Circuit Models for Robust, Adaptive Neural Control

Understanding a nematode's simple circuit could provide a foundation for understanding much more complex behaviors.

February 1, 2019

Air Force Research Laboratory, Arlington, Virginia

This project seeks to reproduce the neural circuits used by the nematode Caenorhabditis elegans for locomotion. Caenorhabditis elegans is a small (~1.2 millimeter) nematode found in rotting fruit in many parts of the world. It feeds on bacteria and is neither parasitic nor pathogenic. Although capable of sexual reproduction, most laboratory strains reproduce primarily as self-fertilizing hermaphrodites, with each adult hermaphrodite producing approximately 300 progeny (Figure 1).

Figure 1. Basic anatomy of an adult hermaphrodite C elegans nematode.

C. elegans is a very simple organism, with only 959 somatic cells in the adult hermaphrodite. Although the total number of cells is small, they are differentiated into the standard array of tissues: 302 neurons, 95 body muscle cells, 32 gut cells, etc. In addition, the position, morphology, and lineage of each cell are reproducible from animal to animal. Because of the small size of the animal, the relatively small number of neurons, and the reproducible nature of the nervous system, it has been possible to provide an almost-complete synaptic connectivity map of the adult hermaphrodite nervous system (Figure 2).

Figure 2. Synaptic connectivity map of C. elegans nervous system.

Utilizing only 113 neurons, this simple circuit drives the 95 body wall muscles to generate surprisingly complex and adaptive locomotion behavior. Recent advances in C. elegans electrophysiological techniques, which have resulted in a surge of new data, have made it possible to build an accurate computational model of C. elegans locomotion.

Taking the perspective that the best way to understand something is to construct it, this interdisciplinary project aims to reproduce the locomotion neural circuitry used by C. elegans to drive a virtual model in a highly detailed 3D C. elegans simulator. The goal of this project, therefore, is to develop an understanding of the basic motifs used by nature in developing complex, adaptive control systems. This goal can be further refined into three specific project objectives:

Develop a biologically accurate computational model of the locomotion circuitry used by C. elegans.
Validate the model by demonstrating that it produces the various locomotion modalities in a physics-based simulation environment.
Verify that the model accurately reproduces the robustness and adaptability seen in the living organism using comparative video analysis.

It is believed that the locomotion circuit used by C. elegans forms the basis for a number of more complex circuits found in higher order organisms. In other words, understanding this simple circuit could provide a foundation for understanding much more complex behaviors.

This work was done by Roger Mailler, The University of Tulsa, for the Air Force Research Laboratory. AFRL-0269