Coevolution of Behavior and Networks

Social structure is both a consequence and a determinant of human behavior. In order to shed light on the problem of the emergence and maintenance of social order, one of the central underlying quests in social science, we need to understand how behavior and structure coevolve.

Under the supervision of Prof. Vincent Buskens at Utrecht University, I investigated how this coevolution process can be modeled with cooperation and coordination games on dynamic networks. I reviewed models from recent analytical, simulation-based, and experimental studies by reconstructing them within a general formal framework. The analysis revealed that in theoretical studies, the relative speed of network update (how often actors reconsider their links compared to their action choice) is one of the factors with the biggest impact on macro-outcomes such as efficiency, hierarchical organization and inequality. However, this effect is conditional on one assumption that is common to all existing models, namely, that players employ the same action against all of their partners. I argued that future research should relate models to applications and experimental tests more adequately, which often implies allowing for discriminatory action.

The unpublished review paper can be found here.