ASPEN: Spectral-Temporal Fusion for Cross-Subject Brain Decoding
This addresses the problem of individual variability in EEG signals for brain-computer interface users, offering an incremental improvement through a novel fusion method.
The paper tackled cross-subject generalization in EEG-based brain-computer interfaces by investigating spectral features for stability and introducing ASPEN, a hybrid architecture that fuses spectral and temporal streams via multiplicative fusion, achieving the best unseen-subject accuracy on three of six benchmark datasets and competitive performance on others.
Cross-subject generalization in EEG-based brain-computer interfaces (BCIs) remains challenging due to individual variability in neural signals. We investigate whether spectral representations offer more stable features for cross-subject transfer than temporal waveforms. Through correlation analyses across three EEG paradigms (SSVEP, P300, and Motor Imagery), we find that spectral features exhibit consistently higher cross-subject similarity than temporal signals. Motivated by this observation, we introduce ASPEN, a hybrid architecture that combines spectral and temporal feature streams via multiplicative fusion, requiring cross-modal agreement for features to propagate. Experiments across six benchmark datasets reveal that ASPEN is able to dynamically achieve the optimal spectral-temporal balance depending on the paradigm. ASPEN achieves the best unseen-subject accuracy on three of six datasets and competitive performance on others, demonstrating that multiplicative multimodal fusion enables effective cross-subject generalization.