Some studies suggest that the prevelance of pedophilia may be between 3% and 5% in the general population (as reviewed by Seto, 2009). In penile plethysmography studies of men with sexual offense histories against children, these prevalences can jump from 30% for men with one offense to 61% for men with 3 or more sexual offenses against children (Blanchard, 2010; Seto, 2009).
Much higher is the prevalence in an older anonymous self-report survey study of 193 healthy male college students: 21% admitted some degree of sexual interest in children, 9% admitted to having sexual fantasies involving children, 5% admitted to masturbating to orgasm through these fantasies, and 9% admitted that they would have sex with a child, if it were guaranteed they would never be caught (Briere and Runtz, 1989). Yet, it is important to note that this study did not specifically investigate the preference of pedophilia, rather sampled fantasy.
Considering the lack of reliable estimates of the prevalence of pedophilia in general, prevalence estimates for the subtypes of pedophilia also remain scarce. The current estimated prevalence of homosexual pedophilia is anywhere between 9 and 40% (Hall and Hall, 2007); the ratio of heterosexual to homosexual pedophiles was approximately 1.4:1 among men with CSA offenses in another study using phallometry (Freund and Watson, 1992). Prevalence estimates of bisexual pedophilia are not yet available due to measurement complexity (Hall and Hall, 2007).
Child pornography use is also strongly related to pedophilia. As a study deriving from the German Dunkelfeld Prevention Project concluded, among 345 pedophiles admitting one or more sexual offenses against children, 37% have solely used child pornography, 21% committed exclusively hands-on sexual contacts with a minor, and 42% have committed both (Neutze et al., 2012).
Self-reported interest in children, child pornography use, and the number of children as sexual victims all uniquely contribute to phallometrically assessed sexual interest in children (Mokros et al., 2012b). Additionally, child pornography users showed a greater phallometric response to sexual child stimuli than non-pedophilic child sexual offenders (CSO), and there was no significant difference within the child pornography group between those who had committed sexual offenses against children and those who had not (Seto et al., 2006).
Eye tracking and pupil dilation may also indicate sexual preference and results show that men react more strongly in these studies than women. Heterosexual men did initially orient to their stated preference and eye fixations were significantly longer than when looking at non-preferred stimuli (Fromberger et al., 2012b). In a study investigating pedophilia, eye tracking produced high sensitivity and specificity, 86.4 and 90.0%, respectively (Fromberger et al., 2012a). Heterosexual women reacted similarly to stimuli of both sexes, whereas heterosexual men, homosexual men, and homosexual women reacted most strongly to their stated partner gender in pupil dilation research (Rieger and Savin-Williams, 2012). However, criticisms have been put forward suggesting that the success seen in heterosexual and homosexual participants to respective stimuli in pupil dilation studies is attributable to factors other than sexual preference, such as luminance, salience of the stimuli, and emotional reaction (Beatty and Lucero-Wagoner, 2000; Rieger and Savin-Williams, 2012). These methods have not yet been used in the sexual age preference measurement of pedophiles, but do hold promise as collateral information for diagnosis.
Kramer (2011) addresses a point that currently many pedophilia researchers are facing: should we continue to classify pedophilia as a separate psychiatric disorder or as a sexual orientation, when patients harbor complaints not only of the preference but of the pressure under which they suffer? This pressure often precedes the onset of psychiatric illness (most often mood or anxiety disorders), which then precedes the decision to seek psychiatric help (Kramer, 2011). Due to a temporal-causal relationship being nearly impossible to determine in these cases, the DSM-5 has differentiated among those who experience the sexual preference but do not suffer and those who do, leading us back to Pedophilia vs. Pedophilic Disorder, regardless of whether or not child sexual offenses have occurred (Kramer, 2011).
In an empirical study comparing 20 forensic inpatients with pedophilia attracted to males or attracted to females (but lacking information of whether they were exclusive or non-exclusive types, respectively) and 24 matched male controls on various psychiatric measures, findings included increased personality subscale scores from the MMPI-2 for psychopathy and paranoia, with enhanced scores for hypochondriasis, depression, hysteria, and masculinity/femininity, psychasthenia, schizophrenia, and social introversion (Kruger and Schiffer, 2011). Furthermore, that study shows that 61.1% of the sample qualified for a personality disorder diagnosis, with Borderline Personality Disorder (22%; from Cluster B) and Avoidant Personality Disorder (33%; from Cluster C) as the two most common (Kruger and Schiffer, 2011). Self-report results in low socioeconomic status individuals, often including non-pedophilic sexual offenders against children and rapists, indicate more social anxiety, less social poise, and a decreased ability to appropriately socially assert oneself, relating to the cognitive distortions seen among these groups of negative attitudes toward women, reinforcing beliefs about sex with children, denial of harm to victims, and misattribution of responsibility of offending (Geer et al., 2000).
Initial studies exploring the neuropsychological correlates of pedophilia often used uncontrolled designs with incarcerated pedophilic men. An initial study by Tarter et al. (1983) among recently incarcerated adolescent offenders and controls, no neuropsychological differences were found among the groups on the Wechsler Adult Intelligence Scale (WAIS) or Pittsburgh Initial Neuropsychological Test System. Furthermore, among incarcerated adult male sex and non-sex non-violent offenders, no differences were seen in any neuropsychological test variables, after age and education status were accounted for (Abracen et al., 1991). Keeping in mind that the subject groups were incarcerated at the time of study and that pedophilia was not explicitly examined, the results are limited in their generalizability and specificity to pedophilia.
In another study, heterosexual and homosexual pedophiles were tested, but each group, plus one control group, had sexual offense histories and were incarcerated at the time of the study. Fully admitting, heterosexual pedophiles had gender differentiation indices (or the erotic sensitivity for the gender-differentiating body shapes that distinguish physically mature males and females), which were greater than for non-admitting heterosexual pedophiles, but no differences were found for either homosexual or bisexual pedophiles. This highlights that fully admitting heterosexual pedophiles prefer the body shapes of female children, whereas the partial- or non-admitting pedophiles do not seem to discriminate between victim body types (Freund et al., 1991; Freund and Kuban, 1993). Gillespie and McKenzie (2000) investigated neuropsychological differences among forensically incarcerated sex offenders and non-sex offenders and found no significant differences on any of their measures, including the WAIS, Trail Making Task, List Learning test, Controlled Oral Word Association test, and National Adult Reading Task (NART). Among personality disordered offenders, violent sex offenders, violent non-sex offenders, and non-violent, non-sex offenders, no differences were noted on any neuropsychological test variables, including the WAIS, Trail Making Task, Face Recognition, Wechsler Memory Scale-Revised, and Wechsler Recognition Memory Task (Dolan et al., 2002). As with previous studies, neither incarceration status nor sexual preference was controlled for, which limits the generalizability results to pedophilic men.
These results suggest that disturbed and prosecuted pedophiles do show deficits in executive functioning, which might be due rather to mental disturbances and not to the sexual preference. This is in contrast to findings from the APSD/psychopathy literature that suggests, at least among community samples measured for psychopathic traits, that these personality traits are linked to deficits in response inhibition and impulsivity, specifically with social deviancy associated with overall deficits in executive functioning and response inhibition, whereas callous-unemotional traits observed in psychopathy are associated with improved executive functioning abilities (Sellbom and Verona, 2007). In ASPD, broad executive function domain deficits have been noted in response inhibition, planning, and rule acquisition, and reversal learning, suggesting that previous studies examining pedophilia may have been measuring ASPD or simply an incarceration stress effect in their incarcerated samples rather than pedophilia. This is in contrast to recent studies that have found processing speed impairments in pedophiles, but few other deficits suggestive more of offense status effects than sexual preference effects (Eastvold et al., 2011; Kruger and Schiffer, 2011; Schiffer and Vonlaufen, 2011; Suchy et al., 2014).
As discussed in a previous review by Seto (2008), there are three major neurobiological theories, which have come to be connected to pedophilia but all have the same shortcoming that they rely on data based on cases of pedophiles who have other psychological disorder diagnoses, are incarcerated or otherwise legally sanctioned, or are not sufficiently diagnostically classified (i.e., not differentiating between the exclusive or the non-exclusive type, etc.).
The third major neurobiological theory holds that differences in the sex dimorphic brain structures affected by the masculinization of the male brain would more strongly affect pedophilia development. Furthermore, the volumes of these structures would be influenced, but the hypothesis failed to state in what direction these changes occur, i.e., either increased or decreased volumes as a result of testosterone exposure. In the frontal and temporal lobes, these differences would be limited to those sexually dimorphic structures, rather than a generalized difference in region volume, but research has not supported the hypothesis (Cantor et al., 2008). 2b1af7f3a8