Population-level encoding of somatosensation in mouse sensorimotor cortex

Abstract

Somatosensation constructs the bodys dynamic sense of state and allows for dexterous and precise movements. The heterogeneous responses of single neurons in sensorimotor cortex to so-matosensation have led to disparate views of the computational role of this brain area in sensori-motor processing. Here, we use population-level analyses of neural activity recorded during passive limb movements to assess the structure and to summarize the properties of neural encoding in sensorimotor cortex. We used 2-photon imaging to record the activity of thousands of neurons in eight anesthetized mice during passive deflections of each limb. We additionally analyzed neu-ral responses to passive limb movements in eight awake mice, sourced from an open dataset [1]. We employed principal component analysis on the neural activity in each dataset and found that a small fraction of principal components explained a large fraction of variance in the neural re-sponses. Low-dimensional representations of limb movements were well conserved across animals, including the orthogonal representations of ipsilateral and contralateral limbs. This organization of somatosensory information mirrors the well-known structure of neural encoding of motor com-mands in sensorimotor cortex. Furthermore, neural populations dually encoded both changes in joint angles during movements and more abstract information, i.e., the direction of limb movement. Increasing the size of the neural population improved encoding of both types of movement infor-mation and better differentiated representations from movements in opposing directions. Together, these results demonstrate that population-level encoding of somatosensory information in mouse sensorimotor cortex is structured to facilitate sensorimotor integration across the brain.