New research using the twin design reveals that anxiety levels in children with callous-unemotional (CU) traits can tell us something important about the origins of these traits. Children with CU traits and high anxiety appear to be more influenced by environmental factors and less by genetic factors. These findings suggest that assessing anxiety levels in children with CU traits may be crucial for tailoring treatment.
What are callous-unemotional (CU) traits?
Some children with conduct problems – a range of persistent oppositional and antisocial behaviors – also show a pattern of traits that we call “callous-unemotional” (CU) traits. These traits include low empathy, a lack of remorse, and limited prosocial emotions such as guilt, compassion, and concern for others. The DSM-5 includes a specifier for the presence of CU traits within the Conduct Problems diagnosis (i.e. “with limited prosocial emotions”). In comparison to conduct problems alone, these traits are more strongly influenced by genetic factors (Moore et al., 2019), appear to have different brain and behavioural correlates, and are associated with more persistent behavioural problems, which makes their presence or absence an important consideration in assessment and treatment planning. There is evidence that standard treatments like parenting training are helpful for young people with conduct problems and CU traits, but these treatments do not bring behaviour into the normative range for those with CU traits (Perlstein et al., 2023). Thus, clinical researchers have begun to develop treatments more specifically targeted to children with CU traits.
The puzzle of anxiety in children with CU traits
A key challenge in developing treatments for children with CU traits is that these children represent a heterogeneous group: while many experience very little anxiety, as might be expected given their low emotional range, a subgroup of children with CU traits experience high levels of co-occurring anxiety. Researchers have called the presentation in this subgroup “secondary” CU traits and hypothesised that children in this group have acquired CU traits as an adaptation to environmental stress, such as trauma and early life adversity, contrary to “primary” CU traits which are thought to have a stronger genetic basis (Todorov et al., 2024). This hypothesis raises a ‘nature vs. nurture’ question: Are CU traits indeed more strongly influenced by environmental factors among children with both CU traits and anxiety?
Our study: Using twin research to understand the origins of CU traits
Twin studies offered an ideal way to investigate this question. Because identical twins share 100% of their DNA and fraternal twins share about 50%, we can use differences and similarities between twin pairs to estimate how much genes and environments (both those experiences that are shared between twins growing up together and those that are different between twins) contribute to CU traits. Importantly, twins can differ in their anxiety levels and environmental experiences, allowing us to examine whether these differences matter. In our study, we used a twin modelling approach called the extended univariate genotype x environment (GxE) interaction model. This model allows us to test whether the genetic and environmental influences on CU traits change depending on anxiety levels. Importantly, this model accounts for genetic and environmental influences shared between CU traits and anxiety, as well as for gene-environment correlations, such as the possibility that children with a genetic predisposition to CU traits might experience more anxiety-provoking environments. Therefore, this model allowed us to confidently determine whether anxiety modulates the relative contribution of nature versus nurture on the development of CU traits.
We first tested this question in a large study of 1,196 twins (age 6-11) in the USA –the Twin Study of Behavioural and Emotional Development in Children (TBED-C; Burt & Klump, 2019), led by coauthors S. Alexandra Burt and Kelly Klump at Michigan State University. We hypothesised that our findings would fit with the theory described above: genetic influences would decrease and non-shared environmental influences would increase as anxiety increased. This would support the idea that “secondary” CU traits, or CU traits co-occurring with high anxiety, had a more environmental basis than “primary” CU, or CU traits with low anxiety. Our findings were generally in line with this hypothesis: there was a large and statistically significant increase in non-shared environmental influences on CU traits as anxiety increased. The moderating effect on genetic influences was also in the direction we expected (decreasing with increasing anxiety), but it was not statistically significant.
Because these findings were novel and might have treatment implications, we decided to see if they replicated in a different group of children. We preregistered a conceptual replication study in an even larger study of twins in the UK, the Twins Early Development Study (TEDS; Lockhart et al., 2023), which included 13,486 twins aged 7 years. Following preregistration, we ran the exact same models in the UK-based cohort as we did in the first, USA-based cohort. We found a very similar pattern of results: again, there was a large and significant increase in non-shared environmental influences on CU traits as anxiety increased. In this larger sample, there was also a significant decrease in genetic influences on CU traits as anxiety increased.
What do these results mean?
These findings provide the first evidence that the relative contribution of nature versus nurture influence on CU traits differs depending on the level of co-occurring anxiety. The strongest evidence across both cohorts was for an increase in non-shared environmental influences as anxiety increased. Non-shared environmental influences include everything that makes twins more different from each other. Although our study did not investigate exactly what those influences are, plausible candidates for future research include individual experiences of adversity or trauma, consistent with the theory of “secondary” CU traits, as well as differences in how sensitive individual children are to environmental stressors.
Implications for practice
The strongest takeaway from our study is that it is important to assess children with CU traits for comorbid anxiety. We know that targeted behavioural treatments are likely the best option for children with “primary” CU traits, or low comorbid anxiety. However, in the presence of anxiety, clinicians may need to assess for previous trauma and, if warranted, use a trauma-informed lens. Alternatively, with higher anxiety, clinicians may need to use a broader treatment that addresses both conduct problems and anxiety, and certainly, we need more research on treatment for young people with CU traits for those who also experience high levels of anxiety. It is worth noting that one trial found that children with secondary CU traits showed faster initial improvements during treatment but quicker deterioration during follow-up (Fleming et al., 2023), highlighting the need for careful treatment planning and potentially longer-term support for this group.
Strengths and limitations
This study has several notable strengths. We used complementary samples from two countries (USA and UK), with different measures of CU traits and anxiety, yet found consistent results. We used advanced twin modelling techniques that account for gene-environment correlation, strengthening our confidence in the findings. However, some limitations should be noted. We relied on maternal report of both CU traits and anxiety, which may introduce rater bias. We also did not measure trauma exposure, so we cannot definitively say whether the environmental influences we detected reflect traumatic experiences specifically. Additionally, both samples were predominantly White, which may limit generalisability to other populations. Finally, these analyses focused on middle childhood; the picture may look different in adolescence or early adulthood.
Conclusions
Our findings across two large, independent twin samples provide compelling evidence that anxiety may signal different developmental pathways of CU traits, with children with CU traits and high anxiety appearing to be more strongly influenced by environmental factors and less by genetic factors. These findings underscore the importance of measuring co-occurring anxiety in children with CU traits, as this may help us understand the origins of the CU traits themselves and guide us towards more personalised interventions.
References
Burt, S.A., & Klump, K.L. (2019). The Michigan State University twin registry (MSUTR): 15 years of twin and family research. Twin Research and Human Genetics, 22, 741–745.
Fleming, G.E., Neo, B., Kaouar, S., & Kimonis, E.R. (2023). Treatment outcomes of children with primary versus secondary callous-unemotional traits. Research on Child and Adolescent Psychopathology, 51, 1581–1594.
Lockhart, C., Bright, J., Ahmadzadeh, Y., Breen, G., Bristow, S., Boyd, A., . . . & Eley, T.C. (2023). Twins early development study (teds): A genetically sensitive investigation of mental health outcomes in the mid-twenties. JCPP Advances, 3, 12154.
Moore, A.A., Blair, R.J., Hettema, J.M., & Roberson-Nay, R. (2019). The genetic underpinnings of callous-unemotional traits: A systematic research review. Neuroscience and Biobehavioral Reviews, 100, 85–97.
Perlstein, S., Fair, M., Hong, E., & Waller, R. (2023). Treatment of childhood disruptive behavior disorders and callous- unemotional traits: A systematic review and two multilevel meta-analyses. Journal of Child Psychology and Psychiatry, 64, 1372–1387.
Todorov, J.J., Kohls, G., Pauli, R., Rogers, J., Bernhard, A., Ackermann, K., . . . & De Brito, S.A. (2024). Maltreatment and parenting in youth with primary and secondary callous- unemotional traits: Anxiety matters. JCPP Advances, 5, 12266.
About the authors
Patrizia is a Lecturer in the Division of Psychology and Language Sciences at University College London. Her research investigates developmental pathways to interpersonal violence and psychopathology, particularly how childhood maltreatment and personality vulnerabilities shape relational and mental health outcomes. Her work combines genetically informed and other quantitative methods with qualitative and co-produced approaches to inform prevention efforts.