Conflict and convention in dynamic networks

Michael Foley, Patrick Forber, Rory Smead and Christoph Riedl
Journal of the Royal Society Interface
15(139) 20170751
March 21, 2018

Abstract

An important way to  resolve games of conflict (snowdrift, hawk-dove, chicken) involves adopting a  convention: a correlated equilibrium that avoids any conflict between  aggressive strategies. Dynamic networks allow individuals to resolve conflict  via their network connections rather than changing their strategy. Exploring  how behavioral strategies coevolve with social networks reveals new dynamics  that can help explain the origins and robustness of conventions. Here we  model the emergence of conventions as correlated equilibria in dynamic  networks. Our results show that networks have the tendency to break the  symmetry between the two conventional solutions in a strongly biased way.  Rather than the correlated equilibrium associated with ownership norms (play  aggressive at home, not away), we usually see the opposite hostguest norm  (play aggressive away, not at home) evolve on dynamic networks, a phenomenon  common to human interaction. We also show that learning to avoid conflict can  produce realistic network structures in a way different than preferential  attachment models.