Global predictions of memory in Alzheimer's disease: Evidence for preserved metamemory monitoring

Previous researchers have argued that there is a metamemory monitoring defi cit in Alzheimer's disease (AD) because patients tend to overestimate their recall performance on a word list. We propose that these previous results are a misleading by-product of the methodologies used, rather than evidence of an underlying metamemory deficit. In two experiments, AD patients and o lder adult controls made predictions of performance both before and after e ncoding a to-be-remembered list. Metamemory function was measured by observ ing the shift in predictions made with, and without, an opportunity to moni tor the list. Experiment I found that although there were differences betwe en the groups ' accuracy for their prestudy predictions of recall, both gro ups were equally accurate after encoding. Experiment 2 explored this using four lists that varied in item difficulty and semantic relatedness. This ex periment replicated the findings of Experiment 1, and it was also found tha t the AD group became more accurate at predicting their performance with mo re exposure to study-test trials. These studies suggest that metamemory mon itoring is intact in AD, because AD patients utilize information gained dur ing processing the to-be-remembered items to revise their predictions of su bsequent performance.