Fast calibration of a thirteen-command BCI by simulating SSVEPs from trains of transient VEPs - towards time-domain SSVEP BCI paradigms

Abstract

A 13-command Brain-Computer Interface (BCI) based on Steady-State Visual Evoked Potentials (SSVEP) is assessed. The SSVEPs are simulated from VEP sequences recorded by electroencephalography (EEG) on the same subjects. SSVEP features extracted in the time domain are averaged over all channels of the occipital region. Most subjects achieved satisfactory classification rate (50~80% correct command detection). A simulated/offline information transfer rate of 60 bits/min is achieved, averaged across the best eight subjects. Online validation was performed on one new independent subject. The calibration procedure, based on VEP recordings, lasts one minute whatever the number of commands. Online information transfer rate of 58 bits/min is achieved.