Mirror Neuron Network Based Motor Imagery Training to Improve Brain Computer Interface Performance in Spinal Cord Injury Patients

Intend to train subjects in motor imagery techniques using an advanced virtual reality based
platform. Hypothesize that this will enhance the cortical signals necessary for operating a
non-invasive EEG-based BCI. Specifically, designed training paradigms will activate "mirror
neurons" associated with performance of movement, observation of movement, and motor imagery.
Mirror neuron network (MNN) activation induces cortical plasticity, and may therefore enhance
cortical signals generated during BCI use. Plan to assess the effect of training paradigm on
cortical signals generated during motor imagery, using EEG. Also, investigate
neuropsychological factors that play a role in a subject's ability to generate vivid motor
imagery.

Eight subjects with spinal cord injury (SCI) will be recruited during the sub-acute phase
(3-6 months following injury). Four participants will immediately undergo motor imagery
training in an immersive virtual reality environment at NASA. EEG assessments and motor
function testing will be performed at intervals during the 3-4 week motor imagery training
period to assess improvement in cortical signature generation and motor recovery. The other
four participants will undergo the same protocol after a 6 week delay, in order to assess the
effect of training timing after injury in both cortical signature generation and motor
function recovery.

Inclusion Criteria:

- sustained incomplete Spinal Cord Injury (American Spinal Injury Impairment Scale) at a
level of C5-7 within three to six months prior to enrollment

Exclusion Criteria:

- unstable or acute medical or psychiatric illness; pregnancy; epilepsy; diabetes;
autism spectrum disorder, history of severe traumatic brain injury or stroke, and
those taking medications that may interfere with EEG readings, including beta blockers
and sedatives.