Neural correlates of human motor planning and visuomotor transformation

Abstract

Goal-oriented coordinated movement is an everyday feat of computation, the mundaneness of which belies its immense complexity. When we lose function in disease or injury, however, we realize that our understanding of the neural bases of coordinated movements is currently insufficient to use engineered solutions to fully restore function. Neural engineering for the restoration of goal-oriented motor function requires a more complete understanding of how context-dependent information might flow throughout the cortico-basal-ganglia motor circuit, and be informed by spatial cues and existing task-oriented memory. Successful design of bidirectional neurotechnology---with both feedforward model predictions and cortical stimulation feedback---may therefore require not only the state of movement neural correlates in the sensorimotor cortex, but also representations of spatial context and goal coordinates informed in part by hippocampal and subiculuar structures. A combination of deep-brain and surface cortical electrodes may be required to engineer functional neurotechnology capable of restoring spatially-aware volitional movement and sensory feedback. We explore some of the cortical dynamics underlying movement visualization - specifically grasp preshaping, and our ability to adapt to rotated movement perspectives in virtual fixed-viewpoint maze navigation in order to explore these complex interactions of movement, imagined movement goals, and task space. We use electrocorticography and stereo-electroencephalography electrodes clinically implanted in patients to interrogate the underlying neural substrates of human sensorimotor computation. We use the state trajectories of the spectral power sampled by these aggregate field potential-coupling electrodes to work towards a more thorough understanding of the neural dynamics underlying movement control, expand our understanding of the complex visual-spatial-motor interactions of the human body, and move towards a more complete implementation of complex, spatially aware neural technology.