认知与发展心理学研究室知识库

情绪预期是个体对未来情景中将要体验到的情绪进行预测的能力，是人类生活中重要的适应功能。研究表明，存在快感缺失和动机缺乏的亚临床人群社会功能受损，更有发展为临床精神疾病患者的风险。分裂型特质个体、孤独特质个体、抑郁状态个体等存在快感缺失和动机缺乏的亚临床人群表现出情绪预期缺损。然而，前人研究仅限于对亚临床人群是否存在情绪预期缺损进行探讨，且研究结论不一；亚临床人群情绪预期缺损背后的加工机制、不同亚临床人群情绪预期缺损的异同之处仍需要进一步探讨。因此，本论文将通过四个研究分别采用自我报告、行为测试、功能影像任务和经验取样的研究方法，从不同层面对亚临床人群情绪预期缺损的特点及其加工机制进行系统的探讨。

研究一采用自评问卷和网络分析的研究方法，在一般大学生群体中探讨了与情绪预期缺损直接相关的亚临床特征。研究发现，分裂型特质、孤独特质、抑郁状态与减少的正性预期情绪直接相关。其中，分裂型特质的人际问题、孤独特质的社交技巧问题与社会性预期情绪的减少有关；抑郁状态与一般性预期情绪的减少有关。

研究二采用社会性情绪预期行为任务探讨分裂型特质、孤独特质、抑郁状态在社会和非社会条件下对情绪预期的影响。研究发现分裂型特质的人际问题能够负向预测社会条件下的预期情绪效价。在一般人群中，个体的心理模拟过程与期待性情绪均可预测情绪预期。分裂型特质的人际问题通过预测期待性情绪来预测预期情绪，而分裂型特质人际得分较高的人群心理模拟能力相对完好。

研究三的实验一采用功能影像任务探讨心理模拟过程的脑激活特点。研究发现个体在进行心理模拟加工时，核心网络(内侧前额叶、后扣带回、楔前叶、海马旁回等)明显激活，内侧前额叶在正性情绪加工时激活显著。分裂型特质个体、孤独特质个体在进行心理模拟加工时，核心网络脑区激活减弱；分裂型特质个体在正性情绪加工时表现出尾状核异常激活。实验二结合静息态功能连接与社会性情绪预期行为任务，探讨以上亚临床群体与情绪预期相关的异常功能连接。研究发现，孤独特质个体的内侧前额叶-额中回功能连接明显增强，且异常增强的功能连接与社会条件下减少的正性预期情绪效价相关显著，与非社会条件下的正性预期情绪效价无关。

研究四基于大样本数据对日常生活中情绪预期的特点及其影响因素进行探讨。研究结果发现，个体在进行情绪预期时高估情绪的强度。当下情绪体验与事件的社会性特点影响情绪预期。抑郁症状更明显的个体预期更强的情绪强度。

综上所述，研究发现分裂型特质个体的缺损主要体现在情绪预期加工的情绪成分(期待情绪)中；孤独特质个体的缺损主要体现在认知成分(心理模拟)中；抑郁状态个体的情绪预期缺损主要体现在日常生活中。分裂型特质个体与孤独特质个体的情绪预期缺损主要存在于社会条件中。本研究强调了社会成分在情绪预期中的重要作用，补充了心理模拟理论的相关脑机制，为非药物干预提供了靶点建议，具有理论意义与临床价值。

Affective forecasting, the ability to predict one’s emotional feedback in the future, is one of the most important adaptive functions in human daily life. Subclinical populations with anhedonia and amotivation show impaired social functioning. Empirical findings suggest that subclinical populations with anhedonia and amotivation such as individuals with schizotypal traits, autistic traits and subsyndromal depression, also exhibit deficit in affective forecasting. However, previous studies have only used self-reported behavioural measures to assess affective forecasting in these subclinical individuals. Findings from the extant literature are inconsistent. The neural correlates of affective forecasting deficits in these subclinical individuals remain to be investigated. To bridge this gap in knowledge, we conducted four studies to systematically investigate the patterns of affective forecasting in individuals with schizotypal traits, autistic traits, and subsyndromal depression and to explore the neural correlates of affective forecasting deficits.

Using network analysis and self-report measures, Study 1 examined the pattern of the relationship between anticipated pleasure and subclinical features in a general college sample. We found that schizotypal traits, autistic traits and depressive symptoms were associated with reduced anticipated pleasure. Moreover, interpersonal features of schizotypal traits and poor social skills of autistic traits showed a strong correlation with reduced social anticipated pleasure and depressive symptom severity was found to be weakly associated with reduced abstract anticipatory pleasure.

the social affective forecasting task, Study 2 investigated the underlying mechanism of reduced anticipated pleasure in individuals with schizotypal traits, autistic traits and depressive symptoms, in both social and non-social conditions. We found that interpersonal features of schizotypal traits negatively predicted anticipated emotion (positive valence) under social conditions. Moreover, both mental simulation and anticipatory emotion positively predicted anticipated emotion (positive valence). Interpersonal features were also found to be correlated with reduced anticipatory pleasure suggesting that interpersonal features might reduce anticipated pleasure by reducing anticipatory pleasure, while individuals with higher levels of interpersonal features showed intact mental simulation.

Experiment 1 in Study 3 investigated the shared and unique neural underpinnings of individuals with schizotypal traits, autistic traits and subsyndromal depression, using the mental simulation imaging task. Regions in the “core network” including the medial frontal gyrus, the posterior cingulate cortex, the precuneus and the parahippocampus were activated during mental simulation in healthy controls. The medial frontal gyrus was activated during positive emotion simulation. Compared with the control group, hypoactivation in the “core network” regions were observed in the autistic traits group and the schizotypal traits group during mental simulation. Compared with healthy controls, individuals with schizotypal traits showed hyperactivation at the caudate during positive emotion simulation. Then, Experiment 2 in Study 3 investigated the altered functional connectivity associated with affective forecasting in these individuals. We found that individuals with autistic traits exhibited significantly stronger functional connectivity between the medial frontal gyrus and the middle frontal gyrus. More importantly, the observed abnormal functional connectivity was negatively correlated with anticipated pleasure in social conditions, but not in non-social conditions.

Using experience sampling methods, Study 4 investigated the pattern of affective forecasting in daily life and explored their corresponding correlations with subclinical features. We found that individuals overestimated the intensity of emotions. Anticipatory valence was positively correlated with anticipated valence, while social condition was correlated with higher levels of anticipated valence. Individuals with more depressive symptoms anticipated stronger emotion intensity than individuals with less depressive symptoms.

together, our findings show that affective forecasting impairment in individuals with schizotypal traits is mainly due to problems in cognitive components (mental simulation), while affective forecasting impairment in individuals with autistic traits is mainly due to problems in emotional components (anticipatory emotion). On the other hand, individuals with subsyndromal depressive symptoms exhibit affective forecasting deficit mainly in their daily life. Lastly, schizotypal traits and autistic traits are associated with social-specific affective forecasting impairment. This study emphasized the importance of social components in affective forecasting. Moreover, it extended the mental simulation hypothesis by showing evidence of their behind neural mechanism. Besides, this study provided valuable suggestions for future intervention that social components might be the target to improve mental health.