Dynamics of Partner Choice and Cooperation

Abstract

Ever since Robert Axelrod initiated his computer tournament aimed at finding the winning strategy in a repeated prisoner’s dilemma game have simulations and tournaments been the preferred methodology of social scientists investigating the phenomena of reciprocity and cooperation. Simulations are attractive because they can often be applied when theoretical deduction is not possible. Even simple models of social interaction often produce complex dynamics that are hard to characterize with mathematical analysis. This is especially likely when the system under study has many elements that interact with each other in many different ways. The dynamics of such a system usually depends on the emerging interaction pattern as well as on the private experiences of each agent. Simulations may yield paths of system dynamics given by any set of behavioral rules and thus provide an insight into effects of any particular modeling assumption. Approaching the same task mathematically is often much more difficult.