Evolving payoff currencies through the construction of causal theories

Payoff-biased cultural learning has been extensively discussed in the literature on cultural evolution, but where do payoff currencies come from in the first place? Are they products of genetic or cultural evolution? Here we present a simulation model to explore the possibility of novel payoff currencies emerging through a process of theory construction, where agents come up with “channels” via which different cultural traits contribute to some ultimate payoff and use such “channels” as intermediate payoff currencies to make trait-updating decisions. We show that theory-building as a strategy is mostly favored when the noise associated with the ultimate-level payoff is high, selective pressures are strong, and the probability of arriving at the right theory is high. This approach provides insights into both the emergence of payoff currencies and the role of cognition for causal model building. We close by discussing the implications of our model for the broader question of causal learning in social contexts.