Counting conformity: evaluating the units of information in frequency-dependent social learning

In a recent study (Aplin et al., 2015), we conducted a large-scale cultural diffusion experiment in which we used trained demonstrator individuals to introduce one of two alternative foraging techniques into five replicate subpopulations of wild great tits, Parus major. Three further subpopulations served as controls. By tracking the spread of these two techniques, we showed that information was acquired through social learning, transmitted through social network ties, and novel behaviours became established in each subpopulation, forming stable arbitrary traditions (for technique A or B). These traditions persisted over generations and were stable despite immigrating and innovating individuals, resulting in a within-group behavioural homogeneity and between-group variation. Most pertinent for this discussion, our experimental design allowed an examination of the interaction between individual decision making and population level outcomes. We found that the population level bias for each introduced technique increased by an average of 14% per day towards the common variant. This was explained both by a tendency for naïve individuals to disproportionately adopt the most common behaviour (conformist transmission) and by a tendency for individuals with experience of both techniques to change their behaviour to match the common variant (conformity).