The effect of emotion-focused orientation at retrieval on emotional memory in young and older adults

Abstract

This study examines how emotion-focused orientation at retrieval affects memory for emotional versus neutral images in young and older adults. A total of 44 older adults (ages 61–84 years, M=70.00, SD=5.54) and 43 young adults (ages 17–33 years, M=20.58, SD=3.72) were tested on their free recall and forced-choice recognition of images. At retrieval the emotion-focused orientation was manipulated by instructing participants to focus on emotion-related information (i.e., emotional content of images and the emotional reactions evoked by the images). In the control conditions participants were either instructed to focus on visual information or not provided any specific orientation instruction. In free recall but not forced-choice recognition, the emotion-focused orientation increased young adults’ positivity bias and thus wiped out their superior negativity bias. However, the emotion-focused orientation did not affect older adults’ emotional memory. The data suggest that young adults activate and prioritise emotional goals in response to external demand during intentional information processing whereas older adults seem to spontaneously tune themselves to emotional goals.

Keywords:

• Emotional memory
• Ageing
• Emotion-focused orientation
• Positivity bias
• Negativity bias

Acknowledgements

This work was supported by grants from the Faculty of Arts SRC research fund of Ryerson University and the Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery Grant (371762-2009) awarded to Lixia Yang. We thank all the research assistants involved in data collection.

Notes

¹Despite age differences in BDI and BAI scores, the correlations between these variables and memory performance (i.e., free recall and recognition) or positivity/negativity bias scores are consistently low and non-significant (rs<.20). In addition, all the critical effects (e.g., age, valence, orientation, and the age×orientation×valence three-way interaction in the overall free recall analysis, as well as the age×orientation interaction in the free recall memory bias analysis) remain significant (Fs>3.66, ps<.05) even after controlling for age differences in BDI and BAI by including them as covariates. In addition, the main critical effects remain significant (Fs>3.25, ps<.05) even after controlling for age differences in all the demographic variables (i.e., education, speed, Shipley vocabulary, and health rating). Taken together, the analyses of covariance suggest that the age differences in all these variables do not account for the main result patterns reported in the article.