Parallel simulations of the iterated n-player prisoner’s dilemma

Abstract

We present in this work some extended results to those originally published in [8], when we simulated the Iterated n-Player Prisoner’s Dilemma in a sequential computer. In [12], we presented a solution to parallel agent-based simulations where agents need to interact with all its neighbours in a von Neumann neighbourhood, aiming to improve the usage of computational resources. By using such parallel techniques, we could better study the effect of several additional parameters of the Iterated n-Player Prisoner’s Dilemma simulation, like the grid dimension and the error rate.