Weighted-elite-memory mechanism enhances cooperation in social dilemmas

Abstract

The issue of how to enhance cooperation has been a hot topic of research in evolutionary games for a long time. A mechanism is proposed to facilitate the cooperation behavior of evolutionary groups on networks in three game models, including prisoner's dilemma, snowdrift game, and stag hunt game. The core of the mechanism lies in: 1) Each player has a length of memory and uses the information of the elite in the memory span to update its strategy. 2) Each player has the chance to game with a certain neighbor more than once in each round. 3) The accumulative payoff of a player consists of two parts, one from playing with elites in memory length and another from playing with current neighbors, and a weight is introduced to adjust these two parts. The findings of the simulation demonstrate that a small weight can significantly enhance cooperation in three typical social dilemmas. Furthermore, the level of cooperation increases at first and then stays stable as the memory length increases.