The Influence Of Second-order Reputation Mechanism On Individual Cooperation And Game Behavior In Spatial Networks

Cooperation is an important factor to promote the progress and development of human society.However,because individuals are usually selfish,how to get each component of the group to reach a consensus and participate in the cooperative process has been a difficult problem in academia.Evolutionary game theory provides a feasible and effective research plan for this problem.Researchers have built different game models to simulate the interaction between individuals in reality,and introduced different mechanisms to explore some methods to promote cooperation.Researches have shown that both network reciprocity and indirect reciprocity are important mechanism to promote cooperation.The combination of evolutionary game theory and complex network theory has provided a new way to study the derivation of cooperation.In recent years,researches on the introduction of reputation mechanism into the model of evolutionary game are mostly carried out on the basis of binary reputation,and on the basis of first-order reputation update rules,that is,only the strategy of the individual in this round is considered when reputation is updated.These models are intuitive and easy to study,but they are too one-sided.For these reasons,the second-order reputation mechanism,which takes the image of the interactive object into consideration,has attracted more and more attention.Based on such second-order reputation mechanism,the research and innovative achievements of this paper include the following two aspects:1)The adaptive reputation mechanism is introduced into the model of public goods game on the spatial lattice network.The reputation of an individual is not only good or bad,but also changes in the range of 1 to 100.There is a reputation threshold in the game environment,which is used to determine the quality of an individual’s image.An individual with a good image has a higher influence and can easily transfer his own strategies to the surrounding individuals.We proposed a second-order reputation update rule according to individual’s own strategy and his interactive object’s image,and individual’s reputation will change differently in different situations.At the same time,we also introduced a new parameter,the reputation update step,which makes the change of reputation value is no longer single,and could be changed with different ranges.Through a lot of simulations,the results show that our second-order reputation update model can significantly improve the frequency of cooperation in the group compared with the traditional game model.2)On the basis of the previous work,the game model is replaced by prisoner’s dilemma game,and all four classical second-order reputation update rules are studied: Stern Judging,Simple Standing,Shunning and Image Scoring.We analyzed the similarities and differences between the four rules in detail.Through a large number of simulation experiments,we can draw the conclusion that all four update rules can improve the group cooperation rate under the framework of adaptive reputation mechanism,and performed better than the traditional prisoner’s dilemma game model.By adjusting the values of different parameters,we can observe the different changes of cooperation rate under the four rules,and analyze the causes of these changes in detail.Finally,it is concluded that the second-order reputation updating rules can promote the cooperation in network groups.