Neural computations in the AOTu, a multi-input visual brain area in fruit flies.

• Applicant: Dr. Inês Ribeiro, Ph.D.
• Subject Area: Cognitive, Systems and Behavioural Neurobiology
• Term: since 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 533863827

Summary: Motion, discrete objects, and other visual features are extracted in early processing centers from 2-dimensional arrays of light intensity fluctuating over time, and relayed to retinorecipient regions the central brain. Most retinorecipient regions in insects lack spatial organization. The exception is the anterior optic tubercle (AOTu), the largest retinorecipient region in Drosophila. The AOTu lateral unit feeds into the anterior visual pathway and the central complex, where relative landmark position is encoded to guide navigation. The central unit of the AOTu receives inputs from different neuron types, the LC10-group neurons and LoPl-LC10. My work revealed that LC10a neurons are essential for female tracking during courtship behavior. In addition, LC10d neurons mediate object avoidance whereas LC10bc neurons are insensitive to objects. The AOTu thus subserves diverse behaviors. The AOTu output neurons stemming from the central unit lose spatial organization, suggesting that they perform de novo computations that require topographic inputs, including object speed and future position. Hypothesis: Based on my work, as well as connectomics data of the fruit fly brain, I hypothesize that the neural circuits stemming from the central unit of the AOTu perform de novo computations based on spatially organized terminals of different input neuron types. These de novo computations transform sensory-correlated information into motor commands and are gated by internal states, that alter neural circuit configurations, to orchestrate diverse behaviors. Key goal: To characterize the neural circuits in the central unit of the AOTu, unravel the computations underlying processing of the diverse visual features reaching the central unit and implementation of different circuit configurations, using social interactions as a naturalistic behavior, for which the fly brain evolved to resolve. The experimental approach will draw on combinations of genetic access to single neuron types to express of transgenes enhancing, disrupting, or sensing neuronal function, robust behavioral paradigms with unbiased analysis, characterization of tuning properties of AOTu output neurons, and identification of neuromodulatory pathways implementing different circuit configurations. Specific Aims are designed to (1) define the neural circuits involved in tracking and avoidance of visual objects, (2) characterize the sensorimotor transformations performed by such circuits, (3) investigate signaling pathways modulating circuit configurations. Impact: Insects solve seemingly complex computational problems, posed by interactions with their environment, with a low number of neurons. An understanding of neural circuits stemming from AOTu central unit will resolve outstanding questions in insect vision, namely how visual information is processed in the central brain to guide behavior towards objects.

• DFG Programme: Research Grants