Proceedings of the Annual Meeting of the Cognitive Science Society

Emergence of collective cooperation in an inherently selfishsociety is a paradox that has preoccupied biologists, sociol-ogists, and cognitive scientists alike for centuries. We pro-pose a computational model and demonstrate through simula-tions how collective cooperation can emerge from selfish inter-ests: the goal of improving each individual’s own rewards. Wealso demonstrate how the same selfish interests lead to the dy-namic emergence of a network of interconnected agents. Ourmodel includes two simple mechanisms: Selfish-Trust (ST)and Selfish-Connection (SC). ST involves the possibility of re-lying on others in a society of agents when it is beneficial tothe individual, and SC involves the possibility of connecting toother agents when those agents help improve the individual’sown benefit. Our simulation results suggest that collective co-operation can emerge from ST and a complex dynamic net-work can emerge from ST and SC. The simulated data demon-strate an important property of many living organisms: pat-terns of temporal complexity, which are essential to transferinformation among agents of any society of living beings.