Abstract Our goal is to provide novel, clinically feasible, and precise diffusion magnetic resonance imaging (dMRI) tech- nologies for investigation of the in-vivo human brain's cellular microstructure in early psychosis. Psychotic dis- orders are devastating brain diseases that include a range of symptoms such as delusions, hallucinations, and thought disorder. A better understanding of the etiology of psychosis can lead to improved diagnosis and treatment. dMRI is a noninvasive imaging method that has identified unique microstructural abnormalities in psychosis. However, as many pathological processes have been proposed to co-exist in psychosis, there is a need for more specific measures derived from dMRI. Current state-of-the-art dMRI encodes (measures) diffusion along a single direction using a technique called single diffusion encoding. We propose q-space trajectory imaging (QTI), a new dMRI method that dynamically changes the measurement orientation during acquisition to better characterize the true complexity of water molecule motion. The novel QTI sequences that we propose allow measurement of microstructural properties of the in-vivo human brain that are invisible using today's scanning methods. These microstructural properties are mathematically expressed in terms of variability in size (CMD), shape (Cµ), and orientation (Cc), extracted from a model representing a mixture of distinct neuronal tissue microenvironments, such as neurites, cellular domains and extracellular spaces. To achieve our goals we propose the following three aims. In Aim 1, we will develop experimental foundations for novel dMRI. We will investigate QTI methodologies to better characterize important pathological features expected in psychosis. We will extend the QTI framework to enable microstructural models and properties that explicitly account for restricted, non-Gaussian, and time-dependent diffusion. The successful endpoint of this aim will provide measures and models for the extraction of new microstructural properties related to psychosis from QTI. In Aim 2, we will develop novel dMRI sequences and standards. We propose to develop robust and fast QTI pulse sequences and scan protocols, with representation of the acquisition parameters in DICOM. The successful endpoint of this aim will be novel, robust, and fast acquisitions (under 15 minutes), enabling QTI for clinical studies. In Aim 3, we propose to study QTI-based microstructure measures in 24 early psychosis pa- tients and 24 matched controls to disentangle pathologies that co-exist in psychosis, such as neurodegenera- tive and neuroinflammatory processes. We expect that upon successful completion of the proposed project, we will have developed novel dMRI to provide increased sensitivity and specificity for the study of the in-vivo human brain's cellular microstructure.