[ Persistent memory impairment is one of the most common complaints after brain injury and is the hallmark of Alzheimer's disease (AD) as well as amnestic Mild Cognitive Impairment (aMCI; a transitional stage between healthy aging and AD). The prevalence of these conditions is increasing, as 40% of the VA population is elderly and age is the greatest risk factor for AD. Head injury also increases the risk of developing AD and the incidence of brain injury in Veterans (and the memory sequelae that accompanies it) has increased in recent years due to the Iraq and Afghanistan wars. Efforts to understand the functional and structural substrates of memory impairment have focused almost entirely on impaired new learning. Impaired remembering of the past (retrograde memory [RM] loss) has received little attention even though this impairment is particularly devastating (e.g., one forgets family relationships, important events, and facts about the world). Although impaired new learning is well studied, RM has received less attention even though Veterans with MCI and AD exhibit severe RM loss. Interestingly, individuals with MCI can exhibit relatively mild impairment in new learning in conjunction with relatively severe RM loss, suggesting that a new tool to detect RM loss could serve as an early estimate of cognitive and neural decline associated with the development of AD. Older Veterans exhibit changes in brain structure and function as the result of aging, and Veterans with aMCI exhibit even more significant changes in the brain. We will study RM in these two groups and identify brain regions (and connections between brain regions) where measures of structure and function are related to performance on RM tests. The two goals of the proposal are 1) to identify which brain regions support RM in cognitively normal older Veterans; and then 2) determine if a RM test could serve as a novel and unique gauge of the cognitive and neural changes associated RM loss in MCI. As time passes after learning, it is thought that the role of the hippocampus and related structures in supporting memory retrieval gradually decreases, whereas the role of the prefrontal cortex and other cortical areas gradually increase (memory consolidation). These observations are typically observed across years and decades in humans but, curiously, across days and weeks in animals. The proposed studies will identify the neuroanatomy of RM for both the short and longer time frames. In addition, we will identify the relevant neuroanatomy using four neuroimaging measures (i.e., 2 focusing on brain regions and 2 focusing on connections between brain regions). For the longer time frame, memory for facts will be tested during functional brain imaging (fMRI). The facts will concern 160 notable news events that occurred 1 to 30 years earlier. For the short time frame, memory will be tested during scanning for 320 fact- like, three-word sentences that were learned 1 hour to 1 month earlier. Measure...