EEG-Based Micro-Expression Recognition: Flexible Brain Network Reconfiguration Supporting Micro-Expressions Under Positive Emotion

Abstract

Purpose: Micro-expression recognition is valuable in clinical, security, judicial, economic, educational, and human-computer interaction fields. Electroencephalography (EEG)-based micro-expression recognition has gained attention for its objectivity and resistance to interference, unlike image-based methods. However, the neural mechanisms of micro-expressions remain unclear, limiting the development of EEG-based recognition technology. Methods: We explored the brain reorganization mechanisms of micro-expressions (compared with macro-expressions and neutral expressions) under positive emotions across global networks, functional network modules, and hub brain regions using EEG, graph theory analysis, and functional connectivity. Results: In global network, micro-expressions demonstrated higher network efficiency, clustering coefficient, and local efficiency, along with shorter average path lengths. In functional network modules, micro-expressions enhanced connectivity between the bilateral superior frontal gyrus (SFG), anterior cingulate cortex, and ventromedial prefrontal cortex (cognitive control), as well as between the left orbitofrontal cortex (OFC), temporal pole (TP), and inferior frontal gyrus (emotional processing). In hub brain regions, micro-expressions increased the hub centrality, information transmission efficiency, and local clustering of bilateral SFG, left OFC, left TP, and left Broca’s area. Conclusion: Micro-expressions require more efficient global communication and specialized emotion and cognitive control modules. Key hub regions supporting positive micro-expressions include the bilateral SFG (inhibitory control), left OFC and TP (emotion processing), and left Broca’s area (language processing).