How do you know that? Sensitivity to statistical dependency in social learning

Social learning has been shown to be an evolutionarily adaptive strategy, but can be implemented via many different cognitive mechanisms. Sensitivity to statistical dependency in the behavior of other people is a factor that discriminates between possible mechanisms: simple rule based strategies may be unaffected by dependency, while more sophisticated social learning strategies should take it into account. We use a Bayesian model to determine how rational agents should incorporate the effects of statistical dependency when learning from other people, conducting two experiments that examine whether human learners behave similarly. We find that people are sensitive to two different patterns of dependency, supporting the use of a sophisticated strategy for social learning.