Psychopathic Traits in Early Childhood: Further Validation of the Child Problematic Traits Inventory

Abstract

The aim was to further test the reliability and validity of a newly developed instrument designed to assess psychopathic personality traits in children, the Child Problematic Traits Inventory (CPTI). Data from the Preschool Twin Study in Sweden were used, a national general population study of 5-year-old twins (n = 1,188, 50.3% girls). Both preschool teachers and parents were used as informants. Confirmatory factor analysis replicated the intended three-factorial structure of the 28 items of the CPTI. Overall, our findings demonstrated good internal consistency and convergent validity, with all the teacher-rated CPTI scores being associated with teacher and parent ratings of externalizing psychopathology, aggressive behavior, fearlessness, and prosocial peer involvement. In conclusion, the CPTI hold promise as a teacher-rated tool for assessing psychopathic traits in childhood, though more research is needed to see if these findings can be generalized to other countries, settings, and older children.

Keywords

grandiose–deceitful callous–unemotional psychopathic traits psychometric properties preschoolers

Psychopathic personality is commonly described as a multifaceted personality disorder comprising a constellation of co-occurring personality traits being captured under at least three dimensions, including an interpersonal, an affective or callous–unemotional (CU), and a behavioral/lifestyle dimension (e.g., Cooke & Michie, 2001; Hare & Neumann, 2008). Many studies indicate that psychopathic traits can be measured in childhood and adolescence, are related to severe conduct problems, delinquency, various forms of aggressive behavior, including relational, reactive aggression (RA), and proactive aggression (PA; e.g., Frick, Ray, Thornton, & Kahn, 2014a; Lynam, Miller, Vachon, Loeber, & Stouthamer-Loeber, 2009; Salekin, 2008; Salekin & Lynam, 2010), and are at least moderately stable over time (for a review, see Andershed, 2010). Psychopathic personality has been described as a developmental disorder present already by the age of 8 years (Blair, 2010), and rooted in early childhood (Raine, 2013). To empirically test the idea that individuals with a psychopathic personality can be identified already in early childhood, a tool designed to assess all relevant psychopathy dimensions in early childhood is needed.

The Need for a New Tool

Various tools already exist that aim to assess psychopathic traits in children, including the Antisocial Process Screening Device (APSD; Frick & Hare, 2001), the Youth Psychopathic Traits Inventory–Childhood Version (van Baardewijk et al., 2008), and the Inventory of Callous–Unemotional traits (ICU; Frick, 2003). However, none of these tools was designed specifically for assessing these personality traits in early childhood (i.e., younger than age 6). Studies that used all three (Dadds, Fraser, Frost, & Hawes, 2005) or only the CU dimension (Kimonis et al. 2006) of the APSD in preschool children reported very low internal consistency of the APSD scores. This is unfortunate because it may reduce the ability to detect underlying mechanisms and risk features involved in the development of psychopathic personality (e.g., Farrington, Ullrich, & Salekin, 2010; Waller, Gardner, & Hyde, 2013). Although prior work showed that the ICU provides a reliable assessment of CU traits in preschoolers (e.g., Ezpeleta, Osa, Granero, Penelo, & Domènech, 2013), the ICU is limited in that it does not assess features relating to the interpersonal and the behavioral dimension of the psychopathic personality.

The few tools that were developed for use with very young children (e.g., Dadds et al., 2005; Scholte & van der Ploeg, 2007; Waller et al., 2013; Willoughby, Waschbusch, Moore, & Propper, 2011) assess only one (e.g., Willoughby et al., 2011 ) or two (e.g., Scholte & van der Ploeg, 2007) of the three aforementioned psychopathy dimensions. This is unfortunate because neither one of these new tools nor the ICU are suited for studying stability and change of psychopathic personality as a multifaceted disorder. Thus, to further our understanding of psychopathic personality as a developmental disorder, it is a necessity to have a tool that assesses psychopathic personality in early childhood in a way that closely resembles how it is often conceptualized in adolescence and adulthood, that is with interpersonal, CU and behavioral dimensions (e.g., Andershed, Kerr, Stattin, & Levander, 2002; Colins, Fanti, Salekin, & Andershed, 2017; Cooke & Michie, 2001). Such a tool would allow researchers to study stability and change of psychopathic personality throughout life without introducing unwanted variance in test scores due to difference in content and factor structure (Byrd, Kahn, & Pardini, 2013; Campbell, Doucette, & French, 2009).

The Child Problematic Traits Inventory

With the intent to provide a reliable assessment of interpersonal, CU, and behavioral/lifestyle psychopathic traits from early childhood onward, the Child Problematic Traits Inventory (CPTI; Colins, Andershed, et al., 2014) was recently developed and validated (Colins, Andershed, et al., 2014). The CPTI was developed for use in 3- to 12-year-old children, and primarily to be a teacher-rated instrument (for details, see Colins, Andershed, et al., 2014). In developing the CPTI, the aim was to assess psychopathic personality in (early) childhood in line with the three-factor model of psychopathic personality in adolescents (Andershed et al., 2002) and adults (Cooke & Michie, 2001). This implies (a) that the 28 CPTI items (see Table 3) were intended to load on three theoretically proposed dimensions or factors, being an interpersonal factor (labelled: Grandiose–Deceitful), a CU factor (labelled: Callous–Unemotional), and a behavioral factor (labelled: Impulsive–Need for Stimulation) and (b) that these three factors load onto an overarching latent psychopathy construct (i.e., Psychopathic Personality). Two principles guided our work in choosing which traits to include in the CPTI (for details, see Colins, Andershed, et al., 2014).

The first principle was that only those traits that have theoretical and/or empirical support for being applicable and assessable in 3- to 12-year-olds within the framework of the three-factor model conceptualization of psychopathy (Cooke & Michie, 2001) should be included. In our literature review (for details, see Colins, Andershed, et al., 2014), we found evidence that the three-factor model traits lying, deceitfulness/manipulation, grandiosity, lack of empathy/callousness, shallow affect, lack of remorse or guilt, impulsivity, need for stimulation, sensation seeking, and proneness to boredom, can be assessed in a meaningful way in early, middle, and late childhood. Therefore, the CPTI includes items that aim to assess these traits (Colins, Andershed, et al., 2014). We found no compelling theoretical or empirical arguments, though, to support the possibility that the three-factor model items parasitic lifestyle, lack of realistic long-term goals, and glibness/superficial charm, can be meaningfully assessed in early childhood. The CPTI, therefore, does not include items that tap these traits.

The second principle was that the CPTI should not include traits that are closely related to or even overlapping conceptually with rule-breaking, conduct problems, and antisocial behavior. This was decided to avoid issues with contamination (e.g., Skeem & Cooke, 2010) when using the CPTI as a measure of psychopathic personality traits in research aimed at understanding the development of conduct problems (e.g., Frick et al., 2014a). This implies that the traits irresponsible behavior and failure to accept responsibility for one’s own actions, and other antisocial features, such as items in the antisocial facet of the Psychopathy Checklist–Revised (Hare, 2003), were also not included in the CPTI (Colins, Andershed, et al., 2014).

Psychometric Properties of the Child Problematic Traits Inventory

The only study so far that examined the psychometric properties of the CPTI was carried out in a Swedish general population sample of 2,056 3- to 5-year-old children (53% boys; Colins, Andershed, et al., 2014). Exploratory factor analyses revealed that the 28 CPTI items loaded distinctively on the three theoretically proposed factors or dimensions. Confirmatory factor analysis (CFA) showed that this three-factor model fitted the data well in the total sample (root mean square error of approximation [RMSEA] = .07; comparative fit index [CFI] = .96; and Tucker–Lewis index [TLI] = .96), across gender (boys/girls: RMSEA = .07/.07; CFI = .97/.96; TLI = .96/.96), and across age groups (3-/4-/5-year-olds: RMSEA = .07/.07/.07; CFI = .98/.97/.97; and TLI = .97/.96/.97). The CPTI total and three-factor scores showed excellent internal consistency in the total sample, across gender, and across the three age groups (alphas from .88 to .96). In terms of external validity, the three factors exhibited positive and significant zero-order correlations with teacher-rated conduct problems, attention-deficit/hyperactivity disorder (ADHD) symptoms, easy temperament, and fearlessness (most rs approximately .60). These correlations remained significant after controlling for various sociodemographic characteristics, including age, socioeconomic status (SES), ethnic origin, and gender. Importantly, these correlations decreased substantially in strength when also controlling for the other two CPTI factors, suggesting that the relation between the CPTI factors and the external constructs at least in part lies in the combination of all three CPTI factors (Colins, Andershed, et al., 2014).

The Present Study

While there is evidence that the CPTI reliably and validly assesses a constellation of traits that is similar to psychopathic personality as manifested in adolescence and adulthood (Colins, Andershed, et al., 2014), future studies are warranted to provide additional support for the psychometric properties of the CPTI and to tackle some of the limitations reported in the only prior study testing the CPTI (Colins, Andershed, et al., 2014). First, the convergent validity of the CPTI scores needs and will be, in the present study, tested by examining its relation with external constructs of interest measured via well-known and already validated instruments. Second, the criterion measures used to test the convergent validity of the CPTI will be extended by testing if the CPTI total and CPTI factor scores are positively related to RA and PA, and to relational aggression, a subtle nonphysical form of aggression that primarily aims to damage relationships and feelings of belonging and friendship (e.g., Czar, Dahlen, Bullock, & Nicholson, 2011; Marsee, Silverthorn, & Frick, 2005). This is relevant because psychopathic traits have related to these forms of aggression (e.g., Blais, Solodukhin, & Forth, 2014; Colins, Andershed, & Pardini, 2015; Frick et al., 2014a; Kimonis et al., 2006). Third, whereas Colins, Andershed, et al. (2014) examined the external validity of teacher-reported CPTI scales in relation to teacher-rated variables, possibly leading to inflated relations among study variables due to information from a single informant, the present study will also test the external validity of the CPTI scores by testing relations between teacher-rated CPTI scores and parent-rated criterion measures.

Aims and Hypotheses

The primary aim of the current investigation was to examine the psychometric properties of the CPTI in a relatively large Swedish general population sample of 5-year-old children. First, we hypothesized that CFA would show acceptable to good model fit indices for the original three-factor structure of the CPTI in the total sample and across gender. Second, we hypothesized that the CPTI total and factor scores would show good to excellent internal consistencies. Gender differences in CPTI scores were also examined and based on prior work with preschoolers (e.g., Colins, Andershed, et al., 2014; Kimonis et al., 2006), elementary school–aged children (e.g., Fanti & Centifanti, 2014), adolescents (e.g., Colins, Noom, & Vanderplasschen, 2012), and adults (e.g., Miller, Watts, & Jones, 2011), we expected higher levels on all three factors in boys than in girls. Third, we hypothesized that zero-order correlations would show that CPTI total and the three-factor scores are significantly and positively correlated to teacher- and parent-rated oppositional defiant problems, conduct problems, ADHD symptoms, RA, PA, relational aggression, and fearless temperament and negatively to prosocial peer involvement. To enable comparison with Colins, Andershed, et al. (2014), we also tested whether these correlations would hold when controlling for different demographic variables, including gender, ethnic origin, and SES. It was expected that this would be the case, a finding that would suggest that the validity of CPTI scores can be generalized across various sociodemographic groups.

The secondary aim was to further unravel the unique relations between CPTI factor scores and the aforementioned criterion measures. Although most researchers refer to psychopathic personality as a constellation of co-occurring traits, numerous studies in child and adolescent literature on psychopathic traits provided data on one single psychopathy factor (i.e., CU traits) separate from other dimensions of psychopathic personality (e.g., Salekin, 2015). As such, it cannot be concluded if unique variance in CU traits or common variance with other dimensions of psychopathy account for the reported results (Frick et al., 2014a). To fill this void, and to enable comparison with the prior CPTI study (Colins, Andershed, et al., 2014), we examined the unique relation of the CPTI factor scores and criterion measures of interest by means of partial correlation analyses. It was hypothesized that the correlation coefficients between a single CPTI factor score and the external criterion measures, when controlling for the other two CPTI factors, would be substantially weaker than the correlations between this single CPTI factor score and these measures, and even may become nonsignificant. This would show that the three CPTI dimensions are at least in part dependent of each other in their relation to the external criterion measures. Put differently, such a finding would not only converge with defining psychopathic personality as a constellation of co-occurring traits but it would also suggest that the CPTI total score is more informative for the study of psychopathic personality than any of its factor scores on their own.

Method

Participants and Procedure

This study uses data from a larger study, the Preschool Twin Study in Sweden (PETSS). This PETSS project was designed to address how genes and environments in early childhood contribute to the development of behavioral problems. Parents of all twins born in Sweden between January 2004 and May 2005 were identified through the Swedish population-based medical birth register and contacted 1 month prior to the twins’ fifth birthday. Thus, all children in the sample were 5 years old at the time of assessment. Questionnaires and consent letters were sent to parents of 1,261 twin pairs. Nonresponders were approached with up to three reminders. Parents were approached separately, resulting in 828 (65%) responses from the mothers and 698 (55%) responses from fathers. Mother ratings or father ratings were available from 879 twin pairs (1,758 twins). With regard to origin, 7.4% and 7.2% of the mothers and fathers from these 1,758 twins, respectively, were born in a country other than Sweden. In terms of education, 2.4% and 4.8% of the mothers and fathers, respectively, had a low level of education (i.e., fewer than 9 years of elementary school). Teachers were only contacted after parental consent was obtained. The teacher-rated questionnaires were distributed to the teachers by the parents and had a response rate of 54% (n = 686 twin pairs).¹ The main focus of present study was the CPTI which was administered only to the teachers. A total of 1,188 children (591 boys and 597 girls) had responses from teachers on the majority of the CPTI items and could thus be included in the majority of the analyses of the present study. The PETSS project was evaluated and approved by an ethics committee at the Karolinska Institutet (#2007-1034). Only the parents received compensation for their participation.

Measures

Both the teacher- and parent-rated questionnaires consisted exclusively of already validated scales. Thus, they are only very briefly described here in the measures section.

Teacher-Rated Measures

Psychopathic traits

The CPTI (Colins, Andershed, et al., 2014) was used to assess psychopathic traits. Specifically, the Grandiose–Deceitful factor score (GD; eight items), the CU factor score (10 items), the Impulsive–Need for Stimulation factor score (INS; 10 items), and the total score (28 items), were calculated by computing the mean scores. The response scale of the CPTI is 1 (does not apply at all), 2 (does not apply well), 3 (applies fairly well), and 4 (applies very well). On the first page of the questionnaire, information is given concerning the overall content of the items and that the rater should assess each item based on how the child usually and typically behaves rather than based on how he or she behaves at the moment. All CPTI items are displayed in Table 3.

Oppositional Defiant and Conduct Problems

These variables were assessed by means of the Diagnostic and Statistical Manual of Mental Disorders (DSM)–oriented Oppositional Defiant Problems (ODP; 5 items, α in the present study = .75) and Conduct Problems scales (CP; 13 items; α = .85) from the Teacher Report Form 6-18 (TRF 6-18; Achenbach & Rescorla, 2001).² These scales use a 3-point response scale ranging from 0 (does not apply) to 2 (applies very well or often). The mean of the items was calculated to create the ODP score and the CP score. Example items for ODP are “argues a lot,” “disobedient at school,” “temper tantrums or hot temper,” and “stubborn, sullen, or irritable.” Example items for CP are “cruelty, bullying, or meanness to others,” “gets in many fights,” and “steals.”

ADHD symptoms

DuPaul’s ADHD Rating Scale–IV 5-17 (DuPaul, Power, McGoey, Ikeda, & Anastopoulos, 1998) was used to assess the 18 (α = .94) DSM-IV criteria of ADHD using a 4-point response scale ranging from 0 (never) to 3 (always). The ADHD score was created by calculating the mean of the 18 items. Example items are “He or she is running around, clutching, or climbing more than what is considered appropriate” and “He or she is inattentive on details or is careless.” The validity of this rating scale has been demonstrated in various samples, including 5-year-old preschoolers (DuPaul et al., 1998).

Reactive and Proactive Aggression

Both concepts were assessed through the RA (6 items, α = .78) and the PA (10 items, α = .86) scales from the Teacher Report of RA and PA (Brown, Atkins, Osborne, & Milnamow, 1996), using a 3-point response scale ranging from 1 (never) to 3 (very often). The mean of the items was calculated to gain the RA and PA scores. Example items for PA are “has hurt others to win a game” and “picks on smaller kids.” Example items for RA are “gets mad when corrected” and “mad when doesn’t get his or her way.”

Relational Aggression

This variable was assessed by means of the six-item Relational Aggression scale (α = .91) from the Preschool Social Behavior Scale–Teacher form (Crick, Casas, & Mosher, 1997). Items were scored on a 5-point response scale ranging from 1 (never applies) to 5 (always applies). The mean of the six items was calculated to gain the Relational Aggression score. Example items are “When mad at a peer, this child keeps that peer from being in the group,” “Tries to get others to dislike a peer,” and “Verbally threatens to keep a peer out of the group if the peer does not do what the child asks.” The validity of this rating scale has been demonstrated in various samples, including 5-year-old preschoolers (e.g., Ostrov & Keating, 2004).

Fearlessness

The Child Fearlessness Scale (see Colins, Andershed, et al., 2014) consists of six items (α = .81) that must be scored on a 4-point response scale, ranging from 1 (does not apply at all) to 4 (applies very well). A total Fearlessness score was created by calculating the mean of the six items. Examples items are “He or she does not seem to be afraid of anything,” “He or she does not seem to be afraid when someone is trying to frighten him or her,” and “He or she never seems to get scared when someone is mad at him or her.” Although this scale has not been extensively validated, prior work showed that this Fearlessness score was negatively related to the CPTI total and factor scores (Colins, Andershed, et al., 2014) and positively related to conduct problems (r = .50, p < .001; Andershed, personal communication, 2015).

Prosocial Peer Involvement

Prosocial Peer Involvement was assessed through the corresponding seven-item (α = .85) subscale from the Child Behavior Scale (Ladd & Profilet, 1996), that uses at 3-point response scale ranging from 1 (does not apply at all) to 3 (applies precisely). The Prosocial Peer Involvement score was computed by calculating the mean of the seven items. Example items are “is kind toward classmates,” “listens to classmates,” and “compromises in conflicts with classmates.” The validity of this scale has been demonstrated in various samples, including 5-year-old preschoolers (e.g., Birch & Ladd, 1998).

Parent-Rated Measures

The test battery described below was completed by both the mother and father of the twins and both of them answered the exact same questions about each child. This test battery was incorporated in a multidisciplinary study that focuses on various constructs other than psychopathic traits. Because parents were restricted in time to complete various tools, the test battery completed by the parents is not as extensive as the one completed by the teachers. In the present study, and unless described otherwise the mother and father ratings were combined (by calculating the mean of the mother and father scale). The rationale to do so was based on the principle not to present more results than needed since other approaches (i.e., keeping mother and father ratings separate and using the highest score prevails method) yielded the same results for all analyses reported in the present study (all details available on request). When only one parent responded, this single rater was used to yield the variable score of interest.

Parent’s SES and origin

SES was measured as the mean of the two parent’s highest finalized educational level ranging from 1 (elementary school fewer than 9 years) to 10 (research exam/PhD). The parents answered a question about the country where they were born. This variable was then coded with 0 (both parents born in Sweden) or 1 (at least one parent born in a country other than Sweden).

Oppositional Defiant and Conduct Problems

These variables were assessed by means of the DSM-oriented ODP (5-items; Mother/father/combined: α = .77/.78/.79) and CP scales (16 items; Mother/father/combined: α = .75/.76/.78) of the Child Behavior Checklist that has a response scale similar to the TRF-6-18. Mean scores were calculated.

ADHD symptoms, Relational Aggression, Fearlessness

These variables were assessed and created by means of the items from the corresponding measures being described in the teacher-rated measures section. Alphas for mother-/father-rated/combined ADHD symptoms were .92/.92/.93; for mother/father/combined relational aggression .84/.85/.86; and for mother/father/combined Fearlessness .85/.85/.86.

Statistical Analyses

We first present descriptive information for all continuous variables used in this study for the total sample and for boys and girls separately, with tests for gender differences. Effect sizes (Cohen’s d) were calculated as indicators of the magnitude of differences between both gender groups. Second, we conducted CFA using the mean and variance adjusted weighted least squares estimator appropriate for use with ordinal items (Flora & Curran, 2004) to test the three-factor model of the CPTI. Model fit was assessed using RMSEA, CFI, and TLI. RMSEA scores below .05 indicated good fit, whereas scores between .05 and .08 indicated acceptable fit. A CFI and TLI score of .95 or above indicated excellent fit, and a CFI and TLI score of .90 or more indicated good fit (Hu & Bentler, 1999). The three-factor model was specified with the 28 items as observed variables and the three factors as latent and correlated constructs. Finally, the model was specified to include an overarching latent psychopathic personality construct joining the three latent factor constructs. For reasons of comparison, and in line with prior CFA work with the CPTI (Colins, Andershed, et al., 2014) and other tools that assess psychopathic traits (e.g., Anestis, Caron, & Carbonell, 2011; Colins, Bijttebier, & Broekaert, Andershed, 2014), we also tested a one-factor structure (with all 28 items in the same factor). Third, to evaluate the internal consistency of the CPTI total and factor scores, Cronbach’s alphas (α) were calculated. Alpha coefficients were interpreted as follows: <.60 = insufficient; .60 to .69 = marginal; .70 to .79 = acceptable; .80 to .89 = good; and .90 or higher = excellent (Barker, Pistran, & Elliot, 1994). Fourth, zero-order correlations and—for the three CPTI factors also—partial correlations (controlling for the other CPTI factors and sociodemographic variables) between the CPTI scores and the external criterion variables, were computed. The correlation coefficients were interpreted as follows: weak ≤ .30; moderate = .30 to .50; and strong ≥ .50 (Cohen, 1988). CFA was conducted in Mplus 6.12 (Muthén & Muthén, 2011). All the other analyses were conducted in SPSS 20.0 with p < .05 as an indicator of statistical significance.

Results

Descriptive Information

Table 1 presents descriptive information for all continuous variables used in this study for the total sample and for boys and girls separately, with tests for gender differences. Boys had significant higher mean level scores than girls on almost all variables, with a few notable exceptions. Girls were significantly higher in teacher-rated prosocial peer involvement and parent-rated relational aggression than boys. No gender differences were revealed for teacher-rated relational aggression and parent-rated oppositional defiant problems. Finally, age and SES were robustly and significantly related to all CPTI scores and most criterion measures of interest, whereas origin only showed a significant correlation with CPTI INS (r = −.07; p < .02; details available on request). To allow comparison with the original CPTI paper (Colins, Andershed, et al., 2014), origin will nevertheless be used in all analyses as a control variable, together with age and SES.

Table 1.

Descriptive Statistics of All Study Variables Across Gender Along With t Tests Testing for Gender Differences.

	Boys		Girls		Gender differences
	N	M (SD)	N	M (SD)	t (df^a)	Cohen’s d
Teacher-rated variables
CPTI Total	591	1.73 (0.55)	597	1.55 (0.48)	5.98*** (1160.40)	.35
GD	591	1.42 (0.52)	598	1.33 (0.49)	3.04** (1187)	.18
CU	591	1.60 (0.65)	598	1.38 (0.51)	6.60*** (1118.74)	.38
INS	591	2.11 (0.72)	598	1.91 (0.66)	5.17*** (1177.38)	.29
Oppositional Defiant Problems	591	1.40 (0.41)	596	1.34 (0.36)	2.89** (1170.01)	.15
Conduct Problems	591	1.16 (0.25)	596	1.06 (0.15)	7.88*** (967.39)	.48
ADHD symptoms	590	0.57 (0.48)	597	0.38 (0.39)	7.55*** (1136.00)	.43
Reactive Aggression	591	1.61 (0.45)	598	1.47 (0.38)	5.53*** (1156.02)	.34
Proactive Aggression	591	1.44 (0.36)	597	1.33 (0.32)	5.19*** (1186)	.32
Relational Aggression	590	1.77 (0.73)	598	1.81 (0.75)	−1.05 (1186)	.05
Fearlessness	587	1.83 (0.60)	595	1.75 (0.57)	2.17* (1180)	.14
Prosocial Peer Involvement	591	2.36 (0.43)	598	2.57 (0.36)	−8.76*** (1148.36)	−.53
Parent-rated variables
CU	821	1.47 (0.46)	823	1.40 (0.41)	3.38** (1617.97)	.16
Oppositional Defiant Problems	822	1.50 (0.39)	821	1.49 (0.38)	0.29 (1641.00)	.03
Conduct Problems	823	1.11 (0.13)	822	1.07 (0.09)	6.43*** (1493.96)	.36
ADHD symptoms	823	0.65 (0.38)	823	0.53 (0.35)	6.81*** (1629.83)	.33
Relational Aggression	820	1.59 (0.47)	822	1.67 (0.50)	−3.33** (1640.00)	−.16
Fearlessness	823	1.86 (0.54)	823	1.76 (0.54)	3.68*** (1644.00)	.15

Note. CPTI = Child Problematic Traits Inventory; GD = Grandiose–Deceitful factor; CU = Callous–Unemotional factor; INS = Impulsive–Need for stimulation factor; ADHD = attention-deficit/hyperactivity disorder.

Degrees of freedom with decimals shows t test adjusted for that boys and girls did not exhibit equal variance on the tested variable.

p < .05. **p < .01. ***p < .001.

Factor Structure of the CPTI

Table 2 shows that, compared with the one-factor model, the three-factor model better fits the data and it fits both boys and girls to an acceptable degree. To test whether the three-factor model was invariant across gender, a multigroup CFA with gender as the grouping variable was conducted. Results showed that the three-factor model fitted to a satisfactory degree across gender (RMSEA = .07; CFI = .95; TLI = .95). Factor loadings of the 28 CPTI items derived from the CFA on the total sample are presented in Table 3. As shown in this table, all items loaded well and statistically significant on the expected CPTI factor and on the latent psychopathy construct.³

Table 2.

Fit Indices for One- and Three-Factor Models.

	RMSEA	CFI	TLI
One-factor model
Total sample	.13	.87	.86
Boys	.13	.87	.86
Girls	.12	.87	.86
Three-factor model
Total sample	.08	.95	.94
Boys	.08	.95	.95
Girls	.07	.95	.95

Note. RMSEA = root mean square error of approximation; CFI = comparative fit index; TLI = Tucker–Lewis index.

Table 3.

Standardized Item Loadings of the Three-Factor Solution in the Total Sample as Shown via a Confirmatory Factor Analysis.

Item	GD	CU	INS
5—Lies often to avoid problems	.87
7—Seems to see himself or herself as superior compared with others	.84
9—Often lies to get what he or she wants	.91
15—Seems to lie more than other children in the same age	.90
18—Is often superior and arrogant toward others	.89
21—To get people to do what he or she wants, he or she often find it efficient to con them	.81
24—Thinks that he or she is better than everyone on almost everything	.78
26—To frequently lie seems to be completely normal for him or her	.93
2—Seldom expresses sympathy for others		.86
4—Usually does not seem to share others’ joy and sorrow		.85
8—Never seems to have bad conscience for things that he or she has done		.83
11—Often seems to be completely indifferent when other children are upset		.84
13—Does not become upset when others are being hurt		.88
17—Seldom remorseful when he or she has done something not allowed		.88
20—Often does not seem to care about what other people feel and think		.93
22—Sometimes seems to completely lack the capability to feel guilt and remorse		.92
25—Never expresses feelings of guilt when he or she has done something not allowed		.91
27—Does not express guilt and remorse to the same extent as other children of the same age		.90
1—Likes change and that things happen all the time			.53
3—Often has difficulties with awaiting his or her turn			.90
6—Seems to do certain things just for the thrill of it			.71
10—Provides himself or herself with different things very fast and eagerly			.83
12—Often does things without thinking ahead			.86
14—Often consumes things immediately rather than saving them			.75
16—Seems to have a great need for change and excitement			.80
19—Does not like waiting			.90
23—Seems to get bored quickly			.86
28—Quickly gets tired of things and wants new things to happen all the time			.86
Loading of factor with latent construct Psychopathic Personality	.91	.85	.78

Note. GD = Grandiose–Deceitful factor; CU = Callous–Unemotional factor; INS = Impulsive–Need for Stimulation factor.

Internal Consistency of the CPTI Scores

The αs for the CPTI total score and three factors were indicative of excellent internal consistency. Specifically, in the total sample, α for the CPTI total score was .96 (boys/girls: α = .96/.95), for the GD factor .90 (boys/girls: α = 89/.91), for the CU factor .94 (boys/girls: α = .95/.93), and for the INS factor .91 (boys/girls: α = .91/.91). Significant zero-order correlations (p < .001) were found between the three CPTI factor scores and the CPTI total score as well as between the three CPTI factor scores. In the total sample, these correlations were r_GD-total = .84; r_CU-total = .86; r_INS-total = .88; r_GD-CU = .66; r_GD-INS = .61; and r_CU-INS = 57. After controlling for sociodemographic characteristics, these correlations remained substantially similar (rs = .84, .86. .88, .66, .60, and .55, respectively).

Validity of the CPTI

Teacher-Rated CPTI Scores in Relation to Teacher-Rated Criterion Measures

At the zero-order level, the CPTI total and factor scores were significantly positively related to teacher-rated oppositional defiant problems, conduct problems, ADHD symptoms, RA, PA, relational aggression, and fearlessness, and negatively related to prosocial peer involvement. These correlations were most often strong to moderate at least (Table 4). These results remained substantially similar after controlling for sociodemographic characteristics (Table 4).

Table 4.

Zero-Order and Partial Correlations Between Teacher-Rated CPTI Scores and Teacher-Rated Variables.

	CPTI Total		GD			CU			INS
Teacher-rated variables	Z-O	Part. 1	Z-O	Part. 1	Part. 2	Z-O	Part. 1	Part. 2	Z-O	Part. 1	Part. 2
Oppositional Defiant Problems	.64***	.63***	.60***	.59***	.31***	.51***	.51***	.13***	.55***	.54***	.25***
Conduct Problems	.68***	.67***	.65***	.65***	.36***	.65***	.63***	.34***	.49***	.47***	.05
ADHD symptoms	.69***	.67***	.48***	.46***	−.03	.53***	.51***	.19***	.71***	.70***	.57***
Reactive Aggression	.65***	.63***	.61***	.61***	.33***	.51***	.50***	.09	.56***	.55***	.26***
Proactive Aggression	.70***	.69***	.72***	.72***	.50***	.55***	.53***	.06	.59***	.57***	.23***
Relational Aggression	.58***	.59***	.64***	.64***	.46***	.44***	.45***	.01	.47***	.47***	.13***
Fearlessness	.45***	.45***	.31***	.31***	−.02	.35***	.35***	.13***	.45***	.46***	.32***
Prosocial Peer Involvement	−.56***	−.53***	−.43***	−.42***	−.02	−.66***	−.65***	−.54***	−.35***	−.32***	.05

Note. CPTI = Child Problematic Traits Inventory; GD = Grandiose–Deceitful factor of the CPTI; CU = Callous–Unemotional factor of the CPTI; INS = Impulsive–Need for stimulation factor of the CPTI; Z-O = zero-order correlations; ADHD = attention-deficit/hyperactivity disorder; Part. 1 = partial correlations partialling out parent’s socioeconomic status, parent’s origin, and child’s gender; Part. 2 = Part. 1 and also partialling out the other two CPTI factors.

p < .05. **p < .01. ***p < .001.

Table 4 also presents partial correlations between each CPTI factor and criterion measures, after controlling for the sociodemographic characteristics and the other two CPTI factors. Overall, the partial correlations were clearly, as hypothesized, lower than the zero-order correlations and for some variables were nonsignificant. Specifically, while all three CPTI factor scores were associated with oppositional defiant problems, only GD and CU scores, score remained significantly related to conduct problems. Furthermore, the INS score remained strongly related to ADHD symptoms after accounting for the shared variance between the three factors, while the other two CPTI factor scores became weakly (CU) or not significantly (GD) related to ADHD symptoms. The CU score was no longer significantly related to aggression, but was the only CPTI factor score that remained significantly and negatively related to prosocial peer involvement. Finally, the CU and INS scores remained significantly related to fearlessness, whereas the GD score did not.⁴

Teacher-Rated CPTI Scores in Relation to Parent-Rated Criterion Measures

At the zero-order level, the teacher-rated CPTI total and factor scores were weakly but significantly positively related to all parent-rated variables presented in Table 5, being oppositional defiant problems, conduct problems, ADHD symptoms, relational aggression, and fearlessness. These results remained substantially similar after controlling for sociodemographic characteristics (Table 5). After also controlling for the shared variance among the teacher-rated CPTI factor scores, the partial correlations were generally lower than the zero-order correlations, with several correlation coefficients being nonsignificant. This was especially true for the teacher-rated CPTI GD factor.

Table 5.

Zero-Order and Partial Correlations Between Teacher-Rated CPTI Scores and Parent-Rated Variables.

	CPTI Total		GD			CU			INS
Parent-rated variables	Z-O	Part. 1	Z-O	Part. 1	Part. 2	Z-O	Part. 1	Part. 2	Z-O	Part. 1	Part. 2
Oppositional Defiant Problems	.25***	.25***	.22***	.21***	.06	.23***	.23***	.10**	.20***	.20***	.07*
Conduct Problems	.30***	.28***	.25***	.23***	.05	.29***	.26***	.12***	.25***	.23***	.08**
ADHD symptoms	27***	.24***	16***	.14***	−.06*	.22***	.19***	.07*	.29***	.27***	.21***
Relational Aggression	.13***	.15***	.13***	.14***	.06	.09**	.11***	.01	.12***	.14***	.06*
Fearlessness	.16***	.15***	.10**	.09**	−.01	.09**	.08*	−.03	.21***	.19***	.17***

Note. CPTI = Child Problematic Traits Inventory; GD = Grandiose–Deceitful factor; CU = Callous–Unemotional factor; INS = Impulsive–Need for Stimulation factor; Z-O = zero-order correlations; Part. 1 = partial correlations partialling out parent’s socioeconomic status, parent’s origin, and child’s gender; Part. 2 = partial correlations partialling out parent’s socioeconomic status, parent’s origin, child’s gender and the other two CPTI factors; all the zero-order and partial correlations were repeated keeping the mother- and father-rated scales separate but also with the principle of combining the two but choosing the highest score on each item from the mother and father, respectively (the highest score prevails method). All these three different sets of analyses yielded the same results as compared with the ones presented in this table. The same correlations were significant and deviations were very small when they existed at all.

p < .05. **p < .01. ***p < .001.

A comparison between Tables 4 and 5 shows that the correlations between the teacher-rated CPTI and parent-rated criterion measures were lower than the correlations between the teacher-rated CPTI scores and teacher-rated criterion measures and that the teacher-rated CPTI INS factor remained significantly related to parent-rated conduct problems, but not to teacher-rated conduct problems (partial correlations).

Discussion

This study aimed to further test the reliability and validity of the CPTI. In line with the only available previous study testing the validity of the CPTI (Colins, Andershed, et al., 2014), CFA indicated that the CPTI assesses a construct of psychopathic personality in early childhood that closely resembles how it is often conceptualized and assessed in adolescence and adulthood. The three-factor structure fit the data well among boys and girls, and better than a one-factor model including all items of the CPTI. The standardized factor loadings were consistently high (most of them in the .80 to .90 range), a feature that is quite unique in the literature on parent- or teacher-rated psychopathic traits (e.g., Dadds et al., 2005; Frick, Bodin, & Barry, 2000), and clearly suggests that the CPTI items coherently and distinctively assess three different but interrelated dimensions resembling the interpersonal, CU, and behavioral/lifestyle dimensions of the psychopathic personality construct. Of note, the model fit of the three-factor model was acceptable in terms of RMSEA, and excellent in terms of CFI and TLI. Although modification indices may suggest that allowing some items to correlate with each other, could improve the model fit terms of RMSEA, such modification indices may be sample specific. Because the other fit indices were in the excellent range, we therefore did not consider it as relevant or necessary to decrease the RMSEA.

The CPTI total and factor scores overall demonstrated excellent reliability in terms of internal consistency, with αs for the CPTI total and three-factor scores between .89 and .96. The excellent internal consistency of the CPTI CU score is particularly interesting in the context of prior CU-related research with young children. Indeed, the α for the CU score of the APSD parent and/or teacher report versions varies from low to acceptable (e.g., Dadds et al., 2005; Fite, Greening, Stoppelbein, & Fabiano, 2009; Kimonis et al., 2006). There is some evidence that the internal consistency of the ICU (Frick, 2003) in early childhood (Ezpeleta et al., 2013) is much better than APSD generated CU scores. Unfortunately, and for reasons detailed elsewhere (Hawes et al., 2014), the factor structure and item content of the ICU has been questioned (e.g., Lahey, 2014) and various alternative ICU versions are currently being proposed (Benesch, Görtz-Dorten, Breuer, & Döpfner, 2014; Feilhauer, Cima, & Arntz, 2012; Hawes et al., 2014). For now, it seems that the CPTI is responsive to previous calls for psychometrically sound and comprehensive assessments of psychopathic traits in children (e.g., Hawes et al., 2014; Waller et al., 2014). However, all available support up to date for the CPTI stems from research with Swedish samples, and these findings need to be replicated in studies conducted in other countries and languages. Studies are on their way that will report on the psychometric properties of the Italian, Dutch, Cypriot, and German versions of the CPTI.

By including both teacher and parent ratings, the present study was responsive to prior calls to use multiple informants when scrutinizing the convergent validity of psychopathy measures (e.g., Assary, Salekin, & Barker, 2014). CPTI scores showed the expected relations with teacher- and parent-rated external criterion measures, even after controlling for sociodemographic characteristics. As such, the current study provides novel evidence in support for the convergent validity of teacher-rated CPTI scores. In line with prior work on psychopathic traits (e.g., Waller et al., 2014), the associations between the teacher-rated CPTI scores and criterion measures were weaker when parents (vs. teachers) were used. Crucial, though, is that the associations between all teacher-rated CPTI scores and parent-ratings were still significant at the zero-order level and after controlling for sociodemographic characteristics. This finding is encouraging given recent work with preschoolers showing that teacher-rated CU traits were not related to parent-rated conduct problems, oppositional defiant problems symptoms, prosocial behavior, and temperament (Ezpeleta et al., 2013).

Although most researchers do agree theoretically that psychopathy is a syndrome consisting of traits that co-occur in the same person, their analyses are predominantly performed on the separate factors (e.g., Colins et al., 2017; Salekin, 2015). Overall, this study showed that each single CPTI factor was substantially less strongly correlated to external criterion measures after controlling for the other two CPTI factors than was the case at the zero-order level. This set of findings, together with the CPTI total score being significantly related to all criterion measures, suggests that the relation between the different CPTI factors and criterion measures in part lies in the combination of the three CPTI factors. Due to multicollinearity issues in the present data set, we were not able to strictly test if the three-way interaction between the CPT factors indeed was more strongly related to criterion measures than any dimension on its own (Colins, Andershed, et al., 2014). An alternative strategy to test this combined effect of all CPTI factors, but that was beyond the scope of this study, would be to identify children who are high on all CPTI factors and compare them with children with other permutations of psychopathic traits (e.g., high on CU traits and relatively low on the other dimensions; Colins et al., 2017). Supportive results would provide evidence that the CPTI not only reliably and validly assesses psychopathy traits but also that it is possible to identify preschoolers with a psychopathic personality, at least for research purposes. The psychopathic personality construct should not yet be used in early childhood for clinical purposes, especially not because stability studies that span different developmental periods in life are still rare and it is still not really clear at what level these traits might be indicative that development goes awry.⁵ Put differently, the CPTI is at present a research tool and should only be used as such.

Our findings revealed that each of the three CPTI factors—after removing their shared variance—is uniquely related to criterion measures of interest. These findings are particularly interesting in light of an increasing number of studies that assess or study CU traits without reference to the other psychopathic traits. It has been recognized that such a methodology cannot provide information as to whether it is the CU construct or the other psychopathy dimensions that account for the results (Frick et al., 2014a). Our study was able to provide this information, and showed that, after controlling for the other two CPTI factors, the CU factor remained significantly related to most of the teacher- and parent-rated criterion measures although most often not as strong as the other CPTI factors. Of note, CU traits was uniquely related to variables associated with behavioral problems, including conduct problems, and fearlessness, a finding that supports the inclusion of a CU-based specifier for conduct disorder in DSM-5, and theories linking fearlessness to the development of CU traits (e.g., Frick et al., 2014b). Also, the CPTI CU factor was the only dimension that was uniquely related to prosocial peer involvement, a finding that dovetails well with evidence that CU traits are associated with greater impairment in peer relationships and peer problems (e.g., Essau, Sasagawa, & Frick, 2006). Importantly, and after controlling for the other CPTI factors, only the CU factor was no longer significantly related to teacher-rated proactive, reactive, and relational aggression. This suggests that the association between CU traits and aggressive behavior identified by prior work might be due to the shared variance with other psychopathic traits. The finding that the other psychopathy dimensions account for the relation between CU traits and aggression, also converges with evidence that interpersonal traits are uniquely related to bullying, relational aggression, and delinquency, sometimes even stronger than CU traits as measured by various tools other than the CPTI (e.g., Colins et al., 2012; Lau & Marsee, 2013; Stellwagen & Kerig, 2013; Vahl et al., 2014). These findings are in need of further replication, especially because studies also showed that none of the three psychopathy factors were related to aggression after controlling for their shared variance (e.g., Marsee et al., 2005). Therefore, it would be beneficial for researchers to report on all factor scores of psychopathy measures as this would allow researchers to examine the unique relation between psychopathy factors and various correlates of psychopathy (see also Salekin, 2015).

Having a tool like the CPTI that provides a reliable and valid assessment of psychopathic traits in preschoolers may advance research on psychopathic traits and psychopathic personality in various ways. First, it may help answering the intriguing question of how many preschoolers with high levels of interpersonal, CU, and behavioral psychopathic traits will remain high on all these dimensions and become our tomorrow’s adult psychopaths. This kind of research is warranted to empirically test the definition of psychopathic personality being a developmental disorder that is present by the age of 8 years (e.g., Blair, 2010). Second, and to the extent that future research will be able to identify preschoolers with a so-called psychopathic personality, the CPTI may help the further validation of differentiating between primary and secondary variants of psychopathy (e.g., Kimonis, Frick, Cauffman, Goldweber, & Skeem, 2012; Skeem, Johansson, Andershed, Kerr, & Louden, 2007). Secondary variants are hypothesized to acquire psychopathy as a consequence of negative environmental influences early in life (e.g., maltreatment), and primary variants mainly as a consequence of genetic vulnerabilities (Skeem et al., 2007). Third, the CPTI may also help study how certain psychopathic traits develop in early childhood. Research recently showed, for example, that parenting might not only be critical in understanding the development of CU traits (e.g., Frick et al., 2014a) but also in unraveling the development of grandiose and manipulative traits (e.g., Horton & Tritch, 2014). Such novel findings may be particularly relevant and informative for developmental theories on psychopathic personality that currently do not account for the onset and development of interpersonal traits (e.g., Blair, 2013). Fourth, the CPTI will also allow to test and inform competing developmental theories to understand the development of serious conduct problems. Although current theories put much emphasis on CU traits as developmental precursor of sever conduct problems (e.g., Frick et al., 2014b), it has been argued and demonstrated that the combination of all three psychopathy dimensions is a better predictor of conduct problems in childhood and psychopathy in adulthood, than one dimension on its own, including CU traits (Colins, Andershed, et al., 2014). As mentioned earlier, we could not test this in the present study but testing this hypothesis in different samples and settings is relevant given that only the CU dimension is in the DSM-5 to identify a severe group of children with conduct disorder.

The present study has several strengths, including a relatively large community-based sample of preschoolers (with an equal proportion of boys and girls), the use of multiple informants (which enabled us to avoid shared method variance in several analyses), and the use of well-validated and commonly used questionnaires. As always, the findings should also be viewed with some limitations in mind. A first limitation is the cross-sectional design of the study, which does not allow testing the usefulness of psychopathic personality traits to predict future problems. Second, we did not include alternative measures to assess psychopathic traits in early childhood, which imply that we do not know how much the CPTI overlaps with other tools that were designed for use or are being used in early childhood. Third, this study relied on twin data, suggesting that there might be some dependence in the data. However, and as mentioned in a footnote, all results presented in this study were replicated in two randomly selected groups, and support for the CPTI’s factor structure was also confirmed by a multilevel CFA that takes into account the correlated twin data. Fourth, the lack of parent ratings for some criterion measures is a limitation. Future studies that address this issue and even include other sources of information (e.g., observations during play) are warranted when testing the validity of teacher-rated CPTI scores, especially because this would help see to what extent validity of CPTI scores is influenced by shared method variance. Fifth, given the modest response rate (see participant section) and given that twin study’s finding may not be generalizable to singletons, it is not possible to make firm conclusions about the generalizability of our findings to the general Swedish population.

In conclusion, this study replicated and substantially extended prior work on the psychometric properties of the CPTI by replicating the three-factor structure, the internal consistency, as well as the external validity of the CPTI total and factor scores. Also, the present study’s findings suggest that the psychometric qualities of the CPTI are not affected by gender, age, origin, and SES. Only two studies so far, including the present one, have tested the CPTI and some issues seem to be present, such as the less than perfect fit of the factor structure in terms of RMSEA. Also, more research is clearly needed to investigate the usefulness of the CPTI in other countries, cultures, settings, and ages during childhood, while also using parents as CPTI raters. Altogether, and for now, the CPTI hold promise as a teacher-rated tool for assessing psychopathic traits in childhood and may further our current understanding of psychopathic personality as a developmental disorder.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a grant from the Swedish Research Council.

Notes

References

Achenbach

T. M.

Rescorla

L. A.

(2001). Manual for the ASEBA school-age forms & profiles. Burlington: University of Vermont, Research Center for Children, Youth, & Families.

Andershed

(2010). Stability and change of psychopathic-traits. What do we know? In Salekin

R. T.

Lynam

D. R.

(Eds.), Handbook of child and adolescent psychopathy (pp. 233-250). New York, NY: Guilford Press.

Andershed

Kerr

Stattin

Levander

(2002). Psychopathic traits in non-referred youths: Initial test of a new assessment tool. In Blaauw

Sheridan

(Eds.), Psychopaths: Current international perspectives (pp. 131-158). The Hague, Netherlands: Elsevier.

Anestis

J. C.

Caron

K. M.

Carbonell

J. L.

(2011). Examining the impact of gender on the factor structure of the Psychopathic Personality Inventory–Revised. Assessment, 18, 340-349.

Arseneault

Kim-Cohen

Taylor

Caspi

Moffitt

T. E.

(2005). Psychometric evaluation of 5- and 7-year-old children’s self-reports of conduct problems. Journal of Abnormal Child Psychology, 33, 537-550.

Assary

Salekin

R. T.

Barker

E. D.

(2014). Big-Five and Callous-unemotional traits in preschoolers. Journal of Psychopathology and Behavioral Assessment, 37, 371-379.

Barker

Pistran

Elliot

(1994). Research methods in clinical and counselling psychology. Chichester, England: Wiley.

Benesch

Görtz-Dorten

Breuer

Döpfner

(2014). Assessment of Callous-Unemotional Traits in 6 to 12 year-old children with oppositional defiant disorder/conduct disorder by parent ratings. Journal of Psychopathology and Behavioral Assessment, 36, 519-529.

Birch

S. H.

Ladd

G. W.

(1998). Children’s interpersonal behaviors and the teacher–child relationship. Developmental Psychology, 34, 934-946.

10.

Blair

(2010). A cognitive neuroscience perspective on child and adolescent psychopathy. In Salekin

R. T.

Lynam

D. R.

(Eds.), Handbook of child and adolescent psychopathy (pp. 156-178). New York, NY: Guilford Press.

11.

Blair

J. R.

(2013). The neurobiology of psychopathic traits in youths. Nature Reviews Neuroscience, 14, 786-799.

12.

Blais

Solodukhin

Forth

A. E.

(2014). A meta-analysis exploring the relationship between psychopathy and instrumental versus reactive violence. Criminal Justice and Behavior. doi:10.1177/0093854813519629

13.

Brown

Atkins

M. S.

Osborne

M. L.

Milnamow

(1996). A revised teacher rating scale for reactive and proactive aggression. Journal of Abnormal Child Psychology, 24, 473-480.

14.

Byrd

Kahn

Pardini

(2013). A validation of the Inventory of Callous-Unemotional Traits in a community sample of young adult males. Journal of Psychopathology and Behavioral Assessment, 35, 20-34.

15.

Campbell

S. B.

Doucette

N. L.

French

(2009). Validity and stability of the youth psychopathic traits inventory in a nonforensic sample of young adults. Journal of Personality Assessment, 91, 584-592.

16.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

17.

Colins

Andershed

Frogner

Lopez-Romero

Veen

Andershed

A.-K.

(2014). A new measure to assess psychopathic personality in children: The Child Problematic Traits Inventory. Journal of Psychopathology and Behavioral Assessment, 36, 4-21.

18.

Colins

Fanti

Salekin

Andershed

(2017). Psychopathic personality in the general population: Differences and similarities across gender. Journal of Personality Disorders, 31, 49–74. doi:10.1521/pedi_2016_30_237

19.

Colins

Noom

Vanderplasschen

(2012). Youth Psychopathic Traits Inventory-Short Version: A further test of the internal consistency and criterion validity. Journal of Psychopathology and Behaviorial Assessment, 34, 476-486. doi:10.1007/s10862-012-9299-0

20.

Colins

O. F.

Andershed

Pardini

D. A.

(2015). Psychopathic traits as predictors of future criminality, intimate partner aggression, and substance use in young adult men. Law and Human Behavior, 39, 547-548. doi:10.1037/lhb0000148

21.

Colins

O. F.

Bijttebier

Broekaert

Andershed

(2014). Psychopathic-like traits among detained female adolescents reliability and validity of the antisocial process screening device and the youth psychopathic traits inventory. Assessment, 21, 195-209.

22.

Cooke

D. J.

Michie

(2001). Refining the construct of psychopathy: Towards a hierarchical model. Psychological Assessment, 13, 171-188. doi:10.1037//1040-3590.13.2.171

23.

Crick

N. R.

Casas

J. F.

Mosher

(1997). Relational and overt aggression in preschool. Developmental Psychology, 33, 579-588.

24.

Czar

K. A.

Dahlen

E. R.

Bullock

E. E.

Nicholson

B. C.

(2011). Psychopathic personality traits in relational aggression among young adults. Aggressive Behavior, 37, 207-214. doi:10.1002/ab.20381

25.

Dadds

M. R.

Fraser

Frost

Hawes

D. J.

(2005). Disentangling the underlying dimensions of psychopathy and conduct problems in childhood: A community study. Journal of Consulting and Clinical Psychology, 73, 400-410.

26.

DuPaul

G. J.

Power

T. J.

McGoey

K. E.

Ikeda

M. J.

Anastopoulos

A. D.

(1998). Reliability and validity of parent and teacher ratings of attention-deficit/hyperactivity disorder symptoms. Journal of Psychoeducational Assessment, 16, 55-68.

27.

Essau

C. A.

Sasagawa

Frick

P. J.

(2006). Callous-unemotional traits in a community sample of adolescents. Assessment, 13, 454-469.

28.

Ezpeleta

Osa

N. d. l.

Granero

Penelo

Domènech

J. M.

(2013). Inventory of callous-unemotional traits in a community sample of preschoolers. Journal of Clinical Child and Adolescent Psychology, 42, 91-105.

29.

Fanti

K. A.

Centifanti

L. C. M.

(2014). Childhood callous-unemotional traits moderate the relation between parenting distress and conduct problems over time. Child Psychiatry and Human Development, 45, 173-184.

30.

Farrington

D. P.

Ullrich

Salekin

R. T.

(2010). Environmental influences on child and adolescent psychopathy. In Salekin

R. T.

Lynam

D. R.

(Eds.), Handbook of child and adolescent psychopathy (pp. 202-230). New York, NY: Guilford Press.

31.

Feilhauer

Cima

Arntz

(2012). Assessing callous-unemotional traits across different groups of youths: Further cross-cultural validation of the Inventory of Callous-Unemotional traits. International Journal of Law and Psychiatry, 35, 251-262.

32.

Fite

P. J.

Greening

Stoppelbein

Fabiano

G. A.

(2009). Confirmatory factor analysis of the antisocial process screening device with a clinical inpatient population. Assessment, 16, 103-114.

33.

Flora

D. B.

Curran

P. J.

(2004). An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data. Psychological Methods, 9, 466-491.

34.

Frick

P. J.

(2003). The Inventory of Callous-Unemotional Traits. Unpublished rating scale, University of New Orleans, New Orleans, Louisiana.

35.

Frick

P. J.

Bodin

S. G.

Barry

C. T.

(2000). Psychopathic traits and conduct problems in community and clinic-referred samples of children: Further development of the psychopathy screening device. Psychological Assessment, 12, 382-393.

36.

Frick

P. J.

Hare

R. D.

(2001). Antisocial process screening device technical manual. Toronto, Ontario, Canada: Multi-Health Systems.

37.

Frick

P. J.

Ray

Thornton

Kahn

R. E.

(2014a). Can callous-unemotional traits enhance the understanding, diagnosis, and treatment of serious conduct problems in children and adolescents? A comprehensive review. Psychological Bulletin, 140, 1-57.

38.

Frick

P. J.

Ray

J. V.

Thornton

L. C.

Kahn

R. E.

(2014b). Annual research review: A developmental psychopathology approach to understanding callous-unemotional traits in children and adolescents with serious conduct problems. Journal of Child Psychology and Psychiatry, 55, 532-548.

39.

Hare

R. D.

(2003). Hare Psychopathy Checklist–Revised (PCL-R). Toronto, Ontario, Canada: Multi-Health Systems.

40.

Hare

R. D.

Neumann

C. S.

(2008). Psychopathy as a clinical and empirical construct. Annual Review of Clinical Psychology, 4, 217-246.

41.

Hawes

S. W.

Byrd

A. L.

Henderson

C. E.

Gazda

R. L.

Burke

J. D.

Loeber

Pardini

D. A.

(2014). Refining the parent-reported inventory of callous–unemotional traits in boys with conduct problems. Psychological Assessment, 26, 256-266.

42.

Horton

R. S.

Tritch

(2014). Clarifying the links between grandiose narcissism and parenting. Journal of Psychology: Interdisciplinary and Applied, 148, 133-143.

43.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1-55.

44.

Kimonis

E. R.

Frick

P. J.

Boris

N. W.

Smyke

A. T.

Cornell

A. H.

Farrel

J. M.

Zeanah

C. H.

(2006). Callous-unemotional features, behavioral inhibition, and parenting: Independent predictors of aggression in a high-risk preschool sample. Journal of Child and Family Studies, 15, 745-756.

45.

Kimonis

E. R.

Frick

P. J.

Cauffman

Goldweber

Skeem

J. L.

(2012). Primary and secondary variants of juvenile psychopathy differ in emotional processing. Development and Psychopathology, 24, 1091-1103.

46.

Ladd

G. W.

Profilet

S. M.

(1996). The Child Behavior Scale: A teacher-report measure of young children’s aggressive, withdrawn, and prosocial behaviors. Developmental Psychology, 32, 1008-1024.

47.

Lahey

(2014). What we need to know about callous-unemotional traits: Comment on Frick, Ray, Thornton, and Kahn (2014). Psychological Bulletin, 140, 58-63. doi:10.1037/a0033387

48.

Lau

K. S.

Marsee

M. A.

(2013). Exploring narcissism, psychopathy, and Machiavellianism in youth: Examination of associations with antisocial behavior and aggression. Journal of Child and Family Studies, 22, 355-367.

49.

Lynam

D. R.

Miller

D. J.

Vachon

Loeber

Stouthamer-Loeber

(2009). Psychopathy in adolescence predicts official reports of offending in adulthood. Youth Violence and Juvenile Justice, 7, 189-207.

50.

Marsee

M. A.

Silverthorn

Frick

P. J.

(2005). The association of psychopathic traits with aggression and delinquency in non-referred boys and girls. Behavioral Sciences & the Law, 23, 803-817.

51.

Miller

J. D.

Watts

Jones

S. E.

(2011). Does psychopathy manifest divergent relations with components of its nomological network depending on gender? Personality and Individual Differences, 50, 564-569.

52.

Muthén

L. K.

Muthén

B. O.

(2011). Mplus (6.12) [Computer software]. Los Angeles, CA: Author.

53.

Ostrov

J. M.

Keating

C. F.

(2004). Gender differences in preschool aggression during free play and structured interactions: An observational study. Social Development, 13, 255-277.

54.

Raine

(2013). The anatomy of violence: The biological roots of crime. New York, NY: Random House.

55.

Salekin

R. T.

(2008). Psychopathy and recidivism from mid-adolescence to young adulthood: Cumulating legal problems and limiting life opportunities. Journal of Abnormal Psychology, 17, 386-395.

56.

Salekin

R. T.

(2015). Psychopathy in childhood: Toward better informing the DSM-5 and ICD-11 conduct disorder specifiers. Personality Disorders: Theory, Research, and Treatment. Advance online publication. doi:10.1037/per0000150

57.

Salekin

R. T.

Lynam

D. R.

(2010). Handbook of child and adolescent psychopathy. New York, NY: Guilford Press.

58.

Scholte

van der Ploeg

(2007). The development of a rating scale to screen social and emotional detachment in children and adolescents. International Journal of Methods in Psychiatric Research, 16, 137-149.

59.

Skeem

J. L.

Cooke

D. J.

(2010). Is criminal behavior a central component of psychopathy? Conceptual directions for resolving the debate. Psychological Assessment, 22, 433-445.

60.

Skeem

J. L.

Johansson

Andershed

Kerr

Louden

J. E.

(2007). Two subtypes of psychopathic violent offenders that parallel primary and secondary variants. Journal of Abnormal Psychology, 116, 395-409.

61.

Stellwagen

K. K.

Kerig

P. K.

(2013). Ringleader bullying: Association with psychopathic narcissism and theory of mind among child psychiatric inpatients. Child Psychiatry and Human Development, 44, 612-620.

62.

Talwar

Lee

(2008). Social and cognitive correlates of children’s lying behavior. Child Development, 79, 866-881.

63.

Vahl

Colins

Markus

Lodewijkcs

Doreleijers

Vermeiren

(2014). Psychopathic-like traits in detained male adolescents: The usefulness of self-report outside a research context. European Child & Adolescent Psychiatry, 23, 691-699.

64.

van Baardewijk

Stegge

Andershed

Thomaes

Scholte

Vermeiren

(2008). Measuring psychopathic traits in children through self-report: The development of the Youth Psychopathic Traits Inventory-Child Version. International Journal of Law and Psychiatry, 31, 199-209.

65.

Waller

Gardner

Hyde

L. W.

(2013). What are the associations between parenting, callous–unemotional traits, and antisocial behavior in youth? A systematic review of evidence. Clinical Psychology Review, 33, 593-608.

66.

Waller

Wright

A. G.

Shaw

D. S.

Gardner

Dishion

T. J.

Wilson

M. N.

Hyde

L. W.

(2014). Factor structure and construct validity of the parent-reported Inventory of Callous-Unemotional Traits among high-risk 9-year-olds. Assessment, 22, 561-580. doi:10.1177/1073191114556101

67.

Willoughby

Waschbusch

Moore

Propper

(2011). Using the ASEBA to screen for Callous Unemotional Traits in early childhood: Factor structure, temporal stability, and utility. Journal of Psychopathology and Behavioral Assessment, 33, 19-30.