A key feature of reactive behaviors is the ability to spatially localize a salient stimulus and act accordingly. Such sensory-motor transformations must be particularly fast and well tuned in escape behaviors, in which both the speed and accuracy of the evasive response determine whether an animal successfully avoids predation [1]. We studied the escape behavior of the fruit fly, Drosophila, and found that flies can use visual information to plan a jump directly away from a looming threat. This is surprising, given the architecture of the pathway thought to mediate escape [2, 3]. Using high-speed videography, we found that approximately 200 ms before takeoff, flies begin a series of postural adjustments that determine the direction of their escape. These movements position their center of mass so that leg extension will push them away from the expanding visual stimulus. These preflight movements are not the result of a simple feed-forward motor program because their magnitude and direction depend on the flies’ initial postural state. Furthermore, flies plan a takeoff direction even in instances when they choose not to jump. This sophisticated motor program is evidence for a form of rapid, visually mediated motor planning in a genetically accessible model organism.

The body of an animal influences how its nervous system generates behavior1. Accurately modeling the neural control of sensorimotor behavior requires an anatomically detailed biomechanical representation of the body. Here, we introduce a whole-body model of the fruit fly Drosophila melanogaster in a physics simulator. Designed as a general-purpose framework, our model enables the simulation of diverse fly behaviors, including both terrestrial and aerial locomotion. We validate its versatility by replicating realistic walking and flight behaviors. To support these behaviors, we develop new phenomenological models for fluid and adhesion forces. Using data-driven, end-to-end reinforcement learning we train neural network controllers capable of generating naturalistic locomotion along complex trajectories in response to high-level steering commands. Additionally, we show the use of visual sensors and hierarchical motor control, training a high-level controller to reuse a pre-trained low-level flight controller to perform visually guided flight tasks. Our model serves as an open-source platform for studying the neural control of sensorimotor behavior in an embodied context.

Preprint: www.biorxiv.org/content/early/2024/03/14/2024.03.11.584515

Many animals rely on acoustic cues to decide what action to take next. Unraveling the wiring patterns of the auditory neural pathways is prerequisite for understanding such information processing. Here we reconstructed the first step of the auditory neural pathway in the fruit fly brain, from primary to secondary auditory neurons, at the resolution of transmission electron microscopy. By tracing axons of two major subgroups of auditory sensory neurons in fruit flies, low-frequency tuned Johnston's organ (JO)-B neurons and high-frequency tuned JO-A neurons, we observed extensive connections from JO-B neurons to the main second-order neurons in both the song-relay and escape pathways. In contrast, JO-A neurons connected strongly to a neuron in the escape pathway. Our findings suggest that heterogeneous JO neuronal populations could be recruited to modify escape behavior whereas only specific JO neurons contribute to courtship behavior. We also found that all JO neurons have postsynaptic sites at their axons. Presynaptic modulation at the output sites of JO neurons could affect information processing of the auditory neural pathway in flies. This article is protected by copyright. All rights reserved.