Cooperation in a network of heterogeneous agents

LUIS G. MOYANO; ANGEL SANCHEZ

Salamanca

Conferencia; XV Congreso de Física Estadística - FisEs08; 2009

Universidad de Salamanca

The question of how cooperation emerged in societies still puzzles researchers, as this behaviour does not seem to fit a priori an evolutionary standpoint. Indeed, cooperation has been thought to be contrary to the idea that individual fitness is the relevant observable to the survival over generations. There is an ongoing effort in several scientific communities to understand this emergent aspect of interacting complex agents, from evolutionary biologists and social scientists to statistical physicists. Cooperation and evolution are often tackled using evolutionary game theory tools. In this context, the Prisoner’s Dilemma has been a prototype to study cooperation: two players may cooperate or not and are rewarded and punished consequently. In the evolutionary Prisoner´s Dilemma (PD) game, agents play with each other and update their strategies in every generation according to some microscopic dynamical rule. In this work we explore mixed PD systems composed of two types of agents, each kind with a representative update rule. Our model shows a rich variety of behaviours, including that cooperation may be greatly promoted. It also shows that certain type of rules normally become extinct while other rules usually prevail.