Mild Autism in Adults: When the Diagnosis Isn’t Believed

Mild autism in adults is a phrase that does a great deal of damage for something that sounds so gentle. It turns up in headlines about rising diagnosis rates, in family conversations after someone is identified at forty, and in the private doubt a person carries home from their own assessment. This post is for adults who were diagnosed autistic or ADHD later in life, and for the partners who pushed for the evaluation and are being told the result does not really count. If you are looking for the wider picture of what changes after a late identification, I have written about that separately in navigating life after a late-in-life autism diagnosis. This one is narrower. It is about the moment the diagnosis itself gets argued with.

A man came to see me two months after his assessment came back. I’ll call him Sean. He was forty-four, and his wife had spent three years reading, asking questions, and eventually booking the appointment he kept postponing. The report was thorough. It named things neither of them had words for. Then they told her mother over dinner, and her mother said, kindly, that Sean seemed like everyone else she knew, and that they were handing out those diagnoses to half the country. By the time they reached my office, Sean had stopped mentioning the report, and his wife had started carrying it in her bag.

In this blog post, I want to look at what the new research actually found, because a study published in September 2026 is being summarized in a way that will reach your kitchen table this month. Then I want to look at what the word “mild” does and does not measure. And then at four situations couples land in during the first year after a diagnosis, each of which I see often enough to recognize on sight. None of them mean the relationship is failing. All of them are easier to move through when you can name them.

3 Key Takeaways

• The September 2026 study measured who is being diagnosed, not how much anything costs the person diagnosed. It did not measure severity at all, and its authors and reviewers said so plainly.
• “Mild” is a description of how a person looks from the outside. It has never been a reliable guide to what a day actually takes out of them.
• A diagnosis does not resolve the arguments a couple was already having. It re-opens them with new language, which is a different and more workable problem.

What the New Study Measured, and What It Did Not

On September 9, 2026, a large Danish registry study was published in JAMA Psychiatry (Tarp et al., 2026). The researchers looked at 71,317 children and adolescents diagnosed with ADHD or autism between 2012 and 2022, each matched against ten peers without a diagnosis, and asked whether the characteristics that preceded a diagnosis had changed across that decade. They found that they had. People diagnosed more recently looked more like the general population beforehand than people diagnosed earlier did. The authors’ own conclusion was careful: rising diagnostic rates may reflect changes in diagnostic practice or in health care capacity rather than changes in underlying risk alone.

That is a finding about the gate, not about the people walking through it. It tells you that the criteria and the capacity to apply them have shifted, which most clinicians would already tell you. It does not tell you that anyone is less affected.

The independent experts who reviewed the study for the UK’s Science Media Centre made exactly that point (Science Media Centre, 2026). Steven Kapp noted that the study “did not measure the degree of core traits (‘symptoms’) or disability.” Uta Frith, who called it a sound population study, said that confirming an increase in mild cases would require measures of symptom severity, which this paper does not have. So the interpretation now travelling through headlines is the one the researchers and the reviewers both said the data cannot support.

There is a second thing worth knowing if you are the one being argued with. Every person in that study was diagnosed before the age of eighteen. The median age at diagnosis was 11.4 years. Adults identified at thirty-five, forty-four, or sixty-one are not in the dataset at all. A study of children says nothing about a man diagnosed at forty-four, and it certainly says nothing about what his Tuesday evenings cost him.

What “Mild” Measures, and What It Does Not

The word “mild autism in adults” is doing a job it was never built for. In ordinary use it means this person seems to be managing. What it actually describes is how legible someone’s difficulty is to the people watching.

This is not a new observation, and it has been studied directly. Research on autistic children at the point of diagnosis found that intelligence quotient was a weak predictor of adaptive functioning, and that for children without intellectual disability, adaptive scores sat significantly below their measured intelligence, with the gap widening as they got older (Alvares et al., 2020). The authors’ conclusion was that “high functioning” is a misnomer. Capability on a test does not tell you what a person can sustain in a life.

What I notice with clients is that the people described as mild are frequently the ones who have paid the most for the description. Masking is expensive, and it is most expensive for the people who do it well. A man who can hold a meeting together for six hours and then cannot decide what to eat is not mildly affected. He is precisely affected — in a way that happens to be invisible to everyone who saw him at the meeting.

So when a family member says he seems like everyone else, they may be entirely sincere and entirely correct about what they have seen. They have seen the performance. The bill arrives later, at home, and usually in front of one person.

When the Diagnosis Becomes a Verdict on Who Was Right

This is the most common one, and the one couples are least likely to say out loud.

For years, one partner has been naming a pattern and the other has been experiencing that naming as criticism. Then a report arrives with a clinician’s signature on it. In the first weeks, it is very hard for that report not to land as a scoreboard. One partner hears I was right, and now a professional agrees with me. The other hears I have been the problem this whole time, and now it is official.

Neither of those is what the assessment said. But both are what the assessment can feel like, and a feeling that strong does not wait for accuracy.

What helps is separating two questions that have become one question. What is happening between us is a question about two nervous systems. Who has been right about it is a question about the last ten years of arguments, and it does not have a useful answer. Damian Milton’s account of the double empathy problem is the clearest framing I know of here: the misunderstanding between an autistic person and a non-autistic person is mutual, a two-way failure of shared reference rather than one person’s deficit (Milton, 2012). Nobody was right. Both people were reading a language they had not been taught.

In a session, this often sounds like moving from “you never told me” to “I said it in a way you could not hear, and you heard it in a way I did not mean.” That is not a softer version of the same sentence. It is a different sentence, and it has somewhere to go.

When the Diagnosis Becomes the Explanation for Everything

The second pattern usually arrives around month three, once the relief has settled and the newly diagnosed partner has read a great deal.

Suddenly there is a reason for everything. The missed anniversary is executive function. The flat tone during an argument is alexithymia. The three-hour silence after the family lunch is a shutdown. Some of that is accurate and long overdue. Some of it is a person who has spent decades with no framework at all now holding one framework and using it on every surface in the house.

The cost lands on the other partner, and it lands quietly. If everything has a neurological explanation, there is nothing left to ask for. She stops raising things, not because she has stopped minding, but because every raise now produces an explanation rather than a response.

Understanding a behavior and being finished with it are different stages. A shutdown after a family lunch is real, and it is also something a couple can plan around: leaving earlier, arriving separately, agreeing in advance on the sentence that signals it is time to go. You might notice which conversations in your house end in an explanation, and whether either of you has been able to ask for anything since.

When the Diagnosis Becomes a Reason to Stop Asking

This one is the mirror of the last, and it comes from the other direction.

The non-autistic partner reads everything she can find, decides the accommodating one has to be her, and starts pre-absorbing. She handles the phone calls. She manages the logistics. She stops mentioning the things that used to bother her, because she has learned that they are harder for him than she realized, and it feels unkind to add to the load.

She is not wrong about any of that. She is also, six months in, doing more than before with less acknowledgment — and quietly furious in a way she would not describe as furious.

Affirming a partner’s neurotype is not the same as removing every expectation from them. A diagnosis explains why something is hard. It does not settle who does it, and it never means one adult holds the whole relationship while the other is understood. Accountability and compassion are not in competition here. The useful question is not is this his autism but what does this specific thing need, and who is actually able to do it this week.

When the Diagnosis Has to Keep Being Defended

And then there is the version Sean and his wife walked in with, which is the one the current headlines will feed.

Someone outside the relationship does not accept the diagnosis. A mother, a brother, a friend who has known him for twenty years. They are not being cruel. They are reporting accurately on a person who has masked in front of them for two decades, and they now have a news story that seems to confirm their reading.

The trap is that the couple starts producing evidence. Anecdotes get gathered. The report comes to dinner. And the diagnosis, which was supposed to be a piece of information the two of them could use, becomes a case they are permanently prosecuting in front of an audience that was never going to be the jury.

Here is the thing I say most often about this. You do not need your family to believe the diagnosis in order to use it. The report is for the two of you. Its job is to change how you plan a Sunday, not to change your mother-in-law’s mind. Some people will come around slowly, by watching the relationship get easier rather than by reading anything. Some will not. Neither outcome changes what the assessment found, and neither is a referendum on whether it was worth doing.

For Sean and his wife, the change was small and not dramatic. She stopped carrying the report. They stopped bringing it up at dinner. What they brought instead was a plan for leaving family gatherings at a reasonable hour, which nobody needed to approve.

What the Diagnosis Is Actually For

A diagnosis is not a verdict, an excuse, a full explanation, or a claim you have to win in public. It is a piece of information about how one nervous system is built, arriving late, into a relationship that formed without it.

That is a smaller thing than it sounds like in the first month and a larger thing than it sounds like in the second year. The couples I see do best with it are not the ones who agree about what it means. They are the ones who stop asking the diagnosis to settle arguments it was never going to settle, and start using it for the thing it is good at: predicting what a given Tuesday will cost, and planning accordingly.

The word “mild” will keep appearing in headlines. It describes a population. It has never described an evening.

If you are in the first year after a diagnosis and it is re-opening more than it resolves, that is ordinary, and it is workable with some help. You can Book a FREE 20-minute “Clarity & Connection” Zoom session and we can talk about what is happening in your house. No preparation needed.

If you have been the one holding the file, booking the appointment, and defending the result to people who were not in the room, there is a free guide for that.

Get the free guide: You’re Not Too Much. You’re Just Alone in the Repair. Eight pages, sent straight to your inbox.

Written by Barbara (Blaze) Lazarony, MA, LMFT #151788, who works with neurodivergent adults and neurodiverse couples.

Client stories referenced in this post are composites, with identifying details changed to protect privacy.

Frequently Asked Questions

Does a “mild autism in adults” diagnosis mean less support is needed?

No. Descriptors like mild and high-functioning track how visible a person’s difficulty is to observers, not how much a day costs them. In my practice, the people described that way are often the ones carrying the highest masking load, because they have become skilled at hiding the effort. Support should be matched to what a person can sustain, not to how well they present.

My partner was diagnosed and now attributes everything to it. Is that normal?

It is very common in the first year, and it usually settles. After decades with no framework, having one is a relief, and a new framework tends to get applied everywhere before it finds its proportions. What matters is whether you can still raise something and get a response rather than an explanation. If you cannot, that is worth naming directly and early.

My family does not believe the diagnosis. What do I do?

You are allowed to stop making the case. The assessment exists so that the two of you can plan differently, and it works whether or not anyone else accepts it. Some family members change their minds slowly, by noticing that things have got easier. You do not have to produce evidence at dinner in the meantime.

Does a diagnosis fix the arguments we were already having?

Not directly, and expecting it to is where a lot of first-year disappointment comes from. What it does is change the material the arguments are made of, which makes them more workable. The same fight about the missed anniversary becomes a conversation about reminders, transitions and capacity, and that one has somewhere to go.

Is it worth getting assessed as an adult if the traits are not severe?

That is a personal decision, and severity is a poor basis for making it. The more useful question is whether having language for how you are built would change how you set up your life, your work and your relationship. For some people it changes a great deal. For others it confirms something they had already arranged around, and that is a reasonable outcome too.

References

Alvares, G. A., Bebbington, K., Cleary, D., Evans, K., Glasson, E. J., Maybery, M. T., Pillar, S., Uljarević, M., Varcin, K., Wray, J., & Whitehouse, A. J. O. (2020). The misnomer of ‘high functioning autism’: Intelligence is an imprecise predictor of functional abilities at diagnosis. Autism, 24(1), 221–232. https://doi.org/10.1177/1362361319852831

Milton, D. E. M. (2012). On the ontological status of autism: The ‘double empathy problem’. Disability & Society, 27(6), 883–887. https://doi.org/10.1080/09687599.2012.710008

Science Media Centre. (2026, September 9). Expert reaction to study of changes in characteristics associated with ADHD and autism spectrum disorder diagnoses over time. https://sciencemediacentre.org/expert-reaction-to-study-of-changes-in-characteristics-associated-with-adhd-and-autism-spectrum-disorder-diagnoses-over-time/

Tarp, M. E., Lousdal, M. L., Rask, C. U., Keyes, K. M., Formánek, T., Brikell, I., Liu, X., Maseras, G. T., & Plana-Ripoll, O. (2026). Changes in characteristics associated with ADHD and ASD diagnoses over time. JAMA Psychiatry. https://doi.org/10.1001/jamapsychiatry.2026.2722

Intention vs. Impact: When You Stop Explaining Yourself

Intention vs. impact in relationships is usually taught as a correction: what you meant matters less than how it landed. For many neurodivergent adults and neurodiverse couples, that lesson arrives so often, and so one-sidedly, that something quietly gives way. Explaining what you meant stops feeling worth the breath. What replaces it is not anger and not indifference. It is a kind of flatness — a private conclusion that meaning does not travel, so there is no reason to send it.

I’ll call her Thandiwe. She is forty-three, identified autistic at thirty-nine, and she came to a session able to describe the pattern with unsettling precision. Her wife would tell her that something she said had hurt. Thandiwe would explain what she had actually meant. Her wife would say that the explanation did not change how it felt. Both were right, and both were tired. By the time she sat down with me, Thandiwe had stopped offering the explanation at all. She described it without self-pity, the way you would describe a road that had washed out: it used to go somewhere, and it does not anymore.

In this blog post, I want to look at the belief sitting underneath that flatness, because it is a belief and not a personality. I will walk through what one well-known study actually found about intention and outcome in moral judgment, including what it did not find, and why the way that research gets summarized has done real damage. I will look at how the same gap runs in both directions in a neurodiverse relationship, not just one. And I will describe what tends to replace “why bother” when it loosens, which is not optimism and not more explaining.

3 Key Takeaways

• The exhaustion people describe as not caring is usually a belief that has hardened: explaining what I meant does not change anything, so there is no point saying it.
• Research on intention and outcome in moral judgment describes small average differences between groups on a rating task. It does not describe a person, and it has often been summarized in ways the findings do not support.
• The gap runs both ways. When each partner is being judged by the part of the exchange they have least access to, both eventually stop trying to bridge it.

The Belief Underneath “Why Bother”

When someone tells me they have stopped explaining themselves, they usually offer it as evidence of a character change. They say they have gotten colder, or lazier, or that they finally accepted reality. What I notice with clients is that something much more specific has happened. A belief has formed, it has been tested repeatedly, and it has held up every time.

The belief is this: what I intended is not admissible. Only the outcome counts. And if only the outcome counts, then explaining my intention is not communication, it is just noise I am adding to a situation that is already bad.

That belief is not irrational. It is what you would conclude from the evidence if the evidence kept pointing the same way. Many people describe years of a particular exchange: something goes wrong, they explain what they meant, and the explanation is received as an excuse, a deflection, or a refusal to take responsibility. After enough repetitions, the cost of explaining exceeds the return, and a reasonable person stops paying it aka saying it!

What makes it so hard to see from the outside is that the person who has stopped explaining looks like the person who has stopped caring. Those two states produce nearly identical behavior and come from opposite places. One is absence of investment. The other is investment that has been spent and not repaid.

What the Research Actually Found About Intention and Impact

There is a study people reach for here, usually secondhand, and it is worth handling carefully because the careless version of it is doing harm.

Moran and colleagues (2011) ran two experiments. In the first, autistic adults and non-autistic adults answered questions about other people’s false beliefs, and the two groups performed the same: no difference in accuracy, no difference in speed. In the second, participants read short scenarios and rated how morally permissible an action was on a seven-point scale. The scenarios varied two things independently: whether the person intended harm, and whether harm actually occurred.

On three of the four scenario types, the groups did not differ. Both condemned deliberate harm. Both were untroubled by neutral acts. Both took a middle position on attempted harm that failed. The difference showed up in one place: accidental harm, where someone with an innocent intention caused a bad outcome. The autistic group rated those actions as less morally permissible than the comparison group did. Put plainly, they were less willing to let the innocent intention outweigh the bad result.

Here is what that finding is, and it matters. It is an average difference between thirteen autistic adults recruited through one regional organization and fifteen non-autistic adults recruited from a university community, on a written rating task, in 2011. Both groups had above-average measured IQ. It is a group average on an artificial instrument, not a property of a person, and it tells you nothing about which side of that average any individual falls on.

The paper is titled “Impaired theory of mind for moral judgment in high-functioning autism.” I want to name that, because reading yourself described that way is itself one more impact-only verdict. The authors measured a difference in how two small groups weighted two kinds of information. The word chosen to describe that difference was impaired. If you have ever wondered where the belief comes from that your intentions are not admissible, some of it comes from here, from the language the literature uses about you before anyone has met you. (Gross, right? )

The Same Gap, Running the Other Direction

A study like that one describes autistic participants weighing someone else’s innocent intention less heavily. Ordinary life in a neurodiverse relationship delivers the mirror image constantly: the autistic partner is the one whose innocent intention is not being counted.

Milton (2012) named this mutual version the double empathy problem. His argument is that the breakdown between autistic and non-autistic people is not one person failing to read another. It is two people with genuinely different ways of processing and expressing experience, each struggling to read the other, with the difficulty distributed across both of them rather than located in one.

That reframe changes what the “why bother” conclusion means. If only one person has a gap, then the fix is for that person to try harder, and giving up is a failure of effort. If the gap is between two people, then both partners are being asked to do something difficult, both are getting it wrong at a similar rate, and both are quietly building the same conclusion on their own side of it.

In my practice, I often see both halves of that in the same room. The autistic partner has stopped explaining what they meant, because it has never once changed the verdict. The non-autistic partner has stopped explaining how something landed, because naming it has only ever produced a defense of intent. Each has concluded that the other cannot receive what they are sending. Each is, on the available evidence, right. What neither can see is that the other one has reached the same conclusion, in the same way, about them.

Where the Belief Comes From

For late-identified adults, this rarely starts in the relationship. It usually arrives already formed.

Leedham and colleagues (2020) interviewed women diagnosed autistic in middle to late adulthood about what the diagnosis changed. What they described was not primarily relief about a label. It was a long, difficult renegotiation of a life story that had been built out of self-criticism, and a slow movement toward something more like self-compassion. That renegotiation is hard precisely because the old story is coherent. Decades of being told that what you meant was beside the point produce a serviceable explanation: I must be the kind of person whose meaning does not matter.

What that looks like by forty is not a wound. It is an operating principle. It is efficient. It saves energy. It prevents a specific humiliation, which is offering an explanation and having it refused.

And it generalizes. People who have stopped explaining what they meant in conflict also tend to stop explaining what they want, what they are looking forward to, what they noticed, what they are afraid of. The channel that carried intention was the same channel that carried everything interior. When it closes, it closes for all of it. That is usually what the partner is reacting to when they say the relationship feels empty. They are not missing the apologies. They are missing everything that used to come through the same door.

What “Why Bother” Is Protecting

I want to be careful not to treat this as a thinking error to be corrected. It is not distorted thinking. It is accurate thinking about a real pattern, and it is doing a job.

Withdrawing the explanation protects against a particular experience: offering the most honest thing you have, watching it be treated as a maneuver, and having no move left. That experience is genuinely costly, and avoiding it is not cowardice. Many people describe the moment they stopped as a relief.

It also protects the relationship in a narrow sense. A great many people stop explaining because explaining had started making things worse, and stopping reduced the number of bad nights. That is a real result, and anyone treating it as avoidance has not sat with what the alternative was costing.

The trouble is the price. The protection works by removing the possibility of being understood — not just the possibility of being misread. And it is invisible to the other person, who does not experience a partner who has protectively withdrawn. They experience a partner who has gone quiet, and they draw their own conclusion from the outcome, which is exactly the move that started all of this.

What Replaces the Belief

The shift here is not from “why bother” to “keep explaining.” Going back to the thing that failed, with more effort, is not a plan.

Why This Happens

The belief formed because intention and impact were being treated as competing claims, with one of them required to win. In that frame, an explanation of intent can only function as a rebuttal to the report of impact, which is why it keeps landing as a defense no matter how gently it is offered. The frame is what is broken, not the delivery.

What This Can Look Like

The belief loosens when the two stop being tried against each other. In practice that often means saying both, in one breath, with no “but” between them. Something closer to: that landed badly and I can see it, and separately, here is what was happening on my end. Not as a defense. As the other half of the information.

Some couples find it easier to split them in time rather than in a sentence. Impact gets named and received on its own, with nothing asked of the other person yet. Intent gets offered later, in a calmer hour, when it is not competing with the injury. Others find the reverse works better. What matters is that neither one is being submitted for a ruling.

What Helps

You might notice which direction is currently closed on your side. Many people can do one and not the other. Knowing which is which is more useful than trying to do both at once.

One option is to lower what an explanation is for. If the purpose is to change the verdict, it will keep failing. If the purpose is only that the other person has the information, it can succeed even on a night when nothing is resolved.

It can also help to say the conclusion out loud rather than acting on it silently. “I stopped explaining because it never changed anything” is information a partner has almost certainly never been given. It is frequently the first thing that lands in years, because it is a report of impact, which is the language the relationship already speaks.

None of this is a sequence to work through, and it is reasonable to try a version of it, find it does not fit, and adapt it. If an approach does not work, that is information about the approach, not about how hard you tried.

Holding Both Things at Once

What I keep coming back to with couples in this place is that nobody in the room has behaved unreasonably. One person concluded that their intentions were not admissible, from evidence. The other concluded that their experience was always going to be argued with, from evidence. Both conclusions were earned. Both are now generating the exact behavior that confirms the other.

That is a better position than it sounds, because a conclusion drawn from evidence can be revised by evidence. It does not require anyone to have been wrong about the past. It requires a few exchanges that go differently, which is slower than a decision and considerably more durable.

It is also worth saying that this can take a while, and that going at it at the speed of your nervous system is not a delay tactic. A belief built over thirty years does not come apart in an evening, and it should not have to.

If you would like support with this, individual and couples work at LOTAS is a place to look at the pattern without either of you having to be the one who was wrong. You can Book a FREE 20-minute “Clarity & Connection” Zoom session and we can talk about what would actually help.

If you have been the one explaining the pattern for years and have started to wonder why you keep bothering, there is a free guide written for exactly that.

Get the free guide: You’re Not Too Much. You’re Just Alone in the Repair. Eight pages, sent straight to your inbox.

Written by Barbara (Blaze) Lazarony, MA, LMFT #151788, who works with neurodivergent adults and neurodiverse couples.

Client stories referenced in this post are composites, with identifying details changed to protect privacy.

Frequently Asked Questions

Does this mean autistic people do not care about other people’s intentions?

No, and the research does not support that reading. In the study most often cited here, autistic and non-autistic participants judged deliberate harm, attempted harm and neutral acts the same way. One scenario type produced a group-average difference on a rating scale. A group average on an artificial task does not describe an individual, and it does not describe caring.

My partner says explaining what he meant is just making excuses. Is he making excuses?

Sometimes an explanation of intent is a way of avoiding responsibility, and sometimes it is a genuine attempt to hand you information you do not otherwise have. The difference is usually whether the impact has been received first. If it has, the explanation is the other half of the picture. If it hasn’t, it will land as a rebuttal regardless of what you meant.

I have stopped explaining myself, and I do not want to start again. Is that a problem?

Not on its own. The question worth asking is whether the same channel was carrying other things: what you wanted, what you noticed, what you were looking forward to. If it closed for all of that too, the cost is probably higher than the protection is worth. If it closed only around conflict, that may be a reasonable arrangement.

We have both stopped. Where does anyone start?

Usually with the one who is closer to the surface, which is not always the one who is more hurt. Naming the conclusion itself tends to be the least risky opening, because it reports an experience rather than requesting a change. “I stopped because it never worked” asks nothing of the other person and is almost always new information.

Is this something therapy can actually change?

It can change, though not quickly, and the change is usually cumulative rather than a breakthrough. What tends to help is a setting where impact and intent do not have to compete, repeated enough times that the old conclusion stops being the most reasonable one available.

References

Leedham, A., Thompson, A. R., Smith, R., & Freeth, M. (2020). “I was exhausted trying to figure it out”: The experiences of females receiving an autism diagnosis in middle to late adulthood. Autism, 24(1), 135–146. https://doi.org/10.1177/1362361319853442

Milton, D. E. M. (2012). On the ontological status of autism: The ‘double empathy problem’. Disability & Society, 27(6), 883–887. https://doi.org/10.1080/09687599.2012.710008

Moran, J. M., Young, L. L., Saxe, R., Lee, S. M., O’Young, D., Mavros, P. L., & Gabrieli, J. D. (2011). Impaired theory of mind for moral judgment in high-functioning autism. Proceedings of the National Academy of Sciences, 108(7), 2688–2692. https://doi.org/10.1073/pnas.1011734108

Stanford Neurodiversity Summit: What Research Misses

Update: my poster from the Summit, The Neurodivergent Spiral of Life, is now online.

The Stanford Neurodiversity Summit runs on September 19, 20 and 21, 2026, and I will be there presenting a poster. If you are autistic, ADHD or AuDHD, or you are partnered with someone who is, a neurodiversity research conference can feel like a room where you are the subject and not one of the people in the conversation. That feeling is worth taking seriously. It also points at something real about how autism research and relationship research get made, and about what tends to fall out of them somewhere between the study and a couple at their own kitchen table.

A woman I will call Zuzanna came to a first session carrying a folder. Printed studies, tabbed and highlighted, four years of reading about autistic adults done at her kitchen table after her husband had gone to bed. She could define alexithymia. She could explain executive function better than some clinicians I have worked with. And near the end of that first hour she said that none of it had told her what to do on a Tuesday night when he goes quiet after dinner and she cannot tell whether he is resting or leaving. The reading had given her a vocabulary. It had not given her a Tuesday.

In this blog post I want to describe what a poster session actually is, why a working couples therapist stands at one, and where the distance opens between what gets published and what gets lived. I will look at two places where autism research changed direction because autistic people insisted that it change. I will be straight with you about what a study can and cannot tell you about your own relationship. And I will say plainly what I am hoping to carry home from Stanford.

3 Key Takeaways

• A poster session is a conversation, not a lecture. It is the part of a research conference where the person who did the work stands beside it and answers questions, which makes it one of the few academic formats a non-researcher can walk into and use.
• Research describes groups. Your relationship is one particular pair of people on one particular evening. Both are real, and neither one replaces the other.
• Several of the most useful ideas in autism research began as things autistic people said about their own lives long before anyone measured them.

What a Poster Session Actually Is

If you have never been to an academic conference, the word “poster” is misleading. It is not a display board that people walk past. It is a format. A few square feet of printed argument, with the person who made it standing beside it for an hour, waiting to be asked about it.

Nobody presents from a stage. There is no slide deck and no question card. You stand there, and strangers stop, read for thirty seconds, and either walk on or start asking. Sometimes the person asking is a graduate student. Sometimes it is an autistic adult who is speaking at the summit themselves and wants to know why you framed something the way you did. Those are usually the most useful conversations of the day, and they are the ones that change how the work is written up afterwards.

This is the seventh annual Stanford Neurodiversity Summit, hosted by the Stanford Center for Neurodiversity and Human Potential and directed by Lawrence Fung, MD, PhD. The theme this year is “Building Strengths through Neurodiversity Innovation.” It runs as a hybrid event, in person at Stanford and online, and the room it sets out to assemble is deliberately mixed: neurodivergent individuals, employers, service agencies, educators and students, parents, and professionals from across the field. It is being planned with Universal Design in mind, which means accessibility is built into the structure rather than requested afterwards.

That structure matters more than it sounds. Most research conferences are built for researchers. This one is built on the premise that all of those people belong in the same rooms and that the exchange runs in every direction. Its stated priority this year is strengths-based, collaborative approaches between neurodivergent and neurotypical individuals, and it says plainly that the aim is action rather than discussion alone. Read that sentence again and it is a description of a couples session. Two people, wired differently, working out how to build something together rather than one of them being corrected into the shape of the other.

It is also explicit about how it wants people to talk to each other. It describes itself as a place where attendees listen to each other especially when their views are not the same, and it calls that active listening an act of growing together as a community rather than only a courtesy. Questions can be submitted to speakers before and during the sessions. And it says it will collect the community’s views on what research and community-based services should be prioritised, and on how they could be designed better. That last one is an open door, and it is why I am going with something to say and not only something to hear.

The keynote list shows what the mixing looks like in practice. Dr. Lynn Koegel is a clinical professor in Stanford’s own department of psychiatry and behavioral sciences. Dr. Eva Loth is a professor of cognitive neuroscience at King’s College London. Dr. Vikram Jaswal is a developmental psychologist and professor of psychology at the University of Virginia. Hari Srinivasan is a PhD candidate in neuroscience and an NSF Graduate Research Fellow. A clinician, a cognitive neuroscientist, a developmental psychologist and a doctoral researcher, on one programme, alongside employers and educators and parents, is not an accident of scheduling. It is the argument of this post arranged as a running order.

The Distance Between a Finding and a Tuesday Night

Here is the thing Zuzanna had run into, and it is not a failure of her reading.

Research describes groups. It says something about what tends to be true across a sample of people, measured on a particular instrument, compared against a particular comparison group. That is a real and valuable kind of knowledge. It is also, structurally, not about you. A finding that holds for most autistic adults in a study says nothing certain about the autistic adult sitting across the table from you, who is also a middle child, who was raised by a mother who never stopped talking, who has a job he cannot leave, and who is very tired on a Tuesday.

In my practice, I often see people arrive holding an enormous amount of accurate general information and almost no traction. They know the concepts. They cannot yet do anything with them at 9pm. The gap between those two states is not a knowledge gap, and reading more rarely closes it.

What closes it, usually, is specificity. Not “autistic people often need recovery time after social demand,” but “when you come home from the office you need forty minutes before you can be spoken to, and she needs to know that the forty minutes is not about her.” The general finding is the raw material. The particular arrangement — in a particular house, between two particular nervous systems — is the work.

I go to research conferences because the raw material matters and I want it to be good. I go to a poster session in particular because it is where the raw material gets argued with, up close, by the people it describes.

When the Research Described the Room I Was Already Sitting In

For a long time, the standard framing of autistic social difficulty put the deficit in one person. One partner was described as having impaired social communication, and the other was, by implication, communicating fine.

That framing does not survive contact with a couples room. What I watch happen, week after week, is two people who are each making entirely coherent sense within their own frame of reference and are each landing wide of the other. He answers the question she asked rather than the one she meant. She reads his flat delivery as indifference. He reads her tears as an accusation. Neither of them is failing at communication in general. They are failing at each other, specifically, and in both directions.

Milton (2012) named this the double empathy problem: a mismatch between two people with different ways of experiencing the world, where the breakdown in understanding runs both ways rather than sitting inside one person. It is a theoretical paper, not a trial, and it was written by an autistic academic describing something autistic people had been saying for years.

I remember reading it and recognizing my own consulting room in it. It was not that I learned something new. I was handed language for a pattern I had been watching for years and describing badly. That is what a good idea does. It does not tell you something you have never seen. It tells you what you have been looking at.

The practical consequence in the room is large. If the problem sits in one person, the work is remediation, and one partner becomes the project. If the problem sits between two people, the work is translation, and both of them have something to learn and something to be understood about. Almost everything I do with couples follows from that second reading.

What Gets Counted, and What Does Not

A study can only find what its design allows it to see. This is not a criticism; it is simply how measurement works. But it has a consequence worth understanding, because it explains a lot of the strangeness in the older literature.

For decades, autistic social ability was measured in mixed pairs: an autistic person interacting with a non-autistic person, with the non-autistic person’s comprehension treated as the standard. Under that design, autistic communication looks impaired more or less by definition, because the design has already decided whose understanding counts.

Crompton and colleagues (2020) changed the design. They used a chain task, passing information from person to person, and compared three kinds of chains: all autistic, all non-autistic, and mixed. Information passed along the all-autistic chains as effectively as along the all-non-autistic chains. It was the mixed chains that lost detail.

That finding does not say autistic people communicate better, and it does not say the difficulty is imaginary. It says the difficulty is located in the crossing, not in one of the two sides. That is close to what Zuzanna’s husband had been trying to tell her for years. It is also the thing the older designs were built in a way that could not see.

I find this useful to say out loud to couples, because it moves the question from “which of us is doing this wrong” to “what happens at the join.” The second question has answers. The first one mostly produces defensiveness.

How to Read Research About Yourself Without Disappearing Into It

Most of the people I work with have already read a great deal before they arrive, and a fair amount of it has done them harm. Not because the studies were bad, but because of what happens when a person reads a description of a group and files themselves inside it wholesale.

A few things I say fairly often, offered here as options rather than instructions.

Notice the comparison group

When a paper reports a rate, ask what it was compared against and who was in the sample. A figure drawn from clinic-based studies of adults says nothing reliable about children, or about people who never reached a clinic. Rates travel badly, and they get quoted far outside the population they came from.

Separate the finding from the frame

The measurement can be sound and the interpretation still carry an old assumption. It is entirely reasonable to accept what a study found and decline the sentence the discussion section wrapped around it.

Let a description be partial

Recognising yourself in something is not the same as being fully accounted for by it. You are allowed to say “that is three-quarters right and the last quarter is not me at all.” In my experience that sentence is a sign of good reading rather than resistance.

Keep hold of where you actually are

The map I use with couples, the Neurodivergent Spiral of Life, exists partly for this reason. A person can enter it at any layer, and knowing which layer you are standing on tends to be more useful than knowing everything about all seven. Reading widely without knowing where you are is how people end up exhausted and no clearer.

What I Am Bringing to Stanford, and What I Hope to Carry Home

I am a clinician, not a laboratory researcher. What I have is a great many hours in rooms with neurodivergent adults and neurodiverse couples, and a stubborn interest as a neurodivergent person myself, in the part of the picture that the literature has less to say about: not diagnosis, not childhood, but two adults trying to stay in contact with each other over years.

The summit names where it wants neurodivergent strengths applied: education, employment, technology, healthcare, and entrepreneurship. Every one of those is a place a person goes. None of them is the person waiting at home when they get back. I do not offer that as a complaint about the programme, and it would be a strange complaint to bring to a summit that asks its attendees what ought to be prioritised next. I offer it because they asked. A great deal of the cost of being chronically misread lands in a marriage, and that room gets less attention in these conversations than the workplace does.

That is what I want to put in front of people at a poster board. Not because clinical observation outranks a controlled study. It does not. But it sees different things, and it sees them earlier. Patterns show up in a caseload well before anyone has an instrument for them. If nobody carries those patterns into the room where the instruments get designed, the gap between the literature and the kitchen table stays where it is.

One of this year’s keynotes is Janet Lawson, an LMFT and the co-founder and CEO of Autistry Studios. I notice that because it is the same license I hold. A marriage and family therapist on the keynote list at a summit like this one is, in itself, a small version of the argument I am making.

What I hope to carry home is the other half of that exchange. Researchers who can tell me where my thinking is loose. Autistic and ADHD attendees who will tell me, without much ceremony, where a framing lands badly. Three days of listening to people who are working on this from angles I do not have.

I will write about what I learn there. Some of it will change how I explain things in sessions, which is the only real test of whether a conference was worth attending.

Holding Both Things at Once

The research is not going to describe your marriage. It was never going to. It describes tendencies across groups of people, most of whom are not you, measured in ways that could not have captured a Tuesday night in your kitchen.

And it still matters, because the frame the research uses becomes the frame your therapist uses, and your doctor, and eventually your mother-in-law. When Milton wrote that the misunderstanding runs both ways, that sentence eventually reaches a couples room and changes who gets treated as the problem in it. Ideas travel. It is worth caring which ones do.

So both things are true. Nothing published this year is going to tell you what to do when he goes quiet after dinner. And it is still worth having people in the room where those ideas are argued about who have sat with a thousand Tuesdays.

If you would like to talk about what any of this looks like in your own relationship, you can Book a FREE 20-minute “Clarity & Connection” Zoom session. It is a conversation, not an intake, and there is no obligation attached to it.

If you have been the one reading everything, quietly, after everyone else has gone to bed, there is a free guide written for exactly that.

Get the free guide: You’re Not Too Much. You’re Just Alone in the Repair. Eight pages, sent straight to your inbox. Read it here.

I work with neurodivergent adults and neurodiverse couples, and I will be at the Stanford Neurodiversity Summit this September with a poster and a great many questions. Barbara (Blaze) Lazarony, MA, LMFT #151788.

Client stories referenced in this post are composites, with identifying details changed to protect privacy.

Frequently Asked Questions

Do I have to be a researcher or a clinician to attend the Stanford Neurodiversity Summit?

No. This is the seventh annual summit and it is built for a deliberately mixed audience: neurodivergent individuals, employers, service agencies, educators and students, parents, and professionals from across the field. The organisers are planning it with Universal Design in mind, so accessibility is part of the structure rather than something you have to ask for. If you have wanted to see how this field talks about itself, it is an unusually accessible place to do it.

What is a poster session, in plain terms?

It is a set of printed research summaries on boards in a large room, each with its author standing beside it for a couple of hours. You read what interests you and ask the person questions directly. There is no audience, no microphone and no formal Q&A, which for a lot of people makes it far easier than a lecture hall.

Is the summit only in person?

No. It runs as a hybrid event across September 19, 20 and 21, 2026, in person at Stanford and online, with a virtual registration tier alongside the in-person and professional ones. The virtual option exists so that cost and travel are not the thing that decides who gets to be in the conversation.

Does autism research apply to me if I am self-identified rather than formally diagnosed?

Largely yes, with a caveat worth knowing. Most published samples are made up of formally diagnosed people, which skews towards those who had the access, money and recognisable presentation to reach a clinic. That means the literature under-describes late-identified adults, women, and people of colour in particular. Read it as partial rather than as not about you.

Why does it matter that some findings started with autistic people rather than researchers?

Because it changes what gets studied at all. Autistic burnout is the clearest example: Raymaker and colleagues (2020) built their definition from autistic adults’ own accounts, and it entered the literature as a community-defined concept rather than a clinical one. It is not a formal diagnosis, and autistic people consistently describe it as distinct from depression. Without the community naming it first, there would have been nothing for anyone to measure.

References

Crompton, C. J., Ropar, D., Evans-Williams, C. V. M., Flynn, E. G., & Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704–1712. https://doi.org/10.1177/1362361320919286

Milton, D. E. M. (2012). On the ontological status of autism: The “double empathy problem.” Disability & Society, 27(6), 883–887. https://doi.org/10.1080/09687599.2012.710008

Raymaker, D. M., Teo, A. R., Steckler, N. A., Lentz, B., Scharer, M. L., Delos Santos, A., Kapp, S. K., Hunter, M., Joyce, A., & Nicolaidis, C. (2020). “Having all of your internal resources exhausted beyond measure and being left with no clean-up crew”: Defining autistic burnout. Autism in Adulthood, 2(2), 132–143. https://doi.org/10.1089/aut.2019.0079