The challenges of a social learning phylogeny: mechanisms, content and context biases and inter-individual differences

Behavioral phylogeny, via comparative studies, has been successfully used to hypothesize when a behavior appeared and to identify the selective pressures at its origin. The application of behavioral phylogeny to social learning is very promising, but proponents of the approach must also address several obstacles before measurably advancing the field of study. The aim of this review is to highlight and to propose solutions for these challenges. First, the term “social learning” includes a diversity of non-exclusive mechanisms from simple enhancement to complex imitation which can be based on different cognitive process. So, a phylogenetic approach of social learning cannot be a unitary one and should be expressed in terms of biases and choices among the varied range of possible mechanisms. However, to ensure comparison of homologous mechanisms in the species of interest, it is necessary to first properly identify mechanisms as well as to identify their underlying structure, their limitations and their adaptive flexibility. Second, any experiment aiming to identify the social learning mechanisms employed by a species should take into account possible “contextual variables” and, consequently, any resulting content and context biases. With respect to content biases, social learning may selectively involve stimuli characteristics when, for example, foraging and distinguishing non-toxic foods, manufacturing and using tools, choosing songs and other behaviors to court a mate, discovering migration paths and recognizing predators. Context biases are dependent on the context in which stimuli occur and are thus reflected by diverse social learning strategies, such as “when and who to copy”. Third, it is important to emphasize that content and context biases can strongly depend on the age, sex, social rank, personality traits and previous experiences (including parental, epigenetic, ecological and cultural effects) of the tested individuals. These factors influence the mechanisms concerned. For example, while a very social individual may learn through imitation because of closer proximity to the demonstrator, a more timid or a lower-ranking individual can become as socially aware of and motivated to experience the same contingencies through respective local or stimulus enhancements. Individuals might even specialize in idiosyncratic learning mechanisms depending on the possible costs of switching strategies. Therefore, in considering social learning mechanisms that are nurtured and shaped by experiences during development, only a holistic approach taking into account content and context biases, as well as physical and psychological characteristics of the individuals tested, will correctly assess social learning capacities and the evolution of those capacities within species.