Eye-gaze independent brain machine interface using amplitude modulation of steady-state visual evoked potential with eyes-closed associated with mental concentration

Brain machine interfaces (BMIs) transform modulation of electroencephalogram (EEG) elicited by cognitive and mental events users voluntarily perform into words and commands in accordance with their intents to communicate with somebody else or machines the users want to control. One of the leading paradigms in BMIs includes a method which utilizes the modulation of a steady state visual evoked potentials (SSVEPs) in EEG with eye-gazing at flickering target users chose. We aim at developing a novel 2-class of BMI which is not only independent of eye-gaze but also available in the eyes-closed state using the modulation of SSVEP associated with mental tasks for patients with visual impairment. The present study found from ten healthy subjects aged 21-24 years old that the amplitude modulation of SSVEP was elicited by mental concentration the flicker stimuli, image recall and mental calculation, respectively. Whether EEG of each run had been recorded during performing a mental task or not was classified on the basis of the amplitude of SSVEP each task, as a pilot study for the development of the proposed BCI on a trial basis and with offline classification. The classification performance depended on the mental tasks, but the classification accuracies for the three mental tasks exceeded 75 % which satisfied the minimum requirement for communication. Stabilization and augmentation of the effect of the mental tasks on the SSVEP may increase the reliability of the proposed paradigm.