Effects of Centrality and Heterogeneity on Evolutionary Games

Abstract

In the evolutionary games based on the heterogeneous populations, recent research has shown that the degree of players in the network plays an important role and often determine the level of cooperation. Yet, the individual influence described by centralities remains inadequate in quantifying the effect of promoting cooperation. In this work we have comprehensively investigated how the representative centrality metrics impact the fate of cooperation on different levels of heterogeneous populations. Simulation results show that on the whole, centrality characteristic is efficient to facilitate cooperation in social dilemmas except the Clustering, and Degree is neither the sole nor the best one. Meanwhile, there is an optimal level of heterogeneity that maximizes the cooperators regardless of the influence of centralities.