Enabling Control in the Minimally Conscious State in a Single Session with a Three Channel BCI

Abstract

The study aimed to detect awareness in a single participant, diagnosed minimally conscious for the past 11 years, using an EEG-based brain-computer interface (BCI), and to determine if real-time feedback enhances our ability to detect awareness in a single session. After 90 trials involving motor imagery (MI) with no feedback hand grasp vs. wiggle toes could be classified with ~82% accuracy with only three EEG channels. In the same session we subsequently provided real-time feedback with two games where the participant was instructed to move a ball and a spaceship, respectively, to reach a target by performing the same MI tasks. ~77% ball and 80% spaceship control was achieved. At the outset of the experiment the participant did not seem attentive or interested however after the feedback runs the participant was noticeably more attentive. Family members in attendance at the experiment commented on the noticeable changes in demeanor of the participant who had provided no overt indication of language comprehension or other cognitive function since diagnosis. The results indicate that real-time feedback should be used in the detection of awareness to inform the completely locked-in of the potential of (BCI) technology as a means of communication (i.e., that it is not just another assessment) and motivate engagement.