Strategy-updating depending on local environment enhances cooperation in prisoner's dilemma game

An evolutionary prisoner's dilemma game is studied on a square lattice when local environment is considered. And a new strategy updating rule is utilized in this game. First, individuals are classified as two classes: stochastic players and intelligent players. The stochastic players are the ones who adopt the rule of replicator dynamics to update strategies, and the intelligent players are those who make a decision through comprehensive considerations of the local environment (namely, the information of different-strategy neighbors' environment). Second, the intelligent player's evaluation is calculated from payoffs and local environment by Dempster-Shafer theory. The results indicate that the new strategy-updating method promotes cooperation when intelligent individuals utilize the information of different-strategy neighborhoods. Moreover, the increasing number of intelligent player promotes cooperation level. Furthermore, it is noteworthy that cooperators will finally coexist with defectors no matter what the proportion of intelligence is, which is worthy to think about.