Quantitatively trained Ph.D, students participated in three replicatio
ns of a 20-trial market entry game in which each agent has two pure st
rategies-to ''enter the market'' or ''stay out of the market''-and the
individual payoff depends on the size of the set of active agents ent
ering the market. With experience subjects adhered more strictly to pu
re strategies, which although markedly different from one to the other
combined to form distributions supporting a Nash equilibrium solution
to the game.