PROJECT ABSTRACT Functional connectivity (FC) has been found to be altered in a wide range of otherwise indistinguishable disease states. The most common tool to non-invasively study the organization of brain-wide FC networks is functional magnetic resonance imaging (fMRI). fMRI relies on an indirect and indiscriminate measure of activity through the blood oxygen level-dependent (BOLD) contrast mechanism. By capturing fluctuations in the BOLD signal fMRI can detect distant synchronization between brain regions either at rest or during the performance of a specific task. These regions are inferred to be functionally connected and are thought to involve the synchrony of neuronal populations involved in a common function that are wired together through plasticity. Interpretation of FC networks derived from BOLD fMRI studies is currently limited by 1) the dependence of fMRI BOLD signals on hemodynamic changes as a proxy for neural activity and 2) a limited understanding of the mechanistic basis for FC in the context of behaviorally relevant longitudinal reorganizations. My long-term goal is to better understand the relationship between BOLD and neural activity in behaviorally relevant multi-regional circuits to advance brain network analysis. Once we know how both vascular and neural changes influence FC, the basis for network dysfunctions can be exploited with fMRI providing more robust fMRI-based disease detection. In this proposed project, I will use EMX1-Cre mice expressing a novel JEDI-1P voltage fluorescent protein in excitatory neurons across dorsal cortex. This voltage sensor has a large spectral band, allowing us to record both slow, subthreshold voltage activity, and fast gamma band activity. Wide-field optical imaging will be combined with fMRI to link neuronal changes in FC to hemodynamic changes measured with fMRI. In Aim 1 I will image these JEDI-1P at resting state to establish the correspondence between membrane voltage dynamics and fMRI for resting state functional connectivity networks. In Aim 2 I will train these mice to perform a sensorimotor task. I will image them during task training to assess changes in FC due to learning. I hypothesize that the correlation between fMRI BOLD and neural activity will be regionally specific for both resting state and task training, the latter will result in a measurable strengthening of FC between task-relevant areas. Completion of these aims will determine how functional networks observed with BOLD relate to neural activity and will provide insights into how FC reflects behaviorally relevant changes in the learning of a sensorimotor task. This results in a sharper understanding of the properties of neural network activity, its dependencies, and how to harness it in future fMRI studies.