Approximately 1.4 million people in the United States have Lewy body dementia, which includes both dementia with Lewy bodies (DLB) and Parkinson's disease (PD) dementia (PDD). Patients with DLB experience cognitive decline similar to Alzheimer's disease (AD), motor changes seen in PD5, behavioral and psychotic features associated with DSM5 Axis I psychiatric disorders, and constitutional and autonomic features that are often missed as early warning signs. Importantly, recent efforts have described prodromal DLB subtypes according to presenting symptoms (e.g., cognition, motor, sleep, behavior). Both AD and PD have benefited from large longitudinal studies that have advanced research, but few have DLB as a focus. Unfortunately, the diagnosis of DLB and its prodromal states can be difficult, with patients seeing more than 3 physicians and experiencing an 18-month delay to diagnosis. While the DLB consensus criteria have excellent specificity, until recently there has been no standardized way to assess signs and symptoms to improve sensitivity. We have led recent efforts to improve diagnosis with the Lewy Body Composite Risk Score (LBCRS) and the DLB-Module (DLB-MOD) for the NIA-funded Alzheimer Disease Research Center Program. These advances hastened the ability to (a) characterize DLB, (b) discriminate DLB from cognitively normal controls and AD, and (c) discriminate mild cognitive impairment (MCI) due to DLB from MCI due to AD. This application will test the HYPOTHESIS that DLB-MCI has unique neuropsychological, neuroanatomic, and neurophysiologic signatures distinct from MCI due to AD or vascular dementia (VCID). We further posit that combining state-of-the-science plasma biomarkers (e.g., amyloid, tau, synuclein) improves detection and permits antemortem characterization of co-morbid pathology and how pathology may drive transition to DLB and rates of progression. We will leverage existing longitudinal cohorts (n=850) of healthy brain aging, MCI, AD, and VCID that use identical data collection platforms to provide robust comparison groups at no cost to this application. Our SPECIFIC AIMS are: (1) Recruit and deep phenotype a DLB-MCI cohort (n=300) systematically characterizing and validating clinical-cognitive-sleep-behavioral-autonomic features, MRI, DAT, qEEG, plasma and genetic DLB biomarkers; (2) Refine phenotypic presentations of prodromal DLB and their associated biomarker signatures to formally test the recently published clinical criteria and leverage archival data from the PI's existing cohorts using identical data platforms to differentiate DLB-MCI as a distinct clinical entity; and (3) Model clinical-cognitive features, genetic, imaging, qEEG, and plasma biomarkers to predict transition and characterize progression to DLB based on biological variables (e.g., sex, ApoE), comorbid pathologies (e.g., amyloid, tau), and symptom presentations (e.g., fluctuations, hallucinations). DLB is the second most common cause of neurodegenerative de...