Computational roles of inhibition in human action control

点击翻译按钮获取中文摘要

ABSTRACT Classic neurocognitive models of cortico-basal ganglia circuits imply excitatory signals drive, and inhibitory signals suppress, behavioral output. However, inhibition can support other computations. Important research on animal motor systems and human sensory systems has shown inhibition shapes the gain and tuning properties of neural populations. Compelling evidence suggests these same two computations support action preparation in the human motor system. The overall objectives of this application are to establish roles for gain modulation and tuning within the human motor system during action preparation and to evaluate to what extent these computations relate to the neurochemical capacity for inhibition. The long-term goals of this work are to characterize physiological and neurochemical contributions to the neural computations underlying human action control and to understand how disease-related disruption of these mechanisms contributes to behavioral impairments. Accordingly, the proposed experiments will test the central hypothesis that gain and tuning within the human corticospinal pathway change dynamically during the preparation of actions and relate to inhibitory neurotransmitter availability in motor cortex. The first specific aim of this application is to test the hypothesis that gain within the human motor system increases during action preparation to facilitate the execution of a selected action. The second specific aim is to test the hypothesis that spatial tuning of motor representations sharpens during action preparation. Sharper tuning of motor representations is expected to promote more fine-grained control by facilitating the selective recruitment of muscles involved in a prepared action. Whereas Aim 1 examines a computational mechanism for facilitating the execution of a selected action, Aim 2 examines a computational mechanism for selecting actions from a pool of neighboring and overlapping representations. We will use non-invasive brain stimulation to examine the patterns of excitability within a given muscle (Aim 1) and across a group of muscles (Aim 2) to detect changes in gain and tuning, respectively, during behavioral task performance. The third specific aim of this application is to explore relationships between gain, tuning, and local availability of the inhibitory neurotransmitter gamma-amino butyric acid (GABA) in the cortex. Individual differences will be examined to test whether the availability of local GABA correlates with the magnitude of gain and tuning changes. We hypothesize individuals with more GABA in motor cortex will exhibit greater increases in gain and sharper tuning during action preparation. The proposed work is innovative because the roles of gain and tuning within the human motor system are crucially understudied and their links to neurochemical content are completely unexplored. This research is significant because the results could change our interpretation of the roles of inhibition in behavioral control. Abnormalities in inhibitory mechanisms are associated with symptoms of Parkinson’s disease, stroke, and dystonia, and knowledge gained from the proposed experiments can help identify new targets for therapeutic interventions.