Cultural differences in categorical memory errors persist with age

ABSTRACT

Objectives: This cross-sectional experiment examined the influence of aging on cross-cultural differences in memory errors. Previous research revealed that Americans committed more categorical memory errors than Turks; we tested whether the cognitive constraints associated with aging impacted the pattern of memory errors across cultures. Furthermore, older adults are vulnerable to memory errors for semantically-related information, and we assessed whether this tendency occurs across cultures.

Methods: Younger and older adults from the US and Turkey studied word pairs, with some pairs sharing a categorical relationship and some unrelated. Participants then completed a cued recall test, generating the word that was paired with the first. These responses were scored for correct responses or different types of errors, including categorical and semantic.

Results: The tendency for Americans to commit more categorical memory errors emerged for both younger and older adults. In addition, older adults across cultures committed more memory errors, and these were for semantically-related information (including both categorical and other types of semantic errors).

Conclusion: Heightened vulnerability to memory errors with age extends across cultural groups, and Americans’ proneness to commit categorical memory errors occurs across ages. The findings indicate some robustness in the ways that age and culture influence memory errors.

KEYWORDS:

• Aging
• culture
• memory
• categorization
• cognition

Acknowledgments

This material is based upon work supported by the National Science Foundation under grant number BCS-1147707 (A. Gutchess) and A. Boduroglu was supported by the BAGEP Award of the Science Academy. The authors gratefully acknowledge experimental assistance from Müge Özbek, Peter Millar, Işılay Konuk, Ryn Flaherty, and Anna Carmichael.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Younger adult (Schwartz et al., 2014) and American data (Carmichael & Gutchess, 2016) were published previously.

2. Results persist with pattern matching and actual age as covariates. See Supplementals.

3. Results persist with pattern matching and actual age as covariates. See Supplementals.

Additional information

Funding

This work was supported by the National Science Foundation [grant number BCS-1147707] and BAGEP Award of the Science Academy.