Age differences in managing response to sadness elicitors using attentional deployment, positive reappraisal and suppression

Abstract

The current study investigated age differences in the use of attentional deployment, positive reappraisal and suppression while regulating responses to sadness-eliciting content. We also tested to what extent these emotion regulation strategies were useful for each age group in managing response to age-relevant sad information. Forty-two young participants (M_age = 18.5, SE = .15) and 48 older participants (M_age = 71.42, SE = 1.15) watched four sadness-eliciting videos (about death/illness, four to five minutes long) under four conditions—no-regulation (no regulation instructions), attentional deployment (divert attention away), positive reappraisal (focus on positive outcomes) and suppression (conceal emotional expressions). We assessed negative emotional experience, expression, skin conductance level (SCL) and visual fixation duration while participants watched the emotional clips and followed the instructions for each condition. Results suggest that older adults were more successful than younger adults at implementing both attentional deployment and positive reappraisal. Ability to suppress emotions appears to remain stable with age. Within age-group comparisons suggested that for the older adults, positive reappraisal was a more useful emotion regulation strategy than the others, while the pattern among younger adults was less conclusive. Age-relevant differences in motivation and successful emotion regulatory efforts based on theoretical and empirical literatures are discussed.

Keywords:

• Ageing
• Sadness
• Emotion regulation
• Psychophysiology

This research was supported by National Institutes of Health [grant number R01 AG026323] to Derek M. Isaacowitz.

This research was supported by National Institutes of Health [grant number R01 AG026323] to Derek M. Isaacowitz.

Notes

¹ There was an additional online monitoring system that we had adopted to check whether the distracter task was successful or not and to minimise variance in mood data due to variability in mood before a new trial. This check insured that the participants were not in a negative mood due to the previous trial before beginning a new trial: right after completing the distracter task, the participants were asked to report their ‘current’ mood. If participants reported their mood lower than 40 (on the mood slider, where 0 was extremely negative and 100 was extremely positive) then they were given a break from the study, so that they can recover from any carry-over negative mood before the next trial was started. However, there were very few trials (less than 5) where the next trial was not started right away. Also the pattern was not specific to a condition or an age group. This further suggests that the distracter task was mostly successful in resetting moods between trials.

² A failure of the equipment to detect signals resulted in missing data: for SCL (one young and old participant), EMG data (two young and one old) and eye data (one young participant). Analyses include all the participants with relevant data.

³ The pattern of findings remained the same even when each measure was standardised on its respective baseline measure using baseline means and standard deviation estimates of the whole sample compared to the current individual-level difference scores.

⁴ The order of clips and instruction conditions was counterbalanced across participants. No significant main effect of counterbalancing order was found for any dependent measures: negative fixation (p = .89), mood (p = .64), SCL (p = .54) and EMG (p = .67). The pattern of findings for Goal 1 and Goal 2 involving the Age × Condition interaction still remained reliable when order was included as a factor in the model. Thus, to have a parsimonious model, we left the order variable from the model while reporting the findings of interest, discussed next.

⁵ Mauchly's test indicated that the assumption of sphericity had been violated (p < .05); therefore, degrees of freedom were corrected using Huynh–Feldt estimates of sphericity.