Sophisticated Feedback Processing in Motor Control
MetadataShow full item record
Recent theories of motor control emphasize the role of sensory feedback in driving motor output and how feedback could be flexibly modulated to adapt to the behavioural requirements of a task. Here, we tested the merits of this theory by studying how sensory feedback is modulated by behavioural context and studied the possible neural circuitry underlying such task dependent processing. Mechanical perturbations were applied to the arms of our subjects (in humans or non-human primates) in different behavioural tasks to quantify task-dependent muscle responses. We also quantified cortical responses in non-human primates to explore the neural basis of this flexible feedback processing. Our first experiment (Chapter 2, human study) shows that introducing redundancy in a bimanual postural control task could rapidly change the magnitude of muscle responses to the perturbations in as little as 50ms. In our second experiment (Chapter 3, NHP study), we show that primary motor cortex (M1) responses to mechanical perturbations were rapidly modulated (in as little as 40ms), when the monkey was engaged or not in a postural control task (task-dependent response). Strikingly, the initial perturbation responses remained identical across tasks (task-independent response). We speculated that different sources of feedback, with different time delays and sensitivity to behavioural tasks, might be driving M1 activity; one driving the early task-independent response and the other driving the late task-dependent response. We therefore, studied perturbation responses in a range of sensory and motor cortices across 3 different behavioural contexts (Chapter 4, NHP study). We found sensory feedback to be rapidly transmitted to all these cortical regions within 25ms of limb disturbance. Furthermore, sensory feedback was differentially modulated across these areas, depending on the behavioural task. For instance, posterior parietal area 5 was the first area to show response modulation with task engagement (same as experiment 2). Primary motor cortex, on the other hand, was the first area to show modulation in perturbation response with different spatial targets (i.e., target selection). These results suggest that a highly distributed neural substrate is involved in processing sensory feedback and each area plays a unique role in context-dependent modulation of feedback responses.