Adaptive brain interfaces for physically-disabled people

This paper presents first results of an adaptive brain interface suitable for deployment outside controlled laboratory settings. It robustly recognizes three purely mental states from on-line spontaneous EEG signals and has them associated to simple commands. Three commands allow to interact intelligently with a computer-based system through task decomposition. Our approach seeks to develop individual interfaces since no two people are the same either physiologically or psychologically. Thus the interface adapts to its owner as its neural classifier learns user-specific filters.