社会与工程心理学研究室知识库

青少年是发生自杀事件的主要群体，同时该年龄段出现自杀意念、非自杀性自伤行为等极端心理问题的现象也十分频繁，因此厘清自杀风险及其作用机制对于自杀预防具有重要意义。以往有大量研究指出了来自社会方面的危险因素对自杀行为有潜在的影响，但当下从社会视角出发的自杀相关研究，通常难以兼具长期纵向可回溯性和对社会价值层面的非客观指标进行观察。基于领域目前的不足方面，本研究依托于社会媒体的海量数据，实现了社会视角下的自杀风险因素影响机制的长期纵向观察，并对社会价值层面的社会道德、社会个体主义/集体主义进行量化计算，纳入自杀行为风险的关联因素观测范围。

为了深入分析自杀风险因素对不同群体的影响机制，本研究针对微博一般人群、青少年群体和高自杀风险青少年，分别开展其自杀风险影响机制的研究。研究发现：

（1）社会变化（包括社会环境变化和社会价值变化）可以视作外界压力源，通过心理疼痛这一中介因子，对微博一般人群的自杀行为风险产生正向或负向的影响。在社会环境方面，经济因素（居民可支配收入，β=-0.044，p<0.001）可以通过负向影响心理痛苦来缓解自杀行为风险；城市化因素（第一产业占比，β=0.037，p<0.001）可以正向预测心理疼痛和自杀行为风险。在社会价值方面，集体主义（β=0.086，p<0.001）的减少或个体主义（β=-0.077，p<0.001）的表达增加与都与低程度的心理疼痛相关，从而促进自杀行为风险的降低；社会道德中的反抗维度（β=-0.026，p<0.05）可以负向预测心理疼痛、肮脏维度（β=0.130，p<0.001）可以正向预测心理疼痛，并进一步与自杀行为风险产生关联。该研究以微博一般人群为对象，从群体视角分析了社会变化（包括社会环境变化和社会价值变化）对一般人群自杀行为风险的影响路径，同时强调了心理疼痛作为中介原因的重要性。

（2）青少年群体的自杀行为风险对社会环境变化不敏感，而受社会价值变化的影响较大。对于青少年群体来说，来自于社会变化（社会环境变化、社会价值变化）的压力源并不能通过诱发心理疼痛而影响该群体的自杀行为风险，更多通过绝望这一中介因素而发挥作用。在社会环境变化方面，离婚率能够降低绝望的表达（β=-0.026，p<0.05），而更低的绝望表达可以进而预测更高的自杀行为风险（β=-0.627，p<0.01）。对于青少年群体来说，绝望与自杀行为风险是负向预测的关系，即高自杀行为风险通常有更低的绝望表达。这可能暗示着在对绝望的表达方面，青少年高自杀风险用户可能偏好更为隐晦的表达方式。

(3）自杀问题严重的高风险青少年可能更少在微博表达心理疼痛和绝望感；而通常发表具有强烈心理疼痛和绝望感受的个体，往往处在较浅的自杀阶段（例如抑郁情绪阶段）。通过高自杀风险青少年所发微博的心理疼痛表达、绝望表达及其目前所处的自杀阶段发现，绝望表达可以直接负向预测用户 3 个月后的心理状态（β=-0.271，p<0.05）；心理疼痛表达可以通过与绝望表达的正向预测关系（β=0.571，p<0.001），间接预测其 3 个月后所处的自杀阶段（β=-0.154，p<0.05），两者存在完全中介效应。这说明，我们不能仅通过微博文字表达出的心理疼痛和绝望的程度高低来判断其距离自杀死亡的远近，因为自杀阶段较严重的青少年，往往不会直观地披露出过多心理疼痛和绝望感。

综上所述，本研究基于海量微博数据库，通过对三个不同层级的群体进行了较为有针对性的自杀风险影响机制分析。从社会群体视角，分别验证了心理疼痛和绝望感作为自杀行为风险前置因素的重要性，同时，发现了高自杀意念青少年自杀严重性与其心理疼痛和绝望感的表达关系。研究发现揭示了社会变化影响下的自杀行为产生动机及其恶化模式，对自杀风险干预提供了有利的科学建议，方便后续进行高自杀风险地区的预警以及自杀预防资源的合理分配。

Adolescents are the main group of suicide. There are extreme psychologicalproblems occur frequently at this age, such as suicidal ideation and non-suicidalself-injury behaviors. Therefore, clarifying suicide risk and the mechanism is of greatsignificance for suicide prevention. In the past, a large number of studies have pointedout the potential impact of social risk factors on suicidal behavior, but the currentsuicide-related research is difficult with both long-term longitudinal retrospective andnon-objective indicators of social value which from a social perspective. Based on thecurrent deficiencies in the field, this study relies on the massive data of social media,realizes a long-term longitudinal observation of the influencing mechanism of suiciderisk factors from a social perspective, and also quantifies social morality and socialindividualism/collectivism to be included in the observation range of factorsassociated with suicidal behavior risk.

In order to deeply analyze the influence mechanism of suicide risk factors ondifferent groups, this study carried out research on the influence mechanism of suiciderisk for the general population of Weibo, adolescent groups, and adolescents withhigh suicide risk. The study found:

(1) Social changes (including social environment and social values) could be regarded as external stressors, which through the intermediary factor of psychache,have a positive or negative impact on the risk of suicidal behavior in the generalpopulation of Weibo. In terms of social environment, economic factors (disposableincome of residents, β=-0.044, p<0.001) has alleviated the risk of suicidal behaviorby negatively affecting psychological distress; urbanization factors (the proportion ofprimary industry, β=0.037, p <0.001) has positively predicted psychache and risk ofsuicidal behavior. In terms of social value, a decrease in collectivism ( β =0.086,p<0.001) or an increase in the expression of individualism (β=-0.077, p<0.001) wereboth associated with lower levels of psychache, and also promoted a higher risk ofsuicidal behavior. The revolt (β=-0.026, p<0.05) in social morality has negativelypredicted psychological pain, and the dirty dimension ( β =0.130, p<0.001) haspositively predicted psychache and is further related to the risk of suicidal behavior.This study took the general population of Weibo as the sample, analyzed the impactpath from social changes (including social environment and social values) to the riskof suicidal behavior from a group perspective, and emphasized the importance ofpsychache as an intermediary cause.

(2) The risk of suicidal behavior in adolescents does not sensitive to the changesin the social environment but is greatly affected by changes in social values. Foradolescents, stressors from social changes (social environment or social values) do notaffect the risk of suicidal behavior by inducing psychache, but rather throughhopelessness. In terms of social environment changes, the divorce rate has reduced theexpression of hopelessness ( β =-0.026, p<0.05), and a lower expression ofhopelessness could in turn predict a higher risk of suicidal behavior ( β =-0.627,p<0.01). For adolescents, hopelessness and risk of suicidal behavior were negativelypredictive, that is, high risk of suicidal behavior generally had lower expressions ofhopelessness. This finding may suggest that adolescent users with high suicide riskmay prefer more subtle expressions of hopelessness.

(3) Adolescents with serious suicide problems may be less express psychacheand hopelessness on Weibo; while those who usually post strong psychache andhopelessness are often in a shallower suicide stage (such as the stage of depressedemotion). Through the psychache and despair expression posted by high suicide riskteenagers, we found that could directly predict the user's psychological state after 3months (β=-0.271, p<0.05 ); the expression of psychache could indirectly predict thesuicide stage after 3 months (β=-0.154, p<0.05) through the hopelessness(β=0.571,p<0.001). Results found that we cannot judge the possibility of suicide only by theexpression frequency of psychache and despair in Weibo text, because adolescentswith more serious suicide stages might not intuitively disclose their feeling ofpsychache and hopelessness.

(4) To sum up, based on the massive Weibo database, this study conducted a more targeted analysis, of the impact mechanism of suicide risk among three groupsat different levels. From the perspective of social groups, respectively verified theimportance of psychache and hopelessness as risk factors for suicidal behaviors. Atthe same time, the relationship between suicide and the expression of psychache andhopelessness in adolescents with high suicidal ideation was found. The findings revealthe motivation and deterioration pattern of suicidal behavior under the influence ofsocial changes, provide favorable scientific advice for suicide risk intervention,facilitate subsequent early warning of high suicide risk areas, and rational allocationof suicide prevention resources.