Are Callous Unemotional (CU) Traits Enough?
A guiding question for this special section, sparked by the increasing number of studies on callous –unemotional traits (CU traits) and the DSM-5 Limited Prosocial Emotion (LPE) specifier, is whether CU traits are enough to identify a clinically meaningful subgroup of youth with conduct problems or whether other psychopathic traits contribute. Seven empirical papers and two commentary papers address this question to varying degrees and in different ways with some papers “competing” various models against one another along with other papers investigating distinct correlates of the psychopathy factors. Findings across the papers suggest that there could be benefits to examining the broader construct of psychopathy in addition to examining its underpinning dimensions rather than CU traits alone. Two commentary papers also shed light on the importance of psychopathic traits in children with Conduct Disorder (CD) and provide insights for future work on this topic. It is hoped that the articles and commentary will generate new ideas for research and practice that could have implications for the treatment of those individuals with CD and psychopathic traits.
KeywordsPsychopathy Child Dimensions GM traits CU traits CD
Youth psychopathy has emerged as an increasingly important concept in understanding conduct problems (CPs) and in subtyping youths with Conduct Disorder (CD). This is evident from the inclusion of callous unemotional (CU) traits into the fifth edition of the Diagnostic and Statistical Manual (DSM-5) through the Limited Prosocial Emotions (LPE) specifier. However, CU traits represent only one component of the multidimensional psychopathy construct first delineated by Hervey Cleckley (1941) in his seminal book The Mask of Sanity. Cleckley (1941) discussed different aspects of psychopathy, such as the interpersonal, affective, and behavioral components of the condition. The multicomponent model for psychopathy was later further emphasized by Robert Hare (1991) when he developed the Psychopathy Checklist-Revised, a systematic tool for indexing psychopathy. The departure from the multicomponent model of adult psychopathy raises many concerns pertaining to the appropriateness of only a single element of psychopathy being included in the DSM-5 and potentially the ICD-11 (Salekin 2016a, b). Aside from the disconnect from the adult literature, Lahey (2014) noted that the criteria that appear in the DSM are not actually the criteria that were tested in the research base originally used to argue for the specifier (Frick & Moffitt, 2010). Moreover, Lahey (2014) expressed concern that the specifier may omit important psychopathic markers such as “unsympathetic concern” and, in their place, include less specific items (e.g. shallow affect) that may miss the intended target. Lahey’s point was that psychopathic youth may have a troubling level of unconcern for the welfare of others (e.g. carefree attitude about seeing others in distress or pain), but exhibit a wide range of emotions for other life events (e.g., experiencing joy when duping unsuspecting others). Thus, items like “shallow affect” may not fully capture in detail how youth with CD and psychopathic traits experience emotion.
A critical issue is whether CU traits, even if properly configured, offer sufficient information to forego use of the other psychopathy factors in the specification of the diagnosis of CD, or whether the unitary construct of psychopathy (as well as its differing dimensions) may offer additional information and thus be beneficial in the prediction of negative outcomes for those youth with conduct problems (Salekin et al., 2018). This special section is designed to add to the literature in this area by comparing various models or ways to subgroup youths with CD symptoms as well as to examine the different dimensions of psychopathy and test their relation to various external constructs. The papers in this special section take a variety of approaches to addressing this issue. Four papers compare CU traits alone against the broader construct of psychopathy. Three other studies examine whether the dimensions of psychopathy differ with respect to their relation to external correlates. We divide the papers into two sections: i) longitudinal papers that compare models, referred to as “competing models,” and ii) papers that examine psychopathy dimensions and the differential relationships with external variables, referred to as “unique relations.”
Prior to reviewing the papers, it is important to note that factor analytic studies have consistently shown that psychopathy is multidimensional (e.g., Harpur, Hare, & Hakstian, 1988). Originally, support was found for a broad two factor model (Factor 1 or interpersonal affective and Factor 2 or impulsive antisocial; Harpur et al., 1989). Later, factor analytic studies yielded three and four facet models (Frick et al., 2000; Hare, 2003). The multidimensional factor structure appears to hold across psychopathy measures (e.g., Andershed et al., 2002; Frick et al., 2000; Hare, 2003), though, the factor terminology can be different. For instance, the PCL-YV refers to the lower order facets as Interpersonal (superficial charm, glib, grandiose), Affective (lack of remorse, lack of empathy, shallow affect), Lifestyle (need for stimulation, proneness to boredom, parasitic lifestyle), and Antisocial (revocation of conditional release, criminal versatility). The YPI refers to the factors as Grandiose Deceitful (GD), Callous Unemotional (CU), and Irresponsible-Impulsive (II). Similar to these latter terms, the terms grandiose-manipulative (GM), callous-unemotional (CU), and daring-impulsive (DI) have recently been introduced (Salekin, 2017). Despite different terminology, the four factors are meant to represent approximately the same content (e.g., interpersonal, affective, lifestyle, and antisocial). The section below contains studies that examine the broader construct of psychopathy against CU traits alone.
Competing Models for Specifying CD
Four papers in this special section compare models of CU against the broader construct of psychopathy. Each study offers conclusions regarding the effectiveness of CU versus the broader concept. Specifically, Colins, Andershed, Salekin, and Fanti (this volume) using a sample of children (321 boys and 369 girls aged 7–12 in Cyprus) examined the conduct problems of these youths at 6- and 12- month longitudinal follow-ups. Results showed that children with CU and Conduct Problems (CP) occasionally are at risk for future and stable CP. Across gender, however, youth with CP and elevations on all three dimensions of psychopathy showed the most robust and highest risk for future and stable CP. Also, those children with CP and CU, at first assessment only, were at similar risk for stable CP as those children who were high in CP only at first assessment, suggesting that simply assessing CU traits as a specifier may pick up on a similar group as CP alone. The authors conclude that CP plus CU, is less effective than the broader multidimensional model for psychopathy in combination with concurrent CP, in predicting future CP. The authors suggest that the broader construct of psychopathy could be helpful in specifying CD.
Andershed, Colins, Salekin, Lordos, Kyranides, and Fanti (this volume) used a community sample of young adolescents (n = 996, 12 year old Cypriot children) and compared CU traits to the broader multidimensional construct of psychopathy, with and without co-morbid concurrent CP, in the prediction of future as well as stable forms of antisocial behavior. The participants were followed for 3 years after initial assessment. The authors found that youth scoring high on all three psychopathic dimensions, and who also exhibited co-morbid concurrent conduct problems at the first assessment showed the highest risk for future and stable (across assessment points) conduct problems, aggression, and substance use, followed by youth who were high on all three dimensions of psychopathy but without co-morbid conduct problems at first assessment. Importantly, youth who exhibited conduct problems and CU traits only, were not the most stable in their CP or concomitant psychiatric problems (e.g., substance use). These authors argue for using the broader multidimensional concept of psychopathy to convey the most information in predicting stable forms of conduct problems, aggression, and substance use.
Frogner, Andershed, and Andershed (this volume) used prospective longitudinal data of 1867 3- to 5-year-old Swedish children followed annually for two years in order to compare groups of children with different combinations of CP and psychopathic traits in relation to fearlessness and symptoms of Attention Deficit Hyperactivity Disorder (ADHD). These authors found that children with CP and psychopathic personality at first assessment had higher baseline and stable levels of fearlessness and ADHD symptoms as compared to children with CP and concurrent CU traits. Children with the broader set of psychopathic traits were also more likely to display stable levels of CP and ADHD symptoms in combination, and the findings were similar regardless of gender. The authors conclude that there are reasons to consider the multidimensional concept of psychopathy as specifiers for CD, rather than using CU traits only, in order to optimize both diagnostic practice and treatment outcome.
Fanti, Kyranides, Lordos, Colins, and Andershed (this volume) compiled data from two large community samples of children (ns 1599 children, and 2719 adolescents) who were followed longitudinally. Both cross-sectional and longitudinal analyses were conducted. Cross-sectional findings indicated that only youth presenting with a combination of all three psychopathy dimensions scored above the clinical cutoff score for CD. The longitudinal findings provided evidence that high GM, CU, DI and CP can lead to clinical levels of CD symptoms. The authors conclude that their findings provide evidence that the additional psychopathy dimensions, over and above CU traits alone, can be helpful to clinicians when it comes to better understanding and treating CD, and thus contend that it would be beneficial to include the additional components of psychopathy in future versions of the DSM and ICD.
Dimensions and their Unique Relations with Key Variables
Recent studies have shown that the dimensions of psychopathy have differing psychological and biological correlates (e.g., Salekin, 2017). Articles in this special section further test such potential relations. Specifically, three studies examine the extent to which the three dimensions are related to various external criteria, or predict negative outcomes. For instance, Gillen, Lee, Salekin, Iselin, Harrison, Clark, Colins, and Salekin (this volume) utilize a sample of 144 US detained adolescents to examine the relation among the PCL-YV interpersonal, affective, and lifestyle traits in relation to emotional skills and recidivism. Gillen and colleagues found that interpersonal traits were positively associated with better face processing (especially processing of fearful and angry faces), whereas affective traits were related to worse face identification of happy faces as well as sad, angry, and fearful voices. Lifestyle traits (DI traits, or II traits) were associated with reduced emotional intelligence. And, the PCL-YV dimensions differentially predicted recidivism. The authors highlight the need to consider the broad construct of psychopathy as well as its underlying dimensions.
Ridder and Kosson (this volume), use a sample of 205 US incarcerated adolescents to cross-sectionally examine psychopathy and its dimensions in connection with several measures of psychopathology, exposure to violence, and antisocial behavior. The authors found that CU traits were important in helping explain the variance in many of the external criterion variables. However, GM and DI traits also explained unique variance in a range of criterion variables including violent and non-violent crime, exposure to violence, impulsivity, and anger expression. The authors conclude that multiple components of psychopathy are likely to be important both in predicting negative outcomes (e.g., recidivism) and with respect to understanding mechanisms involved in the development and maintenance of psychopathic traits.
Finally, Somma, Andershed, Borroni, Salekin, and Fossati (this volume) utilized a sample of 558 Italian high school students and examined the interaction effects between GM traits, CU traits, and II traits in conjunction with delinquency. The authors found that psychopathic traits were related to self-reported delinquency. Moreover, a significant three way interaction was found among the three psychopathy dimensions in the prediction of delinquency. The authors conclude that the three factor model of psychopathy may be quite helpful to clinicians and researchers in the prediction of delinquency, and thus would be important in the specification of CD.
The special section concludes with a discussion of this research by two noted scholars who provide their views as to both the promises and potential limitations of the manner in which psychopathic traits might be employed. Scott O. Lilienfeld (this volume) argues that there are benefits to considering the broader syndrome of psychopathy but also its underpinning dimensions. He questions whether psychopathy can be considered a syndrome in the classic sense or whether it should be considered configural (each individual dimension comes together to create psychopathy). He argues for additional research on the topic and offers five recommendations for future work in this area including to 1) avoid exclusive reliance on total psychopathy scores, 2) connect the psychopathy literature with the literature on general models of personality, physiology and genetics, 3) examine specificity of findings in relation to other disorders (e.g., ADHD and adult personality disorders), 4) examine potential suppressor effects, and 5) examine the statistical interactions among the psychopathy dimensions. Steve S. Lee (this volume) notes the richness of the studies in the special section and argues for the value of contextualization, attention to developmental issues, and the importance of future research in this area to “catalyze deeper thinking” and more refined efforts to understand the origins, development, and nature of psychopathic traits in young people. He notes that these future studies, if carefully crafted, will lead to the elucidation of causal mechanisms and have implications for prevention and intervention.
Summary, Conclusions, and Ways Forward
As Lahey (2014) so poignantly noted, it is not long before the ink dries on one version of a DSM that questions arise about what new set of revisions might be needed to help us better understand the various disorders that are housed within its pages. Specifically, Lahey (2014) noted that each edition of the DSM “freezes scientific thinking” while the accumulation and consideration of evidence is ongoing. Fortunately, however, each new edition also provides the “impetus for the next round of research” (Lahey 2014, p. 58) and diagnostic considerations. This practice is likely advantageous for the fields of psychology and psychiatry if the aim is to improve what we know about classification, etiology, and treatment of various disorders. And, this may be especially true with CD and its more recent specifier of CU traits, which deviates markedly from theoretical models.
Cleckley (1941/1976) and Hare (1991/2003) provided anchors for understanding the psychopathic characteristics which included interpersonal, affective, and behavioral features. Factor analytic research has consistently shown that three or four factor models underpin psychopathy (Frick et al., 2000; Hare, 2003). Child studies, however, have often focused on CU traits alone. As noted above, this singular set of traits does not map well to work with adults where multidimensional models of psychopathic traits are employed. While some proponents of the CU model suggest that CU traits provide enough utility to capture those with psychopathic traits, this proposition lacks scientific support, leading to the question -- are CU traits enough?
The current special section was designed to test whether the overall psychopathy construct might perform better than CU traits alone in the prediction of negative outcomes. The four longitudinal studies showed that there may be advantages to considering the broader multidimensional concept of psychopathy in specifying CD. In all four longitudinal studies, the multidimensional construct was more predictive of negative outcomes than CU traits alone. The outcomes included conduct problems and or delinquency, which suggests that the construct, when broadly measured, is helpful in the prediction of negative outcomes and particularly downstream conduct problems.
There is also a notion among the papers included in this special section that the underpinning dimensions of psychopathy may have differing external correlates. Three studies in the special section showed that the dimensions may signal specific information regarding external correlates including face processing, emotional skill, concomitant mental health problems and delinquency. These important distinctions could highlight different mechanisms that may drive various permutations of the condition.
Overall, the findings of the papers presented in this special section suggest that it will be important to consider the broader multidimensional concept of psychopathy in youth samples, as well as to study the importance of the individual dimensions in conjunction with CD. Two commentary papers provide important insights into the papers in the special section and also shed light on areas in which the studies can improve and direct scientists toward new avenues for research. Although the articles in the special section do not conclusively resolve all the critical issues regarding the classification of youth with conduct problems, they do outline the potential utility of child psychopathic traits for classification systems.
A main goal of the special section was to spark new ideas and new methodological/analytical approaches and open a discussion of the ways in which psychopathy could be investigated in children and adolescents. Much of the work contained in this special section may be worth consideration in the specification of adult Antisocial Personality Disorder. We would argue that several of the included papers indeed use new analytical approaches, making the comparison between various subtyping models possible. It could be important for future studies to investigate if the findings using this type of analytical approach can be replicated in independent, larger samples using various external, clinically important variables. This special section attempted to conduct investigations globally, as it is important to improve the generalizability of the findings and to determine the cross cultural validity of CD with psychopathic traits. In closing, it should be noted that there have been a number of attempts to provide the field with the most defining and even core symptoms of the disorder (e.g. prototypical analyses, item response theory (IRT), and most recently, network analyses); although valuable, these studies will need to be complemented with studies such as those in this special section that compare models. Although a special section on this topic cannot answer all the questions regarding CD and child psychopathy, we believe that the manuscripts of this special section offer a starting point. In our view, examining the broader concept of psychopathy and its underpinning dimensions will be necessary to advance world science and clinical care for youth with various forms of CD.
Compliance with Ethical Standards
Conflict of Interest
Randall T. Salekin, Henrik Andershed, Blair D. Batky and Andrew P. Bontemps declare that they have no conflict of interest.
Ethical approval and participant consent were not sought/obtained for this paper as it did not involve human participants.
- Andershed, H., Kerr, M., Stattin, H., & Levander, S. (2002). Psychopathic traits in non-referred youths: A new assessment tool. In E. Blauuw & L. Sheridan (Eds.), Psychopaths: Current international perspectives (pp. 131–158). The Hague: Elsevier.Google Scholar
- Andershed, H., Colins, O., Salekin, R. T., Lordos, A., Kyranides, M., & Fanti, K. (this volume). Callous-unemotional traits only versus the multidimensional psychopathy construct as predictors of various antisocial outcomes during early adolescence. Journal of Psychopathology and Behavioral Assessment.Google Scholar
- Cleckley, H. M. (1941/1976). The mask of sanity: An attempt to clarify some issues about the so-called psychopathic personality. St Louis: Mosby.Google Scholar
- Colins, O., Andershed, H., Salekin, R. T., & Fanti, K. (this volume). Comparing different approaches for subtyping children with conduct problems: Callous-unemotional traits only versus the mutidimensional psychopathy construct. Journal of psychopathology and Behavioral Assessment.Google Scholar
- Fanti, K. A., Kyranides, M. N., Lordos, A., Colins, O., & Andershed, H. (this volume). Unique and interactive associations of callous unemotional traits, impulsivity (DI) and narcissism (GM traits) with child and adolescent conduct disorder symptoms. Journal of Psychopathology and Behavioral Assessment.Google Scholar
- Frick, P. J., & Moffitt, T. E. (2010). A proposal to the DSM-V Childhood Disorders and the ADHD and Disruptive Behavior Disorders Workgroups to include a specifier to the diagnosis of conduct disorder based on the presence of callous-unemotional traits. Washington, DC: American Psychiatric Association.Google Scholar
- Frogner, L., Andershed, A-K., & Andershed, H. (this volume). Psychopathic personality works better than CU traits for predicting fearlessness and ADHD symptoms among children with conduct problems. Journal of Psychopathology and Behavioral Assessment.Google Scholar
- Gillen, C. T., Lee, Z., Salekin, K. L., Iselin, A-M. R., Harrison, N. A., Clark, A. P., Colins, O., & Salekin, R. T. (this volume). Psychopathic traits in adolescence: The importance of examining components in face processing, voice processing and emotional skill. Journal of Psychopathology and Behavioral Assessement.Google Scholar
- Hare, R. D. (1991/2003). Manual for the revised psychopathy checklist (1st ed.). Toronto: Multi-Health Systems.Google Scholar
- Harpur, T. J., Hakstian, R., & Hare, R. D. (1988). Factor structure of the psychopathy checklist. Journal of Consulting and Clinical Psychology, 56, 741–747.Google Scholar
- Lee, S. S. (this volume). Multidimensionality of youth psychopathic traits: Validation and future directions. Journal of Psychopathology and Behavioral Assessment. https://doi.org/10.1007/s10862-018-9662-x
- Lilienfeld, S. (this volume). The multidimensional nature of psychopathy: Five recommendations for research. Journal of Psychopathology and Behavioral Assessment. https://doi.org/10.1007/s10862-018-9657-7.
- Ridder, K. A., Kosson, D. S. (this volume). Investigating the components of psychopathic traits in youth offenders. Journal of Psychopathology and Behavioral Assessment.Google Scholar
- Salekin, R. T. Andershed, H., & Clark, A. P. (2018). Psychopathy in children and adolescents: Assessment and critical questions regarding conceptualization. In C. J. Patrick (2nd Ed)., Handbook of psychopathy (pp. 479–508). New York: Guilford.Google Scholar
- Somma, A., Andershed, H., Borroni, S., Salekin, R. T., & Fossati, A. (this volume). Psychopathic personality traits in relation to self-report delinquency in adolescence: Should we mind about interaction effects. Journal of Psychopathology and Behavioral Assessment.Google Scholar