香京julia种子在线播放

Sexual victimization is very common in the general population (Sardinha et al., 2022). However, it has long been under-recognized (Leight, 2022) because victims remain silent, due to limitations in being able to report and reports often not being believed. This is where sexual abuse differs from other forms of violence: most victims who report it are met with suspicion and denial and therefore are reluctant to talk about it (Ahrens, 2006; Suarez and Gadalla, 2010; Kennedy and Prock, 2018). While the underlying causes of such negative responses remain to be fully elucidated, they can be explained by the strong persistence of myths about sexual violence (e.g., “women provoke rape by the way they dress,” “men have uncontrollable sexual urges,” “it is not rape because she didn’t say no,” “she falsely accused him of rape because she wants his money,” etc.) (Burt, 1980; Payne et al., 1999; Flood and Pease, 2009). Such beliefs prevent victims from reporting the assaults, and sometimes also prevent them from identifying that the assault was an actual sexual aggression (Weiss, 2009). Indeed, feelings of guilt, denial and shame, as well as the legitimate fear of being blamed and stigmatized, are still extremely prevalent among victims, who find it very challenging to talk about the assault (Ahrens, 2006). Even professionals such as clinicians or law officers are far from being immune to such prejudices and do not offer the safe space that would help victims recover thanks to adapted care and legal action (Oram, 2019).

The last decade, however, has seen a steady change taking place: there is a growing movement of victims who have found strength in numbers through social media. They have become able to face social stigma in order to report what happened to them (e.g., child abuse in church, abuse on women in the entertainment industry, abuse on teenagers in competitive sports, etc.). Such social movements might be beneficial for victims in that they realize that they are not alone and can find support. They might also be beneficial for society in that they help acknowledging the extent of the problem, since the massive amount of testimonies published everywhere demonstrate that sexual abuse victims are much more numerous than previously believed.

Statistics on prevalence in the general population indicate elevated rates of victimization all over the world. The recent WHO publication might be the most reliable reference on this topic (WHO, 2021). It states that “30% of women aged 15 years and older have been subjected to physical and/or sexual violence from any current or former husband or male intimate partner, or to sexual violence from someone who is not a current or former husband or intimate partner, or to both these forms of violence at least once” (p. XVI). In a recent study, Sardinha et al. (2022) ran meta-analyses of 366 eligible studies, representing responses of 2 million women aged 15 and older from 161 countries and areas. They found that “27% (uncertainty interval [UI] 23–31%) of ever-partnered women aged 15–49 years are estimated to have experienced physical or sexual, or both, intimate partner violence in their lifetime, with 13% (10–16%) experiencing it in the past year before they were surveyed” (Sardinha et al., 2022). In addition to instances of sexual assault during adulthood, one must also consider victimization during childhood: rates of sexual violence on children reach 12.7% (18% for girls and. 7.6% for boys), according to a meta-analysis by Stoltenborgh et al. (2015).

Sexual victimization rates represent about one woman out of three and one child out of ten. Readers may not have expected such elevated figures, especially since many previous studies have reported lower rates (although a few studies actually reported higher numbers) (Dworkin et al., 2017). This is because estimates of sexual violence depend on (i) the definition of sexual violence; (ii) the choice of investigation methods (analysis of formal complaints in police stations, clinical reports, telephone interviews, internet surveys, open questions, etc.); and (iii) the country where the investigation took place. Definitions of sexual violence are obviously crucial when assessing levels of victimization. For example, in their review of literature, Dworkin et al. (2017) state that “17–25% of women and 1–3% of men will be sexually assaulted in their lifetime.” In contrast, Cleere and Lynn (2013) report much higher figures in their study of 302 college women, with 60.93% of their sample having experienced some form of sexual aggression, including but not limited to unwanted sexual contact. Regarding country-related variations, rates may differ from one country to another. For example, in France, rates of victimization were higher for men (3.9%) and lower for women (14.5%) as compared to other countries. Such figures resulted from Hamel et al. (2016) the large-scale “Violence and Gender Relations” (VIRAGE) survey that was conducted by the French Institute for Demographic Studies (INED) in order to assess lifetime levels of sexual aggression (rape, attempted rape, unwanted sexual touching) in a sample of almost 16,000 women and 12,000 men, aged 20–69 years old.

Despite this geographical and methodological variability, the WHO (2021) study is actually very consistent with two other large-scale studies that were conducted in the United States of America. The National Crime Victimization Survey (NCVS) was made using two methods of identification: self-report versus interview, randomly assigned to a sample of 11,000 women aged 18–49, in order to assess rates of sexual aggression over the past 12 months. Their overall results show that 8.1% of participants had suffered sexual violence over the past year. Their analyses show that self-report led to slightly higher rates of reported assaults than interviews (Cantor et al., 2021). The other large United States-based study was titled “National Intimate Partner and Sexual Violence Survey.” It was conducted on more than 12,000 persons by the National Center for Injury Prevention and Control (CDC). Phone interviews regarding lifetime experience indicate that rape affects 19.3% of women (and 1.7% of men), other forms of sexual violence affect 43.9% of women (and 23.4% of men) (Breiding et al., 2014).

Several researchers have investigated whether autistic women would exhibit higher levels of victimization than women in the general population. Given that being on the autism spectrum condition is characterized by experiencing difficulties in social communication, such as decoding hidden intentions and emotions of others, understanding implicit communication and elements of context, it is expected that women on the spectrum may be at considerable risk for sexual victimization, a hypothesis confirmed by all published studies on this topic. To this end, we have attempted here to conduct an exhaustive review of such research.

Due to social and communication difficulties, many women on the autism spectrum join online autism communities where they can find help and support from peers and where they may feel more comfortable discussing issues of sexual violence that they have experienced. We have contacted several local organizations dedicated to autism (Association Francophone des Femmes Autistes, Asperger Amitié, Asperger Aide, France Asperger…) and asked them for their support by sharing to their community an internet link to a data collection site.

After a short description of the study, women were asked to participate in this survey if they were identifying as autistic women. Participants provided consent for their data to be used in the study after completing anonymously the online questionnaire. A trigger warning was included before questions regarding victimization, as well as an invitation for victims to contact a health practitioner. After pre-submission, the president of the ethics committee of Paris V decided that it was not necessary to review the study insofar as the participants were strictly anonymous (no identifying data were collected, nor IP addresses of respondents). As per French laws regarding data privacy, the questionnaire was declared to CNIL authorities (National Commission on Informatics and Liberty) under identifier # 2172655.

All participants answered to three questionnaires, for a total of 38 questions, using an online survey system⁴. All questions were multiple choice questions with the exception of questions #8, #13, #33, #36, #37 which were multiple response questions (i.e., “check all that apply”). In addition, questions #8, #9, #13, #33, #34, #36, #37 included an open field question: “Other (please specify).”

Statistical analyses were realized with R (R 4.1.1 2021.08.10). We used the following methods⁵ : descriptive statistics (percentages, means, medians and standard deviations). We used non-parametric statistical analyses because data was mostly non-continuous. We used Chi square test to assess for significant differences between subgroups. When applicable, we also used McNemar test and conditional probabilities in order to confirm Chi square tests’ results, because differences in sample sizes of our subgroups could increase type 1 and type 2 errors. In order to assess whether some characteristics of our sample increased measured risks (e.g., developing Post Traumatic Stress Disorder), we used three methods: two-sample test for equality of proportions, Wilcoxon test and univariate logistic regression with odds ratio and confidence intervals. Unless otherwise specified, we used Yates continuity correction as post hoc correction in all cases. Bonferroni correction was additionally used when applicable.

The description of our population sample is revealing of what autism in women looks like and is in line with previous studies (Lai et al., 2017; Ormond et al., 2018). Participants included in our study (n = 225) were of all ages but predominantly in their thirties. A quarter of participants were self-diagnosed.

We asked participants if they had received IQ testing. Among the 66.2% who answered yes, 82.5%, (n = 123, i.e., more than half of our total sample) obtained a score > 120 and 52.3% (n = 78, i.e., more than a third of our total sample) reported IQ > 130. Of course, we can’t extrapolate such numbers to the whole sample, because of a potential clinical bias: it cannot be excluded that IQ testing was more often realized in individuals displaying signs of high IQ, since clinicians may have found that such measures would prove helpful to this subpopulation. Nevertheless, 34.6% of our total sample scored over 130. If true (we had no direct access to the IQ tests; recall memory bias and social desirability bias cannot be excluded), these results seem to be a rather exceptional situation, and it confirms field observations that report sharp mindedness in the population of ‘‘high-functioning’’⁶ autistic women. Two thirds of these women graduated university, but only half of them were professionally active (work or studies). Despite demonstrably high cognitive abilities and education levels, participants in our study experienced employment difficulties. This discrepancy illustrates the well-documented employment issues endured by autistic persons.

More than half of the participants (57.8%) were involved in a relationship and more than half (55.6%) had been sexually active over the last six months (mostly the same individuals, 77% of them being in a relationship while the remainder was not). Of important note, more than half of our sample had only experienced heterosexual sex over their lifetime, despite the extremely high percentage of declared queer orientations (71.6%). This gap between aspiration and experience might be partly explained by social factors that are non-specific to autistic women (i.e., cultural stereotypes, lower proportion of potential partners…), but could also be linked to autistic characteristics, such as camouflaging and social imitation strategies that have been previously described in autistic women (Lai et al., 2017). Another partial and cumulative explanation could be that since a large majority (60.1%) of our participants were actually bisexual, we may postulate that they simply settled with the most available type of romantic relationship. Of note, our results are very consistent with literature describing that sexual orientation in autistic women is very diverse (Greenberg et al., 2018; Pecora et al., 2019, 2020).

Interrogated about their health, while only 17% of participants reported poor health status, more than 75% reported at least one psychiatric comorbidity, as if mental health disorders comorbidities were not health issues. Depression and anxiety were by far the most prevalent comorbidities (62.2 and 56.4% respectively). PTSD came in third place (26.7%). These results should however be interpreted cautiously because of the absence of structured clinical interviews for psychiatric diagnosis in this study. Considering that all these findings are in agreement with reports and studies on the topic of autism in adult women, it is reasonable to see our sample as reliably representative (Lai and Baron-Cohen, 2015).

In order to confirm autism diagnosis reported by participants, we collected their RAADS scores. We only included in the analyses participants who obtained RAADS score > 14 (merely two participants were excluded solely on that criterion). Mean RAADS in our sample was equal to 34.9 ± 5.2 and median was equal to 36, well above the median = 32 reported in the validation study of that scale (Ritvo et al., 2011). Moreover, the distribution of RAADS scores of self-diagnosed participants reflected that of the clinically diagnosed, confirming that including this sub-group in the analyses was methodologically acceptable.

We assessed victimization status twice: through an open question (Question#13) and through a specific questionnaire (SES-SVF). Both methods yielded extremely high figures: open question resulted in 68.9% victimization, compared to 88.4% with SES-SFV questions. Those results are very high but nonetheless extremely consistent with the findings that we have reviewed earlier: a rate of almost 30% victimization in non-autistic women and a 2 to 3-fold increased victimization of autistic women (Roberts et al., 2015; Bargiela et al., 2016; Weiss and Fardella, 2018; Pecora et al., 2019, 2020; Joyal et al., 2021; Warrier and Baron-Cohen, 2021). Moreover, results indicate that sexual aggressions started while most of the victims were underage or barely able to legally consent: out of 199 victims, more than two thirds (68%, n = 135 victims) where aged 18 or below when the first aggression took place; more than half (56%, n = 112) being at or below age of consent (15 in France).

Regarding ascertainment procedure, we used two methods of case identification. As expected, the SES-SVF questions revealed a significantly higher level of victimization than the open question. This confirms that it is necessary to ask questions about specific situations, rather than using open questions, in order to identify all victims, since some victims may need many years of processing memories about the assault before realizing that the non-consensual sex that happened to them qualifies as a legal offense/crime. One participant wrote this “It happened many times over my lifetime (as child, as a teen, as an adult). It is only 5 years ago, at the age of 50, that I understood that I had been victim of sexual violence.” Another participant wrote “This event remained out of my consciousness’ access for many years.” We indeed found several instances of such confusion in our sample: 34 out of 39 participants who answered “I don’t know” to the open question were nevertheless identified as victims through SES. Furthermore, 5 out 22 participants who responded “non-victim” to the open question were actually victims according to SES. Last, Chi-squared test results indicated that participants were consistent: all but two of those who declared being victim in the open question declared the same thing in SES-SFV questions.

Sexual revictimization among our participants was very important: 84.9% of victims were revictimized (n = 169 out of 199). In other words, 75.1% of all participants reported several aggressions. By contrast, only 13.3% (n = 30) were assaulted just once over their lifetime – with the assault consisting of unwanted sexual touching for most of them (n = 22).

We used four different statistical methods to assess whether being victim of one assault increased the risk of being victim of another one, and all yielded significant results, reaching a clear-cut conclusion that, yes, primo-victimization is a major gateway to further victimization. This is actually also the case for non-autistic women: half of those who were molested as children will suffer from new occurrences of sexual violence later in life (Walker et al., 2019).

Being young made things worse in our sample: young age was a significant factor of revictimization later in life. One possible contributor to this mechanism is that children who are molested often exhibit disturbed behaviors. As Mandell et al. (2005) found out, molested autistic children were ten times more likely than the others to act out in a sexual way. Such reaction is particularly maladaptive, since hyper-sexualized behavior by children facilitates exploitation and abuse by offenders. To make things worse, it happens that autistic girls are entering puberty significantly earlier than non-autistic girls, and earlier than autistic boys as well (Corbett et al., 2020). This precocious puberty is in itself enough to put those girls even more at risk as it was demonstrated in a study by Skoog and Özdemir (2016): “early maturing girls are sexually harassed as a result of natural and normative sexual development, which happens earlier than for most of their peers.” In any case, natural early maturing and maladaptive oversexualization are extremely easy to spot, therefore attracting predators.

As mentioned in the introduction to this text, sexual violence has a very negative impact on health. In our sample, participants who had endured sexual violence exhibited significantly more comorbidities than participants who hadn’t. Further testing suggested that, among all comorbidities, it is specifically PTSD that is increased after the aggression, in accordance with a recent meta-analysis (Hailes et al., 2019). Given the small number of non-victims in our sample, results were not clear cut: test of equal proportions was only significant without post hoc correction and the p-value of logistic regression did not reach significance despite high odds-ratio (3.08) and a confidence interval starting above 1. This is obviously a strong limitation for between-groups comparison. Notwithstanding, in accordance with Roberts et al. (2015) and Kildahl et al. (2019), we postulate that sexual assault actually increases the risk of PTSD in autistic women, as it does in the general population. Within the victim group, however, statistics results are unambiguous. PTSD rates and RAADS scores correlate strongly, as it was described by Roberts et al. (2015), though it is not possible to determine causality. Actually, if there is no doubt that mental health in general and sexual violence victimization are strongly correlated, causality seems to be bi-directional (Machisa et al., 2017; Oram, 2019): victimization affects mental health very negatively on the one hand, and suffering from mental health problems dramatically increases the risk of victimization on the other hand. Such vicious cycle explains well the mechanism of revictimization. In our study though, being young at first aggression was a significant predictor of developing PTSD. In contrast, being victim of several aggressions doesn’t seem to increase the risk of PTSD. Therefore, it seems that for the specific case of autism and PTSD, it is really sexual aggression that induces PTSD in autistic women and not the other way around.

Besides PTSD, victims also reported consequences that occurred within the 6 months following the assault. Half of them suffered from sleep disorder (48.2%) and disgust for sex (46.7%) and third of them attempted self-harm (31.7%). It is specifically rape that induces such deleterious effects, as opposed to victims of attempted rape only or unwanted sexual touching only. Of course, this is to be interpreted with caution because there is a discrepancy in sample sizes: 44 victims did not endure rape while 155 did.

Other major vulnerability factors associated with autism and with mental health issues such as social isolation, social stigma and social rejection should be further studied in order to better describe the mechanisms of abuse (Little, 2009; Edelson, 2010).

In our study, only a third of the victims reported the assault (34.6%, n = 69). For the vast majority of those victims (n = 52, i.e., 75.4% of those who reported), such accusation had strictly no effect (no medico-legal action was taken). Even worse, 18 out of 52 were not even believed. A lot has been written about the devastating effects of denunciating sexual violence only to be ignored or even be considered a liar, adding insult to the injury. This is not at all specific to autism: an abundant literature indicates that victims of sexual violence, whatever their gender, age and social status, often experience similar skepticism and indifference to their suffering (Burt, 1980; Ahrens, 2006; Suarez and Gadalla, 2010; Kennedy and Prock, 2018; Bhuptani and Messman, 2021). Ahrens writes that “Speaking out about the assault may therefore have detrimental consequences for rape survivors as they are subjected to further trauma at the hands of the very people they turn to for help” (Ahrens, 2006). This overpowering phenomenon of disbelief has been shown to be the main reason why most victims do not talk at all about it, internalizing pain and shame instead (Kennedy and Prock, 2018). It also explains why major and coordinated social movements such as #Metoo appear to many victims as the only way toward regaining agency and dignity. Furthermore, in our study, among the very few participants (n = 17) whose testimony was acted upon, only 13 received care and only 12 filed a complaint. How many of those 12 complaints resulted in the perpetrators being actually condemned is here a matter of imagination as such a question was not part of the questionnaire.

We have explored whether autistics traits, translated into metrics thanks to the RAADS scale, could be related to victimization. Results do not indicate clearly that autism is in itself a factor. Indeed, we found only one significant result: logistic regression indicated that the higher someone scores on the “sensory reactivity” sub-scores of the RAADS scale, the more elevated the risk of sexual victimization is. It is impossible to infer the direction of a putative causal effect, or even if it is just a correlation due to a confound yet to be determined. Very reactive individuals are indeed easily identified (they may wear sunglasses or noise canceling headphones) and that may help sexual predators to spot them. However, it could be the opposite: victims could display heightened reactivity due to trauma (a PTSD symptom). And last, it could be two effects of a common cause: for example, the trauma of being severely bullied at school could increase such sensitivity while at the same time making the victims more vulnerable to sexual violence.

However, a true effect of autism cannot be excluded of course. We did find a significant correlation between PTSD and total RAADS score after all, and PTSD was significantly correlated with victimization. But that could be interpreted both ways: (i) high autism traits lead to more victimization, leading to more PTSD, or (ii) PTSD might be the confound mentioned above, with PTSD being at the same time a consequence of sexual violence and the cause of heightened sensitivity in adulthood. In any case, it must be reminded here that the very small number of non-victims, the statistical negative results of our between group comparison are not very reliable. The absence of proof is not proof of absence and further research is needed here.

At that point, it seems important to ponder about the fact that several of the publications that we have cited above mention sexual education as the primary prevention method against sexual violence. Brown-Lavoie et al. (2014) for example, assert that sex education mediates the rates of abuse⁷. However, in the case our sample, such a solution would not be deontologically applicable since half of the victims were below the age of consent, and therefore out of the scope of applied sex education. Even for victims who were aged between 15 and 18, this is not applicable: expecting minors with a disability to protect themselves thanks to education can be equated to another form of victim-blaming. Suggesting to a victim that she should have learned how to better state her personal limits is not fundamentally different than telling her that her skirt was too short. Be it in a more benevolent and well-intended way, it is as hurtful and unfair.

This is not to say that there is no need to be informed. Quite the opposite, sex education is very important for everyone in order to foster safe relationships. But, according to Sala et al. (2019) sex education programs teaching how to stay safe from abuse are poorly validated. More importantly, considering that it is the victim’s duty to prevent the sex crime is without doubt a remnant of rape culture⁸. Regrettably, the opposite proposition is not as often suggested. Educating potential offenders instead of educating potential victims would be more ethical and might prove a more efficient method, although it would be of limited efficiency since offenders are usually not enamored youngsters unable to control their sexual urges, thereby becoming unintentional sex offenders. That is simply another rape myth (Payne et al., 1999). On the contrary, most offenders are smart predators, who are very aware of what they are doing (Burt, 1980; Hanson and Morton-Bourgon, 2005; Methot-Jones et al., 2019; Cardona et al., 2020; Russell and King, 2020). As Russell and King (2020) describe it: “Psychopathy and narcissism are known predictors of sexual violence.” Moreover, the callousness and cynicism of perpetrators is actively supported by a systemic rape culture, that is a “sociocultural context that sexually objectifies the female body and equates a woman’s worth with her body’s appearance and sexual function,” an internalized mechanic called objectification theory (Szymanski et al., 2011). As Methot-Jones et al. (2019) write, “individuals high in psychopathic traits see women as sub-human, this dehumanizing appraisal may be facilitating attitudes and behaviors that are consistent with the idea that women are less than human and deserve to be treated as such.” Objectification theory is particularly well illustrated by the systematic use of rape as a weapon of war in all ages and cultures (Bourke, 2014).

It seems intuitively right that, since autistic girls may misunderstand the complex games of flirting and dating, teaching them on how to behave in romantic relationships would help them achieve a satisfying love life (McMahon et al., 2021). However, this can only be successful in equal, fair and respectful relationships, not in situations where they are preyed upon. As Gotby et al. (2018) wrote in their study: “our findings that there were no specific effect of ASD or ADHD symptoms when controlling for general symptom load, fail to lend support to the hypotheses that social deficits, communication difficulties or impulsivity […] mediates the association between ASD/ADHD and coercive sexual victimization. This suggests that rather than focusing on a specific symptom-cluster or diagnosis, it is the general [neurodevelopmental disorder] phenotype that predicts increased risk of victimization” (Gotby et al., 2018). This is in agreement with research showing that women with ADHD are also at very high risk of being victimized (Snyder, 2015). It means that it is not autism in itself, nor the social communication difficulties associated with autism, that increase vulnerability to sexual violence but the fact of being perceptibly different. Or, alternatively, being different increases the risk of bullying and rejection, and in that case, it would be ostracization and stigma that would be the actual factors increasing victimization. Of course, as stated by Hartmann et al. (2019) sex education may still prove to be marginally protective: “While individuals with ASD are certainly not at fault or responsible for experiences of sexual abuse or victimization, misinformation about healthy sexuality (e.g., personal space, communication of consent, appropriate and inappropriate sexual behaviors) may in some cases contribute to adverse sexual encounters” (Hartmann et al., 2019).

Beside those considerations, another major criticism against the “preventing through sex educating” method is that victims are very often in a state of stupor and dissociation during the assault. They are therefore unable to react, sometimes even unable to comprehend, and no education can prevent that. Let’s remember here that the overwhelming majority of sex crimes are committed by persons who are closely related to the victims, a parent, a teacher, a boyfriend (WHO, 2021; Sardinha et al., 2022). In some cases, the state of psychological shock that accompanies such assault by a loved one can be so massively devastating that the victim is stunned into derealization and cannot understand, or even remember, what is happening. As one of our participants wrote: “There was no use of force, no threats, nothing. It was in the waiting line to the cafeteria. I was frozen.”

The SES-SFV responses provided yet another counterargument to the model of sex education as a prevention tool in that it collects information about the strategies used by offenders. Perpetrators primarily used two strategies to violate and abuse: manipulation/mental ascendency/harassment (“a”) and using surprise to take advantage (“c”). Strategy “c” was the first choice for imposing unwanted sexual contact, while strategy “a” was preferably used for rape attempt and rape. This might prove crucial for understanding why and how sexual violence happens so often in this population. As we have seen, one out of three non-autistic women is a victim of sexual violence. How could autistic women who, by very definition, are ingenuous and cannot guess hidden motives, defend themselves better than non-autistic persons against deviousness, treachery and deception? As Roberts et al. (2015) accurately write: “Deficits in emotional and social cognition, specifically, inability to identify sexually inappropriate behavior […] and inability to identify one’s own discomfort at inappropriate behavior […] increase risk of victimization and characterize persons with autistic traits.”. One participant wrote “It was because I believed that he would love me. I was raped and I never had any other experience since then. I was 19, I am now 36.” Another participant stated that she had “sexual relationships before the age of 15 with an adult over 30 who clearly took advantage of the situation.”

Interrogated about prevention methods, one victim out of ten answered that nothing could have protected them. One out of four considered that their autistic traits made them easy to spot by sexual predators. Four out of ten thought that knowledge about risk and self-affirmation strategies could have prevented the assault. However, as we asserted, the high proportion of underage victims renders such a method irrelevant since it would mean putting the responsibility of avoiding victimization on the shoulders of the minor victims.

In addition, it is very important to remember that, contrary to widespread myths regarding rape, offense is almost always perpetrated by a close person, exploiting ascendancy or hierarchical power over the victim. What could education do against situations such as the ones reported by our participants: “I was repeatedly raped by a relative during infancy and childhood”; “the first time, I was 6 years old”; “someone who had authority over me (father)”? On the contrary, as mentioned by one out of five participants, the most efficient protection, especially in the case of vulnerable individuals, is the presence of a vigilant caregiver. Educating families and professional about the risk of sexual victimization of girls on the autism spectrum should therefore be a priority.

Our research indicates that almost 9 autistic women out of 10 are sexually victimized, at huge costs for their mental and physical health. Of course, one major limitation of our study is that it is based on voluntary participation and that represents a bias since concerned individuals, i.e., victims, may be more willing to contribute to such research than non-victims. However, this bias might not impact results a lot, since our findings are very consistent with literature about sexual violence against autistic women and against women in general. So, what could be done to prevent such generalized issue? The WHO study (WHO, 2021) shows unambiguously that sexual violence is systemic and that vulnerable individuals are preferably targeted by offenders. Therefore, we state that it would be a methodological and deontological mistake to consider that victimization in autistic women is mainly due to autism. On the contrary, we postulate that the main cause of sexual violence against autistic women is their womanhood, since perpetrators preferably target women and girls. Therefore, it is reasonable to consider that autism is not the cause of sexual victimization in autistic women but just a factor increasing their vulnerability.

Prevention is absolutely necessary, certainly not in the sole form of sex education but rather by promoting profound cultural changes such as recommended by the Center for Disease Control (CDC) and World Health Organization (WHO): both organizations indeed strongly affirm that the very root of sexual violence is gender inequality. As Jessica Leight formulates it (Leight, 2022), “the literature analyzing the effectiveness of strategies to prevent and reduce intimate partner violence […] is now encapsulated in the RESPECT framework and its implementation plan.” As a tentative conclusion, we report here the most recent recommendations for preventing sexual violence as published by the CDC and the WHO. From the CDC, the Division of Violence Prevention of the National Center for Injury Prevention and Control has published a technical package to prevent sexual violence, titled the STOP-SV (Basile et al., 2016), that advocates sexual violence prevention along five dimensions:

The WHO has published a framework, very consistent with the above model, titled the RESPECT framework (World Health Organization, 2019).

In her seminal article, “Cultural myths and supports for rape,” Martha R. Burt wrote that “the task of preventing rape is tantamount to revamping a significant proportion of our societal values” (Burt, 1980). The CDC and WHO sexual violence prevention programs, designed four decades after Burt wrote those words, indicate that such profound change is taking place. Vulnerable populations, such as autistic women, shall benefit from this evolution.