EEG Decoding
EEG: Electroencephalography
Overview
EEG decoding studies how electrical activity recorded at the scalp can reveal brain states or support tasks such as motor-imagery control, sleep staging, auditory attention, and speech or image reconstruction. Approaches range from filtering, spectral features, common spatial patterns, and Riemannian covariance methods to CNNs, Transformers, state-space models, and self-supervised foundation models; generative systems also align EEG with text, images, or audio. A central challenge is generalizing across people, sessions, devices, and electrode layouts, since EEG is noisy and individual recordings vary substantially. Foundation models and increasingly broad benchmarks are expanding the toolkit, but their advantages over task-specific models are inconsistent, and leakage, shortcuts, privacy, and realistic cross-subject evaluation remain important concerns.
Latest papers
210 papers27 papers in the last four weeks, down 4% on the four weeks before. 0.2% of all new papers.