Exploring the Effect of Proximity and Kinship on Mutual Cooperation in the Iterated Prisoner’s Dilemma

The iterated prisoner’s dilemma (IPD) has been used as a model for investigating cooperation in nature. Here, we present an analysis of the evolution of reciprocal cooperation in a dynamically simulated environment in which individual agents are free to move in space, interacting with their nearest neighbors in fixed-length IPD games. Agents aim to avoid those against whom they score poorly, and to seek out those against whom they score highly. Individuals are modeled using finite state machines, allowing us to extend previous work on kin group markers. Though they have no direct effect on an individual’s strategy, such markers do lead to the emergence of coherent, mutually-cooperating sub-populations.