S16.1: Relationships between juvenile object exploration and ecological specialisation in songbirds

Juvenile songbirds are often attracted to and manipulate a wide variety of objects – displaying strong neophilia (Greenberg 1984c, 1987, Heinrich 1995). Once mature and independent of parental care, songbirds are more often hesitant to approach a novel object – even when the object is placed near an attractive food source (Greenberg 1983, 1984c, Heinrich 1988). This ontogenetic pattern is not unique to birds, but is found widely among the higher vertebrates (Glickman and Sroges 1966).

The phenomenon of early-life neophilia and intense object play seems to be a costly behaviour and demand some explanation of its ultimate adaptive significance for its widespread maintenance in such a diversity of animal species. The costs incurred are the additional energy demands, and perhaps more importantly the risks incurred. Increased risks have not been measured, but involve increased activity and attraction of predators, as well as encountering dangers within the novel objects they manipulate.

The costs and benefits of neophilia/object play can be assessed at different levels of biological organisation. Since juvenile exploration is so widespread, one could approach the universal benefits conferred across the broad range of animals displaying these behavioural patterns (Glickman and Sroges 1966). Many studies have noted consistent differences in the form and intensity of these behaviours within species. To the degree these differences are genetically based and heritable, researchers can assess the factors that maintain this variation, whether it be continuous or take the form of behavioural polymorphisms. Finally, variation in novelty responses and object manipulation can be species-typical and, to the degree the variation is genetically based, can be seen as adaptations to the ecological challenges facing all individuals of a particular species. In this paper, I will focus on the species-level hypotheses.

In particular, I will focus on the Neophobia Threshold Hypothesis (NTH) which I developed in the late 1980’s to relate the neophilia/neophobia dynamic to the development of ecological plasticity in animals. Much of my work centred on the possible role of the variation in adult neophobia in governing plasticity. However, the factors that govern selectivity during the period of juvenile object exploration is an issue of importance to the NHT hypothesis and I will discuss the small amount of work that has been conducted on this topic as well.

The Neophobia Threshold Hypothesis was developed based on a series of studies of congeneric warblers (Dendroica) and sparrows (Melospiza). The hypothesis (Greenberg 1983) does not address variation in the nature of early exploration, but rather on the effect of interspecific variation in the intensity of adult neophobia given that early exploration conveys a familiarity which has salience into adulthood. I will present the hypothesis as it was fully stated in Greenberg (1990a): (1) birds respond differentially to novel stimuli; (2) adult birds often respond with acute stress, which leads to avoidance and excitement; (3) the attraction a potential food source and the fear response induced by novelty produce a tension, which is ultimately resolved either through habituation of by avoidance. The greater the intensity of the initial fear response, the less likely the individual will explore or feed at the novel stimuli; (4) the fewer novel microhabitats or food items approached, the fewer opportunities will be available for a bird’s foraging repertoire. The result will be observed as reduced ecological plasticity; (4) exploration is largely relegated to the juvenile period, where fledgling birds are attracted to a wide variety of novel objects. Neophilia in juvenile birds makes this period particularly important for shaping the foraging niche of a species. The more intense the neophobia of adults, the more important the period of early exploration will be. (It should be noted that filial imprinting is based on early neophilia followed by extreme neophobia as well – Bateson 1979) and (5) although neophobia may be ameliorated by environmental, social, or internal-state factors, the normative response of individuals is heritable and differences between species tested under similar conditions are largely genetically based.

If supported the NTH provides a relatively simple behavioural mechanism underlying ecological adaptability. Ecological adaptability is driving forces in evolution (Wyles et al. 1983) . After all, new niches and new adaptive zones are generally thought to be explored and incorporated through simple shifts and behaviour long before morphological adaptations can be fine-tuned through natural selection. On the other hand ecological specialisation is the key to success in saturated and stable communities. Specialised animals require mechanisms to maintain specialisations in many circumstances. Most animals lie between these extremes and selection should adjust the equilibrium between exploratory tendencies and stereotypic preference to some optimal level. Still, however elegantly the NHT can explain the evolution of optimal ecological plasticity in birds, it remains largely untested and in need of fine-tuning in the face of a few experimental results.

Most of the research on the NTH has focused on the differential response of birds to approach novel objects – without regard to the role of the exact nature of early exploration. Research on the Dendroica warblers and several experiments on sparrows focused on the neophobic response of birds that grew up in the wild and were captured as immatures in the first autumn migration (Greenberg 1983, 1984b, 1989, 1990b). In the case of a field experiment, the age of birds was often not known – although most birds were immature as well.

The strategy of the experiment to measure feeding neophobia is quite simple and similar to earlier experiments on container neophobia in rodents (Barnett 1958, Mitchell 1976, Cowan 1977). Food is offered to the experimental animal in the presence or absence of a series of unfamiliar objects. The presenting of familiar and attractive food in the presence of an object effectively separates the stimulus of the animal to feed from the response to the novel object. The latency to feed at the object can be compared to a base-line latency to feed at the same food receptacle without the object. In addition, the latency to feed immediately after the object is removed can be used to evaluate the effect of the disturbance associated with the set-up of the experiment. Finally, habituation can always be demonstrated, so that the object once it has become familiar no longer elicits a latency response. In experiments on wild-caught birds the experience of the subjects cannot be known. However, I tested a wide variety of objects that would not be in the natural habitat of any of the species and looked for a consistently greater latency across all objects. I reasoned that some emergent property of the objects, such as novelty, rather than a particular feature of any one object could best explain a general difference in response.

The first experiments establishing a relationship between neophobia and ecological plasticity were conducted on tropical wintering species of Dendroica warblers. Based on extensive field studies in lowland tropical forest of Panama (Greenberg 1984a), I determined that the Chestnut-sided Warbler Dendroica pensylvanica was a relatively specialised foliage-insectivore throughout the non-breeding season, whereas the Bay-breasted Warbler D. castanea displayed a high degree of foraging generalisation and seasonal variability. In addition, this species is highly omnivorous and was observed to employ innovative foraging (ground foraging, feeding at screens under insect lights etc.). Using the above experimental paradigm, I showed that Bay-breasted Warblers approached novel objects more readily and had a far lower latency to feed in the presence of novel objects (Greenberg 1983, 1984b). I proposed that conifer-breeding Dendroica were more likely be omnivorous and generalised in non-breeding foraging behaviour and a comparative experiment with a number of species in the genus supported the hypothesis that the conifer breeders were generally less neophobic than broad-leaf breeding species.

The second set of experiments focused on two species of sparrows in the genus Melospiza. The Song Sparrow M. melodia is a widespread species noted for its highly generalised habitat use and ecology. The Swamp Sparrow M. georgiana is a specialist on marsh habitats. I tested these species by comparing visitation rates to feeders where both species occurred with and without associated novel objects and found a decline in visitation rates by Swamp but not Song sparrows (Greenberg 1989). In the laboratory, I found that wild-caught immature Swamp Sparrows showed a much greater latency to feed at a variety of novel objects than Song Sparrows (Greenberg 1990b)

The role of early object exploration, although integral to the NTH, has received considerably less experimental attention. I conducted two experiments that address this aspect of the hypothesis: one on Chestnut-sided Warblers and one on two species of Melospiza. Both experiments employed the same basic strategy: nestlings were raised and released upon fledging into aviaries with a series of natural and artificial objects selected randomly from a larger set of possible objects. The objects were visited frequently and usually mandibulated. The saliency of novelty was demonstrated for the Chestnut-sided Warblers by periodically placing a new object in the flight cage. In all cases the new object attracted the attention of and visitation by the fledgling warblers. After independence (September-October), mature birds were tested for their response to feeding in the presence of the objects explored during the fledgling period and those never seen before. For all three species tested, latency to feed at the explored objects was significantly lower than for the novel objects (Greenberg 1984c, 1992). These experiments suggest a long-term efficacy of the early exploration. A detailed comparison of the two sparrow species showed no difference in the diversity of objects explored as juveniles (Greenberg 1992) – and yet large differences in neophobia upon achieving independence and maturity. This provides support for placing the emphasis of the NHT on variation in neophobia and not juvenile exploration (an issue to which I will return).

In terms of evolutionary arguments, the sixth point of the NTH – the presence of genetically based variation in neophobia -- is probably the most critical. Yet this is where the experimental results for birds are most problematic. Studies of related species and strains of Rattus strongly suggest that variation in neophobia is heritable. However, the only experiment with which I am familiar for wild birds is one I conducted on the comparative neophobia of the Song and Swamp sparrows (Greenberg 1992). As stated above, birds tested in the field and in the laboratory showed strong and consistent differences in neophobia which are consistent with the predictions based on the ecology of these species. The more specialised Swamp Sparrow was much more averse to approaching novel objects to feed than its ecologically adaptable congener. When young birds were reared under identical conditions and exposed to a similar set of objects during their post-fledging exploratory period, I found that when tested as adults the neophobic response was much greater in the Song Sparrow than the Swamp Sparrow – in direct opposition to the results with wild-caught or wild birds. This result, particularly if replicated in other species-pairs, is suggests that, if anything, that observed differences in neophobia in wild-reared immatures is a consequence and not a cause of different experience. This conclusion, if generally confirmed, is damaging to the NTH in its simplest form since it shows that birds reared under identical conditions do not show the same relative neophobic response as wild-caught immatures (presumably reared under different conditions). Perhaps these results are an artifact of laboratory rearing. In this case, the most convincing test of the possible genetic basis of neophobia would be (the very difficult) cross-fostering experiments.

The importance of neophobia in foraging of adult birds (and many other vertebrates) has been well studied. Studies of birds that have invoked novelty responses as particularly important include a wide taxonomic group, including Blue Jays Cyanocitta cristata, (Coppinger 1970), Snail Kites Rostrhamus sociabilis (Beissinger et al 1994), Moustached Warbler Acrocephalus melanopogon (Raach and Leisler 1989), jacamars (Chai 1996), Common Ravens Corvus corax, (Heinrich et al. 1995) and Zebra finches Taeniopygia guttata, (Coleman and Mellgren 1994) as part of a study of social influences on foraging. For the most part, these studies did not relate the presence of neophobia or its intensity to the degree of foraging specialisation of the study species. In Snail Kites, neophobia was implicated as an important mechanism for maintaining this species extreme dietary specialisation. On the other hand, Mönkkönen and Koivula (1993) demonstrated that naïve individuals of the generalist Willow Tit Parus montanus showed little evidence of neophobia. However, because none of the studies were done in a comparative experimental paradigm, they do not explicitly address the possible role of variation in neophobia and its implications for foraging specialisations which is the core of the NTH. One comparative study of primates that produced results consistent with the hypothesis compared the specialised (and neophobic) marmosets (Saguinus) with the adaptable (and less neophobic) marmosets (Callathrix, Tardiff 1994).

The NTH depends upon exploration of novel objects by young and does not address the potential role that young birds play in actively selecting objects to approach and explore. Innate preferences may play a large role in determining what is visited and hence what becomes familiar in a manner analogous to the role of preference in filial imprinting and song learning (Bateson 1979). Specialised foragers might focus their attention on objects that in some way reflect their adult foraging specialisation. On the other hand, juveniles may possess few consistent preferences or preferences may be expressed idiosyncratically between individuals. In the latter case, the parents may play an active role – as young in most songbirds can be expected to follow adults and therefore be amenable to site and local enhancement. Keep in mind that for most of this period, young are not self-feeding and therefore trial-and-error learning based on foraging success itself cannot play a role. However, it has been suggested that during the period of transition from dependence to self-feeding, manipulation of small objects might be a winnowing process by which birds learn what food items are (Marchetti and Price 1989 and reference therein, Heinrich 1995).

I conducted one comparative study of object exploration using two ecologically similar species displaying great differences in foraging plasticity. As an example of a microhabitat specialist, I selected the Worm-eating Warbler Helmitheros vermivorus, a Neotropical migrant that forages predominantly for arthropods in dead curled leaves that hang in the forest understory (Greenberg 1987) during the non-breeding season. The generalist species was the Carolina Chickadee Parus carolinensis which often forages from dead-curled leaves (among many other microhabitats) and shares the warblers use of hanging and manipulation of substrates.

As in the neophobia experiment, the strategy of the object exploration experiment is quite simple. I released hand-raised fledglings into flight cages with established stations for the presentation of a set of objects. Objects are rotated around the flight cage over a four-week period so that each object is located at every possible site at least three times. Between feeding the still-dependent young birds, observers recorded visits to the objects. The number of independent visits to each object is tallied over the course of the experiment and the relative number of visits is used as a measure of preference.

Three major results came from the experiments on Worm-eating Warblers. First, the object of adult specialisation, dead curled leaves, was strongly preferred as an object of exploration by juveniles. A few comparative experiments with other warblers showed that the Worm-eating Warbler was the only species for which dead leaves was the preferred object. This provides some assurance that the preference is species-typical and related to foraging in adults. Second, overall the preference ranking for all objects (two different sets were used) was strikingly high (r = 0.92). Third, and perhaps most interesting was that individuals raised together or in isolation from each other had similarly high correlations of preference (both r = 0.92). This result shows that the preferences are based on internally derived preferences and not social response to the preference of an individual in the experiment. From these observations, I concluded that the object exploration of juvenile Worm-eating Warblers is highly stereotyped and based on innate biases.

These results are in strong contrast to those obtained for the generalist chickadee. The Chickadees showed no consistent preference for any one object and low overall consistency of ranking (r = 0.62). The mean diversity index for foraging visits was significantly higher for chickadees than warblers (Simpson Index 6.9 versus 5.4). Birds from three groups raised in isolation had high within-group consistency of rankings (r = 0.87) and low between group consistency (r = 0.42). These groups were social – but not genetic broods. I concluded that chickadees have little internal structure to the preferences for exploration but are quite susceptible to social influences in their preference.

Object play has been reported for the young of birds across a broad taxonomic spectrum (Ortega and Beckoff 1987). It has been particularly well documented in raptors, parrots, and passerines. Young birds are attracted and manipulate objects and this tendency is less prone to extinction than in adult birds (Vince 1958, 1960). However, the topic of preference in juvenile object exploration has been the subject of almost no observational or experimental work. However, one recently published study (Negro et al. 1994) on object play in American Kestrels Falco sparverius produced results that are parallel to those described in the warbler/chickadee experiment above. In this experiment hand-raised fledgling kestrels were offered a series of objects and they consistently selected mouse or mouse-like objects to manipulate. In addition, kestrels reared together showed a significant tendency to prefer to play with the same objects – thus suggesting the role of social factors in shaping exploratory behaviour.

Juvenile exploration was examined experimentally for the extreme generalist, the Common Raven Corvus corax (Heinrich 1995). Although it has long been believed – based on anecdotal observation – that young ravens are attracted to certain kinds of objects (e.g., shiny ones), these experiments suggested that they approach and manipulate all novel objects and their preferences are winnowed down as ravens mature. Sub-adult ravens then become highly neophobic to new foods and objects (Heinrich 1988, Heinrich et al. 1995). Heinrich showed that in this species, the degree of distinction of an object and the surrounding objects was a dominant feature determining its attractiveness. In this case, the neophilia-neophobia model, without additional structure to juvenile object exploration, appears sufficient to describe the maturation of behaviour in ravens.

If the period of neophilia is relatively unstructured by innate preference, then the role of learning during the period of natal association with parents and other conspecifics takes on major importance in establishing the precise nature of specialisation in highly neophobic birds. However, the research on Worm-eating Warblers, however, suggests that at least in this species, the period of juvenile neophilia may in fact be structured by innate preferences. To the degree that other specialised species show consistent structure to their juvenile object preferences, it might be more reasonable to suggest that neophobia acts to further constrain innate biases rather than winnowing preferences in the manner suggested by Heinrich (1995). The haphazard or unstructured pattern of object preference found in Ravens and chickadees would be consistent with the NTH as originally formulated. However, such exploration has only been documented for two omnivorous-generalist species.

The results of the Worm-eating Warbler and American Kestrel experiments suggest that beyond any neophilia that might exist, young birds have innate preferences in their choice of object to explore and manipulate. These results cannot be taken too seriously until more work on the functional significance of object play is completed. In the case of the warbler and the raptor, adult foraging requires complex prey or prey-site manipulation. For raptors this is obvious – in the case of the warbler it needs to be pointed out that forging for hidden prey in dead leaves requires special manipulation of the leaf microhabitat. In both cases, the role of practice and exercise in determining object play needs to be considered. West (1977), for example, showed that kittens were attracted initially to novel object, but played longer and used different motor patterns with certain objects conducive to prey-handling type movements

The need for more comparative research

The NHT lacks tests from appropriate comparative studies. This is surprising since in the past two decades, a number of researchers have called for a more ecological approach to the study of learning mechanisms in animals (e.g., Johnston 1981, Kamil and Yoerg 1982). Specifically, they call for more focus on the role of specific aspects of foraging ecology on variation in components of learning, such as memory storage and cognition to complement earlier theories based on general mechanisms of learning and broad taxonomic differences in learning abilities. Although object exploration and neophobia do not involve cognitive behaviour per se, both aspects of behaviour play a central role in how animals learn about their environment and seem quite amenable to a more fine-tuned comparative approach. Therefore, it remains surprising how few explicitly comparative studies have been published on these topics. The natural world is full of examples of closely related species that differ in ecological plasticity and each of these ecological comparisons is a potential gold mine for exploring the hypotheses presented here, as well as those not yet developed.

The NHT has not incorporated analysis of what maintains variation in novelty responses within populations. The focus of this paper has been on differences in species-typical norms of exploratory behaviour and neophobia. Behavioural ecology has focused increasingly in intraspecific variation and selective factors that maintain both continuous behavioural variation and genetically based behavioural polymorphisms. It could then be argued that the NTH and hypotheses attempting to explain differences in juvenile exploration are old-fashion in their exclusive approach to species-typical behaviour.

In fact, intraspecific variation in ‘temperament’ (Kagen et al. 1988) or variation along a shyness-boldness continuum (Wilson et al. 1994) is an active area of both empirical research and theorising. Such variation has been particularly well documented in humans and other primates (Kagen et al. 1988). The concept of intraspecific variation in temperament brings up two issues: First, to what degree are ‘shyness’ and ‘boldness’ unitary phenomena (Greenberg 1995, Drent and Marchetti 1999, K. Coleman and D.S. Wilson unpubl.)? Response to novel objects in the context of foraging may not necessarily be correlated with the boldness in social interactions or the tendency to forage in risky situations (but see Drent and Marchetti 1999). The second issue, focusing on novelty responses alone, is what factors select for stable polymorphism in the response types (for example, see Wilson et al 1994). Can such variation be related to variable foraging strategies that have been hypothesised to characterise groups (e.g., producers versus scroungers – K. Coleman and D.S. Wilson unpubl., Drent and Marchetti 1999). If variation in novelty responses characterises – but is not determined by – social roles, then we can postulate the in social species with generalist resource use, the social system selects for greater variance in neophobia. If such intraspecific variation can be shown to have a strong heritable genetic component, then we would predict that social species could evolve optimal levels of ecological plasticity in new habitats more rapidly than solitary species (Greenberg 1995). The outcome of the recent controversy regarding the genetic basis of novelty-seeking behaviour in humans should be of great interest to behavioural ecologists (Benjamin et al. 1996, Ebstein et al. 1996, Sullivan et al 1998).

It has been over a decade since the NHT was first formulated. My review of the literature turned up relatively few citations on the topic during the ensuing period. I found this surprising considering the central role the novelty responses play in the life of birds, particularly regarding foraging decisions and habitat selection. Although the hypothesis remains largely untested and perhaps naïve in some of the details, it remains one of the few truly evolutionary hypotheses focused on novelty responses capable of generating specific predictions regarding interspecific variation.

The lack of attention is particularly surprising considering the attention that has been paid to the evolution of behavioural (i.e., ecological) plasticity in vertebrates (see Wyle et al. 1989, Lefebre et al. 1997 and references therein) and what seems to me to be the obvious connection between neophobia and plasticity. However, the emphasis of the theoretical and empirical work on ecological plasticity has been almost exclusively on its relationship to variation in cognitive abilities and their morphological correlates in brain (particularly fore-brain). In these discussions, emotional responses – including those to novelty -- are generally not mentioned. One recent review mentioned neophobia only to dismiss it as an artifact of captive experiments, suggesting that captive research was therefore irrelevant to understanding foraging ‘innovations’ in the real world (Lefebre et al. 1997. My speculation is that being uniquely large-brained creatures interested in orienting our science to the evolution of our own attributes, we tend to focus our research on cognitive rather than emotional issues. As comparative psychologists have long known, however, many responses that appear to reflect differences in cognitive ability are actually the result of variation in emotional responses to the task at hand (i.e., Scott and Fuller 1965). Perhaps comparisons within and between populations and between related species should emphasise the study of both temperament and cognitive ability in determining factors that lead to differences in ecological plasticity.

The NTH posits that bird species share a period or intense neophilia followed by adulthood characterised by variable intensity of neophobia. Neophilia is a creative period of exploration and neophobia enforces specialisations developed during juvenile exploration. Differences in ecological plasticity stem from variation in the intensity of adult neophobia. This attractive proposition provides a relatively simple behavioural substrate upon which natural selection can act to regulate ecological plasticity. However, ten years after its development, the NHT remains largely ignored and suffers from three empirical problems that demand further investigation: 1) few studies have tested the variation in neophobia with ecologically and evolutionary appropriate comparisons; 2) the one study that examined the stability of differences in laboratory-reared individuals produced contradictory and difficult to explain results; and 3) in at least two studies, birds have been shown to have strong innate biases in the types of objects explored in captivity. The degree to which such biases shape juvenile exploration will effect the role that juvenile neophilia has in shaping the specifics of adult specialisation. Finally, 4) the basis for intraspecific variation in temperamental variables such as neophilia and neophobia, particularly in social species, needs to be incorporated into models of how these characteristics evolve in populations.

Research was supported by Post-doctoral fellowships from the Smithsonian Institution and Friends of the National Zoo, and NSF Grant BSR-8705003. E.S. Morton provided advice and support throughout.

Barnett, S.A. 1958. Experiments on ‘neophobia’ in wild and lab rats. British Journal of Psychology 49:195-201.

Bateson, P. 1979. How do sensitive periods arise and what are they good for? Animal Behaviour 27:470-486.

Beissinger, S.R., Donnay, T.J. & Walton, R. 1994. Experimental analysis of diet specialisation in the Snail Kite: The role of behavioural conservatism. Oecologia 100:54-65.

Benjamin, J., Li, L., Patterson, C., Greenberg, D.B., Murphy, D.L. & Hammer. D.H. 1996. Population and familial association between the D4 dopamine receptor gene and measures of novelty seeking. Nature Genetics 12:81-84.

Chai, P. 1996. Butterfly visual characteristics and ontogeny of responses to butterflies by a specialised tropical bird. Biological Journal of the Linnean Society 59:37-67.

Coleman, S.L. & Mellgren, R.L. 1994. Neophobia when feeding alone and in flocks in Zebra Finches Taeniopygia guttata. Animal Behaviour 48:903-907.

Coppinger, R.P. 1970. The effect of experience and novelty on avian feeding with reference to the warning coloration of butterflies. II. Reactions of naïve birds to novel insects. American Naturalist 194:323-337.

Cowan, P.E. 1977. Neophobia and neophilia: hew object and new-place reactions of three Rattus species. Journal of Comparative and Physiological Psychology 90:190-197.

Drent, P.J. & Marchetti, C. 1999. Early exploratory behaviour and individuality. In: Adams, N.J. & Slotow, R.H. (eds) Proc. 22 Int. Ornithol. Congr., Durban: 896-914. Johannesburg: BirdLife South Africa.

Ebstein, R.P., Novick, O., Umansky, R., Priel, B., Osher, Y., Blaine, D., Bennett, E.R., Nemanov, L., Katz, M. & Belmaker, R.H. Dopamine receptor (D4DR) exon III polymorphism associated with the human personality trait of novelty seeking. Nature Genetics 12:78-80.

Glickman S.E. & Sroges, R.W. 1966 . Curiosity in Zoo Animals. Behaviour 24:151-188..

Greenberg, R. 1983. The role of neophobia in foraging specialisation of some migrant warblers. American Naturalist 122:444-453.

Greenberg, R. 1984a. The winter exploitation systems of Bay-breasted and Chestnut-sided warblers in Panama. University of California Publications in Zoology 117:1-124.

Greenberg, R. 1984b. Differences in feeding neophobia between two species of tropical migrant warblers (Dendroica castanea and D. pensylvanica). Journal of Comparative Psychology 98:131-136.

Greenberg, R. 1984. Neophobia in the foraging site selection of a Neotropical migrant bird: an experimental study. Proceedings of the National Academy of Sciences, USA 81:3778-3780.

Greenberg, R. 1987. The development of dead leaf foraging in a Neotropical migrant warbler. Ecology 68:130-141.

Greenberg, R. 1989. Neophobia, aversion to open space, and ecological plasticity in Song and Swamp sparrows. Canadian Journal of Zoology 67: 1194-1199.

Greenberg, R. 1990. Ecological plasticity, neophobia, and resource use in birds. Studies in Avian Biology 13: 431-437.

Greenberg, R. 1990. Feeding neophobia and ecological plasticity: A test of the hypothesis with captive sparrows. Animal Behaviour 39: 375-379.

Greenberg, R. 1992. Differences in neophobia between naïve song and swamp sparrows. Ethology 91:17-24.

Greenberg, R. 1995. Novelty responses: the bridge between psychology, behavioural ecology, and community ecology. Trends in Ecology and Evolution 10:165-166.

Heinrich, B. 1988. Why do ravens fear their food. Condor 90:950-952

Heinrich, B. 1995. Neophilia and exploration in juvenile Common Ravens Corvus corax. Animal Behaviour 50: 675-704.

Heinrich, B., Marzluff, J.M. & Adams, W. 1995. Fear and food recognition in naïve Common Ravens. Auk 112:499-503.

Johnston, T. D. 1981. Contrasting approaches to a theory of learning. Brain and Behaviour Sciences 4:125-173.

Kagan, J., Reznick, J.S. & Snidman, N. 1988. Biological basis of childhood shyness. Science 240:167-171.

Kamil, A.C. & Yoerg, S. 1982. Learning and foraging behaviour. Pp. 325-362 in (P. Bateson and P. Klopfer, eds.) Perspectives in Ethology. Volume 5. Plenum Press, New York.

Lefebvre, L., Whittle, P., Lascaris, E. & Finkelstein, A. 1997. Feeding innovations and forebrain in birds. Animal Behaviour 53:549-560.

Marchetti, K & T. Price, T. 1989. Differences in the foraging of juvenile and adult birds: The importance of developmental constraints. Biological Reviews of the Cambridge Philosophical Society 64:51-70.

Mitchell, D. 1976. Experiments on neophobia in wild and laboratory rats: a reevaluation. Journal of Comparative and Physiological Psychology 90:190-197.

Mönkkönen, M. & Koivula, K. 1993. Neophobia and social learning of foraging skills in Willow Tits , Parus montanus. Ardea 81: 43-45.

Negro, J.J., Bustamante, J., Milward, J. & Bird, D.M. 1996. Captive fledgling American kestrels prefer to play with objects resembling natural prey. Animal Behaviour 52:707-714.

Ortega, J.C. & Bekoff, M. 1987. Avian play: evolutionary and developmental trends. Auk 104:338-341.

Raach, A. & Leisler.B. 1989. Auswirkung der Jugenderfahrung auf die Wahl von Habitatstrukturen und auf das Erkundungsverhalten des Mariskensängers (Acrocephalus melanopogon). Journal für Ornithologie 130:256-259.

Scott, J.P. & Fuller, J.L. 1965. Genetics and the social behaviour of the dog. University of Chicago Press, Chicago.

Sullivan, P.F., Fifield, W., Kennedy, M., Mulder, R.T., Sellman, J.D. & Joyce, P.R. 1998. No association between novelty seeking and type 4 dopamine receptor gene (DRD4) in two New Zealand samples. American Journal of Psychiatry 155:6.

Tardiff, S. 1994. Sensitive Saguinus and courageous Callithrix: Differences in adaptability and neophobia in two Callitrichid genera. Proceedings of the American Zoo Association Regional Conference 1994:408-412.

Vince, M.A. 1958. Differences related to age in greenfinches, chaffinches and canaries. Animal Behaviour 6:53-59.

Vince, M.A. 1960. Developmental changes in responsiveness in the Great Tit Parus major. Behaviour 15:219-242.

West, M. 1977. Exploration and play with objects in domestic kittens. Developmental Psychobiology 10:53-57.

Wilson, D.S., Clark, A.B., Colman, K. & Dearstyne, T. 1994. Shyness and boldness in humans and other animals. Trends in Ecology and Evolution 9:442-446.

Wyles, J.S., Kunkel, J.G. & Wilson, A.C. 1983. Birds, behaviour and anatomical evolution. Proceedings of the National Academy of Sciences, USA. 80:4394-4397.