Environmental reputation-based preference selection and strategy learning mechanisms promote cooperation

We introduce a novel framework that integrates environmental reputation-based preference selection with an improved strategy learning mechanism. Specifically, individuals select their neighbors based not only on payoffs but also on their reputational standing relative to the environmental reputation. They then update their strategies through a compatible learning rule that incorporates both factors, moderated by a reputation intensity parameter w. We carried out Prisoner’s Dilemma Game (PDG) simulations of the model on both a square lattice network and a Barabási–Albert scale-free network. The results demonstrate that the model effectively enhances both cooperation density and group reputation. Moreover, when the reputation intensity parameter w is sufficiently large, cooperation can be maintained at a high level even under strong temptations to defect, in sharp contrast to traditional models where cooperation rapidly collapses as defection temptation increases. Our findings further indicate that in square lattice networks, increasing the reputation intensity not only elevates the cooperation density but also improves the overall payoff of the population. In Barabási–Albert scale-free networks, nodes with degrees above a certain threshold consistently emerge as high-reputation cooperators. Overall, the model performs effectively on both network types, confirming its robustness.