Stroke rehabilitation relies on repetitive, task-oriented training to promote neuroplasticity and motor recovery. However, maintaining patient engagement is challenging, as prolonged therapy often leads to mental fatigue, frustration, and boredom, limiting rehabilitation efficacy. Virtual reality (VR) has emerged as a promising tool in stroke rehabilitation, offering immersive and interactive environments that enhance motivation and engagement. Despite its potential, current methods for assessing engagement primarily rely on subjective self-reports, which lack real-time accuracy and fail to capture cognitive changes during therapy. Electroencephalography (EEG)-based brain-computer interfaces (BCIs) provide an objective alternative by monitoring neural activity to assess cognitive states dynamically, allowing rehabilitation tasks to be adjusted in real time to sustain engagement.
This presentation explores the development of a passive BCI using an 8-electrode EEG system to assess cognitive engagement during a 2D computer game. By analysing neural correlates of boredom, engagement, and mental fatigue, the research aims to improve real-time mental state monitoring for adaptive rehabilitation. Participants played a modified Tetris game under varying difficulty conditions while EEG data and subjective workload measures (NASA TLX) were collected. The system was successfully able to monitor different mental states, with qualitative findings aligning with trends observed in existing literature. These findings highlight the potential of EEG-based BCIs in VR rehabilitation programmes by enabling personalised therapy strategies tailored to patients' mental states. This approach holds promise for improving patient outcomes during stroke rehabilitation.