Initial impressions are important for building reputation. While an initial impression does not constitute a complete reputation, it is critical for determining how someone is judged, and how subsequent information about that person is interpreted. Reputation is especially important when modeling repeated games and those models typically assume that actors have an initial set of expectations about their counterpart's strategies. The play of the game, whether repeated or sequential, allows an actor to update those priors and then rely on the counterpart's reputation. In repeated games, the ability to both infer and develop a reputation can be critical for reaching a Pareto-superior equilibrium. However, the source of initial reputation is a matter that is largely unexplored. For single-shot games and the initial play of repeated games, reputational priors are critical. For games with multiple equilibria, the equilibrium that is reached is often characterized as path-dependent, and the initial expectations formed by partners prior to playing the game are important for defining which path is taken.
We focus on one process by which reputations are generated. The settings we investigate are simple, one-shot, two-person sequential games. Each game has a unique subgame perfect equilibrium. A subset of the games contain an outcome that Pareto-dominates the equilibrium, but can only be reached by a sequence of moves that depends on actors trusting one another and engaging in reciprocal altruism. Both trust and reciprocity are grounded in the priors that actors hold about their counterparts in the game; evolutionary psychology provides a basis for inferring the source of an actor's prior.
The paper first sketches findings concerning consistency out-of-equilibrium behavior. In the second section reviews some of the extensive literature on facial expressions, largely drawn from social psychology. The third section presents survey and experimental results that test the impact of facial expressions on expectations and behavior. The final section concludes with a discussion of avenues of future research and the importance of these findings for game theory.