We study repeated interactions among a fixed set of "low rationality" playe
rs who have status quo actions, randomly sample other actions, and change t
heir status clue if the sampled action yields a higher payoff. This behavio
r generates a random process, the better-reply, dynamics. Long run behavior
leads to Nash equilibrium in games with the weak-finite improvement proper
ty, including finite, super-modular games and generic, continuous, two-play
er, quasi-concave games. If players make mistakes and if several players ca
n sample at the same time; the resulting better-reply dynamics with simulta
neous sampling converges to the Pareto optimal Nash equilibrium in common i
nterest games. (C) 2001 Academic Press.