Pedophilia, which is classified in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) 1 as a paraphilia, is one of the few psychiatric disorders for which the symptoms constitute a criminal act. Possibly due to its morally objectionable nature, pedophilia is relatively under-researched, despite the profound social costs of pedophilic behavior. Estimates of the incidence of childhood sexual abuse have ranged from 10%–60%, depending on the population studied and the definition of sexual abuse used. 2,3 Moreover, in victims, such childhood histories often result in a broad range of axis I and axis II disorders. 4,5 Thus there is clear need for greater research into pedophilia.
The definition of child sexual abuse and pedophilia varies across different diagnostic systems, 1,6,7 specifically as to whether or not abusers must actually touch the child or if feeling sexual attraction to prepubescent children is sufficient to make the diagnosis. By DSM-IV criteria, pedophilia is characterized by recurrent sexually arousing urges, fantasies, or behavior involving sexual activity with a prepubescent child, generally age 13 or younger, which persist for a period of at least 6 months. Such fantasies or behavior cause clinically significant distress or functional impairment. 1 Of note, this definition does not differentiate between those who do and those who do not who act upon these urges, which reduces its utility. The DSM-IV text revision (DSM-IV-TR) addresses the frequently ego-syntonic nature of pedophilia, stating that recurrent pedophilic fantasies or arousal need not be experienced as distressing and that those “who have a pedophilic arousal pattern and act on these fantasies or urges with a child qualify for the diagnosis of Pedophilia” (pp. 571–2). 8
Moreover, a number of overlapping but nonetheless distinct terms are used in the discussion of pedophilia. Perpetrator and sex offender are legal terms that, in this context, refer to an individual who has been convicted of sexually abusing a child. Likewise, a child molester refers to anyone who has sexually molested a child, irrespective of legal status or characteristic sexual preference. Pedophilia, in contrast, is a clinical term suggestive of a sexual disorder and of ongoing deviant sexual desires, urges, fantasies and possibly, but not necessarily, behavior. There is, in fact, controversy over whether child molesters and/or sexual offenders against children are necessarily specifically attracted to children. Thus, the study of child molesters may not be identical to the study of pedophiles. Because it is not always possible to precisely differentiate these groups in the current literature, this article will incorporate research on child molesters, child sexual offenders, and pedophiles.
Further complicating assessment and research, pedophiles are broadly recognized to be unreliable historians. 9–11 There are compelling legal reasons for this. Admission of the scope of past pedophilic behavior and the likelihood of future behavior can lead to incarceration. Moreover, robust findings of psychopathic traits 12–14 suggest child molesters are likely to lie when they perceive it to be in their interest. Another central finding is of pervasive cognitive distortions, which further compromise the accuracy of self reports. 11,15
Additionally, most research is performed on convicted sex offenders. 9,16,17 Study of convicted sex offenders allows corroboration of self-report data, but excludes pedophiles who have never been apprehended or who have succeeded in inhibiting their impulses. It also provides a bias towards individuals of lower socioeconomic status (SES), who are disproportionately represented in the criminal justice system. 9,16 Moreover, there is considerable heterogeneity across research samples, with some studies including institutionalized sex offenders 9 (either in prisons or hospitals) and others restricted to outpatient samples, although many such subjects were referred for treatment by the court system. 16,17 Most studies specify the source of recruitment but not always subjects’ legal or behavioral status (i.e., whether they have actually committed a pedophilic act). Finally, many of the comprehensive, large-scale studies are several decades old, highlighting the pressing need for greater funding for research on pedophilia. Despite these limitations, there are a number of well-designed, large-scale studies that allow us to draw some conclusions about the clinical features of pedophilia. In this article, we will review the literature on the clinical features, subtypes, and treatment of pedophilia with the aim of providing a basic overview of current knowledge.
CLINICAL FEATURES OF PEDOPHILIA
Key points concerning the phenomenology and clinical features of pedophilia are summarized in Table 1.
To date there are no prevalence data on pedophilia, partly for some of the reasons mentioned above. In 1985, the American Humane Society estimated 123,000 reported cases of child sexual abuse in the United States (cited in Barnard et al. 2). Most researchers assume that a significant fraction, if not the vast majority, of offenses are not reported. 2,18 In a community sample of 501 adult women, 55% reported at least one sexual incident before the age of 14, with 37% by an unrelated male, 13.4% by an unrelated female, and 11.9% by a male relative. However, more intensive sexual contact was reported by a smaller percentage; 5% reported manual-genital contact, 2% intercourse, and <1% oral sex. 18,19 Based on various incidence reports, McConaghy estimated that about 5% of men and 0.5% of women molest pre-adult females. 18
While there is consensus that the majority of pedophiles and child molesters are male, the incidence of female pedophiles may be much higher than was previously reported (e.g., Gebhard et al. 9). During a 3-year period, the Protective Services Department in Texas identified 29 mother offenders, which constituted 4% of the state’s sexual offender population. 20 Of note, in all but 6 of the 29 cases, a male abuser was also involved. In a large scale study of the clinical and legal records of 4,402 pedophilic offenders (out of 6,000 total sex offenders), there were 21 female offenders (0.4%). 17 It is possible, however, that differential reporting and prosecution rates may mask the true prevalence of female pedophiles.
In a national survey by Finkelhor et al. published in 1990, 27% of females and 16% of males reported sexual experiences prior to the age of 18 which they would now consider sexual abuse. 21 Female perpetrators were reported in 1% of the offenses against girls and 17% of those against boys. In our own study of the sexual history of 20 outpatient male pedophiles, 30% reported sexual advances as a child by at least one adult female. 22 Several studies have specifically investigated female perpetrators. 20,23,24
Age of Onset
Most studies suggest an early age of onset for pedophilia. Some studies report onset in adolescence, 9,10 although incestuous pedophilia may have a later age of onset, particularly in those who abuse their own children. 25 As noted above, studies of child and adolescent sexual offenders who molest other children suggest an early onset of pedophilia in at least a subgroup of pedophiles. 23,24,26
Number of Victims and Frequency of Acts
Reports concerning number of victims and acts also vary by study. Many studies count documented legal convictions, which is presumably much lower than the actual number of acts committed. The 1967 study by Gebhard et al. 9 reported on 376 institutionalized sex offenders who were convicted of a sexual offense against a child 11 years old or younger. The mean number of sexual convictions ranged from 1.57 to 2.24 across subgroups. In contrast, under a federal guarantee of confidentiality that protects data even from a subpoena, convicted child molesters admitted to a far greater number of acts and victims. Using information gathered with such a guarantee, Abel and Osborne reported that 453 pedophilic offenders, recruited through outpatient treatment programs, admitted to an average of 236 acts and 148 victims per offender. 25 Even restricting the sample to acts of physical contact, 371 offenders admitted to an average of 210 acts and 104 victims per offender.
The frequency of acts and victims varies considerably by type of victim. The largest number of victims and acts per offender occurred with those who molested nonrelated boys (median of 10.1 acts). Incest offenders committed a median number of 4.4 acts against female victims and 5.2 acts against male victims. Offenders against unrelated girls committed a median of 1.4 acts per offender. 25
Because there are a small number of offenders with very large numbers of acts and victims, the means are much larger than the medians. Molesters of boys outside the home (i.e., unrelated) averaged 150.2 victims, those who molested girls outside the home averaged 19.8. The number of victims of incest offenders was much lower—1.7 and 1.8 for boys and girls inside the home (Barnard et al. 2 quoting Abel et al. 1987 27).
Type of Acts
With regard to the type of acts, there is consistent evidence that sexual intercourse is far less frequent than fondling and genital contact. In Gebhard et al.’s sample, only 2.3% of the 199 convictions of molesters against girls involved completed intercourse. 9 In violent sexual abuse offenses, however, the incidence of completed intercourse is much higher at 23%. In fact, much of the violence in sexually aggressive offenses involved physically coerced intercourse. Intercourse was also more frequent in incest cases, occurring in 8.7% of the convictions. Of the heterosexual, nonincest convictions, 74% involved genital contact, however, as did 90%–99% of the other child sexual abuse convictions. In Abel and Osborne’s 1992 study of 453 pedophilic offenders, 72.9% of reported acts involved touching, while 27.1% involved non-touching offenses (e.g., exhibitionism, voyeurism). 25
In general, it is likely that pragmatic factors affect the prevalence rates of various types of molestation. It is easier for adult men to gain access to boys than girls, less invasive acts are easier to commit than more invasive acts, and there are far more potential victims outside the home than inside the home. With victims inside the home, however, there is more opportunity for multiple molestations and more invasive sexual abuse (e.g. intercourse).
Although violent child sexual abuse does occur, it is much less frequent than nonviolent child sexual abuse. 18 In Gebhard et al.’s sample, only 25 (6.6%) cases involved the use of significant aggression. 9 In our analysis of intake data from CAP Behavioral Associates, an outpatient clinic for sex offenders, nonviolent methods of coercion were the most commonly listed. 28 According to intake data, 42 (33%) of 127 child molesters used manipulation, 29 (23%) used bribery, and 22 (17%) used no coercive method. Only 2 (1.6%) used assault (defined as use of physical force beyond what was necessary to commit the sexual offense) and 3 (2.4%) used threat, although 29 (23%) used force (physical force used solely to commit the sexual offense). When only the 82 subjects who admitted their offense were analyzed, those listed as using no coercive method dropped to 6 (7.3%). Bribery (27%) and manipulation (42%) were the most common methods, followed by force (22%). Threat and assault were both minimal (1%). 28
However, the subgroup of pedophiles who are violent may differ from nonviolent pedophiles on a number of domains. In Gebhard et al.’s study, the sexually aggressive child molesters had more impulsive offenses, were more likely to molest strangers, and had a higher incidence of completed intercourse. 9
The role of impulsivity in pedophilia is a matter of some controversy. Recent literature on paraphilias and hypersexual disorders has suggested that pedophilia may fall into the obsessive-compulsive spectrum 29,30 or may reflect impulsive-aggressive pathology. 31 It is therefore important to consider the degree of impulsivity involved in pedophilic offenses. Such information has implications for both psychological and pharmacological treatments. 32 In Gebhard et al.’s study, 70%–85% of the child molestation convictions were classified as premeditated. 9 In our study of impulsive-aggressive personality traits in 20 convicted male child molesters versus 24 demographically similar healthy controls, findings were mixed, suggesting that impulsive-aggressive personality traits were present but not predominant in the pedophile sample relative to the controls. 16 Thus, impulsivity per se may not be an intrinsic characteristic of pedophilia. On the other hand, impulse control disorders were diagnosed in 55% of 22 adolescent 26 and 29% of 45 adult pedophilic offenders. 33
Lack of insight, however, does appear to characterize pedophiles, since they routinely deny and minimize the deviant nature of their behavior and its destructive impact on the children involved. 11,15,34 In Gebhard et al.’s study, 19% of all child molesting subjects completely denied their offense to the interviewers. 9 In addition, 28% of the 199 heterosexual child molesters reported a greater level of victim encouragement than was indicated in the legal record. In our analysis of outpatient intake data, of 170 convicted child molesters, 40% denied their offense and 22% minimized it, while only 38% admitted it. 28 Pedophiles’ lack of insight extends beyond the simple denial or minimization of illegal behavior, however. In fact, at least two instruments have been developed to measure pedophiles’ widespread cognitive distortions about pedophilic acts. 11,35
Sex of Victim
The ratio of male to female victims varies across studies and may vary according to situational factors. Nonetheless, it is clear that both male and females victims are well represented. Of 4,381 pedophiles treated in a community-based program, 2,940 (67.1%) had offended against minor females and 1,441 (32.9%) had offended against minor males. 17 In Gebhard et al.’s sample 9 of 295 offenders against children outside the home, 199 of them committed offenses against girls and 96 committed offenses against boys, a 2 to 1 ratio similar to that reported in the Maletzky sample. 17 Of note, neither of these studies addressed the question of group overlap. Interestingly, a similar ratio was reported in a sample of 561 paraphiliac males, with 67.2% who targeted only females, 11.9% who targeted only males, and 20% who targeted both sexes. 10 Although there may be more heterosexual than homosexual pedophiles, there may be more male than female victims. Of the total sample of 453 pedophilic offenders reported by Abel and Osborne in 1992, 54.5% of all victims were girls. 25 On the other hand, of the 77,919 touching acts carried out by 371 offenders, only 38% of the victims were girls.
Age of Victim
There is no lower limit for the age of victims. In fact, the term infantophilia has been proposed for pedophiles who abuse children younger than 5 years of age. 36 Children in middle to late childhood, however, are much more frequently abused than very young children. In Gebhard et al.’s study of offenders against children 11 years of age and under, the median age of the victim was 8 years old, with a range from 3 to 11 years old. 9
Relationship to Victim
Pedophiles molest strangers, acquaintances, friends, and relatives. The nature of the relationship to the victim can affect the frequency and severity of the abuse. In Gebhard et al.’s study, 60%–67% of the 295 male and female nonincestuous victims were known to the offender, whereas 32%–39% were strangers. 9 Among sexually aggressive offenders, however, 70% of the victims were strangers. 9 The proportion of offenders who offend against relatives was 26% in one large study 10 and 19% in another. 9 Likewise, 20% of the pedophilic acts reported under a certificate of confidentiality involved a relative. 25
The question of heterogeneity among pedophiles is often raised. DSM-IV lists several subtypes of pedophilia, including exclusive versus nonexclusive; incestuous versus nonincestuous; and heterosexual, homosexual, or bisexual. 1 As discussed above, there is evidence that pedophiles do vary along a number of these dimensions and that such distinctions may have clinical implications. Nonetheless, there is also evidence of considerable overlap. 25
Exclusive Versus Nonexclusive
DSM-IV differentiates between exclusive versus nonexclusive pedophilia. It appears that the majority of pedophiles are nonexclusive and have frequent sexual relations with adults. In many studies, a large percentage of subjects were married, separated, or divorced. 9,12 In Gebhard et al.’s study, 9 a larger percentage of the homosexual pedophiles were single (53% versus 0%–33% across other subgroups), but this does not address their involvement with adult men. Phallometric measurement of penile tumescence, however, has shown a preferential response to pedophilic versus adult erotic stimuli in most pedophiles 17,37,38 and many child molesters. 39
The distinction between incestuous and nonincestuous offenders is also of interest. Although, as noted above, incestuous pedophilic offenders and incestuous pedophilic acts may both comprise about 20% of the respective totals, 9,10,25 in Abel and Osborne’s study, only 1% of the victims were related to the molester. 25 In addition, offenses of father-daughter incest (including daughters and step-daughters) were almost four times more likely to involve completed intercourse than offenses against nonrelated girls (8.7% versus 2.3%). 9 Thus, while incest victims are far fewer in frequency, they are exposed to multiple offenses and more severe molestation over time. It is therefore not surprising that incest victims are highly represented in clinical settings, suggesting the particularly damaging effects of incestuous sexual abuse. 25 Nonetheless, the distinction between incest abusers and nonincestuous abusers is not entirely clear cut. In a study of 446 nonincestuous sex offenders, 29% had also offended against family members. Furthermore, of 199 incestuous sex offenders, 65.8% had also offended against nonfamily members. 25
Heterosexual, Homosexual, Bisexual
As discussed above, several studies suggest a ratio of approximately 2 to 1 between heterosexual and homosexual pedophiles. 9,17 The number of victims, however, may be much higher for homosexual than heterosexual male pedophiles. Moreover, as with incest, there appears to be considerable overlap. In the 1992 study by Abel and Osborne, 25 22.9% of 489 offenders who had molested females had also molested males. Moreover, 62.6% of those who had offended against males had also offended against females.
True Pedophiles Versus Nonpedophilic Child Molesters
This is possibly the most important distinction regarding assessment and treatment. Are all people who sexually molest children actually pedophiles? Can perpetrators of child sexual abuse be differentiated according to the centrality of pedophilic arousal? Other related classifications have included true versus opportunistic or fixated versus regressed pedophiles. 2,18 Plethysmographic measures of arousal have been used to identify true pedophiles as those who show a preferential arousal response to pedophilic stimuli. 17,38,39
Some investigators posit that variation among pedophiles may reflect a continuum of severity as opposed to two distinct classifications. 17 If so, true pedophiles would be those who exceed a critical threshold of severity. Given the difficulty in obtaining accurate histories, it is not known to what extent evidence of a small number of past sexual encounters with a prepubescent child reflects pervasive and chronic pedophilic proclivities and, likewise, what percentage of convicted child molesters truly meet criteria for pedophilia. On one hand, in the studies by Abel and colleagues, a small subgroup accounted for a large percentage of the reported acts and victims. 25,27 Moreover, in one plethysmographic study of 216 child molesters (189 nonhomicidal and 27 homicidal) and 47 non-offenders, about 50% of convicted child molesters showed a greater arousal to pedophilic auditory stimuli than similar stimuli involving adult women, compared to about 28% of non-offenders. 39 There remains considerable controversy, however, about appropriate methods of plethysmographic administration, 40,41 so such figures must be interpreted with caution. Nonetheless, 34%–56% of convicted child molesters were repeat offenders 9 and 100% of 22 adolescent males who had sexually abused other children met DSM-III-R criteria for pedophilia. 26 Further, the notion of a subgroup of particularly severe pedophiles, those with hundreds of victims, has received little attention but may also merit consideration. Thus we might conclude that distinctions between “true” and “opportunistic” pedophiles are meaningful but far from conclusive and that they need to be clarified.
More recent studies have used structured instruments to characterize psychiatric comorbidity in pedophiles. Findings have been strikingly consistent across studies and suggest that pedophiles may share many psychiatric features above and beyond deviant sexual desire.
Comorbid Axis I Disorders
Affective, substance use, and impulse control disorders. Comorbid axis I disorders include high rates of affective and substance abuse disorders. In Galli et al.’s 1999 study of 22 adolescent offenders, 82% met DSM-III-R criteria for a mood disorder, 55% for an anxiety disorder, and 50% for substance abuse or dependence. 26 In this adolescent sample, at least 55% had an impulse control disorder, including intermittent explosive disorder (45%), kleptomania (18%), and pyromania (23%). In a study of 36 adult sexual offenders by a related research group, substance use disorders were diagnosed in 83% of subjects, mood disorders in 61%, anxiety disorders in 31%, and eating disorders in 17%. 42 Likewise, Raymond et al.’s 1999 study of 45 male pedophilic sex offenders demonstrated similarly high rates of lifetime comorbid substance use disorders (60%), mood disorders (66.7%), and anxiety disorders (64%), as well as a 33.3% lifetime incidence of posttraumatic stress disorder. 33 In this same study, the lifetime incidence of impulse control disorders was 29%, with pathological gambling (11.1%) the most common impulse control disorder diagnosis. The incidence of impulse control disorders appears to contradict the premeditated nature of most pedophilic offenses. We can posit several possible explanations, all of which await further research: 1) a subgroup of pedophiles may be impulsive; 2) impulse control disorders may be nonspecific markers of severe psychopathology, or 3) pedophiles may demonstrate compulsive behavior that is misdiagnosed as impulsive behavior. 16
Comorbid paraphilias. There is consistent evidence of multiple comorbid paraphilias. In Abel et al.’s study of 561 paraphiliacs, 359 of whom had molested children 14 years of age and younger, at least 70% reported multiple paraphilias, including exhibitionism, voyeurism, and frottage. 10 In Galli et al.’s 1999 study of 22 adolescent pedophilic offenders, 86% met DSM-III-R criteria for frotteurism, 50% met criteria for voyeurism, and 41% for exhibitionism. 26 In Raymond et al.’s sample of adult pedophilic offenders, 53% were diagnosed with additional paraphilias. 33 Likewise, comorbidity among multiple paraphilias was also reported in 36 subjects with compulsive sexual behavior (29% with pedophilia). 43 Of note, in McElroy et al.’s 1999 sample, the 58% diagnosed with a comorbid paraphilia were significantly younger at their first offense, offended for longer before being apprehended, had significantly more victims, and were more likely to report a childhood history of sexual abuse. 42
Axis II Pathology
There is pervasive evidence of severe axis II pathology that cuts across all three DSM-IV axis II clusters in individuals with pedophilia. 12,33,43 Although some authors have argued against a specific personality profile for such individuals, 44 there have been consistent findings on a number of personality dimensions.
Psychopathy. Sociopathic traits are consistently found in studies of pedophiles 13,14 and other sexual offenders. 31 On the Minnesota Multiphasic Personality Inventory (MMPI), 45 elevated PD (Psychopathic Deviancy) scores were found in a study of 113 pedophiles. 46 Antisocial personality disorder (ASPD) was the most common Cluster B disorder (23%) and the second most common axis II disorder in a study of 40 pedophiles, with obsessive compulsive personality disorder (25%) the most common axis II diagnosis. 33 In a study of 36 patients with compulsive sexual behavior (29% of whom admitted to pedophilia), 43 14%–35% met criteria for ASPD. Finally, Gebhard et al.’s sample of 376 convicted child molesters averaged 1.5 nonsexual criminal convictions. 9 For a more detailed discussion of psychopathy, the reader is referred to the work of Hare and colleagues 47,48 and of Widiger and Lynam. 49
Impaired assertiveness. A number of older studies have reported shyness, introversion, and lack of assertiveness in pedophiles. 50,51 Howells demonstrated that pedophiles are particularly attuned to children’s lack of dominance, suggesting that pedophiles’ own difficulties with assertiveness may attract them to children’s relative powerlessness. 52 More recent studies have also provided support for anxious, inhibited personality traits in pedophiles. 53 In a study of 40 male pedophiles, 43% met criteria for a Cluster C disorder (i.e., obsessive-compulsive, dependent, or avoidant personality disorder), based on the Structured Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II). 33,54 In a study of 36 patients with compulsive sexual behavior, 19%, 17%, and 28% of the patients met the Structured Interview for DSM-IV Personality (SIDP/SIDP-R) 55 criteria and 15%, 21%, and 15% of the patients met the Personality Diagnostic Questionnaire (PDQ/PDQ-R) 56 criteria for avoidant, obsessive-compulsive and passive-aggressive disorders, respectively. 43,57
Cognitive distortions and denial of the implications of pedophilic behavior have been widely discussed. 34,53 Pedophiles routinely deny and minimize the deviant nature of their behavior and its destructive impact on the children involved, 11,15,35 and such distortions can reach delusional intensity. In fact, consistent findings of elevated Schizophrenia (Sc) scores on the MMPI have led some investigators to postulate that pedophiles are schizotypal. 46 In the study by Raymond et al., 18% of 40 pedophiles met criteria for paranoid personality disorder;33 paranoid (25%–32%) and schizotypal (3%–24%) personality disorders were elevated in 36 patients with compulsive sexual behavior. 43,57 In our own investigation, 20 male pedophiles scored significantly higher than 24 male controls on six personality measures associated with a generalized propensity to distort reality and 30% of pedophiles versus 8% of controls exceeded the Millon Clinical Multiaxial Inventory-II (MCMI-II) 58 cut-off point for a cluster A disorder. 59
While the focus of this article is primarily clinical, several areas of research may have etiological significance and merit some discussion, although an exhaustive review is beyond the scope of this paper. Specifically, there is literature on pedophiles’ own histories of childhood sexual abuse, on their sexual arousal patterns, and on their neuroendocrine and neuropsychiatric function.
Childhood History of Sexual Abuse
Some investigators have posited an “abused abuser theory,”60 in which pedophilia reflects pathological sequelae of pedophiles’ own childhood sexual abuse. Numerous reports do suggest an elevated rate of child sexual abuse in the history of pedophile offenders. 61–63 Moreover, the rate of childhood sexual trauma among pedophiles may be higher than that among other groups of sexual offenders. 60,63
In a study comparing pedophile offenders, nonpedophilic sexual offenders, and healthy controls, Freund and Kuban (1994) found that 43.9% of 83 pedophiles compared to 13.4% of 134 controls and 17.6% of 34 gynephilic sex offenders (sexual offenders against women) reported histories of childhood sexual abuse. 60 Reports of child sexual abuse were higher among heterosexual (28.6% of 77) and homosexual (25.9% of 54) pedophilic offenders than among nonpedophilic child molesters (19.6% of 51), gynephilic offenders (10.7% of 36), and homosexual (11.8% of 51) or heterosexual (10.7% of 75) controls. 61 Further, in a study of 151 male perpetrators, 57% of the child molesters compared to 23% of the rapists reported childhood sexual abuse. 63 On the other hand, in a smaller sample, incarcerated rapists reported higher rates of child sexual abuse (62%, 18/29) than either pedophiles (50%, 8/16) or nonsexual offenders (20%, 4/20). 64 Nonetheless, in a large community sample (n = 750), men who reported multiple sexual contacts in their own childhood were almost 40 times more likely to report adult sexual contact with children aged thirteen and younger (7.7% versus 0.2%). 62
Such data provide support for a central role of childhood sexual abuse in the etiology of pedophilia. In all studies, however, a significant fraction of child molesters deny any childhood sexual abuse. While cognitive distortions and other reporting biases may lead subjects to underreport such histories, it is unlikely that 100% of child molesters were in actuality sexually abused as children. Future research could differentiate the characteristics of pedophiles with and without reported histories of childhood sexual abuse.
On the other hand, it is also likely that convicted sex offenders may overstate histories in order to gain sympathy in the judicial system. Recent polygraphic studies have suggested this is the case. 65 Further research is needed, however, before any conclusions are drawn as to the degree to which self-reported rates of childhood sexual abuse are overstated, since polygraphic results in sex offenders have been shown to be unstable over time 66 and influenced by a number of external variables. 67 Further, the relative consistency of reported rates of childhood sexual abuse across varied study samples supports the relevance of such abuse histories to pedophilic research.
Deviant Sexual Arousal
It appears that pedophiles may demonstrate deviant sexual arousal beyond the abnormal choice of object. Most studies of sexual arousal patterns use plethysmographic or phallometric methods, in which one or more expandable metal rings are placed on the penis. Percent tumescence in response to sexual stimuli in various formats, such as audiotapes, videotapes, slides, or fantasies of erotic encounters, is then recorded. 37,40 This objective measure of sexual arousal is considered far more accurate than pedophiles’ self reports, 34,59 although many subjects are to some degree capable of suppressing erectile responses 37,40 and there remains controversy over administration procedures. 37,40,41
A significant number of pedophiles show preferential responses to pedophilic cues. 17,38–40,59 Several studies have also demonstrated general sexual hyperarousal in pedophiles, evidenced by elevated arousal across stimuli and at baseline. 41,59 It appears that pedophiles may also demonstrate reduced erotic discrimination, so that both heterosexual and homosexual pedophiles demonstrate less response differentiation across varied gender or age groups than either heterosexual or homosexual controls. 68,69 Thus pedophiles demonstrate deviant sexual arousal patterns characterized by general hyperarousal, preferential response to prepubescent children, and decreased erotic discrimination.
Studies of treatment outcome suggest that those who have more specific object choices have better treatment response. 17,70 In fact, in both studies, frequency of pedophilic behaviors was less important in predicting treatment response than the multiplicity of types of pedophilic acts. This may speak to the clinical importance of reduced erotic differentiation.
It is of interest to investigate whether pedophiles’ deviant sexual arousal can be linked to abnormal neuropsychiatric function. The neuroanatomy of human sexual response has not been extensively studied but is known to involve a wide range of cortical, subcortical, and autonomic systems. A number of studies have looked at both neuroendocrine and cortico-limbic function in pedophiles and healthy controls.
Elevation of testosterone levels is associated with sexual arousal and activity in healthy males and females. 71,72 Anti-androgens also serve to decrease libido. Nonetheless, most studies have failed to find differences in baseline levels of testosterone and other neuroendocrines in pedophiles and other sex offenders/paraphiliacs. 2,30,73 although two studies found elevated leutinizing hormone (LH) response to either gonadal releasing hormone 74 or LHRH. 75
Garavan et al. suggested that a nonspecific reward circuitry may mediate sexual response. 76 The authors compared the fMRI responses of cocaine abusers and healthy controls while viewing films with sexual, cocaine, and neutral stimuli. Extensive frontal regions (medial, superior, and inferior frontal gyri), anterior and posterior cingulate, bilateral insula, caudate, thalamic, occipital and cerebellar regions were activated during sexual stimuli in both groups and during cocaine stimuli in the group of cocaine abusers only. A nonspecific reward circuitry was hypothesized, with the frontal regions mediating cognitive components and the anterior cingulate and medial frontal mediating emotional and attentional aspects. It is of interest that, during sexual stimuli, the cocaine abusers had somewhat lowered brain activation compared to controls. The authors suggest that cocaine may “co-opt” the brain’s reward circuitry, making it less responsive to alternative reward stimuli. We can speculate as to whether pedophilic sexual arousal may similarly “co-opt” the brain’s reward system in pedophiles.
As frontal impairment has a robust relationship with behavioral disinhibition, frontal abnormalities in pedophiles are also of interest. Specific frontal involvement is also suggested by Wright et al., 77 who reported smaller left frontal volumes on CAT scan (CT) in a group of pedophiles compared with other sex offenders and controls who were not sex offenders. Flor-Henry et al. found electroencephalogram abnormalities in pedophiles compared with controls and incest offenders, which were particularly evident during a word fluency test, pointing to left frontal dysfunction. 78 Similar findings have also been noted for other paraphilias. 79
The temporal lobes also have been implicated in the mediation of sexual arousal patterns, including erotic discrimination and arousal threshold. 80 Kluver-Bucy syndrome, characterized by hypersexual behavior among other behavioral disturbances, is linked to bilateral lesions in the temporal lobes. 81,82 There are also reports of hyposexuality in patients with temporal lobe epilepsy. 83 Hucker et al. demonstrated left and bilateral temporal abnormalities on CT in pedophiles compared with nonviolent, nonsexual offenders. 38 Wright et al. also reported smaller left temporal volumes on CT in a group of pedophiles compared with other sex offenders and controls who were not sex offenders. 77
In response to such findings, as well as our own preliminary imaging findings, 59 our group has posited a model of pedophilia in which childhood sexual abuse leads to aberrant cortical development. Specifically, abnormalities develop in temporal regions mediating sexual arousal threshold and erotic discrimination, and in frontal regions mediating the ideational aspects of sexual desire 76,84 and behavioral inhibition. 80,85 In effect, pedophiles become imprinted by premature sexual stimulation to develop deviant sexual desire. If significant personality pathology develops, specifically sociopathy and cognitive distortions, there will be a failure to inhibit pedophilic urges (see Figure 1). 59
Of course, any such model must be seen as highly speculative. Even if robust supporting evidence is demonstrated, a significant fraction of child sexual molesters deny any childhood history of sexual abuse. It is possible that pedophiles without a history of sexual abuse may be less specifically attracted to children (i.e., may be opportunistic pedophiles). In these cases, child molestation may be part of a larger pattern of sociopathic and exploitative behavior, etiologically unrelated to childhood sexual experience.
To our knowledge, there is only one other integrated model of pedophilia. Araji and Finkelhor posited a multifactorial model with four main components:86 1) the blockage theory, which holds that pedophiles are attracted to children because they have difficulty or anxiety relating to adult females; 2) the disinhibition theory, in which pedophiles are disinhibited either due to cognitive impairment or a generalized impulse control disorder; 3) the deviant emotional congruence theory, which suggests pedophiles experience sexual relations with children as emotionally satisfying or appropriate due to their own childhood experiences of sexual abuse; and 4) the deviant sexual arousal theory, in which pedophiles’ sexual arousal is abnormal in its choice of object and possibly above and beyond object choice. Of these four components, the deviant sexual arousal and emotional congruence (also known as the “abused abuser” theory) are the best researched and have the most empirical support. 59,86
Ultimately the aim of all clinical research must be to improve treatment and prevention. Although recidivism remains a serious problem, sophisticated treatment programs for pedophiles and other sex offenders have been developed. Treatment for pedophiles is generally intensive and comprehensive, involving pharmacological and psychological interventions over months to years.
Accurate assessment is necessary for effective treatment. Because of the distinct problems with self-report, assessment poses particular difficulties in the treatment of pedophilia. A comprehensive assessment would therefore need to employ multiple methods, including detailed clinical interviews and structured assessment instruments, such as the Abel Assessment for Sexual Interest 87 and the Bumby RAPE and MOLEST scales of cognitive distortions. 35 Objective data are also critical to counteract self-disclosure biases. These include criminal records as well as phallometric measures of pedophilic sexual arousal. The use of polygraphy has gained popularity in recent years and has been shown to increase the admitted number of past pedophilic acts. 65,66 For further discussion of assessment methods, see Krueger and Kaplan. 30
Pharmacotherapy for pedophilia and other paraphilias has focused largely on the global reduction of sexual drive. Anti-androgens have been in use for several decades. While effective in reducing sexual drive, they have considerable side effects. These, along with the deep loss of libido, lead to problems with noncompliance. 30,88 A newer anti-androgen treatment involves the use of gonadotropic releasing hormone (GNRH) analogues, which are longer acting, available in depot format, and associated with fewer side effects. 30,89 SSRIs also offer better-tolerated alternatives, and several open-label trials have suggested efficacy with paraphilias. 90 It is unclear, however, to what extent this efficacy is due to the antidepressant and anti-anxiety effects, sexual side effects, or anticompulsive effects of the SSRIs. In addition, controlled trials of the SSRIs are lacking. 30 For a more detailed discussion of the pharmacological treatment of pedophilia, the reader is referred to reviews by Gijs and Gooren 91 and Rösler and Witztum. 92
Cognitive-behavioral treatments have been employed to reduce pedophilic sexual desire, to increase age-appropriate sexual and affiliative behavior, and to strengthen inhibition of pedophilic behavior. 93,94 Associative conditioning techniques such as covert sensitization and aversive conditioning, as well as plethysmographic biofeedback, and attempts to extinguish desire through masturbatory satiation are used to reduce the reward value of pedophilic arousal. 17,30,70 Training in interpersonal skills, assertiveness, and empathy are used to enhance the reward value of relationships with adults. Finally, confrontation of denial, particularly in group format, cognitive restructuring of cognitive distortions, and training in empathy for victims are all used to strengthen inhibition of pedophilic behavior. In cases where internal motivation is lacking, external restraints, such as surveillance networks or threat of incarceration, are used to bolster inhibition. 17,30,70
Recidivism, drop-out, and treatment noncompliance are significant problems with pedophiles. Furthermore, the efficacy of treatment is difficult to measure, while the costs of treatment failure are extremely steep. Completion of treatment and the absence of further arrests for sexual offenses offer easily measurable but crude indices of treatment efficacy. Plethysmographic measure of pedophilic arousal can also be used to evaluate treatment response. In Abel et al.’s report of voluntary, 30-week outpatient treatment of 192 sex offenders against children and/or adolescents, there was a 34.9% drop out rate. 70 Of the 98 offenders evaluated 1 year post-treatment, 12 (12%) had recidivated as measured by structured clinical interview. Thus, considering the drop-out rate and the recidivism rate together, short-term, voluntary treatment may have a high failure rate. Maletzky conducted an extensive review of the clinical and legal records of 4,381 pedophiles treated over a 20-year period in an outpatient program. 17 Classification of treatment failure involved either treatment drop out, re-charge with a sex offense, self-report of deviant sexual behavior after treatment completion, or deviant plethysmographic arousal greater than 20%. According to this definition, only 6% of 2,940 heterosexual pedophiles and 15% of 1,441 homosexual pedophiles were treatment failures. Thus the treatment success rate in this study, in which many patients were court mandated and treatment lasted an average of 23 months, was quite high. Nonetheless, relapse rates continued to climb even 10 years after treatment, supporting the notion that pedophilia is a lifetime disorder requiring lifetime treatment.
Predictors of Treatment Outcome
In Maletzky’s retrospective study of 4,381 pedophiles over 20 years, situational characteristics were the most powerful predictor of treatment outcome. 17 Those who were well known to the victim, had fewer victims, and who abused victims inside the home were 5–7 times more likely to be classified as treatment successes than those who abused multiple, unknown victims outside the home. Of perpetrator characteristics, the level of denial, lack of stable vocational history and stable relationships, and degree of pedophilic arousal on plethysmography all predicted a poor treatment response. 17
In Abel et al.’s 30 week program, the most potent predictor of treatment failure was the multiplicity of offense types. 70 Those who offended against males and females, against prepubescent and adolescent victims, and against incestuous and nonincestuous victims, accounted for 91.4% of the treatment drop-outs. Of perpetrator characteristics, being pressured to enter treatment and having ASPD also predicted treatment drop-out. Interestingly, in both studies the frequency of pedophilic acts did not predict treatment outcome as much as the type of pedophilic acts. 17,70 These findings may be confounded, however, since pedophiles who are less motivated to change may be more likely to underreport the frequency of their pedophilic behavior.
Hanson and colleagues have conducted a number of studies looking at the rates of and risk factors for recidivism among sex offenders. 95,96 In a study of 197 child molesters released from prison over 16 years, over 42% were reconvicted of sexual crimes, violent crimes or both. 95 Consistent with data from other samples, 9,17,25 offenders against unrelated boys were the quickest to recidivate, followed by offenders against unrelated girls and then incest offenders. Ten percent of the sample was reconvicted at least one decade after release. This study may be skewed toward particularly severe offenders, however, since the sample includes only incarcerated child molesters and recidivism is measured only by reconviction rates. In a meta-analysis of 61 treatment follow-up studies of sexual offenders, the recidivism rate was only 13%, with deviant sexual preferences and prior sexual offenses potent risk factors. 96 Although measurement of recidivism is a complex issue, these two studies do suggest the protective effect of treatment. Nonetheless, both studies also support the serious and long-term risk of recidivism in pedophilic offenders.
We have attempted to clarify current knowledge about the phenomenology, clinical features, and treatment of pedophilia and to point out areas requiring further research. The literature clearly shows that pedophilia is a serious, chronic, and probably life-long psychiatric disorder, with significant comorbidity with other paraphilic disorders as well as other axis I and axis II disorders. Pedophilia is extremely difficult to treat and effective treatment needs to be intensive, long-term, and comprehensive, possibly with lifetime follow-up.
Several areas of controversy remain, the most significant of which involves the nature of subgroups, since there appear to be distinct differences across subgroups regarding numbers and types of victims and acts as well as significant diagnostic overlap across groups that runs as high as 70% in some samples. There are also questions concerning the prevalence and definition of true pedophiles versus nonpedophilic child molesters. Further research is also needed concerning the etiological role and sequelae of pedophiles’ own childhood history of sexual abuse as well as the underlying neurobiology of pedophiles’ deviant sexual arousal and decreased erotic differentiation.
1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edition. Washington. DC: American Psychiatric Association; 1994.
2. Barnard GW, Fuller AK, Robbins L, et al. The child molester: An integrated approach to evaluation and treatment. New York: Bruner/Mazel Clinical Psychiatry Series; 1989.
3. Russell DEH. The prevalence and seriousness of incestuous abuse: Stepfathers vs. biological fathers. Child Abuse and Neglect 1984; 8:15–22.
4. Herman J. Trauma and recovery. New York: Basic Books, 1992.
5. Beck JC, Van der Kolk B. Reports of childhood incest and current behavior of chronically hospitalized psychotic women. Am J Psychiatry 1987; 144:1474–6.
6. Council on Scientific Affairs of the American Medical Association. AMA diagnostic and treatment guidelines concerning child abuse and neglect. JAMA 1985; 254:796.
7. World Health Organization. The ICD-10 classification of mental and behavioral disorders: Clinical descriptions and diagnostic guidelines. Geneva: World Health Organization; 1992.
8. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th edition, text revision. Washington. DC: American Psychiatric Association; 2000.
9. Gebhard PH, Gagnon JH, Pomeroy WB, et al. Sex offenders. New York: Bantam Books, 1967.
10. Abel GG, Becker JV, Cunningham-Rathner BA, et al. Multiple paraphilic diagnoses among sex offenders. Bull Am Acad Psychiatry Law 1988; 16:153–68
11. Abel GG, Gore DK, Holland CL, et al: The measurement of the cognitive distortions of child molesters. Annals of Sex Research. 1989; 2:135–52
12. Cohen LJ, McGeoch P, Watras-Gans S, et al. Personality profile of male pedophiles. J Clin Psychiatry 2002 (in press).
13. Virkkunen M. The pedophilic offender with antisocial character. Acta Psychiatr Scand 1985; 53:401–5.
14. Dorr D: Psychopathy in pedophiles: In: Millon T, Simonsen E, Birket-Smith M, et al., eds. Psychopathy: Antisocial, criminal and violent behavior. New York: Guilford; 1998:304–20.
15. Blumenthal S, Gudjonsson G, Burns J. Cognitive distortions and blame attribution in sex offenders against adults and children. Child Abuse Negl 1999; 23:129–43.
16. Cohen LJ, Watras-Gans S, McGeoch P, et al. Impulsive-aggressive personality traits in male pedophiles: Is pedophilia an impulsive-aggressive disorder? Compr Psychiatry. 2002; 43: 127–34.
17. Maletzky BM. Factors associated with success and failure in the behavioral and cognitive treatment of sexual offenders. Annals of Sex Research 1993; 6:241–58.
18. McConaghy N. Paedophilia: A review of the evidence. Aust N Z J Psychiatry 1998; 32:252–65.
19. Kilpatrick AC. Long-range effects of child and adolescent sexual experiences. Hillsdale, NJ: Lawrence Erlbaum Associates; 1992.
20. McCarty LM. Mother-child incest: Characteristics of the offender. Child Welfare 1986; 65:447–58.
21. Finkelhor D, Hotaling G, Lewis IA, et al. Sexual abuse in a national survey of adult men and women: Prevalence, characteristics, and risk factors. Child Abuse Negl 1990; 14:19–28.
22. Cohen LJ, McGeoch P, Watras Gans S, et al. Childhood sexual history of twenty male pedophiles vs. twenty-three male healthy controls. J Nerv Ment Dis (in press).
23. Faller KC. Women who sexually abuse children. Violence and Victims 1987; 2:263–76.
24. Johnson TC. Female child perpetrators: Children who molest other children. Child Abuse Negl 1989; 13:571–85.
25. Abel GG, Osborne C. The paraphilias: The extent and nature of sexually deviant and criminal behavior. Clinical Forensic Psychiatry 1992; 15:675–87.
26. Galli V, McElroy SL, Soutullo CA, et al. The psychiatric diagnoses of twenty-two adolescents who have sexually molested other children. Compr Psychiatry 1999; 40:85–8.
27. Abel GG, Becker JV, Mittelman M, et al. Self-reported sex crimes of non-incarcerated paraphiliacs. Journal of Interpersonal Violence 1987; 2:3–25.
28. Weaver CJ, Cohen LJ, Boettger S, et al. Admission and coercion among sexually aggressive paraphilics. Poster presented at the 155th Annual Meeting of the American Psychiatric Association, Philadelphia, PA, May 18–23, 2002.
29. Stein DJ, Black DW, Pienaar W. Sexual disorders not otherwise specified: Compulsive, addictive or impulsive? CNS Spectrums: the International Journal of Neuropsychiatric Medicine 2000;5:1, 60–4.
30. Krueger RB, Kaplan MS. Treatment of paraphilic and hypersexual disorders. Journal of Psychiatric Practice 2002; 8: 21–32.
31. Kafka M. Sexual impulsivity. In: Hollander E, Stein D, eds. Impulsivity and aggression. New York: John Wiley & Sons; 1995:201–28.
32. Hollander E, Cohen LJ. The psychobiology and psychopharmacology of compulsive spectrum disorders. In: Oldham J, Hollander E, Skodol A, eds. Impulsivity and compulsivity. Washington DC: American Psychiatric Press; 1996:143–66.
33. Raymond NC, Coleman E, Ohlerking F, et al. Psychiatric comorbidity in pedophilic sex offenders. Am J Psychiatry 1999; 156:786–8.
34. Haywood TW, Grossman LS. Denial of deviant sexual arousal and psychopathology in child molesters. Behav Ther 1994; 25:327–40.
35. Bumby K. Assessing the cognitive distortions of child molesters and rapists: Development and validation of the MOLEST and RAPE scales. Sex Abuse 1996; 8:37–54.
36. Greenberg DM, Bradford J, Curry S. Infantophilia: A new subcategory of pedophilia? A preliminary study. Bull Am Acad Psychiatry Law 1995; 23;63–71.
37. Abel GG, Clanchard EB, Barlow D. Measurement of sexual arousal in several paraphilias: The effects of stimulus modality, instructional set and stimulus content on the objective. Behav Res Ther 1981; 19:25–33.
38. Hucker S, Langevin R, Wortzman G, et al. Neuropsychological impairment in pedophiles. Canadian Journal of Behavioral Science 1986; 18:440–8.
39. Firestone P, Bradford JM, Greenberg, DM, et al. Differentiation of homicidal child molesters, nonhomicidal child molesters, and nonoffenders by phallometry. Am J Psychiatry 2000; 157:1847–50.
40. Hall GC. Sexual arousal as a function of physiological and cognitive variables in a sexual offender population. Arch Sex Behav 1991; 20:359–69.
41. Hall GC, Proctor WC, Nelson GM. Validity of physiological measures of pedophilic sexual arousal in a sexual offender population. J Consult Clin Psychol 1988; 56: 118–22.
42. McElroy SL, Soutullo CA, Taylor P, et al. Psychiatric features of 36 men convicted of sexual offenses. J Clin Psychiatry 1999; 60:414–20.
43. Black DW, Kehrberg LLD, Flumerfelt DL, et al. Characteristics of 36 subjects reporting compulsive sexual behavior. Am J Psychiatry 1997; 154:243–9.
44. Okami P, Goldberg A. Personality correlates of pedophilia: Are they reliable indicators? J Sex Res 1992; 29:297–328.
45. Dahlstrom WG, Welsh GS, Dahlstrom LE. MMPI handbook, vol. II: Research applications. Minneapolis, MN: University of Minnesota Press; 1975.
46. Henderson MC, Kalichman SC. Sexually deviant behavior and schizotypy: A theoretical perspective with supportive data. Psychiatr Q 1990: 61;273–84.
47. Hare RD. Without conscience: The disturbing world of the psychopaths among us. New York: Simon & Schuster; 1993. Reissued 1999 by Guilford Press, New York.
48. Cooke DJ, Forth AE, Hare RD. Psychopathy: Theory, research, and implications for society. Dordrecht, The Netherlands: Kluwer; 1998.
49. Widiger TA, Lynam DR. Psychopathy and the five-factor model of personality. In: Millon T, Simonson E, Birket-Smith M, et al., eds. Psychopathy: Antisocial, criminal, and violent behavior. New York: Guilford; 1998: 171–87.
50. Fisher G. Psychological needs of heterosexual pedophiliacs. Diseases of the Nervous System 1969; 30:419–21.
51. Wilson GD, Cox DN. Personality of paedophile club members. Personality & Individual Differences 1983; 4:323–9.
52. Howells K. Some meanings of children for pedophiles. In: Cook M, Wilson G, eds. Love and Attraction. London: Pergamon Press; 1979:519–26.
53. Fisher D, Beech A, Browne A. Comparison of sex offenders to nonoffenders on selected psychological measures. Int J Offender Ther Comp Criminol 1999; 43:473–91.
54. First MB, Gibbon M, Spitzer RL, et al. User’s guide for the Structured Clinical Interview for DSM-IV Axis II Personality Disorders. Washington, DC: American Psychiatric Press; 1997.
55. Pfohl B, Blum N, Zimmerman M. Structured interview for DSM-IV Personality. Washington, DC: American Psychiatric Press; 1997.
56. Hyler SE. Personality Diagnostic Questionnaire–4. New York: New York State Psychiatric Institute; 1994.
57. Black DW. The epidemiology and phenomenology of compulsive sexual behavior. CNS Spectrums: the International Journal of Neuropsychiatric Medicine 2000:5;1,26–35.
58. Millon T. Manual for the MCMI-II. Minneapolis, MN: National Computer Systems; 1987.
59. Cohen LJ, Nikiforov K, Watras-Gans S, et al. Heterosexual male perpetrators of childhood sexual abuse: A preliminary neuropsychiatric model. Psychiatr Q 2002; 73:313–35.
60. Freund K, Kuban M. The basis of the abused abuser theory of pedophilia: A further elaboration on an earlier study. Arch Sex Behav 1994; 23:553–63.
61. Freund K, Watson R, Dickey R. Does sexual abuse in childhood cause pedophilia? An exploratory study. Arch Sex Behav 1990; 19:557–68.
62. Bagley C, Wood M, Young L. Victim to abuser: Mental health and behavioral sequels of child sexual abuse in a community survey of young adult males. Child Abuse Negl 1994; 18: 683–9.
63. Seghorn TK, Prentky RA, Boucher RJ. Childhood sexual abuse in the lives of sexually aggressive offenders. J Am Acad Child Adolesc Psychiatry 1987; 26:262–7.
64. Dhawan S, Marshall WL. Sexual abuse histories of sexual offenders. Sex Abuse 1996; 8:7–15.
65. Hindman J, Peters JM. Polygraph testing leads to better understanding of adult and juvenile sex offenders. Federal Probation 2001; 65:8–15.
66. Ahlmeyer S, Heil P, McKee B, et al. The impact of polygraphy on admissions of victims and offenses in adult sexual offenders. Sex Abuse 2000; 12:123–38.
67. Kircher JC, Raskin DC. Polygraph techniques: History, controversies, and prospects. In: Suedfeld P, Tetlock PE, eds. Psychology and social policy. Washington, DC: Hemisphere Publishing; 1992: 295–308.
68. Freund K, Kuban M. Deficient erotic gender differentiation in pedophilia: A follow-up. Arch Sex Behav 1994; 22:619–28.
69. Howard RC, Longmore FJ, Mason PA, et al. Contingent negative variation (CNV) and erotic preference in self-declared homosexuals and in child sex offenders. Biol Psychiatry 1994; 38:169–81.
70. Abel GG, Mittelman M, Becker JV, et al. Predicting child molesters’ response to treatment. Ann N Y Acad Sci 1988; 528: 223–34.
71. Dabbs JM, Mohammed S. Male and female salivatory testosterone concentrations before and after sexual activity. Hysiol Behav 1992; 52:195–7.
72. Pirke KM, Kockott G, Dittmar F. Psychosexual stimulation and plasma testosterone in man. Arch Sex Behav 1974; 3: 577–84.
73. Bradford JM, McLean D. Sexual offenders, violence and testosterone: A clinical study. Can J Psychiatry 1984; 29: 335–43.
74. Bain J, Langevin R, Hucker S, et al. Sex hormones in pedophiles: I. Baseline values of sex hormones: II. The gonadotropic releasing hormone test. Annals of Sex Research 1988; 1:443–54.
75. Gaffney GR, Berlin FS. Is there hypothalamic-pituitary-gonadal dysfunction in paedophilia? Br J Psychiatry 1984; 145: 657–60.
76. Garavan H, Pankiewicz J, Bloom A, et al. Cue-induced cocaine craving: Neuroanatomical specificity for drug users and drug stimuli. Am J Psychiatry 2000; 157:1789–98.
77. Wright P, Nobrega J, Langevin R, et al. Brain density and symmetry in pedophilic and sexually aggressive offenders. Annals of Sex Research 1990; 3:319–328.
78. Flor-Henry P, Lang RA, Koles ZJ, et al. Quantitative EEG studies of pedophilia. Int J Psychophysiol 1991; 10:253–8.
79. Flor-Henry P, Lang RA, Koles ZJ, et al. Quantitative EEG studies of genital exhibitionism. Annals of Sex Research 1988; 1:49–62.
80. Stein DJ, Hugo F, Oosthuizen P, et al. Neuropsychiatry of hypersexuality. CNS Spectrums: the International Journal of Neuropsychiatric Medicine 2000;5:1, 36–48.
81. Terzian H, Dalle Ore G. Syndrome of Kluver and Bucy reproduced in man by bilateral removal of temporal lobes. Neurology 1955; 5:373–80.
82. Lilly R, Cummings JL, Benson DF, et al. The human Kluver-Bucy syndrome. Neurology 1983; 33:1141–5.
83. Demerdash A, Shaalan M, Midani A, et al. Sexual behavior of a sample of females with epilepsy. Epilepsia 1991; 32:82–5.
84. Stoléru S, Grégoire M-C, Gérard D, et al. Neuroanatomical correlates of visually evoked sexual arousal in human males. Arch Sex Behav 1999; 28:1–21.
85. Bechara A, Damasio A, Damasio H, et al. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 1994; 50:7–15.
86. Araji S, Finkelhor D. Explanations of pedophilia: Review of empirical research. Bull Am Acad Psychiatry Law 1985; 13: 17–37.
87. Abel GG, Huffman J, Warberg BW, et al. Visual reaction time and plethysmography as measures of sexual interest in child molesters. Sex Abuse 1998; 10:81–95.
88. Bradford J. The treatment of sexual deviation using a pharmacological approach. J Sex Res 2000; 37:248–57.
89. Reilly DR, Delva NJ, Hudson RW. Protocols for the use of cyproterone, medroxyprogesterone and leuprolide in the treatment of paraphilia. Can J Psychiatry 2000; 45:559–63.
90. Greenberg DM, Bradford JMW. Treatment of the paraphilic disorders: A review of the role of the selective serotonin reuptake inhibitors. Sex Abuse 1997; 9:349–60.
91. Gijs L, Gooren L. Hormonal and psychopharmacological interventions in the treatment of paraphilias: An update. J Sex Res 1996; 33:273–90.
92. Rösler A, Witztum E. Pharmacotherapy of paraphilias in the next millenium. Behav Sci Law 2000; 18:43–56.
93. Marshall W, Anderson D, Fernandez Y, eds. Cognitive behavioral treatment of sexual offenders. New York: John Wiley & Sons; 1999.
94. Burton DL, Smith-Darden JP. 1996 nationwide survey: A survey of treatment programs and models serving children with sexual behavior problems, adolescent sex offenders, and adult sex offenders. Brandon VT: Safer Society Press; 2000.
95. Hanson RK, Steffy RA, Gauthier R. Long-term recidivism of child molesters. J Consult Clin Psychol 1993; 61:646–52.
96. Hanson RK, Bussiere MT. Predicting relapse: A meta-analysis of sexual offender recidivism studies. J Consult Clin Psychol 1998; 66;348–62.