An experiment on behavior in social learning games with collective preferences

Abstract

Many environments with social learning involve payoff interdependencies. Until recently, few studies have examined sequentially observational social learning environments with externalities. This paper presents the results of an experiment that examines the theoretical predictions of a simple model of observational learning where individuals care about their own actions and the actions of others (Ali and Kartik, 2012). Consistent with theory, I find that behavior does not change with this class of payoff interdependencies. However, behavior deviates from equilibrium predictions in the same manner both with and without these interdependencies.