Depression and emotion regulation strategy use moderate age-related attentional positivity bias

Abstract

Effective emotion regulation is critical for maintaining emotional health in the face of adverse events that accumulate over the lifespan. These abilities are thought to be generally maintained in older adults, accompanied by the emergence of attentional biases to positive information. Such age-related positivity biases, however, are not always reported and may be moderated by individual differences in affective vulnerabilities and competencies, such as those related to dispositional negative affect and emotion regulation styles. To examine these relationships, we analyzed eye-tracking data from 72 participants (35-74 years; 50 female), 44 without and 28 with a diagnosis of Major Depressive Disorder during a free-viewing task comprising neutral-neutral, negative-neutral, and positive-neutral image pairs. Emotional bias scores were calculated based on the ratio of time spent dwelling on the emotional image vs. the neutral image in each emotional-neutral pair. Results indicate that healthy participants exhibited a stronger positivity bias than a negativity bias, whereas individuals with higher depressive symptom scores showed no difference. Next, we examined how age and emotion regulation strategy use (reappraisal vs. suppression, measured with the Emotion Regulation Questionnaire) impacted these effects. Individuals with Major Depressive Disorder did not exhibit a significant relationship between age and positivity bias. However, for healthy participants who self-reported a preference for using reappraisal in daily life, increased age was associated with an increased positivity bias. These findings indicate that the emergence of the positivity effect in older adults is related to reappraisal regulatory preferences in the absence of depressive symptoms.

Keywords: aging; attentional bias; depression; emotion regulation; eye-tracking.

Figure 1

Task paradigm. Participants were sequentially presented with 48 trials. Each trial consisted of a relax period (4.5 seconds), a fixation period that lasted until the eye tracker successfully captured both eyes looking at the fixation cross, and a pair of images (10 seconds) taken from a 96-item subset of the IAPS repository. Participants could freely gaze between the two images. The 48 pairs consisted of 16 positive-neutral, 16 neutral-neutral, and 16 negative-neutral pairings, presented in random sequence. On each trial, the location of the images on the screen (left or right) was randomly assigned.

Figure 2

Dwell times on images as a function of emotion. (A) Overall, participants (n = 72) spent more time looking at the emotional than the neutral images in each pair, irrespective of spatial location (left or right). We also observed a significant interaction of image emotion and pair condition due to a stronger differential attentional bias toward positive images in the positive–neutral pairs relative to the attentional bias toward negative images in the negative–neutral pairs (middle and right panels). No difference in dwell time was observed as a function of spatial location for neutral-neutral image pairs (left panel). (B) Regression with Beck Depression Inventory (BDI) sum scores revealed that only participants with lower depressive symptoms exhibited a stronger attentional bias for positive–neutral pairs than negative–neutral pairs. On the x-axis, the range of BDI scores for individuals diagnosed with Major Depressive Disorder (MDD) are indicated in purple while those for healthy controls are indicated in orange. All plots depict estimated marginal means from the linear mixed effects model and 95% confidence intervals/bands. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 3

Emotion regulation strategy preferences in healthy and depressed individuals. (A) Reappraisal use is negatively associated with suppression use in healthy controls, but not MDD participants. (B) Healthy controls exhibit a general preference for reappraisal use, while MDD participants exhibit no preference for either strategy (ERQ preference score = reappraisal average – suppression average). (C) For both groups, age was not significantly associated with an ERQ preference for reappraisal or suppression. Plots depict averages for each participant and 95% confidence intervals/bands. **p < 0.01.

Figure 4

Emotion regulation style moderates age-related positivity biases in non-depressed individuals. (A) The relationship between age and dwell times (y-axis) shifts based on ERQ preference scores (x-axis). Regions where the regression line and confidence band do not cross zero indicate a significant (p < 0.05) slope of age. As ERQ preference score increases (more preference for using reappraisal than suppression), the relationship between age and dwell times becomes more positive for positive images, and more negative for negative images, but only in healthy controls. (B) The relationship between age and dwell times at specific levels of ERQ preference, only for healthy controls. Values in parentheses indicate the raw ERQ preference score at each level from −1.5 to +1.5 SD (standardized within controls). Both plots depict estimated marginal trends computed from the higher-level linear mixed effects model and 95% confidence intervals/bands. *p < 0.05, **p < 0.01.