Neural Correlates of Sensorimotor Control in Human Cortex: State Estimates and Reference Frames

Abstract

Interacting with our environment involves multiple sensory-motor circuits throughout the human brain. How do these circuits transform sensory inputs into discernable motor actions? Our understanding of this question is critical to behavioral neuroscience and implementation of brain-machine interfaces (BMIs). In this thesis, we present experiments that explore the contributions of human cerebral cortex (parietal, premotor, and primary somatosensory cortices) to sensory-motor transformations. First, we provide evidence in support of primary somatosensory cortex (S1) encoding cognitive motor signals. Next, we describe a series of experiments that explore contributions of posterior parietal cortex (PPC) to the internal state estimate. Neural correlates for the state estimate are found in PPC; furthermore, it is found to be encoded with respect to gaze position. Finally, we investigate reference frame encoding in regions throughout human cortex (AIP, SMG, PMv, and S1) during an imagined reaching task. We find the greatest heterogeneity among brain regions during movement planning, which collapses to a largely single reference frame representation (hand-centered) during execution of the imagined reach. However, this result is dependent upon brain region. These findings yield new perspectives and evidence on the organization of sensory-motor transformations and the location the human brain’s internal estimate of the body’s state.