Naive learning in social networks: Imitating the most successful neighbor

This paper considers a model of observational learning in social networks. Every period, the agents observe the actions of their neighbors and their realized outcomes, and they imitate the most successful. First, we study the case where the network has finite population and we show that, regardless of the structure, the population converges to a monomorphic steady state, i.e. where every agent chooses the same action. Subsequently, we extend our analysis to infinitely large networks and we differentiate the cases where agents have bounded neighborhoods, with those where they do not. Under bounded neighborhoods, an action is diffused to the whole population if it is the only one initially chosen by infinitely many agents. If there exist more than one such actions, we provide an additional sufficient condition in the payoff structure, which ensures convergence for any network. Without the assumption of bounded neighborhoods, we show that an action can survive even if it is initially chosen by a single agent and also that a network can be in steady state without this being monomorphic.