CRCNS US-French Research Proposal Plastic Neural Circuits in Olfactory Learning

Project Summary An ubiquitous feature of the mammalian brain, which is likely a major factor contributing to its formidable computational power, is the preponderance of extensive top-down feedback. Depending on the tasks and demands at hand, this top-down feedback allows a brain area to control and enrich the processing of information in the very brain areas from which it receives the information. Such feedback is not restricted to high-order brain areas. In fact, in the olfactory system the valence of an odor stimulus and its context affect the processing of odor stimuli already in the olfactory bulb, which is only a single synapse away from the sensory space. So far, it is only poorly understood what functionalities the olfactory system may gain through this top-down influence. The impact of the top-down projections depends critically on their connectivity. Strikingly, in the olfactory system the feedback network exhibits extraordinary plasticity, which includes the cortically controlled persistent turnover of the dominant neuronal population of the olfactory bulb through adult neurogenesis and strongly fluctuating spine dynamics. The functional benefits of this structural plasticity still remain unclear. Intellectual Merit The proposed project will combine electrophysiology, optogenetics, imaging, and behavioral experiments with computational modeling • to characterize experimentally the dependence of spine fluctuations and cell survival on feedforward odor input and cortical feedback and to develop in parallel a computational model for the network evolution incorporating the experimental results, • to analyze computationally the resulting network connectivities and the mechanisms determining them and, guided by the computational model, to assess the functional connectivities experimentally in awake animals by driving conditionally tagged cortical cells, • to exploit the model to suggest functionalities that may emerge from the cortically controlled connectivity. Possibilities indicated by preliminary results using a highly simplified model include – context-enhanced background subtraction and stimulus discrimination, – rapid cortical switching between previously learned bulbar circuits. Functionalities predicted by the newly developed model will be investigated in behavioral experiments in which the top-down feedback will be controlled optogenetically. With this effort the PIs aim to contribute to the understanding of what functions top-down feedback can support, how they may be implemented, and what special features adult neurogenesis contributes to these functions in the olfactory system. Broader Impacts Imbalances between bottom-up and top-down pathways have been associated with psychological disorders like schizophrenia, which manifest themselves also in olfactory dysfunction. Insights into the mechanisms at work in the interaction between these two pathways may aid in understanding such psychological disorders. Chemical sensor technology may benefit from a better understanding of how the brain uses active topdown processes to extract the relevant information from the complex activity patterns generated by the chemical sensor array of the nose. The computational model developed in the project will be made publicly available via the ModelDB database, a public repository of neuronal models. A graduate student and a postdoc will be part of an international and interdisciplinary research team. Through extensive lab visits the student will gain interdisciplinary communication skills and int