The Cybathlon BCI race: Successful longitudinal mutual learning with two tetraplegic users

This research explores how people with severe paralysis can learn to use brain-computer interfaces (BCIs) to control devices, focusing on a method called "mutual learning." Unlike other approaches that heavily emphasize the machine learning aspect, this study prioritized equally improving the user's skills, the machine's ability to understand brain signals, and the design of the application itself. Two participants with spinal cord injuries were trained using this method to control avatars in a virtual race, demonstrating that this comprehensive approach can be very effective in real-world scenarios.

• 1
  The study demonstrated strong and continuous learning effects at all levels (machine, subject, application) for both end-users over several months.
• 2
  The training approach facilitated the emergence of sensorimotor rhythm (SMR) modulations that participants could sustain even under adverse conditions.
• 3
  Refinement of the application's control paradigm significantly benefited subject learning, suggesting a crucial role for application design in facilitating user skill acquisition.