Spontaneous Production and Use of Mnemonic Strategies in Older Adults

Abstract

Mnemonic strategy use in older adults has been measured a number of ways, and although strategy measurement has received considerable attention, little work has been done to compare various types of subjective strategy reporting. The authors compared self-generated and endorsed strategies for memory tasks in a sample of 85 African-American and Caucasian older adults and investigated demographic characteristics related to each strategy measurement and the relationship between strategy use and memory performance. Across memory measures, significantly more strategies were endorsed than self-generated. Race, favoring Caucasians, was the most salient demographic predictor of mnemonic strategy usage. Although strategic behavior was associated with ability performance on most memory measures, specific strategies for optimal performance were identified for number and story recall only. Findings highlight the importance of measuring both self-generated and endorsed strategies and confirm previous work on the relationship between strategy use and memory performance.

A portion of this paper was presented at the 2003 annual conference of the Gerontological Society of America as part of a symposium entitled ‘Strategy use and cognitive performance’ (November, San Diego, CA).

The research reported in this study was funded by grant 1 R03 AG19879-01 from the National Institute on Aging awarded to Dr. George Rebok. The authors would like to thank Emily Anozie, Garrett Booth, Terry Froelich, Lisa McKenzie, Sharon Morley, Diana Potts, Robin Sills, and Janet West for their technical assistance with this project. The authors would also like to gratefully acknowledge the participants in the study.

Notes

Note. ∗p < .05; ∗∗p < .001.

Prevalence rates for each test do not total 100 because not all participants self-generated a strategy and participants were able to endorse as many strategies as they used.

Note. ∗p < .05; ∗∗p < .01; ∗∗∗p < .001.

OR = odds ratio; CI = confidence interval.

Race compares Caucasians (0) to African Americans (1).

Note. ∗p < .05; ∗∗p < .01; ∗∗∗p < .001.

Note. Age and education entered as covariates (N = 85).

∗p < .05; ∗∗p < .01; ∗∗∗p < .001. ^†Unadjusted model. ^‡Final adjusted model.