From Emotion to Error: How Physiological Arousal Drives Emergency Decision Failure in Virtual Mining

This dataset supports the study entitled “From Emotion to Error: How Physiological Arousal Drives Emergency Decision Failure in Virtual Mining – A Study Based on a Multi-Level Moderating Mediation Model.” The data were collected through an immersive virtual reality mine-fire emergency experiment involving 1,000 valid participants nested within 50 work teams . The dataset contains anonymized physiological, behavioral, psychological, and group-level variables, including heart rate, skin conductance, systolic and diastolic blood pressure, reaction time, error rate, negative emotion, task disengagement, decision impulsivity, group safety management climate, group emotion regulation climate, and graded emergency response competence. These data were used for multilevel modeling, mediation and moderation analyses, and interpretable machine-learning analyses. All personally identifiable information has been removed to protect participant confidentiality.本数据集支撑题为《情绪致错：生理唤醒如何驱动虚拟矿山场景下的应急决策失误——基于多层级有调节中介模型的研究》的学术研究。本数据集的数据源自一项沉浸式虚拟现实（Immersive Virtual Reality）矿山火灾应急实验，共纳入50个工作班组嵌套的1000名有效参与者。数据集包含匿名化处理的生理、行为、心理及班组层级变量，涵盖心率、皮肤电导率、收缩压、舒张压、反应时、错误率、负性情绪、任务脱离行为、决策冲动性、班组安全管理氛围、班组情绪调节氛围以及分级应急响应能力。该数据集可用于多层级建模、中介与调节效应分析以及可解释机器学习分析。为保护参与者隐私，所有可识别个人身份的信息均已移除。