Sexual Abuse

This work rests on responses from 219 male sexual assault and rape victims who self-reported their victimization in the 1994-1996 Violence and Threats of Violence Against Women and Men in the United States survey. The authors expected that men who reported being severely assaulted would be more likely than others to seek counseling. They defined severely assaulted as having been penetrated, assaulted with a weapon, threatened, self-reported sustaining physical injuries, sought medical care, and/or reported the assault to the police. However, in their logistic model that explores who sought counseling, only one variable was significant. The odds of seeking counseling for men who reported being penetrated had significantly lower odds of seeking counseling all else equal.

Sex offenders taking part in treatment programs in 10 probations areas of England were asked to undertake polygraph testing on a voluntary basis. Over a 2-year period 347 offenders attended for testing (43% of those eligible). Outcome was compared with offenders from four probation areas where polygraphy was not introduced. Case managers of polygraphed offenders reported new disclosures relevant to supervision being made in 70% of first tests, compared with 14% of case managers of nonpolygraphed offenders who reported new disclosures in the preceding months (odds ratio [OR] = 14.4, confidence interval [CI] = 8.5, 24.5). Of the disclosures made during polygraph testing, 27% were rated as being of “medium” severity and 10% “high.” Polygraph and nonpolygraph case managers reported making revisions in risk assessment with similar frequency, but nonpolygraph case managers were much more likely to consider risk to have reduced while changes in risk assessment made by polygraph case managers were usually upwards (OR = 5.0, CI = 1.7, 14.6). Case managers of polygraph offenders reported more treatment changes than case managers of the comparison group (OR = 3.1, CI = 1.6, 6.0), which were attributable to the polygraph test. Case managers rated polygraphy as “somewhat” or “very” helpful after 93% of tests for which we had information.

Phallometric assessments of single-victim sexual offenders against children have suggested that only about 50% of these men are more attracted to children than they are to adults. This has raised the question of what motivates the other 50% of men to approach young girls for sex. Freund et al. showed that gynephilic men (i.e., men preferentially attracted to adult women) evidenced greater arousal to images of prepubescent girls than to images of males of any age or to nonerotic images, arguing that gynephilic men may approach prepubescent girls as a “surrogate” for their preferred erotic targets (i.e., adult women). One might argue that these phallometric results are artifactual, given that they were obtained in a time period during which images of nudity were far less common than they are today (thus any female nudity might have elicited arousal). To address this issue, the authors examined the sexual arousal patterns of 214 contemporary men who, based on self-report, offense history, and phallometric responses, were purely gynephilic. Results showed the “classical control profile”: the greatest arousal to adult women, systematically decreasing arousal as the female stimuli became younger, and essentially no arousal to any age categories of males or to neutral (nonerotic) stimuli. Arousal to both pubescent and prepubescent girls was significantly greater than to neutral stimuli (p < .001 for both). Thus, Freund et al.’s results still appear to be valid, and the explanation for child molesting that they suggest still seems to be feasible.

This study investigates characteristics differentiating sexually motivated murderers targeting child victims (CV; n = 35) from those with only adult victims (AV; n = 100). In the initial phase, psychiatric court reports were evaluated using standardized instruments (SCID-II, PCL-R, HCR-20, SVR-20, Static-99). In the second phase, data on duration of detention and reconviction rates were obtained from German federal criminal records. The CV group showed more often diagnostic criteria of pedophilia (43% vs. 4%) and less often alcohol abuse and drug dependency (31% vs. 55%), sexual dysfunctions (9% vs. 29%) and narcissistic personality disorder (0% vs. 13%). No significant differences were found regarding PCL-R and total risk assessment scores. Child victim perpetrators were more likely to have committed acts of sexual child abuse before the sexual homicide (46% vs. 16%) but were less likely to have committed rape or sexual assault (17% vs. 42%) or caused bodily injury (26% vs. 50%). The CV group was detained more frequently in forensic psychiatric hospitals (59% vs. 26%), but the two groups showed the same rates of release and reconviction for sexual (22% for both groups), nonsexual violent (CV 25% vs. AV 15%) and nonviolent offenses (CV 63% vs. AV 59%). Although well-known differences between nonhomicidal sexual child abusers and rapists were replicated in this study on sexual homicide perpetrators, the groups showed more similarities than differences. The high prevalence of violence and antisocial personality disorder in both groups seem to be important risk factors for committing a (sexual) homicide and might have outweighed other differences.

This study investigates target selection scripts of 72 serial sex offenders who have committed a total of 361 sex crimes on stranger victims. Using latent class analysis, three target selection scripts were identified based on the victim’s activities prior to the crime, each presenting two different tracks: (1) the Home script, which includes the (a) intrusion track and the (b) invited track, (2) the Outdoor script, which includes the (a) noncoercive track and the (b) coercive track, and (3) the Social script, which includes the (a) onsite track and the (b) off-site track. The scripts identified appeared to be used by both sexual aggressors of children and sexual aggressors of adults. In addition, a high proportion of crime switching was found among the identified scripts, with half of the 72 offenders switching scripts at least once. The theoretical relevance of these target selection scripts and their practical implications for situational crime prevention strategies are discussed.

Most studies of confessions have looked at the influence of individual factors, neglecting the potential interactions between these factors and their impact on the decision to confess or not during an interrogation. Classification and regression tree analyses conducted on a sample of 624 convicted sex offenders showed that certain factors related to the offenders (e.g., personality, criminal career), victims (e.g., sex, relationship to offender), and case (e.g., time of day of the crime) were related to the decision to confess or not during the police interrogation. Several interactions were also observed between these factors. Results will be discussed in light of previous findings and interrogation strategies for sex offenders.