EmoEEG-MC: A Multi-Context Emotional EEG Dataset for Cross-Context Emotion Decoding

作者：Xin XU, Xinke SHEN, Xuyang CHEN, Qingzhu ZHANG, Sitian WANG, Yihan LI, Zongsheng LI, Dan Zhang, Mingming ZHANG, Quanying Liu · 发表于：UC San Diego · 年份：2026 · DOI：10.82901/nemar.on005540 · 研究领域：Psychology、Computer science、Artificial intelligence、Cognitive psychology、Speech recognition

EmoEEG-MC is a multi-context emotional EEG dataset comprising 64-channel EEG and peripheral physiological recordings from 60 participants exposed to video-induced and imagery-induced emotional stimuli across seven emotion categories (joy, inspiration, tenderness, fear, disgust, sadness, and neutral). This dataset addresses the critical gap in cross-context emotion decoding by enabling investigation of how emotional neural responses generalize across different elicitation contexts, with demonstrated classification accuracies of 66.7% for binary emotion classification and 28.9% for seven-category emotion classification using machine learning approaches.