(PS6-8) Emotion Recognition in Children with Callous-unemotional Traits: Does Anxiety Matter?

Callous-unemotional (CU) traits and anxiety have independently been found to be associated with anomalies in facial emotion recognition (FER) patterns in children. By addressing the common co-occurrence of CU-traits and anxiety, which are typically studied in isolation, the study endeavors to provide a comprehensive understanding of their combined roles in FER. Based on prior work, we hypothesized that CU traits are uniquely associated with deficits in FER, particularly for distress expressions (i.e., fear & sadness). Additionally, we hypothesized that anxiety is uniquely linked to the tendency to misinterpret ambiguous emotional expressions (i.e., neutral). Lastly, per research suggesting distinct CU subtypes based on anxiety level, we explored whether child anxiety moderates the link between CU traits and FER. In keeping with ABCT’s commitment to diversity and inclusion, we recruited a diverse sample and explored the potential role of race and ethnicity in these associations.

This study supports ABCT’s mission to advance scientific approaches to behavioral, cognitive, and biological evidence-based approaches to behavioral health. A community child sample ages 6 to 11 years (N = 101, 65% male, 46.5% White, 53.5% Black & Other) completed a computerized FER task displaying 36 faces, 6 per emotional expression, and the Revised Children’s Manifest Anxiety Scale – II for total and specified anxiety (worrying, physical, social) scores. Caregivers reported child CU-traits on the Antisocial Process Screening Device and behavioral/emotional issues on the Child Behavior Checklist. 

CU traits were uncorrelated with child-report anxiety (ps > .10) but positively associated with parent-report anxiety r = .246 p = .013. General linear mixed models (GLMM) indicated that child-reported total anxiety moderated the CU trait association with total FER accuracy (b = - 0.008, p = 0.037). Specifically, at relatively low levels of total anxiety, CU traits were not associated with total FER accuracy (b = 0.014, p = 0.721) but at moderate (b = - 0.071, p = 0,0369) and relatively high levels of anxiety (b = - 0.157, p = 0.001) were negatively associated with total FER accuracy. Child-report physical anxiety moderated the CU traits association with neutral FER accuracy (b = - 0.056, p = 0.047). At relatively low levels of physical anxiety, CU traits were not associated with neutral FER accuracy (b = -0.039, p = 0.651) but at moderate (b = - 0.199, p = 0.011) and relatively high levels of physical anxiety (b = - 0.358, p = 0.001), they were negatively associated with neutral FER accuracy. Child-reported social anxiety moderated the CU-traits association with fear FER accuracy (b = -0.078, p = 0.033). At relatively low (b = 0.204, p = 0.072) and moderate (b = - 0.091, p = 0.323) levels of social anxiety, CU traits were not associated with fear FER accuracy, but were inversely associated with fear FER at relatively high levels of physical anxiety (b = - 0.3869, p = 0.002). For Parent-reported anxiety, CU-traits were uniquely inversely associated with total FER accuracy (b = - 0.091, p = 0.038) but did not interact with parent-reported anxiety. We discuss the nature and potential theoretical and practical implications of these findings, including for intervention.