Abstract
Research shows psychopathic traits in adults are associated with perpetrating intimate partner violence (IPV). Less research, however, has examined this association in adolescents’ relationships. Our study examines whether adolescent psychopathic traits are associated with different forms of IPV and whether early adverse experiences of abuse and/or interparental violence might be implicated in this relationship. A mixed sample of 156 justice-involved and at-risk adolescents were assessed with the PCL:YV, participated in an early adversity interview assessing abuse and interparental violence, and completed IPV measures of physical and psychological aggressive acts. Psychopathic traits were associated with each IPV measure over and above early adversity. Psychopathic traits continued to be associated with psychological but not physical IPV after accounting for sample type. Furthermore, psychopathic traits indirectly linked interparental violence with psychological IPV, suggesting that psychopathic traits may play a role in the intergenerational perpetration of partner violence.
An increasing number of studies show that psychopathic traits (e.g., manipulation, callousness, sensation seeking, criminality) are associated with various forms of intimate partner violence (IPV) in adult populations (Forth et al., 2022; Robertson et al., 2020). Relationship behavior, however, including abusive behavior can begin to develop during the earliest romantic relationships, including during adolescence (Wolfe, 2006). Despite the potential importance of these formative early relationships, few studies have examined how psychopathic traits relate to IPV in adolescent populations (Robertson et al., 2020). There is also a need to consider how the link between psychopathic traits and early adversity (de Ruiter et al., 2022) could play a role in fostering abusive relationships. Those who experience abuse and/or witness violence between parents in their early relationships may go on to perpetrate violence in their own relationships (McKinney et al., 2009; Whitfield et al., 2003)—a process described as the cycle of violence (Widom, 1989, 2017). Because not all young people who experience early adversity go on to perpetrate violence themselves (Murphy & Blumenthal, 2000), individual differences like psychopathic traits may play an important mediating role in this process. Our goals in this study are to examine in adolescence (a) whether psychopathic traits are associated with different forms of IPV beyond the effects of early adversity and (b) whether psychopathic traits may play a mediating role in the association between early adversity and IPV perpetration.
Psychopathic Traits in Adolescence
Psychopathic traits in adolescence include interpersonal (e.g., manipulation, impression management), affective (e.g., callousness, lack of remorse), and behavioral features (e.g., impulsivity, sensation seeking) that collectively combine with antisocial features (e.g., poor anger control, criminal versatility) to reflect the prototypical personality of psychopathy (Forth et al., 2003; Neumann et al., 2007). The Psychopathy Checklist: Youth Version (PCL:YV; Forth et al., 2003) is the most common clinical measure used to assess adolescent psychopathy and has been shown to be associated with similar outcomes that the Psychopathy Checklist–Revised (PCL-R; Hare, 2003) shows in adults, including violence and recidivism (Forth & Book, 2010; Schmidt et al., 2011). Previous concerns that psychopathic traits might reflect normative changes in adolescent personality may be unfounded (e.g., Seagrave & Grisso, 2002). Although adolescence is a developmental period of increased adventurousness and experimentation (Hill et al., 2012), recent research shows that the personality traits underlying psychopathy themselves do not generally increase across adolescence but rather stay relatively stable (Lynam et al., 2009; Vachon et al., 2018). There also appears to be moderate stability of psychopathic traits from adolescence into adulthood (Hemphälä et al., 2015; Lynam et al., 2007).
Several studies show that psychopathic traits exist on a continuum in adults and adolescents (Murrie et al., 2007; Walters, 2014; Walters et al., 2007). Mean level differences are typically found across different populations including community, psychiatric, and justice-involved populations (Coid et al., 2009; Hare, 2003; Skeem & Mulvey, 2001), but these mean level differences do not capture qualitatively distinct populations of psychopathic versus nonpsychopathic individuals. Rather, the differences in psychopathic traits are a matter of degree rather than kind, ranging from very low to moderate to very high (Murrie et al., 2007). Hence, we take the approach as others have (Vitacco et al., 2010) of not labeling youth as a “psychopath” or not and instead refer to youth who display elevated or higher levels of psychopathic traits compared with those who display lower levels.
Psychopathic Traits and IPV
Because of the antisocial collection of traits underlying psychopathy, it has been examined for its role in IPV specifically (Humeny et al., 2021). A meta-analysis assessing 43 studies found that psychopathic traits were associated with IPV, with a medium effect size (Robertson et al., 2020). Some evidence suggests psychopathic traits increase the risk of premeditated rather than impulsive forms of IPV (Juodis et al., 2014; Stanford et al., 2008) and those higher in psychopathic traits may feel less concern for the harm caused to partners (Costa & Babcock, 2008; LaMotte et al., 2019). Most of this research, however, has been conducted with adult populations, suggesting the need to examine whether the same associations carry over to adolescents (Robertson et al., 2020). High levels of IPV can occur during adolescence (Halpern et al., 2001), yet adolescence also might be a time when relationship patterns are more malleable, which means IPV tendencies at these ages could be more subject to intervention (Wolfe et al., 2004).
Two studies that we are aware of have examined IPV and psychopathic traits assessed with the PCL:YV in adolescence, both using the Pathways to Desistance Study (Shaffer et al., 2021; Sweeten et al., 2016). These studies used person-centered analyses to examine IPV trajectories and found that adolescents categorized as perpetrating the highest amount of physical and psychological IPV in adolescence that subsequently declined into adulthood as well as those who engaged in a smaller number of physical and psychological IPV—but consistently over time—both scored higher on psychopathic traits compared with adolescents who did not perpetrate IPV. Another study that assessed psychopathic traits at age 16 with the Antisocial Process Screening Device (APSD; Frick & Hare, 2001) showed that adolescent psychopathic traits were predictive of IPV perpetration in young adulthood (Goodnight et al., 2017). This study, however, did not examine whether psychopathic traits were associated concurrently with IPV perpetration during adolescence. Overall, the limited data suggest psychopathic traits may influence IPV in adolescents as it does in adults. Studies have yet to compliment the person-centered analyses, however, with variable-centered analyses, both of which are important in research on psychopathy (Neumann et al., 2016).
Different Forms of IPV
As shown in the above studies using the Pathways study, IPV is not a unitary phenomenon but rather involves different forms of abuse, including sexual, physical, and psychological. Physical abuse can cause serious injuries as well as psychological harm to victims (Straus, 1979). Psychological abuse often involves subtler behaviors that may be more difficult to identify (Wolfe, 2006). Some forms of psychological abuse include dominating, isolating, or controlling partners into doing things they otherwise would not want to do; other forms include emotional, verbal, or manipulative abuse (Tolman, 1989). Understanding the individual factors that might impact perpetrating these different forms of partner abuse in adolescence is critical for identifying who may be at the greatest risk for continuing violence in future relationships (Jung et al., 2019). Although adult research suggests psychopathic traits may be a risk factor for multiple types of partner abuse (Robertson et al., 2020), it is important to examine whether psychopathic traits are associated with different forms of partner abuse in adolescence.
Early Adversity, IPV, and Their Association With Psychopathic Traits
Although adolescent psychopathic traits might capture one individual difference factor involved in perpetrating multiple forms of IPV, other factors have been shown to be important as well. Early adversity such as early abusive experiences and witnessing interparental violence are two such factors that may be important in predicting partner abuse (Herrenkohl et al., 2008), which is in line with the cycle of violence (Widom, 2017). Research has shown that early adverse experiences have an impact on relationship quality (reviewed in Zamir, 2022). Those who experience abuse and/or witness interparental violence may have their capacity to function in healthy relationships challenged (Herrenkohl et al., 2008). Experiencing early adversity has been shown to longitudinally predict poorer partner relationships (McLeod et al., 2014) and more divorce (Colman & Widom, 2004). There also is some evidence that early adversity may disproportionately affect men’s relationship functioning (Paradis & Boucher, 2010; Story et al., 2004), which may be important for psychopathy, where mean levels of psychopathic traits for boys and men tend to exceed those of girls and women (Beryl et al., 2014; Tsang et al., 2015). It is unclear whether psychopathic traits in adolescence might predict IPV beyond these other known risk factors for IPV.
Several studies have provided evidence of the intergenerational cycle of violence from early abuse and interparental violence to perpetrating IPV (Herrenkohl et al., 2004; Jung et al., 2019; McKinney et al., 2009; Whitfield et al., 2003). In addition to documenting the direct link between early adversity and later IPV, researchers have also examined mediating influences that may explain the link, especially considering that not all people who experience early adversity go on to perpetrate violence themselves (Kalmuss, 1984). Some mediating factors that may be important include trauma symptoms, interpersonal problems, and social information processing (DiLillo et al., 2009; Dodge et al., 1990; Murphy & Blumenthal, 2000). Thus, early adversity that includes abuse and witnessing interparental violence might affect the perpetration of later partner abuse via these (and possibly other) individual differences. We examined whether psychopathic traits may be one of these mediating factors.
At least two studies using adult samples can inform the possibility of a mediating link between early adversity and IPV via psychopathy (Robertson et al., 2020). Swogger et al. (2012) showed that higher levels of some psychopathic traits measured in adulthood combined with early abuse experiences were associated with an increased risk of engaging in IPV, suggesting that those with psychopathic traits might be at-risk of turning their own childhood abusive experiences into the abuse of partners. Hilton et al. (2019) found that men convicted of IPV charges scored higher in both psychopathy and adverse childhood experiences relative to those who had non-IPV-related offenses, suggesting again that partner abusers were both higher in psychopathy and had experienced more early adversity. No studies, however, have examined these associations in adolescent boys with psychopathic traits. Given the links between early adversity and psychopathic traits (de Ruiter et al., 2022) and the importance of mediating factors in the cycle of violence research (Dodge et al., 1990), we examine psychopathy as a mediating factor between early adversity and later perpetration of abuse in the present study.
The Present Study
Our study examines in a diverse sample of adolescent boys the associations between psychopathic traits as measured by the PCL:YV, early adversity, and different forms of IPV. As mentioned above, mean levels of psychopathic traits can range drastically across settings but scores on measures of psychopathic traits are continuous, suggesting that sampling from single settings could affect capturing a limited range of scores along the continuum. Thus, we combined two samples of youth to increase the range of psychopathic traits obtained. These samples included secure custody facilities where youth had been charged for a variety of offenses and community locations that had an overrepresentation of at-risk youth (e.g., employment centers). Based on our literature review, we posit two main hypotheses:
Given that early versus late adolescents may have had less versus more opportunities for engaging in romantic relationships, we included age as a covariate. We also considered sample type (custody vs. community) as a control variable in our analyses as youth who come from these different samples may have differential risks for IPV and/or early adversity.
Method
Participants
Adolescent boys (N = 156) between the ages of 13 and 19 (M = 17.44, SD = 1.17) participated in the study. Of these boys, 106 came from one of two secure custody facilities where youth were serving a sentence for committing offenses and 50 came from community centers where there was likely to be an overrepresentation of youth from at-risk backgrounds such as employment, recreation, and after-school centers. The subsamples were matched on age, whereas ethnicity was only measured in the secure custody subsample, which were White (80%), Indigenous (14%), Black (4%), and Asian (2%).
Measures
Psychopathic Traits and Early Adverse Experiences
Psychopathic traits were assessed with the PCL:YV (Forth et al., 2003), a 20-item expert-rating scale that captures the broad range of traits underlying the four-factor model of the psychopathy construct in youth. The PCL:YV was assessed using a semi-structured interview and file review. Each of the 20 items are rated on a 3-point scale (0 = item does not apply; 1 = item applies to a certain extent; 2 = item applies). The four facets include traits characterized as Interpersonal (e.g., manipulation), Affective (e.g., lack of remorse), Lifestyle (e.g., stimulation seeking), and Antisocial (e.g., poor anger control). Scores on the Interpersonal and Affective facets range from 0–8 and scores on the Lifestyle and Antisocial facets range from 0 to 10. The scale reliabilities were good for the facets (range α = .80–.88) and the total score (α = .93).
Childhood and adolescent experiences of abuse and interparental violence were taken from a semi-structured interview that assessed early adverse experiences of participants, which was separate from the PCL:YV interview. Both early experience variables were assessed with six items each. Abuse included whether and how often participants were physically abused by their parents and/or caregivers growing up. Specific behaviors were referenced and included questions such as, “When you were growing up did your mother, father, or caretaker ever hit you with a fist?” and, “Did your mother, father, or caretaker ever hit you with a belt, cord, hanger, wooden spoon, or any other object?” Items were scored by the interviewer on a 5-point scale, with the response options of 0 (no), 1 (only once or twice), 2 (occasionally, over a period of years from ____ to ____), 3 (occasionally, throughout childhood and adolescence), or 4 (often, throughout childhood and adolescence). All items were summed to create an overall score of abuse, which had acceptable scale reliability (α = .74). Interparental violence included whether and how often participants experienced their parents engage in verbal or physical conflict with one another. We inquired about specific conflictual behavior between parents, which included questions such as, “How often did your parents get into verbal fights?” and “Did your parents ever hit or punch each other?” Items were scored by the interviewer on the same 5-point scale as for abuse. All items were summed to create an overall score of interparental violence, which also had acceptable scale reliability (α = .75).
Intimate Partner Violence
Participants completed a shortened version of the Psychological Maltreatment of Women Inventory (PMWI; Tolman, 1989) to assess the frequency of psychological intimate partner abuse. The measure consists of 29 self-report items, responded on a scale ranging from 1 (never) to 5 (very frequently). Participants were asked to consider their behavior across all of their romantic relationships. Youth who had not yet had a romantic relationship were instructed to skip these questions (n = 1). The PMWI has two subscales, which included 15 items that were summed to form the Dominance-Isolation factor (DI; e.g., “I ordered my partner around”) and 14 items summed to form the Emotional-Verbal factor (EV; e.g., “I brought things up from the past to hurt her feelings”). The subscale reliabilities were α = .80 for DI and α = .89 for EV.
The Conflict Tactics Scale (CTS; Straus, 1979) was used to assess physical acts of violence toward an intimate partner. The CTS consists of 18 items that ask about frequency of using different acts (e.g., “Threw something at them” and “Beat them up”). We selected only the items that involved physical acts of violence, which included 13 items. For each act, participants chose among response options of 0 times, 1 time, 2 times, 5 times, 10 times, 20 times, or 21+ times. We summed across all items to capture an overall level of physical abuse reported across all romantic relationships. The overall scale reliability was acceptable (α = .72).
Procedure
Adolescents from secure custodies were recruited via notices placed at the institutions. The Ministry of Correctional Services provided consent for their participation and adolescents provided written assent to take part in the study. No incentives were offered for secure custody participants. Youth were told also that the information from the study would not be used in any way to affect their treatment or release decisions, nor that the institutional staff would have access to the information. Interviews and surveys were completed individually and in a private room in the institution with one of the research assistants.
Adolescents from the community were recruited via notices placed in locations where a higher proportion of at-risk youth could be obtained, such as community outreach centers and employment centers. For adolescents under 18 years old, written assent and parental consent were obtained and for those over 18 years old, written consent was obtained. Community participants completed the interview and surveys in a private room at the university and were compensated CAD$25. The research was approved by the Ontario Ministry of Corrections and the university’s research ethics board.
Interviews for both subsamples took approximately 90 to 180 minutes, which was followed by completing the self-report surveys. Due to scheduling limits, some of the secure custody adolescents completed the self-report surveys during a separate session, typically the next day. As part of the PCL:YV assessment, participants in the community sample nominated a collateral source (mostly a friend or family member).
Data Analysis
We first obtained descriptive statistics (mean, standard deviations, range) and bivariate correlations for the overall combined sample and then for the separate subsamples. We then conducted three hierarchical multiple regressions, one for each IPV measure as the outcome. Age and early adversity (i.e., abuse and interparental violence) were entered in Step 1 and total psychopathic traits were entered in Step 2. To control for sample, we included a dummy-coded variable of sample type in Step 3. Because psychopathic traits are best viewed on a continuum rather than categorical (Murrie et al., 2007), we determined that this approach was appropriate to (a) examine using a broad range of PCL:YV scores its relation to IPV in adolescence and (b) assess whether being recruited from either sample (custody vs. community) accounts for the variance explained by PCL:YV scores in the previous step. We then examined Hypothesis 2 by using atemporal mediation models of the relations between the early adversity measures of abuse and interparental violence and each of the three IPV outcomes through total psychopathic traits while controlling for sample type and age. Analyses were also conducted on the facets of the PCL:YV and are included in the Supplemental Material.
Results
Table 1 shows the descriptive statistics and bivariate correlations for the combined sample. Total psychopathic traits and the facets were associated with abuse, interparental violence, and each IPV measure across the combined sample. Table 2 shows descriptives and correlations for the separate subsamples, as well as the group comparisons on model variables. Although the custody sample scored significantly higher on all variables compared with the community sample (see bottom row and far right column in Table 2), the associations between psychopathic traits and other variables were generally stronger in the community compared with the custody sample. The differences across sample type on all model variables suggests the importance of controlling for this variable in our analyses.
Descriptive Statistics, Bivariate Correlations, and Internal Consistencies Across Combined Sample
Note. Cronbach’s α internal consistencies are on the diagonal. PCL:YV = Total Score of Psychopathy Checklist: Youth Version (Forth et al., 2003); INT = Interpersonal psychopathic traits; AFF = Affective psychopathic traits; LIF = Lifestyle psychopathic traits; ANT = Antisocial psychopathic traits; PMWI = Psychological Maltreatment of Women Inventory (Tolman, 1989); DI = Dominance-Isolation abuse; EV = Emotional-Verbal abuse; CTS = Conflict Tactics Scale (Straus, 1979); Parent Viol = Witnessing Interparental Violence.
p < .05. **p < .001.
Descriptive Statistics, Bivariate Correlations, and Group Differences of Study Variables Across Subsamples (Secure Custody Correlations Before Slash, Community After Slash)
Note. PMWI = Psychological Maltreatment of Women Inventory (Tolman, 1989); DI = Dominance-Isolation abuse; EV = Emotional-Verbal abuse; CTS = Conflict Tactics Scale (Straus, 1979); PCL:YV = Total Score of Psychopathy Checklist: Youth Version (Forth et al., 2003); INT = Interpersonal psychopathic traits; AFF = Affective psychopathic traits; LIF = Lifestyle psychopathic traits; ANT = Antisocial psychopathic traits; Parent Viol = Witnessing Interparental Violence.
p < .05.
Hierarchical multiple regressions of each IPV outcome regressed on the predictors are displayed in Table 3. Dominance-Isolation and CTS physical abuse were associated with experiencing interparental violence in Step 1, but these associations dropped to nonsignificant in Step 2 when psychopathic traits were added. Psychopathic traits were associated with each IPV outcome as shown in Step 2, including Dominance-Isolation, Emotional-Verbal, and CTS physical abuse, explaining 7%, 8%, and 3% of the variance in each measure, respectively. In the last step of the regression, sample type was not associated with any IPV outcome beyond the effects of psychopathic traits. However, psychopathic traits continued to be associated with Dominance-Isolation and Emotional-Verbal forms of IPV in Step 3 but was no longer associated with CTS physical abuse. Hence, the association between CTS physical abuse and psychopathic traits may be explained by youth in the custody sample having generally higher CTS scores, but the associations between Dominance-Isolation and Emotional-Verbal abuse and psychopathic traits were robust after controlling for sample type. We conducted a post hoc power analysis to determine how much power we had given our relatively small, combined sample size. With five predictors and our effect sizes, we had an achieved power of 90%.
Hierarchical Multiple Regressions of IPV Types on Early Adversity and Psychopathic Traits, Controlling for Sample in the Final Step
Note. Bold faced values indicate statistically significant at .05 level. Exact p-values are shown in the table. IPV = intimate partner violence; PMWI = Psychological Maltreatment of Women Inventory (Tolman, 1989); CTS = Conflict Tactics Scale (Straus, 1979); Parent Viol = Witnessing Interparental Violence; PCL:YV = Total Score of Psychopathy Checklist: Youth Version (Forth et al., 2003).
Community = 0, Custody = 1.
Overall model explains a significant amount of variance (p < .05).
We repeated the above regression models, replacing total psychopathic traits with the PCL:YV facets (see Supplemental Material for full results). None of the facets were uniquely associated with the Emotional-Verbal form of IPV, suggesting that the full psychopathy construct may be important in its prediction. In contrast, the interpersonal facet was uniquely associated with Dominance-Isolation and the antisocial facet was uniquely associated with CTS physical abuse.
We next examined Hypothesis 2 by conducting atemporal mediation models of the two early adversity measures (i.e., abuse and interparental violence) and their associations with each IPV type through total psychopathic traits. After controlling for the effects of age and sample type, the association between interparental violence and Dominance-Isolation abuse was significantly mediated by total psychopathic traits, b = .06, SE = .03, β = .04, 95% CI [.01, .09], and the association between interparental violence and Emotional-Verbal abuse was significantly mediated by total psychopathic traits, b = .10, SE = .05, β = .05, 95% CI [.01, .10]. There were no indirect effects between early adversity measures and CTS physical abuse via psychopathic traits (see Supplemental Material for all indirect effects). Hence, higher levels of interparental violence were associated with higher levels of psychopathic traits, which in turn were associated with higher levels of both Dominance-Isolation and Emotional-Verbal forms of psychological partner abuse.
Discussion
Our study assessed two hypotheses regarding adolescent psychopathic traits, early adversity, and IPV. First, we examined whether psychopathic traits in adolescent boys would be associated with different forms of IPV. In line with Hypothesis 1, psychopathic traits were associated with each IPV outcome after controlling for the effects of early adversity; and psychopathic traits continued to be associated with psychological forms of IPV after controlling for sample type, while the association with physical abuse fell to nonsignificance. Regarding Hypothesis 2, we found that the associations between interparental violence and the psychological forms of IPV were atemporally mediated by psychopathic traits, suggesting some evidence that psychopathic traits may play a mediating role between early adversity and engaging in some forms of IPV during adolescence.
Research on IPV in adolescent populations with psychopathic traits is relatively scarce (Robertson et al., 2020). Our study adds to this research, suggesting that psychopathy is an important individual difference variable that may influence engaging in a range of abusive acts toward intimate partners in the earliest intimate relationships. Psychopathic traits are also increasingly found to have links with early adversity (de Ruiter et al., 2022), which we also found in our data (Table 1). Early adversity can also affect perpetrating IPV (McKinney et al., 2009), but our findings showed that psychopathic traits were associated with the different forms of IPV in adolescents beyond these factors. The findings are in line with previous studies using the Pathways to Desistance Study that showed adolescents who were classified as engaging in IPV over time (both low-level and high-level decreasing) were higher in psychopathic traits than those who did not engage in IPV (Shaffer et al., 2021; Sweeten et al., 2016). Our study adds to this research by adopting a variable-centered analysis of psychopathic traits and IPV. The findings also add to the broader literature on psychopathy and IPV in adults, which shows that men with elevated psychopathic traits from different populations tend to engage in various forms of partner abuse, including psychological, physical, and sexual abuse (Brazil et al., 2023; Cunha et al., 2021; Hilton et al., 2008; Robertson et al., 2020).
Mean-level differences in psychopathic traits are often found among different populations (Coid et al., 2009; Hare, 2003; Skeem & Mulvey, 2001). Our study also found large mean differences in PCL:YV scores across our two subsamples of secure custody youth and youth from the community, even though community youth were sampled in higher risk locations. Despite these large differences, psychopathic traits tended to be more highly correlated with the other predictor variables and IPV outcomes in the community sample. Our group sample sizes may be too small and unequal to provide strong conclusions about the differences among the justice-involved and community populations, but future work should examine how psychopathic traits presents risk of IPV across different settings. We also found that even after controlling for sample type in the regressions, that psychopathic traits continued to be associated with both the Dominance-Isolation and Emotional-Verbal forms of IPV, but not physical abuse from the CTS. Thus, even after taking into account the differences among custody versus community youth, psychopathic traits may be a risk factor for various forms of psychological partner abuse, whereas the links between psychopathic traits and physical abuse may be explained by the differences these two samples showed on those variables (e.g., youth from custody tend to have both higher psychopathic traits and physical partner abuse). Overall, our use of a combined sample—with the appropriate controls applied and limitations acknowledged—allowed us to examine a broader range of scores (PCL:YV range: 0–37) on the continuous spectrum of psychopathic traits in adolescence (Murrie et al., 2007).
The analysis regarding Hypothesis 2 showed that early adversity was indirectly linked to some forms of IPV via higher levels of psychopathic traits. This link, however, was specific to witnessing interparental violence rather than abuse, and it was specific to the psychological forms of IPV rather than physical IPV. Some research suggests that witnessing interparental violence may be more strongly linked to continuing the cycle of violence in one’s relationships than is abuse (Kalmuss, 1984). Our findings further suggest that psychopathic traits may be one mediating factor involved in this intergenerational IPV perpetration (Murphy & Blumenthal, 2000). Thus, in partial support of Hypothesis 2, adolescents who are elevated on psychopathic traits may engage in higher levels of psychological IPV, in part, because of having experienced more interparental violence at home. More research should examine the links between psychopathic traits, witnessing interparental violence, and their combined influence on relationship outcomes (Dargis & Koenigs, 2017).
A further question from this study that we might ask is what makes psychopathic traits a possible influential individual difference factor in the intergenerational perpetration of violence? What mechanisms could be at play? There are several explanations that could be explored in future studies. One possibility is poor relationship functioning in psychopathic individuals as a result of trauma and/or early relationship issues (de Ruiter et al., 2022; Schimmenti et al., 2021; Zamir, 2022). Some researchers have found trauma experiences in boys specifically predicted engaging in psychological abuse of partners over a 1-year period (Wolfe et al., 2004). Linking this finding with the present study, do youth with psychopathic traits have enduring trauma that impacts their ongoing abusive treatment of partners? Rather than confronting trauma and acknowledging vulnerable emotions, individuals with psychopathic traits might be prone to turn their attention away from emotionally evoking stimuli (Kosson et al., 2018; Vitale et al., 2018), gradually becoming calloused and numbed to negative feelings (Kerig et al., 2012). But the unresolved trauma may still impact their relationships, including engaging in IPV. Future research should examine this as a possible mechanism.
Another potential mechanism is how individuals with psychopathic traits process emotions (Hoppenbrouwers et al., 2016). For example, some studies suggest that the emotion of shame could be important in psychopathic traits, including the role it plays in violence (Ribeiro da Silva et al., 2019). In particular, externalized coping of shame might be prominent in psychopathy (e.g., lashing out against others when feeling negative emotions). Similarly, psychopathy might also involve higher levels of anger (Kosson et al., 2020), which could reduce the threshold for engaging in partner violence. Combined, externalized shame coping and anger could be a powerful recipe for violence in general and IPV in particular. Applying these emotional factors to our study, psychopathic traits might play a mediating role in the early adversity–IPV link because of an increased tendency for externalized shame coping and anger in those with psychopathic traits. How these processes are developmentally related to early adverse experiences such as witnessing interparental violence and how these all combine to influence IPV and other relational problems in those with psychopathic traits should be a target of future research and interventions.
Limitations and Future Directions
Our study has limitations that future studies can help address. First, our small sample size—although providing sufficient power for our purposes—should be followed-up with studies that have larger numbers of youth. Given that psychopathic traits were more strongly associated with IPV in our community subsample, future research could examine how psychopathy may be more differentiating for youth who engage in IPV from community settings. Second, although our model was consistent with a mediation effect of psychopathic traits on the link between early adversity and IPV, we did not have longitudinal data and thus cannot draw definitive causal conclusions. Future studies should track over time how these variables might be related (e.g., more interparental violence affects increases in psychopathic traits, which affects increases in IPV). Another limitation is our reliance on retrospective report from our adolescent participants. Although self-report can provide additional information that might be hard to capture in official records of child maltreatment (Colman & Widom, 2004), it would also be important to capture early adversity with other measures, such as parent report and observation methods. In addition, it would be informative to have dyadic data involving couples that include partner report and/or observations in addition to the self-report IPV measures used in this study.
Implications
Our study has implications for practice and policy involving youth psychopathic traits, family violence, and IPV. More resources should be directed toward trying to prevent the circumstances that foster and promote the development of psychopathic traits in the first place, before they become a problem (Reidy et al., 2015). As part of this effort, our study suggests that early family experiences such as witnessing interparental violence might be a risk factor for higher levels of psychopathic traits in youth across different samples, which is consistent with what others have found (de Ruiter et al., 2022; Schimmenti et al., 2021). This suggests the importance of including family interventions in the treatment and possible prevention of psychopathic traits (Donohue et al., 2021; Kimonis et al., 2019). A benefit of identifying these associations earlier during adolescence is that it captures a time during development where more change may be possible (Backman et al., 2018; Collins et al., 2009).
Our study also suggests the inclusion of psychopathic traits as an individual difference factor that should be considered in treatments of youth involved in dating violence (Wolfe, 2006). Similar to prevention, adolescent IPV intervention may benefit from focusing on several key relationships in adolescents’ environment (De Koker et al., 2014). However, adults with psychopathic traits involved in IPV present unique treatment challenges (Huss & Langhinrichsen-Rohling, 2000; Rock et al., 2013), which suggests psychopathic traits also could present unique challenges to adolescent IPV treatment. Interventions that target a broader range of treatment principles such as risk, need, and responsivity (or RNR) have shown some promise where psychopathic traits are a concern (Olver et al., 2013; Wong & Gordon, 2013). These treatments may be leveraged to reduce IPV-specific behavior as well. In general, more research is needed on treatment involving psychopathic traits and IPV both in adults and adolescents but targeting adolescent populations could help mitigate the damage that psychopathic traits ultimately cause in relationships and society (Caldwell et al., 2007).
Conclusion
Overall, our study suggests that psychopathic traits in adolescents are associated with different forms of IPV, after accounting for early adversity and the sample type where youth were recruited. Psychopathic traits may also play a mediating role between interparental violence and psychological partner abuse, suggesting its importance as an intergenerational factor involved in partner violence. These results can stimulate future research into how psychopathic traits are involved in IPV during adolescence and provide guidance for interventions that target IPV prevention.
Supplemental Material
sj-docx-1-cjb-10.1177_00938548231219810 – Supplemental material for Adolescent Psychopathic Traits, Early Adversity, and Intimate Partner Violence
Supplemental material, sj-docx-1-cjb-10.1177_00938548231219810 for Adolescent Psychopathic Traits, Early Adversity, and Intimate Partner Violence by Kristopher J. Brazil and Adelle E. Forth in Criminal Justice and Behavior
Footnotes
Authors’ Note
The study was performed in line with the principles of the Declaration of Helsinki and informed consent was obtained from all individual participants included in the study. Adelle E. Forth receives royalties from the sale of the Psychopathy Checklist: Youth Version.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
