Olfactory learning alters navigation strategies and behavioral variability in C. elegans.

Area of Science:

• Neuroscience
• Computational Biology
• Animal Behavior

Background:

• Flexible brain computation is vital for survival, enabling animals to adapt behavioral responses to sensory input based on past experiences.
• The nematode *C. elegans* exhibits learned odor navigation, modulating its response to butanone based on prior associations with food or starvation.
• Understanding the precise changes in navigation strategy following learning is crucial for deciphering adaptive behavior.

Approach:

• Developed a novel descriptive model combining biased random walk and weathervaning strategies to analyze worm navigation.
• Applied the model to precise experimental measurements of worm trajectories and odor concentration data.
• Inferred strategy weights and analyzed behavioral variability changes post-learning.

Key Points:

• Appetitive training up-regulates the biased random walk strategy, while aversive training down-regulates weathervaning in *C. elegans*.
• The statistical model accurately predicts training conditions (>90%) and outperforms classical chemotaxis metrics.
• Learning reduces behavioral variability in worms compared to naive individuals.

Conclusions:

• The study provides a new quantitative paradigm for flexible navigation algorithms in *C. elegans*.
• Identified learning-dependent alterations in navigation strategies and behavioral variability.
• Investigated neural substrates downstream of the AWC$^\mathrm{ON}$ olfactory neuron involved in learned odor navigation.