Real-time emotional state recognition from neural activities has great potentials, e.g., enabling closed-loop systems to treat neuropsychiatric disorders. Besides its utility in identifying epileptogenic brain regions, electrocorticography (ECoG) provides the opportunity to study brain activities during various emotional events. In this study, we aim to test if the second order statistics such as the electrode network connectivity are relevant in discriminating different emotions. Specifically, we adopt the statistical dependence as the connectivity measure and use sparse logistic regression for classification based on the connectivity features. The practical issues of limited samples and imbalanced data are addressed. We show that the data with the highest synchronization in the γ band provide the most competitive performance. The identified brain regions involved in the most discriminative connectivity patterns are consistent with the previous findings in the literature.
Using one of the most comprehensive multimodal biosensing datasets in affective research, densely sampled within individuals, we asked how emotions are instantiated in each person’s brain and body. Four participants viewed 168 emotion-evoking videos during 128-channel EEG and 28 autonomic, ocular and facial recordings. Cross-validated elastic-net classifiers distinguished each of 15 positive and neutral emotions from baseline in every participant (mean accuracy 87-97%). We found that no emotion was predicted by a single marker and each reflected a multimodal ensemble of cortical and autonomic activity whose composition was largely person-specific. Yet physiology-derived groupings converged selectively with self-report; Interest-Awe-Amusement-Satisfaction and Entrancement-Aesthetic Appreciation recurred across individuals based on physiology. Across emotion groups, frontal and posterior delta, posterior alpha and electrodermal activity formed a shared backbone, plausibly reflecting affective and sensory engagement, whereas cardiovascular and facial signals varied between people. This framework yields individualized, objective neurophysiological indicators of emotion, potentially aiding personalized diagnosis.