Efficacy of brain-computer interface training with motor imagery-contingent feedback in improving upper limb function and neuroplasticity among persons with chronic stroke: a double-blinded, parallel-group, randomized controlled trial.

The sample was small and sized as a pilot: 27 were recruited, but 25 were analysed after two contingent-feedback participants dropped out. It was a single rehabilitation hospital, so results may not generalise. Only chronic stroke (at least 6 months after onset) was included, so early stroke recovery is not addressed. The sample size did not allow meaningful subgroup analysis by lesion location or size. MEP/TMS data were available for only five participants and were not statistically analysed, so corticospinal excitability evidence is very limited. One participant's EEG data were excluded for insufficient artifact-free epochs, leaving 24 for functional connectivity analysis. Both groups received the same BCI hardware, electrical stimulation, virtual feedback and conventional therapy, so the effect of contingency alone cannot be separated from those components. Participants had severe wrist extensor weakness (MRC score 2 or less), so findings may not apply to milder or more functional patients.

Funded by the Translational Research Program for Rehabilitation Robots, National Rehabilitation Center, Ministry of Health and Welfare, Republic of Korea.

Do not conclude that BCI training alone caused the wrist strength gain. Both groups received the same BCI hardware, electrical stimulation, virtual hand feedback and 30 minutes of conventional upper-limb therapy, and only wrist extensor strength differed significantly between groups; active wrist extension range and Fugl-Meyer outcomes did not show a clear contingent-feedback advantage.