Emergent Spatial Characteristics from Strategic Games Simulated on Random and Real Networks

TL;DR

This paper investigates how initial network structures, including real and random networks, influence the evolution of strategies and emergent spatial patterns in iterated prisoners' dilemma games, highlighting the roles of network size and small-world properties.

Contribution

It introduces a simulation framework that examines the impact of network topology on strategy dynamics in the prisoners' dilemma, emphasizing the effects of network size and small-world features.

Findings

Network size affects convergence rate.

Small-world properties influence dominant node status.

Initial network structure impacts strategy evolution.

Abstract

Complex networks are a great tool for simulating the outcomes of different strategies used within the iterated prisoners' dilemma game. However, because the strategies themselves rely on the connection between nodes, then initial network structure should have an impact on the progression of the game. By defining each interaction in terms of a prisoner's dilemma and using its payoff matrix as a basis for investigation, we implemented players with various interaction and edge attachment strategies, and ran this dynamic process on real and random networks with varying network structure. We found that, both network size and small world properties played an important role in not only deciding the convergence rate of the simulation but also the dominant status of nodes, under the conditions where identical strategies are employed by every player.