Evolutionary dynamics and individual heterogeneity in multi-agent networking systems

Cooperative behaviors lie at the basis of human societies, yet its evolutionary origin still remains a perplexing problem. There exists a large evolutionary game literature exploring effects of population structures on evolutionary outcomes, but in most cases structured population models only employ one game alone. A relevant question is covered in this study thus concerns the effect of game diversity occurring along the edges linking two players situated on the nodes. The basic structure of our mechanism provides more than one type of games available for the players. In our game setting, the game model is assumed as a fixed property of each edge and will remain unaltered through game process. As preliminary initiatives, we only consider two typical games (Prisoner's dilemma game and Snowdrift game), and all possible mixtures between them controlled by a parameter. The performed simulations and theoretical analysis help us observe the remarkable improvement on cooperation levels in suitable parameter regions when the population are simultaneously provided with more than one type of games. Still, network heterogeneity helps to bring into existence of cooperator clusters and thus favors the survival of cooperators.