近年来，随着“北京中关村二小校园欺凌事件”的爆出，同伴侵害现象逐渐走入了公众的视野，引发了国家和公众，尤其是学生家长的广泛关注。同伴侵害指的是个体遭到来自同伴直接或间接攻击的经历，是全世界范围内均普遍存在的公共健康问题。近三十年来的大量研究发现，同伴侵害对儿童的身心健康发展，尤其是内外化问题存在非常严重的消极影响。但是，探讨同伴侵害对儿童生理应激系统功能的研究正处于萌芽阶段，更鲜有研究考察生理应激系统功能在同伴侵害与儿童内外化问题间可能具有的中介作用。以往为数不多关于同伴侵害和儿童生理应激系统功能关系的研究也主要是在西方文化背景下进行的，得出的结论是否适用于集体主义文化背景下的中国儿童也尚未可知。因此，本研究拟在中国文化背景下，探讨同伴侵害的两种典型类型——关系侵害和身体侵害与儿童生理应激系统功能——下丘脑-垂体-肾上腺（hypothalamic-pituitary-adrenal, HPA）轴功能的关系，并进一步考察HPA轴功能是否会在两种同伴侵害类型与儿童的内外化问题间起到中介作用。 本研究以北京市某小学五、六年级共115名儿童（N=115，9.25–13.67岁）为被试，使用儿童版特里尔社会应激测试（Trier Social Stress Test for Children, TSST-C）诱发儿童的应激反应，并在应激任务前后测查6次皮质醇水平以计算应激下皮质醇反应。关系侵害和身体侵害为儿童自我报告，内化问题和外化问题由儿童家长进行报告。为了充分地考察同伴侵害（即关系侵害和身体侵害两种类型）与儿童应激下皮质醇反应及内外化问题的关系，我们不仅以同伴侵害水平这一连续变量为指标，分析同伴侵害水平与儿童发展结果的关联，也筛选出了遭受过较为严重侵害的儿童，比较遭受过同伴侵害与未遭受过同伴侵害的儿童发展结果的差异。 结果发现：（1）TSST-C诱发了中国儿童显著的皮质醇反应，儿童在应激结束后10分钟的皮质醇水平均显著大于基线阶段。（2）总体同伴侵害水平与应激下皮质醇反应不存在关联。但是，相比于未遭受过同伴侵害的儿童，遭受过同伴侵害的儿童表现出迟钝的应激下皮质醇反应。在控制人口学变量后，此差异依旧存在。（3）关系侵害与儿童迟钝的应激下皮质醇反应相关联。具体而言，关系侵害水平与应激下皮质醇反应存在负相关，且在控制人口学变量和身体侵害的情况下，关系侵害水平依旧可以显著预测迟钝的应激下皮质醇反应。此外，相比于未遭受过关系侵害的儿童，遭受过关系侵害的儿童表现出更加迟钝的应激下皮质醇反应，在控制人口学变量和身体侵害的情况下，此差异依然边缘显著。而身体侵害水平与应激下皮质醇反应不存在关联，遭受过身体侵害与未遭受过身体侵害的儿童，应激下皮质醇反应也不存在差异。（4）迟钝的应激下皮质醇反应在关系侵害水平和内外化问题间具有中介作用。具体体现为，在控制人口学变量和身体侵害水平的情况下，关系侵害可以通过预测儿童迟钝的应激下皮质醇反应，进而预测儿童出现更多的内外化问题。此外，相比于未遭受过关系侵害的儿童，遭受过关系侵害的儿童也表现出迟钝的应激下皮质醇反应，进而出现更多的内外化问题。然而，我们并未发现应激下皮质醇反应在身体侵害（包括水平和组别）与内外化问题间的中介作用。不过，身体侵害水平对儿童的外化问题具有直接的正向预测作用。且相比于未遭受过身体侵害的儿童，遭受过身体侵害的儿童有更多的外化问题。 本研究采用主观心理指标和客观生理指标测查、自评和他评问卷相结合的研究方式，精细地考察了两种不同的同伴侵害类型与儿童发展之间的关系。首次在中国文化背景下，探讨了关系侵害对儿童生理应激系统功能所具有的重要和独特的作用，以及生理应激系统功能在关系侵害与儿童内外化问题间所起到的中介作用。更为重要的是，我们的研究揭示了关系侵害和身体侵害与儿童内外化问题间的关联具有不同的作用机制。研究结果提示我们，未来的研究有必要单独考察关系侵害和身体侵害两种不同的同伴侵害类型对儿童发展所具有的不同及独特作用，这可以提供给我们更多、更丰富的信息。此外，研究结果对于未来针对不同同伴侵害类型和内外化问题的预防和干预工作也具有一定的参考价值。未来对于内外化问题的预防和干预中，不仅要关注内外化问题现象本身，还要结合行为和生理层面的病因学，对儿童进行有针对性的预防和干预。

In recent years, with the case of bullying at Beijing Zhongguancun No. 2 Primary School of the outbreak, the phenomenon of peer victimization has gradually entered the public's field of vision, which has attracted considerable attention from the government and the public, especially from the parents. Peer victimization refers to the experience of being a target of aggressive behaviors by similar-aged peers, which is a common public health problem all over the world. Over the past three decades, a large body of concurrent and longitudinal work has shown that children experienced peer victimization demonstrated increased risk for psychopathology. However, research on the effects of peer victimization on children's physiological stress systems is still in its infancy, and the physiological mechanisms linking peer victimization to child psychopathology remain inadequately understood, especially in a non-Western context. This study examined the relationship between two typical forms of peer victimization (i.e., relational victimization and physical victimization) and hypothalamic-pituitary-adrenal (HPA) axis response to stress and, furthermore, explored whether HPA axis response to stress mediated the associations of exposure to different types of peer victimization with child psychopathology (i.e., internalizing and externalizing problems) in a sample of Chinese children. The sample included 115 (range 9.25–13.67) healthy children of Grade 5th and Grade 6th recruited from an elementary school in Beijing, China. Participants reported experiences of relational victimization and physical victimization that occurred over the past year, and participated in a standardized laboratory psychosocial stress task (i.e., Trier Social Stress Test for Children, TSST-C), during which six segments of salivary cortisol samples were collected. One of participant’s parents reported on measures of child internalizing and externalizing problems in the past year. To fully investigate the associations among peer victimization (relational and physical victimization), cortisol response and child internalizing and externalizing problems, we not only examined the relationships of levels of peer victimization and child development, but selected children experiences of severe peer victimization to compare the developmental outcomes of victims of peer victimization with non-victims. Results revealed that: (1) TSST-C induced a significant cortisol response in Chinese children. The cortisol levels assessed 10 min after the task were significantly higher than those at baseline. (2) The association between levels of overall peer victimization and cortisol response failed to reach significance. However, in comparison with non-victims, children exposed to peer victimization showed blunted cortisol response, even after controlling for covariates. (3) Relationship victimization was associated with blunted cortisol response. Specifically, levels of relational victimization were associated with blunted cortisol response. After controlling for the impact of covariates and physical victimization, the unique effect of relationship victimization remained significant. Additionally, in comparison with non-victims, victims of relational victimization displayed a blunted cortisol response. And the difference in cortisol response remained marginally significant after controlling for the impact of covariates and physical victimization. However, the association between levels of physical victimization and cortisol response was non-significant, and there was also no difference in cortisol response between victims of physical victimization and non-victims. (4) The indirect effects of relational victimization on internalizing and externalizing problems through blunted cortisol response were significant. After controlling the effects of covariates and physical victimization, levels of relational victimization were associated with blunted cortisol response, which in turn related to more internalizing and externalizing problems. And children experiences of relational victimization exhibited a blunted cortisol response compared to non-victims, thus were prone to more internalizing and externalizing problems. However, cortisol response did not mediate the relationship between physical victimization and internalizing and externalizing problems. Notably, we found that levels of physical victimization were directly associated with more externalizing problems, and victims of physical victimization exhibited higher levels of externalizing problems. In the present study, we adopted subjective psychological and objective physiological measures, as well as self-report and other-report questionnaires to examine the associations among two different types of peer victimization and child development. For the first time in the Chinese context, we found the important and unique role of relational victimization on children's HPA axis stress response, and the HPA axis stress response mediating the relationships between relational victimization and child internalizing and externalizing problems. More importantly, our study revealed different mechanisms by which different forms of peer victimization relate to psychopathology. This result offers important evidence that risks for psychopathology are at least partially distinct with regard to experiences of different subtypes of peer victimization and underscores the benefit of distinguishing between relational victimization and physical victimization for future research and intervention. Distinguishing distinct pathways through which different types of peer victimization influence psychopathology may yield valuable implications for targeting intervention approaches in terms of psychopathology.