![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
ABSTRACT
Two experiments examined age differences in the effect of a sad mood induction (MI) on attention to emotional images. Younger and older adults viewed sets of four images while their eye gaze was tracked throughout an 8-s presentation. Images were viewed before and after a sad MI to assess the effect of a sad mood on attention to positive and negative scenes. Younger and older adults exhibited positively biased attention after the sad MI, significantly increasing their attention to positive images, with no evidence of an age difference in either experiment. A test of participants’ recognition memory for the images indicated that the sad MI reduced memory accuracy for sad images for younger and older adults. The results suggest that heightened attention to positive images following a sad MI reflects an affect regulation strategy related to mood repair. The implications for theories of the positivity effect are discussed.
Acknowledgments
We thank two anonymous reviewers for their excellent feedback and suggestions. We also thank Kristin Newman for her assistance with the training and mentorship of the student coauthors. Zorry Belchev is now at the University of Toronto and Stephanie Korol is now at the University of Regina. This research was supported by a grant from the Natural Sciences and Engineering Research Council: [Grant Number RGPIN 203664-2013] and a grant from Alberta Innovates-Health Solutions: [Grant Number 10002614] to Christopher Sears.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. A statistical power analysis was used to calculate the power to detect the three-way interaction. Assuming an effect size identical to the effect size of the two-way interaction between mood induction and image type (partial η2 = .12) and a total sample size of 43, achieved power was calculated to be at least 93% (using the G*Power 3.1 software package; Faul, Erdfelder, Lang, & Buchner, Citation2007).
2. The statistical power to detect the three-way interaction was at least 79% assuming the identical effect size as the mood induction × image type interaction (partial η2 = .12).
3. To more closely examine possible age differences in the “sure old,” “guess old,” “sure new,” and “guess new” responses, each of these responses was analyzed separately using a 2 (age group: younger adults, older adults) × 2 (study mood: neutral mood, sad mood) × 3 (image type: sad, positive, neutral) mixed-model ANOVA, with study mood and image type being within-subject factors. For the “sure old” responses, there was an effect of image type, F(2, 52) = 18.10, p < .001, partial η2 = .41, with more “sure old” responses for sad images (44.2%) and positive images (36.2%) than neutral images (21.3%), reflecting participants superior memory for emotional images (there were significantly more “sure old” responses for sad and positive images than neutral images, ps < .001, whereas sad and positive images were only marginally different, p = .07). There was also a marginally significant interaction between age group and image type, F(2, 52) = 3.02, p = .06, partial η2 = .10. The general pattern in the data indicated that younger adults had more “sure old” responses for sad images (48.7%) and positive images (42.9%) than older adults (39.7% and 29.4%, respectively), whereas for neutral images, there was a comparatively small difference between younger and older adults (18.9% vs. 23.6%). The analysis of the “guess old” responses produced no significant effects (all ps > .10). The analysis of the “guess new” responses produced a main effect of image type, F(2, 52) = 7.65, p = .001, partial η2 = .23, with significantly (p < .01) more “guess new” responses for neutral images (32.2%) than for sad images (20.3%) and positive images (24.5%), which reflects poorer memory for neutral images (i.e., participants had viewed the image but guessed that they had not). There were also two interactions: one between age group and image type, F(2, 52) = 3.87, p = .027, partial η2 = .13, and the other between study mood and image type, F(2, 52) = 6.10, p = .004, partial η2 = .19. The interaction between age group and image type reflected the large difference in “guess new” responses for neutral images between younger and older adults (39.6% vs. 24.8%), relative to the much smaller group differences for sad images (19.8% vs. 20.9%) and positive images (25.1% vs. 23.9%). The interaction between study mood and image type reflected the different impact of the sad MI on memory for sad and positive images: there were more “guess new” responses for sad images when they were viewed after the sad MI than when they were viewed before the sad MI (24.4% vs. 16.2%), t(27) = 2.52, p = .018, which reflected poorer memory for sad images after the sad MI. The opposite was true for the positive images: there were fewer “guess new” responses for positive images when they were viewed after the sad MI than before (21.5% vs. 27.6%), t(27) = 2.46, p = .02, which reflects better memory for these images when they were viewed after the sad MI. For neutral images, the percentage of “guess new” responses for images viewed before and after the sad MI was similar (30.7% vs. 33.7%), t(27) = 1.11, p = .27. Finally, for the “sure new” responses, there was a main effect of image type, F(2, 52) = 18.32, p < .001, partial η2 = .41, with significantly (p < .01) more “sure new” responses for neutral images (25.8%) than for sad images (10.3%), and positive images (16.2%), consistent with the previous analyses showing better memory for emotional images. There was also a marginally significant main effect of age group, F(1, 26) = 3.87, p = .06, partial η2 = .13, which reflected the tendency for older adults to respond “sure new” more often than younger adults (22.7% vs. 12.2%), reflecting older adults’ poorer memory for the images overall.