Abstract:

Third-party punishment (TPP) is individuals punish the norm violator as unaffected third parties even at a personal cost. Many studies have provided insight into the neural mechanisms underlying TPP behavior from evidence at the electrophysiological and functional imaging levels. However, this evidence has been restricted to a single component or has focused only on the results of activation in independent brain regions. Moreover, there is still a lack of holistic understanding regarding the connections between the cognitive processes underlying TPP behavior and the functional brain networks. Therefore, this paper reviews the research related to TPP in the past decade. First, we summarize theories that can explain TPP behavior in order to deepen the understanding of the theoretical dimensions of TPP behavior. These theoretical models include the reciprocity model, which reflects individual preferences for cooperation and fairness, the emotion model of intuitive processing, and the dual-systems model, which integrates emotional and cognitive factors under a reinforcement learning perspective.

Second, we conduct a review of functional neuroimaging and electrophysiological evidence that pertains to TPP, with a particular emphasis on the inter- and intra-network connectivity within the brain. Taking into account the functions and activation patterns of the relevant brain networks in previous studies, we suggest that the generation of TPP behavior is divided into three phases: emotion generation, responsibility assessment and punishment selection. The corresponding brain networks are salient network, default mode network, and central executive network. In addition, the reward network collaborates in TPP processing, mainly playing the role of value representation and expected reward.

Finally, in order to explain the occurrence mechanism of TPP behavior from a more comprehensive perspective, we integrate the results of previous studies and propose a cognitive neural network model of TPP. In this model, the affective system and the reward system jointly function as the motivation system for TPP, playing a role in generating motivation for TPP behavior. The corresponding brain networks associated with these systems are the salience network and the reward network, respectively. The cognitive system consists of two subsystems: the social cognitive system and the executive control system. These subsystems play a role in two phases of TPP: responsibility assessment and punishment selection. The default mode network and the central executive network are the respective brain networks associated with these two phases. The components of the model cooperate and interact with each other, and ultimately the executive control system makes the decision of whether to punish and the intensity of punishment. The feedback information generated in turn influences the internal loop, enabling the individual to learn and refine their behavioral strategy based on each feedback. Over time, this process leads to the development of a stable behavioral pattern. The model establishes the connection between TPP behavior-related research at the psychological and cognitive-neural level. Moreover, it provides a more holistic and comprehensive explanation of the mechanisms of TPP behavior and suggests that TPP is a dynamic process with feedback and reinforcement involvement.

In the future, researchers can further explore TPP behavior from the following perspectives: (a) Starting from a more microscopic perspective such as neurotransmitters and hormones to reveal the neurophysiological mechanisms of TPP behavior; (b) Incorporating individual differences to explore the relationship between neurophysiological representations of TPP behavior and personality trait variables; (c) Introducing machine learning algorisms to further optimize and develop the relevant models to provide quantitative explanations and predictions of TPP behavior; (d) Utilizing meta-analysis to provide quantitative data support for the models to increase their reliability; (e) Exploring third-party intervention preferences and the underlying cognitive neural mechanisms in different contextual information or more complex social contexts.

Key words: third-party punishment, cognitive neural mechanisms, brain network, fMRI