The evolution of cooperation with memory, learning and dynamic preferential selection in spatial prisoner's dilemma game

Abstract

The prisoner's dilemma is widely used to research the evolution of cooperation. Here we suppose players use non-discriminative strategies within their neighborhoods and they have different memories to remember the previous interaction histories. In each round, a player copies his most successful neighbor's strategy or most successful strategy in his neighborhood within memory history as his strategy in the coming round, and the player uses preferential selection rule to select a neighbor to learn from. By the use of MC simulations, research results have been given as following: the preferential selection rule greatly promotes the cooperation level in heterogeneity networks while it inhibits the emergence of cooperation in homogeneous regular network; and different learning rules and memory sizes strongly affect the evolution of cooperation especially in homogeneous case.