Optimal population turnover regimes for cultural evolution depend on network size, density and behavioral transmissibility

A change to a populations social network is a change to the substrate of cultural transmission, affecting behavioral diversity and adaptive cultural evolution. While features of network structure such as population size and density have been well studied, less is understood about the influence of social processes such as population turnover-- or the repeated replacement of individuals. Experimental data has led to the hypothesis that naive learners can drive cultural evolution by being better samplers, although this hypothesis has only been expressed verbally. We conduct a formal exploration of this hypothesis using a generative model that concurrently simulates its two key ingredients: social transmission and reinforcement learning. We explore how variation in turnover influences changes in the distributions of cultural behaviors over long and short time-scales. We simulate competition between a high and low reward behavior, while varying turnover magnitude and tempo. We find optimal turnover regimes that amplify the production of higher reward behaviors. We also find that these optimal regimes result in a new population composition, where fewer agents which know both behaviors, and more agents know only the high reward behavior. These two effects depend on network size, density, behavioral transmissibility, and characteristics of the learners. Our model provides formal theoretical support for, and predictions about, the hypothesis that naive learners can shape cultural change through their enhanced sampling ability, identified by previous experimental studies. By moving from experimental data to theory, we illuminate an under-discussed generative process arising from an interaction between social dynamics and learning that can lead to changes in cultural behavior.