Evolutionary game analysis of individual behaviors with reputation-driven dynamics

Reputation plays an essential role in shaping human interactions and sustaining cooperative norms in complex societies. To examine its influence on strategic decision-making, we develop an evolutionary Prisoner’s Dilemma (PD) model that incorporates reputation as a bounded, strategy-dependent, and socially conditioned dynamic variable. In this framework, reputation not only contributes to individual fitness but also functions as an attention-allocation mechanism, increasing the likelihood that agents imitate high-reputation neighbors. We analyze the model through large-scale simulations on both ER random networks and BA scale-free networks. The results demonstrate that reputation substantially promotes cooperation and stabilizes social order across different topologies. In particular, balanced reward–punishment mechanisms sustain higher levels of cooperation and reputation, while the weight assigned to reputation in fitness evaluation and the substitutability parameter govern the trade-off between payoff-driven and reputation-driven incentives. Moreover, increasing the bias toward high-reputation individuals consistently accelerates cooperative diffusion and reputation accumulation, with scale-free networks exhibiting greater robustness than random networks. By jointly examining the co-evolution of cooperation density and average reputation, our findings highlight reputation as a social amplifier that links local behaviors with collective outcomes. This study offers new insights into the mechanisms by which reputation fosters cooperation and provides theoretical guidance for designing incentive schemes and governance policies in complex social systems.