Sexual Paraphilias

Cindy M. Meston & Penny Frohlich 


According to the DSM-IV, in order to be diagnosed with a paraphilia, one must demonstrate the following features:

  • “Recurrent, intense sexually arousing fantasies, sexual urges, or behaviors generally involving 1) nonhuman objects, 2) the suffering or humiliation of oneself or one’s partner, or 3) children or other nonconsenting persons, that occur over a period of at least 6 months.”
  • The behavior, sexual urges, or fantasies cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

The DSM-IV lists eight types of paraphilic disorders but in practice, individuals displaying one paraphilia very often also exhibit other paraphilic behaviors. Incarcerated pedophiles often report, for example, that they have also engaged in other paraphilic behaviors (e.g., exhibitionism, voyeurism) and that deviant sexual behaviors other than pedophilia are their primary interest. The presence of paraphilic behavior may represent an underlying sexual impulsivity disorder that is characterized by sexual compulsivity and hypersexuality, and in some cases, aggression (Kafka, 1997).

Back to top


According to the DSM-IV, fetishism involves “recurrent, intense sexually arousing fantasies, sexual urges, or behaviors involving the use of nonliving objects” as sexual stimuli (American Psychiatric Association, 1994). Most fetishists are male and nearly one in four are homosexual. Common fetish items include shoes and lingerie and common materials include rubber and leather. Fetishists become aroused by stealing the object, viewing the object, or masturbating with the object. Most fetishists are aroused by a number of different objects. The etiology of fetishism is not known. Two reported cases of fetishism have been associated with abnormalities in the temporal lobe. In one case the patient had temporal lobe epilepsy and in the other the fetish behavior was linked to the development of a temporal lobe tumor (Wise, 1985). Some evidence suggests that fetishism may be a learned behavior that results when a normal sexual stimulus is paired with the fetish item. Seven heterosexual males free from any prior fetish were repeatedly shown erotic stimuli paired with a slide of a black knee-length women’s boot. When the slide of the boot was later shown alone, five of the seven men demonstrated penile erection, indicating that a boot fetish had been conditioned. The conditioned fetish was shown to generalize to other types of shoes in three of the men. That is, the men also became aroused when shown a slide of a high-healed black boot and a low-healed black shoe. They did not become aroused to a slide of a short brown boot, a brown string sandal, or a golden sandal, suggesting that the fetish only generalized to similar types of shoes (Rachman & Hodgson, 1968). A similar study was conducted in women to determine whether women could also be conditioned to become sexually aroused to a stimulus. Subjects were randomly assigned to repeatedly view an erotic film paired with a light stimulus versus an erotic film alone. No significant differences where found in physiological sexual arousal between the experimental and control groups when a light stimulus was later presented alone (Letourneau & O’Donohue, 1997). Meston and Rachman (1994) tried to condition sexual arousal to the sound of a male’s voice. Even after repeated pairings of erotic video clips and the male’s voice, later presentation of the male’s voice alone did not produce sexual arousal. This suggests that sexual arousal is not readily classically conditioned in women and may explain why, like other paraphilias, fetishism occurs almost exclusively in men.

Back to top


Transvestic fetishism is diagnosed in heterosexual males who experience “recurrent, intense sexually arousing fantasies, sexual urges, or behaviors involving cross-dressing” (American Psychiatric Association, 1994). A distinction is drawn between transvestism (cross-dressing) and transvestic fetishism. A variety of people cross-dress but the behavior is not considered a fetish unless the cross-dressing is associated with sexual feelings. For example, transsexuals, or people who feel that their external sex does not match their internal gender identity, may cross-dress in order to feel more congruent with their gender identity but do not find the cross-dressing sexually arousing. Similarly, homosexual males may cross-dress (e.g., drag-queens), but the cross-dressing is not considered to be a fetish unless it is sexually arousing.

Very few studies have been published regarding transvestic fetishism and those that have often grouped transvestic fetishists with transvestites who experienced little to no sexual arousal from cross-dressing. Doctor and Prince (1997) surveyed 1,032 male transvestites between 1990 and 1992. They found that 40% of respondents found cross-dressing “often” or “nearly always” sexually exciting but only 9% described themselves as a “fetishist [who] favored women’s clothing.” While keeping in mind that it is unclear what percentage of subjects would meet DSM-IV criteria for transvestic fetishism, the following characteristics were reported. Respondents ranged in age from 20 to 80 years of age, lived throughout the United States, and reported a range of religious affiliations (24% were Catholic, 38% were Protestant, 3% were Jewish, 10% were agnostic, and 25% were with other religious affiliations). The majority of respondents were well educated (65% had at least a B.A.), in committed relationships, and had children. Of those currently married, 83% reported that their wives were aware of their transvestic tendencies at present, but only 28% accepted the behavior. The vast majority reported a heterosexual orientation (87%) although 29% reported having had homosexual experiences. The majority of respondents began cross-dressing before age 10 (66%) or between age 10 and 20 (29%), had been raised by both parents (76%), and reported that their father “provided a good masculine image”(76%)

A few cases have reported of men with transvestic fetishism who had fathers or brothers who also cross-dressed. Since so few cases of familial co-occurrence have been reported in the literature, and because the occurrence of transvestic fetishism in the general population is not known, it is not clear whether family environment and/or genetics contributes to the likelihood of developing a cross-dressing fetish. Transvestic fetishism is associated with learning disabilities, and a few cases of transvestic fetishism have been associated with temporal lobe abnormalities (Zucker & Blanchard, 1997).

A number of studies have been published examining psychosocial causes of transvestic fetishism but most have serious methodological flaws that limit drawing confident conclusions. Some such studies suggest that adolescents with transvestic fetishism tendencies may have a history of separation from and hostility towards their mothers. The cross-dressing may serve as a means to make a connection with females, even if that connection often involves some expressions of anger and hostility (Zucker & Blanchard, 1997).

Back to top


Pedophilia is defined as intense and repeated sexually arousing fantasies, urges, or behaviors involving sexual activity with children, typically less than 14 years old (American Psychiatric Association, 1994). Since few pedophiles are likely to openly admit their preference, it is difficult to estimate the prevalence of pedophilia in the general population. Furthermore, individuals who feel sexual attraction to children may resist the temptation due to societal pressures, yet may nonetheless experience sexual fantasies involving children. Recent evidence suggests that pedophilia may be associated with homosexuality, mental retardation, and high maternal age. Homosexuality in the general population is estimated at 2% while homosexuality in pedophiles is estimated at up to 40% .When sexual orientation, intellectual functioning, and maternal age were measured in 991 male sex offenders, high maternal age and low intellectual functioning were significantly associated with homosexual pedophilia. The association between low intelligence and pedophilia suggests that pedophilia may reflect a developmental disorder. The association between high maternal age and pedophilia is unclear, although is may reflect differences in birth order as homosexuality is associated with being later born (discussed below under gender identity disorder)(Blanchard et al., 1999).

Some researchers have speculated that a childhood history of sexual abuse contributes to an adult preference for sexual activity with children. In a large sample of men who were child sex offenders, Freund et al. (1990) found that heterosexual and homosexual pedophiles were significantly more likely to report childhood sexual abuse by a male abuser (versus female abuser) as compared to controls. Freund and Kuban (1994) classified child sex offenders according to whether they demonstrated phallometric (increased penile volume) preference to photographs of nude children versus adults. They found that child sex offenders who demonstrated preference for children were significantly more likely to have a childhood history of sexual abuse. It should be noted that although reports indicate approximately 49% of pedophiles have a history of childhood sexual abuse, very few people with a history of childhood sexual abuse become pedophiles (Freund & Kuban, 1994).

Pedophiles may have difficulty with gender differentiation. Freund et al. (1991) showed slides of nude male and female children and adults to pedophiles and controls, and measured penile volume changes. The pedophiles demonstrated less differentiation between stimuli containing males versus females as compared to non-pedophiles. Although this pattern of undifferentiated arousal has also been noted in a case study of a 20-year-old woman with multiple paraphilias (Cooper et al., 1990), few cases of female pedophilia have been reported in the literature.

Pedophiles may differ from non-pedophiles on several physiological dimensions as well. Baseline plasma cortisol, prolactin, and body temperature were significantly higher in pedophiles than controls. When both groups were administered a serotonin agonist, mCPP, versus placebo, plasma cortisol levels were more elevated and remained elevated longer for pedophiles compared to controls. The pedophiles reported experiencing side effects (e.g., dizzy, restless) of mCPP administration while the controls did not. Consistent with these findings, some researchers have speculated that pedophilia may be associated with disturbances in serotonin-related aggression and impulsivity (Maes et al., 2001). It has also been suggested that pedophilia may be a subtype of obsessive-compulsive disorder; a problem that is marked by repetitive, irrepressible behavior associated with serotonin disregulation (Balyk, 1997).

Back to top


The DSM-IV defines sexual masochism as “recurrent, intense sexually arousing fantasies, sexual urges, or behaviors involving the act (real, not simulated) of being humiliated, beaten, bound, or otherwise made to suffer” (American Psychiatric Association, 1994). In 1886, Krafft-Ebing coined the term, masochist, after Leopold von Sacher-Masoch, who wrote novels depicting men being humiliated and bound by females. Sexual sadism is characterized by “recurrent, intense sexually arousing fantasies, sexual urges, or behaviors involving acts (real, not simulated) in which the psychological or physical suffering (including humiliation) of the victim is sexually exciting to the person” (American Psychiatric Association, 1994). The term, sadism, was derived from writings of the Marquis de Sade, an 18 th century author who wrote stories depicting sexual torture and brutality. A distinction is drawn between minor versus major sexually sadistic acts. Minor sexually sadistic acts would include, for example, humiliation and bondage of a willing sexual masochist while major sexually sadistic acts would involve acts such as sexual torture and rape of an unwilling participant. The key distinction here is whether the victim was consenting or not.

The practice of sadomasochism (referred to as S&M), or the consensual participation between sexual sadist and sexual masochist, involves carrying out predetermined sexual scenarios. These scenarios commonly involve several themes: flagellation (usually on the buttocks), bondage, “water sports” (urophilia – attraction to urine, coprophilia – attraction to feces, and mysophilia – attraction to filth), and penis and nipple torture (Arndt, 1991). Sadomasochists interviewed in New York and San Francisco between 1976 and 1983 reported S&M activities that included elements of dominance and submission, role-playing (e.g., master and slave), consensuality (i.e., both participants were willing), and were of sexual context (i.e., the role-playing was sexual)(Weinberg et al., 1984). Commonly reported S&M role-play themes include: “severe boss and the naughty secretary,” “the queen and many slaves,” “ the male barber and his customer,” and “arrest scenes and military training” (Sandnabba et al., 1999). Although the sexual sadist appears to be in control, often the degree of domination and humiliation is agreed upon earlier, and it is the sexual masochist who indicates with a predetermined cue when he/she has reached his/her limit (Arndt, 1991).

Female sexual masochists and sadists are outnumbered by male sexual masochists and sadists and in many cases, the females are prostitutes who specialize in sadomasochism. One study found that approximately a quarter of female sexual sadists are prostitutes (Breslow et al., 1985). Approximately 80% of sadomasochists reported that they were regularly engaging in sadomasochistic activities by age 30 years (Sandnabba et al., 1999). Spengler (1977) obtained questionnaire data from 245 male sadomasochists recruited through S&M magazine advertisements and via S&M clubs. The majority of respondents reported that they met partners through sadomasochism advertisements, clubs, or bars. The sample contained 30% heterosexual sadomasochists, 31% bisexual sadomasochists, and 38% homosexual sadomasochists. The respondents came from all ages, socioeconomic backgrounds and levels of education. In most cases, the families knew little if anything about the respondents’ S&M activities; 41% of married respondents (n=109) reported that their wives knew nothing about the sadomasochistic activity. When queried whether they thought the sadomasochistic behavior was acceptable, 70% indicated acceptance of the behavior, 85% reported that they “want to do it again,” “it was fun” (84%), and “sexually satisfying” (79%). Although many of the respondents reported that they enjoyed non-sadomasochistic sexual activity, they reported being more likely to orgasm with sadomasochistic activity (79%) than without (45%). About a third of respondents reported fetishisms (e.g., boots and leather).

Very few studies have been conducted examining sexual sadists who target unwilling victims. Seto and Kuban (1996) examined penile volume changes in seven sadistic rapists compared to 14 non-sadistic rapists and 20 controls. The subjects were presented audiotapes depicting five different scenarios: (1) nonviolent, nonsexual interaction with a female, (2) consensual sexual activity with a female, (3) nonsexual violence against a female, (4) rape, and (5) violent rape. Compared to controls, the sadistic rapists and non-sadistic rapists were equally aroused by the different types of sexual contact – they were less likely to differentiate between consensual sexual activity, rape, and violent rape.

A subset of sexual sadists may have abnormal endocrine activity although hormone levels typically do not differ between sexual sadists and controls. In a review of individual cases, one sexual sadist had unusually high levels of luteinizing hormone (stimulates progesterone secretion) and follicle-stimulating hormone (stimulates estradiol in women and sperm development in men), another had low testosterone levels and another Klinefelter’s syndrome (XXY chromosomes rather than the typical XY male pattern). Gross examination of brain functioning revealed no differences between sexual sadists and controls, but more careful examination revealed a subtle but significant difference in the right temporal lobe. Forty-one percent of the sexual sadists had a slightly dilated right temporal horn, compared to 13% of controls. One sexual sadist had a slow growing tumor in the left frontal-temporal lobe, likely present since childhood. Another had enlargement of the ventricles, a condition typically associated with schizophrenia and suggestive of overall brain atrophy. In short, temporal lobe abnormalities may be implicated in sexual sadism, but more information is needed before any strong conclusions can be made (Langevin et al., 1988).

Serial killing, which is often reported in the media and dramatized in movies, may reflect comorbid sexual sadism and antisocial personality disorder. Geberth and Turco (1997) examined records of 387 serial murderers within the United States and found that 248 had sexually assaulted their victims. These included famous cases of serial killing, such as Theodore (Ted) Bundy and the Green River Killer. Of these, they determined that 68 met DSM-IV criteria for both sexual sadism and antisocial personality disorder (in other cases, sufficient data was not available to make a determination). These 68 individuals displayed a pattern of behavior characterized by childhood aggressiveness and antisocial behavior, and a pattern of killing involving sexual violence, humiliation, domination and control. Examination of their records suggests that these 68 individuals engaged in sexual violence and killing because they derived pleasure from it.

Back to top


Voyeurism, exhibitionism, and frotteurism may be different behavioral expressions of a single underlying courtship disorder. The overt behaviors differ, but can also be conceptualized as different stages on a continuum – different degrees of proximity to the victim. Voyeurism involves viewing the victim from a distance, exhibitionism involves approaching the victim, and frotteurism involves physically touching the victim. The preference for rape over consensual sexual activity (termed the preferential rape pattern) may represent the forth phase in the courtship disorders (Freund et al., 1983). A common etiological factor has not been identified although evidence indicates that the courtship disorders are associated with a preference for eliciting an alarmed reaction from an unfamiliar target rather than any lack of interest in intercourse (Freund & Watson, 1990). A high degree of comorbidity exists between these disorders and even when no overt comorbid behavior is present, some evidence suggests that presence of one disorder predisposes to another such disorder (Freund et al., 1983).

Back to top


The DSM-IV defines voyeurism as “recurrent, intense sexually arousing fantasies, sexual urges, or behaviors involving the act of observing an unsuspecting person who is naked, in the process of disrobing or engaging in sexual activity” (American Psychiatric Association, 1994). Most men, if given the opportunity to view a woman disrobing, would not avert their eyes. A man who engages in an opportunistic “peep,” is not a voyeur, the peeping must be recurrent and the urges to do so intense. Voyeurs tend to be the youngest child in the family. Compared to other sex offenders and controls, voyeurs have fewer sisters, have a good relationship with both parents, but have parents who do not have a good marital relationship. Voyeurs are often underdeveloped socially and sexually. They tend to engage in sexual activity later than other groups, and are less likely to marry than controls and other sex offenders (Smith, 1976). The more sexually experienced a voyeur, the more frequently he is likely to engage in peeping behavior (Langevin et al., 1985). Some evidence suggests that voyeurs may be predisposed to other paraphilias as well (e.g., sadomasochism, zoophilia)(Langevin et al., 1985).

Although voyeurism is rare in women, some evidence suggests that women have similar “peeping” urges as men. Friday (1975) interviewed women from all ages (teen to retirement) and walks of life and found that women expressed fantasies about peeping and, in some cases, engaged in actual peeping behavior.

Learning theorists have suggested that voyeurism develops when the subject is provided a voyeuristic opportunity, and then subsequently masturbates while fantasizing about the experience. Some evidence supports this hypothesis; 50% of voyeurs reported that prior to the onset of their peeping behavior they believed that normal sexual relations were not likely to be an option for them, and so they fantasized about scenarios they believed to be more obtainable, such as peeping. In addition, 75 % of voyeurs reported that the sexual scenario they envision while masturbating reflected their first peeping experience (Smith, 1976).

Back to top


Exhibitionism is defined as “the exposure of one’s genitals to an unsuspecting stranger” (American Psychiatric Association, 1994) and involves some form of sexual gratification. Exhibitionism occurs almost exclusively in men. A very few cases of female exhibitionists have been reported in the literature, but the characteristics of these women differed from typical male exhibitionists. Male exhibitionists tend to be timid and unassertive men who have underdeveloped social skills and who are uncomfortable with angry or hostile feelings. Some studies suggest that exhibitionists were more likely to have been raised in a sexually puritanical background. The few female exhibitionists described in the literature, and studies examining female strippers, would suggest that the majority of female exhibitionists gain no pleasure from exposing their genitals but do so either to gain money or attention (Blair & Lanyon, 1981).

Behavioral theory proposes that exhibitionism develops as a result of a learned behavior that is subsequently reinforced. This theory has been applied successfully to the treatment of exhibitionism (i.e., a learned behavior can be replaced with a more socially acceptable behavior) but it is not clear whether this reflects the actually etiology of exhibitionism. Attempts to identify a physiological cause of exhibitionism have thus far been unsuccessful.

Back to top


Frotteurism involves “intense sexually arousing fantasies, sexual urges, or behaviors involving touching and rubbing against a nonconsenting person” (American Psychiatric Association, 1994). The majority of published articles on this disorder group frotteurism with other paraphilic disorders or report cases of men with multiple paraphilias, including frotteurism. Abel et al. (1987) examined 62 males diagnosed with frotteurism, as well as other paraphilic disorders, and found that, at the time of the interview, they had committed an average of 849 frottage acts. Rooth (1973) interviewed 561 nonincarcerated men with paraphilias and found that of those exhibiting frotteurism, 79% had other paraphilias, with an average of 4.8 paraphilias each.

It is unclear whether true frotteurism in women exists, perhaps in part because of the decreased likelihood that male victims would view the behavior as unwelcome or threatening. A handful of case reports of sexual molestation of men by women have been reported in the literature. The molestation typically occurred subsequent to erectile failure or inhibited desire (Sarrel & Masters, 1982). Although these cases do not represent female frotteurism, they suggest that it is feasible that rare cases of female frotteurism may exist, but are rarely reported.

Back to top


In the mid 1900’s, some European countries used castration as a means of treating exhibitionism, pedophilia, and other forms of sexual crimes. In West Germany, psychosurgery, which involved removing the nucleus ventromedialis of the hypothalamus, was used as a treatment for male sex offenders. Published reports of these practices rarely provided sufficient information to determine whether this intervention was successful in eliminating the inappropriate sexual behavior. Of course there are serious consequences to performing such extreme and permanent techniques.

Cognitive-behavioral therapies, such as aversion therapy, are often used to treat paraphilias. The arousing stimulus is paired with an aversive stimulus such as a shock or noxious odor until the paraphilic behavior no longer produces sexual arousal. A review of the handful of studies and case reports published, suggest that aversion therapy alone is effective in reducing arousal, but that relapse rates are high (Kilmann et al., 1982). More recently, other forms of cognitive-behavioral therapy such as covert sensitization or orgasmic reconditioning are being used. Orgasmic reconditioning involves fantasizing about the paraphilic behavior while masturbating, and at the moment just before orgasm, switching the fantasy to a more acceptable stimulus, such as one’s partner. The belief is that orgasm, being an intensely pleasurable sensation, will serve to reinforce the more accepted sexual fantasy. Few well-controlled treatment outcome studies have been published, however, making it difficult to determine whether these types of interventions are effective. Covert sensitization involves fantasizing about the paraphilic behavior followed by imagining a noxious scenario, such as vomiting, or an undesirable consequence such as being discovered by one’s family. It is not yet clear how successful these techniques are in eliminating the behavior although a few reports indicate that they can be highly successful for some patients.

Pharmacological interventions include hormonal supplements or psychotropic medications. Hormonal treatments are designed to inhibit deviant sexual behavior by reducing sexual drive and sexual arousal. They include the following: (1) estrogen; (2) medroxyprogesterone acetate (MPA), which lowers plasma testosterone and reduces gonadotropin secretion; (3) luteinizing hormone-releasing hormone agonists (LHRH agonists), which produce the pharmacological equivalent of castration by significantly inhibiting gonadotropin secretion; and (4) antiandrogens such as cyproterone acetate (CPA), which blocks testosterone uptake and metabolism. Treatment outcome studies suggest that these treatments are effective in reducing deviant sexual behavior provided that the treatment regimen is maintained, although more well-controlled treatment outcome studies are needed before the true effectiveness of these treatments can be determined. Psychotropic medications that affect the serotonin systems have recently been used to treat paraphilias. Clinical studies suggest that SSRIs such as Prozac are effective in reducing paraphilic arousal and may be effective in reorienting arousal to more socially acceptable scenarios. The effectiveness of SSRIs in reducing paraphilic fantasies and behaviors suggests that these disorders may have an obsessive-compulsive component, as SSRIs are often used to treat obsessive-compulsive disorders. As with hormone treatments, however, more well-controlled treatment outcome studies must be conducted before the true effectiveness of these treatments can be determined (Bradford, 2000).

Back to top