Poster #24 - Self-Regulation Facets as Predictors for Developmental Trajectories of Internalizing Symptoms From Childhood to Adolescence

• In Event: Poster Session 06

Abstract

Introduction: Internalizing symptoms (INT) are among the most common psychological symptoms in childhood and adolescence. Current research discusses the role of various facets of self-regulation (SR) in symptom genesis and considers lower capacities for SR to be associated with higher INT. The first objective of this study was to examine trajectories of INT from middle childhood to early adolescence whereby at least two stable and one increasing trajectories were expected. Second, we aimed to investigate whether a) lower levels of SR facets predict unfavorable trajectories of INT, that is increasing or stable high trajectories, and b) SR facets incrementally contribute to the prediction of these trajectories when other risk factors are taken into account.
Methods: The present project is based on a longitudinal data set with three measurement points and a sample of 1.657 children (52% female) drawn from German schools. At t1, children were 6-10, at t2 7-11 and at t3 9-13 years old. Assessments were self-, parent- and teacher-rated and included a variety of basal (working memory, inhibition, cognitive flexibility, emotional reactivity, heart-rate variability) and complex (planning behavior, affective decision-making, anger regulation, delay of gratification, behavioral control) SR facets, and well-established risk factors of INT (gender, self-esteem, socioeconomic status, family adversity, peer problems). INT was measured by the emotional problems scale of the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997), parent-report.
Results: Using Mplus (version 8.8) we estimated growth mixture models of INT and identified four distinct trajectory classes (see Figure 1): Class 1 with “stable low” symptoms (n = 1.061), class 2 with “increasing low to high” symptoms (n = 126), class 3 with “decreasing moderate to low” symptoms (n = 207), and class 4 with “decreasing high to moderate” symptoms (n = 59). Next, we conducted a multinomial-logistic regression analysis with SR facets at t1 as independent variables and class membership as dependent variable, class 1 serving as reference group (see Table 1). The membership to all classes vs. class 1 was significantly predicted by emotional reactivity and planning behavior. Furthermore, working memory significantly predicted the membership to class 2 vs. class 1. After adding risk factors as independent variables all the previously reported predictors remained significant except planning behavior for classes 2 and 3 vs. class 1. Significant risk-factors were socioeconomic status (for class 2 and 3 vs. class 1), family adversity and peer problems (both for class 4 vs. class 1).
Conclusion: In contrast to our expectations, we did not find a stable high trajectory but one increasing, two decreasing and one stable low class, which might be due to our community-based sample. The SR facets emotional reactivity, working memory and planning behavior could differentiate between participants with moderate to high symptoms at t1 or increasing symptoms and those with low symptoms even when well-established risk-factors (e.g., socioeconomic status) were included. Regarding emotional reactivity a dose-response-pattern could be observed, that is higher symptoms at t1 were associated with higher emotional reactivity at t1. High emotional reactivity may be a promising target for prevention.

Authors

• Johanna Klinge, International Psychoanalytic University
  Presenting Author

• Petra Warschburger, University of Potsdam

• Annette M Klein, Universität Leipzig