH and H groups were compared. Because the novel objects were not presented in random order, there is a possibility that the data may have been confounded by the birds reactions to different types of objects. However, the fact that H and P/PH birds exhibited very different responses to the first three objects (Fig. 1), suggests that this effect, if present, was probably minor.

9 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) Fig. 3. Neophobia scores (mean ± S.E.) when parent-reared (P), parent-reared with human handling (PH) and hand-reared (H) birds were presented with a novel object at 12 months of age. A score of 0 represents immediate approach to the toy, while a score of 5 represents frantic attempts to avoid the toy. Differences between the groups were not significant Novel object presentation at one year of age At 1 year of age, P, PH, and H birds displayed similar levels of neophobia. When presented with a brightly colored wooden parrot toy (Avitech Large Chili Strand), most birds avoided it. The average neophobia scores for the three groups were nearly identical (Fig. 3), although the PH birds displayed more variability in their response to the toy. After 30 min, only three birds were observed to have any contact with the toy (one each from the P, PH, and H groups), and most birds were still positioned in corners or at the bottom of the cage, generally away from the toy. 4. Discussion The aim of this study was to elucidate the contribution of the early environment to behavioral development in Amazon parrot chicks. As has been shown in rats, parrot chicks early experience can affect behavioral development. However, parental grooming behavior appears to be less significant in shaping developmental outcomes in Amazon parrots than in rats. For parrots, the level of novelty in the early environment may exert a stronger influence on behavior. In rats, maternal grooming behavior is largely responsible for determining the stress reactivity (and neophobia) of offspring (Francis et al., 1999; Vazquez, 1997; Levine, 2001; Meaney, 2001). Neonatal handling of rat pups increases maternal licking and grooming

10 126 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) behavior (Francis et al., 1999; Meaney, 2001). If a similar effect occurred in orange-winged Amazons, one would expect that handling of nestlings be associated with overgrooming of the chicks, and that overgrooming should be associated with measurable differences in neophobia relative to nonhandled birds. Handling of nestlings was associated with lower plumage quality in 3-month-old birds, since PH birds had significantly lower feather scores than either P or H birds. Because most of the chicks had feathers removed on the backs of their heads, we assume that featherpicking by the parents was at least partly responsible for the lower plumage quality. We note that the association between handling of nestlings and decreased plumage quality may be coincidental, since further data have not been collected to determine whether handling of nestlings is consistently associated with featherpicking of the chicks. Regardless of the basis for the lower plumage scores, differences in parental grooming behavior did not result in measurable differences in neophobia. There were no significant differences between the PH birds and P birds in tests of neophobia, despite the fact that PH birds were significantly more featherpicked. These results suggest that parental grooming behavior is relatively unimportant to the development of neophobia in parrots. Meaney (2001) suggests that the increase in maternal grooming behavior induced by neonatal handling of rat pups may be a mechanism to compensate for short periods of maternal separation. Unlike rats, orange-winged Amazons do not experience significant periods of maternal separation during the first few weeks of life; the female remains in the nest and is fed by the male (Fox, unpublished observation). This difference in parental care may account for the apparent insensitivity of the parrot chicks to neonatal handling and/or overgrooming by their parents. In rats, maternal grooming behavior during the first few weeks of life is crucial to the regulation and normal development of the hypothalamic pituitary adrenal (HPA) axis (Vazquez, 1997; Levine, 2001; Meaney, 2001). Because long periods (5 24 h) of maternal separation prevent rat pups from experiencing normal grooming behavior, maternal separation induces dramatic changes in HPA axis regulation and reactivity to acute environmental stressors in rat pups (Vazquez, 1997; Levine, 2001; Meaney, 2001). Therefore since hand-rearing of parrot chicks involves a prolonged maternal separation, one would expect to see significant differences in neophobia in H chicks relative to P/PH chicks. In fact, hand-rearing significantly reduced neophobia in H chicks until approximately 6 months of age. However, evidence suggests that this difference is unrelated to the absence of parental care. In rats, both maternal separation and low levels of maternal grooming can cause an increase in HPA axis responsiveness to acute environmental stressors (e.g., saline injection, introduction into an open-field apparatus) and result in higher levels of neophobia (cf. Vazquez, 1997; Meaney, 2001), while high levels of maternal grooming behavior decrease HPA axis responsiveness and neophobia in rats. In the orange-winged Amazons we observed that hand-rearing led to a significant, transient decrease in neophobia, but overgrooming of chicks by their parents had no measurable impact on neophobia. Obviously, some aspect of the hand-rearing treatment influences the neophobic behavior of H chicks, but the insensitivity of P/PH chicks to variation in grooming suggests that the mechanism driving differences in neophobia is different for parrots than for rats. We propose that the high levels of environmental novelty experienced by the H chicks may be at least partly responsible for their lower levels of neophobia during the first 6 months

11 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) of life. Normally, Amazon parrot chicks spend the first 2 months of their life in the dimly lit, relatively constant environment of the nestbox. In contrast, hand-reared birds experience high levels of environmental novelty in the nursery; they are exposed to several different experimenters, a large variety of objects, such as syringes and cups, and frequent feedings and bedding changes. Thus, hand-reared birds may have a generalized habituation to novelty that accounts for their very low levels of neophobia. This habituation to novelty may be similar to a phenomenon described by Francis et al. (2002), in which environmental enrichment reverses the effects of maternal separation on stress reactivity in rats. Although the increased stress reactivity, related to maternal separation in rats, generally persists well into adulthood (Meaney, 2001; Levine, 2001; Francis et al., 1999; Vazquez, 1997), maternally separated rats, housed in enriched conditions after weaning, exhibit decreased basal corticosterone levels and fearful behavior in adulthood (Francis et al., 2002). In fact, the behavior of maternally separated/enriched rats is comparable to that of neonatally handled rats (Francis et al., 2002), suggesting that high levels of environmental novelty can also cause significant behavioral changes in young animals. Environmental enrichment during periadolescence also reverses the reduction in play behavior and exaggerated HPA-axis response to restraint stress characteristic of rats exposed to prenatal stress (Morley-Fletcher et al., 2003). Laboratory rats housed with environmental enrichment showed significantly lower HPA axis responses to saline injection than rats housed without enrichment, suggesting that enrichment reduces animals reactivity to acute stressors, which may include novelty (Rhodes et al., 2003). Meehan and Mench (2002) also demonstrated that increasing the physical complexity of the cage by providing environmental enrichment reduces fear in juvenile orange-winged Amazons. The low levels of neophobia exhibited by the hand-reared birds were not persistent. Hand-reared birds exhibited moderate levels of avoidance of novel objects at 7 months of age, and by 12 months of age, they behaved in a manner almost indistinguishable from parent-reared birds in response to a novel object. This suggests that neophobia may be quite plastic in Amazon parrots, at least in juveniles. The plasticity of neophobia in older juveniles may also be related to environmental novelty. After weaning, both P/PH and H birds were housed under identical conditions. Enrichments were not changed more frequently than twice a month, and although the tamer birds were handled occasionally, hand-reared and parent-reared birds older than 3 months of age typically were visited by experimenters no more than twice per day, 4 5 days per week. During these visits, most birds were not handled, although they did have auditory and visual contact with the experimenters. Other than brief visits to check food and water, birds older than 3 months had no contact with humans on weekends. During hand-rearing, hand-reared birds had more frequent contact with experimenters. Although hand-reared birds typically were only directly handled by experimenters during feeding and bedding changes (each bird was probably handled for less than 3 min per feeding), the frequency of feedings meant that hand-reared birds in the nursery had min of auditory and visual contact with humans up to 5 times per day, 7 days/week. Thus, the level of environmental novelty experienced by the P/PH and H birds in post-weaning housing was probably lower than the level of novelty experienced by H birds in the nursery. Possibly, this decrease in novelty may have induced a de-compensation for environmental novelty, essentially the reverse situation to that described by Francis et al. (2002) for juvenile rats in an enriched environment.

12 128 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) Plasticity in neophobic behavior may be relevant ecologically. Mettke-Hoffman et al. (2002) argue that on a species level, neophobia and exploratory tendency in parrots is related to the costs (e.g., predation risk, risk of eating noxious food) and benefits (e.g., finding unexploited foraging sites, nest sites, or mates) of exploration in a particular environment. Parrot species that inhabit highly variable environments tend to be less neophobic (Mettke-Hoffman et al., 2002). Similar effects may also be important within species, with birds tending to be less neophobic during periods of environmental variability and more neophobic when the environment is relatively constant. This plasticity would allow parrots to more finely tune their behavior to environmental conditions. Plasticity may also be relevant to captive parrot welfare, since it suggests that fearful behavior in captive birds may be managed by manipulating the level of novelty in the birds environments. If adult parrots also respond to high levels of environmental novelty by decreasing neophobia, reducing neophobia in captive parrots may be a simple matter of frequently rotating enrichments in their cages and aviaries. Recent work by Meehan and Mench (2002) suggests that environmental enrichment may be a viable means of reducing fearfulness in captive parrots. 5. Conclusion Studies of rats have shown that early experiences can have a potent effect on the development of neophobia (Vazquez, 1997; Meaney, 2001), and the same appears to be true of orange-winged Amazon parrots. However, the mechanisms regulating the development of neophobia in rats and orange-winged Amazons appear to be different, while stress responsiveness is strongly correlated with maternal grooming in rats, there appears to be no association between grooming and neophobia in orange-winged Amazons. Until 6 or 7 months of age, hand-reared birds are significantly less neophobic than parent-reared birds, regardless of whether those parent-reared birds have been handled by humans. However, the fact that the differences between hand-reared and parent-reared birds in neophobia were not persistent suggests that neophobia may be quite plastic and that orange-winged Amazons may remain sensitive to environmental change well after weaning. A similar phenomenon has been seen in rats, whereby housing in enriched conditions after weaning can reverse the behavioral deficits (including high levels of neophobia) associated with maternal separation (Francis et al., 2002). These results suggest that it may be possible to manage neophobia in captive parrots with environmental enrichment. Acknowledgements Support for this project was provided by an NSF predoctoral fellowship to Rebecca Fox and by the Psittacine Research Project at the University of California at Davis. The authors would also like to acknowledge the contributions of the undergraduate research assistants who assisted with hand-rearing, parrot care, and data collection: Kyshia Davis, Daniela Muhawi, Amber Steinhauer, Lacy Hartford, Cam Pham, Jennifer McNulty, Kristy Smith, Jessica Dais, Sumi Seto, Caitlin Kelly, and Ryan Edwards.

13 R.A. Fox, J.R. Millam / Applied Animal Behaviour Science 89 (2004) References Benus, R.F., Bohus, B., Koolhaas, J.M., van Oortmerssen, G.A., Heritable variation for aggression as a reflection of individual coping strategies. Experientia 47, Capitanio, J.P., Personality dimensions in adult male rhesus macaques: prediction of behaviors across time and situation. Am. J. Primatol. 47, Carere, C., Welnik, D., Drent, P.J., Koolhaas, J.M., Groothius, T.G.G., Effect of social defeat in a territorial bird (Parus major) selected for different coping styles. Physiol. Behav. 73, Francis, D.D., Diorio, J., Plotsky, P.M., Meaney, M.J., Environmental enrichment reverses the effects of maternal separation on stress reactivity. J. Neurosci. 22, Francis, D.D., Diorio, J., Liu, D., Meaney, M.J., Nongenomic transmission across generations of maternal behavior and stress responses in the rat. Science 286, Gosling, S.D., John, O.P., Personality dimensions in nonhuman animals: a cross-species review. Curr. Dir. Psych. Sci. 8, Levine, S., Primary social relationships influence the development of the hypothalamic pituitary adrenal axis in the rat. Physiol. Behav. 73, Mason, W., Ontogeny of social behavior. In: Vandenbergh, P.M.a.J.G. (Ed.), Handbook of Behavioral Neurobiology: Social Behavior and Communication, vol. 3. Plenum Press, New York. Meaney, M.J., Maternal care, gene expression, and the transmission of individual differences in stress reactivity across generations. Annu. Rev. Neurosci. 24, Meehan, C.L., Mench, J.A., Environmental enrichment affects the fear and exploratory responses to novelty of young Amazon parrots. Appl. Anim. Behav. Sci. 79, Meehan, C.L., Garner, J.P., Mench, J.A., Foraging opportunity and increased physical complexity both prevent and reduce psychogenic featherpicking by young Amazon parrots. Appl. Anim. Behav. Sci. 80, Millam, J.R., Neonatal handling, behavior, and reproduction in orange-winged Amazons and cockatiels Amazona amazonica and Nymphicus hollandicus at the Department of Animal Science, University of California. Davis. Int. Zoo Yearbook 37, Morley-Fletcher, S., Rea, M., Maccari, S., Laviola, G., Environmental enrichment during adolescence reverses the effects of prenatal stress on play behavior and HPA-axis reactivity in rats. Eur. J. Neurosci. 18, Myers, S.A., Millam, J.R., Roudybush, T.E., Grau, C.R., Reproductive success of hand-reared versus parent-reared Cockatiels (Nymphicus hollandicus). Auk 105, Rhodes, M.E., Belz, E.E., Kennell, J.S., Czambel, R.K., Rubin, R.T., Effects of environmental enrichment on stress-responsive hormones in laboratory rats. FASEB J. 17, Vazquez, D.M., Stress and the developing limbic hypothalamic pituitary adrenal axis. Psychoneuroendocrinology 23, Verbeek, M.E.M., Drent, P.J., Wiepkema, P.R., Consistent individual differences in early exploratory behavior of male great tits. Anim. Behav. 48, Verbeek, M.E., Boon, A., Drent, P.J., Exploration, aggressive behavior, and dominance in pair-wise confrontations of juvenile male great tits. Behaviour 133, Verbeek, M.E., DeGoede, P., Drent, P.J., Wiepkema, P.R., Individual behavioral characteristics and dominance in aviary groups of great tits. Behaviour 133, Wilson, D.S., Adaptive individual differences within single populations. Phil. Trans. R. Soc. London Ser. B: Biol. Sci. 353,