Older adults have especially poor recognition memory for word pairs, and recent research suggests this associative deficit manifests primarily in older adults’ higher rates of false alarms compared to younger adults. This could result from older adults either failing to generate meaningful (deep) mediators at study, or failing to benefit from having generated deep mediators at test. Younger and older adults performed a recognition memory task for words and word-pairs. A think-aloud analysis of their spontaneous encoding strategies (e.g., repetition, shallow mediators, and deep mediators) revealed that generation of deep mediators did not differ between younger and older adults, and was associated with high hit rates for items and associates in both age groups. However, generation of deep mediators was inversely related to false alarm rates in younger adults but not older adults. A trial-level analysis of encoding strategies and recognition responses revealed that younger adults benefited from having generated deep mediators when presented with corresponding recombined pairs at test as shown in their lower false alarm rates. In contrast, older adults who generated deep mediators during study (e.g., to blanket-figure) did not benefit from having done so when they encountered the corresponding recombined pairs at test (blanket-summer and district-figure): Their false alarm rates to pairs at test were unrelated to generation of deep mediators at study. These results suggest that many older adults have difficulty retrieving their mediators when presented with recombined pairs at test, older adults’ mediators are not distinct enough to individuate intact pairs from recombined pairs at test, or some combination of both.

Document Type

Post-print Article

Publication Date


Publisher Statement

Copyright © 2016 American Psychological Association. Article first published online: AUGUST 2016.


The definitive version is available at:

Full citation:

Mark C.Fox, Zachary Baldock, Sara P. Freeman, and Jane M. Berry. "The Role of Encoding Strategy in Younger and Older Adult Associative Recognition: A Think-Aloud Analysis." Psychology and Aging, 31(5) (August 2016), 471-480.