Implicit memory and Alzheimer's disease neuropathology.

Explicit memory failure is the defining cognitive feature of Alzheimer's disease and relates to the hallmark neuropathological features (plaques and tangles) of this illness. However, a pattern of preserved and impaired implicit memory has been found in Alzheimer's disease patients that may be explained by the association between the processing demands of certain implicit tests and the level of regional Alzheimer's disease neuropathology. In this study, we tested the hypothesis that these neuropathological features are related to implicit memory--measured by repetition priming--in a test that emphasized conceptual (or meaning-based) cognitive processing, and that the pathological changes are not related to implicit memory in a repetition priming test that emphasized perceptual (or sensory-based) cognitive processing. Subjects were older nuns, priests and brothers participating in the Religious Orders Study who agreed to annual neurological and neuropsychological evaluation for Alzheimer's disease and common neurological conditions of ageing, and brain autopsy at time of death. Explicit memory was measured by seven tests of episodic recall and recognition and converted to a previously established summary measure. Implicit memory was measured by four repetition priming tests. One test, category exemplar priming, emphasized conceptual, or meaning-based cognitive processing. A second test, word-identification priming, emphasized perceptual, or sensory-based cognitive processing. Two additional priming tests, picture-naming and word-stem completion, invoke both conceptual and perceptual processes. Neuritic and diffuse plaques, and neurofibrillary tangles identified by Bielschowsky silver stain, were quantified from five regions separately (frontal, parietal, temporal, entorhinal cortex and the hippocampus) and converted to a previously established summary measure. In linear regression analyses--controlling for age, sex and education--higher levels of Alzheimer's disease neuropathology were related to lower levels of explicit memory proximate to death. Higher levels of neuropathology were also related to lower levels of priming on the category-exemplar test, but were not related to levels of priming on the word-identification, picture-naming, or word-stem completion tests. The results suggest that hallmark indices of Alzheimer's disease neuropathology are associated with performance on priming tests to the extent that conceptual, but not perceptual, processing resources are required.