Imitation dynamics in a game of traffic

Abstract

We study a model of traffic where drivers adopt different behavioral strategies. These can becooperative or defective according to a driver abiding or not by a traffic rule. Drivers can change their strategy by imitating the majority, with a rule that depends on the strategies with which they have interacted. These interactions occur at intersections, where vehicles pay a temporal cost according to their strategy. We analyze the conditions under which different strategy compositions represent an advantage in the system velocity. We found that the cooperators? mean speed is higher than the defectors? even when the vehicle density is large. However, defectors can obtain benefits in their mean speed when they are a minority in an essentially cooperative population. The presence of a core of educated drivers, who persist firmly in a cooperative behavior, optimizes the speed in the system, especially for intermediate values of vehicular density and higher temporal costs.