BCIwiki (bciwiki.com) — Five distinct cognitive processes triggered by the same visual stimulus can be told apart from electroencephalogram (EEG) signals with 91.72% ± 6.10% accuracy, Ju J and Li H reported in IEEE Transactions on Bio-Medical Engineering on January 16, 2026 (PMID 41543945, DOI 10.1109/tbme.2026.3654639).
The authors note that multi-task decoding is valuable for brain-computer interface (BCI) use in naturalistic settings, but most studies decode clearly different tasks, leaving the range of cognitive responses to a single stimulus underexplored. Their paradigm used 1 common visual stimulus to elicit single reach, interception reach, sequence reach, attention reach and inhibition reach.
EEG signatures were analyzed with temporal and spectral methods, showing significant activation differences across tasks and brain regions (p < 0.05). A regularized linear discriminant analysis (RLDA) classifier was trained on temporal and event-related spectral perturbation (ERSP) features. With ERSP features it classified the 5 cognitive states at 91.72% ± 6.10% accuracy. In the sequence reach task, temporal features separated normal from catch trials at 77.96% ± 7.03%.
The authors say the results show the potential of EEG-based BCIs to distinguish diverse cognitive states elicited by identical stimuli. They plan to improve decoding performance and extend the work to online applications.