An experiment on learning in a multiple games environment

We study how players learn to make decisions if they face many different games. Games are drawn randomly from a set of either two or six games in each of 100 rounds. If either there are few games or if extensive summary information is provided (or both) convergence to the unique Nash equilibrium generally occurs. Otherwise this is not the case. We demonstrate that there are learning spillovers across games but participants learn to play strategically equivalent games in the same way. Our design and analysis allow us to distinguish between different sources of complexity and theoretical models of categorization.