Backward Induction, Normal Form Perfection and Explicable Equilibria

A weakening of D. Kreps and R. Wilson's (1982) notion of sequential rationality is presented. The motivation stems from the difficulty in justifying sequentially rational behavior in subgames reachable only through a violation of sequential rationality. Although the present notion of weak sequential rationality is based upon extensive form considerations, it bears a close relation to R. Selten's (1975) normal form perfect equilibria. Backward induction outcomes can be achieved in generic games of perfect information with additional restrictions on beliefs. An example with imperfect information shows that sequential rationality is not the consequence of equilibrium play and the absence of incredible threats. Copyright 1992 by The Econometric Society.