Recent estimates from the Centers for Disease Control (CDC) and Prevention suggest that 1:150 children have autism or an Autism Spectrum Disorder (CDC, 2008). The rate at which autism is diagnosed has been steadily increasing in the past twenty years, with reported increases in autism ranging from three- to twenty-fold in that time (Waterhouse, 2008). Originally identified by Kanner (1943), autism has been characterized by challenges in communication, social ability, and behavior (American Psychiatric Association [APA], 2000), although there can be great variability in the extent to which difficulties are manifested. In addition to autism, there is a spectrum of disorders related to autism including Asperger's Disorder, Rett's Syndrome, and Childhood Disintegrative Disorder (APA, 2000). In total, these four disorders are referred to as Autism Spectrum Disorders (ASDs), and approximately 560,000 individuals in the U.S. between the ages of 0 to 21 years meet criteria for one of the ASDs (CDC, 2008).

In the past few decades, there has also been a steady increase in the number of firsthand accounts written by self-advocates who self-identify as individuals on the autism spectrum (e.g., Grandin & Scariano, 1986; Mukhopadhyay, 2008; Newport & Newport, 2002; Prince-Hughes, 2002). These accounts show not only that these challenges are often present because the social world is designed for typical individuals, with few adaptations for those who are not typical, but also how difficulties with sensory processing and overwhelming anxiety have a significant impact on their experience. As is illustrated in these accounts, autism and ASDs are heterogeneous in presentation. Additionally, although not considered as part of the autism diagnosis, cognitive ability in individuals with autism can also vary greatly; some individuals with autism have mental retardation while others score in the average, above average, or superior range of intelligence (Edelson, 2006). The heterogeneity in symptom presentation and severity, the heterogeneity in cognitive abilities, and the fact that even individuals with superior intelligence may not be able to decode and/or engage in typical social interaction can result in considerable variability in the ability of individuals with autism to interact and communicate successfully with others.

Most typical children have social, communication, and cognitive skills that allow them to navigate the complexities of the social world with success. Despite these skills, there are some children who will be victims of unwanted and harmful social interactions such as child sexual abuse. Current estimates suggest that 1:3 girls and 1:10 boys will be sexually abused by the time they are 18 years old (Tang, Freyd, & Wang, 2007). Given the nature of sexual abuse and the hesitancy to disclose its occurrence (Alaggia, 2004), the rates of child sexual abuse are likely underreported (Tang et al., 2007). Furthermore, when sexual abuse does occur, the sexual offender is usually someone who is known and trusted by the child (Cavanagh Johnson, 1999).

Although there are no empirical data assessing the frequency with which children with autism specifically are sexually abused, there is information about those with developmental disabilities in general. Mansell, Sobsey, and Moskal (1998) note that the rates of sexual abuse for children with developmental disabilities are nearly two times greater than for typical children. Moreover, Mansell et al. suggest that the effects of sexual abuse in developmentally disabled individuals may be exacerbated by social isolation and alienation.

When there are concerns that a typical child may have been sexually abused, there are protocols for how to evaluate whether or not abuse has occurred. Children's Advocacy Centers (CACs) or Child Abuse Assessment Centers (CAACs) that offer medical examinations of children and forensic interviews to make a determination regarding sexual abuse are often utilized (Cronch, Viljoen, & Hansen, 2006; Walsh, Jones, & Cross, 2003). These determinations are based on medical evidence obtained, which is rare in cases of sexual abuse (Finkel & Dejong, 1996; Myers, 1998); a previous history provided by the child and his/her family; and statements the child makes during the evaluation. In order for a valid determination to be made regarding whether or not sexual abuse has occurred, the child has to be able to participate effectively in the entire evaluation. Some children with autism may have difficulty with the models currently used to assess whether sexual abuse has occurred because of the use of lengthy, one-time interviews and the need for sustained reciprocity and verbal exchanges (Cronch et al., 2006; Cross, Jones, Walsh, Simone, & Kolko, 2007). For these reasons, it is important to develop protocols for use with individuals with autism that are sensitive to the way in which individuals with autism most easily interact and communicate with others.

The goal of the present paper is to highlight the reasons why children and adolescents with autism may be vulnerable to sexual abuse and to raise awareness about the lack of adequate protocols for evaluating children and adolescents with autism when there are concerns of sexual abuse. For the purposes of this paper, the discussion will be limited to those with autism or Asperger syndrome only, not the other ASDs. In the current paper, two main points will be argued: (a) that children with autism are at risk for sexual abuse and may have challenges being understood if they make a disclosure of abuse should it occur; and (b) that when children with autism are sexually abused, they may show this in ways that get missed or misattributed to autism rather than to possible sexual abuse. Furthermore, when there are concerns about possible sexual victimization, there are challenges in evaluating individuals with autism due to the unique ways in which individuals with autism communicate, making the use of traditional strategies for assessing whether or not a child has been sexually abused inadequate. This issue will be briefly addressed in the conclusion of the paper.

Characteristics of Autism and Risk for Sexual Abuse

Individuals with autism encounter behavioral, social, and communicative challenges (APA, 2000) largely because the social world is designed for typical individuals. Although not an issue for all individuals with autism, certain social-emotional and communication challenges, when present, may be interpreted by sexual offenders as vulnerabilities that they can exploit. The current section highlights the increased risk for sexual abuse that might be present for those children with autism who have the specific social-emotional and communication challenges discussed.

For example, interpreting the emotions of others may help a child identify safe from unsafe individuals. Although some self-advocates describe a keen ability to process and intuit others' emotions (Mukhopadhyay, 2008), other self-advocates (Prince-Hughes, 2002) describe their own significant challenges in this area. Research has also shown that emotional processing can be difficult for some individuals with autism. Begeer, Koot, Rieffe, Meerum Terwogt, and Stegge (2008) conducted a review of the literature related to the emotional competence in children with autism with regard to four areas required to be successful in social interactions: (a) expression of emotion; (b) perception of emotion; (c) responding to emotion; and (d) understanding emotion. Begeer et al. (2008) noted that laboratory studies of those with autism generally reported that those who have few symptoms of autism are able to express simple emotions and respond to others' emotions, whereas those who have many symptoms of autism are more likely to encounter difficulties in emotional processing. Begeer et al. also found that in natural settings, many individuals with autism may encounter challenges in identifying emotions and responding empathically to others.

It can be even more difficult for individuals with autism to understand the emotions of others when the emotions expressed are deceptive (as may be the case when interacting with a possible sexual offender). Dennis, Lockyer, and Lazenby (2000) found that high functioning children with autism were less able to identify facial expressions that depicted deceptive emotions and were less able to understand the reasons why someone would display a deceptive facial expression compared with age- and gender-matched control children. Offenders attempt to gain trust from potential victims and often do so by being deceptive. Therefore, they may display deceptive emotions that may not be recognized by some children with autism.

In addition to difficulty with emotional processing, children with autism may encounter communication challenges that may make them particularly desirable targets of sexual offenders because of the perception that they would be unable to disclose the abuse. Research indicates that up to 50% of children with autism are functionally nonverbal (APA, 2000). Although there are alternative and augmentative methods used by many children with autism to communicate effectively, the seeming inability of nonverbal children with autism to communicate may increase the likelihood that sexual offenders would target them for abuse.

Even verbal children with autism may have difficulty reporting abuse if they have certain communication difficulties. For example, Dahlgren and Dahlgren Sandberg (2008) examined referential communication in children with autism and ASDs. Referential communication requires a speaker to provide enough specific information to a listener so that the listener knows to what the speaker is referring. This skill is especially important in communicating information not already known by another party, as in the case of a sexual abuse disclosure. Dahlgren and Dahlgren Sandberg found that children with autism or ASDs had greater difficulty communicating relevant information about a referent and were less efficient referential communicators than typical children. Thus, some children with autism who attempt to disclose sexual abuse may not have the skills to effectively communicate what happened to them in a way that will be understood by others. Furthermore, Hale and Tager-Flusberg (2005) note that some children with autism have difficulties with the pragmatic use of language and in the ability to maintain social discourse with others. These difficulties are especially likely to be manifested in conversation, again increasing the likelihood that some children with autism may be unable to understand the nuances of reciprocal conversation needed to disclose sexual abuse should it occur.

Social-emotional and communication challenges are just part of the reason why some children with autism may be at risk for sexual abuse. Stevens (1997) studied the selection techniques predatory rapists used to target victims. He classified the selection characteristics into one of four broad categories: (a) "easy prey" (e.g., vulnerable victims such as being young and female); (b) victim attributes (e.g., sexual desirability); (c) situational characteristics (e.g., opportunity); and (d) circumstance or manipulation (e.g., the use of victim manipulation such as violence or intimidation prior to the sexual assault). Because children with autism may be seen as "easy prey," may be easily accessible to offenders, and may be easily manipulated or intimidated because of social challenges related to autism (APA, 2000), they may be seen as particularly desirable targets of sexual abuse by offenders.

Moreover, sexual offenders who target children often have cognitive distortions that allow them to justify their offending and not identify the offending as "wrong" or "harmful" to the child (Burn & Brown, 2006). The offenders' cognitive distortions serve to justify their offending by minimizing or rationalizing the offending behavior (Burn & Brown, 2006). In the adult sexual assault literature, it has been shown that one cognitive strategy employed by sexual offenders to "allow" them to offend is the "objectification" of their victims, viewing them as objects rather than people (Russell, 1998). Some children with autism may exhibit certain repetitive or stereotyped behaviors that seem unusual to others (APA, 2000). Therefore, a sexual offender may find it much easier to objectify a child who engages in these behaviors than to objectify a typical child.

According to Cavanagh Johnson (1999), there are two main types of child sexual offenders. The first is the offender who "grooms" the child prior to offending. Grooming behaviors have the function of introducing the child to pleasant forms of physical contact and of establishing a positive relationship with the child over time to mold the child into a potential victim. By grooming the child, the offender is able to test whether or not the child will resist or disclose the abuse early in the offending process. A child who resists grooming efforts is typically discarded by the offender as a potential victim because he/she is perceived to be a risk to disclose the abuse. In this instance, the tactile defensiveness experienced by some children with autism (Grandin & Scariano, 1986) might work in the children's favor; however, children with autism who do not speak may not be perceived by offenders as carrying the same risk of disclosure as a typical child, and therefore, offenders may choose not to engage in grooming behaviors. The second type of child sexual offender is the "opportunistic" offender, who takes advantage of opportunities to offend vulnerable children (Cavanagh Johnson, 1999). Both the social-emotional and communication challenges previously discussed place children with autism at increased risk of sexual abuse by opportunistic offenders and may make children with autism particularly desirable — or even "ideal" — targets for opportunistic offenders.

Children with autism may also be at greater risk of being sexually abused than typical children because of the increased contact with opportunistic offenders who are service providers. Goldman (1994) cites evidence that over 50% of offenders of individuals with developmental disabilities had contact with their victims through some type of disability services with which they were involved. The specific nature of the offenders' contact with their victims included serving as paid service providers, as foster care providers, and as transportation providers. Because children with autism often require specialized services such as those cited by Goldman (1994), they may come into frequent contact with potential abusers. Moreover, although there are not data specific to autism, those with developmental disabilities who live in institutional settings may be at even greater risk for sexual abuse than those who reside with their nuclear family (Goldman, 1994). This is likely due to increased contact with opportunistic offenders in the institutional milieu.

Finally, regardless of offender type, children with ASDs may be at increased risk of being sexually abused compared with typical children because of the desire to be accepted socially despite the social challenges they often face. If a sexual offender presents him/herself as a "friend," the child may see the relationship with the perpetrator as an opportunity to have the social relationship he/she desires. Just as is the case for typical children, a child with an ASD might become the victim of an offender who initiates sexually inappropriate behaviors in order to keep the "friendship." Similarly, due to a lack of proper sex education, which is often not provided to children with autism due to an erroneous belief that children with autism are asexual (Irvine, 2005), a child with an ASD may not recognize that the offender's behaviors are, in fact, inappropriate. This risk is noted by Newport and Newport (2002) who state, "the relative naiveté of autistic girls or their possible wish to trade sex for 'popularity' may initiate them far earlier [into sexual activity] but rarely in a healthy way" (p. 34).

Given the increased risk of sexual abuse that children with autism may face, it is important to identify when sexual abuse has occurred. However, due to the constellation of symptoms associated with autism, children with autism who are sexually abused may not be identified as abuse victims. The next section details why behavioral signs of sexual abuse in children with autism may be missed or misattributed to the child's autism.

Misattributed or Missed Behavioral Signs of Sexual Abuse in Children with Autism

Children with autism sometimes display self-stimulatory behaviors, self-injurious behaviors, and stereotypic and repetitive behaviors (APA, 2004; Cunningham & Schreibman, 2008). Should a child with autism be sexually abused, the child's attempts to cope with or make sense out of that abuse may lead to an increase in the intensity and frequency of these behaviors or to the development of new behaviors that were not previously present.

Research suggests that children with autism who are nonverbal exhibit more behavioral difficulties than those who have verbal communication abilities (Dominick, Davis, Lanihart, Tager-Flusberg, & Folstein (2007). This may relate to frustration caused by the inability of others to understand what the child is trying to communicate. For example, Dominick et al. (2007) found that there was a significant inverse relationship between the display of self-injurious behaviors and expressive verbal language ability in a sample of children with autism. For children with autism who wish to disclose their abuse, behavioral reactions to sexual abuse may develop if others cannot understand their communication about the abuse, but these behaviors may be misinterpreted by others as merely a manifestation of autism. Therefore, the fact that the child was, or continues to be, sexually abused may be missed.

Some have suggested that the presence of sexualized behaviors is indicative of sexual abuse. For example, Bow, Quinnell, Zaroff, and Assemany (2002) note that the presence of sexualized behaviors occurs more frequently in sexually abused children than non-sexually abused children. However, researchers have also found that sexualized behaviors can occur in response to physical abuse, not just sexual abuse (Merrick, Litrownik, Everson, & Cox, 2008). Additionally, the presence of sexualized behaviors does not necessarily mean that any abuse has occurred (Cavanagh Johnson, 1999). Cavanagh Johnson (2002) discusses a continuum of sexual behaviors that children can display, including typical sexual behaviors; sexually-reactive behaviors; excessive, but mutual, peer sexual behaviors; and sexually abusive behaviors. The first category on the continuum is developmentally normative, and the other three categories can develop in reaction to traumatic events in general or to over-stimulating environmental experiences, not just in reaction to abuse (Cavanagh Johnson, 2002).

Historically, those with developmental disabilities were not believed to have sexual feelings (Irvine, 2005). Nario-Redmond (in press) conducted a study that examined empirically the cultural stereotypes individuals have of individuals who are and are not disabled. Part of her investigation sought to identify which characteristics of individuals with disabilities would be offered spontaneously by the participants, some of whom had disabilities and some of whom did not. Consistent with the historical views of individuals with disabilities, Nario-Redmond found that the three most commonly offered stereotypes of both men and women with disabilities were that they were dependent, incompetent, and asexual.

Nario-Redmond (in press) notes that there are challenges present for individuals who differ from the norm not because of any biologically-based disabilities they may manifest but because the environments and policies which they encounter can "systematically exclude" them from full participation in the world. Because of this exclusion and the stereotype that individuals with disabilities are asexual, individuals with autism may not be given opportunities for appropriate displays of, or education related to, sexual behaviors. Thus, they may manifest sexually inappropriate behaviors that others may misattribute as indicative of sexual abuse. Furthermore, children who are sexually abused do not always display sexualized or concerning behaviors at all (Kendall-Tackett, Meyer Williams, & Finkelhor, 1993). Therefore, the presence or absence of sexualized behaviors cannot be used as a marker for whether or not a child has been sexually abused.

Unfortunately, there is no research on the behavioral manifestations of sexual abuse in children with autism. In fact, a PsycInfo search attempting to obtain literature on the sexual abuse of children with autism revealed no empirical articles on this topic. Sexuality, in general, has been rarely discussed in the scholarly literature on autism as well; only four references were found when doing a combined search for sexuality and autism (see Gabriels & Van Bourgondien, 2007; Koller, 2000; Rhodes, 2006; Stokes & Kaur, 2005). There has been slightly more attention paid to sexuality in the non-scholarly literature but not much.

Jerry and Mary Newport, a married couple with autism/Asperger syndrome, have written a book that provides information and practical advice on sexuality given their experiences (Newport & Newport, 2002). The Newports provide pragmatic information on developing social and sexual relationships, how to address the first sexual feelings, and how parents should talk about sexuality with their children with ASD. They also have a chapter on rape, molestation, and abuse. The Newports are evidence that individuals with ASDs are sexual and can and do encounter multiple kinds of sexual abuse.

The scholarly literature on sexuality in individuals with autism that does exist focuses mainly on the perceptions and concerns of parents with regard to sexual education. In one of the few studies on sexuality in autism, Ruble and Dalrymple (1993) analyzed 100 surveys of parents with children with autism from 9 to 38 years of age, assessing the parents' (usually mothers') views of their children's sexual awareness, education, and behaviors. The survey results revealed that the more verbal the child, the more the parents reported that the child had knowledge of body parts and functions, understood the difference between public and private behaviors, and had received some form of sex education. It is possible that these results were obtained because parents of children with greater verbal abilities had talked with their children more about sexuality than did parents of children with less developed verbal skills. Ruble and Dalrymple also found that the more verbal the child, the more the parents reported that the child displayed inappropriate sexual behaviors, with 66% of the parents of verbal children with autism observing at least some inappropriate sexual behaviors in their children. As is the case with those with developmental disabilities in general (Irvine, 2005), this may be due to the lack of opportunity for appropriate sexual behaviors, possibly because of the stereotype that individuals with autism are asexual (Nario-Redmond, in press).

In Ruble and Dalyrmple's (1993) study, parents of children with autism were concerned about their child being taken advantage of sexually, experiencing unwanted pregnancy and STDs, having sexual behaviors misunderstood, and questioning whether sexual relations were even relevant for individuals with autism. However, most parents did not have concerns related to typical sexual development in their child, again possibly due to a reflection of the societal view that individuals with autism are asexual.

Sexualized behaviors may appear at various stages of sexual development for typical children (Cavanagh Johnson, 1999) and may seem more pronounced in children with autism because the ages at which children with autism reach various developmental stages may be delayed compared to typical children. For example, although it is fairly common for preschool children to explore and stimulate their own bodies, sometimes in public (Cavanagh Johnson, 1999), children and adolescents with autism may also engage in these behaviors although at an older age. The presence of these behaviors may then be misinterpreted as signs of sexual abuse, especially if parents maintain the belief that children with autism are asexual (see Ruble & Dalrymple, 1993). Conversely, there may be times when sexualized behaviors do indicate sexual abuse, but parents and professionals may instead conclude that the behaviors are just part of a delayed progression of typical sexual development. It is, therefore, easy to note why it may be difficult to determine if a child with autism has been sexually abused on the basis of observed behaviors.

In addition to the difficulty in determining whether or not a child with autism has been sexually abused based solely on behavior, there is also the potential for behavioral signs of sexual abuse to be misattributed as signs of autism. There is evidence in the psychiatric literature that when individuals have a mental illness, their behavior may be interpreted in light of their disorder (Rosenhan, 1973). Rosenhan (1973) conducted a classic study in which he sent "pseudopatients" into psychiatric facilities complaining of hearing existential voices saying "empty," "hollow," or "thud." With the exception of masking the fact that the pseudopatients worked in the mental health field, all other personal information provided to the psychiatric facilities was accurate. All pseudopatients were deemed mentally ill (most diagnosed as having schizophrenia) and admitted to a psychiatric facility. However, once admitted, the pseudopatients no longer complained of hearing voices and, with the exception of note-taking to document the results of the study, did not act in any way different from how they typically acted.

Among other interesting results that Rosenhan (1973) noted was that the note-taking was assumed to be a manifestation of their schizophrenia. Rather than question a behavior such as note-taking in a psychiatric facility, the mental health professionals merely saw it as a symptom of the patient's disorder. Even the pseudopatients' personal histories were interpreted in a way that seemed to support their diagnoses. According to Rosenhan (1973), "one tacit characteristic of psychiatric diagnosis is that it locates the sources of aberration within the individual and only rarely within the complex of stimuli that surrounds [the person]. Consequently, behaviors that are stimulated by the environment are commonly misattributed to the patient's disorder" (p. 253).

In the field of autism, there have been many historical examples where environmental conditions led to assumptions about the abilities of individuals with autism. Perhaps the best example of this is the oft-reported belief that the majority of individuals with autism are mentally retarded despite a lack of evidence for these claims (Edelson, 2006). The assumption of mental retardation was often made when communication, behavioral, or attention challenges prevented examiners from obtaining valid estimates of intelligence. Researchers would attribute low test scores to the intellectual abilities of the children with autism rather than to the fact that the measures used to assess intelligence were not appropriate for the children or that the examiners did not account for the symptoms of autism when attempting to determine intelligence (Edelson, 2006). Similarly, it is quite probable that a child with autism who has been sexually abused and subsequently displays behaviors deemed concerning by others may have those behaviors misattributed to his/her autism.

There have been a number of individuals with autism who have been able to share their frustrations when their behaviors have been misattributed, misunderstood, or pathologized. Temple Grandin, an adult with autism, has written a number of books about what it is like to have autism. In her book, Emergence labeled autistic, she describes many challenging situations when she was a child and teased because of her autism. One example she recalls was a time when a girl in her junior high school called her a "retard." Grandin relates how she became so angry that she hurled her history book at the girl and, in the process, hit the girl in the eye. Her principal, Mr. Harlow, expelled her from school following this incident. As she relays the situation, Temple notes, "anger and frustration surged through me and I trembled, sick to my stomach. Mr. Harlow hadn't even asked to hear my side of it. He just assumed that since I was 'different' I was entirely to blame" (Grandin & Scariano, 1986, p. 64). Just as Mr. Harlow attributed the negative interaction with the school girl to Temple's autism, there are many times when researchers, mental health professionals, teachers, and perhaps even parents attribute behaviors seen in the child to his/her autism rather than to a myriad of other factors that may account for these behaviors.

More recently, there have been self-advocates who have discussed the challenges they have faced when their behaviors have been misunderstood or pathologized by typical individuals. Tito Rajarshi Mukhopadhyay has written a number of books detailing his experiences as someone with autism who is required to interact in the typical world. In his most recent book, How can I talk if my lips don't move? Inside my autistic mind (Mukhopadhyay, 2008), Tito describes his frustration at how his mother viewed autism as something to be cured, rather than as something to be accepted. Tito states, "How could she [my mother] participate in a system that classified me as sick? Did Mother really think I was less of a person?" (p. 176). Often, typical individuals will characterize the behaviors of individuals with autism as "pathological" or "sick." Not only is this damaging to the individual, it may result in the misattribution of concerning behaviors to the "sickness" of autism rather than to an environmental cause for the behaviors, such as sexual abuse.

Finally, some individuals with autism may feel they need to adapt behaviors that are comforting to "fit in" with the typical world. For example, in her edited book, Aquamarine blue: Personal stories of college students with autism, Dawn Prince-Hughes shares the stories of many adults with autism and ASDs (Prince-Hughes, 2002). One person whose story she shares is Darius who describes his frustration with the fact that he must "disguise" self-stimulating behaviors because of the pressures by those in the typical world who do not understand the "reassuring feeling" such behaviors can provide. Darius describes an episode at school when he was 11, and he was bouncing with his back to the wall. His teacher told him not to bounce, and Darius states, "I remember not understanding why I could not bounce, as it was such a reassuring feeling. I had already decided to stop publicly engaging in some of the more clearly autistic 'stimming' behaviors and only did them in my room. This was the last one to go. I had by that time learned to 'disguise' some of the 'stimming' and repetitive behavior" (Prince-Hughes, 2008; p.13). Darius' story reflects the conflict between engaging in behaviors that help allay anxiety (as was the case with his "bouncing") and hiding those behaviors because others did not understand them. It is possible and, in fact, likely that a child with autism who is sexually abused may turn to behaviors that provide comfort such as "stimming" or self-soothing behaviors, the reasons for which may be misattributed to an increase in the severity of the child's autism rather than to the abuse. Moreover, the pressures to "cure" these behaviors may increase which may obfuscate the search for any environmental reasons for these behaviors.

There may also be misattributions with regard to the origins of offending behaviors sometimes seen by individuals on the autism spectrum. Most victims of child sexual abuse do not become sexual offenders; however, some offenders do have a history of child sexual abuse victimization (Burn & Brown, 2006; Coxe & Holmes, 2001; Glasser, Kolvin, Campbell, Glasser, Leitch, & Farrelly, 2001). This is especially true for male victims of child sexual abuse (Glasser et al., 2001). Coxe and Holmes (2001) found that male child sexual offenders were nearly twice as likely to offend a child under the age of 10 years old if they had a history of being sexually abused as a child compared to offenders who did not have a history of childhood victimization. Therefore, it is essential to identify those children who have been sexually abused so that they both can obtain treatment to help them heal from the abuse and not become offenders of young children themselves.

There is recent literature to suggest that some adults with high functioning autism or Asperger's Disorder engage in offending behavior, although the frequency with which this occurs has been the subject of debate (Allen, Evans, Hider, Hawkins, Peckett, & Morgan, 2008). Allen et al. studied a small group of adults with Asperger's Disorder and obtained data from informants about the sample's offending behaviors. Among other findings, the informants offered predisposing factors that they believed led to the offending, most of which were attributed to the Asperger's Disorder. The predisposing factors included such variables as social naiveté, lack of awareness of outcome, and misinterpretation of rules.

What is of interest in the Allen et al. (2008) study is the failure to note the possible role of sexual victimization as a predisposing factor for offending. Consistent with Rosenhan's (1973) assertion, the informants' attributions for the offending behaviors were congruent with the symptoms associated with Asperger's Disorder, and there was little attempt to consider other explanations for why the adults, all of whom were male, may have engaged in offending behaviors. Clearly, the majority of individuals with Asperger's Disorder do not engage in offending behavior. Therefore, it is logical to suspect that, at least some of the time, variables unrelated to Asperger's Disorder symptomatology account for offending when it occurs. Because of the history of child sexual abuse victimization in some adult offenders (Burn & Brown, 2006; Coxe & Holmes, 2001), it is reasonable to assume that this link might exist in some individuals with autism and Asperger's Disorder who offend. It is, therefore, vital that sexual abuse of children with Asperger's Disorder and autism be identified so that appropriate intervention can help children heal without developing offending behaviors.

Conclusions and Implications

The world of autism research and education has devoted little attention to sexuality in general and the possibility of sexual abuse in particular. The lack of research does not mean, however, that the issue does not exist. Due to the particular manifestations of ASDs, children on the spectrum are likely to be at greater risk for sexual abuse than other children. Because of this risk, it is incumbent upon researchers to identify strategies to prevent sexual abuse, to develop protocols to assess accurately if abuse has occurred, to educate people with ASDs about sexual health and abuse, to ensure children with autism are taught to use augmentative and alternative means of communication, and to develop methods to help children heal so that they do not develop offending behaviors themselves in response to abuse.

When sexual abuse is suspected in children with autism, there must be valid protocols established to assess whether or not it has occurred. There are many models used to assess typical children when there are concerns of sexual abuse. Cronch, Filjoen, and Hansen (2006) reviewed different techniques frequently used in forensic interviews to determine whether or not sexual abuse has occurred. Some of these include the use of cognitive interviews, anatomically detailed dolls, and structured interviews.

Unfortunately, these techniques may not work well for children with autism and ASDs. First, sexual abuse evaluations are often one-time experiences in which a child meets with a previously unknown person. Many children with autism prefer consistent routines and may have difficulty with new environments (Richler, Bishop, Kleinke, & Lord, 2007) and/or unfamiliar people. Temple Grandin describes difficulty with changes in routines and the anxiety new situations or people would cause her as a child with autism (Grandin & Scariano, 1986). Thus, the one-time nature of the evaluation may be problematic. Second, the standards of practice in forensic interviewing are based on the utilization of structured protocols with an emphasis on open-ended questions designed to elicit free narratives (Cronch et al., 2006). These protocols require a child to have sufficient verbal skills and the ability to engage in referential communication and conversational discourse which some children with autism may not be able to do (Dahlgren & Dahlgren Sandberg, 2008; Hale & Tager-Flusberg, 2005). Tito Rajarshi Mukhopadhyay discusses the difficulty he has with facial perception and recognition. In fact, Tito describes how he creates stories of the sensory experiences he has when talking with others, sensory experiences often relating to vibrant colors because of the fact that he experiences synesthesia. Tito notes that "without those stories, recognizing and recalling a person or a situation is very difficult" (Mukhopadhyay, 2008; p. 109). The currently utilized protocols for sexual abuse assessment would not be reliable or valid for a person like Tito, given his way of recalling the "stories" of the people in his world. Therefore, the creation of new protocols that reliably enable individuals with autism to disclose sexual abuse is imperative. Firsthand accounts of people with ASDs, such as Temple Grandin, the Newports, Tito Rajarshi Mukhopadhyay, and the adults whose stories appear in Dawn Prince-Hughes' book, some of whom have been sexually abused, must inform how these protocols are structured.

The ineffectiveness of current protocols may also be due in part to the fact that children with autism, like typical children, have short attention spans and have not encountered situations in the real world that mirror a forensic interview. An extended forensic evaluation model has been suggested by the National Children's Advocacy Center (Cronch et al., 2006) in which multiple interviews are used to address those with shorter attention spans and/or those who need to establish a rapport with their communication partners before meaningful and personal communication will occur. Although the impetus of the extended forensic evaluation model was the need to find a process that worked better for young children than those currently employed, it could be adapted for use with children with autism. Certainly, modifications to address the various challenges discussed in this paper would be necessary. Difficulty identifying faces and places (Mukhopadhyay, 2008), the tendency to shut down all communication if the interviewer is condescending (Mukhopadhyay, 2008; Prince-Hughes, 2002), and a significant lack of body awareness (Grandin & Scariano, 1986; Newport & Newport, 2002; Prince-Hughes, 2002), are additional challenges individuals on the autism spectrum often face, and these also must be recognized and incorporated in the establishment of new protocols. Moreover, firsthand accounts of, and feedback from, self-advocates who identify as being on the spectrum will help design the protocols most likely to enable reliable disclosure of sexual abuse by both speaking and non-speaking children with ASDs.

Without an acknowledgement that sexual abuse is a real risk for children with autism, there cannot be adequate measures taken to ensure the safety of these children, to help those who have been sexually abused heal from the abuse, and to prevent possible future victimization of other children. Angie, an adult with autism who scores in the "very superior" range on intelligence tests, recounts the aftereffects of psychological, physical, and sexual abuse as a child (Prince-Hughes, 2002). She states, "I am frightened to be put into a situation where I have to explain anything to anyone…. Most of the time I just keep it to myself because I just make too many enemies when I say something…. I am not really interested in anything anymore (although I once had the remarkable ability to be interested in anything). In fact, I truly wish I had mental retardation because most people get what the hell that is and my life probably would have turned out better" (Prince-Hughes, 2002; pp. 77-78). Angie is able to articulate both the frustration at not being understood because of her autism and the despair that abuse as a child has caused or at least contributed to. For individuals with autism like Angie who can articulate the effects of abuse and for those with autism who cannot, it is imperative that we as a community of researchers, educators, parents, and self-advocates find a way to increase the awareness of the risk of sexual abuse for those with autism and ASDs, to allow for a diversity of communication styles and voices to "hear" when abuse has happened, and most importantly, to prevent abuse from occurring in the first place.