ABSTRACT (PROJECT SUMMARY) Despite decades of research, schizophrenia remains the most devastating of all psychiatric disorders. Schizophrenia is characterized by significant deficits in neural and cognitive processing, shortened lifespans, and tremendous social, personal, and economic burden—for patients, their families, and society. The development of cognitive markers that can track clinical and neural progression is a promising avenue for monitoring illness burden and effective treatment response. Yet, despite enthusiasm for cognitive markers of progression, clinically useful markers have proven difficult to identify. One reason for this may be the lack of fidelity between potential cognitive markers and neuropathology. Cognitive markers that are causally linked with brain pathology are needed to overcome this obstacle. We propose that relational memory – a form of memory dependent on the hippocampus – is ideally suited as a marker of neuropathology in schizophrenia. Hippocampal atrophy is one of the most significant brain abnormalities in chronic schizophrenia but is less pronounced in the prodromal and early stage of psychosis. Progressive microstructural changes, resulting from an excitation/inhibition imbalance advancing from the hippocampal head to encompass the hippocampal body and tail has been identified as a plausible mechanism for the hippocampal atrophy observed over illness stages in schizophrenia. We propose to use a novel paradigm developed and validated in hippocampal amnesia patients to quantify performance across three types of relational memory – associative, temporal, and spatial – in schizophrenia. As distinct relational memory types have been shown to segregate along the hippocampal longitudinal axis, we propose that impairment within relational memory types will map to dysfunction in distinct hippocampal divisions of the head (associative memory), body (temporal memory), and tail (spatial memory). Further, we will test the relationship between relational memory impairments and hippocampal pathology over illness stages. To test our hypotheses, we will examine hippocampal volume and microstructure in patients ranging from early to chronic illness stages. The proposed combination of a multi- type, hippocampal-dependent relational memory assessment with multi-dimensional hippocampal neuroimaging will allow us to characterize the relationship between relational memory function and hippocampal structural abnormalities. The establishment of a cognitive marker sensitive to hippocampal structure will significantly improve our ability to track, intervene, and prevent the progression from early psychosis to schizophrenia.