Evidence for Social Learning in Wild Lemurs (Lemur Catta)
Kendal, Rachel L., Custance, Deborah M., Kendal, Jeremy R., Vale, Gillian, Stoinski, Tara S., Rakotomalala, Nirina Lalaina, Rasamimanana, Hantanirina, Learning & Behavior
Interest in social learning has been fueled by claims of culture in wild animals. These remain controversial because alternative explanations to social learning, such as asocial learning or ecological differences, remain difficult to refute. Compared with laboratory-based research, the study of social learning in natural contexts is in its infancy. Here, for the first time, we apply two new statistical methods, option-bias analysis and network-based diffusion analysis, to data from the wild, complemented by standard inferential statistics. Contrary to common thought regarding the cognitive abilities of prosimian primates, our evidence is consistent with social learning within subgroups in the ring-tailed lemur (Lemur catta), supporting the theory of directed social learning (Coussi-Korbel & Fragaszy, 1995). We also caution that, as the toolbox for capturing social learning in natural contexts grows, care is required in ensuring that the methods employed are appropriate-in particular, regarding social dynamics among study subjects. Supplemental materials for this article may be downloaded from http://lb.psychonomic-journals.org/content/supplemental.
Social learning, or learning from others, is currently of widespread interest because it potentially provides a means by which animals can acquire adaptive information about their environment rapidly and efficiently. Social learning is thought to underlie the rapid diffusion of novel behavioral variants, interpopulation variation in behavior, and cultural traditions in animals from fishes to apes (Lefebvre & Palameta, 1988; Rendell & Whitehead, 2001; Warner, 1988; Whiten, 2009). Interest in animal social learning has also been fueled by reports of intra- and interpopulation variation in the behavioral repertoires of animal populations, spawning claims of culture in apes (McGrew, 1998; van Schaik et al., 2003; Whiten et al., 1999), cetaceans (Krützen et al., 2005; Rendell & Whitehead, 2001), and monkeys (Leca, Gunst, & Huffman, 2007; Perry et al., 2003). However, claims that these data demonstrate animal cultures remain controversial because alternative explanations to social learning, such as genetic proclivities or ecological differences, remain difficult to refute (see Laland, Kendal, & Kendal, 2009), despite innovative work in captivity (e.g., Whiten et al., 2007). Moreover, since learning is frequently functional, adaptive, based on genetic proclivities, and responsive to ecological resources, the current ethnographic method, which proclaims culture where the alternatives can be dismissed, is vulnerable to excluding genuine cases of social learning. Thus, in contrast to the controlled laboratory study of social learning, and despite pioneering work with apes and cetaceans (see Sargeant & Mann, 2009; Whitehead, 2009; Whiten et al., 1999), there is a dearth of tools for capturing compelling evidence of social learning in natural contexts, either in the wild or in captivity.
Recently, however, several statistical methods have been created to aid in the task of identifying social learning in naturalistic contexts (e.g., Boogert, Reader, Hoppitt, & Laland, 2008; Franz & Nunn, 2009; Hoppitt, Boogert, & Laland, 2010; J. R. Kendal, Kendal, & Laland, 2007; R. L. Kendal, Kendal, Hoppitt, & Laland, 2009; Matthews, 2009). We presented a solution to the problem in the form of a method known as option-bias analysis (R. L. Kendal, Kendal, et al., 2009). The method is based on the well-established premise of social learning research-that is, when ecological and genetic differences are accounted for, social learning can generate greater homogeneity in behavior between animals than would be expected in its absence (but see Thornton & Malapert, 2009). For example, when probing for termites in their mound, chimpanzees are reported to use either a short- or long-twig method (Whiten et al., 1999), and when manufacturing wide Pandanus leaf tools, New Caledonian crows (Corvus moneduloides) have three variants available to them (Holzhaider, Hunt, & Gray, 2010). …