Outcome measures for recovery

How recovery is quantified determines what counts as success, and rehabilitation BCI is easy to over-claim without careful endpoints. Measures span the levels of the WHO ICF framework: impairment (the Fugl-Meyer Assessment, MRC strength, and physiological markers such as TMS motor-evoked-potential amplitude and corticomuscular coherence), activity/function (the Action Research Arm Test, box-and-block, gait speed), and participation/quality of life. A credible claim of recovery — as opposed to compensation or a device-on effect — requires gains that persist after the device is removed, exceed the minimal clinically important difference, transfer to untrained real-world tasks, and beat a dose-matched sham or active control.

Honesty: a recurring problem is the gap between neural/impairment change and meaningful function — a device can shift an evoked potential or a Fugl-Meyer score without improving what the patient can actually do. Small samples, heterogeneous stroke, lack of blinding, and short follow-up further weaken many reports. Predictive biomarkers (for example the proportional-recovery rule and corticospinal-tract integrity) are used to stratify who can respond and to interpret trials honestly.