A neural geometry for forelimb proprioception in the cervical spinal cord

Area of Science:

• Neuroscience
• Motor Control
• Spinal Cord Physiology

Background:

• Somatosensory feedback is crucial for coordinated limb movement.
• The neural code and computational mechanisms underlying this feedback remain largely unknown.
• Understanding spinal cord processing of sensory information is key to explaining motor control.

Purpose of the Study:

• To investigate how the cervical spinal cord represents the forelimb's kinematic state.
• To identify the specific sensory inputs (muscle, tendon, cutaneous) involved in this representation.
• To determine the computational role of spinal networks in real-time motor control.

Main Methods:

• Recording neural activity from cervical spinal cord neurons in mice.
• Classifying neuron response properties related to limb movement (speed, position, direction).
• Perturbing muscle, tendon, and cutaneous sensory afferents and observing effects on neural representations and movement accuracy.

Main Results:

• Identified two classes of movement-responsive spinal neurons encoding limb kinematics.
• Demonstrated that population activity forms a low-dimensional manifold representing limb position and velocity.
• Showed that ablating muscle and tendon afferents, but not cutaneous, disrupts this neural manifold and causes movement errors.

Conclusions:

• Spinal cord networks, close to sensory origins, perform complex computations for forelimb movement representation.
• Muscle and tendon afferents are essential for constructing the spinal neural code for limb kinematics.
• This spinal processing is critical for precise, real-time motor control and coordinated movement.