Show the full module text

Module 11 — ASIM-24 — Sex, Intimacy and Sensory Reality

A reverse-scored questionnaire about sexual knowledge, sensory comfort, boundaries and communication — what the research on autistic adults and intimacy actually shows, and why our own instrument has no validation behind it.

Autistic Self-Discovery · Part Four — Feeling, Naming, Connecting · 13 min read, about 26 min with the workbook

The big idea

If you have ever said yes because saying no would have taken more words than you had at that moment, or left an evening early because of a smell nobody else could detect — this module is about the part of adult life that gets the least honest information. It is also about a questionnaire we wrote ourselves, which has never been tested, and which runs backwards.

Hold the two halves of this together. The elevated rates are real and they are not your fault. What the evidence points at is a set of conditions — sex education that never arrived, environments that will not accommodate, compliance trained in by decades of being corrected, and social networks too thin to pass on a warning. Conditions can be changed. That is the entire reason this section is in the module.

Step 1 — The lesson

A. Low is the difficult end

Start with the thing that will otherwise make this whole module useless to you. The ASIM-24 runs in the opposite direction to almost every other questionnaire in this course. On the RAADS-14 or the AQ-50, a higher number means more traits and more reason to look further. Here, 24 is the hard end and 120 is the comfortable one.

A total of 34 is not a small result. It is close to the largest signal the questionnaire can give you that some part of this area is costing you something. A total of 112 is not a warning. Read it the way you would read a battery indicator rather than a thermometer.

Diagram — A · Low is the difficult end. These three bands are the ones the live screener uses, word for word. The colors run the opposite way to every other screener on the site, which is the single most likely thing to go wrong when you read your result. If you have taken two questionnaires this week, your eye will get this backwards at least once.

The mechanics are simple. Twenty-four questions, five to seven minutes, a total between 24 and 120, and four subscale scores that each run from 6 to 30. What the live page does not tell you is what scale the individual items use — how many points each answer is worth, or what the endpoints are called. We are not going to guess on your behalf, because a screener that will not state its own scoring is a screener you should hold loosely.

None of that makes the questions worthless. It does mean the number on the end of them is the least interesting thing the exercise produces. The four subscale scores, and the sentences you find yourself saying while you answer, are where the value sits.

B. Four parts, and what each one is actually asking

The questionnaire splits into four named areas. They are not arbitrary. Each one corresponds to something the published research on autistic adults and intimacy has found repeatedly, and it is worth knowing what sits behind each heading before you answer anything.

Sexual Knowledge & Understanding. When 95 autistic adults were compared with 117 matched adults without a diagnosis, the autistic group had drawn far less of what they knew from social sources — parents, teachers, friends — and far more from non-social ones such as television, the internet and pornography. Their measured knowledge was lower as a result: a mean of 17.44 against 23.54.¹ That is not a difference in capacity. It is a difference in who bothered to explain anything.

A separate study of 20 autistic young adults and 20 comparison participants found both groups rated the sex education they had received about equally, but the autistic group scored significantly lower on sexual awareness. Every autistic participant described school-based teaching as inadequate preparation, against three in four of the comparison group.² The teaching was the same. What it left behind was not.

Diagram — B · Four parts, and what each one is asking. Four subscales of 6 to 30 each, adding to the total. Because the parts can pull in opposite directions — strong knowledge and low sensory comfort, say — a mid-range total can hide a very specific problem. Write down all four before you look at the sum.

Sensory Comfort & Regulation. In 24 in-depth interviews with autistic adults aged 18 to 61, sensory dysregulation during partnered sex came up as one of the two most common concerns, alongside the difficulty of courtship itself. The same participants described what they had done about it: sensory accommodations, deliberate planning of when and how, explicit negotiation with partners, and scripts that departed from the neurotypical version entirely.³

A qualitative study drawing on written narratives from more than 120 autistic people found the same thing running in both directions. Some described touch, light, sound or smell making closeness unbearable when they wanted it. Others described seeking out stronger sensation, or finding physical closeness genuinely regulating — the calmest their nervous system got all day.⁴ Sensory difference in this area is not a synonym for aversion.

Boundaries & Consent Self-Advocacy. Forty-three autistic adults who wrote about being exploited by people they knew described something the researchers named heightened compliance: going along with unreasonable requests, sometimes from fear, often simply to avoid conflict. Several described difficulty recognizing, at the time, that a line had been crossed at all.⁵ If you have spent thirty years being told that your read of a situation is wrong, compliance is a trained response, not a personality.

Relationship Communication Confidence. Pooled across 15 studies, roughly half of autistic participants met the threshold for alexithymia — difficulty identifying and describing your own feelings — against about 5 per cent of comparison participants.⁶ Consent is usually taught as though the only hard part were saying the word. If the feeling itself arrives unlabelled, or arrives four hours later in the car, the hard part is upstream of the word.

C. The finding that has to be said plainly

There is a body of research here that gets either sensationalised or quietly left out. It deserves neither. Autistic adults, and autistic women in particular, report sexual victimisation at substantially higher rates than non-autistic comparison groups, and the finding has held across different countries, samples and methods.

In that study of 95 autistic adults, 78 per cent reported at least one experience of sexual victimisation, against 47.4 per cent of the comparison group. The autistic group were around three times more likely to report unwanted sexual contact, 2.7 times more likely to report sexual coercion and 2.4 times more likely to report rape. The part that matters most is what came next in the analysis: the link between being autistic and being victimised was partly explained by lower actual sexual knowledge.¹

Diagram — C · The finding, in one picture. Read this as a fact about circumstances rather than about people. The same study that produced the 78 per cent found that knowledge accounted for part of the gap — and knowledge is the one variable on that list that somebody else decided not to give you.

The pattern repeats. Among 118 autistic and 110 non-autistic Australian adults asked about violence since the age of 15, 56.8 per cent of the autistic group reported sexual violence against 28.2 per cent of the comparison group, and the usual sex differences in who experiences what did not appear in the autistic group at all. Of the incidents described, only 13.9 per cent had been reported to police.⁷

A comparison of 134 autistic and 161 non-autistic women found higher rates of unwanted sexual experiences, with the risk concentrated in particular subgroups rather than spread evenly.⁸ An earlier study by the same group, comparing 135 autistic women with 96 autistic men and 161 non-autistic women, found autistic women reported less sexual interest yet more sexual experience, and more regretted or unwanted encounters than either comparison group.⁹ In a set of in-depth interviews with 14 women diagnosed late, nine described experiences of sexual abuse, and described the same mechanisms each time: uncertainty about unspoken social rules, difficulty reading intent, isolation from the peer network that hands other people the warnings.¹⁰

The most carefully designed study in this area is a Swedish population-based twin study of 4,500 children, with neurodevelopmental symptoms rated at ages 9 to 12 and coercive sexual experiences reported by age 18. Autistic traits were associated with roughly a threefold increase in risk among girls. But when the researchers controlled for overall neurodevelopmental symptom load, no single diagnosis predicted victimisation on its own, and the association ran through shared genetic factors rather than through anything specific to autism.¹¹ Whatever produces this risk, it is not a property of being autistic as such.

D. What the number does not mean

Here is the honest accounting, and it is short, because there is not much to account for.

The ASIM-24 was written in-house by our own team. It has never been validated. The live screener page says so in its own words: "This tool is still in development and has not been validated through formal scientific research." There is no published paper, no normative sample, no reliability coefficient, no factor analysis and no study comparing its scores against anything else. Nobody has checked whether the four subscales are actually four separate things.

Diagram — D · What the number does not mean. The band edges at 56 and 88, and the subscale edges at 12 and 17, were chosen editorially. They are not thresholds derived from data, because there is no data. Treat them as three rough drawers to file an answer in, not as lines with meaning on either side.

So a score of 61 and a score of 66 are the same score. A move from 54 to 58 across two months is not improvement, it is noise. And the total is not a percentage of anything: 40 out of 120 does not mean that 33 per cent of your intimate life is working.

There is a deeper limitation, and it is the one most likely to mislead you. The questions assume a particular shape of adult life. In a survey of 1,183 autistic and 1,203 non-autistic adolescents and adults, 10.05 per cent of autistic participants identified as asexual against 1.51 per cent of the comparison group; 63.03 per cent identified as heterosexual against 83.08 per cent; and 12.60 per cent identified as transgender against 2.00 per cent.¹² A large multi-dataset study of 641,860 people found transgender and gender-diverse participants had between three and six times the odds of an autism diagnosis.¹³

Put plainly: if you are asexual, or your relationships do not look like the ones the questions have in mind, you can score in the lowest band while nothing at all is wrong. The questionnaire has no way to distinguish "this area is difficult for me" from "this area is not part of my life and I am content with that". Only you can make that distinction, and the module would rather you made it out loud.

One more correction to the assumption people arrive with. Among 129 single autistic adults, reported sexual functioning was broadly positive. The participants who had never been in a relationship reported more sexual anxiety and less dyadic desire than those who had — a difference that tracks experience and opportunity rather than any capacity for intimacy.¹⁴

INSIDE THE INSTRUMENT

ASIM-24 — the Autistic Sexual Intimacy Measure, New Path's own instrument

Where it came from

The ASIM-24 was written by our own clinical team at the Adult Autism Assessment Center, part of the New Path Family. It was built because the published options were unsuitable: the best-established measures in this literature are research instruments such as the Sexual Behaviour Scale-III, which are not licensed for free public self-administration and were not designed to hand an adult a usable map of their own situation. There is no development paper, no item-generation protocol and no manual.

What it is made of

Twenty-four items, four subscales — Sexual Knowledge & Understanding, Sensory Comfort & Regulation, Boundaries & Consent Self-Advocacy, Relationship Communication Confidence — each scoring 6 to 30, summing to a total of 24 to 120. Administration is 5 to 7 minutes. The scale is reverse-oriented: 24 to 56 is High Support Needed, 57 to 88 is Growth Zone, 89 to 120 is Strong Foundation, with subscale breaks at 12 and 17. The live page does not state the item response scale, and we have not inferred one.

How well it performs

Unknown, in every respect that can be measured. No Cronbach alpha, no test-retest interval, no exploratory or confirmatory factor analysis, no convergent validity against the SBS-III or any other published measure, no sensitivity, no specificity, no cut-off derivation. The live screener carries the disclaimer that the tool "is still in development and has not been validated through formal scientific research". That sentence is the complete psychometric record.

Where it was validated

Nowhere. There is no normative sample, no clinical sample, no comparison group and no translation. Scores are raw, not standardised, and not comparable to any published threshold. The three band labels are editorial judgments about how to talk to a reader, not empirical classifications.

What it cannot do

It cannot screen for autism, and it cannot screen for anything else. Every domain it touches is influenced by sex education, relationship history, sensory environment, mood, medication, trauma and the person you happen to be sitting next to. A low total is consistent with a dozen situations that call for entirely different responses.

It must not be used as a risk instrument. The victimisation literature summarized above is strong,¹^,⁷^,¹¹ and none of it licenses using an unvalidated 24-item self-report to estimate an individual's risk of harm. A low Boundaries & Consent Self-Advocacy subscale is a prompt for a careful conversation and a proper history, not a flag.

The reverse orientation is a live data-entry hazard. Every other screener in this course scores upward for difficulty. A clinician transcribing several totals into one record will eventually reverse this one, and the resulting error points in the most dangerous direction — reading acute need as a strong foundation. Record it with its label attached, never as a bare number.

It conflates state with trait. Sensory comfort in intimacy fluctuates with cumulative load, sleep and safety; the qualitative evidence describes exactly this variability.³^,⁴ A single administration captures one fortnight of one life.

It cannot distinguish absence of interest from absence of comfort. Given that around one in ten autistic adults in a large survey identified as asexual, against roughly one in sixty-six of the comparison group,¹² a low total may describe a life the person is entirely content with. The instrument has no item that asks.

What a clinician does with it

Use the four domains as an agenda, not a score. Each maps onto a distinct and evidenced intervention target: accurate adult sex education where knowledge is the gap;¹ ^,² sensory accommodation and planning where comfort is the gap;³^,⁴ explicit, rehearsed consent language where compliance is the pattern;⁵ and work on interoception and emotional labeling where naming the feeling in real time is the obstacle.⁶ An international consensus process involving 493 participants, 353 of them autistic adults, reached agreement on eight of eleven recommendations, including that education about sexuality, relationships and gender diversity should be provided and that services should combine gender-related expertise with victimisation prevention.¹⁵ Take the ASIM-24 as an opening, and let the literature carry the weight.

Validity tier: 4 — built in-house by New Path. Honest and useful, not normed, no published validation. The subject matter is well evidenced and badly served; the measurement is not evidenced at all, and no decision should rest on the number it returns.

Diagram — E · Many roads, one low score. A low total is a reading, not a diagnosis of the cause. Two of the roads on this diagram — being asexual, and being with the wrong person — require no work on yourself whatsoever. Only a conversation can tell you which road you are on.

STRENGTHS LENS

Explicit is not a downgrade. It is the advanced version.

The autistic adults in the interview studies were not muddling through. They were planning the timing, adjusting the room, saying out loud what they needed and what they did not want, and building scripts that worked for them instead of copying ones that did not. Most couples never get anywhere near that level of directness, and a good deal of ordinary unhappiness comes from the assumption that it should not be necessary.

If you have ever apologized for needing to spell something out, notice what you were apologizing for: clear communication about consent and comfort, which is the thing every sex educator in the world is trying to teach. You arrived at it by a harder route. It is still the better method.

F. What helps

Five things, in rough order of how much difference they make relative to the effort. None of them requires you to change what you want.

1. Read your number the right way round, once, out loud.

Write the label next to the figure every time you record it — "68, Growth Zone" rather than "68". This sounds trivial. It is the single most common error with this questionnaire, and getting it backwards turns a signal that you could use support into a reassurance that you do not need any.

2. Fill the education gap deliberately, as an adult.

The knowledge difference in the research came from where information was available, not from any difference in ability to learn it. Pick accurate adult sources rather than absorbing it by osmosis: a good evidence-based book on anatomy and consent, a sexual health clinic, a neurodivergent-affirming therapist. This is the one variable that measurably tracked risk, and it is fixable at any age.

3. Treat the sensory side as logistics, not as a verdict on desire.

Lighting, fabric, temperature, scent, sound, and above all timing — the end of a long sensory day is the worst possible slot and it is the one most people default to. Change the conditions before you conclude anything about wanting. The interview studies are full of people who solved this with a lamp, a different sheet and a rescheduled hour.

4. Write your three sentences in advance and keep them flat.

One for yes, one for no, one for pause — "I want to stop now, nothing is wrong" is a complete sentence and needs no justification after it. Rehearsing them matters because they have to be available at the exact moment when word-finding is hardest. Flat and plain beats subtle every time; hinting is a system that fails you.

5. Run the compliance check, and treat a partner's reaction as data.

Ask yourself: would I have agreed to this on a Tuesday morning, in daylight, with no one waiting on an answer? If the honest answer is no, that is the compliance pattern rather than a decision. A partner who can hear an explicit conversation about pace and comfort without taking offence is showing you something reliable. One who cannot is showing you something equally reliable.

Step 2 — Take the screener

Twenty-four questions, five to seven minutes, free and confidential. You get a total between 24 and 120 plus four subscale scores — sexual knowledge, sensory comfort, boundaries and consent, and relationship communication. One steer before you start: this scale runs backwards, so a low score means more support could help, and the four subscales tell you far more than the total does.

Take the ASIM-24

Before you start

The ASIM-24 was written in-house by New Path and its own page states that it is still in development and has not been validated through formal scientific research. It has no published psychometrics, no normative sample and no tested cut-offs, so the band edges are editorial judgments rather than thresholds derived from data. It is reverse-scored: a low total means more support may help, which is the opposite of every other screener on this site. It is a structured way to think about an area adults are rarely given good information about, not a test, and it cannot screen for autism or for anything else. Only a clinician can diagnose.

Step 3 — Your workbook

Your answers save to this device only — we cannot see a word of what you write. This module lets you record the total and all four subscale scores the right way round, then turn the lowest one into a specific education, sensory or communication change you can actually make.

Your ASIM-24 results

Took the screener? Put your numbers in below. This scale runs backwards — a LOW score means more support needed, so write the band label next to every figure. Entirely optional — skip it if you would rather just read.

Score bands: 24–56 = High Support Needed; 57–88 = Growth Zone; 89–120 = Strong Foundation

Fields: ASIM-24 · Autistic Sexual Intimacy Measure; Total score (24–120) — remember, LOW means more support needed; My total (enter 24–120); Sexual Knowledge & Understanding (6–30); Sexual Knowledge & Understanding (enter 6–30); Sensory Comfort & Regulation (6–30); Sensory Comfort & Regulation (enter 6–30); Boundaries & Consent Self-Advocacy (6–30); Boundaries & Consent Self-Advocacy (enter 6–30); Relationship Communication Confidence (6–30); Relationship Communication Confidence (enter 6–30)

Which of the four is lowest, and what it looks like

Section B. The four parts are the useful output; the total flattens them. Pick the lowest one and describe it in ordinary language rather than in the questionnaire's words.

Fields: Lowest subscale: Sexual Knowledge & Understanding / Sensory Comfort & Regulation / Boundaries & Consent Self-Advocacy / Relationship Communication Confidence / Two or more are level; What that actually looks like in my life; Does the low score describe a difficulty, or a life I am content with?: A difficulty I would like to change / Something I am content with / Genuinely not sure yet

The education nobody gave you

Section F, item 2. In the research, the knowledge gap came from where information was available, not from ability. This is the one variable that measurably tracked risk.

Fields: Things I was never actually told, and worked out late or not at all; One accurate adult source I will use; Tick: I would rather do this with a clinician than alone

Sensory logistics

Section F, item 3. Lighting, fabric, temperature, scent, sound, and above all timing. Conditions first, conclusions later.

Fields: What reliably makes closeness harder; What reliably makes it easier, including time of day; One condition I will change first

Your three sentences

Section F, item 4. One for yes, one for no, one for pause. Flat and plain, written now, so they are available when word-finding is hardest.

Fields: My yes; My no; My pause

The compliance check

Section F, item 5. Would I have agreed to this on a Tuesday morning, in daylight, with nobody waiting on an answer?

Fields: A time I agreed to something I did not want, and what made saying no feel unavailable; What I want a partner to do differently, in one sentence they could act on; Who I would talk to about any of this

Appendix — Research companion

Peer-reviewed research

10. Bargiela S, Steward R, Mandy W (2016). The experiences of late-diagnosed women with autism spectrum conditions: an investigation of the female autism phenotype. Journal of Autism and Developmental Disorders, 46(10), 3281-3294. DOI 10.1007/s10803-016-2872-8. View the paper Framework analysis of in-depth interviews with 14 women aged 22 to 30 diagnosed in late adolescence or adulthood. Four themes emerged, including pretending to be normal and a move from passive to assertive; nine of the 14 participants described experiences of sexual abuse, and linked this to difficulty reading others' intentions, uncertainty about unspoken social rules including consent, isolation from peers who might have passed on warnings, and a strong wish for acceptance. The authors call the finding a priority for further research and for training professionals to recognise autism in women. Limitation: 14 cognitively able UK women recruited through late-diagnosis routes, diagnoses not verified from clinical records, no comparison group and no male participants, so the rate cannot be read as a prevalence estimate.

1. Brown-Lavoie SM, Viecili MA, Weiss JA (2014). Sexual knowledge and victimization in adults with autism spectrum disorders. Journal of Autism and Developmental Disorders, 44(9), 2185-2196. DOI 10.1007/s10803-014-2093-y. View the paper Compared 95 autistic adults aged 19 to 43 with 117 adults without an autism diagnosis on sources of sexual knowledge, measured knowledge and victimisation. The autistic group drew less knowledge from social sources and more from non-social sources such as television and the internet, scored lower on actual knowledge (mean 17.44 versus 23.54), and reported far higher victimisation: 78% reported at least one experience against 47.4% of the comparison group, with roughly three times the odds of unwanted sexual contact, 2.7 times the odds of sexual coercion and 2.4 times the odds of rape. Reduced actual knowledge partially mediated the relationship between autism and victimisation risk. Limitation: convenience online sample with diagnoses confirmed by self-report and the Autism Spectrum Quotient rather than clinical assessment, cross-sectional design that cannot establish causation, and victimisation measured only from age 14 onward.

14. Byers ES, Nichols S, Voyer SD (2013). Challenging stereotypes: sexual functioning of single adults with high functioning autism spectrum disorder. Journal of Autism and Developmental Disorders, 43(11), 2617-2627. DOI 10.1007/s10803-013-1813-z. View the paper Online survey of 129 single community-dwelling adults with high-functioning autism or Asperger syndrome (61 men, 68 women), assessing autistic characteristics, psychological wellbeing and several dimensions of sexual functioning. Participants reported broadly positive sexual functioning overall; those who had never been in a romantic relationship reported significantly greater sexual anxiety, lower arousability, lower dyadic desire and fewer positive sexual cognitions than those who had, and men reported better functioning than women on several dimensions. The paper is included as a corrective to the assumption that autistic adults are uninterested in or incapable of sexual wellbeing. Limitation: volunteer online sample recruited through autism organisations, all single at the time of the survey, cognitively able, with no non-autistic comparison group.

7. Gibbs V, Hudson J, Pellicano E (2023). The extent and nature of autistic people's violence experiences during adulthood: a cross-sectional study of victimisation. Journal of Autism and Developmental Disorders, 53(9), 3509-3524. DOI 10.1007/s10803-022-05647-3. View the paper Cross-sectional survey of 118 autistic adults (104 clinically diagnosed, 14 self-identified) and 110 non-autistic adults reporting experiences of sexual harassment, stalking, sexual violence and physical violence since the age of 15. Autistic participants reported significantly higher rates of every category, including sexual violence at 56.8% against 28.2%, and were more likely to report multiple types and repeated incidents; the usual sex differences seen in the non-autistic group were absent in the autistic group. Of the incidents described, only 13.9% had been reported to police. Limitation: a modest self-selected Australian sample recruited online, retrospective self-report of events across decades, and no measurement of the risk or protective factors that would explain the difference.

2. Hannah LA, Stagg SD (2016). Experiences of sex education and sexual awareness in young adults with autism spectrum disorder. Journal of Autism and Developmental Disorders, 46(12), 3678-3687. DOI 10.1007/s10803-016-2906-2. View the paper Twenty young adults with an autism spectrum diagnosis and 20 typically developing comparison participants completed the sexual knowledge, experiences, feelings and needs questionnaire and a sexual awareness questionnaire, followed by semi-structured interviews. Both groups rated the sex education they had received similarly, but the autistic group scored significantly lower on sexual awareness, and all autistic participants described school teaching as inadequate preparation against three of four comparison participants. Four interview themes emerged: dissatisfaction with school instruction, confusion about sexual orientation, social interaction difficulty, and negative intimate experiences. Limitation: only 20 participants per group, all cognitively able young adults recruited in the UK, with self-report measures and no follow-up on whether tailored education changes outcomes.

6. Kinnaird E, Stewart C, Tchanturia K (2019). Investigating alexithymia in autism: a systematic review and meta-analysis. European Psychiatry, 55, 80-89. DOI 10.1016/j.eurpsy.2018.09.004. View the paper Systematic review and meta-analysis of 15 studies using the Toronto Alexithymia Scale to compare autistic and neurotypical groups. Autistic participants scored significantly higher on the total scale and its subscales, and approximately 50% met the threshold for alexithymia against approximately 5% of comparison participants. The authors conclude that alexithymia is a common co-occurring condition in autism rather than a core feature of it, and that autistic people with alexithymia may form a distinct subgroup with distinct clinical needs. Limitation: only 15 studies with small individual samples, all relying on self-report of emotional experience in a population whose defining difficulty is describing emotional experience.

11. Ohlsson Gotby V, Lichtenstein P, Langstrom N, Pettersson E (2018). Childhood neurodevelopmental disorders and risk of coercive sexual victimization in childhood and adolescence - a population-based prospective twin study. Journal of Child Psychology and Psychiatry, 59(9), 957-965. DOI 10.1111/jcpp.12884. View the paper Prospective population-based study of 4,500 participants in the Child and Adolescent Twin Study in Sweden, with parent-rated neurodevelopmental symptoms at ages 9 to 12 and self-reported coercive sexual victimisation by age 18, analysed with bifactor modelling and genetic decomposition. In females, autism spectrum disorder was associated with roughly three times the risk and ADHD with roughly twice the risk; associations in males ran in the same direction but were not significant. When general neurodevelopmental symptom load was controlled, no specific diagnosis independently predicted victimisation, and the association between the general factor and victimisation was attributable to shared genetic influences. Limitation: parent-rated childhood symptoms rather than clinical diagnoses, self-reported victimisation collected retrospectively at age 18, and a Swedish twin cohort whose findings may not transfer to other countries or to adults.

8. Pecora LA, Hancock GI, Hooley M, Demmer DH, Attwood T, Mesibov GB, Stokes MA (2020). Gender identity, sexual orientation and adverse sexual experiences in autistic females. Molecular Autism, 11(1), 57. DOI 10.1186/s13229-020-00363-0. View the paper Compared 134 autistic females (mean age 26.2) with 161 non-autistic females (mean age 22.0) using the Sexual Behaviour Scale-III. Autistic participants were more likely to report a transgender identity and a non-heterosexual orientation, and unwanted sexual experiences were elevated, with the risk concentrated in autistic homosexual females (odds ratios of at least 3.29 against autistic heterosexual females, and at least 2.38 against non-autistic comparison groups). The authors argue that orientation and identity, not autism alone, structure who is most exposed. Limitation: self-selected online samples with a large age difference between groups, only 40 transgender participants in total, fixed-format response items, and an unusually high 45.6% non-heterosexual rate in the comparison group that limits how far the contrasts generalise.

9. Pecora LA, Hancock GI, Mesibov GB, Stokes MA (2019). Characterising the sexuality and sexual experiences of autistic females. Journal of Autism and Developmental Disorders, 49(12), 4834-4846. DOI 10.1007/s10803-019-04204-9. View the paper Compared 135 autistic females with 96 autistic males and 161 typically developing females on sexual functioning and experience. Autistic females reported less sexual interest yet more sexual experience than autistic males, and higher rates of regretted or unwanted sexual behaviour and unwanted advances than either comparison group, which the authors interpret as elevated exposure to negative sexual experiences including victimisation and abuse. Limitation: self-report questionnaires completed online by volunteers recruited through autism organisations, cross-sectional, with no measure of context, partner behaviour or the education participants had received.

13. Warrier V, Greenberg DM, Weir E, Buckingham C, Smith P, Lai MC, Allison C, Baron-Cohen S (2020). Elevated rates of autism, other neurodevelopmental and psychiatric diagnoses, and autistic traits in transgender and gender-diverse individuals. Nature Communications, 11, 3959. DOI 10.1038/s41467-020-17794-1. View the paper Analysis of five independent datasets totalling 641,860 people, comparing transgender and gender-diverse participants with cisgender participants. Transgender and gender-diverse individuals were between 3.03 and 6.36 times more likely to report an autism diagnosis after adjusting for age and educational attainment, scored higher on self-report measures of autistic traits, systemising and sensory sensitivity, and lower on empathy, with elevated rates of ADHD, depression, OCD, bipolar disorder, learning disorders and schizophrenia also reported. Limitation: internet-based convenience samples with self-reported diagnoses, likely under-representation of people with intellectual disability or severe mental illness, and inconsistent handling of intersex participants across datasets.

12. Weir E, Allison C, Baron-Cohen S (2021). The sexual health, orientation, and activity of autistic adolescents and adults. Autism Research, 14(11), 2342-2354. DOI 10.1002/aur.2604. View the paper Anonymous online survey of 2,386 adolescents and adults, comprising 1,183 autistic (746 female, 437 male) and 1,203 non-autistic participants. Autistic participants were 8.1 times more likely to identify as asexual (10.05% versus 1.51%), less likely to identify as heterosexual (63.03% versus 83.08%), and more likely to identify as transgender (12.60% versus 2.00%); autistic females were more likely to identify as homosexual (7.59% versus 2.44%) and autistic males as bisexual (9.40% versus 3.48%). Autistic participants also reported lower rates of sexual activity, and the authors conclude that sex education for autistic people remains inadequate. Limitation: a self-selected, English-language, predominantly white and UK-based online sample, with a comparison group recruited through autism-interested channels, so the percentages are not population prevalence estimates.

Clinical frameworks and position statements

15. Dewinter J, Giwa Onaiwu M, Massolo ML, Caplan R, Van Beneden E, Brormann N, Crehan ET, Croen LA, Faja S, Gassner DL, Graham Holmes L, Lai MC, Noens I, Pukki H, Stokes MA, Strang JF, van der Miesen AIR (2024). Short report: recommendations for education, clinical practice, research, and policy on promoting well-being in autistic youth and adults through a positive focus on sexuality and gender diversity. Autism, 28(3), 770-779. DOI 10.1177/13623613231188349. View the source Community-driven consensus project in which an international group of researchers, clinicians and autistic self-advocates drafted recommendations, then tested them in an online survey of 493 participants including 353 autistic adults. Eight of eleven recommendations reached consensus, defined as 80% of respondents rating them 8 or higher on a 0 to 10 scale, covering education about sexuality, relationships and gender diversity, accessible healthcare that combines gender-related expertise with victimisation prevention, and meaningful inclusion of autistic people in future research. Limitation: a short report rather than an evidence review, with a self-selected survey sample skewed towards people already engaged with autistic community organisations, and no data on whether implementing the recommendations improves outcomes.

Lived experience

3. Barnett JP, Maticka-Tyndale E (2015). Qualitative exploration of sexual experiences among adults on the autism spectrum: implications for sex education. Perspectives on Sexual and Reproductive Health, 47(4), 171-179. DOI 10.1363/47e5715. View the source Thematic analysis of 24 in-depth internet-facilitated interviews with self-identified autistic adults aged 18 to 61 in the United States, conducted between November 2012 and May 2013. Participants reported little extrafamilial support, lower rates of heterosexuality and gender conformity than the general population, and later romantic and sexual debut, with courtship difficulty and sensory dysregulation during partnered sex the two most common concerns. They described active solutions: sensory accommodations, planning the timing and method of sex, explicit negotiation with partners, and adopting scripts that depart from neurotypical expectations. Limitation: 24 self-identified rather than clinically diagnosed participants recruited from online autistic community spaces, all US-based and able to communicate in writing, so the accounts are rich but not representative.

4. Gray S, Kirby AV, Graham Holmes L (2021). Autistic narratives of sensory features, sexuality, and relationships. Autism in Adulthood, 3(3), 238-246. DOI 10.1089/aut.2020.0049. View the source Qualitative study drawing on the written words of more than 120 autistic people: 72 unique authors across five books and 13 online forums, plus 49 open-ended survey responses, analysed with deductive coding against Dunn's sensory model and inductive thematic analysis. Sensory features shaped sexual and relationship experience in both directions: some participants described touch, sight, sound and smell producing avoidance and distress despite wanting closeness, while others sought heightened sensation or found physical intimacy calming and regulating. Reported strategies included communicating with partners about sensory needs and, in some accounts, substance use. Limitation: relies on text already written for other purposes and on responses to a single survey question, with no demographic control over who was represented and no quantitative estimate of how common each pattern is.

5. Pearson A, Rees J, Forster S (2022). This was just how this friendship worked: experiences of interpersonal victimization among autistic adults. Autism in Adulthood, 4(2), 141-150. DOI 10.1089/aut.2021.0035. View the source Qualitative online study in which 43 autistic adults wrote about experiences of being victimised or exploited by people they knew, analysed with inductive thematic analysis from a critical realist perspective. Two themes emerged: cycles of victimisation, with many participants reporting polyvictimisation across several close relationships, and perceptions of victimisation, including difficulty trusting their own judgement, difficulty recognising abusive behaviour as it happened, and heightened compliance with unreasonable requests driven by fear or by avoidance of conflict. Limitation: written online accounts with minimal demographic information collected, no data on race or ethnicity, nothing recorded about the people who caused harm, and no comparison group.

Up next

Module 12 - ABTI-24 — The Traits That Make Autistic Burnout Likely

All modules in Autistic Self-Discovery

This is a normal thing to want help with. This course was built by clinicians who are part of the New Path Family. Intimacy, consent and sensory comfort are ordinary clinical territory, and talking them through with someone who works with autistic adults is a great deal more useful than a score from an unvalidated questionnaire. Therapy for clients in California and coaching worldwide, all by telehealth, are offered by our sister company New Path Family of Therapy Centers, Inc. A conversation costs nothing and there is no pressure. Saving this for later counts too. Talk with the New Path team

NEW PATH FAMILY