Publication: Remapping and transformation of neural codes in the Drosophila navigation system
Open/View Files
Date
Authors
Published Version
Published Version
Journal Title
Journal ISSN
Volume Title
Publisher
Citation
Abstract
The central complex is a Drosophila brain region important for navigation behavior. One central complex structure, the ellipsoid body (EB), contains compass neurons (E-PG neurons) that encode the heading direction of the fly. In the first study, we use in vivo two-photon calcium imaging during tethered walking behavior to examine how the internal representation of heading is guided by sensory input. We find persistent remapping of compass neuron coding following changes in visual experience, suggesting a model of associative plasticity of sensory inputs onto the compass network. In the second study, we identify two cell types downstream of compass neurons, called PFNd and PFNv, which conjunctively encode the heading direction and body-centric translational velocity of the fly. Together, PFNd and PFNv neuron populations comprise a Cartesian system of bodycentric velocity. We identify the origins of conjunctive coding in PFNd neurons as a combination of heading inputs inherited from the compass system and inhibitory velocity-tuned scalar inputs from premotor regions. We also show that inhibition of PFNd neurons can disrupt path integration behavior in freely-walking flies. PFNd and PFNv neurons project to another central complex structure called the fan-shaped body (FB), where they synapse onto a cell type called hΔB. Using connectomics analysis and computational modeling, we find that the pattern of PFN projections onto hΔB neurons pools together heading and translation direction combinations corresponding to the same movement in world-centric space. This network motif effectively rotates the brain’s representation of body-centric translational velocity according to the current heading direction. Consistent with our predictions, we observe that hΔB neurons form a representation of velocity in world-centric coordinates. Our work demonstrates the neural circuit mechanisms of coordinate transformation and vector computation in the fly.