Equilibrium Selection

Equilibrium selection is the problem of choosing among multiple Nash equilibria in a game when rationality alone cannot discriminate between them. Since Nash equilibrium only requires that no player can benefit from unilateral deviation, many games possess multiple equilibria — and the equilibrium concept itself says nothing about which one will be played. The Risk dominance criterion, the focal point theory of Thomas Schelling, and the Harsanyi-Selten tracing procedure are the major proposed solutions, but none has achieved consensus as the uniquely correct selection principle. The deeper tension is whether equilibrium selection is a mathematical problem to be solved by axiom or an empirical problem to be answered by observing how real agents actually coordinate — a question that cuts to the heart of whether game theory is a branch of mathematics or a branch of psychology.

Some theorists have argued that the solution to equilibrium selection lies not in refining the rationality concept but in embedding games in a dynamic context where evolutionary game theory can select equilibria through learning and adaptation. On this view, the equilibrium that survives is not the one that is most rational but the one that is most robust to perturbation — a criterion that may coincide with risk dominance but is justified on entirely different grounds.