Skip to content
Search resources Open menu

Autism Sex Offense

Knowledge and Intent in Online Cases Involving Autistic Defendants

Published

Snap Shot

  • The question: When can autism evidence bear on the knowledge an online CSAM offense actually requires?
  • What the law requires: Federal offenses require knowledge of possession, of the explicit nature of the material, and of the minority of those depicted (X-Citement Video), not knowledge of illegality or appreciation of harm. Naivety about legality or harm goes to charging, plea, and sentencing; what files were present, what they depicted, the ages shown, and what the software was sharing go to elements.
  • What the research shows: Theory-of-mind and literal-thinking arguments are clinical hypotheses, and the theory-of-mind construct itself is contested. Concealment applied selectively to CSAM is more informative than uniform privacy habits; deletion is ambiguous.
  • What it does not show: No controlled study has examined autistic people charged with CSAM offenses, and the suggestion that autistic people have particular difficulty judging age from images has not been tested. No test can determine what a defendant knew at the time.
  • The law: Rule 704(b) bars opinion on the defendant’s own mental state but not testimony about a group (Diaz). Knowing distribution matters for peer-to-peer charges (Shaffer, Budziak). Military law permits mental-condition evidence on a state of mind that is an element of the offense; Arizona limits expert evidence on mens rea (Mott, Clark).
  • Issues for each side: Prosecutors should focus on the required knowledge and its digital proof while recognizing that knowledge of age and of file-sharing can be genuinely uncertain in some cases; defense counsel should tie autism evidence to specific elements and the record, not group-level claims about understanding harm.

When an autistic man is charged with receiving, possessing, or distributing child sexual abuse material (CSAM), the defense sometimes argues that he did not know what he had, did not understand that it was illegal, did not realize that file-sharing software was making it available to others, or did not appreciate that the images recorded the abuse of real children. Prosecutors respond that the digital record shows deliberate searching, selection, and concealment, and that the diagnosis does not change what the evidence shows he knew. Both positions can be supported or undercut by the facts of a particular case, and the research does not settle the question for any individual.

The issue reaches court at several points: in charging and plea decisions, in motions about the admissibility of expert testimony, at trial on the knowledge elements, and at sentencing, where an autistic defendant’s understanding of the harm may bear on culpability even when it is no defense to the charge. This paper separates the mental states the law actually requires from those it does not, summarizes the clinical arguments about autism and knowledge, describes what the research does and does not show about those arguments, and reviews the federal, military, and Arizona law.

CSAM records the sexual abuse of real children, and its circulation continues that harm. Autism and sexual interest in children can co-occur, and a diagnosis opens questions; it does not answer them. The paper uses “child sexual abuse material (CSAM)” as its default term; statutes and courts use “child pornography,” and the paper uses that term when discussing the law. Many autistic adults prefer identity-first language (“autistic person”), and the paper uses it alongside “person with autism.”

What the Statutes Require

The federal offenses of receiving, distributing, and possessing child pornography, and of knowingly accessing it with intent to view, require that the defendant act knowingly (18 U.S.C. §§ 2252, 2252A). The Supreme Court has held that the knowledge requirement extends both to the sexually explicit nature of the material and to the minority of the persons depicted (United States v. X-Citement Video, Inc., 1994). The government must therefore prove that the defendant knew he received or possessed the material, knew it was sexually explicit, and knew that it depicted minors. It need not prove that he knew the conduct was illegal, understood the statute, or appreciated the harm to the children depicted. Ignorance of the law is not a defense.

That distinction shapes how autism evidence can be used. Arguments that an autistic defendant did not understand that the material was illegal, did not grasp that viewing it harmed children, or did not see the moral significance of what he was doing do not address any element of the offense. They may be relevant to charging discretion, plea negotiation, and sentencing, but not to guilt. Arguments that he did not know particular files were on his device, did not know what they depicted, did not know that the persons depicted were minors, or did not know that his software was sharing files do address elements, and they are the arguments on which evidence about the individual and the digital record is most important.

Courts infer knowledge from circumstantial evidence such as search terms, file names, the handling and organization of files, repeated viewing, and the defendant’s statements. Files found only in temporary caches or in unallocated space, without evidence that the defendant knew of them or could access them, have not supported possession or sentencing counts in some cases (United States v. Kuchinski, 2006; United States v. Flyer, 2011). Those cases turn on the digital evidence and apply to any defendant. The Online Sex Offense resource discusses CSAM law and digital evidence in detail.

Age of the Persons Depicted

Under federal law, a “minor” for purposes of the child pornography statutes is any person under 18 (18 U.S.C. § 2256(1)). Arizona defines a minor for purposes of sexual exploitation as a person under 18 at the time the depiction was created (A.R.S. § 13-3551). The Manual for Courts-Martial uses the same under-18 definition for the Article 134 child pornography offense. The age of consent to sexual activity is a different question governed by different statutes. Federal law, for offenses within federal jurisdiction, and many states set it at 16; Arizona’s offense of sexual conduct with a minor applies to persons under 18 (A.R.S. § 13-1405), and the Uniform Code of Military Justice defines a “child” for sexual abuse offenses under Article 120b as a person under 16. A 16- or 17-year-old’s legal capacity to consent to sexual activity in some jurisdictions does not make a sexually explicit image of that person lawful to receive, possess, or distribute. Material depicting anyone under 18 is CSAM under all three bodies of law, and minors cannot consent to their sexual exploitation in such material.

What the defendant knew about age is part of the knowledge element (X-Citement Video). In Arizona, the trier of fact may infer that a participant is a minor if the depiction, through its title, text, or visual representation, depicts the participant as a minor (A.R.S. § 13-3556). A claim that the defendant believed the persons depicted were adults is most plausible when the persons were near 18 and the files were labeled or presented as adult material, and least plausible when the material depicts prepubescent children or was found through searches for child-specific terms. Some practitioner writing suggests that autistic people may have particular difficulty judging age from appearance. That suggestion has not been tested in published studies, and judging age from images is difficult for anyone when the person depicted is an adolescent. The “Estimating Age from Images” paper in the Online Sex Offense resource discusses that research.

Theory-of-Mind Arguments and Their Limits

The most common clinical argument about autism and CSAM draws on theory of mind, the capacity to infer the thoughts, feelings, and intentions of others. In chapters written for defense practitioners, Mesibov and Sreckovic argued that differences in theory of mind may make it hard for some autistic people to recognize distress in the faces of children in images, to infer that the children did not consent and were harmed, and to connect viewing with the exploitation that produced the material. They also argued that literal thinking may lead some autistic people to assume that material freely available online must be legal (in Dubin & Horowitz, 2017). Mogavero (2016) and Allely and Dubin (2018) likewise described ways in which autism symptoms might contribute to online offending.

These are clinical hypotheses, not tested findings. Allely and Dubin (2018) stated that no empirical studies had examined autism and online child pornography offending and called for research. No controlled study has measured whether autistic CSAM users understand the harm to children less well than non-autistic users, whether that understanding relates to their offending, or whether it changes after explicit instruction. The theory-of-mind construct itself is contested. Gernsbacher and Yergeau (2019) reviewed the research and argued that the claim that autistic people lack a theory of mind rests on tasks of doubtful validity, that many autistic people pass such tasks, and that group differences do not support the conclusion that autistic people as a class cannot understand others’ mental states. A defendant’s performance on theory-of-mind tasks after arrest, even if measured, would not establish what he understood when he viewed the material.

The theory-of-mind argument also has a logical limit. Its strongest form concerns appreciation of harm, which is not an element of the offense. It says little about whether a person knew that the images showed children engaged in sexual conduct, which is the knowledge the statutes require, and difficulty reading facial expressions does not by itself prevent a person from recognizing that a depicted person is a child. In a review of English appellate cases, Creaby-Attwood and Allely (2017) found that convictions were set aside where judges had failed to direct juries adequately about a defendant’s autism and its effect on thought and behavior, particularly in offenses that required proof that the conduct was sexually motivated. Those cases involved offenses whose elements included a sexual purpose. They illustrate the point that autism evidence has the most legal force when it bears on a mental state the offense actually requires.

Naivety About Legality

A related argument is that some autistic people, especially those with limited sex education and heavy reliance on the internet for information about sexuality, may not realize that material accessible on the open web or through common file-sharing programs is illegal. In interviews with UK practitioners, Davy and colleagues (2025) described pathways to offending by autistic people that included limited tailored sex and online-safety education and naivety online, while cautioning that autistic people are not more likely to offend and that the study could not establish causes. That research comes from practitioner perceptions, not from measurement of what offenders knew. Naivety is a hypothesis to examine in the individual case, not an assumption.

Even when credible, naivety about legality is not a defense to the federal, military, or Arizona offenses. Its relevance lies elsewhere: in assessing culpability, in shaping supervision conditions and treatment that teach explicit rules, and in evaluating whether a defendant’s account of his conduct is consistent with the record. The record often answers the question. Concealment applied selectively to CSAM suggests awareness that the material was wrong; uniform privacy habits are less informative.

Concealment and Awareness

Concealment is the most frequently cited counterpoint to claims of naivety. Steel (2016), a practitioner framework rather than an empirical test, written from the perspective of a digital forensic investigator, argued that autistic people and CSAM offenders share some behaviors, such as limited victim empathy, poor intuition about social norms, and repetitive collecting, but that no causal link between autism and CSAM offending has been shown (Behavioural Commonality section), and that concealment can indicate awareness of wrongdoing. His examples are encrypting CSAM but not other files, using private browsing only for that material, and giving it innocuous names while naming other files accurately (Digital Forensics section). The behaviors he lists bear on awareness but are not decisive alone.

The informative pattern is selective concealment: steps taken to hide CSAM that were not taken with other material. Uniform privacy habits are less informative, because some autistic adults keep their computer use private as a matter of routine, and embarrassment about pornography of any kind does not distinguish CSAM from adult material. Deletion is ambiguous. In a survey of people convicted of CSAM offenses, about three-quarters reported having deleted their entire collection at least once (Steel et al., 2021). Purging may reflect guilt or an attempt to stop, which suggests awareness that the material was wrong. Understanding that something is forbidden or shameful is not the same as appreciating that it records the abuse of a child, and an evaluation can address both.

Peer-to-Peer Distribution and Understanding Software

Distribution charges and enhancements often rest on peer-to-peer software that makes downloaded files available to other users by default. Courts have held that knowingly placing files in a shared folder, with awareness that others can download them, is distribution when another user does download them (United States v. Shaffer, 2007; United States v. Budziak, 2012), and the federal Sentencing Guidelines limit the two-level distribution enhancement to a defendant who “knowingly engaged in distribution” (U.S.S.G. § 2G2.2(b)(3)(F)). The question is what the defendant understood about the software.

Here autism evidence may bear directly on an element. Some autistic people have strong technical skills and understand file-sharing software well; others use programs without exploring how they work. Neither tendency can be assumed from a diagnosis. The relevant evidence includes whether the defendant changed default settings, moved files out of shared folders, limited upload speeds, discussed sharing in chats, or used the software for other material, along with his history with technology. An evaluator can describe the defendant’s cognitive profile and how he approaches technology, but the inference about what he knew must rest on the forensic evidence. The Psychosexual Risk Evaluations resource’s paper “Clients on the Autism Spectrum” discusses evaluation of autistic clients more generally.

Limits of the Evidence

There are no controlled studies of autistic people charged with CSAM offenses. The arguments about theory of mind, literal thinking, naivety about legality, and age judgment come from clinical experience, practitioner chapters, and reviews, and the practitioner study by Davy and colleagues records perceptions of professionals rather than measured knowledge. Research on concealment and deletion comes from investigators’ observations and from surveys of convicted offenders, not from autistic samples. The case-law analysis describes how appellate courts in England and Wales have treated autism, under statutes that differ from American law.

No test can determine what a particular defendant knew at the time of the offense. Tests of theory of mind, social cognition, or legal knowledge administered after arrest measure present performance under different conditions. Group findings about autism cannot establish whether an individual knew that files depicted minors or that his software was sharing them. What an evaluation can do is describe the person’s history, cognitive and social profile, sources of sexual knowledge, technological habits, and understanding of the material, and set those against the digital record, stating plainly where the evidence points in different directions.

Questions to Ask About This Research

When theory-of-mind arguments (Mesibov and Sreckovic, in Dubin & Horowitz, 2017) are offered to show that a defendant did not appreciate the harm to the children depicted, ask whether the claim has been tested, since Allely and Dubin (2018) found no empirical studies of autism and online child pornography offending, and whether a theory-of-mind task given after arrest can show what he understood at the time. When practitioner accounts of naivety online (Davy et al., 2025) are cited, ask what exactly was counted: the perceptions of professionals, not measured knowledge. When the finding that about three-quarters of people convicted of CSAM offenses had deleted their entire collection at least once (Steel et al., 2021) is offered as evidence of awareness, ask who was studied and whether it applies to this person, since the survey did not involve people identified as autistic and deletion alone is ambiguous. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.

Knowledge, Intent, and the Law

Federal and Military Standards

Expert testimony about autism and knowledge must satisfy Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993), Kumho Tire Co. v. Carmichael (1999), and Federal Rule of Evidence 702, which since its December 1, 2023 amendment requires the proponent to show by a preponderance of the evidence that the opinion rests on sufficient facts and reflects a reliable application of reliable methods. An opinion that a defendant did not understand what he possessed must rest on evidence about the individual and the digital record, not on generalizations about autism. Rule 704(b) bars an expert from stating an opinion on whether the defendant had the mental state that constitutes an element of the crime. The Supreme Court has held that the rule bars only opinions about the defendant himself, not testimony about what most people in a group know or intend (Diaz v. United States, 2024). An expert may therefore describe autism, the defendant’s characteristics, and the research, but the conclusion about whether he knew must be left to the trier of fact.

Under the Insanity Defense Reform Act, a mental disease or defect is not otherwise a defense (18 U.S.C. § 17). The Third Circuit held that the Act does not bar psychiatric evidence offered to negate the mental state required for an offense, while requiring that such evidence be tied to the specific mental state at issue (United States v. Pohlot, 1987); other circuits have addressed the question, the rule varies by circuit, and current authority in the relevant circuit should be verified. At sentencing, autism evidence is most often presented under 18 U.S.C. § 3553(a), and United States v. Crosby (2024) shows that a sentence resting largely on expert evidence linking the offense to autism will be vacated if the court does not also address the other sentencing purposes.

In courts-martial, the Article 134 child pornography offense in Part IV, paragraph 95, of the Manual for Courts-Martial requires knowing and wrongful conduct, and the Manual’s explanation provides that an accused may not be convicted without awareness that the images were of minors, or what appeared to be minors, which may be inferred from circumstantial evidence such as file names, search terms, and the number of images. Rule for Courts-Martial 916(k)(2) provides that a mental condition not amounting to a lack of mental responsibility is not an affirmative defense; the Discussion to the rule states that evidence of such a condition may be admissible as to whether the accused entertained a state of mind necessary to be proven as an element of the offense (see also Ellis v. Jacob, 1988). Military Rule of Evidence 702 incorporates the 2023 federal amendment (Joint Service Committee on Military Justice, 2025). The Court of Appeals for the Armed Forces has held that viewing images on a computer without dominion or control is not possession (United States v. Navrestad, 2008), which, as in civilian court, makes the defendant’s relation to the files a matter of proof.

Arizona

Sexual exploitation of a minor under A.R.S. § 13-3553 requires knowing conduct, and when the minor is under 15 the offense is a dangerous crime against children sentenced under A.R.S. § 13-705, with consecutive sentences. Arizona’s definition of “exploitive exhibition” in § 13-3551 includes the purpose of sexual stimulation of the viewer, which builds a purpose element into some charges and may make evidence about the defendant’s understanding of the material more relevant than under federal law. The § 13-3556 inference of minority applies to the age element.

Under State v. Mott (1997) and Clark v. Arizona (2006), Arizona does not allow expert evidence of a mental disorder short of insanity to negate the mental-state element; the scope of evidence about behavioral tendencies is for counsel and the court. In Clark, the Supreme Court upheld that rule, distinguishing “observation evidence” about a defendant’s characteristic behavior, which Arizona permits on mens rea, from expert evidence of mental disease and capacity, which it channels to the insanity question. In practice, this limits expert testimony about autism on knowledge at an Arizona trial, although lay and expert testimony about what the defendant actually did and said, and how he characteristically behaves, may be admissible. Because the per-count structure and consecutive sentencing leave little room at sentencing, the defendant’s knowledge and understanding are often most consequential at charging and plea negotiation, and at sentencing under A.R.S. § 13-701(E) within the ranges the statutes allow.

Special Issues in Legal Practice

The principles described above apply with equal force to every party in a legal proceeding. The science operates the same way regardless of who is asking the questions, and the research summarized here does not by itself favor either side. What differs between the parties is not the science but the vantage point — the particular places where these issues tend to surface, and the particular errors each party is best positioned to guard against. The two summaries below identify those recurring issues. They describe how each side commonly uses this research and where its limits lie, for orientation only; in every case the controlling question is what the evidence in that case will, and will not, support, and the expert’s role is to inform the trier of fact rather than to advocate for either side.

Special Issues for Prosecutors

The research and case law support a prosecutor’s focus on the knowledge the statutes actually require and on the digital evidence that proves it: CSAM-specific search terms, file names and labels indicating age, selection and repeated viewing, changes to file-sharing settings, chats about sharing or about children, and concealment applied selectively to CSAM. Arguments that a defendant did not understand illegality or harm do not address an element, and the defense theory-of-mind literature consists of untested hypotheses. Under X-Citement Video, Diaz, and Rule 704(b), a defense expert may describe autism but may not tell the jury what the defendant knew; in Arizona, Mott and Clark limit such testimony further. Both sides may find it useful to ask whether the evidence of selection and concealment is thin or substantial and whether the defendant’s account is consistent with the record, because the answer bears on charging, plea terms, and supervision conditions written as explicit rules.

The research also counsels caution. Knowledge of age is an element, and in cases involving older adolescents depicted in material labeled as adult, age judgment can be genuinely uncertain for anyone. Knowledge of peer-to-peer sharing is an element of distribution and of the distribution enhancement, and some users of any background do not understand that file-sharing software shares files by default; what an individual understood must be shown from the record rather than assumed from a diagnosis. Uniform privacy habits and wholesale deletion are weak evidence of awareness that CSAM was wrong. A charging analysis that treats the diagnosis as irrelevant to these questions, like one that treats it as dispositive, can overlook evidence bearing on what the defendant knew.

Special Issues for Defense

The research and case law support defense presentation of autism evidence that addresses the elements and is tied to the individual and the record: what the client knew about the files on his device, what he understood about how his software worked, whether he believed the persons depicted were adults, and how his communication style affected his statements to investigators. Evidence of that kind can bear on knowledge and on the distribution enhancement, and evidence about the client’s understanding of harm and legality, while not a defense, is relevant to charging, plea negotiations, treatment planning, and sentencing. Admissibility questions turn on whether testimony fits the specific mental state the offense requires; testimony about autism in general does not address that question.

The research also counsels caution about overclaiming. Claims that autistic people as a group cannot understand that CSAM harms children rest on untested hypotheses and a contested account of theory of mind, and they do not address the knowledge the statutes require. Selective concealment, CSAM-specific searches, and focus on particular ages will weigh heavily against claims of naivety, and an opinion that does not address them is open to challenge on that point. In Arizona, Mott and Clark limit expert testimony on mental state at trial, and in federal court Rule 704(b) bars the ultimate opinion. Testimony that acknowledges the harm to the children depicted and the limits of the research is better supported than testimony that presents the diagnosis as an explanation for knowledge.

This paper is for educational purposes only and is not legal advice. The author is a psychologist, not an attorney; summaries of law are general and must be verified against current authority or with a licensed attorney.

References

Allely, C. S., & Dubin, L. (2018). The contributory role of autism symptomology in child pornography offending: Why there is an urgent need for empirical research in this area. Journal of Intellectual Disabilities and Offending Behaviour, 9(4), 129–152. https://doi.org/10.1108/JIDOB-06-2018-0008

Creaby-Attwood, A., & Allely, C. S. (2017). A psycho-legal perspective on sexual offending in individuals with autism spectrum disorder. International Journal of Law and Psychiatry, 55, 72–80. https://doi.org/10.1016/j.ijlp.2017.10.009

Davy, D., Brown, S., Vinter, L. P., Allely, C., Mann, N., Orofino, E., & Lundrigan, S. (2025). Understanding vulnerabilities and “pathways” in the commission of sexual offences by autistic individuals: Perspectives from UK-based practitioners. Psychology, Crime & Law. Advance online publication. https://doi.org/10.1080/1068316X.2025.2526604

Dubin, L. A., & Horowitz, E. (Eds.). (2017). Caught in the web of the criminal justice system: Autism, developmental disabilities, and sex offenses. Jessica Kingsley Publishers.

Gernsbacher, M. A., & Yergeau, M. (2019). Empirical failures of the claim that autistic people lack a theory of mind. Archives of Scientific Psychology, 7(1), 102–118. https://doi.org/10.1037/arc0000067

Joint Service Committee on Military Justice. (2025). Amendments to the Military Rules of Evidence since 1984 (G. E. Maggs, Ed.; 2025 ed.). https://jsc.defense.gov/Portals/99/MRE%20Amendments%202025.pdf

Mogavero, M. C. (2016). Autism, sexual offending, and the criminal justice system. Journal of Intellectual Disabilities and Offending Behaviour, 7(3), 116–126. https://doi.org/10.1108/JIDOB-02-2016-0004

Steel, C. (2016). The Asperger’s defence in digital child pornography investigations. Psychiatry, Psychology and Law, 23(3), 473–482. https://doi.org/10.1080/13218719.2015.1080150

Steel, C. M. S., Newman, E., O’Rourke, S., & Quayle, E. (2021). Collecting and viewing behaviors of child sexual exploitation material offenders. Child Abuse & Neglect, 118, Article 105133. https://doi.org/10.1016/j.chiabu.2021.105133

Legal Authorities

Clark v. Arizona, 548 U.S. 735 (2006).

Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993).

Diaz v. United States, 602 U.S. 526 (2024).

Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999).

United States v. X-Citement Video, Inc., 513 U.S. 64 (1994).

United States v. Budziak, 697 F.3d 1105 (9th Cir. 2012).

United States v. Crosby, 119 F.4th 1239 (10th Cir. 2024).

United States v. Flyer, 633 F.3d 911 (9th Cir. 2011).

United States v. Kuchinski, 469 F.3d 853 (9th Cir. 2006).

United States v. Pohlot, 827 F.2d 889 (3d Cir. 1987).

United States v. Shaffer, 472 F.3d 1219 (10th Cir. 2007).

Ellis v. Jacob, 26 M.J. 90 (C.M.A. 1988).

United States v. Navrestad, 66 M.J. 262 (C.A.A.F. 2008).

State v. Mott, 187 Ariz. 536, 931 P.2d 1046 (1997).

18 U.S.C. § 17.

18 U.S.C. § 2252.

18 U.S.C. § 2252A.

18 U.S.C. § 2256.

18 U.S.C. § 3553(a).

U.S. Sentencing Guidelines Manual § 2G2.2(b)(3)(F) (U.S. Sentencing Comm’n 2025).

10 U.S.C. § 920b (Article 120b, Uniform Code of Military Justice).

10 U.S.C. § 934 (Article 134, Uniform Code of Military Justice).

Manual for Courts-Martial, United States, pt. IV, ¶ 95 (2024 ed.).

R.C.M. 916(k)(2).

Ariz. Rev. Stat. Ann. § 13-701(E).

Ariz. Rev. Stat. Ann. § 13-705.

Ariz. Rev. Stat. Ann. § 13-1405.

Ariz. Rev. Stat. Ann. § 13-3551.

Ariz. Rev. Stat. Ann. § 13-3553.

Ariz. Rev. Stat. Ann. § 13-3556.

Fed. R. Evid. 702.

Fed. R. Evid. 704(b).

Ariz. R. Evid. 702.

Mil. R. Evid. 702.

Related topic papers: Distinguishing Autism-Related Conduct from Sexual Interest in Children · Autism, the Internet, and Collecting · Counterfeit Deviance: Origin, Use, and Limits · Police Interviews, Miranda, Compliance, and Suggestibility in Autism · Demeanor, Voir Dire, and Educating the Fact-Finder

Related topics from Arizona Forensic Services: Online Sex Offense · Psychosexual Risk Evaluations · Courts-Martial