Cooperation in Heterogeneous Populations

ÄBSTRACT This paper deals with stochastic reactive strategies for the Iterated Prisoner's Dilemma. It considers populations of individuals meeting randomly, and noisy interactions. Both the analysis of monomorphic and heteromorphic populations show that the reciprocal strategy Tit For Tat acts like a pivot: it triggers an evolution towards cooperation, but is not the ultimate beneficiary of such an evolution. Ever since the early days of game theory, the Prisoner's Dilemma (PD) has been viewed as a major paradigm for discussing cooperation between selfish in-dividuals in the absence of constraints. From 1979 onwards, the topic received a further strong impetus by Axelrod's computer tournaments (see Axelrod, 1984)