How Cultural Transmission Through Objects Impacts Inferences About Cultural Evolution

The cross-fertilisation between biological and cultural evolution has led to an extensive borrowing of key concepts, theories, and statistical methods for studying temporal variation in the frequency of cultural variants. Archaeologists have been among the front-runners of those engaging with this endeavour, and the last 2 decades have seen a number of case studies where modes of social learning were inferred from the changing frequencies of artefacts. Here, we employ a simulation model to review and examine under-discussed assumptions shared by many of these applications on the nature of what constitutes the 'population' under study. We specifically ask (1) whether cultural transmission via 'objects' (i.e. public manifestations of cultural traits) generates distinct patterns from those expected from direct transmission between individuals and (2) whether basing inference on the frequency of objects rather than on the frequency of mental representations underlying the production of those objects may lead to biased interpretations. Our results show that the rate at which ideational cultural traits are embedded in objects, and shared as such, has a measurable impact on how we infer cultural transmission processes when analysing frequency-based archaeological data. At the same time, when cultural transmission is entirely mediated by the material representation of ideas, we argue that copying error should be interpreted as a two-step process which may occur in either one or both of embedding information in objects and retrieving it from them.