The Emergence of Cooperation in the well-mixed Prisoner's Dilemma: Memory Couples Individual and Group Strategies (2402.03890v1)

Abstract: Exploration of mechanisms underlying the emergence of collective cooperation remains a focal point in field of evolution of cooperation. Prevailing studies often neglect historical information, relying on the latest rewards as the primary criterion for individual decision-making-a method incongruent with human cognition and decision-making modes. This limitation impedes a comprehensive understanding of the spontaneous emergence of cooperation. Integrating memory factors into evolutionary game models to formulate decision criteria with delayed effects has shown potential in unraveling cooperation mechanisms. However, this comes at the significant cost of heightened computational complexity. In this paper, we propose an experiential decision-making method based on reinforcement learning. Utilizing this method, we construct a multi-agent system to engage in the evolutionary Prisoner's Dilemma game. Simulation results indicate that memory establishes a coupling relationship between individual and group strategies, fostering periodic oscillation between cooperation and defection in a well-mixed group. Specifically, defection loses its payoff advantage over cooperation as the group cooperation rate decreases. Conversely, the cooperative behavior gains reinforcement with an increase in the group cooperation rate, overcoming defection as the dominant strategy for individuals. This coupling between individual and group strategies fundamentally bridges the gap between individual and group interests, integrating a multitude of known factors and elucidating the fundamental mechanism of cooperation emergence in the face of social dilemmas.