Community connectivity and heterogeneity: clues and insights on cooperation on social networks

Abstract

While studies on the emergence of cooperation on structured populations abound, only few of them have considered real social networks as the substrate on which individuals interact. As has been shown recently [Lozano et al., PLoS ONE 3(4):e1892, 2008], understanding cooperative behavior on social networks requires knowledge not only of their global (macroscopic) characteristic, but also a deep insight on their community (mesoscopic) structure. In this paper, we look at this problem from the viewpoint of the resilience of cooperation, in particular when there are directed exogenous attacks (insertion of pure defectors) at key locations in the network. We present results of agent-based simulations showing strong evidence that the resilience of social networks is crucially dependent on their community structure, ranging from no resilience to robust cooperative behavior. Our results have important implications for the understanding of how organizations work and can be used as a guide for organization design.