ABSTRACT Schizophrenia is a heterogeneous disorder that likely involves multiple underlying pathological mechanisms, which has plagued attempts to identify rational therapeutic targets. All available antipsychotic drugs (APD) are dopamine receptor antagonists, but clinical response is variable, with a third of patients being partial responders, and a third non-responders. Arguably, those who respond well to APD have primarily dopaminergic abnormalities but it is imperative to also characterize the specific underlying pathologies in those with poor response in order to unravel the heterogeneity of psychosis and effectively develop new treatments. We propose to longitudinally follow treatment response to APD for eight months in medication-naïve first episode psychosis (FEP) subjects using complementary brain imaging techniques. We already have identified provisional markers for several different pathophysiological mechanisms underlying psychosis, including abnormalities in glutamate, brain connectivity, and neurodevelopment that we can track with brain imaging. In addition, we propose to study the changes that occur in the brain in early compared to delayed treatment responders and changes that occur over time in response to treatment. By characterizing treatment trajectories and their relationship to baseline pathophysiologic alterations, we will further complement our mechanistic understanding of the heterogeneity of psychosis. We propose to study 60 well-characterized FEP subjects who are medication naïve and treat them with the most frequently used APD for 32 weeks. We will follow a rigorous longitudinal design to capture treatment response whereby those without an adequate response after 16 weeks of treatment will be switched to another APD for 16 weeks. All patients will be scanned four times: at baseline and after 6, 16, and 32 weeks of treatment. We will use (1) proton MR Spectroscopy (MRS), (2) task and resting state functional MRI and (3) MRI and diffusion weighted imaging (DWI) to measure brain biochemistry, function and structure. Using several imaging modalities has the potential to interrogate different neurobiological aspects of treatment response and will offer greater opportunities for clustering the patterns and combinations of the underlying pathologies in those with poor response. Deconstructing the heterogeneity of psychosis has broad implications for the identification of specific targets for drug development, and to lay the groundwork needed to conduct therapeutic trials on patients characterized by their specific underlying psychopathology.