Culture In Nonhuman Organisms Research Paper

1. Social Learning In Animals

Underlying culture in nonhuman animals is a capacity for ‘social learning.’ The term social learning refers to learning that is influenced by observation of, or interaction with, another animal (typically of the same species) or its products. Social learning is frequently contrasted with individual or asocial learning, in which animals learn exclusively on the basis of their own personal experience, without recourse to social cues or guidance. In any instantiation of social learning, the learner or receiver of information is generally de-scribed as the ‘observer,’ while the transmitter of information is called the ‘demonstrator.’ Social learning occurs when the communication of information between two animals results in the observer learning from the demonstrator. Frequently the observer learns through social interaction to perform the same behavior as that exhibited by the demonstrator and, if repeated amongst other animals, this process can result in the spread of a particular behavior pattern and behavioral conformity in a population. Less frequently, animals may learn by observation to perform a different behavior from that exhibited by the demonstrator.

Social learning among animals has been of interest to scientists from a number of different disciplines. Psychologists traditionally have been interested in animal learning. Contemporary animal learning theory describes much learning in animals as resulting from associations formed between two external stimuli (stimulus-stimulus, or classical conditioning), or be-tween the behavior of an animal and an external stimulus (responsereinforcer, or operant conditioning). Arguments abound as to whether there will be anything different or special about social, compared with asocial learning, and whether social learning can be adequately explained in conditioning terms. Many psychologists have suggested those more complex forms of social learning might be indicative of the animal concerned possessing unusual or sophisticated psychological capabilities (Heyes 1994). For instance, it has been suggested that for an animal to imitate the motor patterns of another animal it must be conscious, must be capable of taking the other animal’s perspective, must be unusually intelligent, or must be capable of intentional action. All such claims are hotly disputed.

At the same time, social learning has been of interest to ethologists and behavioral ecologists because it seems to allow animals to learn about their environments rapidly and efficiently, without having to engage in potentially costly or hazardous learning trials, or expend considerable time and energy exploring the environment. Animals can learn which foods to eat, acquire food processing skills, learn to identify predators, learn which members of the opposite sex to mate with, or develop songs and calls, by exploiting the knowledge base of more experienced conspecifics (Heyes and Galef 1996, Zentall and Galef 1988). Social learning can be regarded as a short cut to learning about the environment, but may incur a cost if the acquired knowledge is inappropriate or outdated.

Evolutionary biologists, biological anthropologists, and archeologists have also studied animal social learning because it is accorded a prominent role in the evolution of human culture. Although definitions of human culture vary, one central feature is the trans-mission of acquired information between individuals through social learning processes. Many researchers believe that an understanding of the ways in which social learning operates in animal populations will generate insights into the evolutionary roots of human culture. Social learning is regarded as pivotal to a number of social intelligence hypotheses, which maintain that the evolution of high intelligence or large brains was driven by a capacity for social learning and for proto-culture (Byrne and Whiten 1988).

2. History Of Research Into Animal Social Learning

The study of social learning in animals dates back to Darwin, and flourished as a result of attempts to demonstrate mental continuity between humans and animals in order to prove that humans had evolved. For many Victorian scientists, demonstrations of imitative learning in animals was seen as providing evidence of an evolutionary origin of the higher mental faculties of humans. Imitation was regarded as intermediate in psychological complexity between animal instinct and human reasoning (Boakes 1984). The verb ‘to ape’ and the phrase ‘monkey see monkey do’ are manifestations of the then commonly held view that simian primates are frequent imitators of each other’s behavior.

While numerous observations of imitative behavior in primates and other animals were reported in the late nineteenth century, attempts to demonstrate imitation proved problematic. It became apparent that there are many processes that can result in two animals ex-pressing similar behavior. Researchers began to realize that anecdotal stories concerning animals in uncontrolled environments would not provide satisfactory evidence to address this issue. In the first decades of the twentieth century, Edward Thorndike, a leading psychologist, attempted to introduce some scientific rigor into investigations by carrying out controlled experiments exploring imitation in cats, dogs and monkeys (Galef 1988). Thorndike’s experiments led him to the controversial conclusion that there was no evidence that animals could ‘from an act witnessed learn to do an act.’ Out of the controversy there emerged two distinct research traditions in the study of social learning: a laboratory experimental tradition, dominated by experimental psychology, and a second tradition that carried out observations of natural populations of animals, dominated by ethologists and primatologists. The laboratory approach has focused on investigating the psychological mechanisms under-lying social learning. In contrast, field studies have reported behavior patterns spreading through animal populations, or cultural traditions. Recent years have seen the development of experimental approaches to the study of the spread of learned information through animal populations, including transmission chain and diffusion studies (Laland et al. 1993). In addition, mathematical approaches have explored the conditions under which social learning is expected to evolve, and the evolutionary consequences of social learning. (See Galef 1988 for more on the history of research into animal social learning.)

3. Laboratory Studies Of Animal Social Learning

Modern laboratory experiments investigating social learning in animals commonly train animals (called demonstrators) to perform a target behavior (e.g., to push a lever to get a food reward), pair these demonstrators with untrained experimental subjects (called observers) that watch the demonstrator per-form the behavior, and then test the observers alone to establish whether they have learned to perform the target behavior (Galef 1988). Frequently, the rate at which the subjects learn to perform the behavior is compared with control subjects, which have received different forms of observational experience. Such experiments have established that there are a number of ways that animals can learn socially, the variants reflecting the nature of the observer’s learning and the type of information transmitted from demonstrator to observer. For instance, in the lever-pressing example, the observers might learn at an accelerated rate relative to controls, (a) because they learned of the presence or significance of the lever, (b) because they learned of the presence of food, (c) because their observational experience allowed them to form an association between the depressed lever and the appearance of food, or (d) because they formed an association between the action of pressing and the appearance of food. In addition, there are other nonsocial-learning processes that could result in an elevation in the observers’ performance, for example, if observation changes its motivational state.

Such experiments have engendered a fissioning of imitative phenomena, resulting in a multitude of terms to describe the different processes that can result in social learning and spawning a host of classificatory schemes. At present there is little consensus as to the use of terms, and little agreement as to which classification is superior. Whiten and Ham (1992) distinguish four classes of social learning. ‘Stimulus enhancement’ occurs when the observer learns from the demonstrator to which object or location its behavior should be oriented. For instance, British birds learned to peck open milk bottles and drink the cream by having their attention drawn to milk bottles by other feeding birds (Hinde and Fisher 1951). (The term ‘local enhancement’ is sometimes used to describe similar phenomena). ‘Observational conditioning’ occurs when the observer learns to what circumstances a behavior should be a response. For example, infant rhesus monkeys learned to respond to the presence of a snake with fear and flight behavior through observations of the screams and avoidance behavior of adults. ‘Imitation’ occurs when the observer learns from the demonstrator to perform a specific motor pattern in a particular context. For instance, orangutans have been reported to go through the motions of washing clothes in the presence of human caretakers that were doing the laundry. Finally, ‘goal emulation’ refers to instances where the observer learns from the demonstrator the goal to pursue or the result that can be achieved. For example, juvenile chimpanzees that observed a stick being used to rake in out-of-reach food appeared to learn that the food could be gained using the stick but developed their own means of doing so. Earlier Galef (1988) identified seven processes that could result in social learning. Other researchers (e.g., Heyes 1994) have argued that social learning can be best described by using the same conceptual tools as are employed by psychologists to describe asocial learning. In general, there remain considerable doubts as to whether any of the current classification schemes are comprehensive, or whether each of a given classification’s categories is mutually exclusive. The plethora of competing terms and usage reflect an ignorance of the biological basis of social learning. At present there is little understanding of the genetic, neural or physiological processes that underlie the processing of sensory information, the associationist or transformational organization of this information, and the production of motor patterns, during social learning. Nonetheless, there is little doubt that the laboratory experimental approach has greatly enhanced under-standing of the processes that can result in social learning, and developed the methodologies and conceptual tools to further that understanding.

4. Field Transmission Chain And Diffusion Studies

In a population of Japanese monkeys, a young female began washing sweet potatoes provisioned on a beach before eating them. This previously unobserved behavior was subsequently recorded to increase in frequency in the troop, with 90 percent of individuals eventually adopting the potato-washing behavior, presumably by learning from each other. There are hundreds, perhaps thousands, of such reports of the spread of novel behavior patterns in populations of animals. Well-known examples include the transmission of dietary preferences by rats attending to cues on each other’s breath, the transmission of predator recognition in birds, mammals, and fish, the learning of vocal dialects in song birds, feeding on pine cones among black rats, and tool-using traditions in primates (see Bonner (1980) for a readable overview). Probably all vertebrates and many invertebrates will eventually be found to exhibit some form of social learning. In the main, these reports are observations of acquired behavior patterns spreading through single populations. In other cases, distinct cultural variations may characterize the behavioral repertoires of different populations of the same species. For instance, much birdsong is thought to be learned from older males, and frequently regional song dialects characterize neighboring populations. Undoubtedly the most impressive case of local cultural traditions is provided by the different tool-using repertoires of populations of common chimpanzees. For instance, termiting (using a stalk to fish for termites) is reported in some but not all chimpanzee populations and is thought to be maintained within populations through social learning processes. A recent study documented 39 different behavior patterns, including tool usage, grooming, and courtship behaviors, that were habitual in some chimpanzee communities and absent in others, where ecological explanations had been discounted (Whiten et al. 1999).

These field reports are of value in documenting the breadth of animal cultures and throwing light on the contexts in which social learning occurs. However, it is difficult in the field to establish reliably whether or not a behavior genuinely is socially learned (as opposed to learned independently by each animal), to determine whether local traditions reflect different cultural repertoires or varying ecological resources, or to investigate the psychological and social processes underlying the diffusion.

More recently there have been attempts to investigate animal cultural traditions experimentally, under controlled conditions. ‘Diffusion studies’ are characterized by the introduction of a trained demonstrators into captive populations, and a careful monitoring of the increase in the target behavior over time (Lefebvre and Palameta 1988). ‘Transmission chain studies’ involve the gradual replacement of a founder population of trained demonstrators with untrained animals, and explore whether the target behavior remains in the population when the founders have been removed (Laland et al. 1993). These methods have been valuable in determining how processes of social interaction impede, or propitiate, the diffusion of information. It has become apparent that the structure of animal populations (dominance relations, spatial positioning, sex, age, and size differences) strongly influences the pathways by which learned information spreads, a phenomenon known as ‘directed social learning’ (Coussi-Korbel and Fragaszy 1995). For instance, novel skills often spread at different rates among juveniles and adults, among the two sexes, or within and between family groups.

Typically the new behavior pattern found in an animal culture will have been invented by a single individual. There has been comparatively little investigation of the process of innovation. It is not clear whether innovation should be characterized as a personality trait (associated with clever, creative, or nonconformist individuals), a state-dependent variable (e.g., hunger may drive foraging innovation), or resulting from environmental changes (such as a drought). Preliminary studies suggest that the adage ‘necessity is the mother of invention’ may be appropriate, with innovation appearing at elevated frequencies in hungry, low status, and small individuals. Across primate species, incidence of innovation co-varies with incidence of social learning (also with tool use and deception), suggesting that innovation and social learning may be manifestations of a general capacity for behavioral plasticity. However, within a given species it is not known whether innovative individuals tend to be good, average, or poor social learners.

5. The Adaptive Value Of Animal Social Learning

Theoretical analyses have explored the circumstances under which natural selection will favor reliance on social learning, as opposed to asocial learning or evolved nonlearned behavior (see Laland et al. 1996, for an overview of this literature). A consensus has been reached that the issue hangs, in part, on patterns of variability in the environment. In an environment that is changing comparatively slowly, or that exhibits relatively little spatial heterogeneity in resources, populations are able to evolve appropriate behavior patterns through natural selection, and learning is of little adaptive value. In contrast, in rapidly changing or highly variable environments asocial learning pays. Here natural selection cannot track environmental fluctuations quickly enough, while social learning is unreliable because it may lead individuals to acquire outdated or locally inappropriate behaviors. It is intermediate rates of change and patterns of spatial heterogeneity that favor social learning, and within this window vertical transmission of information (social learning of offspring from parents) is thought to be an adaptation to slower rates of change than horizontal transmission (social learning among unrelated individuals of the same cohort). Although these findings are theoretically robust, a number of questions remain. For instance, do contemporary populations of animals capable of social learning switch between social and asocial learning depending on the pattern of environmental variability they experience in pertinent resources? Which other factors affect reliance on social learning? For instance, is social learning more likely when individuals are confronted with an unfamiliar problem, or a threatening environment?

6. Social Learning And Social Intelligence

Social learning has been central to a number of social intelligence hypotheses, which suggest that complex social behavior drove brain evolution and intelligence. Allan Wilson’s (1985) concept of ‘behavioral drive’ is the hypothesis that most explicitly emphasises social learning. Wilson noted a positive relationship between relative brain size and rate of evolution in vertebrates. He explained this by arguing that big-brained species exhibit greater behavioral plasticity, are more innovative, and are more capable of social learning than small-brained species. Cultural traditions, he argued, allow populations of animals to open up new niches, and exploit novel resources, thereby exposing them-selves to new selection pressures, and accelerating the rate of fixation of mutations. Wilson’s hypothesis is supported by recent evidence that there is a significant correlation between relative brain size and incidence of both innovation and social learning in primates.

Bibliography:

1. Boakes R 1984 From Darwin to Behaviourism. Cambridge University Press, Cambridge, UK
2. Bonner J T 1980 The Evolution of Culture in Animals. Princeton Science Library, Princeton, NJ
3. Box H, Gibson K 1999 Mammalian Social Learning. Comparative and Ecological Perspectives. Cambridge University Press, Cambridge, UK
4. Byrne R W, Whiten A 1988 Machiavellian Intelligence: Social Expertise and the Evolution of Intellect in Monkeys, Apes and Humans. Oxford University Press, Oxford, UK
5. Coussi-Korbel S, Fragaszy D M 1995 On the relation between social dynamics and social learning. Animal Behaviour 50: 1441–53
6. Galef B G Jr 1988 Imitation in animals: History, definition, and interpretation of data from the psychological laboratory. In: Zentall T R, Galef B G Jr (eds.) Social Learning. Psychological and Biological Perspectives. Erlbaum, Hillsdale, NJ
7. Heyes C M 1994 Social learning in animals: Categories and mechanisms. Biological Reviews 69: 207–31
8. Heyes C M, Galef B G Jr 1996 Social Learning in Animals. The Roots of Culture. Academic Press, San Diego, CA
9. Hinde R A, Fisher J 1951 Further observations on the opening of milk bottles by birds. British Birds 44: 393–6
10. Laland K N, Richerson P J, Boyd R 1993 Animal social learning: towards a new theoretical approach. Perspectives in Ethology 10
11. Laland K N, Richerson P J, Boyd R 1996 Developing a theory of animal social learning. In: Heyes C M, Galef B G Jr (eds.) Social Learning in Animals. The Roots of Culture. Academic Press, San Diego, CA
12. Lefebvre L, Palameta B 1988 Mechanisms, ecology, and population diffusion of socially-learned, food-finding behavior in feral pigeons. In: Zentall T R, Galef B G Jr (eds.) Social Learning. Psychological and Biological Perspectives. Earlbaum, Hillsdale, NJ
13. Whiten A, Goodall J, McGrew W C, Nishida T, Reynolds V, Sugiyama Y, Tutin C E G, Wrangham R W, Boesch C 1999 Cultures in chimpanzees. Nature 399: 682–5
14. Whiten A, Ham R 1992 On the nature and evolution of imitation in the animal kingdom: Reappraisal of a century of research. Advances in the Study of Behavior 21: 239–83
15. Wilson A C 1985 Scientific American 253: 148–57
16. Zentall T R, Galef B G Jr 1988 Social Learning. Psychological and Biological Perspectives. Earlbaum, Hillsdale, NJ