Memory processing in Alzheimer's disease

Abstract

The aim of the present research was to examine automatic and controlled influences on memory processing in patients with Alzheimer's disease using the process-dissociation procedure. In Experiment 1, a source recognition procedure was used, and the patients were found to have significantly reduced estimates of automatic processing and capacity to recognise words seen during the study phase of the procedure. In Experiment 2, a detection of repetition procedure was used to determine whether automatic influences on memory decline as a dementia progressed. The patients showed the expected inability to detect repetition in their responding, but there was no evidence that estimates of automatic processing were predicted by mental status scores or by ratings of the severity of dementia. In the third study, a novel method for estimating parameters in the process-dissociation model, developed from the task used in Experiment 2, was tested in a student sample. In this procedure, participants first produce semantic associates with either high or low relatedness to a list of cue words. These responses are subsequently used in a paired associate learning paradigm to determine independent estimates of recollection and automatic processing. Evidence for the validity of this procedure was found in Experiment 3 and the procedure used to examine memory processing in a sample of persons with dementia (Experiment 4). The patient group was found to have a substantial deficit in controlled recollection and a reduced capacity for automatic memory processing.