Project Summary Perceiving multisensory information and responding with appropriate, real-time behaviors is critical for normal communication and interaction with the environment. Past studies have investigated general brain regions as well as specific cells that fire in response to more than one type of sensory cue, yet have not pinpointed their presynaptic unimodal partners (inputs), meaning that the studied cells may have not been the direct points of multisensory convergence. Additionally, these neurons themselves did not then drive responsive and innate motor behavior. This has left gaps in understanding (1) how multisensory neurons acquire their multi-modal feature detection properties directly from unimodal inputs at the cellular and circuit levels, (2) how they integrate multimodal signals over time, and (3) how they then transform those signals into dynamic motor responses, all during ethologically relevant and innate interactions. Neural circuits that govern Drosophila melanogaster courtship serve as a well-developed model for sensory processing and real-time behavioral responses: during fly courtship, males sing to females, and females perceive and respond to song, which in turn alters males’ own courting behavior. This forms a complex “conversation” that emulates properties of many animals’ social interactions. Within the courtship circuit, I have discovered two direct unisensory convergence points onto multicellular cells and circuits, which have themselves been shown to be necessary and sufficient to drive and modulate robust, measurable, and innate behavior in Drosophila females, providing an unprecedented opportunity to address the gaps described above. The convergence points were identified via analysis of novel whole-brain and half-brain electron microscopy datasets at synaptic resolution. This proposal will directly elucidate principles of multisensory integration at the cellular, circuit, and behavioral levels. Through high resolution behavioral tracking assays, Aim 1 will determine how, by integrating natural combinations of audiovisual information, two multisensory neurons drive an ethologically relevant behavior in Drosophila females. Aim 2 will determine, using calcium imaging in behaving flies in a courtship virtual reality, how those neurons integrate auditory and visual signals from three of their unisensory inputs. Taken together, this study will significantly expand understanding of how brain cells and circuits process multisensory signals and transform them into dynamic motor responses, contributing to a foundation for long-term understanding of normal and disordered sensorimotor function. Additionally, this scientific proposal, along with the outstanding training environment in the Murthy lab and at the joint MD/PhD program at Princeton and Rutgers, will provide exceptional foundational training in preparation for my career as an independent physician scientist with my own laboratory.