Stochastic motor control of search behaviour in the model organism C. elegans
CEAB and CREAF, Spain
Previous research on C. elegans search behaviour has been largely undertaken in short spatiotemporal scales and mainly focused on stereotyped turning behaviour. However, the structural organization of C. elegans search trajectories is more complex and spans a wide range of scales, so classic experiments are not able to explore generative mechanisms in whole detail. Here we expanded such classic scales up to 90 min and 24.5 mm2 in an effort to capture high-order trajectory dynamics and patterning. Our results show that C. elegans unfolds looping behaviour covering a wide range of sizes and in a time-ordered manner. Based on a microscopic (Langevin) stochastic model we are able to explain part of the observed variability in terms of speed and curvature motor control, and analyze how the movement of C. elegans departs from an elementary stochastic generative process, suggesting that neuromotor control of key movement variables at the microscopic scales modulate a program of (macroscopic) trajectory motifs, such as loops of multiple sizes and variability. We use these results to discuss the theoretical impact of the observed motor control and looping patterns under the perspective of sampling behaviour and search theory.