GRANT T. HARRIS MARNIE  E. RICE

 

Can  psychopaths  be  treated?  In  this  chapter, we evaluate the empirical evidence on the treatment of psychopaths. We concentrate  on  treatment  for  criminal  psychopaths and intervention strategies in which efforts to re- duce criminal and violent behavior are at least part of the protocol. Without denying the importance of other psychopathic char- acteristics, criminal and violent behaviors  are clearly the most important outcomes from a social policy perspective.

We do not discuss treatment for various types of psychopaths, although there has been considerable discussion about the clinical and theoretical significance of psychopathy sub- types. Prototypical (sometimes called pri- mary) psychopaths present as callous and un- emotional, whereas secondary psychopaths seem more emotionally labile, angry, or anx- ious (Poythress & Skeem, Chapter 9, this vol- ume; Skeem, Poythress, Edens, Lilienfield, & Cale, 2002). It has been hypothesized that one form of psychopathy is primarily a heritable condition while another is due mainly to envi- ronmental influences, particularly abuse dur- ing childhood (Mealey, 1995). Whether the primary–secondary distinction maps onto the genetic–environmental distinction is unclear. Nevertheless, subtypes of psychopathy might require different therapies (Skeem, Poythress, et al., 2002). However, until there is more evi- dence that it matters to prognosis (criminal outcome, response to treatment), the exis- tence of subtypes cannot have much relevance to treatment.

 

TREATMENT OF PSYCHOPATHIC OFFENDERS AND PSYCHOPATHIC FORENSIC  PSYCHIATRIC PATIENTS

 The clinical literature has been quite pessi- mistic about the outcome of therapy for psy- chopaths. Hervey Cleckley, in his several edi- tions of The Mask of Sanity (1941, 1982), described psychopaths as neither benefiting from treatment nor capable of forming the emotional bonds required for effective ther- apy. In contrast, some early studies claimed positive effects of psychotherapy (Beacher, 1962; Corsini, 1958; Rodgers, 1947; Rosow,

1955; Schmideberg, 1949; Showstack,  1956;

Szurek,  1942;  Thorne,  1959).   However,   all these were uncontrolled case reports. Reviewers before 1990 concluded, as had Cleckley, that there was no evidence for the efficacy of treatment with adult psychopaths (Hare,   1970;   McCord, 1982).

 

Therapeutic  Communities

 One of the most popular treatments for psy- chopathy has been the therapeutic com- munity. Hare (1970) suggested that the reshaped social milieu of a therapeutic com- munity might alter the basic personality characteristics and social behavior of psy- chopaths. Although lacking  comparative data for untreated psychopaths, there were several early positive reports (Barker & Mason, 1968; Copas, O’Brien, Roberts, & Whiteley, 1984; Copas & Whiteley, 1976; Kiger, 1967). Based on these, Rice, Harris, and Cormier (1992) evaluated an intensive therapeutic community for mentally disor- dered offenders thought to be especially suit- able for psychopaths. It operated for over a decade in a maximum security psychiatric hospital and drew worldwide attention for its novelty. The program was described at length by Barker and colleagues (e.g., Barker, 1980; Barker & Mason,  1968;  Barker, Mason, & Wilson, 1969; Barker & McLaughlin, 1977) and elsewhere (Harris, Rice,     &     Cormier,     1994;     Maier,    1976; Nielson, 2000; Weisman, 1995). Briefly, the program was based on one developed by Maxwell Jones (1956, 1968). It was largely peer operated and involved intensive group therapy for up to 80 hours per week. The goal was an environment that fostered empa- thy and responsibility for  peers.

The evaluation (Rice et al., 1992) was quasi-experimental in which 146 treated of- fenders were matched with 146 untreated of- fenders on variables related to recidivism (age, criminal history, and index offense). Al- most all offenders had a history of violent crime and were scored on the Psychopathy Checklist—Revised (PCL-R; Hare, 1991, 2003). Although the two groups were not explicitly matched on the PCL-R, the aver- age score in each was 19. Because the PCL-R was scored using file information only, the cutoff score for classifying offenders as psy- chopaths was set at 25 rather than the customary 30. The results of a follow-up conducted an average of 10.5 years after completion of treatment showed that, com- pared to no program (in most cases, un- treated offenders went to prison), treatment was associated with lower violent recidivism for nonpsychopaths but higher violent recid- ivism for psychopaths. Psychopaths  showed poorer adjustment in terms of problem be- haviors while in the program, even though they were just as likely as nonpsychopaths to achieve positions of trust and early recom- mendations  for release.

Why did the therapeutic community pro- gram have such different effects on the two offender groups? We speculated that both  the psychopaths and nonpsychopaths who participated in the program learned more about the feelings of others, taking others’ perspective, using emotional language, be- having in socially skilled ways, and delaying gratification. For the nonpsychopaths, these new skills helped them behave in prosocial and noncriminal ways. For the psychopaths, however, the new skills emboldened them to manipulate  and  exploit others.

In another therapeutic community, Ogloff, Wong, and Greenwood (1990) reported on the behavior of psychopaths and nonpsycho- paths defined by criteria outlined in an early version of the Psychopathy Checklist (Hare & Frazelle, 1985). Compared to nonpsycho- paths, psychopaths showed less motivation, were discharged earlier (usually because of lack of motivation or security concerns), and showed less improvement. Similar results were reported for a therapeutic community in England’s Grendon prison in (Hobson, Shine, & Roberts, 2000), where poor adjust- ment to the program was likewise associated with higher PCL-R scores. A recent study of  a therapeutic community for female sub- stance abusers (Richards, Casey, & Lucente, 2003) reported that, although none of the of- fenders scored over 30 on the PCL-R, higher psychopathy scores were nevertheless associ- ated with poorer treatment response indi- cated by failing to remain in the program, rule violations, avoiding urine tests, and spo- radic attendance.

Despite evidence that therapeutic commu- nities are ineffective with psychopaths, they remain popular in prisons, secure hospitals, and other institutions in Europe in which some participants are likely to be psycho- paths (Dolan, 1998; McMurran, Egan, & Ahmadi, 1998; Reiss, Meux,  &  Grubin, 2000). Even in North America, therapeutic communities are advocated for people with substance abuse problems (e.g., Knight, Simpson, & Miller, 1999; Wexler, Melnick, Lowe, & Peters, 1999), some of whom are likely to be psychopaths. Few studies of therapeutic communities outside North America, and only one for substance abusers (Rich- ards et al., 2003), have used PCL measures that would allow estimating the prevalence of psychopathy.

 

Other Treatment Approaches

Besides therapeutic communities, cognitive- behavioral therapy is often recommended for psychopathic offenders. Andrews and Bonta (1994), Brown and Gutsch (1985), Serin and Kurychik (1994), and Wong  and  Hare (2005) all suggested that intensive cognitive- behavioral programs targeting “crimino- genic needs” (i.e., personal characteristics correlated with recidivism) might be effec- tive. For example, Wong and Hare recom- mended relapse prevention in combination with cognitive-behavioral programs. How- ever, doubts as to the efficacy of this treat- ment with psychopaths arose from an evalu- ation of a cognitive-behavioral and relapse prevention program for sex offenders con- ducted by Seto and Barbaree (1999). High psychopathy offenders who were rated as having shown the most improvement (as measured by conduct during the treatment sessions, quality of homework, and thera- pists’ ratings of motivation and change) were more likely to reoffend than other partici- pants, particularly in violent ways. The treat- ment followed the principles of good correc- tional treatment (Andrews & Bonta, 1994; Andrews et al., 1990): It was highly struc- tured and cognitive-behavioral, best match- ing the learning style of most offenders, including psychopaths. Moreover, psycho- paths are high-risk offenders with many criminogenic needs (Zinger & Forth, 1998), and thus the program targeted deviant sexual preferences and antisocial attitudes (Barbaree, Peacock, Cortini, Marshall, & Seto, 1998). In view of these features, the re- sults pertaining to psychopaths are especially notable.

Further doubts regarding the efficacy of cognitive-behavioral treatment for psycho- paths emerge from other outcome studies. Among participants in a program for men- tally disordered offenders in a secure psychi- atric hospital, Hughes, Hogue, Hollin, and Champion (1997) found that PCL-R score was inversely correlated with therapeutic gain,    even    though    patients    with  PCL-R

scores over 30 were excluded. In another study, Hare, Clark, Grann, and Thornton (2000) evaluated cognitive-behavioral prison programs for psychopathic and nonpsycho- pathic offenders. After short-term anger management and social skills training, 24- month reconviction rates for 278 treated and untreated offenders yielded an interaction between psychopathy and treatment out- come similar to that reported by Rice and colleagues (1992). Whereas the program had no demonstrable effect on nonpsychopaths, treated offenders who scored high on Factor 1 of the PCL-R had significantly higher rates of recidivism than high-scoring but un- treated offenders.

In short, the few available empirical re- sults regarding the effectiveness of treatment with psychopathic offenders are dismal, leading some to suggest that one should dis- cuss management rather than treatment for psychopathic offenders (see Lösel, 1998). It may be that the very highest-risk offenders (i.e., psychopaths) might not be treatable even with very intensive and carefully de- signed and implemented programs. Of even more concern, perhaps, is the possibility that programs that might be beneficial for other offenders actually increase the risk repre- sented  by psychopaths.

 

Meta-Analysis of Research

on the Treatment of Psychopathy

Traditionally, in a review of the evidence per- taining to a particular question, commenta- tors summarize studies and derive an infor- mal summary of the state of knowledge. This summary is usually accompanied by specula- tion about possible sources of apparent con- flict in findings across studies. However, a more systematic way to resolve apparent in- consistencies in research findings is to use meta-analysis. This statistical approach al- lows the combination of research results from many studies, permitting conclusions about the likelihood that a group difference or relationship exists, how large it is, and why some studies find it and others do not. Research on the treatment of psychopathic offenders might seem particularly fruitful for meta-analysis because studies in this area of- ten use small samples, such that effects might go undetected due to low statistical power. Studies  also  differ  in  the  measures  of psychopathy, kinds of treatment provided, cri- teria by which candidates are assigned to treatments, and procedures used to evaluate outcomes. Meta-analysis offers a solution to the problem of small sample sizes in individ- ual studies, as well as a methodology for test- ing hypotheses about the sources of differ- ences in findings across  studies.

Of course, meta-analysis cannot overcome general deficits. For example, if very few studies of psychopathy treatment used the PCL-R, meta-analysis could not  examine  it  in moderating treatment effects. A meta- analysis also cannot make up for method- ological inadequacies in the literature as a whole. For example, one of the most serious problems in this literature is the scarcity of well-controlled studies, especially those us- ing random assignment. By contrast, there is an increasing trend toward evidence-based medicine in the treatment of physical and mental health problems in general, which has resulted in the Cochrane Database of systematic reviews—a collection of meth- odologically adequate studies on various diseases and conditions (www.update-soft- ware.con/cochrane/). Studies using random assignment are heavily weighted in this data- base and few other designs are considered strong  enough  to  be informative.

A good illustration of the limitations of meta-analysis was afforded by a recent meta- analysis of research findings on the treat- ment of psychopathy (Salekin, 2002).1 Salekin provided a quantitative review of 42 studies he identified as having evaluated the effectiveness of some form of therapy for psychopaths. Salekin reported that the mean rate of successful intervention across all treatment studies was .62, p  <  .01.  This  was the proportion of treatment candidates judged to have “improved”2 minus the pro- portion expected to have improved without treatment (the latter proportion was calcu- lated, according to the author, by averaging the improvement of untreated subjects in eight studies identified as including compari- son or control groups). Salekin concluded that the prevailing pessimism about the treatment of psychopaths was unfounded.

Several aspects of this meta-analysis are noteworthy: The mean intensity of treatment was approximately four sessions per week over a year; only four studies employed the Hare   PCL-R;   only   eight   studies   included

comparison subjects3; few studies (< 20%) assessed outcome in terms of criminal behav- ior, and even fewer (< 10%) mentioned vio- lence or aggression; the most effective treat- ment was found to be psychodrama; and the evaluation of effectiveness was most often (> 70%) based on therapists’ impressions. In an effort to improve the rigor of studies without control groups, Salekin stated that he used averaged data from the “controlled” studies to estimate an effect of nontreatment for all studies. However, for reasons articulated later, we consider this method of calculating the improvement of control subjects to be problematic.

Our opinion, based on a variety of consid- erations, is that no firm conclusions can be drawn from this meta-analysis. In particular, we maintain that only controlled studies can be informative regarding treatment efficacy, and no conclusions can be drawn from uncon- trolled studies. Because we consider control groups to be essential, we turn our attention first to the eight studies Salekin identified as controlled. We begin with Rice and colleagues (1992), which was discussed at some length earlier in this chapter, and then consider each of the other seven studies in turn. Rice and col- leagues reported that 78% of the treated psy- chopaths committed a new violent offense during the follow-up compared to 55% of un- treated psychopaths. Salekin’s summary of Rice and colleagues stated that 22% of the psychopaths “benefited” from treatment compared to 20% who would have “bene- fited” without the program, for a net benefit of 2%. Salekin considered that psychopaths who did not violently reoffend during the fol- low-up “benefited” from treatment even though untreated psychopaths exhibited less violent recidivism. The 20% figure was the weighted average proportion of psychopaths he calculated as having improved without treatment from the eight studies considered to be controlled. For each study in the meta- analysis, he subtracted this 20% figure from the percentage he considered to have bene- fited from treatment to compute net benefit. We believe the Rice and colleagues study shows why this method is problematic.

Craft, Stephenson, and Granger (1964) compared 50 severely delinquent boys alter- nately assigned to either a group psychother- apy unit or an “authoritarian” unit. No ac- cepted measure of psychopathy was used.

The former program was new and incorpo- rated many components of Jones’s therapeu- tic community. In the latter, “authoritarian” program, patients were told on admission that “noise and disarray would not be toler- ated and peace and quiet would be enforced by putting offenders to bed, fines, [and] de- privation of privileges . . . [combined with] “superficial psychotherapy” (p. 546). It was described as “standard” treatment at the time (1958). The authors had clearly ex- pected that group psychotherapy would emerge as the superior program, but the re- sults favored the authoritarian program. Sig- nificantly fewer offenses were committed by boys from that program in the follow-up pe- riod than by boys from the group psycho- therapy program. Psychological test results also clearly favored the authoritarian pro- gram. The authors concluded that no conclu- sions could be drawn about the effectiveness of either treatment, as there was no un- treated control group. They stressed that their study yielded no evidence to support the prevalent view among therapists that psychotherapy was more effective than stan- dard treatment.

Salekin categorized this study as contain- ing two treated groups—“therapeutic com- munity” (the group psychotherapy program) and “cognitive-behavioral” (the authoritar- ian regime). The term “cognitive-behavior- al” did not appear in the original study and cognitive-behavioral therapy was not devel- oped until approximately a decade after this study was completed (Friedman, 1970). Salekin reported that this study showed posi- tive results for both programs because 63% benefited (i.e., had no convictions in the fol- low-up) from the cognitive-behavioral pro- gram, and 43% benefited from the thera- peutic community program. However, a different interpretation was given by the study’s original authors: “Both treatments may have been better than nothing; both . . . may have worsened the boys—we do not know” (Craft et al., 1964, p. 553). We think a fairer interpretation is that this study yielded results similar to those of Rice and colleagues (1992), inasmuch as the therapeu- tic community increased recidivism relative to a standard, more custodial, approach.

Ingram, Gerard, Quay, and Levinson (1970) compared 20 juvenile delinquents treated   in   an   “action-oriented” program

with 41 youths admitted either before the program began or after it ended. All were categorized as psychopathic according to an instrument developed by one of the authors. Treated youths had fewer assaultive offenses during the program (.25 per youth) com- pared to controls (.50 per youth), although the difference was nonsignificant. None of  the treated youths were reported to have made a negative institutional adjustment af- ter transfer to another institution, compared to 21% of controls. Salekin reported this study as demonstrating that 75% of treat- ment participants had benefited in terms of reduction of institutional aggression, and 100% had benefited in terms of improve- ment  in  community adjustment.

Korey (1944) studied delinquent boys in a training school. No objective measure of psychopathy was used, although all partici- pants were diagnosed as “constitutional psy- chopathic inferiors” with “severe delinquent and behavior problems” (p. 127). Seven boys (the experimental group) received benzedine sulfate, and five boys (the controls) received a placebo. Outcome was measured by thera- pist opinion regarding improvement in vari- ous aspects of institutional adjustment. Sig- nificantly more (N = 4) boys given the drug were judged to have improved than boys given placebo (none of whom were judged   to have improved). Korey cautioned that benzedine left the boys’ underlying personal- ities untouched and that it should be part of  a more comprehensive treatment. Salekin re- ported that 57% of the treated boys in this study benefited.

Maas (1966) studied 46 adult female of- fenders classified as unsocialized on Gough’s socialization continuum. Half were assigned to group therapy emphasizing psychodrama, and the others were assigned to an untreated control group. The outcome measure was self-reported ego identity. No actual data were presented indicating how many offend- ers improved, but the authors stated that there was a significant difference in favor of the psychodrama group. Salekin summarized this study, stating that 63% of the treated subjects improved.

Persons (1965) compared 12 inmates ran- domly assigned to treatment with 40 inmates randomly assigned to no treatment. Treat- ment was eclectic counseling twice a week for   10   weeks.   All   52   inmates   were psychopaths according to a self-report ques- tionnaire. Self-report and therapist ratings showed significantly more improvement for treated offenders. Treated offenders also had significantly fewer disciplinary reports over the 10 program weeks. Salekin reported that 92% of the treated inmates benefited, al- though (as is the case with the Maas study described previously) it is unclear how this figure was obtained, as no such data were in the  original article.

Skolnick and Zuckerman (1979) com- pared 59 male drug abusers treated in a ther- apeutic community with 37 untreated male drug abusers of similar IQ who spent an equivalent period in prison. The article nei- ther mentions psychopathy nor how many subjects were classified as psychopaths. The main outcome variables were changes on Minnesota Multiphasic Personality Inven- tory (MMPI) scales and three other self- report personality measures administered upon admission and again 6 to 8 months later. Although treated subjects decreased significantly more than controls on several measures of psychopathology, Salekin re- ported a negative effect of treatment in this study, presumably because the number of treated subjects who had 49 or 94 high peak codes on the MMPI increased significantly, whereas there was no increase in the com- parison group. The authors pointed out that the increase in treatment participants with 49 or 94 high peak codes was due to de- creases in the other scales rather than the re- sult of an absolute increase in four and nine scale scores.

Finally, Woody, McLellan, Luborsky, and O’Brien (1985) studied 30 opium-dependent men diagnosed with personality disorder. Some (N = 17) had an additional diagnosis  of depression. Some received drug counsel- ing alone while others received counseling plus professional psychotherapy. The out- come variable was change in problem sever- ity measured before and after treatment via structured clinical interviews. Some positive changes were reported for the depressed men, but the other men “showed little evi- dence of improvement” (Woody  et al.,  1985,

1. 1064). No comparison of the two treat- ments was reported, and it is unclear how Salekin could have considered this a con- trolled Nevertheless, he reported  that

80% of the treated men benefited from treat- ment.

One study in Salekin’s meta-analysis was not classified as controlled, but we believe it should have been. Miles (1969) compared 40 male adolescents admitted to a therapeutic community with 20 control patients in the same hospital (described as a “psychiatric hospital for the subnormal,” p. 23)  who  were not offered the therapeutic community. The two patient groups were similar on age, IQ, and social class. Although Cleckley’s  work is cited, no mention is made of how many patients were psychopaths. Sociometry was used to measure outcome, and there was a net improvement in acceptance in 70% of the therapeutic community subjects com- pared to 10% of the comparison subjects. The authors concluded that the therapeutic community “increased the ability of the pa- tients to accept their fellows more than did the traditional treatment” (p. 35). Salekin re- ported that the therapeutic community bene- fited 65% of the patients on measures that included “improved empathy,” although the authors stressed that they used no measure of  empathy.

How can we summarize these “con- trolled” studies of treatment outcome? We note that only one study (Rice et al., 1992) used the PCL-R, which is the contemporary standard (and most empirically valid) mea- sure of psychopathy. Only two employed objective measures of criminal recidivism (Craft et al., 1964; Rice et al., 1992). Inter- estingly, our interpretation of both of these is that the treated group exhibited higher rates of recidivism than the control group. Our reading of the “controlled” studies in the Salekin meta-analysis is that there is abso- lutely no basis for optimism regarding treat- ment to reduce the risk of criminal or violent recidivism.

Other problematic aspects of the meta- analysis cast further doubt on the author’s optimistic conclusion. As mentioned earlier, most studies in the meta-analysis relied on therapists’ ratings to measure outcome. We consider this inadequate, especially for psy- chopaths. Note that Seto and Barbaree (1999) examined the recidivism of sex of- fenders as a function of psychopathy and progress in treatment, with progress assessed via eight structured therapist ratings. Based on these ratings, which showed good interrater agreement and were undoubtedly more reliable than unstructured impres- sions of therapeutic progress, those offenders with better than average progress were more likely to recidivate violently, and this was es- pecially true for psychopaths. In our opin- ion, therapists’ impressions of clinical prog- ress cannot be defended as an index of treatment effectiveness for offenders, espe- cially psychopaths. Independently measured criminal conduct must be at least part of the outcome for an evaluation of treatment for psychopaths. This requirement eliminates all but a handful of the studies in the Salekin meta-analysis.

Several other categorizations in the Salekin meta-analysis were problematic. For example, Salekin categorized a study by Glaus (1968) as involving cognitive-behav- ioral therapy, with three psychopaths (de- fined by Cleckley’s criteria) all reported to have improved as a function of the therapy. Compared to the 20% Salekin estimated would have improved without treatment,  this was reported as a net treatment benefit of 80%. However, a careful reading of Glaus reveals that the author reported on the his- tory and follow-up of 1,000 criminal psy- chopaths, of which 31 were “fully recovered and socialized” (p. 30). Glaus reported that many more might have improved, but he was unable to find more information (presum- ably despite follow-up efforts). Glaus de- scribed the three aforementioned positive- outcome cases in detail but made no claim that these were representative. Cleckley’s cri- teria were never mentioned, nor was cogni- tive-behavioral therapy (which was only in its infancy in 1968; see Friedman, 1970); the therapy provided was so briefly described that it is impossible to categorize it. The journal editor noted that “the percentage of favorable results observed is low (over 3 per- cent), but the author’s standards of follow- up and cure are unusually high” (p. 35). There is a huge discrepancy between the original author’s report of just above 3% benefit and Salekin’s report of 100%. In  sum, close scrutiny of the studies in the Salekin (2002) meta-analysis reveals a vari- ety of methodological weaknesses that cast serious doubt on its salutary conclusions. Most important, we think more random assignment treatment studies are required be- fore meta-analysis can be informative.

 

Treatment  for  Nonforensic Psychopaths

Few studies reviewed by Salekin (2002) in- cluded offenders or forensic patients. Even   if one could overlook the methodological weaknesses of the meta-analysis and studies included therein and accept its conclusions,  it cannot tell us much about the population  of primary interest—psychopathic offenders. Nonetheless, to be complete, we describe here findings from a recent evaluation of treatment for nonforensic “potentially psy- chopathic” patients (Skeem, Monahan, & Mulvey, 2002) not available at the  time  of the Salekin meta-analysis. Data from the MacArthur Risk Assessment Study were  used to examine the interrelationships among psychopathy (assessed by the PCL- SV), self-reported involvement in treatment (mostly unspecified verbal therapy with or without drugs), and serious subsequent vio- lence (almost all of which was undetected by the criminal justice system). The MacArthur methodology entailed interviews conducted every 10 weeks over a period of 1 year dur- ing which released civil psychiatric patients were asked about their involvement with treatment and violent behavior in the preced- ing period. Skeem and colleagues examined the relationship between violence in each tar- get period and self-reported treatment in the previous period. They  concluded  that,  in  the first 10 postdischarge weeks, potentially psychopathic patients (> 12 on the PCL: Screening Version [PCL:SV]) who partici- pated in more than 6 sessions of therapy (with an average of 11) exhibited less subse- quent violence than those who participated in fewer sessions (the average was 3).

Recognizing that treatment was not as- signed at random, the authors attempted to compensate by deriving a multivariate “pro- pensity for treatment score” based on nine variables associated with the likelihood that subjects would report they had attended more treatment. This score was used as covariate in the aforementioned analysis. The inclusion of the “propensity score” at- tenuated the apparent treatment effect, but it remained statistically significant. While ac- knowledging several limitations of this study,

Skeem and colleagues (2002) inferred that the results provided evidence of an effect of mental health treatment as usual on reducing the violence associated with psychopathy, thus supporting the conclusions of Salekin’s (2002) meta-analysis.

In our view, several methodological prob- lems compromise the conclusions of this study regarding the effectiveness of treat- ment for psychopaths, despite efforts to correct for nonrandom assignment. First, psychopathy, treatment involvement, and vi- olence were all assessed in the same inter- views, leaving open the possibility of un- intended measurement bias in all three constructs. A second issue concerns the num- ber of bivariate comparisons performed in seeking evidence of a treatment effect. Skeem and colleagues reported 10 bivariate com- parisons (two cutoff scores for psychopathy by four time periods, plus the entire follow- up period, presumably), only one of which yielded a statistically significant (p < .10) result4 after the incorporation of the “pro- pensity” covariate. One significant result in 10 is exactly as anticipated by chance alone.

Moreover, Skeem and colleagues’ (2002) use of a “treatment propensity” covariate is questionable in its own right. Miller and Chapman (2001) critiqued the use of covari- ance analysis on the grounds that it capital- izes on regression to the mean, and they as- serted that its use as a method to equate nonrandomly assigned groups was  inappro-

priate. They did acknowledge that a propen- sity score approach (Rosenbaum & Rubin, 1984) might be of assistance but noted that it could not address unobserved differences be- tween groups. Skeem and colleagues cited Rubin (1997) as a source for “propensity” analysis, but did not employ a key aspect of

the method, which involves disaggregating the subjects into subgroups defined by the propensity  variable  or function.

In our view, the Skeem and colleagues (2002) study probably exhibits “creaming intervention selection bias” (Larzelere, Kuhn, & Johnson, 2004), whereby  patients  of lower risk are more likely to receive treat- ment. Moreover, even if one accepts its find- ings, there are other concerns. Skeem and colleagues acknowledged that the civil pa- tients scoring over 12 on the PCL:SV were only “potentially” psychopathic. The study provided    no    information    about effective

components of treatment, and the conclusion that a dozen hours of unspecified therapy re- duced serious violence by psychopaths seems highly questionable. We conclude that this study offers little guidance to those wonder- ing about the efficacy of treatment for psy- chopathy or what therapy is indicated.

 

ALTERNATIVE CONCLUSIONS REGARDING  THE  EFFECTS OF TREATMENT

 

Given that it is such a serious and long-rec- ognized problem, it is surprising that there has been so little good evaluation research on the treatment of psychopathy. Con- sidering the available treatment literature, several alternative conclusions might be en- tertained:

 

• Alternative Conclusion 1. There have already been satisfactory demonstrations of effective treatment(s) for psychopaths (i.e., therapy that causes decreases in criminal and violent behavior), and the appropriate course is to provide such treatment(s) with intensity and integrity to as many psycho- paths as possible. From this perspective, pressing research questions would pertain to the investigation of the conditions that en- sure the successful export and adoption of such treatment(s) throughout the world’s criminal justice systems, and modifications required to apply such treatment to noncrim- inal  and  youthful
• Alternative Conclusion 2. There  have not   been   any   satisfactory demonstrations, but only because adequate and persuasive evaluation work has yet to be done. Effective interventions for psychopaths have already been discovered and applied; it is the persua- sive demonstrations that are lacking. For ex- ample, psychopathic offenders benefit from treatments already shown to be effective for offenders in general, but they require unusu- ally high doses and intensities of such treat- ments in order for them to be effective. From this perspective, the obvious research prior- ity is for rigorous and persuasive empirical demonstrations of the effectiveness of avail- able treatments with psychopaths (with the next step being broader dissemination; viz. Alternative  Conclusion  1).

• Alternative Conclusion    There have been no satisfactory demonstrations because an effective clinical intervention is lacking. Psychopaths are fundamentally different even from other serious offenders, so that— despite available knowledge of what meth- ods are effective for getting nonpsychopathic offenders to desist—no effective interven- tions yet exist for psychopaths. Indeed, some treatments that are effective for nonpsy- chopaths actually increase the risk of repre- sented by psychopaths. Furthermore, the fact that psychopaths and nonpsychopaths are mixed together in most studies is the main reason why it has been so difficult to demon- strate effective treatment for adult offenders overall (i.e., positive treatment effects for nonpsychopaths are diluted or even negated by null or negative effects for the psycho- paths). From this perspective, detailed analy- sis of the characteristics of psychopaths (in- side and outside the laboratory) is needed to inform the design of new and effective inter- ventions tailored to this unique population.

• Alternative Conclusion 4. No clinical intervention will ever be effective. Psycho- paths are  qualitatively  different  from other offenders but do not have deficits or impair- ment in any standard clinical sense. From this standpoint, the entire clinical enterprise is fundamentally unsuited to interventions to reduce the harm perpetrated by psychopaths. All that can be hoped for is a set of strategies to limit the harm by psychopaths by con- straining their activities and  opportunities.

 

It should be noted that these alternatives are not entirely mutually exclusive. For ex- ample, even if one concluded that a dozen sessions of mental health service as usual (Skeem et al., 2002) had actually reduced psychopathic violence (Alternative Conclu- sion 1), one would be unable specify the op- erative elements of that treatment, which would necessitate following the implications of Alternative Conclusion 2. Similarly, the enterprise that follows from Alternative Conclusion 3 of finding new therapies founded on an examination of the funda- mental features of psychopathy could still be worthwhile even if some effective treatments had already been discovered. However, to the extent that one accepts Alternative Con- clusions 1 or 2, one would probably assign lower priority to this task of developing new therapies.

In the final analysis, we adopt a blend of Alternative Conclusions 3 and 4. We believe, as outlined in Alternative Conclusion 4 (and explained further later), that the available evidence implies that psychopaths do not have deficits in the biological or medical sense. We propose that findings from outside the literature on treating psychopathy war- rant serious consideration in designing inter- ventions for psychopaths. We believe the evi- dence favors applying behavioral principles to reducing the harm occasioned by psy- chopathy. Our belief is based partly on empirical evidence that this approach has worked with some offender and violent pop- ulations (although effectiveness with psycho- paths remains to be demonstrated). Our be- lief in the value of behavioral strategies for treating psychopaths also reflects a theoreti- cal perspective that views psychopathy as a nonpathological condition, a reproductively viable life strategy. Next, we outline our evolutionary perspective on psychopathy to highlight  implications  for interventions.

 

A NONPATHOLOGICAL, SELECTIONIST ACCOUNT OF PSYCHOPATHY

There is evidence that psychopathy, unlike many psychological constructs, is underlain by a natural discontinuity or taxon (Ayers, 2000; Harris, Rice, & Quinsey, 1994, Has- lam, 2003; Skilling, Harris, Rice, & Quinsey, 2002; Skilling, Quinsey, & Craig, 2001). By this view, scores on the best measure of psy- chopathy, the PCL-R, appear continuous be- cause the identification of indicators and scoring are imperfect. Perfect measurement would, in theory, reveal just two possibili- ties—an individual either is or is not a true psychopath. Although not unanimous (Marcus, John, & Edens, 2004), the evidence supports the idea that psychopathy is a taxon.

The evidence on taxonicity, our research on treatment and the prediction of recidi- vism (Harris & Rice,  in  press;  Harris,  Rice, & Cormier, 1991; Rice & Harris, 1995) all suggest to us that psychopathy exists because it was a reproductively viable life strategy during human evolution. Adaptations (in- cluding those with psychological effects) were selected because they increased inclusive fitness in ancestral environments. For example, being in a cohesive, mutually sup- portive (“reciprocally altruistic”) group was adaptive and heritable inclinations favoring group solidarity and adherence to rules have been  associated  with  human  reproduc-  tive success (Dawkins, 1978; Ridley, 1997). However, we (Harris, Skilling, & Rice, 2001; see also Mealey, 1995; Seto & Quinsey, Chapter 30, this volume) hypothesize that such a general strategy created a niche for  an alternative cheating (i.e., psychopathic) strategy. When effective, this strategy is es- pecially selfish, callous, manipulative, and lacking in empathy. If many people were cheaters, however, the strategy would lose its effectiveness due to the difficulty finding co- operators to exploit and the increased vigi- lance of remaining cooperators. Thus, the two strategies are expected to be frequency dependent, with cheating/psychopathy at low prevalence.

We hypothesize that high mating effort (i.e., promiscuous sexual behavior and many short-term marital relationships), and espe- cially the willingness to employ deception and coercion, glibness, and charm, were (and are) also part of the psychopathic life strat- egy. Belsky, Steinberg, and Draper (1991) ar- gued that a high mating effort life strategy is characterized by insecure attachment to par- ents and childhood behavior problems, fol- lowed by early puberty and precocious sex- ual behavior, and then unstable adult pair bonding and low parental investment. Psy- chopathy, we suggest, represents a geneti- cally determined life strategy that has been maintained in the population through its re- lationship with reproductive success (Barr & Quinsey, 2004; Harris, Skilling, &  Rice,  2001; Lalumière, Harris, Quinsey, & Rice, 2005;  Rice,  1997).

The evidence on the neurocognitive char- acteristics of psychopaths (reviewed in this volume) reveals the condition to be an en- during set of traits that can be conceived of as aspects of personality or as differences in the form, manner, and relative speed of pro- cessing information. Key for this selectionist account is that these traits endure from situation to situation across the lifespan. Situations vary in the degree to which they differentiate between psychopaths and non- psychopaths, but, by this account, reinforce- ment  and  punishment  operate  for  psychopaths as they do for everyone else, although what constitute reinforcers and punishers might  differ.

Because psychopathy exhibits substantial heritability (reviewed in Waldman & Rhee, Chapter 11, this volume), the most straight- forward and parsimonious version of the evolutionary account is that psychopaths have executed a “healthy” (in the biomed- ical5 but not moral sense) obligate strategy. Subtle neuroanatomical and neurochemical differences (without gross lesions) are consis- tent with this hypothesis. As well, it is ex- pected that special tests would reveal that psychopaths act relatively impulsively, fear- lessly, and unempathically and are resistant to punishment under some laboratory condi- tions but are not grossly disadvantaged. Psy- chopathy should also be associated with en- hanced performance on some tasks. This account of psychopathy is consistent with  the observation that it is peculiar for disor- ders to enhance any ability (such as conning and manipulation, Blair, personal communi- cation,   May 2000).

We have tested this account by examining several indicators of neurodevelopmental problems associated with psychiatric dis- orders (obstetrical and perinatal problems, medical problems in infancy, learning dis- ability, etc.) and found them to be related to violent crime but unrelated or inversely related   to   violent   offenders’    PCL-R scores (Harris, Rice, & Lalumière, 2001; Lalumière, Harris, & Rice, 2001). Although each of neurodevelopmental problems and psychopathy were associated with having had antisocial, negligent, and abusive par- ents, each appeared to be an independent cause of violent crime. Nonpsychopathic of- fenders exhibited more fluctuating asymme- try (an index of biomedical health) than psy- chopaths who themselves were not different from healthy volunteers (Lalumière et al., 2001). Finally, among sex offenders, those who preferentially target “reproductively viable” victims (i.e., postpubertal females) have significantly higher PCL-R scores than those who target all other classes of people (Harris, Hilton, Lalumière, Quinsey, & Rice, 2004). We are unaware of another hypothe- sis about sex offenders or psychopathy that accounts for this widely known difference.

Thus, there might be two distinct paths to serious, chronic criminality—one  associated with psychopathy and one (associated with less extensive crime and for which some treatments are effective) caused by develop- mental neuropathology and low embodied capital. If this nonpathological interpreta- tion of psychopathy is correct, there are im- plications  for intervention.

 

INTERVENTIONS FOR  PSYCHOPATHS?

 Is psychopathy likely to respond to very in- tense forms of the treatment that works with nonpsychopaths? The most straightforward implication of a dimensional view of psy- chopathy is that a high-intensity version of what has been shown to be effective with of- fenders in general would be effective for psy- chopaths. This would amount to a cognitive- behavioral program incorporating relapse prevention to combat substance abuse, anger management to control expressive aggres- sion, prosocial modeling to break down an- tisocial thinking and values, and motiva- tional interviewing to enhance commitment to treatment (Wong & Hare, 2005). The em- pirical literature supporting this approach for seriously violent adult offenders (Rice & Harris 1997) is as yet quite limited (and non- existent for psychopaths). Thus, this ap- proach needs to be further implemented and evaluated, specifically with psychopathic of- fenders. However, by our taxonic, nonpath- ological account of psychopathy, we believe more success might come from identifying different approaches. These are described in the remaining subsections in this chapter.

 

Behavior  Modification

Meta-analyses of intervention studies have been informative with regard to the treat- ment of offenders. Lipsey (1992; see also Lipsey & Wilson, 1998) examined almost 400 evaluations of interventions for juvenile delinquency and reported a small statistically significant effect. Effects were larger to the extent that interventions were behavioral and oriented toward building skills. Even more broadly, Lipsey and Wilson (1993) conducted a meta-analysis of over 300 meta- analytic evaluations of human service inter- ventions. Again, there was a moderate signif- icant overall effect size, and, as far as can be

determined, behavioral interventions yielded effects larger than average and larger than the average for medical interventions. That properly implemented behavioral contingen- cies cause parallel changes in behavior is in- controvertible.6 There are debates concern- ing the mechanisms underlying punishment, the best ways to promote generalization, the effect of reinforcement on intrinsically re- warding behavior, and so on, but there is no doubt that behavior (whether  pathological or not) responds predictably to its con- sequation (e.g., Corrigan & Muesser, 2000; Foxx, 2003; LePage et al., 2003; Lovaas,  1987; Paul & Lentz, 1977; Stein,  1999; Wong, Woolsey,  Innocent,  &  Liberman, 1988; a longer list is available from the au- thors) while contingencies are in effect.

In no sense are we arguing that any of the foregoing provides evidence for a treatment effect among psychopaths. However, in gen- eral, behavioral treatments have the virtue of being explicitly designed for use under con- ditions in which the cause of the distressing behavior is unknown or cannot be specified (or is known, but cannot be altered). Furthermore, there is a technology that facil- itates the implementation of behavioral treatment across an entire facility or agency—namely, the token economy system (Morris & Braukmann, 1987). Unlike other therapeutic approaches, psychopathy does not appear to present special problems for the effectiveness of a token economy (Pickens, Erickson, Thompson, Heston, & Eckert,  1979).

 

Multisystemic Therapy

The second impressive and persuasive litera- ture on interventions for offenders concerns multisystemic therapy (MST) for juvenile de- linquents (Brown, Borduin, & Henggeler, 2001; Brown et al., 1997; Randall & Cunningham, 2003). Theoretically, adoles- cent criminality is a systems problem: Ado- lescents engage in crime when responding naturally to the systems in which they op- erate. Dysfunctional families, ineffective schools, and antisocial peers combine to pro- duce the obvious result—delinquency. MST seeks to alter each system to build functional school and family systems. In practice, MST  is very individual and flexible with several general  features:  building  skills,   especially for parents; emphasis on monitoring and consequation both for adolescents and par- ents; behavioral principles (positive rein- forcement; promoting behaviors incom- patible with antisociality; emphasizing specific, observable, active behaviors; con- cern with generalization); and ensuring ther- apeutic integrity and adherence (Henggeler, Cunningham, Pickrel, Schoenwald, & Bron- dino, 1996; Henggeler, Melton, Brondino, Scherer, & Hanley, 1997; Henggeler, Schoenwald, & Pickrel, 1995). Most impor- tant, MST has yielded large treatment effects in  randomized  controlled  trials   (Borduin et al., 1995; Borduin, Schaeffer, Ronis, & Scott,  2003).

For our present purposes, we recognize that the work on MST provides no evidence of a treatment effect for psychopaths or for adult offenders. In fact, its developers ac- knowledge that it cannot easily be applied to adults (Borduin, personal communication, August 2003). Moreover, our selectionist hy- pothesis about psychopathy (Harris, Skil- ling, & Rice, 2001) assigns little direct causal influence to antisocial peers: Psychopaths have more antisocial friends, but as a result of psychopathy, not as a cause. However, our hypothesis does maintain that psychopathic behavior is occasioned by opportunities fa- vorable for its occurrence and that behavior- al monitoring and consequation could re- duce antisocial conduct by psychopaths by reducing its payoff. Our point here is that the evidence supporting MST as a treatment for delinquency is so promising that we can look past its theoretical underpinnings (Burns, Schoenwald, Burchard, Faw,  &  Santos, 2000; Huey, Henggeler, Brondino, & Pickrel, 2000) and move on to evaluate its efficacy when applied to offender groups for which it was  not  specifically designed.

 

Institutional  and  Community Programs

Where psychopaths have already committed serious offenses and exhibit evidence of high risk for future violence, we favor the use of selective incapacitation in the form of long- term institutionalization. Regardless of the duration of incapacitation, some organiza- tional system must be in place within the in- stitution. To this end, we favor the appli- cation   of   a   sophisticated   token  economy

incorporating four main features. First, the program is completely explicit and concen- trates on reinforcement of behaviors incom- patible with psychopathic conduct (i.e., de- laying gratification, telling the truth, being responsible, being helpful and cooperative— each tied to an appropriate operational defi- nition) and penalties for impulsive, dishon- est, aggressive, irresponsible, and, of course, criminal actions. Second, there is no expecta- tion that the program will be completed or withdrawn; the program  is  only  expected  to be efficacious under conditions of con- tinuous administration. Third, contingencies are tightly monitored by institutional staff, based only on observed, overt behavior, and never based on what inmates report about thoughts, feelings, or conduct. Fourth, sys- tems are in place to monitor and consequate performance by front-line and supervisory program  staff.

It must be recognized that societal and economic conditions would permit use of  this incapacitation strategy with a minority of psychopaths (and a small minority of of- fenders). For most psychopathic offenders, release to the community in the form of pa- role or probation is inevitable. In our opin- ion, greater prospects for effective interven- tion lie in applying continuing behavioral principles to psychopaths under conditional release. Quite clearly, it will not be easy to design and implement a behavioral program for the institutional management of psycho- pathic offenders. It is to be expected that psychopathic  offenders  would  resist  such  a program, break the rules in unexpected ways, seek to undermine institutional secu- rity, and engage in attempts to deceive and manipulate staff, supervisors, volunteers, the media, and members of the public.

The challenges associated with operating a program for psychopathic offenders should not be underestimated, but  implementing  an institutional program will be straight- forward compared to delivering a similar behavioral intervention for psychopathic of- fenders under community supervision. We suggest, however, that the same principles should apply to community-release pro- grams—behavioral monitoring, positive consequation, ensuring program integrity, and an emphasis on observable behavior. Participation  in  such  programs  would need to be a condition for release; otherwise, few psychopaths would volunteer for and persist in such a program. We anticipate that pro- grams of this sort will require more resources than customary parole or probation services, especially because the program is expected to be efficacious only as  long  as  it  continues to be administered. Nevertheless, given the broad societal harm caused by psychopaths, we believe an evaluation of such a program could show it to be cost-effective.

 

Protecting  Potential Victims

The aforementioned suggested interventions are expected to reduce the violent and crimi- nal behavior of psychopaths by  shrinking  the behavioral niche. By our selectionist ac- count, psychopaths (like everyone) are sensi- tive to the features of the interpersonal envi- ronment favoring one behavior over another. To the extent that a particular behavior does not (or appears unlikely to) pay off, we ex- pect its frequency to decline. Because hu- mans exhibit excellent discrimination, we do not expect such behavioral changes to gener- alize to a postprogram environment because it would be obvious that the niche had changed. However, one might also ask: Rather than simply addressing the behavior of psychopaths, why not change the social environment itself? Some approaches of this kind have been tried with other populations.

Wassermann and Miller (1998; see also Catalano, Arthur, Hawkins, Berglund, & Olson, 1998) evaluated outcome data for several universal programs for preschool and school-age children and concluded that such programs can positively affect outcomes plausibly or empirically related to later antisociality. Programs targeting at-risk ado- lescents appear to reduce delinquent conduct (e.g., Tolan & Guerra, 1994). Similarly, in- creasing school supervision, boosting police patrols, installing surveillance cameras, us- ing metal detectors, promoting neighbor- hood watch and citizen patrols, restricting access to firearms, increasing access to abor- tion, restricting citizens’ freedom to move to relocate, and so on (cf. Catalano et al., 1998) can all be expected to shrink the opportunity for harm due to psychopathy. Of course, no one can say whether the reductions in antiso- cial  conduct  achieved  by  such broad-based

interventions reflect differences in the small minority of youth who become psychopaths. Nevertheless, on theoretical grounds, pop- ulation-based interventions that (whatever else they do) decrease the opportunity for psychopathic aggression and exploitation can be expected to be worthwhile.

Finally, one might advocate explicit teach- ing about psychopathy in school and in pub- lic education campaigns. Such campaigns do appear to have had salutary effects in im- proving safety-related behaviors (safe sex, seatbelt use, decreasing smoking, increasing cancer screening, etc.). What is somewhat less obvious, however, is the specific content of training aimed at reducing the harm caused by psychopaths. For example, effec- tively instructing people to distrust strangers, telling young women that young men only want one thing, and advising everyone that leopards never change their spots are all ap- proaches that might decrease the niche for psychopathy, but at such large social costs that benefits would be outweighed. More fo- cused instructional approaches are probably desirable. Of course, similar concerns apply to tactics described in the previous para- graph. For example, how much police sur- veillance should law-abiding citizens tolerate in order to diminish the harm caused by psy- chopaths and other offenders? In our view, there is probably a trade-off in that restric- tions on law enforcement agencies’ security precautions necessarily increase the niche fa- vorable  to psychopathy.

Perhaps the following words of guidance, which we would give to novice forensic clini- cians, could be a starting point for all safe re- lationships:

 

1. Read Hare (1998).
2. Reputation matters; leopards seldom change their
3. Never take an offender’s word at face value; always check his assertions against the record and with other
4. Don’t just attend to how he behaves to- ward you; carefully observe how he treats everyone—peers and other
5. Beware of
6. Be very suspicious if an offender asks you to break a rule, no matter how minor, or to keep  an  illicit
7. Talk to   a   colleague   about   your   relationship with him; if your trusted col- league says things don’t sound right, be- ware.

 

CONCLUSIONS

 We believe there is no evidence that any treatments yet applied to psychopaths have been shown to be effective in reducing vio- lence or crime. In fact, some treatments that are effective for other offenders are actually harmful for psychopaths in that they appear to promote recidivism. We believe that the reason for these findings is that psychopaths are fundamentally different from other of- fenders and that there is nothing “wrong” with them in the manner of a deficit or im- pairment that therapy can “fix.” Instead, they exhibit an evolutionarily viable life strategy that involves lying, cheating, and manipulating others.

Although no therapy has yet been shown to reduce the likelihood of future violence or crime among psychopaths, this does not mean that nothing can help. The best avail- able evidence for effective intervention co- mes from the application of social learning principles in the form of behavioral pro- grams and from MST. We believe that the strongest evidentiary support exists for insti- tutional incapacitation where practical, and in tightly controlled behavioral programs with contingencies that remain in effect both inside and outside the institution. We can also conceive of societal changes that might reduce the behavioral niche for psychopathy, but such changes inevitably carry some nega- tive impact with respect to the personal lib- erty of all citizens. Finally, none of these  ideas comes close to a solution or cure for  the societal harm caused by psychopathy. It is to be expected from our nonpathological, selectionist perspective that psychopaths will attempt to subvert harm reduction strategies employed by nonpsychopaths. In the ongo- ing arms race, the existing literature only suggests ways to limit psychopaths’ advan- tages. More complete solutions lie in inter- ventions based on future advances in basic neuroscience and molecular genetics (see MacDonald & Iacono, Chapter 19, and Seto  &  Quinsey,  Chapter  30,  this  volume).

NOTES

 

1. One other study reported a meta-analysis of treatment for psychopaths (Garrido, Esteban, & Molero, 1995). The authors said there were two separate meta-analyses. The first included 34 studies that examined treatment outcomes for psychopaths compared to nonpsychopaths and purportedly showed that outcomes for psychopaths were worse than those for non- psychopaths. The second included 19 studies that examined pre- and posttreatment studies of psychopaths and purportedly showed that psychopaths “are able to improve in behavior- al and psychological functioning” (p. 59). Be- cause no references were included in the arti- cle, there is no way to critically examine the methodology.
2. Salekin’s definition of “improved” was some- what unusual. For example, in the case of criminal behavior, he counted those who did not recidivate in the follow-up period as hav- ing “improved” regardless of how long the fol- low-up period
3. Salekin does not name the eight studies he counted as “controlled” in the meta-analysis. In a personal communication (May 2004), he advised that the eight were Craft, Stephenson, and Granger (1964); Ingram, Gerard, Quay,  and Levinson (1970); Korey (1944); Maas, (1966); Persons (1965); Rice et al. (1992); Skolnick and Zuckerman (1979); and Woody, McLellan,  Luborsky,  and  O’Brien  (1985).
4. Skeem et    reported  a  chi-square  value  of

3.31 as significant, p < .05. However,  the use   of a one-tailed procedure is clearly unwar- ranted in examining therapy that according to the authors themselves is of doubtful effective- ness, and might even in some instances be harmful.

5. This argument relies on a particular definition of pathology or “disorder” (Wakefield, 1992) which says disorders involve the failure of a mechanism to perform as designed by natural selection. Because this account asserts that it exists because it has been reproductively suc- cessful (i.e., it was designed by natural selec- tion) psychopathy is, by definition, not a
6. Readers might wonder why we consider single case studies persuasive regarding the effects of behavior modification but not with respect to the benefits of psychodrama. The reason is that single-case designs typical of the eval- uation of behavioral treatment incorporate considerable methodological control (e.g., ob- jective measurement, multiple baselines, and reversal designs) rarely seen in the informal, impressionistic evaluation of nonbehavioral therapies.

 

 

 

REFERENCES

 Andrews, D. A., & Bonta, J. (1994). The psychology of criminal conduct. Cincinnati, OH: Anderson.

Andrews, D. A., Zinger, I., Hoge, R. D., Bonta, J., Gendreau, P., & Cullen, F. T. (1990). Does correc- tional treatment work? A clinically relevant and psy- chologically informed meta-analysis. Criminology, 28, 369–404.

Ayers, W. A. (2000). Taxometric analysis of borderline and antisocial personality disorders in a drug and al- cohol dependent population. Dissertation Abstracts International: Section B: The Sciences and Engi- neering,  61, 1684.

Barbaree, H. E., Peacock, E. J., Cortini, F., Marshall, W. L., & Seto, M. (1998). Ontario penitentiaries’ pro- gram. In W. L. Marshall, Y. M. Fernandez, S. M. Hudson, & T. Ward (Eds.), Sourcebook of treatment programs for sexual offenders. New York: Plenum Press.

Barker, E. (1980). The Penetanguishene Program: A per- sonal review. In H. Toch (Ed.), Therapeutic commu- nities in corrections (pp. 73–81). New York: Praeger.

Barker, E. T., & Mason, M. H. (1968). Buber behind bars. Canadian Psychiatric Association Journal, 13, 61–72.

Barker, E. T., Mason, M. H., & Wilson, J. (1969). De- fence-disrupting therapy. Canadian Psychiatric Asso- ciation Journal, 14, 355–359.

Barker, E. T., & McLaughlin, A. J. (1977). The total en- counter capsule. Canadian Psychiatric Association Journal, 22, 355–360.

Barr, K. N., & Quinsey, V. L. (2004). Is psychopathy  a pathology or a life strategy? Implications for so- cial policy. In C. Crawford & C. Salmon (Eds.), Evolutionary psychology, public policy, and per- sonal decisions (pp. 293–317). Hillsdale, NJ: Erl- baum.

Beacher, A. I. (1962). Psychoanalytic treatment of a sociopathy in a group situation. American Journal of Psychotherapy,  16, 278–288.

Belsky, J., Steinberg, L., & Draper, P. (1991). Childhood experience, interpersonal development, and repro- ductive strategy: An evolutionary theory of socializa- tion. Child Development, 62, 647–670.

Borduin, C. M., Mann, B. J., Cone, L. T., Henggeler, S.

W., Fucci, B. R., Blaske, D. M., et al. (1995). Multisystemic treatment of serious juvenile offend- ers: Long-term prevention of criminality and vio- lence. Journal of Consulting and Clinical Psychology, 63, 569–578.

Borduin, C. M., Schaeffer, C. M., Ronis, M., & Scott, T. (2003). Multisystemic treatment of serious behavior in adolescents. In C. A. Essau (Ed.), Conduct and oppositional defiant disorders: Epidemiology, risk factors, and treatment (pp. 299–318). Mahwah, NJ: Erlbaum.

Brown, H. J., & Gutsch, K. U. (1985). Cognitions asso- ciated with a delay of gratification task: A study with

psychopaths and normal prisoners. Criminal Justice and  Behavior,  12, 453–462.

Brown, T. L., Borduin, C. M., & Henggeler, S. W. (2001). Treating juvenile offenders in community set- tings. In J. B. Ashford, B. D. Sales, & W. H. Reid, Treating adult and juvenile offenders with special needs (pp. 445–464). Washington, DC: American Psychological  Association.

Brown, T. L., Swenson, C. C., Cunningham, P. B., Henggeler, S. W., Schoenwald, S. K., & Rowland, M.

238. (1997). Multisystemic treatment of violent and chronic juvenile offenders: Bridging the gap between research and practice. Administration and Policy in Mental Health, 25, 221–238.

Burns, B. J., Schoenwald, S. K., Burchard, J. D., Faw, L., & Santos, A. B. (2000). Comprehensive community- based interventions for youth with severe emotional disorders: Multisystemic therapy and the wrap- around process. Journal of Child and Family Studies, 9, 283–314.

Catalano, R. F., Arthur, M. W., Hawkins, D. J., Berglund, L., & Olson, J. J. (1998). Comprehensive community- and school-based interventions to prevent antisocial behavior. In R. Loeber & D. P. Farrington (Eds.), Seri- ous and violent juvenile offenders: Risk factors and successful interventions (pp. 248–283). Thousand Oaks, CA: Sage.

Cleckley, H. (1941). The mask of sanity. St. Louis, MO: Mosby.

Cleckley, H. (1982). The mask of sanity (6th ed., rev.).

St.  Louis,  MO: Mosby.

Copas, J. B., O’Brien, M., Roberts, J., & Whiteley, J. S. (1984). Treatment outcome in personality disorder: The effect of social psychological and behavioural variables. Personality and Individual Differences, 5, 565–573.

Copas, J. B., & Whiteley, J. S. (1976). Predicting success in the treatment of psychopaths. British Journal of Psychiatry,  129, 388–392.

Corrigan, P. W., & Mueser, K. T. (2000). Behavior ther- apy for aggressive psychiatric patients. In M. L. Crowner (Ed.), Understanding and treating violent psychiatric patients (pp. 69–85). Washington, DC: American  Psychiatric Association.

Corsini, R. J. (1958). Psychodrama with a psychopath.

Group Psychotherapy, 11, 33–39.

Craft, M., Stephenson, G., & Granger, C. (1964). A controlled trial of authoritarian and self- governing regimes with adolescent psychopaths. American Journal of Orthopsychiatry, 543–554.

Dawkins, R. (1978). The selfish gene. London: Paladin

Books.

Dolan, D. (1998). Therapeutic community treatment for severe personality disorders. In T. Millon, E. Simonsen, M. Birket-Smith, & R. D. Davis (Eds.), Psychopathy: Antisocial, criminal and violent behav- ior. New York: Guilford Press.

Foxx, R. M. (2003). The treatment of dangerous behav- ior.  Behavioral  Interventions,  18, 1–21.

Friedman, P. H. (1970). Limitations in the conceptual- ization of behavior therapists: Toward a cognitive- behavioral model of behavior therapy. Psychological Reports,  27, 175–178.

Garrido, V., Esteban, C., & Molero, C. (1995). The ef- fectiveness in the treatment of psychopathy: A meta- analysis. Issues in Criminological and Legal Psychol- ogy,  24, 57–59.

Glaus, A. (1968). The handling of criminal psychopaths in Switzerland. International Journal of Offender Therapy,  12, 29–36.

Hare, R. D. (1970). Psychopathy: Theory and research.

New  York: Wiley.

Hare, R. D. (1991). The Psychopathy Checklist— Revised. Toronto, ON, Canada: Multi-Health Sys- tems.

Hare, R. D. (1998). Without conscience: The disturbing world of the psychopaths among us. New York: Guilford Press.

Hare, R. D. (2003). The Psychopathy Checklist—Re- vised (2nd ed.). Toronto, ON, Canada: Multi-Health Systems.

Hare, R. D., Clark, D., Grann, M., & Thornton, D. (2000). Psychopathy and the predictive validity of the PCL-R: An international perspective. Behavioral Sciences and the Law, 18, 623–645.

Hare, R. D., & Frazelle, J. (1985). Some preliminary notes on a research scale for the assessment of psychopathy in criminal populations. Unpublished manuscript, University of British Columbia, Vancou-

ver.

Harris, G. T., Hilton, N. Z., Lalumière, M. L., Quinsey,

1. L., & Rice, M. E. (2004). Precocious coercive sex- uality in life course persistent antisociality. Manu- script in preparation.

Harris, G. T., & Rice, M. E. (in press). Psychopathy re- search at Oak Ridge: Skepticism overcome. In H. Herve & J. C. Yuille (Eds.), Psychopathy: Theory, re- search, and implications for society. Mahwah, NJ: Erlbaum.

Harris, G. T., Rice, M. E., & Cormier, C. A. (1991). Psy- chopathy and violent recidivism. Law and Human Behavior, 15, 625–637.

Harris, G. T., Rice, M. E., & Cormier, C. A. (1994). Psy- chopaths: Is a therapeutic community therapeutic? Therapeutic Communities, 15, 283–300.

Harris, G. T., Rice, M. E., & Lalumière, M. (2001). Criminal violence: The roles of psychopathy, neuro- developmental insults and antisocial parenting. Criminal Justice and Behavior, 28, 402–426.

Harris, G. T., Rice, M. E., & Quinsey, V. L. (1994). Psy- chopathy as a taxon: Evidence that psychopaths are a discrete class. Journal of Consulting and Clinical Psy- chology, 62, 387–397.

Harris, G. T., Skilling, T. A., & Rice, M. E. (2001). The construct of psychopathy. In M. Tonry (Ed.), Crime and justice: An annual review of research (pp. 197– 264). Chicago: University of Chicago Press.

Haslam, N. (2003) The dimensional view of personality

disorders: A review of the taxometric evidence. Clini- cal  Psychology  Review,  23, 75–93.

Henggeler, S. W., Cunningham, P. B., Pickrel, S. G., Schoenwald, S. K., & Brondino, M. J. (1996). Multisystemic therapy: An effective violence preven- tion approach for serious juvenile offenders. Journal of Adolescence,  19, 47–61.

Henggeler, S. W., Melton, G. B., Brondino, M. J., Scherer, D. G., & Hanley, J. H. (1997). Multisystemic therapy with violent and chronic juvenile offenders and their families: The role of treatment fidelity in successful dissemination. Journal of Consulting and Clinical Psychology, 65, 821–833.

Henggeler, S. W., Schoenwald, S. K., & Pickrel, S. G. (1995). Multisystemic therapy: Bridging the gap be- tween university- and community-based treatment. Journal of Consulting and Clinical Psychology, 63, 709–717.

Hobson, J., Shine, J., & Roberts, R. (2000). How do psychopaths behave in a prison therapeutic commu- nity? Psychology, Crime and Law, 6, 139–154

Huey, S. J., Henggeler, S. W., Brondino, M. J., & Pickrel,

467. G. (2000). Mechanisms of change in multisystemic therapy: Reducing delinquent behavior through ther- apist adherence and improved family and peer func- tioning. Journal of Consulting and Clinical Psychol- ogy, 68, 451–467.

Hughes, G., Hogue, T., Hollin, C., & Champion, H. (1997). First-stage evaluation of a treatment programme for personality disordered offenders. Journal of Forensic Psychiatry, 8, 515–527.

Ingram, G. L., Gerard, R. E., Quay, H. C., & Levinson,

30. B. (1970). An experimental program for the psy- chopathic delinquent: Looking in the “correctional wastebasket.” Journal of Research in Crime and De- linquency, 7, 24–30.

Jones, M. (1956). The concept of the therapeutic com- munity. American Journal of Psychiatry, 112, 647– 650.

Jones, M. (1968). Social psychiatric in practice, Harmondsworth, UK: Penguin Books.

Kiger, R. S. (1967). Treating the psychopathic patient in a therapeutic community. Hospital and Community Psychiatry,  18, 191–196.

Knight, K., Simpson, D. D., & Miller, M. L. (1999). Three-year incarceration outcomes for in-prison therapeutic community treatment in Texas. The Prison Journal, 79, 337–351.

Korey, S. R. (1944). The effects of Benzedine sulfate on the behavior of psychopathic and neurotic juvenile delinquents. Psychiatric Quarterly, 18, 127–137.

Lalumière, M. L., Harris, G. T., Quinsey, V. L., & Rice,

1. E. (2005). The causes of rape: Understanding in- dividual differences in the male propensity for sexual aggression. Washington, DC: American Psychologi- cal Association.

Lalumière, M. L., Harris, G. T., & Rice, M. E. (2001). Psychopathy and developmental instability. Evolu- tion and Human Behavior,  22, 75–92.

Larzelere, R. E., Kuhn, B. R., & Johnson, B. (2004). The intervention selection bias: An underrecognized con- found in intervention research. Psychological Bulle- tin, 130, 289–303.

LePage, J. P., DelBen, K., Pollard, S., McGhee, M., VanHorn, L., Murphy, J., et al. (2003). Reducing as- saults on an acute psychiatric unit using a token economy: A 2-year followup. Behavioral Interven- tions,  18, 179–190.

Lipsey, M. W. (1992). The effect of treatment on juve- nile delinquents: Results from meta-analysis. In F. Loesel & D. Bender (Eds.), Psychology and law: In- ternational perspectives (pp. 131–143). Oxford, UK: Walter  De Gruyter.

Lipsey, M. W., & Wilson, D. B. (1993). The efficacy of psychological, educational, and behavioral treat- ment: Confirmation from meta-analysis. American Psychologist,  48, 1181–1209.

Lipsey, M. W., & Wilson, B. D. (1998). Effective inter- vention for serious juvenile offenders: A synthesis of research. In R. Loeber & D. P. Farrington (Eds.), Se- rious and violent juvenile offenders: Risk factors and successful interventions (pp. 313–345). Thousand Oaks,  CA: Sage.

Lösel, F. (1998). Treatment and management of psycho- paths. In D. J. Cooke, R. D. Hare, & A. E. Forth (Eds.), Psychopathy: Theory, research, and implica- tions for society. Dordrecht, The Netherlands: Kluwer.

Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in young au- tistic children. Journal of Consulting and Clinical Psychology,  55, 3–9.

Maas, J. (1966). The use of actional procedures in group psychotherapy with sociopathic women. Interna- tional Journal of Group Psychotherapy, 16, 190–197. Maier, G. J. (1976). Therapy in prisons. In J. R. Lion &

1. J. Madden (Eds.), Rage, hate, assault and other forms of violence (pp. 113–133). New York: Spec- trum.

Marcus, D. K., John, S. L., & Edens, J. F. (2004). A taxometric analysis of psychopathic personality. Journal of Abnormal Psychology, 113, 626–635.

McCord, J. (1982). Parental behavior in the cycle of ag- gression. Psychiatry, 51, 14–23.

McMurran, M., Egan, V., & Ahmadi, S. (1998). A ret- rospective evaluation of a therapeutic community for mentally disordered offenders. Journal of Forensic Psychiatry, 9, 103–113.

Mealey, L. (1995). The sociobiology of sociopathy: An integrated evolutionary model. Behavioral and Brain Sciences,  18, 523–599.

Miles, A. E. (1969). The effects of therapeutic commu- nity on the interpersonal relationships of a group of psychopaths. British Journal of Criminology, 9, 22– 38.

Miller, G. A., & Chapman, J. P. (2001). Misunderstand- ing analysis of covariance. Journal of Abnormal Psy- chology, 110, 40–48.

Morris, E. K., & Braukmann, C. J. (Eds.). (1987). Behavioral approaches to crime and delinquency: A handbook of application, research, and concepts. New York: Plenum Press.

Nielsen,  R.  F.  (2000).  Total encounters:  The  life and times of the Mental Health Centre Penetanguishene. Hamilton, ON, Canada: McMaster University Press.

Ogloff, J., Wong, S., & Greenwood, A. (1990). Treating criminal psychopaths in a therapeutic community program. Behavioral Sciences and the Law, 8, 81–90. Paul, G. L., & Lentz, R. J. (1977). Psychosocial treat- ment of chronic mental patients: Milieu versus social learning programs. Cambridge, MA: Harvard Uni-

versity Press.

Persons, R. W. (1965). Psychotherapy with sociopathic offenders: An empirical evaluation. Journal of Clini- cal  Psychology,  21, 205–207.

Pickens, R., Errickson, E., Thompson, T., Heston, L., & Eckert, E. D. (1979). MMPI correlates of perfor- mance on a behavior therapy ward. Behavior Re- search and Therapy, 17, 17–24.

Randall, J., & Cunningham, P. B. (2003). Multisystemic therapy: A treatment for violent substance-abusing and substance-dependent juvenile offenders. Addic- tive Behaviors, 28, 1731–1739.

Reiss, D., Meux, C., & Grubin, D. (2000). The effect of psychopathy on outcome in high security patients. Journal of the American Academy of Psychiatry and Law,  28, 309–314.

Rice, M. E. (1997). Violent offender research and impli- cations for the criminal justice system. American Psy- chologist,  52,  414–423.

Rice, M. E., & Harris, G. T. (1995). Psychopathy, schizophrenia, alcohol abuse and violent recidivism. International Journal of Law and Psychiatry, 18, 333–342.

Rice, M. E., & Harris, G. T. (1997). Treatment of adult offenders. In D. Stoff, J. Breiling, & J. D. Maser (Eds.), Handbook of antisocial behavior (pp. 425– 435). New York: Wiley.

Rice, M. E., Harris, G. T., & Cormier, C. (1992). A fol- low-up of rapists assessed in a maximum security psychiatric facility. Journal of Interpersonal Vio- lence, 5, 435–448.

Richards, H. J., Casey, J. O., & Lucente, S. W. (2003). Psychopathy and treatment response in incarcerated female substance abusers. Criminal Justice and Behavior, 30, 251–276.

Ridley, M. (1997). The origins of virtue. New York: Vi-

king Press.

Rodgers, T. C. (1947). Hypnotherapy in character neu- roses. Journal of Clinical Psychopathology, 8, 519– 524.

Rosenbaum, P. R., & Rubin, D. B. (1984). Reducing bias in observational studies using subclassification on the propensity score. Journal of the American Sta- tistical Association, 79, 516–524.

Rosow,  H.  M.  (1955).  Some  observations  on  group therapy with prison inmates. Archives of Criminal Psychodynamics, 1, 866–897.

Rubin, D. B. (1997). Estimating causal effects from large data sets using propensity scores. Annals of In- ternal  Medicine,  127, 757–763.

Salekin, R. T. (2002). Psychopathy and therapeutic pes- simism clinical lore or clinical reality? Clinical Psy- chology  Review,  22,  79–112.

Schmideberg, M. (1949). Psychology and treatment of the criminal psychopath. International Journal of Psychoanalysis,  20, 197.

Serin, R. C., & Kuriychuk, M. (1994). Social and cogni- tive processing deficits in violent offenders: Implica- tions for treatment. International Journal of Law and Psychiatry,  17, 431–441.

Seto, M. C., & Barbaree, H. (1999). Psychopathy, treat- ment behavior, and sex offender recidivism. Journal of Interpersonal Violence, 14, 1235–1248.

Showstack, N. (1956). Treatment of prisoners at the California medical facility. American Journal of Psy- chiatry,  112, 821–824.

Skeem, J. L., Monahan, J., & Mulvey, E. P. (2002). Psy- chopathy, treatment involvement, and subsequent vi- olence among civil psychiatric patients. Law and Hu- man Behavior, 26, 577–603.

Skeem, J. L., Poythress, N., Edens, J. F., Lilienfeld, S. O., & Cale, E. M. (2002). Psychopathic personality or personalities? Exploring potential variants of psy- chopathy and their implications for risk assessment. Aggression and Violent Behavior, 8, 513–546.

Skilling, T. A., Harris, G. T., Rice, M. E., & Quinsey, V.

38. (2002). Identifying persistently antisocial offend- ers using the Hare Psychopathy Checklist and DSM antisocial personality disorder criteria. Psychological Assessment, 14, 27–38.

Skilling, T. A., Quinsey, V. L., & Craig, W. A. (2001). Evidence of a taxon underlying serious antisocial behavior in boys. Criminal Justice and Behavior, 28, 450–470.

Skolnick, N. J., & Zuckerman, M. (1979). Personality change in drug abusers: A comparison of therapeutic community and prison groups. Journal of Consulting and Clinical Psychology, 47, 768–770.

Stein, D. B. (1999). Outpatient behavioral management of aggressiveness in adolescents: A response cost par- adigm.   Aggressive  Behavior,   25, 321–330.

Szurek, S. A. (1942). Notes on the genesis of psycho- pathic personality trends. Psychiatry, 5, 1–6.

Thorne, F. C. (1959). The etiology of sociopathic reac- tions. American Journal of Psychotherapy, 13, 319– 330.

Tolan, P. H., & Guerra, N. G. (1994). Prevention of de- linquency: Current status and issues. Applied and Preventive Psychology, 3, 251–273.

Wakefield, J. C. (1992). The concept of mental disorder: On the boundary between biological facts and social values.  American Psychologist,  47, 373–388.

Wasserman, G. A., & Miller, L. S. (1998). The pre- vention of serious and violent juvenile offending. In

1. Loeber & D. P. Farrington (Eds.), Serious and violent juvenile offenders: Risk factors and success- ful interventions (pp. 197–247). Thousand Oaks, CA: Sage.

Weisman, R. (1995). Reflections on the Oak Ridge ex- periment with mentally disordered offenders, 1965– 1968. International Journal of Law and Psychiatry, 18, 265–290.

Wexler, H. K., Melnick, G., Lowe, L., & Peters, J. (1999). Three-year reincarceration outcomes for am- ity in-prison therapeutic community and aftercare in California. The Prison Journal, 79, 321–336.

Wong, S., & Hare, R. D. (2005). Guidelines for a psy- chopathy treatment program. Toronto, ON, Canada:

Multi-Health  Systems.

Wong, S. E., Woolsey, J. E., Innocent, A. J.,  & Liberman, R. P. (1988). Behavioral treatment of vio- lent psychiatric patients. Psychiatric Clinics of North America, 11, 569–581.

Woody, G. E., McLellan, T. A., Luborsky,  L.,  & O’Brien, C. P. (1985). Sociopathy and psychotherapy outcome. Archives of General Psychiatry, 42, 1081– 1086.

Zinger, I., & Forth, A. E. (1998). Psychopathy and Ca- nadian criminal proceedings: The potential for hu- man rights abuses. Canadian Journal of Criminology, 40, 237–276.