This essay discusses some recent work on 'learning in games'. We explo
re nonequilibrium theories in which equilibrium emerges as the long-ru
n outcome of a dynamic process of adjustment or learning. We focus on
individual level models, and more specifically on variants of 'fictiti
ous play' in two-player games. We discuss both the theoretical propert
ies of the models and their relationship to regularities observed in g
ame theory experiments. (C) 1998 Elsevier Science B.V. All rights rese
rved.