Felipe A. Cini (Luke Remage-Healy's Lab)
Neural population dynamics in vestibular nuclei during locomotion in primates
Ruihan Wei (Kathleen Cullen's Lab)
Department of Biomedical Engineering at Johns Hopkins University, U.S.A
Our study combines advanced high-density neuronal recording, electromyographic measurements, behavioral analyses, and computational approaches to investigate the vestibular nuclei in macaques during locomotion. It reveals how neuron population dynamics stabilize posture through multi-sensorimotor integration, emphasizing the critical role of neuron-specific modulation in maintaining balance and coordinating voluntary movement.
The vestibular system detects the motion of the head in space and transmits this self-motion information to the brain in order to generate stabilizing reflexes. In particular, vestibulospinal reflex pathways play a critical role in maintaining head and body posture by stabilizing the head and body relative to space during locomotion. These pathways receive input from vestibular-only (VO) neurons, a population within the vestibular nuclei which are insensitive to eye movements. To investigate the signals encoded by vestibular pathways in nonhuman primates during locomotion, we recorded from the vestibular nuclei of both normal and bilateral vestibular loss monkeys using 128-channel electrodes. We tracked head and trunk positions with a 6D motion sensor and marker-based systems, and captured limb motion using high-speed, markerless cameras. Our study began by assessing responses to passive movements to characterize vestibular responses. We then examined head stabilization and neuron sensitivity to multisensory inputs during 1) treadmill locomotion with fixed and free head positions at various speeds, and 2) during free overground movement. We also analyzed the population dynamics of VO neurons across these locomotion settings. In normal animals, VO neurons differentially encoded passive vestibular stimulation versus head movements during locomotion. A significant fraction of VO neurons modulated with gait cycle, and our findings reveal that at the population level, VO neurons exhibit a low-tangling geometry, indicative of robustness to noise. Furthermore, these neurons exhibit context-specific geometry that coherently changes across different treadmill speeds and demonstrating distinct differences between treadmill and overground walking. In contrast, in BVL animals, VO neurons did not exhibit phasic modulation with gait, and the resulting population activity structure demonstrated significantly more tangling. Finally, by using innovative dynamical systems analysis tools to distinguish the contributions of multisensory and motor signals, we find that VO neurons increasingly incorporate non-vestibular information in the absence of vestibular input. This research underscores the critical role of neuron-specific modulation at the level of the vestibular nuclei in coordinating voluntary movement and maintaining posture during locomotion. Furthermore, it sheds light on the brain’s strategies for balance control and gait adjustment, enhancing our understanding of how posture is maintained and adjusted during daily activities.