Predicting violence and criminal behavior is a particularly challenging endeavor for the criminal justice system and forensic mental health clinicians alike. Psychopathy, which is sometimes referred to as the psychopathic personality, is a psychological construct that has been extensively examined with respect to its utility to predict violence and crime. Psychopathy is a personality disorder characterized by a set of maladaptive traits that impact emotional and interpersonal functioning and behavior. These traits can include callousness and lack of empathy, interpersonal antagonism, fearlessness and lack of anxiety, impulsivity, irresponsibility, and criminal and aggressive behavior (among others). These traits, collectively labeled psychopathy, appear to have significant implications in the criminal justice system. Research has repeatedly shown that mock juries rate individuals described as psychopathic as much more dangerous. Consequently, the use of the term psychopath or psychopathy carries significant weight in how an offender is viewed by the criminal justice system with respect to sentencing and amenability for treatment interventions. Interestingly, the construct has been argued by both prosecuting and defense attorneys as an aggravating and mitigating factor for the court to consider prior to sentencing. Prosecutors have argued that criminal defendants labeled psychopaths, or exhibiting strong psychopathic traits, pose a danger to society and therefore should be subjected to longer and harsher sentences (including ineligibility for parole and the death penalty). Conversely, defense attorneys have pointed to the intractable nature of psychopathy as a neurologically based disorder outside the control of the offender.
Assessment of Psychopathy
Several assessment tools have been developed to assess psychopathy. The most popular and widely studied assessment instrument is the Psychopathy Checklist–Revised (PCL-R), developed by Canadian psychologist Robert Hare. Originally developed in the 1980s, the PCL-R was published in 1991 and has been the subject of extensive scientific evaluation. The PCL-R is a clinician rating consisting of 20 characteristics related to the construct of psychopathy. Ratings for these characteristics are assigned with information gathered during a semistructured interview with the individual and a review of records. Scores on the PCL-R have been examined in a number of ways, including a total score as a reflection of global level of psychopathic traits as well as various factor analyses that have predominately focused on interpersonal and affective traits (e.g., callousness, manipulativeness, superficial charm, deceitfulness) as a marker of primary psychopathy and behavioral and social deviance as the secondary psychopathy (e.g., poor behavioral controls, criminal behaviors, history of juvenile delinquency). Hare and his colleagues developed a self-report variant of the PCL-R called the Self-Report Psychopathy Scale (SRP), its fourth edition (SRP4) published in 2016. The SRP-4 has the same underlying structure as the PCL-R.
Another widely researched assessment measure, the Psychopathic Personality Inventory (PPI), was developed by Scott Lilienfeld. Like the SRP-4, the PPI is a self-report instrument that assesses underlying traits of psychopathy, such as rebellious nonconformity, fearlessness, stress immunity, and blame externalization (among others). The PPI has been examined in correctional, community, and university settings. The eight primary scales of the PPI have been examined through factor analyses, wherein three factors emerge: Fearless Dominance, Self-Centered Impulsivity, and Coldheartedness. Fearless Dominance reflects a tendency to remain calm in intense (and potentially dangerous) situations, a bold interpersonal style, and increased ability to recover from stress and anxiety. Self-Centered Impulsivity captures tendencies to act in a rebellious, impulsive manner without regard for social standards. Coldheartedness, as the term implies, is linked to deficiencies in empathy and interpersonal affiliation. It should be noted that there are other self-report measures of psychopathy aside from the tools already mentioned, such as the Levenson SRP, the Triarchic Psychopathy Measure, and the Dark Triad measure. These measures are primarily used in research and generally have not been utilized in forensic psychological evaluations aimed at assessing risk for violence and criminal behavior. The SRP and PPI are commercially available for psychologists to use in psychological evaluations, which make them more admissible in court and much more likely to be utilized in forensic mental health assessments.
Risk Assessment
Predicting future behavior is a difficult task, given the complex interplay between internal (e.g., personality, cognitive abilities) and situational factors. When the target of prediction is violence and criminal behavior, the task becomes even more challenging, given the consequences for both false positives (e.g., restrictions on personal freedoms and liberties) and false negatives (e.g., future victims of violence). The fields of forensic psychology and psychiatry have identified a number of factors that are empirically associated with future criminal and violent behavior. Many of these factors are historical in nature and include things such as a history of violence, antisocial and criminal behaviors, severe mental illness, substance abuse, and previous poor outcomes while under community supervision (e.g., rearrested while on probation or parole). Forensic mental health evaluations that target risk for violence and criminal behavior often assess these factors. Psychological factors are also considered, such as a history of maladaptive functioning and current symptoms of mental illness, including personality psychopathology. Along these lines, many psychological evaluations aimed at predicting risk for criminal behavior and violence include an assessment of psychopathic traits. Whether considered globally (i.e., overall degree of psychopathy displayed by the individual) or as underlying constituent traits associated with psychopathy (e.g., callousness, impulsivity, irresponsibility), assessment of psychopathy is often a central component of a criminal risk evaluation. There is much debate in the field of psychology about what exactly constitutes the construct of psychopathy. The field has generally moved away from considering psychopathy as a homogeneous construct, instead focusing on various factors of psychopathy as well as the underlying traits that make up this disorder. For instance, some operationalizations of psychopathy emphasize antagonism and lack of conscientiousness (i.e., disinhibition), whereas others also include a focus on a bold and fearless temperament.
Research examining the utility of psychopathy as a predictor of violence and criminal behavior has generally focused on three questions:
- To what extent is psychopathy predictive of violence and crime?
- Are specific facets or traits of psychopathy particularly predictive of violence and crime?
- How do psychopathy measures (e.g., PCL-R, PPI-Revised) compare to other violence risk measures in the prediction of violence?
Is Psychopathy Predictive of Violence and Crime?
Regarding the first point, a number of empirical studies have directly examined the ability of psychopathy, often conceptualized with the PCL-R (and to some extent, the PPI), to predict violence and criminal behavior. Research clearly shows that the PCL-R is at least moderately linked with violence, criminal recidivism (reoffending), and institutional behavioral problems for incarcerated offenders and forensic psychiatric patients. One of the most extensive examinations of violence risk assessment was the MacArthur Study of Mental Disorder and Violence by John Monahan and colleagues. They captured the construct of psychopathy with a Screening Version of the PCL and found that it was a robust predictor of violence among forensic inpatients who had been discharged back into the community, even after controlling for other various risk factors. Given the vast amount of research that has examined the relationship between psychopathy and various antisocial outcomes (e.g., crime, violence), meta-analytic reviews are helpful to consider in summarizing the findings from various studies. Anne-Marie Leistico, Randall Salekin, Jamie DeCoster, and Richard Rogers completed a meta-analytic review of 95 studies investigating the PCL in relation to antisocial behavior. They found that the total score of the PCL-R was a significant predictor of violence and antisocial behavior.
However, the PCL-R has been the subject of controversy with respect to its use in risk assessments. All psychological tests and measures are subject to psychometric evaluation, which generally involves examination of the test’s reliability and validity. In this respect, the PCL-R is generally considered to be a reliable instrument and one that has amassed an incredible amount of validation. However, one particular form of reliability, known as field reliability, is focused on examining how reliable the instrument is outside the lab when it is used in the field, such as for forensic mental health assessments. Specifically, this form of reliability assesses whether a test is reliable in the hands of clinicians in various clinical circumstances rather than in the strictly controlled nature of the lab, where most validation research is conducted. The issue of field reliability became a focus of the PCL-R in a series of studies by Marcus Boccaccini, Daniel Murrie, and John Edens (and others), who examined the use of the measure in various types of risk assessments. Boccaccini and Murrie and their colleagues examined sexual offender risk evaluations in Texas, which provides a unique opportunity to evaluate field reliability since the state of Texas required that every convicted sexual offender’s risk for recidivism (which includes the PCL-R) be evaluated by a forensic expert retained by the state and one retained by the defense. They found that differences between state-appointed experts (who generally reported higher PCL-R scores) and defense-appointed experts (who generally reported lower PCL-R scores) for the same defendant exceeded what would be expected based on the reliability of the measure (the standard error of measurement). These authors examined their findings within the framework of a particular bias called adversarial allegiance, which implies that an evaluator may be biased to form an opinion in line with the side that retained the expert. Thus, defense-retained experts may be more likely to report lower levels of psychopathy, and prosecution-retained experts may be more likely to find higher levels of psychopathy. Contrary to these findings, a number of studies in Canada suggest that the field reliability of the PCL-R is good in Canadian criminal proceedings.
Regardless of its underlying cause (e.g., adversarial allegiance), the issue of field reliability is gaining a lot of attention in the research literature and continues to be an important topic in risk assessment. It should be noted that the issue of field reliability and adversarial allegiance are not unique to the PCL-R and psychopathy and has been examined in other forensic instruments (e.g., Static-99) and contexts (e.g., sexual offender evaluations). Researchers examining the utility of the PCL-R in forensic mental health assessment of violence risk have suggested that forensic examiners obtain certifications of their competency before completing assessments for the courts. Moreover, forensic examiners should be mindful of the reliability of their findings and note this for the court to consider in light of a risk assessment.
For a variety of practical reasons, self-report measures of psychopathy are increasingly being used in forensic mental health evaluations. While there is not as much research examining the utility of self-report measures in predicting violence and criminal behavior (in comparison to the PCL-R), there are some preliminary findings that suggest they may have utility in this area. Michael Vitacco, Craig Neumann, and Dustin Pardini examined an earlier version of the SRP (SRP-III) in a moderately sized sample of community-residing males. Utilizing official criminal records, criminal offending was assessed on an average of 3.5 years after the SRP-III was administered. Scores on the SRP-III were able to predict charges for violent and serious criminal charges, even after controlling for other risk factors for violence and antisocial behavior.
Are Particular Facets or Traits of Psychopathy Particularly Predictive of Violence?
Research has shown that most of the predictive ability of psychopathy measures, such as the PCL-R and PPI, are due to their components that capture behavioral deviance and antisocial behavior. On the PCL-R, the second factor, labeled Social Deviance, comprises items that assess a history of poor behavioral control, criminal versatility (i.e., how many types of criminal activities engaged in by the offender), juvenile delinquency, and poor outcomes when previously placed on community control (e.g., probation). This factor of the PCL-R largely accounts for a history of externalizing behavior. The PCL-R meta-analysis mentioned earlier found that the social deviance component of psychopathy (i.e., Factor 2) was a stronger predictor of antisocial behavior for psychiatric inpatients than incarcerated offenders, whereas the interpersonal and affective components of psychopathy (i.e., Factor 1) were stronger predictors of antisocial conduct in samples containing a higher percentage of females. Factor 2 scores on the PCL-R were also better at predicting antisocial conduct at longer follow-up periods.
Studies examining the PPI in relation to the prediction of violence have generally found that Self-Centered Impulsivity, the factor of the PPI that captures an individual’s tendency to act impulsively without consideration of others, accounts for most of the predictive capacity of the measure to predict violence. Edens and Barbara McDermott, for example, found that scores on the PPI Self-Centered Impulsivity scale were significantly associated with two measures of violence risk in a sample of forensic inpatients: the Violence Risk Appraisal Guide and the Historical Clinical-Risk Management-20. Other research has found that PPI Self-Centered Impulsivity was significantly related to numerous markers of antisocial behaviors, including domestic violence and drug and alcohol use (among others).
Although some have argued that the bold or fearless dominance component found in the PPI may not be as central to the psychopathy construct as interpersonal antagonism and impulsivity, recent research has found that the former component is associated with criminal recidivism and treatment failure in a sample of domestic violence batterers. This research by Martin Sellbom and his colleagues found that fearless dominance moderated the relationship between poor impulse control and failure of domestic violence treatment. This means that the combination of PPI Fearless Dominance and Self-Centered Impulsivity increased treatment failure. Conceptually, it makes sense that perpetrators of domestic violence who show traits such as boldness and fearless dominance (which is linked to traits such as pathological narcissism) are more likely to reject the idea that they need intervention.
Some studies have found that the presence of primary psychopathic traits, such as callousness and lack of remorse, is related to one’s risk for future violence, particularly interpersonal and relational violence and instrumental aggression (i.e., violence directed toward obtaining a goal). For instance, sexual offenders who commit aggressive and violent rapes often exhibit significant primary psychopathic traits. Moreover, these offenders tend to show higher rates of sexual and general recidivism among sexual offenders as a whole.
How Does Psychopathy Compare to Other Violence Risk Measures in the Prediction of Violence?
Psychopathy assessed by the PCL-R and other instruments has been compared with other, violence-specific risk measures in a variety of settings. The results have generally been mixed, with some studies showing equivalent predictive capacity for PCL-R scores and other violence risk instruments. Other studies have shown lower predictive ability for the PCL-R in relation to specific risk assessment instruments. The most comprehensive meta-analytic review of the PCL-R in relation to other risk assessment instruments was completed by Min Yang, Stephen Wong, and Jeremy Coid. These authors found little difference between the PCL-R (particularly Factor 2) and eight other violence risk measures (e.g., Historical-Clinical-Risk Management-20, Violence Risk Appraisal Guide) in the prediction of violence. Similar to findings discussed earlier, they found that Factor 1 generally did not contribute to the prediction of violence risk, although the authors noted that it may play a role in moderating the risk of sexual violence among moderate- to high-risk sexual offenders.
Psychopathy and Response to Violence Interventions
Criminal risk assessments typically involve consideration of an individual’s likelihood for violence and/or recidivism. Frequently, those assessments include consideration of efforts that might be employed by mental health professionals to mitigate, or lower, one’s risk. Part of that consideration involves assessing the individual’s amenability for intervention. In this respect, the assessment of psychopathy can play an important role. In addition to examining its role as a predictor of violence and risk for criminal behavior, psychopathy is also considered in terms of its impact in treatment settings. Following adjudication in criminal cases, many offenders are referred to mandatory treatment programs. These are most prevalent with respect to domestic violence and batterer’s interventions and sexual offender treatment. Consequently, the role that psychopathy plays in relation to treatment and interventions for violence and criminal behavior is important to consider in a risk assessment. Psychopathic traits have the potential to impact the offender’s participation in and response to treatment. For example, the interpersonal components of psychopathy (e.g., manipulativeness, conning, inflated ego) can interfere with efforts in treatment. Participants in these settings with strong psychopathic traits can manipulate the treatment provider and other members of the group. Indeed, positive impression management through the use of charm, conning, and manipulation can leave treatment providers questioning the need for intervention.
A major component of offender-based interventions (such as batterer’s interventions and sexual offender treatment programs) is the development of empathy. Many of these programs focus on improving offenders’ ability to appreciate the impact they have on their victims. Psychopathic traits can interfere with these treatment goals because psychopathic offenders often lack the capacity for moral reasoning and empathy.
Ongoing Role of Psychopathy in Risk Assessment
While the specific role that psychopathy plays in violence risk assessment continues to be debated and studied, it is clear that this construct will remain an important topic in the violence risk literature. Future assessment of psychopathy will include greater contributions from neuroscience and neuroimaging techniques. The criminal justice system will have to find ways to incorporate these types of data and findings in light of the burgeoning neuroscience in the area of psychopathy. Moreover, the criminal justice system will have to contend with adversarial arguments about the aggravating and mitigating nature of psychopathy as a factor to consider in sentencing convicted offenders. Regardless, it is clear that psychopathy will continue to play a significant role in violence risk assessment.
References:
- Camp, J. P., Skeem, J. L., Barchard, K., Lilienfeld, S. O., & Poythress, N. G. (2013). Psychopathic predators? Getting specific about the relation between psychopathy and violence. Journal of Consulting and Clinical Psychology, 81, 467–480. doi:10.1037/a0031349
- Hare, R. D., & Neumann, C. S. (2009). Psychopathy: Assessment and forensic implications. The Canadian Journal of Psychiatry, 54, 791–802.
- Skeem, J. L., Polaschek, D. L., Patrick, C. J., & Lilienfeld, S. O. (2011). Psychopathic personality bridging the gap between scientific evidence and public policy. Psychological Science in the Public Interest, 12, 95–162. doi:10.1177/1529100611426706