Understanding Neural Activity Behind Locomotion in C. Elegans

Locomotion in Caenorhabditis elegans, a species of multicellular roundworms, can be described by switches between forward and reverse states punctuated by turns. These simple movements help the organism find food, avoid predators, and navigate its environment. When a worm swaps direction, this is referred to as a forward to reversal switch.

For the study of neural activity, C. elegans is a great model organism; each worm only has about 302 neurons, and the entire nervous system of the organism has been mapped. Each organism is also nearly identical. “They have a fully stereotyped connectome, where the same neurons are connected in the same way across all organisms,” Megan Morrison of Illinois State University said. Although the system has been widely studied, the network is still incredibly complex, consisting of an intricate network of sensory neurons, interneurons, and motor neurons. “They can produce this nice, wide variety of different behaviors over long and short periods of time,” Morrison said. 

Morrison aims to understand the neural motivations behind each movement choice in C. elegans. In her minisymposium presentation at the 2026 SIAM Conference on the Life Sciences—which is currently taking place in Cleveland, Ohio, concurrently with the 2026 SIAM Annual Meeting—Morrison explains how she studies locomotion in C. elegans and the neural activity that drives decisions surrounding movement. “We want to understand how the C. elegans neural network makes these behavioral decisions,” Morrison said. 

Using data from whole-brain calcium imaging of 15 neurons that are particularly important for forward and reverse movement, Morrison built a mathematical model to analyze neuron activity (see Figure 2). The model analyzes three components: intrinsic dynamics of individual neurons, where each neuron has its own tendency to be active or inactive; gap junctions, which provide bidirectional connections between neurons and help with activity synchronization; and synapses, which create directional connections between neurons. “Our simulation uses a 15-dimensional ordinary differential equation with a forcing function applied to it,” Morrison said. “To validate our model, we looked to see if the forward to reversal switches occurred at the same time in the simulation as they did in the imaging data, and if the probability of the distribution functions looked the same.” 

Once the model simulations were confirmed to match experimental data, they were able to further explore the connections between neurons and the mechanisms behind locomotion. “We know that these switches happen, but we wanted to know why, and what sustained each respective state,” Morrison said. “We looked at what neurons were active during the forward state and sorted them by how much they either promoted the forward state or suppressed the reversal state. We did this for the reversal state as well.”

What they found demonstrated that a switch from forward to reversal could occur through either exciting a cluster of neurons associated with reverse movement or inhibiting a forward cluster. This evidence led to their hypothesis that the reversal to forward switches are driven by forward motion, and that the two movements are in competition with each other, with reversal potentially acting as a braking mechanism for forward motion. 

Morrison also wanted to evaluate the importance of gap junctions and synapses. Using their model, they systematically removed gap junctions or synapses—or both—from the simulation (See Figure 3). “What we found was that with only gap junctions, switches are very synced but they’re not as strong, however with only synapses, switches are very strong but not coordinated,” Morrison said. “This gave us the idea that while both of these connections are important, synapses may be more important for instigating switches, while gap junctions are more important for coordinating a single movement.”

Although Morrison’s model results are computational, these hypotheses could be verified experimentally and serve as a launching point for future projects. Going further, Morrison simplified her model such that the analysis was on overall forward or reverse activity, as opposed to focusing on individual neurons. This low-dimensional simplification provided an easier way to understand the mathematics behind these behavioral decisions. “With this model, we can look at the input and really understand what’s coming to the network,” Morrison said.

While these results focus on the neural mechanisms behind the movements of a small creature, building mathematical models of brain activity could pave the way for predictive neural models and generate hypotheses for future projects in the laboratory. “We build these models to predict switches, but the usefulness of these models is you can understand the mechanisms behind these switches," Morrison said. "You see the switch and can go back into the model to see exactly how the switch is happening, which can hopefully give us a better understanding of how these systems work."

References
[1] Atanas, A.A., Kim, J., Wang, Z., Bueno, E., Becker, M., Kang, D., … Flavell, S.W. (2023). Brain-wide representations of behavior spanning multiple timescales and states in C. elegans. Cell, 186(19), 4003-4256. 
[2] Morrison, M. & Young, L. (2025) A data-driven biology-based network model reproduces C. elegans premotor neural dynamics. PLoS Comput. Biol., 21(12): e1013818.
[3] Zhao, B., Khare, P., Feldman, L., & Dent, J.A. (2003). Reversal frequency in Caenorhabditis elegans represents an integrated response to the state of the animal and its environment. J. Neurosci., 23(12), 5319-5328.