Handling Kinematic Features in an Action Observation Task to Optimize a Brain Computer Interface-Based Rehabilitation Training

Brain-Computer Interface (BCI) technology promotes neuroplasticity mechanisms which favor the functional motor recovery in stroke survivors. Patients' residual motor abilities determine the intention, which, once detected by the BCI is fed back via an effector. The majority of studies aimed at optimizing the feedback branch, but not enough attention has been posed to supporting patient in the movement intention that should be detected by the BCI system. The inclusion of a visual motor priming (observed action before a task) in a BCI could promote the retrieval of movements from the patient's own impaired motor repertoire. None of the motor priming proposed until so far have been tailored to the patients' residual motor ability, although it is well known that the human brain recognizes movements closer from a kinematic perspective to its own repertoire more easily. The aim of this study was to investigate how individual motor style in an action observation task would affect the observer's cortical excitability. EEG signals were recorded from 10 individuals during an action observation task where different levels of motor distance between the observer and the agent were modulated. EEG-based group spectral activations shown an involvement of bilateral parietal areas in beta band in case of more unpredictable kinematics. The results would open the way for the design of a kinematic-based visual motor priming to be embedded in a BCI system for post-stroke rehabilitation.