Skip to main content


Although the idea of using brain-computer interfaces (BCI) for motor neuro-rehabilitation is around for more than a decade, its use in a wider clinical practice is limited and in my opinion, even at the academic level still has not reached expected outcomes. Transferring this paradigm into virtual or mixed reality (VR) may not represent a substantial technical burden, but the crucial element of adequately identifying and monitoring brain processes associated with movement remains open also in VR environments. For example, there is a growing body of evidence supporting the existence of a subject-specific narrow-band, spatially distributed cortical sources associated with real or imagery movement. This differs from BCI concepts where the goal is to find any discriminatory signal. Focus on individuality represents the second major building block of this effort. Obviously, once these aspects are better understood, a wider space for the development of a variety of BCI-VR environments affecting different aspects of human cognition and behavior is open. Advanced machine learning and AI tools may help, but we should be careful when trying to step with these tools too far from the known evidence.
I will discuss lessons we learned from building cognitive training and motor neuro-rehabilitation BCI systems in VR, as well as conceptual approaches we used and validated.