Project Summary: Impairments in visual processing and perceptual decision-making are a significant cause of morbidity in neuropsychiatric disorders such as Alzheimer’s disease and schizophrenia. The primary goal of this training proposal is to characterize a potentially novel node in the distributed network underlying visual perceptual decision-making - the midbrain reticular nucleus (MRN). Classical models of perceptual decision-making ascribe decision formation to forebrain sensorimotor regions, with the midbrain existing at the end of a feedforward pathway to relay motor commands. However, there is substantial evidence that the midbrain, especially the superior colliculus (SC), participates in the visual decision process. Additionally, our recent work has shown that a variety of coding, from movements, to action selection, to even abstract cognitive representations such as cue valuation, is simultaneously distributed in multiple regions throughout the brain. We hypothesize that this coding scheme applies to visual decisions as well, with midbrain regions including MRN and SC participating in a network with regions in the cortex and basal ganglia to form the decision. The MRN has traditionally been thought of as a purely motor-related region. Challenging this notion, we recently found visual and action selection signals in MRN during a visual decision-making task. Questions remain however regarding how the diverse coding in MRN is organized topographically, whether visual responses are a result of task learning, the encoding of abstract decision signals in MRN, and how MRN interacts with the broader perceptual decision-making circuitry. This training proposal addresses each of these questions by leveraging the scale of Neuropixels 2.0 probes and a novel visual reverse contingency task I have designed and implemented to distinguish sensory, motor, and decision signals. Aim 1A uses Neuropixels 2.0 recordings across MRN in mice viewing a variety of stimuli to build a topographic map of sensory and motor coding, and Aim 1B uses similar recordings in task-naive and trained mice passively perceiving task stimuli to assess visual plasticity in MRN. Aim 2A uses recordings across MRN during task performance to establish the presence of abstract decision signals, and Aim 2B uses multi-probe recordings throughout the visual decision network to establish inter-area interactions during decision formation. Together these aims will advance our understanding of the distributed regions and the complex computations that give rise to visual decision-making, thereby improving our ability to precisely target key circuits in disorders where this process goes awry. During my tailored training period, I will learn Neuropixels electrophysiology, rodent behavioral task design and implementation, whole-brain histological processing and imaging, and advanced computational analysis techniques under the guidance of experts in a supportive training environment. These skill...