Based on brain Event-Related Potentials (ERP) obtained under tasks with designed memory demands, the aim is to improve understanding and foster early interventions by 1) developing non-invasive brain activity marker(s) for Alzheimer's disease (AD) early in its course, and 2) predicting with such marker(s) which patients with mild cognitive impairment (MCI) will go on to develop Alzheimer's disease. About half of MCI patients develop AD within a few years. The overall goal is to determine early enough which patients might benefit from interventions and to understand better the brain functions altered in these diseases. We record ERP's while participants perform two simple cognitive tasks. In each of these tasks, ERP components differ under conditions which demand different use of memory and processing resources. The two tasks are (1) a number-letter paradigm that involves comparisons of numbers or of letters within each trial of 4 sequential stimuli (working memory); and (2) a pattern paradigm which involves symmetry judgments of visual patterns, some of which are repeated over trials (implicit memory). The research entails comparison of tasks-related ERPs of MCI patients, AD patients early in the course of AD, and appropriately aged controls. Multivariate statistics are used to measure the ERP components and to develop functions that discriminate among the subject groups and predict AD among the MCI patients. ? ?
Showing the most recent 10 out of 11 publications
