Perceptions of Social Conflicts among Incarcerated Adolescents with Callous-Unemotional Traits: “You’re Going to Pay. It’s Going to Hurt, but I Don’t Care.”

Abstract

Background

Delinquent youth with callous-unemotional (CU) traits may have a unique social-cognitive processing pattern that perpetuates their violent behavior. The current study examined the association between CU traits and the endorsement of deviant social goals during peer conflicts as well as expectancies and values regarding victim suffering following aggression.

Methods

Participants included 156 (84 males, 72 females) adjudicated juveniles residing at two gender-specific residential facilities in an urban city within the southeastern United States. The association between CU traits and participants' ratings of their social goals in hypothetic conflict situations and outcome expectancies/values regarding victim suffering were examined after controlling for prior violence, intelligence, and demographic covariates.

Results

CU traits were associated with an increased endorsement of social goals associated with revenge, dominance, and forced respect in social conflict situations. Adjudicated youth with CU traits were also less likely to endorse conflict avoidance and friendship building as important social goals when provoked by peers. There was no association between CU traits and expectations for victim suffering following aggression, but CU traits were significantly associated with lower levels of concern about victim suffering. These findings were significant after controlling for participants' prior history violence, intelligence, and demographic covariates.

Conclusions

Adjudicated youth with CU traits tend to emphasize power-oriented goals when provoked by peers and have little interest in rectifying social conflicts to build potential friendships with others. Juveniles with CU traits seem to be aware that their aggressive behavior will cause others to suffer, but they do not care when it does.

The presence of callous-unemotional (CU) traits may distinguish a group of delinquent youth with a particularly severe form of antisocial behavior with unique etiological origins (Frick & Dickens, 2006). Some evidence suggest that youth with CU traits may have a fairly unique social schema that serves to perpetuate their violent behavior over time (Pardini, Lochman, & Frick, 2003). In the most basic sense, schemas are defined as an individual’s consistent beliefs or ways of thinking about relationships with others that organize and transform interpretations and predictions about events in the world (Lochman & Pardini, 2008). It is generally thought that schemas can influence behavior by modifying certain social processing variables such as social goals during ambiguous conflicts, expectations regarding the consequences of aggression, and concerns about the possible outcomes of aggression. Emerging evidence suggests that delinquent adolescence with CU traits may exhibit a unique set of beliefs that emphasize aggression as an effective means for obtaining positive outcomes (Pardini et al., 2003). However, it is unclear whether antisocial youth with CU traits downplay the likelihood that their aggressive behavior will result in victim suffering versus simply not caring when it does. Studies have also not examined whether delinquent juveniles with CU traits endorse power-oriented social goals and minimize relationship building goals following minor peer provocation. It also remains unclear whether there are social-cognitive processing patterns that are unique to delinquent youth with CU traits or whether they are more generally associated with violent youth regardless of their level of CU traits.

Social Goals and Violent Behavior

Social goals represent the end state of the problem solving process that is important to understanding the motivating force behind an individual’s behaviors (Lochman, Wayland, & White, 1993). Reactions to relatively minor conflicts can be significantly influenced by the relative importance children place on various potential outcomes. For example, youth who emphasize the importance of de-escalating potential conflicts or working out misunderstandings with the goal of friendship building should be less likely to retaliate aggressively following relatively minor provocation from peers. In contrast, juveniles involved in conflicted social interactions who become fixated on goals such as vengeance, dominance, and forcing others show “respect” are likely to perpetually engage in acts of violence in order to achieve these goals.

Research has supported the importance of social goals in understanding aggressive behavior in children and adolescents. Youth with a dominance-oriented goal pattern in interpersonal conflicts tend to exhibit higher levels of aggression and criminal behavior than their peers (Lochman et al., 1993). Aggressive ADHD boys have been shown to endorse global social goals involving domination, disruption, and trouble-making more often than nonaggressive ADHD boys and normal comparison peers (Melnick & Hinshaw, 1996). Although we know of no published studies investigating the social goals of children with CU traits, one study found that childhood conduct problems were associated with the generation of fewer prosocial and more antisocial solutions to hypothetical social problems involving sharing and friendship making only for children with low CU traits (Waschbusch, Walsh, Andrade, King, & Carrey, 2007). However, what remains unclear is whether delinquent youth with CU traits downplay the importance of social goals associated with conflict avoidance and relationship building, while overemphasizing the importance of more power-oriented goals, when provoked by their peers.

Consequences of Aggression: Expectations and Concerns

Aggressive behavior in children and adolescents may be partially driven by an expectation that attacking others will result in more positive outcomes rather than negative outcomes. Studies have shown that aggressive children are more confident that the use of aggression will result in positive outcomes (Perry, Perry, & Rasmussen, 1986), and this may be particularly true of juveniles with high levels of CU traits (Pardini et al., 2003) and those exhibiting proactive aggression (Dodge, Lochman, Harnish, Bates, & Pettit, 1997). However, an issue that has not been sufficiently examined is whether delinquent youth with CU traits have particular difficulty anticipating the negative effects their aggressive behavior may have on others. It is possible that adolescents with CU traits are less likely to believe their aggressive actions will result in victim suffering, which serves to perpetuate continued violent behavior.

In the adult literature, a historically held assumption is that psychopathic offenders, who prototypically exhibit high levels of callous-unemotional traits, do not have difficulties recognizing the effect their behavior has on others (Cleckley, 1976). However, explicit tests of this assertion have produced somewhat mixed results. While several studies have found that male and female psychopathic offenders have difficulties recognizing distress cues (i.e., fear, sadness) in others (Blair et al., 2004; Eisenbarth, Alpers, Segre, Calogero, & Angrilli, 2008; Hastings, Tangney, & Stuewig, 2008), some studies report no emotion recognition deficits among adults with psychopathic traits (Dolan & Fullam, 2004; Glass & Newman, 2006). Using hypothetical vignettes, Dolan and Fullam (2004) found that both psychopathic and non-psychopathic offenders exhibit deficits in their ability to determine the impact that hurtful or insulting comments may have on others relative to healthy controls.

Similar to the adult literature, conflicting evidence has emerged related to the association between CU traits and anticipating the suffering of others in children and adolescents. One study found that children with psychopathic traits exhibit more aggression on a computerized game unless they are explicitly told their behavior is causing their victim distress (van Baardewijk, Stegge, Bushman, & Vermeiren, 2009), suggesting they may not fully appreciate the negative impact their behavior has on others. Although higher levels of psychopathic features have also been associated with a reduced ability to recognized distress cues in samples of children and adolescents (Blair, 1999; Blair, Budhani, Colledge, & Scott, 2005; Blair, Colledge, Murray, & Mitchell, 2001; Stevens, Charman, & Blair, 2001; Woodworth & Waschbusch, 2008), children with CU traits do not seem to have difficulties recognizing how a protagonist's actions may influence others in hypothetical vignettes (Woodworth & Waschbusch, 2008). In addition, Dadds and colleagues (2009) found that difficulties with emotional understanding were associated with a combined measure of CU traits and antisocial behavior in early childhood, but not in late childhood, suggesting that some of these problems may dissipate over time. However, this study suffered from several limitations, including the use of a normative sample, the failure to look at CU traits and antisocial behavior separately, and the use of a parent-report measure to assess emotional understanding skills in children.

Juvenile delinquents with elevated CU traits also tend to be less concern about the negative consequences of aggression (i.e., outcome values) compared to delinquents without CU traits (Pardini et al., 2003). Importantly, the value children and adolescents place on the consequences of aggression can be disparate from their expectations regarding the likelihood of the outcome occurring. For example, two youth may equally expect that their aggressive behavior would result in victim suffering, but they may differ in terms of the concern they have about the suffering of their victim when it occurs (Boldizar, Perry, & Perry, 1989). In this sense, outcome values about victim distress are consistent with conceptualizations of affective empathy, which have been associated with CU traits in children (van Baardewijk et al., 2008) and adolescents (Pardini et al., 2003). While there is some evidence that aggressive youth are less likely than their nonaggressive peers to be concerned about the feelings of others in aggressive conflicts (Hall, Herzberger, & Skowronski, 1998), it remains unclear if this is more pronounced or unique to antisocial youth with CU traits. One would anticipate that adjudicated youth with CU traits would care little about the suffering of others following aggressive acts, and this lack of concern would not be characteristic of violent youth after controlling for co-occurring CU traits.

Current Study

The primary focus of this study is to examine the social cognitive processing patterns associated with CU traits in adjudicated juveniles. First, the notion that adjudicated youth with elevated CU traits will endorse social goals associated with dominance and revenge and minimize social goals associate with conflict avoidance and friendship building when provoked by their peers will be examined. We will also examine whether there is a tendency for adjudicated juveniles with CU traits to underestimate the probability that their aggressive behavior will cause victim suffering, as well as whether juveniles with CU traits exhibit little concern when their aggressive behavior results in victim suffering. Because social-cognitive processing problems have been associated with violent behavior in children and adolescents in several previous studies (Dodge et al., 1997; Hall et al., 1998; Lochman & Dodge, 1994; Lochman et al., 1993; Melnick & Hinshaw, 1996; Perry et al., 1986), analyses will also examine whether CU traits are uniquely associated with a specific pattern of social information processing even after controlling for a history of violent behavior.

Method

Participants

This sample consisted of 156 (84 males, 72 females) adjudicated juveniles with complete data on all measures used in the current study residing at two gender-specific residential facilities in an urban city within the southeastern United States. The majority of the participants were African-American (n = 107, 69%), with the remainder being Caucasian (n = 46, 30%), Native American (n = 1, .6%), Hispanic (n = 1, .6%), and "Other" unspecified (n = 1, .6%). Most juveniles were in their mid teens (M = 15.83, SD = 1.30, range 11.34–18.71) and had been incarcerated for several months (M = 5.78, SD = 6.09). According to court records, participants tended to have several prior criminal charges (M = 7.87, SD = 5.01), with the first charge occurring in early adolescence (M = 12.79, SD = 1.93). Approximately one-third had been incarcerated at least one other time in the past (n = 47, 30%).

Procedure

Because juveniles residing at the two detention facilities were subjected to an initial 30-day orientation program, only adolescents who had been incarcerated for longer than 30 days were invited to participate in this study. Potential participants were read a consent form that described the basic procedures of the study, explained that participation was voluntary and that answers would be kept confidential. After the consent form was read, youth were allowed to ask questions about the study before agreeing to participate. All females offenders agreed to participate and only two of the male offenders declined participation. Written informed consent was obtained from all participants prior to data collection.

To minimize problems with low reading abilities that frequently occur with delinquent populations, all items were read to participants while they responded on their own questionnaires. Similar procedures for administering social cognitive measures have been successfully used in a court ordered treatment program for chronic violent offenders who had low intellectual abilities (Dodge et al., 1997) and outpatient psychiatric clinics (Matthys, Cuperus, & Van Engeland, 1999). After administration of the questionnaires, a detailed review of each participant’s file was used to gather demographic characteristics, criminal history information, and psychoeducational testing results.

Measures

Demographic and Background Variables

Information on the participant’s age, race, and gender was gathered from a demographic form included in the detention facility client files. For analysis purposes, race was coded as Caucasian (1) versus all other racial groups (0).

Intellectual abilities

Upon admission to the detention facility each juvenile was administered either the Wechsler Adult Intelligence Scale-3^rd Edition (WAIS-III; Weschler, 1997) or the Weschsler Intelligence Scale for Children – 3^rd Edition (WISC-III; Weschsler, 1991). To control for the potential confounding effects of IQ, each participant’s full scale IQ score was recorded.

Antisocial Processes Screening Device (APSD; Frick & Hare, 2001)

The 20-item rating scale was designed to measure psychopathic features in children and adolescents. The self-report version of the APSD has been used to identify subgroups of serious and persistent offenders in several previous studies with adolescents (for review see Frick & Dickens, 2006) and has show evidence of significant temporal stability from childhood to adolecence (Munoz & Frick, 2007). The callous-unemotional (CU) dimension was used in the current study, which consists of 6 items related to a lack of remorse and guilt, shallow and constricted affect, a lack of empathy, apathy toward school, and superficial charm (Frick, Obrien, Wootton, & Mcburnett, 1994). Participants rate each item on a 3-point scale (0 = “not true at all” to 2 = “definitely true"), with some items being reverse scored. As with prior studies (Frick & Dickens, 2006; Munoz & Frick, 2007), the internal consistency of the self-reported CU scale was relatively modest (α= .52).

Prior Violence

Two sources of information were used to index participants' history of violent behavior. First, participants were asked to report whether they had ever engaged in nine different violent acts included as part of a modified version of the Self-Report of Delinquency (SRD; Loeber, Farrington, Stouthamer-Loeber, & Van Kammen, 1998). Items included acts such as gang fighting, attacking someone to seriously hurt or kill them, robbery, and attacking school staff. The internal consistency of the scale was moderate (α = .70). The second source of violence history information was extracted from the official criminal record information kept in each participant's client file. Specifically, the number of prior violent charges the youth had received was collected, including simple assault, aggravated assault, robbery, murder, rape, and kidnapping. An index of prior violence was calculated for each participant by summing the total number of prior violent charges and the number of different violent acts they endorsed on SRD.

Social Goal Measure (SGM; Lochman et al., 1993)

This adaptation of the SGM consisted of 4 hypothetical vignettes designed to assess children’s social goals in benign conflict situations. Participants were presented with a depiction of a relatively minor provocation from a same-sex peer (e.g., kid laughs after you trip accidentally) and are then asked to indicate the importance of various social goals on a four point scale (1 = “not important” to 4 = “very important”). The five goals assessed were avoiding conflict (i.e., “avoid problems with him/her”), dominance (i.e., “let the guy/girl know who is in charge or who’s boss”), revenge (i.e., “get back at him/her”), forced respect (i.e., “make him show you some respect”), and reconciliation (“Work things out with him/her so you could possibly be friends”). Participants’ ratings were summed across the four stories to produce a total score for each possible social goal, with higher scores indicating greater importance being place on that particular goal. The internal consistency of the scales ranged from acceptable to good (αs = .61–.80).

After rating the five social goals, participants were asked to indicate which would be their primary goal for each vignette. A variable representing an overall tendency to focus on escalating conflict was created by summing the number of times youth reported a deviant primary goal (i.e., revenge, dominance, forced respect) and subtracting the number of times they reported prosocial goals (i.e., avoiding conflict, reconciliation).

Outcomes Expectations Questionnaire (OEQ; Pardini et al., 2003)

This version of the OEQ was adapted from a measure developed by Perry and colleagues (1989) and consists of eight brief vignettes designed to measure juveniles’ expectations that aggressive behavior against a same-sex peer would result in various outcomes. In the vignettes, participants imagined using aggressive behavior to obtain objects and retaliate against aversive actions from a hostile peer. After hearing each vignette, participants are asked to rate the likelihood that various outcomes would occur on a four-point Likert scale (1= “very sure the outcome would not occur” to 4 = “very sure the outcome would occur”). For each vignette participants were asked to rate the likelihood that their aggressive behavior would cause suffering in the victim. Items on the scale were summed, with higher scores indicating increased expectations that a particular outcome would occur. The internal consistency for the scale was moderate (α = .72). Studies using similar measures were able to discriminate antisocial youth from controls (Perry et al., 1990; Hall et al., 1998).

Outcomes Values Questionnaire (OVQ; Pardini et al., 2003)

This version of the OVQ was adapted from a measure developed by Perry and colleagues (1986) and includes eight brief vignettes designed to assess the values that children place on the outcomes of aggression against a same-sex peer. Although the stories in these vignettes were different that those used in the OEQ, the vignettes also described the use of aggression to obtain objects and retaliate against aversive peer actions. After each vignette, participants rate how much they would care if various outcomes occurred using a four-point Likert scale (1 = “not care at all” to 4 = “really care a lot”). Just like the OEQ, participants rated how much they would care if their behavior caused suffering in the peer. Items making up the scale were summed, with higher scores indicating increased importance being placed on the outcome. The internal consistency for the scale was high (α = .92). Similar measures have discriminated between aggressive and non-aggressive youth (Hall et al., 1998).

Results

Descriptive statistics and correlations for study variables are presented in Table 1. As expected, a moderate correlation was found between CU traits and prior violence. Both CU traits and prior violence were significantly correlated with all social goal outcomes, including a greater endorsement of goals related to dominance, revenge, forced respect, and overall conflict escalation, as well as lower endorsement of goals related to conflict avoidance and relationship building. While neither CU traits nor prior violence were significantly associated with expectations that aggressive acts would cause victim suffering, they both were significantly correlated with a decreased concern about the victim suffering as the result of aggression.

Table 1.

Correlations and Descriptive Statistics for Study Variables

		1	2	3	4	5	6	7	8	9	10	11	12	13	14
1. Age
2. Female		.36***
3. Caucasian		.23**	.22**
4. Full Scale IQ		.15	.21**	.46***
5. CU traits		−.14	−.11	−.05	−.08
6. Prior Violence		−.09	−.20*	−.21*	−.14	.33***
7. Goal: Conflict Avoidance		.05	.07	−.16*	−.26**	−.24**	−.31***
8. Goal: Dominance		−.01	.04	.05	−.08	.35***	.29***	−.30***
9. Goal: Revenge		.01	.02	.02	−.05	.31***	.27**	−.33***	.73***
10. Goal: Forced Respect		−.15	−.12	−.07	−.03	.25**	.25***	−.36***	.66***	.57***
11. Goal: Relationship Building		−.11	.00	.01	−.13	−.26**	−.16*	.42***	−.46***	−.51***	−.47***
12. Primary Goal: Conflict Escalation		.14	.01	.03	.11	.27***	.24*	−.47***	.52***	.56***	.60***	−.66***
13. Expectation: Victim Suffering		.05	−.10	.16*	.14	.08	.06	−.10	−.12	−.01	−.12	−.01	−.01
14. Value: Victim Suffering		.15	.17*	.08	.03	−.58***	−.32***	.25*	−.31***	−.29***	−.33***	.36***	−.30***	−12
M/%		15.83	46%	29%	78.94	5.59	5.99	9.32	9.40	8.88	10.92	7.83	2.34	14.49	17.88
SD/N		1.30	72	46	14.15	2.20	2.51	3.29	3.13	3.09	3.28	3.54	1.28	4.17	7.50

^*p < .05;

^**p < .01;

^***p < .001

Regression analysis was used to examine the association between CU traits and prior violence and each of the social goal variables after controlling for overlap between the constructs as well as potential confounds (i.e., gender, race, age, Full Scale IQ). Similar to correlation analysis, both CU traits and prior violence significantly predicted social goals related to conflict avoidance, dominance, revenge, forced respect, and overall conflict escalation (see Table 2). However, only CU traits significantly predicted the social goal of relationship building, with higher levels of CU traits predicting lower relationship building scores.

Table 2.

Callous-Unemotional Traits and Prior Violence Predicting Social Goals and Outcome Expectations/Values

	Predictors
	CU Traits			Prior Violence
Outcomes	B	SE	β	B	SE	β
Social Goals
Avoid Conflict	−0.24	0.12	−0.16*	−0.41	0.10	−0.31***
Dominance	0.41	0.11	0.28***	0.29	0.10	0.23**
Revenge	0.35	0.11	0.25**	0.27	0.10	0.20**
Forced Respect	0.26	0.12	0.17*	0.24	0.11	0.18*
Relationship Building	−0.41	0.13	−0.26**	−0.13	0.12	−0.09
Primary Goal: Conflict Escalation	0.29	0.09	0.24**	0.18	0.08	0.18*
Outcome Expectation
Victim Suffering	0.14	0.16	0.07	0.08	0.14	0.05
Outcome Values
Victim Suffering	−1.81	0.24	−0.53***	−0.38	0.21	−0.13

Note. All effects depicted are after controlling for gender, minority status, and full scale IQ. CU = callous-unemotional traits; B = unstandardized coefficient; β = standardized coefficient.

Analysis predicting outcome expectancies and outcome values related to victim suffering following aggression are presented in Table 2. While CU traits were not significantly related to expectations that aggressive behavior would result in victim suffering, they were significantly related to less concern about victim suffering following aggression. On the other hand, prior violence did not significantly contribute to expectations or values regarding victim suffering in the regression analysis after controlling for co-occurring CU traits and potential confounds.

Gender Differences

Because there are relatively few studies of social-cognitive processes in adjudicated girls, interactions between gender and both CU traits and prior violence in predicting all social cognitive variables were tested. Only one significant interaction was observed, namely an interaction between female gender and prior violence predicting the social goal of forced respect (B = .47, SE = .22, p < .05). Probing of this interaction revealed that there was a significant relation between prior violence and an increased endorsement of the social goal of forced respect for females (B = .52, SE = 17, p < .01), but this association was non-significant for males (B = .06, SE = .14, p = .70). No other significant interactions were found (all ps ≥ .10).

CU Traits by Prior Violence Interactions

To examine whether the associations between CU traits and the social-cognitive variables examined were more or less pronounced in juveniles with a history of violent behavior, the analyses were re-run to include an interaction between CU traits and prior violence. None of the interaction terms were statistically significant in these models (all ps > .42). This indicates that the associations between CU traits and the social cognitive variables observed did not differ for individuals with high versus low levels of prior violence.

Discussion

The current study was designed to examine the association between CU traits and adjudicated juveniles' social perceptions of peer conflicts after controlling for co-occurring violence. Findings indicated that juveniles with elevated CU traits are more likely to endorse social goals associated with revenge and dominance following minor provocation from peers, even after controlling for prior violent behavior. In addition, a relative lack of relationship building social goals following peer conflict seemed to be unique to juveniles exhibiting CU traits, rather than common to all violent delinquents. The findings also clearly illustrate that among adjudicated youth, those with CU traits do not seem to have particular difficulties judging whether their aggressive behavior will cause victim suffering. Instead, they simply do not care when it does.

Juveniles with CU traits generally seemed to focus on social goals aimed at escalating conflict with peers in order to assert a position of power. However, this social schema was also characteristic of juveniles with a prior history of violence even after controlling for levels of CU traits. This is consistent with studies indicating that aggressive and violent youth tend to endorse a dominance-oriented goal pattern more often than their peers (Lochman et al., 1993; Melnick & Hinshaw, 1996). The current study expands on these findings by demonstrating that juveniles who are high on both CU traits and violent behavior have a particularly deviant goal-set in conflict situations that may serve to perpetuate their violent behavior over time. Given that many of the social goals endorsed involve attempts at revenge and dominance, these juveniles may be prone to engage in premeditated forms of retaliatory violence following even slight provocation from peers. There was also a tendency for juveniles with CU traits to be particularly unlikely to endorse relationship building as a social goal in minor conflict situations, but this was not characteristic of violent offenders after controlling for co-occurring CU traits. Therefore, youth with CU traits tend to have little desire to develop meaningful relationships with others, especially those peers who provoke them in relatively minor ways.

While there has been some speculation that delinquents with CU traits may have difficulties understanding the consequences their behavior has on others (Dadds et al., 2009) and recognizing distress cues (Blair, 1999), the current results provided no evidence of such a deficit in terms of anticipating victim suffering. When delinquents were asked to rate how likely it was that their aggressive behavior would produce victim suffering across vignettes, those with higher CU traits did not exhibit a decreased expectation that this outcome would occur. While this runs counter to evidence that young children with CU traits may have difficulties understanding emotions in others, once children enter adolescence this deficit may dissipate (Dadds et al., 2009). Among the juvenile delinquents, there was also no evidence that violent offenders exhibited a reduced expectation that aggressive behavior would cause victim suffering. In contrast to findings regarding expectations, juvenile delinquents with CU traits exhibited an substantial decrease in concern about the suffering of their victims following acts of aggression. As expected, this lack of concern was unrelated to a history of violent offending after controlling for co-occurring CU traits. In sum, one of the core deficits in adjudicated juveniles with CU traits is that they care very little about the suffering of others, not that they have difficulty anticipating the extent to which their aggression will cause others to suffer.

Limitations

These findings need to be interpreted cautiously due to several limitations. First, the cross-sectional nature of the data makes it impossible to fully test models regarding the temporal ordering of CU traits, violent behavior, and the social cognitive variables measured. Also, most of the variables used in this study were assessed through self-report, which may have artificially inflated variable associations due to shared method variance. In contrast, some of the subscales used in this study had relatively low internal consistencies, so the significance tests reported should be viewed as conservative estimates. While the self-report scale of CU traits used in the current study has historically exhibited lower that desirable levels of internal consistency, it has been shown to predict serious antisocial behavior, program non-compliance, and recidivism in previous studies (Frick & Dickens, 2006). Another limitation is that this study was conducted solely with incarcerated adolescents, so the results cannot be generalized to youth exhibiting antisocial behavior in the community. This sampling scheme also prevented us from examining how delinquents and non-delinquents may differ on the social cognitive variables assessed in the study. For example, it is possible that delinquent youth are less likely to expect that aggressive behavior would cause victim suffering when compared to non-delinquent youth. Lastly, the current study did not assess relational aggression, which is postulated to be more relevant to understanding antisocial behavior in girls and may be associated with different social information processing deficits than overt violence (Crick, Grotpeter, & Bigbee, 2002). While the current study found that violent and callous-unemotional behavior in boys and girls are associated with similar types of maladaptive social information processing, future studies must examine whether this also applies to more relational forms of aggression.

Potential Clinical Implications

The presence of CU traits seems to designate a subgroup of delinquent youth who have a number of maladaptive social-cognitive processing characteristics. Although firm recommendations for dealing with juvenile offenders with CU traits can not be made based on the current findings, these juveniles may be particularly difficult to treat given their fixation on revenge, dominance, and forced respect following relatively minor conflict situations. Because youth with CU traits are less likely to be motivated to resolve conflicts as a way to develop and maintain relationships with others, therapeutic interventions may have to emphasize more self-serving reasons for not attacking others. It also seems that interventions designed to teach juvenile offenders to more fully understand the negative effect their aggression has on others will in all likelihood not be effective for juveniles with CU traits. Children with CU traits may be fully aware that their victims will suffer when they are attacked, but they do not care when it does. While getting these youth to care about the suffering of others is a critical element to their rehabilitation, treatments designed to instill feelings of concern for others may only be effective if implemented early in development when CU features are initially manifested.

Key Points

• CU traits may be useful for identifying adjudicated youth with a particular social-cognitive processing pattern underlying their antisocial behavior.

• Adjudicated adolescents with CU traits tend to endorse social goals associated with dominance and revenge in peer conflict situations, while downplaying conflict avoidance and friendship building as social goals.

• Adjudicated juveniles with CU traits did not appear to have difficulties judging whether their aggressive behavior would result in victim suffering, they simply did not care when it did.

• CU traits were associated with deviant social goals in conflict situations and reduced concern about the suffering of others, even after controlling for participants' prior history of violent behavior.

Abbreviations

CU

Callous-Unemotional