本研究考察内源性羞耻影响对内外群体陌生人的亲社会行为。159 名大学生随机分配到中性组和内源性羞耻组，在两类亲社会情境里完成助人意愿的自我评定任务。结果显示，当外群体成员求助时，内源性羞耻组被试具有更强的助人意愿，且内源性羞耻可以消除内群体偏好效应。研究支持内源性羞耻的修复性动机影响亲社会行为。

In recent years, many researchers have gradually paid more attention to shame as a self-conscious emotion. Shame is a kind of emotions that is closely related to positive behavior. Shame has an important influence on social development of individuals. In the existing literature, there are two different perspectives about the effect of shame on individual behavior. One is the negative effect of shame, and shamed individuals are often accompanied by feelings of significance, helplessness, and incompetence, hiding, escaping, etc. The other is the positive role of shame, supported by the theory of Commitment Device, and motive of shame. Shame could promote individuals’ moral behavior. The positive role of shame may vary due to shame type, such as endogenous shame or exogenous shame. Endogenous shame has a special effect on behavior, and involving the role of stranger help-seekers, such as ingroup or outgroup. In order to clarify the role of shame in the prosocial behavior of individuals, this study examines the effect of endogenous shame and group identity on prosocial behavior, and reveals the restore motive of shame.

Fifty-five college students participated in the preliminary experiment, and completed the evaluation materials. Emotion induction materials were adapted from typical neutral and shame events. After 2 minutes of light music for relaxation, participants imagined themselves as the protagonist in the situations, and then rated the shame intensity on 7-point scale. Two neutral emotion materials included building introduction and manual of produce, and shame materials included failure of exam and littering in public.

In the formal experiment, 159 college students were randomly assigned to neutral group and endogenous shame group. The validity was tested. Deleting the participants whose reaction was more than 3 standard deviations, the data of 138 participants were analyzed. The entire experiment consisted of two consecutive parts: an emotion inducement stage and prosocial behavior stage. All participants listened to the same soothing music (“Dancing with the Neon Light”) for one minute before the experiment. Participants were randomly assigned to neutral mood group, and endogenous shame group. The neutral mood group read neutral materials, and the endogenous group read shame materials. After instructions onset, a fixation point was shown on the computer screen for 500 ms, followed by the emotion induction materials. Participants read the material, imagined themselves as the protagonists in the situations, and rated their own intensity of shame on a 4-point scale; The larger the number, the stronger was the intensity of shame. After a fixation point was shown for 500 ms, participants read their experience of getting help. Then, daily prosocial behavior and donating situations were presented; The order of the two situations was counterbalanced. The recipients in the situations sought help from the participants. The strangers in the situation asked for help from the participants. The participants randomly met different types of recipients, namely, the fellow-townsman (members of ingroup) and the non-fellow-townsman (members of outgroup). Participants were required to complete the choice of "whether or not they want to help" and "to what extent they are willing to help" (on 5-point scale). The former recorded the reaction time as the decision-making time of participants, while the latter recorded the self-determination score as the willingness to help.

The results showed that participants in the endogenous shame group demonstrated a stronger willingness to help others, when the outgroup asked for the help. Endogenous shame could eliminate the ingroup favoritism. Therefore, this study supports the restore drive of shame enhancing prosocial behavior.