Micro-expression is a brief,subtle facial expression,commonly occurs in high-stakes scenarios.It can help understand an individual’s intentions and motivation,reflecting the natural mental process for communication.Thus,micro-expression provides essential clues for lie detection.Recently,automatic micro-expression analysis has developed rapidly.However,there is a lack of research on the micro-expression cognitive mechanism,the automation of micro-expression analysis requires more theoretical support.Therefore,by manipulating the expressions induced by high-low risk scenarios under the paradigm of mock crime,this study explores the differences of micro-expressions generated by different risk scenarios,so as to further explore the generation mechanism of micro-expressions.Our experimental results and quantitative analysis proved the following conclusions:(1)The physiological indicator of electrodermal activity(EDA)showed significant differences between the two risk-level scenarios we set.(2)The number of micro-expressions expressed in the high-stakes scenario was more than that in the low-stakes scenario.(3)Expression type and risk environment each had distinct effects on EDA,and there was an interaction effect between these two factors.In brief,as an empirical study providing the first evidence on the mechanism of micro-expression generation,our study demonstrated that micro-expressions occur primarily in high-stakes scenarios.

微表情是一种简短且微妙的面部表情。它可以帮助理解个体的意图和动机,反映真实的心理过程。近年来,微表情的自动化分析发展迅速,但对微表情认知机制的研究实际上仍十分缺乏。通过研究模拟犯罪范式下高低风险场景产生的表情,探讨高低风险场景产生微表情的差异,从而进一步研究微表情行为背后的生成机制。实验结果显示:(1)高风险场景下产生的微表情数量大于低风险场景下的数量。(2)皮肤电活动(EDA)在风险场景(高风险/低风险)和表情类型(微表情/宏表情)下都存在显著差异。(3)EDA在表情类型和风险场景之间存在交互作用。结果表明,相比低压场景,微表情更容易在高压下产生。研究首次为微表情产生机制提供了实证研究。