亲社会行为是指那些给他人带来利益的行为，包括合作、分享、助人、捐赠、志愿等行为。这些行为是维持人与人之间良好关系的重要基础，也是建立公正、和谐社会的重要保障，具有重大的进化意义和社会意义。多年来，心理学家从各个角度探讨亲社会行为的发生与发展。首先，在测量方式上，研究者根据研究内容和研究兴趣发展出了多种测量方式。其次，在理论解释上，研究者不断拓展、完善亲社会理论，以期对各种亲社会行为的发生做出解释。最后，在影响因素上，研究者从影响亲社会行为的前端因素、过程因素以及结果因素上进行了系统而深入的探讨。但在这些因素中，结果因素得到的关注较少。以往对影响亲社会行为的结果因素的少量探讨主要关注助人者对自身结果的预期可能会产生的影响，如助人者获益或损失所产生影响，再如对执行或不执行亲社会行为之后产生的情绪预期可能产生影响。但是亲社会行为还涉及受助者这一对象，以往研究未曾关注受助者的结果类型对助人者的亲社会行为的影响。受助者的结果包括获得积极结果和避免消极结果两种，本文主要关注个体在两种受助者结果类型下的亲社会表现差异。并以此为核心，探讨这种差异的发展特点，以及助人者结果类型、助人者调节聚焦对这种差异的影响。这些研究问题通过五项研究来解决。研究一主要关注在受助者获得积极结果和受助者避免消极结果两种条件下大学生的亲社会表现差异，通过三个实验证明了使受助者避免消极结果这一条件比使受助者获得积极结果这一条件诱发更高的亲社会表现。其中实验一通过让106名大学生直接表达其亲社会倾向，发现相对于帮助他人获得积极结果，更多人选择帮助他人避免消极结果。实验二在实验一的基础上考察了60名大学生对受助者所获得的不同结果类型表达的亲社会意愿程度的差异，结果发现被试帮助他人避免消极结果的意愿程度显著高于帮助他人获得积极结果的意愿程度。实验三以32名大学生为被试，在实验室情境中再次验证了该现象，并且除了验证个体在两种受助者结果类型中的亲社会意愿差异，还通过资格测试的形式验证了个体在两种受助者结果类型中的亲社会行为差异。结果发现，无论是亲社会意愿还是亲社会行为，使受助者避免消极结果都比使受助者获得积极结果诱发更高的亲社会表现。这些结果证明了经济领域的损失规避原则（损失比同等程度的获得诱发更高的情感体验，在获益时，相对于获得，人们更倾向于避免损失）同样适用于亲社会领域，即存在亲社会损失规避现象。研究二考察受助者结果类型对亲社会表现的影响的发展特点。该研究通过自由表达亲社会意愿、强行分配亲社会意愿以及亲社会情绪三个指标考察了92名二、四、六年级儿童亲社会损失规避的发展特点。结果发现在自由表达意愿上，二年级儿童帮助他人获得积极结果的亲社会意愿程度高于帮助他人避免消极结果的亲社会意愿程度。四年级儿童的三个亲社会指标在两种受助者结果类型下都没有显著差异。六年级儿童在强行分配亲社会意愿中，更多人选择帮助他人避免消极结果；在亲社会情绪上，六年级儿童在帮助他人避免消极结果时更高兴。总的来说，随着年龄的增长，小学儿童的亲社会损失规避现象越来越明显。研究三以100名大学生为被试，同时考察受助者结果类型和助人者结果类型如何影响亲社会表现。研究发现在助人者获益条件下，个体的亲社会表现高于无损益条件和助人者损失条件。并且在助人者获益条件下，使受助者获得积极结果和使受助者避免消极结果诱发的亲社会表现无显著差异；在助人者损失条件下，使受助者避免消极结果诱发的亲社会表现高于使受助者获得积极结果。助人者损失条件下的亲社会损失规避程度与无损益条件无显著差异，但高于助人者获益条件。研究四同时考察受助者结果类型和助人者调节聚焦对亲社会表现的影响。该研究分为两个实验，实验一对62名大学生的特质性调节聚焦进行测量，然后根据得分将被试划分为特质性促进聚焦和特质性防御聚焦两组。实验二以106名大学生为被试，通过实验情境分别诱发情境性促进聚焦和情境性防御聚焦。然后分别比较两组被试在不同受助者结果类型下的亲社会表现。结果发现，无论是特质性调节聚焦还是情境性调节聚焦，在促进聚焦条件下，使受助者获得积极结果和使受助者避免消极结果诱发的亲社会表现无显著差异；在防御聚焦条件下，使受助者避免消极结果诱发的亲社会表现高于使受助者获得积极结果条件。防御聚焦条件下的亲社会损失规避程度与基线条件无显著差异，但高于促进聚焦条件。研究五以184名大学生为被试，同时考察受助者结果类型、助人者结果类型以及情境性调节聚焦对个体亲社会表现的影响。结果发现三阶交互作用不显著，在二维交互中，只有受助者结果类型和助人者结果类型对个体亲社会表现的交互作用显著。并且通过比较四组条件2（调节聚焦：促进聚焦vs. 防御聚焦）×2（助人者结果类型：获益vs. 损失））与基线组的亲社会损失规避程度差异，发现只有诱发防御聚焦、助人者获益组的亲社会损失规避行为显著低于基线组。进一步分析显示，诱发防御聚焦、助人者获益组个体帮助他人获得积极结果的亲社会行为显著高于基线组，诱发防御聚焦、助人者获益组个体帮助他人避免消极结果的亲社会行为与基线组无显著差异。综上，本论文发现相对于帮助他人获得积极结果，人们更倾向于帮助他人避免消极结果，从而证明了亲社会情境中的损失规避原则和框架效应。并且，亲社会损失规避程度随着年龄的增长而越来越明显。另外，亲社会损失规避受到助人者结果类型和助人者调节聚焦的影响。

Prosocial behavior refers to a broad category of actions that benefit other people or society, such as helping, comforting, sharing, cooperation, philanthropy, and community service. These actions are the basis of interpersonal relationship and harmonious society. Psychologists have studied prosocial behavior from multiple perspectives. Firstly, there are many kinds of operational definition for prosocial behavior. Secondly, various theories and models are used to explain the prosocial behavior. Lastly, a growing stream of research has investigated the factors that influence individuals’ prosocial behavior. These factors can be divided into three categories, i.e., individual differences, situational variables, and outcome-related variables of prosocial behavior. Previous research on the determinants of prosocial behavior has primarily focused either on the demographic and individual characteristics of the helper or on situational factors; however, less attention has been given to the roles of outcome-related variables.Many theories and studies have proved that expected outcomes influence individuals’ motivation and behavior. For example, some researchers found that anticipated regret at not donating blood in the future strengthen individuals’ blood donation intentions. However, these theories and studies addressed the role of expected outcome in term of the helpers.In fact, prosocial behavior creates a link between the helper and the beneficiary. Although there were a few studies explored the effect of expected outcome on prosocial performance in term of the helper, to date, little work has investigated the role of outcome in term of the beneficiary. In general, there are two types of outcome for the beneficiary, obtaining positive outcomes and avoiding negative outcomes. Although the two types of prosocial outcome are equal in the valence, the effects of two types of outcome may be different. The hypothesis about the difference derived from the theory and studies about loss aversion. The main purpose of the present research was to examine the effect of outcome-types for the beneficiary on prosocial performance, and then to examine the influence of age, outcome-types for the helper, and regulatory focus. Three experiments in the first study were designed to examine the difference of prosocial performance between attaining positive outcomes (prosocial gains) and avoiding negative outcomes (prosocial non-losses). We observed a phenomenon of prosocial loss aversion as evidence by the finding that prosocial non-loss outcomes induced greater prosocial performance than did prosocial gain outcomes. In Experiment 1, more students chose to help others avoid negative outcomes. In Experiment 2, the prosocial willingness of prosocial non-losses was higher than that of prosocial gains. Experimental situation and quantity test was used in Experiment 3. We found that prosocial non-loss outcomes induced greater prosocial performance than did prosocial gain outcomes.The second study examined the development of prosocial loss aversion. The children who were second-graders, fourth-graders, and sixth-graders respond on three index of prosocial performance. For the first index, the prosocial willingness of the second-graders was higher for attaining positive outcomes than that for avoiding negative outcomes. The fourth-graders showed no different prosocial performance on three prosocial indexes between two types of outcome for the beneficiary. More sixth-graders chose to help others avoid negative outcomes, and helping others avoid negative outcomes made them happier. In study 3, both the effects of outcome-types for the beneficiary and for the helper on prosocial performance were examined. The results showed that the prosocial performance was higher in the helper-gain condition than that in the helper-loss condition. The difference of prosocial performance between prosocial gains and prosocial non-losses was not significant in the helper-gain condition, whereas prosocial non-losses induced higher prosocial performance than prosical gains in the helper-loss condition. These results meant that the prosocial loss aversion in the helper-loss condition was higher than that in the helper-gain condition.Study 4 was designed to examine the influence of regulatory focus on the relationship between outcome-types for the beneficiary and prosocial performance. There were two experiments in this study. The chronic regulatory focus was measured by a scale in Experiment 1, and all the participants were classified into promotion focus and prevention focus according to their scores. The situational regulatory focus was primed by different situations in Experiment 2. Both for Experiment 1 and Experiment 2, the difference of prosocial performance between prosocial gains and prosocial non-losses was not significant in the promotion-focused condition, whereas prosocial non-losses induced higher prosocial performance than prosocial gains prevention-focused condition. These results meant that the prosocial loss aversion in the prevention-focused condition was higher than that in the promotion-focused condition.The last study examined the effect of situational regulatory focus, the outcome-types for the beneficiary and for the helper on prosocial performance. The results showed that the interaction between the outcomes-types for the beneficiary and for the helper was significant. The result of simple effect analysis was similar to that in Study 3. The results also showed that the prosocial loss aversion in the prevention-focus-primed and helper-gain condition was significantly smaller than that in the base-line condition because helper-gain increased the prosocial performance for helping others attaining positive outcomes.All the studies implied that there was a phenomenon of prosocial loss aversion as evidence by the finding that prosocial non-loss outcomes induced greater prosocial performance than did prosocial gain outcomes, and prosocial loss aversion was influenced by age, the outcome-types for the helper, and regulatory focus.