Neural activity profiles reveal overlapping, intermingled subpopulations spanning area borders in mouse sensorimotor...

Area of Science:

• Neuroscience
• Systems Neuroscience
• Computational Neuroscience

Background:

• Cortical control of movement involves distributed neural computations across interconnected brain regions.
• Existing anatomical atlases provide a coarse understanding of function, but detailed organization of behaviorally relevant neural activity remains unclear.

Purpose of the Study:

• To investigate the large-scale spatial organization of neural activity across sensorimotor cortex during a reach-to-grasp task.
• To characterize neuronal response properties and identify distinct subpopulations within sensorimotor areas.

Main Methods:

• Cellular-resolution, two-photon calcium imaging of over 39,000 layer 2/3 neurons in five sensorimotor cortical regions of head-fixed mice.
• Mice were trained on a 15-target reach-to-grasp task.
• Characterization of neuronal activity using interpretable metrics and nonlinear dimensionality reduction.

Main Results:

• Neuronal response properties exhibited large-scale spatial structure, with abrupt shifts at anatomical borders between motor and somatosensory areas.
• Motor areas showed sharper tuning and more linear relationships with target location compared to heterogeneous responses in somatosensory areas.
• Nonlinear dimensionality reduction revealed distinct neuronal subpopulations with characteristic activity patterns, spatially intermingled across multiple cortical areas.

Conclusions:

• Sensorimotor cortex displays novel activity structures with distinct, overlapping neuronal subpopulations during reach-to-grasp behavior.
• Understanding these subpopulations and their distributed nature is crucial for deciphering motor control mechanisms.
• The findings challenge simplistic area-specific functional models and highlight a more complex, integrated neural architecture.