Brain-wide hierarchical and multiplexed encoding of behaviors in C. elegans

A fundamental problem in neuroscience is how neuronal activity in the brain generates organized and stable behaviors at multiple timescales. Recent research in many model organisms, e.g. in monkeys, led to the observations that the neuronal encoding of individual actions is achieved by large neuronal populations but often lies on low-dimensional manifolds. However, it is unclear if such manifolds extend beyond the single-action level into coordinating entire action sequences or fine-tuned movements. In order to address these questions, we established a whole brain single-cell resolution recording and analysis pipeline in freely crawling C. elegans nematodes, which display natural long time-scale behavior sequences and switches between them. Extending previous findings, we observed a structured manifold resulting from brain-wide coordinated neuronal activity. Here, we developed a principled method for investigating it. Using a global-residual signal decomposition approach, we find that the manifold corresponds to globally shared population activity correlating with longer time-scale action switching. Our analysis approach further revealed hierarchically nested residual activity in many individual neurons that relates to fine-tuned movements. In addition, we show that a fraction of neuron classes that participate in the global manifold encode re-afferent perception. Thus, the manifold is a composition of both globally shared internal action commands and re-afferent sensory information. In conclusion, we propose that neuronal manifolds provide shared context about the current behavioral state to orchestrate other faster time-scale movement patterns and to process movement-related sensory information. We propose such hierarchical multiplexing as an organizational principle relevant for larger animals as well.