中国科学院心理健康重点实验室知识库

We introduce a novel multimodal emotion recognition dataset that enhances the precision of Valence Arousal Model while accounting for individual differences. This dataset includes electroencephalog raphy (EEG), electrocardiography (ECG), and pulse interval (PI) from 64 participants. Data collection employed two emotion induction paradigms: video stimuli that targeted different valence levels (positive, neutral, and negative) and the Mannheim Multicomponent Stress Test (MMST), which induced high arousal through cognitive, emotional, and social stressors. To enrich the dataset, par ticipants’ personality traits, anxiety, depression, and emotional states were assessed using validated questionnaires. By capturing a broad spectrum of affective responses while accounting for individual differences, this dataset provides a robust resource for precise emotion modeling. The integration of multimodal physiological data with psychological assessments lays a strong foundation for personal ized emotion recognition. We anticipate this resource will support the development of more accurate, adaptive, and individualized emotion recognition systems across diverse applications.