Cooperator-driven and defector-driven punishments: How do they influence cooperation?

TL;DR

This paper investigates how cooperator-driven and defector-driven punishments influence cooperation in different population structures, revealing that mixed punishments are more effective in well-mixed populations, while cooperator-driven punishments excel in networked populations.

Contribution

It introduces a game model analyzing six evolutionary scenarios with combined punishments and develops a semi-analytical method to estimate phase boundaries, highlighting the differential effects of punishment types in population structures.

Findings

Mixed punishments outperform single punishments in well-mixed populations.

Cooperator-driven punishment is more effective in networked populations.

Network structure enhances cooperation through reciprocity and mutualism.

Abstract

Economic studies have shown that there are two types of regulation schemes which can be considered as a vital part of today's global economy: self-regulation enforced by self-regulation organizations to govern industry practices, and government regulation which is considered as another scheme to sustain corporate adherence. An outstanding problem of particular interest is to understand quantitatively the role of these regulation schemes in evolutionary dynamics. Typically, punishment usually occurs for enforcement of regulations. Taking into account both types of punishments to curve the regulations, we develop a game model where six evolutionary situations with corresponding combinations of strategies are considered. Furthermore, a semi-analytical method is developed to allow us to give an accurate estimations of the boundaries between the phases of full defection and nondefection. We find that, associated with the evolutionary dynamics, for infinite well-mixed population, the mix of both punishments performs better than one punishment alone in promoting public cooperation, but for networked population the cooperator-driven punishment turns out to be a better choice. We also uncover monotonous facilitating effects of synergy effect, punishment fine and feedback sensitivity on the public cooperation for infinite well-mixed population. Conversely, for networked population an optimal intermediate range of feedback sensitivity is needed to best promote punishers' populations. Overall, networked structure is overall more favorable for punishers and further for public cooperation, because of both network reciprocity and mutualism between punishers and cooperators who do not punish defectors.