Issues in violent risk assessment: lessons learned and future directions

Realizing that the assessment of dangerousness with a yes/no format as a poor form of violent risk assessment has been the most important lesson learned about violence in the last 20 years. Further examining (a) what outcome and (b) how the indicators of the outcome should be measured has resulted in better violent risk assessment. The most promising methodological innovation in violent risk assessment has been the introduction of the receiver operating characteristic to assess the efficiency of risk prediction. In the future, the tension between prediction and explanation of violence will need to be maintained while focusing on a mechanism-driven strategy of risk management.     

The Empirical Limits of Forensic Mental Health Assessment

This article introduces a special issue of Law and Human Behavior, including five articles describing the limits of forensic mental health assessments of (a) risk of violence in female adolescents, (b) sexually violent predators, (c) dangerousness in capital murder cases, (d) child sexual abuse, and (e) PTSD litigants. Knowing the limits of forensic mental health assessment methods is essential in order to recognize their strengths, increase the credibility of forensic mental health assessment, and drive research that will enhance the value of assessments for the courts.     

A Comparative Test of Clinical Judgment Versus Actuarial Prediction of Future Violence

This study tested how graduate level psychology graduate students (n = 20) using the Adapted Violence Risk Appraisal Guide (Adapted VRAG) would do relative to practicing psychologists/psychiatrists (n = 16) using clinical judgment when predicting violence in 10 narratives from the MacArthur Violence Risk Assessment study (Monahan et al., The MacArthur violence risk assessment study. . Retrieved 10 Oct 2005). Results indicated that the practicing psychologists/psychiatrists made significantly more correct predictions than the master’s level students. The professional group demonstrated sensitivity levels of 77.7% and specificity of 96.3%. For the use of the adapted VRAG method by the graduate student group, specificity levels were modest at 54.0%. Sensitivity levels, however, were lower than earlier demonstrated levels at 58.0%. These findings are at variance with earlier reports comparing clinical and actuarial methods. The results may reflect the short amount of time the master’s level students were trained using the Adapted VRAG as well as the small number of participants in this study. Additional research comparing other professions is recommended, as well as examining if experience in the forensics field would affect one’s ability to predict violence.     

Dangerousness

Saleem Shah played a major role in the development of both policy and research concerning the prediction of dangerousness to others over two decades. During that period his commitment to the welfare of people with mental illness led him first to challenge the legitimacy of such predictions and later to support research to improve them. His ideas were central to most of the research that has been done in this area. This article suggests a direction for future research concerning the prediction of dangerousness to others that elaborates and extends his thinking by focusing on the research necessary to help individuals with mental illnesses to minimize their involvement in violence.     

Prediction Versus Management Models Relevant to Risk Assessment: The Importance of Legal Decision-Making Context

Most of the theoretical and empirical literature on violence risk to date has focused on the task of predicting who will behave violently. In the present article, it is argued that at least two models of risk assessment may be applied to the varying legal decisions in which violence risk is a consideration: prediction (with an emphasis on overall accuracy) and management (with an emphasis on risk reduction). These two models are described, and discussed in the contexts of the literatures on forensic assessment and therapeutic jurisprudence. The implications for research, policy, and practice are considered.     

Predicting violence among forensic-correctional populations: the past 2 decades of advancements and future endeavors

Research on violence prediction during the past 2 decades has evolved appreciably in terms of depicting determinants of violence and developing psychometrically sound actuarial measures to predict the probability of future violent behavior. This article provides a brief synopsis of information on predicting violence gained in the past 2 decades, the methodological advances evolved, and finally significant gaps in violence prediction that still exist for future research endeavors.     

Continuity and Pathways from Aggression in Childhood to Family Violence in Adulthood: A 30-year Longitudinal Study

Literature suggests that early patterns of aggressive behavior in both girls and boys are predictive of later violent behavior, including violence that takes place within family contexts. Utilizing the Concordia Longitudinal Risk Project, a study of individuals recruited as children in the 1970s from inner-city schools in Montreal, this study examined different pathways whereby aggressive behavioral styles in childhood may place individuals at risk for continuing patterns of violence towards children and spouses. Childhood aggression directly predicted self-reported violence towards spouse for both sexes, with indirect routes through lowered educational attainment and marital separation. Aggression in childhood was also found to predict parents’ self-reports of using violence with their children. For mothers, educational attainment and current absence of the biological father from the child’s home also played important roles in predicting violent behavior towards offspring. These findings provide evidence of both continuity of aggressive behavior and indirect risk paths to family violence, via lower educational attainment and parental absence. In both men and women, childhood aggression may be an identifiable precursor of family violence and child abuse.     

Perceived Risk for Future Intimate Partner Violence among Women in a Domestic Violence Shelter

Research investigating women’s risk assessments for intimate partner violence (IPV) shows that women can predict future violence with relative accuracy. Limited research has investigated factors that are associated with perceived risk and the potential behavioral consequences of victim risk perception. Results from a survey of women in a domestic violence shelter (N = 56) indicated that women perceive lower risk of future violence if the abusive relationship were to end and higher risk of violence if it were to continue. Certain abuse experiences were related to elevated perceptions of personal risk for future violence. Further, perceived personal risk predicted the women’s intention to terminate their relationship upon leaving shelter. Results are discussed as they may inform interventions preventing IPV.

Typologies of Intimate Partner Violence: Evaluation of a Screening Instrument for Differentiation

Given the prevalence and impact of intimate partner violence (IPV) in both community and therapeutic settings, it is vital that the varying typologies of IPV be identified and treated accordingly. The present study sought to evaluate the efficacy of a screening instrument designed to differentiate between characterologically violent, situationally violent, and distressed non-violent couples; focus was placed on identifying situationally violent couples so that they could be invited to participate in a conjoint pyschoeducational workshop. Couples from two samples were assessed to achieve this goal. Situationally violent couples (N = 115) from Sample 1 were screened into the study via a phone interview and participated in an in-home assessment, which assessed self-reported relationship violence. These couples were compared to a previously collected sample (Sample 2; Jacobsen et al. 1994) of characterologically violent, distressed non-violent, and situationally violent couples. The main hypotheses stated that couples from Sample 1 would report less severe relationship violence than characterlogically violent couples from Sample 2, and would report greater amounts of low-level violence than distressed non-violent couples from Sample 2. Additionally, similar rates of both self-reported violence would be seen for situationally violent couples from Samples 1 and 2. Multivariate analyses supported this with the exception that situationally violent couples from Sample 1 did not differ significantly across all domains from distressed non-violent couples in Sample 2. Implications for the screening instrument’s utility in clinical and research settings are discussed.     

Assessing the Risk of Future Psychological Abuse: Predicting the Accuracy of Battered Women’s Predictions

Building on a handful of studies demonstrating battered women’s accuracy in assessing their risk of being physically reabused, this study examined how accurately victims assess their risk of future psychological abuse. Participants’ ratings of the likelihood that their partner would engage in controlling/dominance behaviors or efforts to humiliate/degrade them in the coming year and their reports 18 months later of whether this had actually occurred were used to create a four category version of accuracy (true positive, false positive, true negative, false negative). Victims were more likely to be right than wrong in their assessments of risk; PTSD symptoms, the recency of physical violence, and the degree of stalking and psychological abuse in the relationship predicted membership in the four accuracy categories. These findings overlap considerably with those examining victim accuracy in predicting physical abuse and inform ongoing debates about the value of incorporating victims’ insights into risk assessment efforts.

Mothering and Domestic Violence: A Longitudinal Analysis

The quality of the relationships that parents, particularly mothers, have with their children is a potent predictor of children’s future development. Recent research suggests that mothers of preschool-age children exposed to domestic violence may be more sensitive and responsive to their children than other parents. This heightened sensitivity and responsiveness in relationships may be key to the successful development of some children exposed to domestic violence; however this theory remains unexplored. The Canadian National Longitudinal Survey of Children and Youth provided a large sample size suitable for longitudinal analysis of the relationship between exposure to family violence and parenting behaviors. Analysis revealed that mothers of children exposed to domestic violence may indeed compensate for exposure to violence in their parenting interactions with their 2- to 12-year-old children.     

Decision-Making About Volitional Impairment in Sexually Violent Predators

The Kansas v. Hendricks (1997) decision, in which the Supreme Court authorized post-sentence civil commitment for certain sex offenders, appeared to be constitutionally legitimized by limiting the class of offenders eligible for this special form of civil commitment to those who are "unable to control" their dangerousness. Nowhere in the available record, however, did the Court elucidate what they meant by this notion of volitional impairment. This study sought to examine factors that legal professionals (n=43), psychologists (n=40), and mock jurors (n=76) deem most relevant to a determination of sex offender volitional impairment. Participants, who were randomly assigned to a sexual predator commitment or an insanity hearing context, read a series of 16 vignettes that described a pedophilic offender and included combinations of variables hypothesized to be related to judgments of volitional impairment. Results suggested that participants, who as a group made remarkably high estimates of likelihood of future sexual violence, considered verbalization of control, history of sexual violence, and the context of the hearing as highly relevant to determinations of volitional impairment. Implications for policy and practice are explored.     

Gender and Risk Assessment Accuracy: Underestimating Women’s Violence Potential

Understanding factors that contribute to mental health professionals (MHPs) accuracy in assessing patients risk of violence can inform efforts to improve accuracy and to integrate risk assessment technology with practice. Based on a sample of 147 clinicians who assessed 680 patients in a psychiatric emergency room, this study investigates the influence of patient gender, MHP gender, and their potential interaction on MHPS risk assessment accuracy. The results indicate that MHPs of both genders are particularly limited in their ability to assess female patients risk of future violence. This finding was not limited to a particular professional group and was not attributable to gender-related differences in violence. Implications for future research on the judgment processes that may underlie MHPs limited accuracy with women and for training programs in violence risk assessment are discussed.     

略论人身危险性与犯罪定量的关系

社会危害性是指行为对特定社会所保护的社会关系造成的现实的或可能的危害的属性,是事实评价与价值评价的统一,是主观与客观的统一,是现实性和可能性、已然性和未然性的统一。人身危险性是初犯可能与再犯可能的统一,是特定人格与非特定人格的统一。因此,两者的关系就表现为人身危险性是社会危害性的重要内容,人身危险性是表征社会危害性的主观范畴。人身危险性的大小是衡量社会危害性程度的一个重要因素。从犯罪定量的角度看,如果犯罪数额、犯罪结果等因素是衡量犯罪的社会危害性程度的客观因素,那么,人身危险性则可以与主观恶性、罪过心理等主观因素一起被视为犯罪定量的主观因素。     

Assessing the Spatial–Temporal Relationship Between Disorder and Violence

The relationship between disorder and violence has generated much debate in the field of criminology. While advocates of the broken windows thesis believe disorder is the root cause of crime, other researchers view both disorder and crime as analogous behaviors resulting from the breakdown of collective efficacy. Scholars from both sides of this debate, however, assume a long-term correlation between disorder and crime at places. This assumption has not been tested with a longitudinal dataset at a relatively small geographic unit of analysis. The current study used data collected in Seattle, Washington and utilized Group-based Trajectory Analysis and Joint Trajectory Analysis to explore the longitudinal relationship between disorder and violence. The results showed that disorder, just like crime, concentrates in a few “hot spots.” Additionally, the results showed that while the lack of disorder problems guarantees places to be violence free, having high levels of disorder predicts having violence problems only about 30% of time. As such, these findings point out the need for future theorization efforts on the disorder-violence nexus to include contextual factors which could explain this imperfect association between the two.     

Predictive Efficacy and the Preventive Detention of Dangerous Sexual Offenders: Contributions from Nonlinear Dynamical Systems Theory

Behavioral scientists are regularly summoned by the legal system to make predictions about the likelihood of future dangerous behavior (i.e., violence prediction/risk assessment). In recent years, such efforts have been subjected to considerable scrutiny as, despite the evolution of risk assessment instruments, social scientists have yet to demonstrate the ability to predict violence with a reasonable degree of accuracy. Yet concerns with predictive efficacy may be rooted in the limitations of prediction itself. Nonlinear dynamical systems theory (i.e., chaos theory) and quantum physics jointly paint a picture of human behavior as fundamentally unpredictable; in their light, we are asked to question whether future violence can ever be predicted with a tolerable level of error. In this article, I provisionally explore this very issue, particularly as it appears in the context of decisions concerning the disposition of criminal sexual offenders.     

Probability,danger, and coercion

Young adults were shown hypothetical stimulus vignettes describing mental patients and were asked to judge (a) the probability that the patient would harm someone else, (b) whether or not the patient should be categorized as “dangerous,” and (c) whether coercion should be used to ensure treatment. Probability and dangerousness judgments were systematically related and were predictive of the judged necessity for coercion. However, judged probability was strongly dependent on the form of the response scale, suggesting that probability was not represented consistently and quantitatively in participants' minds. Study 2 replicated these findings with forensic clinicians as participants. These results underscore the importance of violence to others in mental health law and have important implications for the manner in which risk assessments are formulated for use by the legal system.     

Psychopathy and predatory violence in homicide,violent, and sexual offences: Factor and facet relations

Purpose. Evidence has been found pointing to a relationship between psychopathy and predatory violence. The present study investigated the violence mode and its relationship with psychopathy on factor as well as on facet levels. It was hypothesized that predatory violence would be related to the affective and/or interpersonal facets of psychopathy. A subsidiary hypothesis was that sexual offences could also be differentiated by means of violence mode. Methods. Participants were 82 male inmates convicted of a violent crime, a sexual crime, or a homicide. Psychopathy was assessed with the Hare PCL‐R2 and the violence mode was assessed by means of Cornell's Aggressive Incident Coding Sheet, based on interviews with the offenders and a review of the official record. Results. On the psychopathy subcomponent level, only the interpersonal facet was positively related with predatory violence. This association makes sense considering that psychopaths' interactions with others are defined by gradients of power and control and narcissistic gratification, rather than by attachment patterns. By contrast, the antisocial facet was associated negatively with predatory violence. Our subsidiary hypothesis concerning the possibility of differentiating sexual violence on the basis of the two violence modes was not confirmed. Conclusions. The present results add to the growing evidence that predatory violence is related to the personality traits of psychopathy rather than to its life‐style and antisocial characteristics. Therefore, a risk analysis of future predatory violent behaviour might benefit from the inclusion of the assessment of the personality facets of psychopathy instead of focusing solely on the antisocial behaviour.