Curvilinear Relations Between Preschoolers Effortful Control and Their Internalizing, Externalizing, and Total Problems

• In Event: 1-007 - Developmental Correlates of Effortful Control Across Childhood and Adolescence

Integrative Statement

Most developmental studies solely test for linear associations between children’s self-regulation and socioemotional problems; however, researchers have reported curvilinear effects with community samples. These findings suggest low-risk preschoolers’ extremely high or low levels of self-regulation more strongly contribute to their socioemotional problems than intermediate levels. Murray and Kochanska (2002) found a U-shaped association between preschoolers’ total problems and effortful control (EC), the ability to choose a course of action under conditions of conflict, detect errors, and plan (Rothbart, 2007). Carlson and Wang (2007) found preschoolers’ emotion regulation was most effective when they possessed intermediate levels of inhibition. White and colleagues (2011) recently reported that high inhibitory control increased risk of anxiety for preschoolers showing high behavioral inhibition. This paper extends this small body of literature to clarify functional associations between self-regulation and mental health in a low-risk sample, in which rates of psychopathology are often underestimated. We hypothesize that preschoolers’ EC will be (1) negatively correlated with and (2) have curvilinear effects on internalizing, externalizing, and total problems, and (3) curvilinear effects will remain significant after accounting for demographic and psychological correlates of self-regulation and adjustment, not controlled for in previous work.

Sixty-six preschoolers (M age = 37.97 months; 56.06% female; 55.38% non-Hispanic White) and parents (M age = 36.41 years; 95.38% college graduates; 59.02% $100,000+ annual household income) participated in 90-minute lab visits. Children participated in an 11-task EC behavioral battery adapted from Kochanska et al. (1996), which we recorded and later coded (kappa = .87) to create an EC score (alpha = .65). We administered receptive vocabulary, block design, and picture memory subtests of the Wechsler Preschool and Primary Scale of Intelligence–4th Edition (Wechsler, 2012) to create a mean IQ score (alpha = .59). Parents provided sociodemographic information and rated children’s internalizing (alpha = .85), externalizing (alpha = .89), and total problems (alpha = .95) on the Child Behavior Checklist Ages 1.5–5 (Achenbach & Rescorla, 2000).

Partly supporting hypothesis 1, Table 1 shows child EC is negatively correlated with externalizing but unrelated to internalizing and total problems. Supporting hypothesis 2, the product of child EC and itself (i.e., EC-squared, which tests quadratic effects) is correlated with internalizing and total problems. For hypothesis 3, we estimated hierarchical regressions separately for internalizing, externalizing, and total problems with variables entered in three steps: 1) Child age, sex, IQ, household income, parent age, education; 2) Child EC; and 3) EC-squared. We find no linear or curvilinear effects of EC on externalizing and total problems. The final regression for internalizing problems improves on simpler models (changeR2 = .09, p = .014) and is significant, F(8, 49) = 2.72, p = .015, R2 = .31. Only parent age (beta = -.31, p = .039) and EC-squared (beta = -.32, p = .014) predict internalizing problems. Unexpectedly, children with moderate levels of EC show the most internalizing symptoms (see Figure 1). Further analyses explore curvilinear relations between self-regulation and subdomains of socioemotional problems in low-risk preschoolers to elucidate their unique vulnerabilities to psychopathology.

Authors

• Daniel Ewon Choe, University of California, Davis
  Presenting Author

• Amanda Lawrence, University of California, Davis
  Non-Presenting Author