My long term goal is to elucidate the neural mechanism underlying reaching behavior and apply this scientific knowledge to a neural prosthesis to restore reaching for amputees and paralyzed patients. The posterior parietal cortex (PPC) and dorsal premotor cortex (PMd) are anatomical nodes located in the parieto-frontal network implicated in visually guided reaching. Recently, it was demonstrated that an abstract reaching plan is represented in these areas and that the reaching goal can be read out from both areas before the movement starts. However, it is still unclear what the respective roles of PPC and PMd are in visually guided reaching and how they are functionally interconnected, answers to which could help a prosthesis designer to determine from where and how to decode reaching intentions. To this end, I aim to investigate these areas with the following specific hypothesis: for visually guided reaching, the medial intraparietal area (MIP) in PPC computes a default reach plan, i.e., reaching for a salient object, upon arrival of the visual stimulus information, and this plan is passed to PMd which selects an action between the default reach plan formed in MIP and a non-default plan formed in the frontal area using imposed cognitive rules, e.g., 'green means go and red means stop'. Once PMd resolves the action selection and forms an actual reach plan, this plan is fed back to MIP which then reflects the actual impending reach plan to serve eye-hand coordination and online control of the reaching movement. My hypothesis is based on the following observations. First, MIP neurons represent the location of an eccentric visual stimulus for a brief period upon stimulus onset. Second, inactivation of PMd induces selective deficits in a task requiring action selection based on cognitive rules. Third, MIP neurons represent the location of the upcoming reach target instead of the location of the salient stimulus as time approaches the movement onset. One clear experimental prediction from my hypothesis is that a default reach plan (bottom-up information flow) will be detected in MIP first and a non-default reach plan (top-down information flow) will be detected in PMd first. I will test this prediction by comparing the time at which each plan arises in MIP and PMd. Another prediction is that lesion of MIP will disturb eye-hand coordination and online control of reaching movements. During the mentored phase, I will test the second prediction using a reversible inactivation of MIP. During the independent investigator phase, I will expand the focus to include PMd and test the first prediction using a multi-areal recording under an intact condition. In addition, to further confirm the directional influence between MIP and PMd, the altered neural response in one area by the inactivation of the other will be examined.
The scientific knowledge acquired from this study will not only advance our understanding of the brain but also provide essential information for neural prosthetic applications, e.g., the ideal target brain area to implant the prosthetics and the optimal signal processing scheme to decode the intention of reaching. Considering the importance of reaching in our daily activities, the successful application of the acquired knowledge to a neural prosthesis will bring a significant improvement to the quality of life for the patients who lost reaching abilities.