The evolution of cooperation in partner and stranger networks”
Associate Professor at the Burgundy School of Business
Many interactions qualify as social dilemmas, where cooperation between individuals is needed to achieve outcomes that are optimal for a group or society at large. One key element of the social interactions that has been shown to influence cooperative decisions is the composition of the groups. Specifically, there exists mounting evidence from laboratory and field studies showing that individuals cooperate more when the composition of the group is stable over time, that is when individuals interact repeatedly with the same partners. Whereas the existing evidence relies on studies of how individuals behave within closed groups, real-life interactions are more complex. In fact, most of our interactions take place within social networks where one’s behavior can be contagious at the whole population level. We extend the standard partner-stranger comparison to networks and show that the standard result that cooperation is higher in partner interactions greatly depends on the number of direct connections between network members. Partner matching yields higher cooperation rates only when individuals have direct connections to a small number of others. When the number of direct connections increases, partners do not cooperate more than strangers. There are methodological and institutional implications from our work. In particular, we show that it is important to consider both the sorting procedure (partners vs strangers) and a network’s connectivity when designing institutions to increase social cooperation.
Dates & time