Repertoires of emotion regulation: A person-centered approach to assessing emotion regulation strategies and links to psychopathology

Abstract

Despite growing research on emotion regulation (ER) strategies and psychopathology, research has primarily focused on identifying one-to-one associations between ER strategies and symptoms. Thus, little is known about how patterns in the repertoires of ER strategies are associated with different mental disorders. We utilised latent class analysis to identify distinct repertoires of ER strategies, and their links with various psychopathology domains (i.e., anxiety, depression, disordered eating, borderline personality). Participants (N = 531) reported on their use of seven ER strategies in six recalled stressful contexts, as well as on their symptoms of psychopathology. We identified five classes of ER strategies: Low Regulators (n = 168), High Regulators (n = 140), Adaptive Regulators (n = 99), Worriers/Ruminators (n = 96) and Avoiders (n = 28). Generally, High Regulators and Worriers/Ruminators endorsed greater levels of psychopathology, relative to Low and Adaptive Regulators. Our findings underscore the importance of characterising the dynamics of ER repertoires when seeking to understand links between ER strategies and psychopathology.

Keywords:

• Emotion regulation
• Regulation strategies
• Psychopathology
• Repertoire
• Transdiagnostic

The authors would like to thank Nicole Weiss for her helpful comments on an earlier version of this manuscript.

The authors would like to thank Nicole Weiss for her helpful comments on an earlier version of this manuscript.

Notes

¹ Comparable classifications emerged controlling for site, gender, age and ethnicity, χ²(16) = 1144.82, Cramer's V = .74, p < .001, indicating that class membership did not vary as a function of demographic variables. When we examined the GEE model with ER Class, Psychopathology, and ER Class × Psychopathology as fixed factors and including age, site, gender and ethnicity as covariates, this pattern of findings did not change.

² Comparable classifications emerged controlling for site, gender, age, and ethnicity, χ²(16) = 1144.82, Cramer's V = .74, p < .001, indicating that class membership did not vary as a function of demographic variables.

³ In light of the unexpected null findings with regard to disordered eating, we examined the subscales most theoretically associated with anorexia nervosa (dieting; oral control was not included due to low internal consistency in the present study) and bulimia nervosa symptoms separately. This GEE model (QICC = 3029.36) revealed significant effect of ER Class, Wald χ²(4) = 26.00, p < .001, and ER Class × Psychopathology interaction, Wald χ²(20) = 196.25, p < .001, but no significant effect of Psychopathology, Wald χ²(5) = 0.98, p = .96. Whereas the remaining Bonferroni-corrected pairwise comparisons mirrored those of the previous model, Low Regulators had greater bulimia nervosa symptoms than High Regulators, Adaptive Regulators, and Avoiders, ps < .001, and High Regulators had fewer symptoms than Low Regulators and Avoiders, ps ≤ .005.

⁴ The low levels of psychopathology symptoms among the Low Regulators class raises the question of whether low average use is reflective of greater variability across strategies or lower need to regulate. Thus, we conducted multivariate analysis of variance with ER Class as the independent variable and standard deviations of each ER strategy as dependent variables. Overall, the effect of ER Class was significant, F (28, 1972) = 6.04, p < .001, ηp2 = 0.08. Specifically, Bonferroni-corrected pairwise comparisons revealed Adaptive Regulators had greater variability in reappraisal than Low and High Regulators (ps < .01). In terms of worry/rumination, Worriers/Ruminators reported greater variability than Low and High Users, ps < .004. With regard to self-criticism, expressive suppression, and experiential avoidance, Low Regulators reported less variability relative to all other groups, ps < .05; Adaptive Regulators exhibited less variability in self-criticism than High Regulators and Worrier/Ruminators, and less variability in expressive suppression than Avoiders, ps < .05. Meanwhile, Avoiders demonstrated greater variability in experiential avoidance than High Regulators, p < .04.

⁵ See footnote 3 above.