Objective: Episodic memory impairment is at the core of amnestic mild cognitive impair- ment (aMCI) and Alzheimer’s disease (AD). The origin of memory deficits may result either from an encoding deficit or from an accelerated decline of the memory trace. The present study explores these two hypotheses.
Method: We used the delayed-matching-to sample 48 items (DMS-48) memory test in a group of controls, aMCI patients, and AD patients (n= 16 in each group). The DMS-48 offers an incidental learning phase followed by three forced-choice recognition tests at three-minute, one-hour, and one-week delays. Moreover, the forced-choice test distinguishes three kinds of couple items: abstract (meaningless), paired (two similar exemplars), and unique (two different objects) items.
Results: As predicted by the accelerated forgetting hypothesis, patients showed a decrease in recognition performance over time. Controls also exhibited a similar decline in performance. As predicted by the encoding deficit hypothesis, abstract items were the most poorly recognized in AD, at both the three-minute and the one-week delays. In AD, recognition of the paired items also dropped after the one-hour delay, followed by unique items after a one-week delay. Patients with aMCI exhibited a performance that was similar to controls, except for abstract items, which dropped at the one-week delay.
Conclusions: These results are discussed in light of a third hypothesis, the memory strength hypothesis, in order to better account for the progressive decline in memory performance as a function of the item type in AD.