Bounded Rationality and Social Cognition: A Computational Study

This paper examines the implications of a 'strong' version of bounded rationality popular within computational and evolutionary game theory in which agents are represented by finite automata. It is argued that this view has unrealistic features in that agents of this kind are unable to distinguish between opponent or game types. An alternative view of bounded rationality from social cognition accommodating opponent-specific choices is presented and analyzed by computer simulation.