In the early 1920s, blue tits across Britain were observed stealing milk from the bottles on people’s doorsteps by prizing up, or pecking open, their foil tops (Fisher and Hinde 1949). While this behavior alone is interesting enough, its spread through blue tit populations is striking, because birds all over the country adopted the behavior so rapidly that only some form of social transmission could explain it. Indeed, subsequent research into blue tit milk-bottle-opening has shown that the transmission of the trait between individual birds, along with its maintenance in populations over multiple generations, was1 dependent on a form of social learning – specifically, local enhancement (Sherry and Galef 1990; Sherry and Galef 1984).2
Local enhancement is a widespread and relatively “simple”3 form of social learning. It “occurs when, after or during a demonstrator’s presence, or interaction with objects at a particular location, an observer is more likely to visit or interact with objects at that location” (Hoppitt and Laland 2013: 64; Thorpe 1963). In mediating the transmission of learned traits between individuals, it can allow behavioral innovations to spread through animal populations. To illustrate, in the case of blue tit milk-bottle-opening, it is thought that an individual bird (or, more likely, birds (Lefebvre 1995)) hit upon the initial milk-bottle-opening innovation via a combination of luck and asocial learning (such as trial-and-error learning).4 Following the initial innovation, naïve individuals were then drawn to the milk bottles by local enhancement (i.e. they were more likely to interact with milk bottle tops, having seen the successful demonstrators do so). The naïve birds then, through their own trial and error, learned to open the milk bottles themselves. Given that there are very few possible successful milk-bottle-opening techniques, the same techniques then proliferated through the population. In this way, the milk-bottle-opening trait became widespread and persisted until milk bottle delivery was phased out some years later. Importantly, although asocial learning plays a significant role here, the persistence of the milk-bottle-opening behavior over multiple generations is heavily reliant on social learning (Sherry and Galef 1990; Sherry and Galef 1984).5
Blue tits are not the only nonhuman species able to transmit behavior over multiple generations through “simple” forms of social learning such as local enhancement. Stimulus enhancement, observational conditioning, response facilitation, social enhancement, observational R-S learning, emulation, opportunity providing, and inadvertent coaching6 have also been shown to mediate the transmission and maintenance of various traits in many animals not usually considered particularly cognitively complex, including other species of birds (Krebs and Kroodsma 1980), fish (Brown and Laland 2003; Laland et al. 2003), and rodents (Eisner and Terkel 1991). Importantly, what these cases – which I will broadly call “animal traditions”7 – demonstrate is that social transmission is not restricted to so-called “smart” animals such as chimpanzees and cetaceans. Indeed, recent laboratory work eliciting the social transmission of a learned skill over multiple generations in bumblebees (Alem et al. 2016) suggests that the basic cognitive ingredients for animal traditions are relatively unsophisticated, and exist far outside those lineages in which we would typically imagine them to reside.
The significance of such (nonhuman) animal traditions is debated (Laland and Galef 2009; Avital and Jablonka 2000; Fragaszy and Perry 2003). Historically, however, those interested in animal cognition, and behavioral and cognitive evolution, in animals (comparative psychologists, behavioral ecologists, ethologists, and animal behaviorists), have tended to overlook them. While there is, of course, much discussion in the literature on social learning and its role in the individual life histories of animals, the type of transgenerational transmission which characterizes animal traditions is often assumed to be of little explanatory importance outside of the primate lineage. In this short chapter, I consider why this has been the norm, and offer some reasons for why animal traditions are far from unimportant in understanding both the evolution of cognition and its proximate mechanisms. To begin with, we shall look at two key claims which motivate the historical view. The first relates to how well simple forms of social learning transfer behaviors between individuals.
The “simple” mechanisms of social learning underwriting most animal traditions are widely viewed to be incapable of accurately transmitting traits across multiple generations (Laland and Janik 2006; Laland and Galef 2009; Avital and Jablonka 2000; Laland and Hoppitt 2003). We can illustrate this with the blue tit case. Local enhancement does not facilitate the transfer of a particular milk-bottle-opening behavior between individuals; rather, it increases the likelihood of naïve birds hitting upon the milk-bottle-opening behavior via their own asocial learning. While in this circumstance there are very few ways to open a milk bottle, so we see the spread of a relatively homogenous behavior throughout the population,8 it is easy to imagine cases where this would not be so. For example, imagine if, rather than requiring an imprecise peck on the lid of the bottles, opening the milk bottles involved a precise or unobvious sequence of behaviors. In such a circumstance, we could reasonably predict that mere local enhancement would be insufficient for milk-bottle-opening to persist in a population. In effect, the animals in question would be in a game of telephone, in which the information required to open the milk bottles is eroded from the population with each imperfect transfer. Not only would it appear impossible for any complex behavioral innovations that arise in such populations to persist via social learning alone, but also unlikely that complex behavioral traits could evolve through any cumulative evolutionary process (such as natural selection) in which social learning acted as a route of multigenerational inheritance.
Expanding a little on this point, the evolution of complex adaptations via natural selection usually involves the gradual accumulation of small fitness-enhancing innovations over multiple generations. It is only via such gradual change that a process like genetic mutation, which is blind to adaptive benefit, is likely to generate complex adaptations. This is because large, undirected changes to the phenotype of an organism (through, say, the duplication or translocation of a large swathe of DNA) are more likely to reduce, rather than increase, the adaptive fit of that organism with their environment. Griffiths and Sterelny (1999) illustrate this nicely with the analogy of trying to crack a combination lock. For a combination lock with six wheels numbered 0–9, for example, there are 151,200 different possible combinations of the wheels if each number can only be used once, and a million different possible combinations if each number can be used more than once. If a safecracker were to attempt to open the lock by getting every number in the combination correct in one go, they would far more likely than not take a very, very long time to succeed. Imagine, however, that there is a “click” when each wheel of the lock is turned into the right position. By carefully turning the first wheel and waiting for the click, and then the second, and so on, the safecracker could quite quickly and easily determine the correct combination to open the lock, and surely in far less time than if they went with the aforementioned wholesale approach. As Griffiths and Sterelny (1999) note, natural selection works something like a safe cracker listening for clicks. Just as there are many more incorrect than correct states of the wheels of a combination lock, there are many more ways for organisms to be maladapted to their environments, and only a few ways to be adapted. By retaining small adaptive innovations (like the safe cracker retaining “one wheel right”) and building upon them, however, populations under natural selection gradually build up the required innovations for complex adaptations to evolve. In effect, natural selection solves the problems in the environment through a process of trial and error coupled with the ability to retain small beneficial adaptations, just like the safecracker turning each wheel in turn, listening for clicks.
Although the various mechanisms of asocial learning seen in the animal world are not completely blind to adaptive benefit in the way that genetic mutation is, the most common form of asocial learning – trial-and-error learning – is not “insightful”. Animals learning via trial-and-error are not solving the problems they face “in their heads” and then executing the most adaptive behavior the first time. Rather, they are innovating in a relatively random manner and repeating those behaviors which produce a reward. In a manner akin to that described for mutation and selection above, this means that asocial learning is unlikely to facilitate the development of complex behaviors in organisms through anything but an incremental process. Although large adaptive behavioral innovations are, in principle, possible via trial and error, as with genetic mutations, there are many more ways in which an organism can be behaviorally maladapted than adapted to their environment. Small variations on what is currently working are more likely to be, if not adaptive, neutral with respect to survival and reproduction, than large innovations.
Returning again to the hypothetical “complex” milk-bottle-opening technique case again, it is now clear why fidelity of transfer is important. Recall, here we are concerned with a non-obvious and complex behavioral trait. For such a trait to arise and persist, it seems correct to assume that the small innovations in the milk-bottle-opening technique that individuals hit upon via their own asocial learning must be propagated and maintained in the population via an accurate mechanism of transmission between individuals (i.e. an accurate social-learning mechanism). If innovations are not able to be transferred between individuals in the population with sufficient fidelity for them to be gradually and cumulatively built upon over time, however, adaptive complexity would be highly improbable.
If this pessimistic, traditional picture of the limited power of simple forms of social learning is correct, any behaviors, skills, or information that is transmitted trans-generationally via “simple” social-learning mechanisms are best understood as temporary effects of the environment. On such a view, any adaptive benefit that such traditional resources confer on organisms is easily accounted for by reference to the fitness of the genes producing the cognitive capacities that explain them given a particular environmental context. In short, although socially learned behaviors, skills, and information can no doubt be fitness enhancing, when we think about how that fitness benefit is transmitted between generations, it is the genetic inheritance of the social-learning capacity that should be invoked rather than the inheritance via social learning of the particular behavior, skill, or bit of information.
An example of this disinterest in animal traditions on the grounds that they lack any long-term evolutionary significance can be seen in the discussion of putative cases of animal culture. As outlined by Ramsey and Goodrich (Chapters 32 and 33, respectively, in this volume), many of those interested in cultural evolution, cultural inheritance, and the cognitive underpinnings of culture have eschewed animal traditions as being different in kind to “true” culture (i.e. the type of complex customs, tools, and structures seen in human societies). Culture, it is claimed, is complex and cumulative, relying heavily on the presence of cognitively demanding forms of social learning, such as explicit teaching and copying. Animal traditions, in contrast, can only ever be rudimentary in nature because the “simple” learning mechanisms that underwrite them are unable to facilitate the persistence of cumulative or complex traits. Many advocates of this type of view go on to argue that humans are unique in having culture and in undergoing any form of culturally mediated evolution (Laland and Galef 2009).
To illustrate, Michael Tomasello (2014, 2009) (a prominent player in these debates) argues that only humans are capable of cumulative culture, as it is reliant on a unique human capacity for joint attention (the ability to share attention on an object with another individual by, for example, pointing). Without joint attention, he claims, humans would be unable to engage in the type of imitative and linguistically based learning that is required for the cumulative, complex culture we only see in our own species. Entailed by this account of human culture is the assumption that nonhuman animals (lacking in joint attention, and thus imitation and linguistic social learning) are incapable of cumulative, complex cultures. Bennett Galef (2012, 2009) expresses a similar view, arguing that imitation and teaching are far more developed in humans than any other species (and it is only this more developed sort of social learning that can support what we would call culture).
This orthodoxy has some intuitive pull. As I have outlined above, we have good reason to be cautious about the long-term stability and evolutionary potential of animal traditions. There is, however, some evidence that the intuitive orthodoxy is mistaken. Specifically, there are examples of what appear to be relatively complex behaviors being transferred via “simple” forms of social learning. To illustrate, black rats in the pine forests of Jerusalem learn from their parents via local enhancement to extract the seeds from pine cones (Zohar and Terkel 1991; Aisner and Terkel 1985, 1992). Unlike in the case of the milk-bottle-opening behavior, which involves the transmission of a relatively simple and imprecise extractive foraging technique (the milk bottles are relatively easy to peck open and do not need to be pecked open in a very precise manner – simply pecking with determination will do), accessing the pine cone seeds requires that rat pups execute a precise sequence of behaviors. Specifically, the scales of the cones of Jerusalem pine are very tightly packed and can only be removed (and the seeds inside them thus obtained) by being stripped from the shaft sequentially from one particular end. To be able to successfully learn the adult technique, the naïve pups must learn a number of pieces of information: that pine cones are a food source; how to access the seeds from within pine cone scales; and that to get the scales off the cone, they must be stripped from the correct end. Naïve rat pups learn this by being exposed to partially stripped pine cones and scales by their parents as well as observing their behavior. In a two-stage process involving local enhancement similar to that already described in blue tits, the rat pups are drawn to the pine cones via social learning, and, once there, asocially innovate. In this manner, a relatively complex extractive foraging behavior is transmitted by a “simple” mechanism of social learning over multiple generations.
There is evidence of similar mechanisms being involved in other cases of much more complex animal traditions, such as the transmission of the manufacture and design of stick tools and their use in the New Caledonian crow (Hunt and Gray 2003; Kenward et al. 2006; Holzhaider et al. 2010a, 2010b). New Caledonian crows manufacture and use stick- and leaf-based tools, which they use to “fish” insect larvae out of holes. Both the manufacture of the tools and their use requires a reasonably precise set of actions (for example, for stick tools, the sticks must be broken in just the right places to fashion a “hook”, and the right “hooked” end of the tool must be inserted in the hole to successfully obtain food). As with the black rat pine-cone-stripping, the activity of adult crows scaffolds the learning of juvenile crows in important ways. The juvenile crows learn to use tools by interacting with the discarded tools of adult crows. They are interested in them and pick them up and carry them about. Coupled with local enhancement, this facilitates asocial learning in juvenile crows regarding both how to appropriately use tools, and the types of features of good tools. Once again, contra the intuitive orthodoxy, a relatively simple form of social learning is playing a central role in the maintenance of a behavior in a population over multiple generations. These types of examples undermine the view that cumulative behavioral evolution driven by social learning is a purely human trait (though I do not deny that the prodigious human capacity for cumulative cultural evolution suggests something importantly different is happening in the human case – human culture doesn’t seem to just be an aggregation of a lot of animal traditions).
A second common concern regarding the significance of animal traditions (closely related to the first) is that, even if cases like that of the black rat and New Caledonian crow suggest that quite simple forms of social learning could in principle facilitate the veridical, multigenerational transmission of relatively complex traits within populations, the persistence of such traits remains too sensitive to context to be of any significance on longer (evolutionary) timescales. In short, even if the game of telephone described above is overcome, a relatively small environmental change is all that is needed for these behavioral traits to be lost from populations forever. In the case of the blue tit milk-bottle-opening, for example, when the delivery of milk bottles ceased, unsurprisingly so too did the milk-bottle-opening behavior (McCarthy 2003). The behavior simply died with those birds that carried it, as its transmission to naïve individuals was reliant on the presence of milk bottles. Similarly, the pine-cone-opening of the black rats of the Jerusalem pine forests would likely be lost to the population were there no longer pine cones to open. In the evolutionary context, this issue is particularly worrisome, and is made stark when we compare the transmission of traits via social learning to that via genetic means.
Genetic traits are transmitted across generations regardless of whether they are expressed. The classic recessive gene cases we all learned at school, such as blue eyes, cystic fibrosis, and Huntington’s disease, are all examples of traits that can be transmitted from parent to offspring without expression in the parent. In contrast, no such “silent” transmission is possible for animal traditions. Being transmitted via social learning, they must be present in each generation, making them sensitive to a slew of environmental effects that genetic inheritance is not troubled by (Laland and Janik 2006). While this sensitivity to environmental change does pose an issue for the significance of animal traditions to those interested in cognitive and behavioral evolution, a similar thing can be said for any traits transmitted via social learning, human cultural traits included.
In response to the worries above, one might suggest that the challenge of fragility is only of issue if we expect large-scale and persistent environmental changes to be frequent, because social learning itself offers some robustness to the traits it fosters in populations. By facilitating the propagation of traditions and cultural traits within populations, social learning makes the persistence of those traits less sensitive to small-scale, chancy events, such as predation and localized or temporally restricted environmental change. One bad season where a particular foodstuff is scarce is not enough to lose the specific extractive foraging tradition relevant to that food, so long as some individuals live to see multiple seasons. Only a few individuals with the behavior need survive long enough to see the return of the previous environmental conditions to “reseed” the subsequent naïve generation and for the behavior to carry on.
While both the sensitivity to environmental change and the limited fidelity of transfer of the simple forms of social learning involved in animal traditions present challenges to the general view I am advocating here, they are not insurmountable. The human cultural case gives us some good pointers for the types of things that can mitigate the issues which we might expect to see in animal populations with traditions.
Humans intentionally and unintentionally modify their environments to make them more amenable to the transmission of cultural information from one generation to the next. For example, we provide children with toys, games, and situations that mimic those they will face in adult life (e.g. consider the ubiquity of dolls, play kitchens, doll houses, and dress-ups in the lives of our children). By provisioning our offspring with situations in which they can have cultural information presented and modeled to them, and where they can test what they have learned, we increase the likelihood that our accumulated cultural knowledge will be passed on to them (Sterelny 2012). This type of “epistemic engineering” or “downstream epistemic niche construction” also features in the lives of animals with traditions. As already mentioned, naïve black rats in the Jerusalem pine forests and the New Caledonian crows learn specific extractive foraging techniques from their parents by interacting with discarded bits of an adult “kit”. In leaving their half-stripped pine cones and tools about, they are epistemically engineering the learning environment of their offspring, increasing the likelihood that their particular extractive foraging traditions will be passed on. Other forms of epistemic engineering in animals are as simple as older individuals exhibiting an increased tolerance for the presence of juveniles when hunting, mating, food processing, and so forth, thereby enabling them to learn social information about those practices. Of course, this type of engineering is very simple and far removed from that seen in humans, but this is not unexpected given the great differences between human and nonhuman cultures.
The human case offers us a further pointer regarding how animal traditions might have a lasting impact on populations, despite the challenges of fragility and fidelity in genetic assimilation. Print reading is a human cultural trait for which there are no specialized neurocognitive mechanisms, having arisen only five to six million years ago. Rather, print-reading capacities are the product of careful tuition and effortful social learning (Heyes and Frith 2014). What we do see, however, are perceptual and attentional biases that dispose human children to develop print reading (without these, learning to read can be very difficult, such as in dyslexia (Paracchini et al. 2007)). While working out the evolutionary history of these biases is difficult (did they come before or after print reading, and in what form?), it is not unreasonable to think that they have been strengthened by selection for reading capacity. Specifically, following cultural innovations for reading, those individuals with perceptual and attentional biases that make them more capable of learning to read or capable of learning to read faster would plausibly have had fitness advantages over those without them (or with more rudimentary forms of them). This would have resulted for selection for such biases and a further strengthening of them in the human population over time. We expect that such processes of “genetic assimilation” are reasonably common in human populations (indeed, much of gene culture–co-evolution literature is predicated on such an assumption). Importantly for my purposes here, the process of genetic assimilation suggests that even if it turns out that animal traditions are only ephemeral (fleeting, lasting for only relatively brief evolutionary timescales and not particularly stable), they would still be of evolutionary importance.
In particular, just as human cultures can influence genetic evolution, so could animal traditions. When fitness-enhancing behavioral innovations arise in populations and are maintained via social learning, there is competition within individuals to gain the behavior, and gain it faster and more reliably than others. This plausibly results in selection for genetic adaptations that make the inheritance of the trait more reliable (Tebbich et al. 2016). We see this playing out in the case of the finches of the Galapagos.
Evolutionary biologists have long been impressed by the diversity of beak morphologies and foraging techniques seen in the numerous finch species found on the Galapagos Islands (often known as Darwin’s finches). This diversity is even more impressive given that it has its origins in a relatively small founding flock of finches blown from Central or South America (Sato et al. 2001; Vincek et al. 1997). It is also curious, as other similar bird species endemic to the Galapagos have not undergone the same level of diversification. Recent work (Tebbich et al. 2010) suggests that the differences in diversification are likely the consequence of an increased disposition to foraging innovation in the finches. Capacities for social and asocial learning allowed ancestral finch populations to take advantage of novel food resources that other birds did not utilize. This occurred over multiple generations, altering the selective regime that these ancestral populations were under, and ultimately “driving” finch populations to further adaptation to the novel foraging niches and diversification. The sharp-beaked ground finch serves to illustrate this well. This species of finch makes use of the abundance of sea birds on the Galapagos. Sometimes called “vampire finches,” they peck out the feathers of sea birds and drink blood from the wound they create. They also make use of the sea birds’ eggs by kicking or pushing the large, heavy eggs over ledges or onto rocks and breaking them open to access the nutritious yolk inside. Both of these foraging behaviors are enhanced by a series of morphological adaptations seen only in the sharp-beak ground finch. In particular, they have pointier and stronger beaks and stronger legs than other finches of the Galapagos (Schluter and Grant 1984). The unique morphological adaptations of the vampire finch appear to have arisen only after the unique foraging behavior (or some rudimentary form of it) had evolved. This is because, before the behavioral foraging innovation occurred and its spread in finch populations, there would have been no selective advantage to the morphological adaptations in question, and they are costly to maintain. Given this, it has been claimed (Tebbich et al. 2010) that the innovative foraging behavior arose first, most likely by accident or learning-based exploration, and was then reinforced by further success. The novel behavior then proliferated through the population, most likely through a combination of social and asocial learning (Tebbich et al. 2016). Only once this had occurred was there sufficient selective pressure in place for morphological adaptation to occur. In this way, a behavioral tradition can be said to have “driven” the evolution of the beak morphology of the sharp-beaked ground finch via genetic assimilation.
In this short chapter, I have offered a number of examples of the multigenerational transmission of traits in nonhuman animals via so-called “simple” forms of social learning, such as stimulus enhancement (also known as “animal traditions”). Although there are questions surrounding the evolutionary significance of animal traditions relating to their long-term robustness, I have a variety of reasons to think that they may be important in explaining aspects of both the behaviors and morphologies of many animal populations.
1 As the daily delivery of fresh milk has dwindled, so too has the milk-bottle-opening behavior, and in areas where milk bottle delivery has subsequently recommenced, the milk-bottle-opening trait has not resurfaced (McCarthy 2003). This is not unsurprising if social transmission (rather than genetic inheritance) was playing a central role in the maintenance of the behavior in tit populations.
2 There is also some recent evidence (Aplin et al. 2013) that cultural conformity plays a role in the transmission of behaviors in blue tit populations. Although this has not been shown for the milk-bottle-opening behavior, it is reasonable to believe that it was in play in that case also.
3 The distinction between so-called “simple” and “complex” forms of social learning in comparative psychology is a fraught one (see Meketa 2014, and Mikhalevich (nee. Meketa) Chapter 41 in this volume, for a useful discussion of the challenges to a single objective index for assessing the “simplicity” of cognitive structures, processes, or mechanisms). Here it is intended to delineate between explicit teaching or “true” imitation and other mechanisms; whether there is a “real” difference between them is, however, not particularly important to the claims being made.
4 This is not unsurprising given that blue tits are attracted to shiny objects. As foil milk-bottle lids are shiny, and were very commonly in the blue tit’s environment during this period, the eventual interaction between a blue tit and a milk-bottle lid was reasonably likely.
5 It is, of course, also reliant on asocial learning, both with respect to the original innovation, and in the transmission of the milk-bottle-opening behavior. Naïve individuals ultimately use asocial learning to acquire the milk-bottle-opening technique, but only after having been drawn to the bottles via social learning.
6 See Hoppitt and Laland (2013: 63–4) for definitions of the various forms of social learning. As noted by Ramsey and Goodrich (Chapters 32 and 33, respectively, in this volume), whether all these forms of social learning are importantly different, mechanistically, from asocial learning is unclear, but for my purposes here this distinction is not important.
7 The term “traditions” has been employed in various ways in the literature on the transmission of traits via social learning in nonhuman animals (see Laland and Janik (2006), Ramsey (Chapter 32 in this volume), and Goodrich (Chapter 33 in this volume)). Here, I am using a very broad definition of the term to refer to “distinctive behavior patterns, pieces of information or knowledge that are shared by two or more individuals in a social unit, which persist over time and that new practitioners acquire in part through socially aided learning” (Hoppitt and Laland 2013: 4).
8 As already mentioned above, cultural conformity plays a role in the transmission of behaviors in blue tit populations. This could offer a further reason for observing homogeneity in the socially transmitted behaviors in this instance.
There are two key anthologies and a monograph discussing animal traditions that further develop many of the ideas here: The Question of Animal Culture (edited by Kevin Laland and Bennett Galef), The Biology of Traditions (edited by Dorothy Fragaszy and Susan Perry), and Animal Traditions (by Eytan Avital and Eva Jablonka). Those interested in social learning in animals will find Social Learning by Kevin Laland and William Hoppitt a very useful resource. For a detailed discussion of genetic assimilation, social learning, and innovation in nonhuman animals, see Tebbich et al. (2016).
Alem, S., Perry, C. J., Zhu, X., Loukola, O. J., Ingraham, T., Sovik, E., and Chittka, L., 2016. “Learning and cultural transmission of string pulling in an insect”. PLoS Biology, 14(10), p. e1002564.
Aisner, R., and Terkel, J., 1985. “Habitat exploitation through cultural transmission: Pine cone feeding behaviour in black rats (Rattus rattus)”. Proceedings of the 19th Inter-National Ethological Conference. Toulouse: France.
Aisner, R., and Terkel, J., 1992. Ontogeny of pine cone opening behaviour in the black rat, Rattus rattus. Animal Behaviour, 44, pp. 327–336.
Aplin, L. M., Sheldon, B. C., and Morand-Ferron, J., 2013. Milk bottles revisited: Social learning and individual variation in the blue tit, Cyanistes caeruleus. Animal Behaviour, 85(6), pp. 1225–1232.
Avital, E., and Jablonka, E., 2000. Animal traditions: Behavioural traditions in evolution, Cambridge: Cambridge University Press.
Brown, C., and Laland, K. N., 2003. Social learning in fishes: A review. Fish and Fisheries, 4(3), pp. 280–288.
Fisher, J., and Hinde, R. A., 1949. The opening of milk bottles by birds. British Birds, 42, pp. 347–357.
Fragaszy, D. M., and Perry, S., 2003. The Biology of Traditions, New York: Cambridge University Press.
Galef, B. G., Jr., 2009. Culture in Animals? In K. N. Laland and B. G. Galef Jr., eds. The Question of Animal Culture. Cambridge, MA: Harvard University Press, pp. 222–246.
Galef, B. G., 2012. Social learning and traditions in animals: Evidence, definitions, and relationship to human culture. WIREs Cognitive Science, 3, pp. 581–592.
Griffiths, P. E., and Sterelny, K., 1999. Sex and Death: An Introduction to Philosophy of Biology. Chicago: The University of Chicago Press.
Heyes, C. M., and Frith, C. D., 2014. The cultural evolution of mind reading. Science, 344(6190), pp. 1243091–1243091.
Holzhaider, J. C., Hunt, G. R., and Gray, R. D., 2010a. Social learning in new Caledonian crows. Learning & Behavior, 38(3), pp. 206–219.
Holzhaider, J. C., Hunt, G. R., and Gray, R. D., 2010b. The development of pandanus tool manufacture in wild New Caledonian crows. Behaviour, 147(5/6), pp. 553–586.
Hoppitt, W., and Laland, K. N., 2013. Social Learning, Princeton, NJ: Princeton University Press.
Hunt, G. R., and Gray, R. D., 2003. Diversification and cumulative evolution in New Caledonian crow tool manufacture. Proceedings: Biological Sciences, 270(1517), pp. 867–874.
Kenward, B., Rutz, C., Weir, A. A. S., & Kacelnick, A., 2006. Development of tool use in New Caledonian crows: Inherited action patterns and social influences. Animal Behavior, 72, pp. 1329–43.
Krebs, J. R., and Kroodsma, D. E., 1980. Repertoires and geographical variation in bird song. Advances in the Study of Behavior, 11, pp. 143–177.
Laland, K. N., Brown, C., and Krause, J., 2003. Learning in fishes: From three-second memory to culture. Fish and Fisheries, 4(3), pp. 199–202.
Laland, K. N., and Galef, B. G., 2009. The Question of Animal Culture, Cambridge, MA: Harvard University Press.
Laland, K. N., and Hoppitt, W., 2003. Do animals have culture? Evolutionary Anthropology: Issues, News, and Reviews, 12(3), pp. 150–159.
Laland, K. N., and Janik, V., 2006. The animal cultures debate. Trends in Ecology & Evolution, 21(10), pp. 542–547.
Lefebvre, L., 1995. The opening of milk bottles by birds: Evidence for accelerating learning rates, but against the wave-of-advance model of cultural transmission. Behavioural Processes, 34(1), pp. 43–53.
McCarthy, M., 2003. Blue tits lose their bottle as milk thieves. The Independent. Available at: www.independent.co.uk/environment/blue-tits-lose-their-bottle-as-milk-thieves-84526.html.
Meketa, I., 2014. A critique of the principle of cognitive simplicity in comparative cognition. Biology & Philosophy, 29(5), pp. 731–745.
Paracchini, S., Scerri, T., and Monaco, A. P., 2007. The genetic lexicon of dyslexia. Annual Review of Genomics and Human Genetics, 8(1), pp. 57–79.
Sato, A., Tichy, H., O’Huigin, C., Grant, P., Grant, R. B., and Klein, J., 2001. On the origin of Darwin’s finches. Molecular Biology and Evolution, 18(3), pp. 299–311.
Schluter, D., and Grant, P. R., 1984. Ecological correlates of morphological evolution in a Darwin’s finch, Geospiza difficilis. Evolution, 38(4), p. 856.
Sherry, D. F., and Galef, B. G., Jr, 1984. Cultural transmission without imitation: Milk bottle opening by birds. Animal Behaviour, 32(3), pp. 937–938.
Sherry, D. F., and Galef, B. G., Jr, 1990. Social learning without imitation: More about milk bottle opening by birds. Animal Behaviour, 40(5), pp. 987–989.
Sterelny, K., 2012. The Evolved Apprentice, Cambridge, MA: The MIT Press.
Tebbich, S., Griffin, A. S., Peschl, M. F., and Sterelny, K., 2016. From mechanisms to function: An integrated framework of animal innovation. Philosophical Transactions of the Royal Society B: Biological Sciences, 371(1690), p. 20150195.
Tebbich, S., Sterelny, K., and Teschke, I., 2010. The tale of the finch: Adaptive radiation and behavioural flexibility. Philosophical Transactions of the Royal Society B: Biological Sciences, 365(1543), pp. 1099–1109.
Thorpe, W. H., 1963. Learning and Instinct in Animals, New York: Taylor & Francis.
Tomasello, M., 2009. The Cultural Origins of Human Cognition, Cambridge, MA: Harvard University Press.
Tomasello, M., 2014. A Natural History of Human Thinking, Cambridge, MA: Harvard University Press.
Vincek, V., O’Huigin, C., Satta, Y., Takahata, Y., Boag, P. T., Grant, P. R., Grant, B. R., and Klein, J., 1997. How large was the founding population of Darwin’s finches? Proceedings of the Royal Society B, 264(1378), pp. 111–118.
Zohar, O., and Terkel, J., 1991. Acquisition of pine cone stripping behaviour in black rats. International Journal of Comparative Psychology, 5(1), pp. 1–6.