The nature, extent, and cause of semantic processing deficits in Alzheimer's disease (AD) is an important practical and theoretical issue that is still unresolved. So-called """"""""semantic"""""""" priming tasks (typically involving lexical decision or work pronunciation) have been widely utilized in recent years in a attempt to assess the integrity of semantic knowledge. However, the entire enterprise of using essentially non-semantic tasks to tap semantic knowledge is problematic; in a recent study of young normals, it was found that automatic """"""""semantic"""""""" priming effects were really lexical, not semantic. The proposed study will evaluate whether AD patients' pattern of performance on a traditional """"""""semantic"""""""" priming task (pronunciation of prime and target words presented serially on a computer screen ) reflects lexical or semantic processing. It is hypothesized that AD patients and age-matched normal controls will show priming effects for lexically- associated prime-target pairs (as determined by free association norms, e.g., """"""""butter-bread""""""""), but will not exhibit priming for semantically- related prime-target work pairs that are lexically unassociated (e.g., """"""""robin-chicken""""""""). For SAD patients, it is also predicted that for the lexically-associated paires, priming effects will be abnormal for items that were responded to abnormally in a previous word association task. These findings would indicate that traditional """"""""semantic"""""""" priming tasks tap lexical processing more than semantic processing in AD patients and elderly normals, and that speculating about the status of semantic knowledge ont eh basis of these priming effects is mistaken. AD patients and controls will also be given a priming task (using the same prime-target words used in the pronunciation task) in which they are explicitly required to make a semantic judgment (i.e., indicating whether each work presented on the computer screen is a living thing). It is predicted that AD patients will show abnormal priming effects on this task, and that these abnormalities will be most striking for items that were not named correctly on a previously-administered naming task. Comparisons of the patterns of results for the two priming tasks (which use the same stimulus words) should yield important clarification about lexical and semantic impairments in AD patients.
Showing the most recent 10 out of 48 publications
