Electroencephalogram analysis of finger proprioception in healthy and stroke subjects

Abstract

The main goal of this project is to increase our fundamental understanding of proprioception, by analyzing electroencephalogram (EEG) signals measured during the execution of a robotic proprioceptive task, the Criss Cross task. The task involves the FINGER robot crossing the index and middle fingers of the participants, who then have to press a button when they perceive the fingers to be overlapping, without visual cues. We are interested in using this assessment to study neural processing related to proprioception, as proprioception plays a key role in evaluating our motor performance and may be predictive of post-stroke individuals responsiveness to motor therapy. This study, conducted with 9 healthy and 9 stroke subjects performing the assessment with their right hand, has enabled the identification of EEG features related to proprioception in the P3 and Cz channels. The analysis of healthy subjects revealed characteristics in the event- related potential and event-related desynchronization of EEG data associated with propriocep- tion when participants were actively engaged in making task-related decision based on pro- prioception (Criss Cross Pressing Button task) compared to when they were not (Criss Cross Non-Pressing Button task), indicating that active engagement of proprioception is necessary for the task. Furthermore, greater proprioceptive effects were observed when the robot moved the fingers at a higher speed. The comparison of healthy subjects to stroke subjects before and after proprioceptive training also revealed differences in brain response localization, with greater focalized response in healthy subjects. In particular, the P3 channel displayed greater amplitude in healthy subjects compared to stroke subjects activity measured before training. Slight increases in response amplitude were observed in the stroke population after training, suggesting possible improvement in proprioception among stroke patients after rehabilitation. The results obtained in this project are promising and show a clear line of research for the relationship between proprioception and motor recovery in stroke patients. The next task to reinforce these results would be to perform the study with a larger sample size, which means a larger number of patients submitted to the experiment