Anterograde amnesia is a specific memory impairment that can result from damage to the medial temporal (MT) lobes -- including hippocampus -- as well as from damage to the basal forebrain. Clinically, both MT and basal forebrain amnesics appear similar, which has led many researchers to posit a unified """"""""organic amnesia"""""""" syndrome following either etiology. The proposed study will more closely examine the underlying brain mechanisms of this memory impairment and offer an alternate model resulting in the dissociation of these two etiologies. Based on evidence from both animal studies and computational modeling of hippocampal-basal forebrain interaction, we propose to differentiate distinct patterns of learning and memory impairments between persons with MT and basal forebrain amnesia. Specifically, our developed computational model predicts that simple associative learning should generally be spared following MT damage but slowed (although not abolished) following basal forebrain damage. Conversely, our model predicts that more complex forms of associative learning, which require hippocampal-region mediation in animals, may be abolished in MT amnesia but remain intact in basal forebrain amnesia. Thus, the model predicts that specific areas of memory are differentially affected in different forms of amnesia. Initial pilot work to test our model has provided further evidence for this dissociation, whereby a simple form of associative memory (classical delay eyeblink conditioning), which has previously been shown to be spared in persons with MT amnesia, was demonstrated to be severely disrupted in persons with ACoA amnesia. The proposed study seeks to extend these findings and provide further evidence for this dissociation in human amnesia by using a battery of iteratively acquired associative learning tasks to extend the pilot findings to a variety of other learning paradigms, comparing performance among individuals with MT amnesia, individuals with basal forebrain amnesia subsequent to ACoA aneurysm, and matched controls. If our proposed dissociation holds, the findings will not only expand our understanding of hippocampus and basal forebrain interactions, but may also provide a foundation for the development of useful therapeutic interventions. Specifically, the identification of distinct patterns of memory impairments between these two groups may suggest future approaches for optimizing patient rehabilitation by tailoring therapy based on an amnesic individual's specific pattern of impairment.
