Cognitive reappraisal attenuates the association between depressive symptoms and emotional response to stress during adolescence

ABSTRACT

Depression is associated with increased emotional response to stress. This is especially the case during the developmental period of adolescence. Cognitive reappraisal is an effective emotion regulation strategy that has been shown to reduce the impact of emotional response on psychopathology. However, less is known about whether cognitive reappraisal impacts the relationship between depressive symptoms and emotional responses, and whether its effects are specific to emotional reactivity or emotional recovery. The current study examined whether cognitive reappraisal moderated the relationship between depressive symptoms and trait or state measures of emotional reactivity and recovery. A community sample of 127 adolescents (M-age = 15.28; 49% female, 47% Caucasian), at an age of risk for depression, completed self-report measures of trait emotional responding and depressive symptoms. In addition, they completed an in vivo social stress task and were assessed on state emotional reactivity and recovery from the stressor. Findings suggested that cognitive reappraisal was associated with an attenuated impact of depressive symptoms on trait and state emotional recovery. These results provide evidence that cognitive reappraisal may be an effective strategy for improving some aspects of emotional responding in relation to depressive symptoms among adolescents.

KEYWORDS:

• Emotional reactivity
• emotional recovery
• cognitive reappraisal
• depressive symptoms
• adolescence

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Benjamin G. Shapero http://orcid.org/0000-0002-8764-6730

Notes

1 Given known sex differences in depression rates among youth (Breslau et al., 2017), and the finding that girls had higher trait emotional reactivity, exploratory analyses examined whether sex moderated the relationship between depression symptoms and emotional responding. Sex did not moderate the relationship between depressive symptoms and trait emotional sensitivity (t = 1.44, p = 0.15) or trait emotional persistence (t = 0.85, p = 0.39). Sex also did not moderate the relationship between depressive symptoms and state emotional reactivity (t = 0.81, p = 0.42) or state emotional recovery (t = 0.33, p = 0.32), when controlling for baseline levels of distress. In addition, sex did not moderate the relationship between cognitive reappraisal and trait emotional sensitivity (t = .34, p = 0.73) or trait emotional persistence (t = .56, p = 0.57). Further, sex did not moderate the relationship between depressive symptoms and state emotional reactivity (t = 0.09, p = 0.93) or state emotional recovery (t = .10, p = 0.92), when controlling for baseline levels of distress.

2 An alternative analysis to examine emotional recovery is to control for the degree of reactivity (change in self-reported distress from Time 1 to Time 2) as opposed to controlling for Time 1 self-reported distress. We conducted this alternative analysis and found that the interaction between depressive symptoms and reappraisal was marginally significant in predicting state emotional recovery using this approach (t = 1.87, p = .06). Consistent with the previous model, the form of this interaction was such that depressive symptoms predicted significantly higher distress at Time 3 (indicative of poorer emotional recovery) among individuals with less use of reappraisal (β = .49, t = 3.92, p < .005, 95% CI [0.24, 0.74), but not among individuals with greater use of reappraisal (β = .10, t = 0.63, p = .52, 95% CI [−0.22, 0.41]).

3 Given possible differences in emotion regulation styles and emotional responses (Garnefski et al., 2004; Neumann et al., 2010), we explored whether reappraisal’s attenuation of associations between depression symptoms and emotional responding may further depend on sex. Three-way interactions were examined, and indicated that there were no significant sex differences in the aforementioned moderation analyses.

Additional information

Funding

This work was supported by National Institute of Mental Health (NIMH) [grant number MH79369] and [grant number MH101168] to Lauren B. Alloy, [grant number MH099764] to Benjamin G. Shapero, [grant number MH099761] to Jonathan Stange, and a Social Sciences and Humanities Research Council (SSHRC): Banting postdoctoral Fellowship to Brae Anne McArthur.