Percolation and cooperation with mobile agents: geometric and strategy clusters

Area of Science:

• Evolutionary Game Theory
• Agent-Based Modeling
• Complex Systems

Background:

• The Prisoner's Dilemma game models strategic interactions with potential for conflict and cooperation.
• Understanding persistent cooperation is crucial in various biological and social systems.
• Previous models often assumed non-spatial or lattice-based interactions.

Purpose of the Study:

• To investigate conditions for sustained cooperation in an off-lattice mobile agent model.
• To analyze the impact of agent size and movement on cooperation dynamics.
• To explore the role of network structure in maintaining cooperative or defecting populations.

Main Methods:

• Utilized an off-lattice agent-based model simulating the Prisoner's Dilemma.
• Incorporated agent exclusion radius (population viscosity) and interaction radius (contact network).
• Analyzed phase transitions and geometric percolation of interaction networks.

Main Results:

• Finite-sized agents introduce a coexistence phase of cooperators and defectors, alongside absorbing phases.
• Network percolation, irrespective of strategy, enhances cooperation, similar to lattice models.
• Defector survival is contingent on their network percolating; isolated defector groups face extinction.

Conclusions:

• Agent size and mobility significantly alter cooperation dynamics compared to non-spatial models.
• Network percolation is a critical factor for the persistence of defectors.
• Cooperator clusters can persist without percolation, while defector clusters require it to survive.