Identifying stable variants of callous-unemotional traits: A longitudinal study of at-risk girls Article

Goulter, N, Kimonis, ER, Hawes, SW et al. (2017). Identifying stable variants of callous-unemotional traits: A longitudinal study of at-risk girls . DEVELOPMENTAL PSYCHOLOGY, 53(12), 2364-2376. 10.1037/dev0000394

cited authors

  • Goulter, N; Kimonis, ER; Hawes, SW; Stepp, S; Hipwell, AE



  • Callous-unemotional (CU) traits have proven important for designating children and adolescents showing a pattern of particularly severe, stable, and aggressive antisocial behaviors (Frick, Ray, Thornton, & Kahn, 2014). Individuals with secondary CU traits represent a subpopulation that are distinguished from those with primary CU traits by their high anxiety levels and marked histories of social/environmental adversity; however, evidence is largely based on cross-sectional male samples and this study is the first to examine stable trajectories of CU variants among an all-girl population. Using longitudinal data from the Pittsburgh Girls Study (N = 1,829), we examined whether valid, stable primary and secondary variants of CU traits can be identified among girls using CU traits and anxiety scores, and whether they predict poor adolescent mental health outcomes. Separate trajectory analyses conducted from ages 7 to 15 years indicated an optimal 4-class solution for CU traits (high, moderately high, moderately low, low) and 3 classes for anxiety (high, moderate, low). Classes of interest were combined; those girls with high-anxious secondary CU traits (n = 139) reported significantly greater harsh parental punishment, depression, and less self-control at age 7, and at age 16 were distinguished by greater symptoms of depression, borderline personality disorder (BPD), and conduct disorder (CD), compared with those with primary CU traits (n = 59) and low CU girls (n = 326). Findings improve current understanding of female CU traits by supporting the possibility of multiple developmental pathways, and extend it by identifying possible factors for targeted intervention among this understudied population.

publication date

  • December 1, 2017

published in

Digital Object Identifier (DOI)

start page

  • 2364

end page

  • 2376


  • 53


  • 12