A positivity effect in autobiographical memory, but not phonemic fluency, in older adults

ABSTRACT

To investigate whether processing fluency or cognitive control processes underlie aging-related positivity effects in memory, we compared retrieval of words on a fluency task, and of events on an autobiographical task, in younger and older adults. No positivity effect in word output was found on the fluency task, though older adults output more neutral words. For our autobiographical task, participants wrote descriptions of personal events related to cue words (3 each of positive, negative, neutral). They then classified their memories by valence, and subsequently rated how they ‘felt now’ about each. Older adults output more autobiographical memories classified as positive, and rated their memories more positively than did younger adults. We suggest the aging-related positivity effect emerges in service of emotion regulation, and is primarily observed when the cognitive task allows for personal evaluation and/or engages a reflective style of processing, as on an autobiographical but not a fluency task.

Keywords:

• Aging
• Emotion
• Autobiographical memory
• Phonemic fluency
• Positivity effect

Acknowledgments

This research was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC) discovery grant to M.A.F. and an NSERC scholarship to J.C.T. We thank Michelle Manios, Cheryl Lee, Jackelyn Chung, Wei Min Lim, Christine Lo, Gillian Adams, and Mark Laframboise for help with data collection and/or entry.

Notes

¹ Consistent with literature indicating greater vocabulary with aging (e.g., Verhaeghen, 2003), the older adult group had a higher full scale IQ (FSIQ), t(108) = -8.06, p < .0005, as estimated by the National Adult Reading Test – Revised (NART-R, Nelson, 1992) than the younger adult group. (In the NART-R, participants are asked to read irregularly spelled words out loud.) Older adults also had a greater number of years of education, t(63) = -2.08, p < .05 (though years of education data were missing for four older adults).