APA Journals Article Spotlight^®

For some people, age brings mild cognitive impairment (MCI) and Alzheimer's dementia (AD). Current research into preventing or minimizing these conditions is ongoing, but being able to predict them would render preventative approaches much more helpful.

To these points, Brainerd et al. (Journal of Experimental Psychology: Learning, Memory, and Cognition, 2014) (PDF, 232KB) pursued two goals as they applied dual-process models of retrieval to the study of neurocognitive impairment.

First, they sought to establish the connection between dual-retrieval models and neurocognitive impairment by using the models to identify the nature of memory declines in MCI and AD. Second, they looked to see whether dual-retrieval processes can predict who will develop MCI and AD.

Brainerd et al. conducted three studies (see Table 11).

View larger image

In Study 1, they analyzed 6 years of data from 842 older adults of differing levels of cognitive ability (healthy controls, subjects with MCI diagnoses, and subjects with AD diagnoses) who completed neuropsychological tests, received medical examinations, and provided buccal tissue samples for genotyping. Participants' reconstruction and recall test results were run through the dual-retrieval model to identify memory declines and differences in performance related to age. Brainerd et al. found that declines in reconstruction were the greatest indicators of transitions from one category to another (i.e., healthy control >> MCI and MCI >> AD). Further, increases in reconstructive forgetting were the hallmarks of transitions to new cognitive categories. Finally, changes in reconstruction and familiarity judgments were specific and sensitive to neurocognitive impairment.

In Study 2, Brainerd et al. investigated whether measurements of dual-retrieval processes could predict longitudinal transitions over short (16–18 months) and long (4.5–6 years) periods. They used dual-process models to analyze data from the same database subject pool as in Study 1 but from different waves of testing. They found that measures of reconstructive retrieval from the first data wave could predict which subjects with MCI were likely to transition to AD over the next 16–18 months as well as over the next 6 years. Performance on familiarity judgment and recollective forgetting also predicted which healthy control subjects were likely to transition to categories of neurocognitive impairment over the next 6 years. The genetic predictor did not share this accuracy, but it did provide an indication of conversion from healthy control to MCI.

In Study 3, Brainerd et al. used data from a different database and a different test with an eye toward fitting the model to the performance of individuals. As in Study 1, reconstructive retrieval differed for all three diagnostic groups, and recollective forgetting results separated healthy controls from individuals with MCI and AD. Recollective retrieval also separated the diagnostic groups, and recollective forgetting differed between subjects with MCI who did not convert to AD and subjects with AD.

Brainerd et al. drew five significant conclusions:

• Dual-retrieval models can extract reliable measurements of underlying retrieval processes from simple instruments.
• Declines in reconstructive retrieval were the prime memory marker in these studies: Shifts from healthy aging to neurocognitive impairment were accompanied by declines in nonrecollective processes.
• Dual-retrieval processes consistently predicted future emergence of neurocognitive issues, with sensitivity and specificity values well above chance.
• The genetic data added nothing to the dual-retrieval processes when predicting future disease.
• Dual-retrieval processes remained stable in healthy controls between the ages of 70 years and the early 90s; although there are substantial declines in episodic memory among healthy individuals prior to 70 years of age, subsequent declines seem to be associated with disease rather than with aging.

Citation:
Brainerd, C. J., Reyna, V. F., Gomes, C. F. A., Kenney, A. E., Gross, C. J., Taub, E. S., Spreng, R. N., & Alzheimer's Disease Neuroimaging Initiative. (2014). Dual-retrieval models and neurocognitive impairment. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40(1), 41–65. doi: 10.1037/a0034057

Return to APA Journals - Article Spotlight homepage

Note: This article is in the Basic/Experimental Psychology topic area. View more articles in the Basic/Experimental Psychology topic area.