Frequency-Aware Spatial-Temporal Attention Explainable Network for EEG Decoding

• Published in: IEEE journal of biomedical and health informatics Vol. 29; no. 10; pp. 7175 - 7185
• Main Authors: Jin, Luyao, Song, Yonghao, Zhao, Huan, Cao, Junyi, Cheung, Vincent C. K., Liao, Wei-Hsin
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.10.2025
• Subjects: Adult; Brain - physiology; Brain modeling; brain-computer interface (BCI); Brain-Computer Interfaces; Decoding; Discrete Fourier transforms; EEG; Electroencephalography; Electroencephalography - methods; Emotion recognition; Emotions - physiology; Entropy; Feature extraction; Filter banks; Frequency-aware; Humans; Motors; Neural Networks, Computer; Signal Processing, Computer-Assisted; transformer; Transformers
• ISSN: 2168-2194; 2168-2208; 2168-2208
• DOI: 10.1109/JBHI.2025.3576088

Representation learning in spatial and temporal domains has shown significant potential in EEG decoding, advancing the field of brain-computer interfaces (BCIs). However, the critical role of frequency information, closely tied to the brain's neurological mechanism, has been largely neglected. In this paper, we propose FSTNet, which integrates frequency-spatial-temporal domains synergistically. The network allows broadband EEG signals as input and adaptively learns informative frequency signatures. A frequency-aware module emphasizes the importance of frequency information by selectively assigning weights to latent representations in the frequency space. Subsequently, self-attention captures spatial and temporal dependencies, extracting discriminative neural signatures for EEG decoding. We conducted extensive experiments on EEG datasets for motor imagery and emotion recognition, achieving superior results on SEED, PhysioNet, and OpenBMI datasets in both individual and cross-subject scenarios. Additionally, visualization reveals that the network captures informative frequency ranges and spatial patterns associated with specific tasks, aligning with known physiological mechanisms. This enhances the transparency of the network's learning process. In conclusion, our method exhibits the potential for decoding EEG and advancing the understanding of neurological processes in the brain.