PROJECT SUMMARY Alzheimer’s disease (AD) is the most common form of dementia with significant impact on patients, families and the public health system. At the time of clinical manifestation of dementia, significant irreversible brain damage is already present, rendering the development of cost-effective, biologically informed assessment tools for early detection of the disease an urgent prerequisite for potential therapies to delay or prevent symptoms. While significant advances have been made in characterizing early stages of AD for research, current standard-of-care measures used for clinical diagnosis of prodromal AD lack the ability to identify AD in early stages. The overarching goals of the proposed study are to integrate VR, advanced neuroimaging technologies, and computational techniques to refine, optimize, test and validate a VR-based assessment tool that is cost-effective and ecologically and neurobiologically valid for early detection of AD. Particularly, we will integrate performance on a VR-based multidomain cognitive assessment battery combined with real-time performance data across multiple sources (e.g. kinematic) to identify subtle factors underlying VR task performance that contribute the most to sensitive detection of prodromal AD. Most importantly, using advanced quantitative MR measures of brain microstructure, we will test the association between VR measures and early markers of microstructural changes in brain networks to identify the most biologically valid VR measures of prodromal AD. Our central hypothesis is that VR measures of episodic memory, spatial navigation, and visuospatial skills are most sensitive in detecting prodromal AD, and that these VR measures predict AD- related brain pathology in medial temporal and posterior parietal cortical regions as well as in cingulum and hippocampal white matter fascicles. Our preliminary data using an in-house, novel, multidomain VR assessment battery on a sample of individuals with amnestic mild cognitive impairment (aMCI) and older healthy controls (HC) (N = 23, 17 with aMCI) supported our hypothesis. We propose to refine, optimize, test and validate our suite of VR measures in a larger sample (N = 50 total, 30 aMCI) to accomplish the following Specific Aims: To optimize and test a suite of VR assessments for sensitive detection of prodromal AD (Aim 1) and to test the biological validity of the proposed VR measures in detecting prodromal AD pathology (Aim 2). Successful validation of the proposed VR battery may lead to development of a cost-effective, ecologically and biologically valid assessment tool for early detection of AD. Further, these measures can potentially be used as sensitive behavioral markers for monitoring the response to experimental AD treatments and predicting cognitive and clinical trajectories.