Mean-field interactions in evolutionary spatial games

We introduce a mean-field term to an evolutionary spatial game model. Namely, we consider the of Nowak and May, based on Prisoner's dilemma, augment rules by self-consistent term. This way, agent operates local information from its neighbors nonlocal via coupling. simulate model construct steady-state phase diagram, which shows significant new features due term: while for steady states are characterized constant mean density cooperators, contains with continuous dependence payoff parameter. Moreover, changes nature transitions discontinuous jumps in first derivative. The main effects observed stationary states, parametrically close chaotic states: coupling drives such into chaos. Our approach can be readily generalized broad class games deterministic stochastic decision rules.