We exploit a unique opportunity to study how a large population of players in the field learn to play a novel game with a complicated and non-intuitive mixed strategy equilibrium.  We argue that standard models of belief-based learning and reinforcement learning are unable to explain the data,...

This paper studies the extent to which diffusion approximations provide a reliable guide to equilibrium selection results in finite games. It is shown that they do for a class of finite games with weak learning provided that limits are taken in a certain order. The paper also shows that making...

We study decentralized learning dynamics for the classic assignment game with transferable utility.  At random points in time firms and workers match, break up, and re-match in the sesarch for better opportunities.  We propose a simple learning process in which players have no knowledge about...

Social norms are patterns of behavior that are self-enforcing at the group level: everyone wants to conform when they expect everyone else to conform.  There are multiple mechanisms that sustain social norms, including a desire to coordinate, fear of being sanctioned, signaling membership in...

We study evolutionary dynamics in assignment games where many agents interact anonymously at virtually no cost.  The process is decentralized, very little information is available and trade takes place at many different prices simultaneously.  We propose a completely uncoupled learning process...

We examine some of the factors that might influence the quality of information produced in discussion groups on the internet, such as USENET and the WELL. In particular, we look at the impact of various different pricing structures, and compare regimes in which anonymity is enforced with regimes...

The diffusion of an innovation can be represented by a process in which agents choose perturbed best responses to what their neighbors are currently doing.  Diffusion is said to be fast if the expected waiting time until the innovation spreads widely is bounded above independently of the size...

A long-standing open question raised in the seminal paper of Kalai and Lehrer (1993) is whether or not the play of a repeated game, in the rational learning model introduced there, must eventually resemble play of exact equilibria, and not just play of approximate equilibria as demonstrated...

This paper studies the extent to which diffusion approximations provide a reliable guide to equilibrium selection results in finite games. It is shown that they do for a class of finite games with weak learning provided that limits are taken in a certain order. The paper also shows that making...

This paper demonstrates that inertia driven by switching costs leads to more rapid evolution in a class of games that includes m x m pure coordination games. Under the best-response dynamic and a fixed rate of mutation, the expected waiting time to reach long-run equilibrium is of lower order in...