Aging, source memory, and the experience of “remembering”

ABSTRACT

In a previous study, we found source memory for perceptual features to differentiate between younger but not older adults’ reports of recollective (“remember”; R) and “know” (K) experiences. In two experiments with younger (17–30 years) and older (64–81 years) participants, we examined whether memory for meaningful speaker sources would accompany older adults’ recollective experience. Indeed, memory for male and female speakers (but not partial memory for gender; Experiment 1) as well as bound memory for speakers and their facial expressions (Experiment 2) distinguished between both younger and older adults’ RK reports. Thus, memory for some sources forms a common basis for younger and older adults’ retrieval experience. Nonetheless, older adults still showed lower objective source memory and lower subjective source-attribution confidence than younger adults when reporting recollective experiences, suggesting that source memory is less relevant to their retrieval experience than for younger adults.

KEYWORDS:

• Cognitive aging
• source memory
• binding
• remember/know
• recollection

Acknowledgments

We thank Thorsten Meiser for providing the voice recordings used in both experiments. We also thank Prisca Lankhuijzen, Franka Vontz, and Qizhi Yu for their help with participant recruitment and data collection.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Given our interest in within-comparisons between R and K judgments, it was not possible to include a few participants who reported only one type of experience in the repeated-measures analyses and we regard the assumption of observations missing at random for an alternative mixed-model analysis as violated in this case. Although it would be possible to include their data in the model-based analyses, including them would bias our analyses because they would only contribute to the parameter estimates for one type of retrieval experience (i.e., only for K experiences) but not for the other. Hence, differences between R and K parameters (e.g., the crucial d^K = d^R comparison) could be due to some participants only contributing data for one kind of retrieval experience rather than true R vs. K differences.

2. Not all participants provided data for all four accuracy × retrieval experience cell and were hence not included in the repeated-measures analysis. In Experiment 1, this was the case for five younger and four older adults and in Experiment 2 for one younger and three older adults.