Do Early Callous-Unemotional Traits and Low Prosociality Differentially Predict Externalizing and Relational Adolescent Outcomes?

• In Event: Poster Session 03
  In Poster Session: PS 03 Topic - Developmental Psychopathology

Abstract

Background
Recent research has highlighted the relevance of taking a more developmentally informed approach to increase our understanding of childhood callous-unemotional traits (CU) (e.g., Frick et al., 2014b). One such approach focuses on better identifying important correlates or components of childhood CU, such as guilt, empathy, and prosociality. Prosociality, in particular, is an important target given its demonstrated relevance across development (for meta-analysis see Waller et al., 2020), and the theoretical conceptualization of CU as similar to low prosociality. Research thus far has established childhood CU and prosociality as related but distinct constructs (e.g., Willoughby et al., 2015) that function as risk and protective factors for relational (e.g., Fontaine et al., 2018; Eisenberg et al., 2006) and externalizing behaviors (e.g., Frick et al., 2018; Nantel-Vivier et al., 2014). Nothing, however, is known regarding the extent to which this holds true into adolescence, and whether they independently predict behavior. Given that they show similar behavioral correlates, it is important to know whether interventions would benefit from targeting each separately, or addressing what is common to both.
Methods
Using a longitudinal sample of children assessed at ages 2 and 3 years, and followed-up in adolescence (M age=14.5; N=246; 46% male), we examined the extent to which early CU and early prosociality independently predicted later relational (bullying, victimization, cooperation) and externalizing (aggression, hyperactivity) outcomes in adolescence. Early childhood behavior was assessed via parent-report, and adolescent behavior via parent- and self-report for externalizing behaviors and self-report for relational behaviors.
Results
There were modest to moderate correlations between early CU and prosociality and later outcomes, with the exception of bullying and cooperation. When CU and prosociality were simultaneously included as early predictors of self-reported relational outcomes in a regression, they did not uniquely predict later behavior, suggesting that what predicts later victimization is what is common to both. Results for aggression and hyperactivity differed by rater. Post-hoc bifactor models (Figures 1 and 2) removing rater-specific variance revealed an overall factor representing variance common to parent- and self-report. Moreover, for both aggression and hyperactivity, this factor was uniquely predicted by CU and prosociality.
Conclusions
The current study supports the long-term developmental significance of early childhood CU into adolescence, particularly as a risk factor in predicting later externalizing behaviors. Moreover, this work provides further support to the argument that prosociality should be considered separately as a unique resilience factor against later externalizing problems. Though what is common to both appears to explain the modest associations with adolescent self-reported victimization, both independently accounted for variation in adolescent aggression and hyperactivity when tapping what is common across parent- and self-report. Thus, targeting both CU and low prosociality in interventions in early childhood, rather than one or the other, will likely be most beneficial in preventing later externalizing problems in adolescence.

Authors

• Megan Flom, Boston University
  Presenting Author

• Nicholas James Wagner, Boston University
  Non-Presenting Author

• Kimberly J Saudino, Boston University
  Non-Presenting Author