Environmental and life history factors, but not age, influence social learning about food: a meta-analysis

Social learning is widespread across the animal kingdom and used by individuals to acquire adaptive behaviour across a range of contexts, from mate choice to migration. However, while multiple studies have demonstrated the presence of social learning in different species across several taxa, less attention has been paid to explaining variation in this phenomenon between populations or individuals. Theoretical modelling has suggested age should be important, with social learning most adaptive during the juvenile phase. Yet, empirical studies testing the interaction between age and social learning are equivocal and scattered in the literature. Accordingly, we present here what we believe to be the first meta-analytical review of this body of work. Collating 57 experimental studies that tested social learning in a foraging context, we conducted a two-fold analysis. First, we examined whether learning success or rates were correlated with five possible variables: phylogeny, dispersal system, age at first reproduction (a conservative measure of extended juvenile periods), type of task or testing condition (captive individuals, wild individuals or wild individuals brought into captivity for testing). Second, we tested for interactions between age and these variables (28 studies). Our results did not support a clear direct relationship between age and social learning abilities. However, life history factors indirectly related to age (age at first reproduction and dispersal system) affected social learning propensity. Finally, social learning success was also correlated with factors related to the experiment itself (captive or wild origin of the animals and the type of task used to assess social learning).