A spinal substrate for modular control of natural behavior.

Natural behavior unfolds as coordinated sequences of body movements. This organization suggests that behavior may be built from discrete motor patterns, yet how such arrangements are implemented by neural circuits remains unknown. Here, we combined kinematic analysis, muscle recordings, genetically identified cell types, and closed-loop optogenetic perturbations to examine the organizational logic of natural gap-crossing jumps in mice. Jumping was characterized by a series of precisely defined phases and their associated modular motor patterns. The core phases, propulsion and flight, exhibited distinct signatures of neural control, including unique bursts of coordinated hindlimb muscle activity, differential tuning strategies for jump distance, and active requirements for spinal neural drive. Mapping activity across lumbar interneuron populations and functionally screening candidate cell types for their ability to drive coordinated movement revealed that a population of dorsal excitatory dILB6 neurons can autonomously evoke coordinated multi-joint hindlimb flexion characteristic of the jumping flight phase, across behavioral contexts. These findings provide a specific cellular substrate for the long-standing concept of spinal modular motor control: a flexible, preconfigured motor template that the mammalian CNS can recruit and modulate to meet the demands of natural behavior.