Answers are grounded in this paper and cite their sources.
Machine learning predicts callous-unemotional traits in early adolescents from various parenting features, suggesting intervention strategies.
Background Parenting is related to the development of callous-unemotional (CU) traits (i.e. low empathy and restricted guilt), making it an important target of interventions for childhood conduct problems (CPs). However, the relative importance of different parenting features in relation to the development of CU traits remains unclear. This study used machine learning to examine multiple parenting features assessed across infancy and early childhood as predictors of CU traits and CPs in early adolescence. Methods Data were from the Family Life Project ( N = 1,292; 49% female, 41% Black, and 28% below the poverty line). Seventy-four parenting predictors were assessed at eight time points between children aged 6–90 months using parent-reported questionnaires and observer ratings of videotaped interactions and home visits. CU traits and CPs were assessed via parent-reported questionnaires in preadolescence (12–14 years). Results Parenting features explained 8.2% of CU traits variability in preadolescence, with top predictors including early sensitive parenting and later behavior management and scaffolding practices. Prediction of CPs was weaker, with parenting explaining 4.5% of the variability. Conclusions Results highlight that disruption in close and sensitive early parent–child relationships is relevant to the development of CU traits. Results from the prediction of CPs indicate a more heterogeneous etiology. Findings support targeting parental sensitivity and behavior management within preventative interventions for CU traits and CPs.
No takes yet. Share an insight, caveat, or question.
Paz et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: