The long-term objective of this research is to evaluate the use of the electrical activity recorded on the surface of the brain (i.e. electrocorticograms or ECoGs) as a way to """"""""decode"""""""" intended movement and use those signals to control the movements of an assistive device in the severely paralyzed. Although brain surface recordings can convey some information about an intended movement, the naturally occurring signals may not convey as much detailed information as is needed for robust command of more complex assistive devices such as a robotic arm or a Functional Electrical Stimulation (FES) system that activates the muscles in a person's own paralyzed arm. However, evidence suggests that paralyzed individuals can learn to produce more reliable and complex command signals with regular practice.
Specific Aim 1 is to quantify how well individuals can learn to produce better command signals by increasing the spatial, temporal and/or frequency band resolution of the recorded signal with regular practice. Because we cannot invasively record brain signals from paralyzed individuals until there is better evidence to suggest the benefits will outweigh the risks, we will first study short-term learning effects in able-bodied patients who are having brain surface electrodes implanted for 1-2 weeks for other medical reasons. These patients will practice using their brain activity to control the movements of a virtual arm and hand in a 3D brain controlled Video game. The second complimentary Aim is to quantify similar improvements in brain signals with long-term retraining in a paralyzed animal model that is operantly conditioned to produce increasingly useful brain signals.
The third Aim i s to assess in the animal model if there is any effect of the time post-injury on the ability to learn to produce better control signals from the brain. ? ?
