Declines in representational quality and strategic retrieval processes contribute to age-related increases in false recognition.

Abstract

In a Yes/No object recognition memory test with similar lures, older adults typically exhibit elevated rates of false recognition. However, the contributions of impaired retrieval, relative to reduced availability of target details, are difficult to disentangle using such a test. The present investigation sought to decouple these factors by comparing performance on a Yes/No (YN) test to that on a Forced Choice (FC) test, which minimizes demands on strategic retrieval processes, enabling a more direct measure of the availability of object details. Older adults exhibited increased lure false recognition across test formats (Experiment 1), suggesting a decline in the availability of object details contributes to deficits in performance. Manipulating interference by varying the number of objects studied selectively enhanced performance in the FC test, resulting in matched performance across groups, whereas age differences in YN performance persisted (Experiment 2), indicating an additional contribution of impaired strategic retrieval. Consistent with differential sensitivity of test format to strategic retrieval and the quality of stimulus representations among older adults, variability in the quality of object representations, measured using a perceptual discrimination task, was selectively related to FC performance. In contrast, variability in memory control processes, as measured with tests of recall and executive function, was related to performance across test formats. These results suggest that both declines in the availability of object details and impaired retrieval of object details contribute to elevated rates of lure false recognition with age, and highlight the utility of test format for dissociating these factors in memory-impaired populations.

Neural evidence for age-related differences in representational quality and strategic retrieval processes

“…Here we provide neural evidence for both declines in the differentiation of stimulus representations during encoding, as well as reductions in the recruitment of controlled retrieval processes to reject recombined pairs, coupled with ageinvariant mechanisms for endorsing intact pairs. These results complement our previous behavioural data (Trelle et al, 2017) in suggesting a contribution of age-related changes in representational quality and strategic retrieval processes to episodic memory decline with age. Moreover, the present findings help to explain disproportionate deficits in recall-to-reject relative to recall-to-accept processes among older adults, and suggest that age-related declines in recollection-based retrieval processes vary as a function of demands on controlled retrieval processes.…”

“…This type of strategy may have been facilitated by the correspondence between the stored representation and the retrieval cue (in this case, most likely based on semantic rather than perceptual information due to the absence of the pictorial stimuli at test). Such a possibility would be consistent with some existing behavioural observations of increased ability to access target details when endorsing intact pairs, as compared to rejecting recombined pairs (Cohn et al, 2008;Trelle et al, 2017). Future studies that employ more direct measures of the retrieval of target details, such as the explicit report of target details from encoding not provided in the test cue, the use of remember/know judgments at test, or multivariate metrics of cortical reinstatement of target features during retrieval, are required to provide more direct evidence for this possibility.…”

“…We obtained evidence for age-related reductions in the differentiation of stimulus representations during encoding, as well as age-related changes in the neural mechanisms associated with rejecting recombined pairs (likely to reflect recall-to-reject strategies), coupled with ageinvariant mechanisms for endorsing intact pairs (likely to reflect recall-to-accept strategies). These results suggest that age affects both representational quality and retrieval control processes, consistent with existing behavioural evidence (Trelle et al, 2017), but using fMRI to more directly measure these constructs, and provide insights into the neural mechanisms that support successful memory retrieval in older adults.…”

“…This pattern suggests at least two separate factors contribute to age-related memory impairment: a reduction in the availability of high fidelity event representations that can effectively disambiguate between events with overlapping elements; and a decline in the accessibility of these representations, or at least the ability to intentionally retrieve and evaluate these details in a goal-directed manner using controlled retrieval processes. Recent findings from our research group lend support to this twofactor account (Trelle et al, 2017). We found that reducing demands on either the precision of representational content or the recollection-based retrieval processes alone did not eliminate agerelated memory deficits, but if both were reduced, age-related differences in memory performance were no longer observed, suggesting a contribution of both of these factors to memory decline in older adults.…”

“…In particular, the match between the studied pair and the test cue is said to facilitate access to and retrieval of stored detailsor recall-to-acceptand so minimize demands on retrieval cue elaboration as well as post-retrieval monitoring and evaluation processes (Rotello & Heit, 2000;Lepage et al, 2003;Cohn & Moscovitch, 2007). Notably, age-related differences in behavioral measures of recall-to-accept are often smaller in magnitude compared to measures of recall-to-reject (Cohn et al, 2008;Trelle et al, 2017). This pattern suggests that instead of a generalized deficit in recollection-based retrieval processes, older adults' ability to recall stored details may vary according to demands on strategic processes at test.…”