PROJECT SUMMARY / ABSTRACT Lewy body dementia (LBD) is the second most common cause of dementia and is associated with unrelenting cognitive decline, profound caregiver burden, and higher healthcare costs compared to Alzheimer disease (AD). Despite its prevalence and worse prognosis, patients with LBD suffer from misdiagnosis and delayed diagnosis, which may result in use of medications with increased morbidity and delays in appropriate clinical care. Thus, accurate and early diagnosis of LBD is a serious unmet need. To support earlier and accurate diagnosis, we seek to develop a physiological biomarker of cognitive fluctuations, a characteristic feature of LBD that differentiates it from other dementias. This goal addresses the NIH Alzheimer’s Disease-Related Dementias (ADRD) Summit recommendation to develop biomarkers in LBD. Cognitive fluctuations (CF) are defined as spontaneous periods of impaired attention and reduced arousal that correspond to variability in measures of sustained attention. CF are a core clinical feature of cognitive impairment in LBD and are associated with greater impairment in activities of daily living and worse quality of life. Despite their prevalence, CF are challenging to diagnose clinically, because commonly used formal assessments require administration by an experienced clinician and the presence of a reliable informant. An accurate and widely available biomarker of CF would greatly advance the ability to diagnose CF and thus LBD. EEG differences in LBD are well-documented, but there is currently insufficient evidence to consider specific EEG findings as an indicative biomarker of LBD. A few studies found that both lower dominant frequency and higher dominant frequency variability on EEG correlated with CF. These EEG correlates of CF are thought to reflect dysregulation of cortical synchronization, but perturbations in dominant frequency have not been temporally linked to short-term clinical fluctuations in sustained attention. Demonstrating that EEG correlates of CF in LBD correspond to differences in sustained attention over short timescales will provide evidence of the clinical relevance of EEG as a biomarker. We hypothesize that reduced dominant frequency and higher dominant frequency variability on EEG can accurately identify CF in LBD patients and that dominant frequency on EEG is temporally associated with performance on a test of sustained attention. In Aim 1 we will elucidate the EEG signature underlying CF in LBD by comparing the EEG features of LBD with CF to LBD, AD, Parkinson disease and healthy controls without CF. In Aim 2 we will correlate EEG features of CF in LBD with impairment in sustained attention. We anticipate that this proposal will result in a clinically relevant physiological biomarker of CF in LBD that can then be validated in a large-scale multi-center study. The fact that EEG is non-invasive, relatively inexpensive, and widely available will support its adoption as a biomarker for CF...