Replicator dynamics on heterogeneous networks.

Area of Science:

• Evolutionary Game Theory
• Network Science
• Social Behavior Modeling

Background:

• Networked evolutionary game theory studies social behavior in structured populations.
• Existing research often limits to 2-strategy games on heterogeneous networks or n-strategy games on regular networks.
• This study addresses n-strategy games on arbitrary networks.

Purpose of the Study:

• To analyze n-strategy games on arbitrary networks using the pairwise comparison rule.
• To investigate how network structure, specifically degree distribution, influences evolutionary dynamics.
• To identify conditions favoring specific strategies, like cooperation, in evolutionary games.

Main Methods:

• Approximation of stochastic evolutionary processes using replicator equations under weak selection.
• Transformation of payoff matrices incorporating average degree and degree distribution variance.
• Analysis of evolutionarily stable strategies in minimum-effort and Prisoner's Dilemma games.
• Validation through agent-based simulations on diverse network types (quasi-regular, exponential, scale-free).

Main Results:

• Weak selection allows approximation of short-term evolutionary dynamics by replicator equations.
• Network heterogeneity, characterized by degree distribution variance, can promote payoff-dominant strategies.
• Cooperative equilibria are favored in n-strategy games when networks have low average degree and high degree variance.
• Simulations confirm the accuracy of the replicator equation approximation for various network structures.

Conclusions:

• The study provides a refined model for evolutionary game dynamics on complex networks.
• Network topology, particularly degree heterogeneity, plays a crucial role in shaping social behavior evolution.
• The findings offer insights into the stability of cooperation in structured populations, relevant for understanding social dilemmas.