The Roles Of Anterior Insula And Dorsolateral Prefrontal Cortex In Fairness Norm Enforcement

Since 1990s,functional magnetic resonance imaging(fMRI)has been widely used in the fields of medicine,neuroscience and psychology to help researchers understand brain functions.In the thesis,VIRI was used to explore the roles that anterior insula(AI)and dorsolateral prefrontal cortex(DLPFC)played during fairness norm enforcement.Fairness norm is a key social norm.When people are treated unfairly,they would sacrifice their own interests to punish norm violators(second-party punishment).Even if it is others rather than themselves who are treated unfairly,they would still do so(third-party punishment).Inflicting punishment on norm violators are important ways for people to enforce fairness norm.The Ultimatum Game(UG)and the Third-party Dictator Game(TP-DG)are typical paradigms used to explore second-party and third-party fairness norm enforcement.In UG,there are two players.The proposer decides how to split some money and the recipient(second-party)responds with acceptance or rejection.Acceptance leads to the suggested split.Rejection makes both players receive nothing,which means punishing the proposer.In TP-DG,there are three players.The recipient has no other choice but to accept the proposer's offer,whereas the third-party could punish the proposer at the cost of his/her own loss.Behaviorally,abundant studies revealed that when unfair offers were proposed,both the recipient in UG and the third-party in TP-DG would undertake their own loss to punish unfair proposers,which were against the standard notions of money-maximizing preference.These results indicated that in order to maintain fairness norm,people would rather sacrifice their own interests.Neurally,AI and DLPFC were found to be engaged in fairness norm enforcement.In Study 1,the role of AI in fairness norm enforcement was explored.Previous research revealed that the engagement of AI in UG was related to detecting expectation biases caused by offers which violated recipients,expectations.However,when multiple expectations were violated,how AI would respond was a question which needed to be answered.In Study 1,the information about how others were treated was introduced in UG.Participants acted as the recipient and received fair and unfair offers from proposers whose offers for others were also presented.The results showed that AI was more active when participants received unfair relative to fair offers and when participants were offered more or less than others compared with being offered equally to others.Besides,the highest level of AI activaton was observed when participants received offers which were both unfair and unequal to others'.These results indicated that AI was sensitive to not only expectation biases associated with receiving unfair offers but also expectation biases associated with being not treated equivalently as others,and that AI activation was related to the quantity of expectation biases.Taken together,these results suggested that AI played a general role in detecting expectation biases.The role of DLPFC in fairness norm enforcement was explored in Study 2 and Study 3.Some researchers suggested that the engagement of DLPFC in rejection of unfair offers was associated with overriding selfish impulse,whereas recent studies on third-party punishment in legal decision making assumed that DLPFC was involved in integrating criminal information with context-specific punishment rules and selecting appropriate punishment.In Study 2 and Study 3,we further explored whether DLPFC was associated with overriding selfish impulse or integrating information and selecting appropriate responses by manipulating participants' power to punish norm violators in second-party and third-party fairness norm enforcement.In Study 2,two different power contexts were created according to whether recipents could punish proposers by rejection.UG served as the context where participants had power to punish proposers.The Impunity Game(IG)served as the context where participants had no power to punish proposers since recipients' rejection in IG had no influence on proposers but reduced their own income to zero.Base on the "overriding selfish impulse" assumption,greater DLPFC activity was expected during rejection relative to acceptance of unfair offers in IG since rejection would only reduce participants' own income to zero.However,based on the "integration and selection" assumption,similar strong DLPFC activities were expected during both acceptance and rejection of unfair offers in IG due to the tradeoff process during response selection,since participants might be neither willing to make rejections which only punished themselves nor accept offers they perceived unfair.In Study 3,the influence of third-party's loss on proposers' income was manipulated to creat a high power context where the loss ratio between the third-party and the proposer was 1:6 and a low power context where the loss ratio was 1:3.Based on the "overriding selfish impulse"assumption,if participants undertook more loss to punish extreme-unfair proposers than moderate-unfair proposers,greater DLPFC activity was expected during punishing extreme-unfair proposers in both contexts.However,based on the "integration and selection"assumption,greater DLPFC activity was expected during punishing moderate-unfair proposers in the high power context.This was because that in the high power context,partcipants might inflict excessive punishment on moderate-unfair proposers due to their great influence on proposers' income.Thus,the seletion of appropriate punishment on moderate-unfair proposers might involve more tradeoff process,which might lead to stronger DLPFC acitivity during punishing moderate-unfair proposers.The results of Study 2 showed that DLPFC indeed exhibited similar strong activities during acceptance and rejection of unfair offers in IG.The results of Study 3 showed that in the high power context,participants undertook more loss to punish extreme-unfair proposers whereas DLPFC was more active during punishing moderate-unfair proposers.These results were in accordance with our predictions based on the "integration and selection" assumption,providing evidence for the role that DLPFC played in integrating information and selecting appropriate responses.Overall,in the thesis,three fMRI studies were conducted to explore the roles of AI and DLPFC in fairness norm enforcement.The studies revealed that AI and DLPFC played an important role in fairness norm enforcement.Furthernore,the results provided empirical evidence for the general role that AI played in detecting expectation biases and confirmed the engagement of DLPFC in integrating information and selecting appropriate responses during fairness norm enforcement.