Skip to content
Search resources Open menu

Autism Sex Offense

Police Interviews, Miranda, Compliance, and Suggestibility in Autism

Published

Snap Shot

  • The question: Does autism make a suspect’s rights waiver or statement less reliable?
  • What the research shows: Controlled studies of intellectually able autistic adults have not found greater interrogative suggestibility, and one meta-analysis found lower suggestibility. Two of three controlled studies found higher compliance, and one did not. In one study, truthful autistic interviewees were rated as more deceptive and less credible, and in a mock-suspect study innocent autistic participants offered fewer details supporting their innocence.
  • What it does not show: No study has measured false-confession rates among autistic suspects in real cases, and no empirical study of Miranda comprehension in autistic adults was identified; the one located comparison of autistic and non-autistic adults’ understanding of a rights advisement, concerning the Canadian right to silence, found no group difference, with poor understanding in both groups. Autism is not established as an independent risk factor for false confession, and no test can show whether a particular statement is true or a particular warning was understood.
  • The law: Autism alone does not make a statement involuntary without coercive police conduct (Connelly), but it can bear on whether a waiver was knowing and intelligent (Moran v. Burbine) and on the weight of a statement (Crane). In courts-martial, Griffin and Bresnahan address expert opinions and expert assistance. Arizona permits general “cold” testimony that satisfies Rules 702 and 403 (Salazar-Mercado).
  • Issues for each side: Prosecutors should ensure plain-language warnings, understanding checked in the suspect’s own words, and full recording; defense counsel should build on the client’s individual profile and the recorded interview rather than group claims of suggestibility.

Many investigations of online sexual offenses include an interview with the suspect, often during or soon after the search of a home. That interview may produce the most important evidence in the case. A suspect may describe how he found the material, whether he knew what it was, whether he shared it, and whether he understood that it was illegal. When the suspect is autistic, both sides have reason to ask how the interview was conducted and how well he understood it. Did he understand the rights advisement? Did he agree with suggestions because he was anxious to end the encounter? Did he answer literally questions the interviewer meant figuratively? Did his flat affect or avoidance of eye contact lead the interviewer to read him as evasive?

These questions reach court in motions to suppress, in challenges to a waiver, in requests for expert assistance, and in trial testimony about the weight a statement deserves. Investigators and prosecutors need statements that are accurate and will survive review; defense counsel need to know whether a statement’s reliability can fairly be questioned and what evidence that would take.

This paper summarizes the research on interrogative suggestibility, compliance, false confessions, understanding of rights advisements, and the way autistic suspects present and are perceived in interviews. The research is mixed, and the paper gives equal attention to findings of vulnerability and to findings of no difference. It then reviews the federal, military, and Arizona law that governs these questions. Many autistic adults prefer identity-first language (“autistic person”), and the paper uses it alongside “person with autism.” Where the paper refers to the material at issue in these investigations, it uses “child sexual abuse material (CSAM)”; statutes and courts use “child pornography,” and the paper uses that term when discussing the law. The general research on false confessions and suggestibility is covered in the False Confessions resource, and this paper gives only the core points needed to place the autism research in context. The custody, rights-advisement, and corroboration questions that arise when any suspect is questioned at home while a search warrant is executed are discussed in Interviews at Warrant Execution and Confession Reliability in the Online Sex Offense resource.

Why Interviews Are Demanding for Autistic Suspects

A police interview places heavy demands on social communication. The suspect must follow rapid questions, infer what the interviewer is getting at, interpret indirect or figurative language, manage anxiety, and decide what to say in a setting where the stakes are high and the rules are unfamiliar. Autism involves differences in each of these areas, though the size of the differences varies greatly from person to person.

Many autistic suspects are undiagnosed when they reach custody. When people detained at six London custody centers over eight weeks were offered screening, 216 of 303 eligible arrestees agreed, and about 5% of those without an existing diagnosis (drawn from the 216 who agreed) screened above the threshold of a 10-item autism-trait questionnaire (50% screened positive for ADHD traits on a separate screen); screening was done by healthcare professionals, detention officers, or arresting officers (Brown et al., 2025). A positive screen indicates a need for assessment, not a diagnosis, and the figure is a screening result in one setting, not an estimate of how common autism is among suspects.

Studies of autistic people’s experiences with police describe these demands. In a survey of 394 police officers in England and Wales, 42% were satisfied with how they had worked with autistic people, and 37% had received any autism training. Autistic adults and parents who responded were largely dissatisfied with their contacts with police (Crane et al., 2016). In interviews with 12 autistic adults who had been detained as suspects in England, participants described difficulty understanding their arrest and detention, difficulty coping with the custody environment, and a strong wish to get out of custody that affected how they made decisions and engaged in the interview (Holloway-George et al., 2025). These are small and self-selected samples, and self-reports after the fact are not independent evidence of what happened in a given interview. They do identify the issues that recur in practice.

Experimental work provides more controlled evidence. In a mock-suspect study, 32 autistic and 33 non-autistic adults, matched for age and IQ, took part in a virtual burglary scenario as either innocent or guilty suspects and were then interviewed. Innocent autistic participants reported fewer details supporting their innocence than innocent non-autistic participants. However, both groups reported similar amounts of investigation-relevant information, and their statements were similarly consistent with the evidence. Autistic participants, whether innocent or guilty, reported more difficulty understanding the questions, more anxiety, and a less supportive interview (Bagnall et al., 2023). The finding cuts in two directions: autistic suspects were not less accurate overall, but innocent autistic suspects did less to help themselves. An earlier experiment points the same way. Autistic and non-autistic adults listened to hypothetical scenarios in which police wrongly suspected them of a crime, each containing a fact that would show they were not involved; autistic adults were markedly less likely to identify and offer that fact, and perspective-taking ability explained a significant share of the difference beyond verbal IQ and memory (Young & Brewer, 2020). The scenarios were heard, not lived, and no interviewer was present, so the study shows a difference in a laboratory task rather than in real interviews.

Interrogative Suggestibility

Interrogative suggestibility is the tendency to accept misleading information in questions and to change answers under negative feedback. It is usually measured with the Gudjonsson Suggestibility Scales, in which a person hears a short story, recalls it, answers leading questions, is told that he made errors, and is questioned again. Suggestibility is distinct from compliance, discussed below, and the distinction matters because the research on the two points in different directions.

The studies of autistic adults have not found greater interrogative suggestibility. North, Russell, and Gudjonsson (2008) compared 26 intellectually able autistic adults with 27 controls matched for gender and IQ and found no significant group differences in suggestibility. They noted that suggestibility scores in the autistic group were bimodally distributed, meaning that some autistic participants scored high and others low, and they cautioned that individual characteristics should be assessed rather than assumed. Maras and Bowler (2012) likewise found no differences on any of the suggestibility measures between autistic and comparison adults, even though the autistic group reported more paranoia.

A meta-analysis of research on suggestibility and false memory went further. Griego and colleagues (2019) found that people with intellectual disability showed greater susceptibility to suggestion and false memory than comparison groups, but that people with autism showed lower suggestibility and were less likely to develop false memories. The meta-analysis pooled a limited number of studies with varying methods and age groups, and its findings concern memory tasks rather than police interviews. It does show that the vulnerability long documented in intellectual disability cannot simply be transferred to autism. Some practitioner literature has treated the two populations as parallel. The evidence does not support that equation for suggestibility, although an autistic person who also has intellectual disability may share the vulnerabilities associated with it.

Compliance

Compliance is the tendency to go along with requests or demands to avoid conflict or please others, even while privately disagreeing. It is usually measured with the self-report Gudjonsson Compliance Scale. Compliance is relevant to interviews because a compliant person may agree to propositions or sign statements he does not believe, in order to end an uncomfortable encounter.

Here the findings are mixed. North and colleagues (2008) found that autistic participants were rated significantly more compliant than controls on both self-report and parent report, and also had higher scores on measures of anxiety, depression, fear of negative evaluation, and paranoia. They suggested that autistic people may be more eager to avoid conflict and confrontation. Maras and Bowler (2012), using the same compliance measure, found no difference between autistic and comparison adults and concluded that autistic people may be no more likely than others to give way to interrogative pressure.

Chandler, Russell, and Maras (2019) added a behavioral measure. With 26 autistic and 26 typically developing adults, they found that autistic participants were more compliant both on the self-report scale and on an experimental task in which participants were asked to agree to a request after first refusing a larger one. Autistic participants also reported more childhood bullying, more anxiety, and lower self-esteem. Self-esteem was the only unique predictor of self-reported compliance, and none of the predictors, nor the self-report scale itself, predicted behavior on the experimental task, which the authors suggested may depend more on situational and motivational factors.

Taken together, two of three controlled studies found higher compliance in autistic adults, and one did not. All three used small samples of intellectually able adults recruited for research, not suspects. The defensible statement is that some autistic adults are more compliant than their peers, that the tendency appears to be associated with anxiety, self-esteem, and social history, and that it has to be assessed in the individual. A compliance score does not establish that a particular statement was the product of compliance.

False Confessions

No study has measured how often autistic suspects make false confessions in actual cases. The closest evidence comes from self-report surveys of prisoners. In a study of 386 male prisoners in Scotland, a third reported having made a false confession to police at some point in their lives, most often to cover for someone else. Conduct disorder, ADHD, psychological distress, psychiatric symptoms, and compliance were all significant predictors, and conduct disorder and ADHD were the strongest. Autism was among the conditions examined but was not reported as one of the significant predictors (Gudjonsson et al., 2021). The study relied on prisoners’ accounts of their own past confessions, which were not verified against records.

The general research on false confessions identifies both situational risk factors, such as lengthy interrogation, presentation of false evidence, and minimization of the consequences of confessing, and dispositional risk factors, such as youth, intellectual disability, mental illness, and personality traits including compliance and suggestibility (Kassin et al., 2010; Kassin et al., 2025). Autism is not established as an independent dispositional risk factor in that literature. A careful analysis therefore asks whether a particular autistic suspect showed the traits that are documented risk factors, such as high compliance, marked anxiety, co-occurring ADHD or intellectual disability, or a strong wish to escape the interview, and whether the interview used the techniques that increase risk. The False Confessions: Suggestibility resource discusses the measurement of these traits.

Understanding Miranda and Article 31 Rights

Research on rights comprehension in general populations shows that understanding is often weaker than courts assume. Rogers and colleagues (2007) analyzed 560 Miranda warnings used across the United States and found that their reading level ranged from below third grade to postgraduate, and that warnings differed in what they told suspects, such as when an attorney would be available and whether legal help was free. In a later study, the same research group documented widespread mistaken beliefs about Miranda rights among adults, including pretrial defendants, contrary to the common assumption that everyone knows them (Rogers et al., 2010). Those findings concern suspects generally.

Research specific to autism is much thinner. Salseda and colleagues (2011) reviewed the Miranda-comprehension literature and argued that features of autism, including literal interpretation, difficulty with abstract concepts, and differences in social cognition, could impair understanding of the rights and of the consequences of waiving them. Their article is a review that develops hypotheses; it is not an empirical test. Controlled studies of Miranda comprehension in autistic adults are scarce. The only located empirical comparison of autistic and non-autistic adults’ understanding of a rights advisement concerns the Canadian caution on the right to silence. In an online simulation, 32 autistic and 215 non-autistic adults, most recruited from a university participant pool and with autism self-reported (62.5% reporting a professional diagnosis), imagined being arrested while innocent, watched an actor deliver the standard caution, and were then questioned. The groups did not differ in their understanding of the caution or in how often they made a statement; understanding was poor in both groups, 83% made a statement, and only first language predicted understanding, while higher agreeableness, conscientiousness, and negative emotionality were associated with making a statement. In written explanations from a subset, autistic and non-autistic participants gave somewhat different reasons for their choices (Bjornson et al., 2026). The study involved a hypothetical, low-stakes scenario and a small autistic group, and Canadian wording and procedure differ from Miranda practice, so any application to American warnings must account for those differences. It does, however, offer no support for the claim that autism alone produces poorer comprehension of a caution. Claims that autistic suspects as a group do not understand the warnings therefore go beyond the evidence. Equally, the absence of such studies does not show that understanding is adequate in a particular case.

In practice, the relevant questions are individual. Did the suspect understand the words and the concepts behind them, such as the right to stop answering at any time? Did he take a polite request (“Would you mind talking with us?”) as a directive? Did he answer “yes” to “Do you understand?” because the question invited that answer? Standardized measures of rights comprehension exist, and an evaluator can use them with a developmental history and a review of the recording. Strong verbal ability may lead anyone, including investigators, to assume that an autistic suspect understands more than he does, although this has not been studied directly; in other cases, autistic adults understand the rights well.

Demeanor and Perceived Deception

Autistic adults may be misread in interviews. In a study in which 1,410 participants viewed video-recorded interviews of 30 autistic and 29 non-autistic adults, autistic interviewees telling the truth were rated as more deceptive and less credible than non-autistic interviewees. The effect was driven by the interviewee’s overall presentation rather than by any single behavior, such as gaze aversion, repetitive movement, literal interpretation, or flat affect (Lim et al., 2022). A later study of 43 autistic and 41 non-autistic adults interviewed online about a simulated chatroom found only minor differences in individual behaviors: greater difficulty with figurative language, longer hesitations, and different intonation. It also found wide variation within both groups, and behaviors were neither universal nor consistently displayed together (Logos et al., 2026). Two further experiments showed recorded online interviews of autistic and non-autistic people suspected of computer hacking in a simulated chatroom to more than 600 autistic and non-autistic observers. Autistic male suspects, and autistic suspects with more pronounced speech hesitations, were evaluated more harshly on suspiciousness, credibility, and guilt. The observers’ own diagnostic status made no difference, which did not support the hypothesis that autistic observers would judge autistic suspects more favorably; what predicted harsher judgments, for autistic and non-autistic suspects alike, was the degree to which a suspect’s behavior departed from what observers expected (Brewer et al., 2026). The observers were members of the public recruited online, not police officers or jurors, and the interviews were simulated.

Interviewers who treat gaze aversion, hesitation, or unusual affect as signs of deception may misjudge autistic suspects, and accusatory questioning that follows raises the situational risk discussed above. At the same time, a diagnosis does not make a suspect’s account more or less truthful. The research addresses how an interview should be interpreted, not whether a statement is true.

Adjusting the Interview

Professional guidance increasingly recommends that investigators identify possible autism early and adjust the interview. A UK consensus statement prepared by clinicians, lawyers, and researchers, with input from autistic people, recommends identification at the earliest stages of the criminal process and adjustments to communication and environment throughout (Woodhouse et al., 2024). Practitioner reviews recommend plain and concrete questions, avoiding figurative language and compound questions, explaining the process and its purpose, allowing extra processing time and breaks, reducing sensory stressors, and confirming understanding by asking the person to explain in his own words rather than by asking “Do you understand?” (Murphy, 2018). In England and Wales, the Police and Criminal Evidence Act codes of practice provide for an appropriate adult for vulnerable suspects. That framework has no force in American courts, but it illustrates one model, and the research on it is a caution against assuming that a support person solves the problem. In a survey of 55 trained appropriate adults, respondents knew some key features of autism but were less aware of how those features could affect an interview; they described 15 incidents in which autistic characteristics, such as rigid or repetitive behavior, disrupted the interview, and reported that they had responded effectively, a self-assessment that was not checked against the recordings (Richards & Milne, 2020).

Training for American officers has been studied mainly through surveys before and after a course. In a pilot survey of 51 officers, 56.9% rated their knowledge of autism as low, and those with prior training or personal experience reported more comfort and knowledge (Christiansen et al., 2023). After autism-specific training, officers’ self-rated knowledge and confidence increased, in one study equally whether the presentation was given alone or combined with virtual-reality or live simulation (Gardner & Campbell, 2020, 2025). In a nonrandomized study of 438 officers in California and Florida, overall stigma scores fell after training, but item-level changes were mixed: officers endorsed more neurodiversity-affirming attitudes alongside greater endorsement of conventional behavioral expectations and less trust in information provided by autistic people (Gardner et al., 2026). A scoping review found only five evaluated training programs (Sreckovic et al., 2023). None of these studies had a control group or measured whether training changes how officers conduct interviews or whether it affects the reliability of the statements obtained.

Adjustments serve both sides: a recorded interview that explains rights plainly and checks understanding produces a statement that is more reliable and more likely to be admitted.

Limits of the Evidence

The controlled studies of suggestibility and compliance in autism are few and small, typically with 25 to 35 autistic participants per study, and they involve intellectually able adults recruited for research rather than suspects in custody. Their results conflict on compliance. The meta-analytic finding of lower suggestibility rests on a limited number of studies, many using memory tasks rather than interrogation analogues. No study has measured the rate of false confessions among autistic suspects in real cases, and no empirical study of Miranda comprehension in autistic adults was identified for this paper; the one located study of rights comprehension concerns the Canadian caution, used a hypothetical online scenario, and had 32 autistic participants with self-reported diagnoses. The experiential studies of custody and policing are small, self-selected, and from the United Kingdom. The studies of how observers judge autistic suspects used members of the public and simulated interviews, and the police-training studies rely on officers’ self-reports immediately after training, without control groups.

No test can determine whether a particular statement is true, whether a particular suspect understood a particular warning, or whether he agreed with a suggestion because of compliance. Suggestibility and compliance scales measure tendencies under standardized conditions, administered after the fact when the person knows what is at stake. What an evaluation can do is describe the person’s communication and cognitive profile, measure relevant traits, compare them with the recorded interview, and say how strongly the evidence supports any conclusion about his understanding.

Questions to Ask About This Research

When the meta-analytic finding of lower suggestibility in autism (Griego et al., 2019) is offered to show that a statement is reliable, ask how much evidence it rests on and whether it applies to this person: it pooled a limited number of studies that used memory tasks rather than police interviews, and North, Russell, and Gudjonsson (2008) found that suggestibility scores in their autism group were bimodally distributed. When higher compliance findings (North et al., 2008; Chandler et al., 2019) are offered to show that a statement was the product of compliance, ask how well the finding has held up and whether the conclusion follows from the design, since one of three controlled studies found no difference (Maras & Bowler, 2012), the samples were small groups of research volunteers rather than suspects, and the self-report scale did not predict behavior on the experimental task. When studies of how autistic suspects are judged (Lim et al., 2022; Brewer et al., 2026) are cited, ask who did the judging: members of the public watching recorded simulations, not officers conducting or jurors weighing a real interview. When the 5% positive-screen figure from London custody (Brown et al., 2025) is cited, ask what exactly was counted: a positive screen on a 10-item trait questionnaire, which calls for assessment and is not a diagnosis. These questions are explained in “Reading Research Claims: A Guide for Attorneys” on this site.

Police Interviews and the Law

Federal and Military Standards

A statement obtained through custodial interrogation is admissible only if the suspect was warned and validly waived his rights (Miranda v. Arizona, 1966). A valid waiver must be voluntary, in the sense of being free of intimidation, coercion, or deception, and knowing and intelligent, made with awareness of the nature of the right and the consequences of abandoning it (Moran v. Burbine, 1986). The government must prove waiver by a preponderance of the evidence, and coercive police activity is a necessary predicate for finding a confession involuntary under the Due Process Clause or a waiver involuntary under Miranda (Colorado v. Connelly, 1986). Waiver may be implied from a suspect’s conduct after he has been warned and has understood the warnings, and a suspect who wishes to remain silent must say so unambiguously (Berghuis v. Thompkins, 2010). The result is that a suspect’s autism does not by itself make a statement involuntary under Connelly, but it can bear directly on whether a waiver was knowing and intelligent and on whether police conduct that would not overbear most people was coercive as applied to this person.

When a statement is admitted, the defendant may present evidence about the physical and psychological circumstances in which it was obtained, because those circumstances bear on its credibility (Crane v. Kentucky, 1986), and the jury is to give the statement the weight it deserves (18 U.S.C. § 3501(a)). Expert testimony about autism, suggestibility, or compliance must satisfy Federal Rule of Evidence 702 as amended December 1, 2023, which 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, along with Daubert v. Merrell Dow Pharmaceuticals, Inc. (1993) and Kumho Tire Co. v. Carmichael (1999). Courts have admitted and excluded such testimony depending on the showing made; the Seventh Circuit, for example, found error in excluding false-confession testimony on the ground that jurors need no help (United States v. Hall, 1996).

In the military, Article 31(b), UCMJ, requires that a person subject to the Code who questions an accused or suspect first inform him of the nature of the accusation, that he need not make a statement, and that any statement may be used as evidence against him at court-martial. Article 31(d) excludes statements obtained in violation of the article or through coercion, unlawful influence, or unlawful inducement. Military Rules of Evidence 304 and 305 implement these protections and the Miranda counsel rights, and Rule 304 requires independent corroboration of a confession. Article 31(b) applies whether or not the questioning is custodial, which makes it relevant to the common situation of a service member questioned at work or during a search. The Court of Appeals for the Armed Forces has upheld exclusion of a defense psychologist’s opinion that a confession was false or coerced when the opinion lacked a reliable fit to the accused (United States v. Griffin, 1999), and has upheld denial of government-funded expert assistance on false confessions when the defense did not show necessity, such as evidence of a condition affecting the reliability of the confession (United States v. Bresnahan, 2005). A documented autism diagnosis, with evidence of how it affected the interview, is the kind of showing those decisions contemplate. Military Rule of Evidence 702 now incorporates the 2023 amendment to the federal rule (Joint Service Committee on Military Justice, 2025).

Arizona

Article 2, section 10 of the Arizona Constitution protects against compelled self-incrimination, and Arizona courts apply Miranda and the federal voluntariness standards. Under Arizona Rule of Criminal Procedure 16.2(b), the State must prove by a preponderance of the evidence that its evidence was lawfully obtained, once the defendant makes a prima facie case as the rule requires. Once a confession is admitted, the jury may hear evidence on voluntariness and is instructed to give the confession such weight as it deserves under all the circumstances (A.R.S. § 13-3988).

Arizona Rule of Evidence 702 was conformed to the 2023 federal amendment effective January 1, 2024. State v. Salazar-Mercado (2014) permits educational, “cold” expert testimony that explains general principles without applying them to the parties, if it satisfies Rules 702 and 403. Testimony about the research on autism, compliance, and perceived deception may take that form. 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. That rule concerns the elements of the offense. Expert evidence offered on whether a waiver was knowing and intelligent, or on the circumstances bearing on a statement’s reliability, addresses a different question, though its admissibility remains subject to Rules 702 and 403.

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 supports a prosecutor’s position that autism does not, as a group finding, make a suspect more suggestible, and that a meta-analysis found lower suggestibility in autistic samples. It also supports the point that controlled studies of compliance conflict, that no study shows a higher rate of false confessions among autistic suspects, and that Connelly requires coercive police conduct before a statement is involuntary. A statement is most defensible when the record shows the warnings were given in plain language, understanding was checked by having the suspect explain the rights in his own words, the questions were concrete and non-leading, breaks were offered, and the entire interview was recorded. Where autism is known or suspected early, these adjustments, and corroboration of key admissions through the forensic examination, protect a conviction on review and in a court-martial satisfy the corroboration requirement of Military Rule of Evidence 304.

The research also counsels caution. Compliance findings are mixed in small samples, with two of three controlled studies finding higher compliance; innocent autistic mock suspects in one study offered fewer details supporting their innocence, and in a scenario-based study autistic adults were less likely to identify the fact that would have cleared them (Young & Brewer, 2020). In one study of lay raters viewing recorded interviews, truthful autistic interviewees were judged less credible than others; that finding concerns observers’ impressions, not any individual’s truthfulness. Literal and unguarded answering can produce candid admissions as well as inaccurate ones, depending on the question and the setting; each answer should be checked against records and informants. Interview techniques that treat gaze aversion, hesitation, or flat affect as signs of deception, or that rely on leading or accusatory questions, increase the risk of an unreliable statement and affect how much weight the statement receives. A “yes” to “Do you understand your rights?” is weak evidence of understanding for any suspect; asking him to explain the rights in his own words is more informative, and this matters especially when the suspect is autistic or anxious.

Special Issues for Defense

The research supports defense examination of whether a particular autistic client’s waiver was knowing and intelligent and whether his statement deserves the weight the government gives it. That examination is better supported when it rests on the client’s individual profile: a diagnosis supported by developmental history, measured compliance and anxiety, any co-occurring ADHD or intellectual disability, and a close comparison of the recorded interview with that profile. Evidence that he misunderstood figurative or indirect language, that he expressed a wish to end the interview, or that leading questions supplied the details he later adopted is better supported than the diagnosis alone. Crane protects the right to put the circumstances of the statement before the jury, and Bresnahan indicates what showing supports a request for expert assistance in a court-martial.

The research also counsels caution about overclaiming. The controlled studies did not find greater suggestibility in autistic adults, the meta-analysis found less, and compliance findings are mixed. An expert who testifies that autistic people as a group are suggestible or prone to false confessions is likely to be contradicted by the controlled studies reviewed here, and an opinion that a particular statement is false is beyond what any test can show and, in courts-martial, risks exclusion when it lacks a reliable fit to the accused (Griffin). Where admissions are corroborated by the forensic evidence, as they often are in CSAM cases, the statement’s reliability may matter less to the outcome than other issues, such as sentencing.

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

Bagnall, R., Cadman, A., Russell, A., Brosnan, M., Otte, M., & Maras, K. L. (2023). Police suspect interviews with autistic adults: The impact of truth telling versus deception on testimony. Frontiers in Psychology, 14, Article 1117415. https://doi.org/10.3389/fpsyg.2023.1117415

Bjornson, S. E., Moore, T. E., & Perry, A. (2026). You are not obliged to say anything: A comparison of autistic and non-autistic adults’ comprehension of the Canadian right to silence. Journal of Police and Criminal Psychology, 41(3), 687–698. https://doi.org/10.1007/s11896-026-09811-8

Brewer, N., Lucas, C. A., Logos, K., Young, R. L., & Lim, A. (2026). How behavioral differences between autistic and nonautistic suspects in forensic interviews affect observers’ evaluations: Evidence for expectancy violations theory but not the double empathy hypothesis. Law and Human Behavior. Advance online publication. https://doi.org/10.1037/lhb0000679

Brown, D., Procyshyn, T. L., Allison, C., Neyroud, E., & Baron-Cohen, S. (2025). Neurodiversity in custody: Screening results for ADHD and autistic traits in individuals arrested by the London Metropolitan Police. Criminal Behaviour and Mental Health. Advance online publication. https://doi.org/10.1002/cbm.70018

Chandler, R. J., Russell, A., & Maras, K. L. (2019). Compliance in autism: Self-report in action. Autism, 23(4), 1005–1017. https://doi.org/10.1177/1362361318795479

Christiansen, A., Minich, N. M., & Clark, M. (2023). Pilot survey: Police understanding of autism spectrum disorder. Journal of Autism and Developmental Disorders, 53(2), 738–745. https://doi.org/10.1007/s10803-021-04968-z

Crane, L., Maras, K. L., Hawken, T., Mulcahy, S., & Memon, A. (2016). Experiences of autism spectrum disorder and policing in England and Wales: Surveying police and the autism community. Journal of Autism and Developmental Disorders, 46(6), 2028–2041. https://doi.org/10.1007/s10803-016-2729-1

Gardner, L., & Campbell, J. M. (2020). Law enforcement officers’ preparation for calls involving autism: Prior experiences and response to training. Journal of Autism and Developmental Disorders, 50(12), 4221–4229. https://doi.org/10.1007/s10803-020-04485-5

Gardner, L., & Campbell, J. M. (2025). Training law enforcement officers about autism: Evaluation of adding virtual reality or simulation to a traditional training approach. Journal of Autism and Developmental Disorders, 55(8), 2643–2650. https://doi.org/10.1007/s10803-024-06383-6

Gardner, L., Powell, L., Montoya, K., Ku, T., Lott, A., & Campbell, J. (2026). Law enforcement officers and autism stigma: Changes associated with autism training. Journal of Autism and Developmental Disorders. Advance online publication. https://doi.org/10.1007/s10803-026-07544-5

Griego, A. W., Datzman, J. N., Estrada, S. M., & Middlebrook, S. S. (2019). Suggestibility and false memories in relation to intellectual disability and autism spectrum disorder: A meta-analytic review. Journal of Intellectual Disability Research, 63(12), 1464–1474. https://doi.org/10.1111/jir.12668

Gudjonsson, G. H., Gonzalez, R. A., & Young, S. (2021). The risk of making false confessions: The role of developmental disorders, conduct disorder, psychiatric symptoms, and compliance. Journal of Attention Disorders, 25(5), 715–723. https://doi.org/10.1177/1087054719833169

Holloway-George, C. A., Munro, N., Jackson, J., & Ropar, D. (2025). Experiences of autistic people in police custody: The need for adjustments to improve participation in the custody process. Journal of Autism and Developmental Disorders. Advance online publication. https://doi.org/10.1007/s10803-025-06952-3

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

Kassin, S. M., Cleary, H. M. D., Gudjonsson, G. H., Leo, R. A., Meissner, C. A., Redlich, A. D., & Scherr, K. C. (2025). Police-induced confessions, 2.0: Risk factors and recommendations. Law and Human Behavior, 49(1), 7–53. https://doi.org/10.1037/lhb0000593

Kassin, S. M., Drizin, S. A., Grisso, T., Gudjonsson, G. H., Leo, R. A., & Redlich, A. D. (2010). Police-induced confessions: Risk factors and recommendations. Law and Human Behavior, 34(1), 3–38. https://doi.org/10.1007/s10979-009-9188-6

Lim, A., Young, R. L., & Brewer, N. (2022). Autistic adults may be erroneously perceived as deceptive and lacking credibility. Journal of Autism and Developmental Disorders, 52(2), 490–507. https://doi.org/10.1007/s10803-021-04963-4

Logos, K., Lim, A., Brewer, N., & Young, R. L. (2026). The behavioral presentation of autistic adults in a forensic interview. Journal of Autism and Developmental Disorders, 56(9), 3503–3522. https://doi.org/10.1007/s10803-025-06805-z

Maras, K. L., & Bowler, D. M. (2012). Brief report: Suggestibility, compliance and psychological traits in high-functioning adults with autism spectrum disorder. Research in Autism Spectrum Disorders, 6(3), 1168–1175. https://doi.org/10.1016/j.rasd.2012.03.013

Murphy, D. (2018). Interviewing individuals with an autism spectrum disorder in forensic settings. International Journal of Forensic Mental Health, 17(4), 310–320. https://doi.org/10.1080/14999013.2018.1518939

North, A. S., Russell, A. J., & Gudjonsson, G. H. (2008). High functioning autism spectrum disorders: An investigation of psychological vulnerabilities during interrogative interview. Journal of Forensic Psychiatry & Psychology, 19(3), 323–334. https://doi.org/10.1080/14789940701871621

Richards, J., & Milne, R. (2020). Appropriate adults: Their experiences and understanding of autism spectrum disorder. Research in Developmental Disabilities, 103, Article 103675. https://doi.org/10.1016/j.ridd.2020.103675

Rogers, R., Harrison, K. S., Shuman, D. W., Sewell, K. W., & Hazelwood, L. L. (2007). An analysis of Miranda warnings and waivers: Comprehension and coverage. Law and Human Behavior, 31(2), 177–192. https://doi.org/10.1007/s10979-006-9054-8

Rogers, R., Rogstad, J. E., Gillard, N. D., Drogin, E. Y., Blackwood, H. L., & Shuman, D. W. (2010). “Everyone knows their Miranda rights”: Implicit assumptions and countervailing evidence. Psychology, Public Policy, and Law, 16(3), 300–318. https://doi.org/10.1037/a0019316

Salseda, L. M., Dixon, D. R., Fass, T., Miora, D., & Leark, R. A. (2011). An evaluation of Miranda rights and interrogation in autism spectrum disorders. Research in Autism Spectrum Disorders, 5(1), 79–85. https://doi.org/10.1016/j.rasd.2010.06.014

Sreckovic, M. A., Kenney, C. K., & Wallace, M. (2023). Autism training for law enforcement officers: A scoping review. Journal of Autism and Developmental Disorders, 53(10), 3835–3846. https://doi.org/10.1007/s10803-022-05692-y

Woodhouse, E., Hollingdale, J., Davies, L., Al-Attar, Z., Young, S., Vinter, L. P., Agyemang, K., Bartlett, C., Berryessa, C., Chaplin, E., Deeley, Q., Freckelton, I., Gerry, F., Gudjonsson, G., Maras, K., Mattison, M., McCarthy, J., Mills, R., Misch, P., . . . Allely, C. (2024). Identification and support of autistic individuals within the UK criminal justice system: A practical approach based upon professional consensus with input from lived experience. BMC Medicine, 22, Article 157. https://doi.org/10.1186/s12916-024-03320-3

Young, R. L., & Brewer, N. (2020). Brief report: Perspective taking deficits, autism spectrum disorder, and allaying police officers’ suspicions about criminal involvement. Journal of Autism and Developmental Disorders, 50(6), 2234–2239. https://doi.org/10.1007/s10803-019-03968-4

Legal Authorities

Berghuis v. Thompkins, 560 U.S. 370 (2010).

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

Colorado v. Connelly, 479 U.S. 157 (1986).

Crane v. Kentucky, 476 U.S. 683 (1986).

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

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

Miranda v. Arizona, 384 U.S. 436 (1966).

Moran v. Burbine, 475 U.S. 412 (1986).

United States v. Hall, 93 F.3d 1337 (7th Cir. 1996).

United States v. Bresnahan, 62 M.J. 137 (C.A.A.F. 2005).

United States v. Griffin, 50 M.J. 278 (C.A.A.F. 1999).

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

State v. Salazar-Mercado, 234 Ariz. 590, 325 P.3d 996 (2014).

U.S. Const. amend. V.

Ariz. Const. art. 2, § 10.

18 U.S.C. § 3501(a).

10 U.S.C. § 831 (Article 31, Uniform Code of Military Justice).

Ariz. Rev. Stat. Ann. § 13-3988.

Fed. R. Evid. 702.

Mil. R. Evid. 304.

Mil. R. Evid. 305.

Mil. R. Evid. 702.

Ariz. R. Crim. P. 16.2(b).

Ariz. R. Evid. 702.

Related topic papers: Competency, Effective Participation, and Accommodations · Demeanor, Voir Dire, and Educating the Fact-Finder · Autism in Adults: Diagnosis, Terminology, and Common Misunderstandings · Screening and Diagnosing Autism in an Adult Defendant · Autism Evaluations in Courts-Martial

Related topics from Arizona Forensic Services: False Confessions · False Confessions: Suggestibility · Online Sex Offense · Courts-Martial