A claim that not all incongruence carries the same standing
Alvin van Asselt and colleagues, writing in Autism in Adulthood, have published an editorial with a narrow and consequential argument. Four of the five authors have lived experience of autism or ADHD; one does not, and they say so explicitly, along with the note that they challenged each other’s interpretations during drafting. The target is a research practice that has become routine: coding autism and ADHD content on social media as misleading whenever it fails to match current diagnostic criteria or current scientific understanding.
They open by conceding ground that matters. Some content circulating in neurodivergent online communities is, as they put it, demonstrably false and harmful — they name claims linking vaccines to autism. They accept that concept creep is real, that the expansion of diagnostic concepts toward increasingly broad phenomena carries risks including self-pathologisation and people seeking diagnosis for experiences that do not meet established criteria. They accept that some content overgeneralises particular experiences as universal features of autism or ADHD, obscuring how heterogeneous these populations are. None of this is a defence of the discourse wholesale.
The argument is about what follows. Diagnostic criteria are historically and socially situated; scientific understanding of neurodivergence is still moving; therefore not all incongruence with current understanding has the same epistemic standing. They propose four distinguishable categories: claims that are demonstrably false, recurring experiential claims, claims with debated or preliminary empirical support, and claims consistent with well-established consensus. A coding scheme that collapses the second category into the first is not being rigorous. It is being imprecise in a specific and consequential direction — treating the fact that the science has not yet described something as evidence that it is not there.
Concepts the community named before the field could see them
What makes the editorial hard to dismiss is that it produces receipts, and the receipts are the concepts the field now takes seriously.
Sensory hyper- and hypo-reactivity appeared extensively in autobiographical and first-hand accounts long before they were incorporated into the DSM-5 criteria. Autistic masking was named and described within autistic communities before it became a research construct. Autistic burnout emerged from lived-experience accounts and online community discussion, and was formally defined by researchers only after that. Monotropism is an autistic-led theory — developed by autistic people, published in 2005, and only recently receiving the empirical attention it warrants. Distinctive autistic menopausal experiences are being studied now, having been described by autistic women first. In ADHD, community discourse around emotional dysregulation preceded wider recognition of its lived impact, and rejection sensitive dysphoria — a community-originated concept — has stimulated research interest in responses to rejection and criticism.
Now run the counterfactual, because this is where the argument lands. Take any one of those concepts at the moment it was circulating in community accounts but before the research existed. Apply the coding practice under examination. Every single one fails. None matched the diagnostic criteria of its day. None was supported by established scientific understanding. Each would have been categorised as misleading — not because it was wrong, but because the field had not got there yet.
The concepts that would have been discarded are not marginal. Masking, monotropism, autistic burnout and sensory reactivity are, between them, a substantial part of how autism is now understood at all — and a coding scheme with a two-decade lag would have labelled the lot as misinformation. That is not a hypothetical failure mode. It is a description of what the practice does when applied to the historical record, and it should be sobering for anyone building the coding frameworks currently in use.
The authors supply a live example rather than leaving it in the past. One study categorised claims involving the term ADHD paralysis as misleading — a term that describes recurring executive functioning difficulty, which is itself a well-established feature of ADHD. The experiential vocabulary was unfamiliar, so the claim was coded as false, despite the phenomenon underneath it being about as uncontroversial as anything in the literature. The mismatch was between community language and clinical language, and the codebook read that mismatch as inaccuracy.
A holding pen dressed as a category
The mechanism by which this happens is the most useful thing in the editorial, because it is not crude dismissal — it is something subtler that looks like inclusion.
The authors note that some studies evaluating this content do recognise the value of experiential claims, and include a “personal experience” category in their coding schemes. But then, in at least some cases, content in that category gets reclassified as misleading when it contains any element incongruent with current clinical and scientific understanding. So the category exists, and then empties into the adverse one whenever the content does what experiential content characteristically does: describe something the criteria have not caught up with.
A category that cannot survive contact with novelty is not a category for lived experience. It is a holding pen — a place where an account waits until a clinician or a researcher validates it, at which point it is released, and if it is not validated it is reclassified as misinformation. The appearance of epistemic hospitality is preserved while the actual authority remains entirely with the existing framework. And the reverse move completes it: the authors note that content consistent with current scientific evidence has been coded under the positively framed label “useful.” Once “matches the science” is the criterion for useful and “does not match the science” is the criterion for misleading, the hierarchy is no longer an interpretive stance. It is written into the instrument.
The authors reach for Miranda Fricker’s framework here, and it fits precisely. Testimonial injustice occurs when a speaker’s account is given less credibility than it deserves because of who they are — in this case, when autistic and ADHD experiential knowledge is systematically weighted below clinical and scientific knowledge. Hermeneutical injustice occurs when a group lacks the shared concepts to make sense of their own experience — which is what happens when the concepts they have generated to describe their lives are not recognised as legitimate. Both are operating, and the authors note the risk falls hardest on multiply marginalised groups whose perspectives are least represented in research to begin with.
The harm does not stay in the literature. Studies of this kind usually carry careful caveats about interpretive complexity and the evolving state of understanding. Those caveats do not survive the journey into public communication, where the same work becomes headlines about mental health minefields and claims that autism and ADHD content is full of misinformation. What was a methodological limitation in a paper becomes, downstream, a general licence to disbelieve neurodivergent people about their own lives.
Two different things travelling on the same channel
This argument cuts against something I published a fortnight ago, and rather than leave that unaddressed I want to hold both, because holding both is the honest position and the more useful one.
That piece took a paper on diagnostic inflation seriously — it argued that the inflation is real, that its driver is not loose clinical criteria but an identity made culturally available and taken up without the integrative work it requires, and I named that pre-packaged self a mana-personality. I stand by that. Concept creep exists, this editorial’s own authors say so, and the phenomenon of a category expanding until it describes ordinary human variation is not a fiction invented by gatekeepers.
But both things are travelling on the same channel, and that is precisely what makes the discrimination difficult and necessary. The same social media ecosystem that produced a ready-made identity to step into also produced masking, monotropism, autistic burnout and rejection sensitivity — concepts the field either has adopted or is in the process of adopting. Community discourse is simultaneously the most generative source of new understanding in this field and a vector for expansion that dilutes meaning. Neither observation cancels the other. What is wrong is treating either as the whole story: dismissing the discourse wholesale because some of it inflates, or defending it wholesale because some of it is prophetic.
The two failure modes have a common structure worth naming. Blanket dismissal is a failure of discrimination in one direction: everything that does not match current criteria gets coded as noise, and genuine signal is lost for however long the field takes to catch up. Blanket endorsement is a failure of discrimination in the other: everything a community says about itself is treated as beyond examination, and the expansion goes unchallenged until the category means little. Both are ways of not doing the work — the work being the actual, case-by-case labour of telling one kind of claim from another.
Which is why the four-category framework is the substantive contribution here rather than the rhetorical framing around it. Demonstrably false, recurring experiential, debated or preliminary, established consensus — that is a taxonomy that lets you say the vaccine claims are false, the ADHD paralysis material describes something real, monotropism has moved from experiential to empirically supported, and some diagnostic expansion is running ahead of anything that could be called evidence. All in the same analysis, without a blanket verdict on the channel.
Discrimination as the actual work
The authors’ recommendations follow from the framework and are unglamorous in the right way. Adopt coding schemes that distinguish between claim types rather than collapsing them. Involve lived-experience perspectives throughout the research process rather than as a consultation at the end. Treat recurring incongruent experiential claims as phenomena warranting investigation rather than dismissing them prematurely. Communicate findings with proportionate caution, particularly when applying labels like misleading or useful.
The third of those is the one that would change the field. It reframes the encounter between community discourse and research from adjudication to enquiry: when a claim keeps recurring across many independent accounts and does not fit current understanding, the interesting question is not whether it is permitted but what is generating it. That is how monotropism became a research programme. It is how autistic burnout became a defined construct. It is how sensory reactivity got into the DSM.
And it should be said plainly that the standard being asked for is not special treatment, but the ordinary standard of evidence applied consistently. A recurring, cross-corroborated observation that does not fit the existing model is, in every other domain of science, a lead. It is only in this domain that it becomes a coding error, and the reason is not methodological — it is that the observers are the population under study, and their testimony is weighted accordingly. What this editorial asks for is that the weighting be removed, and that the claims be assessed on what they are rather than on who is making them. Some will not survive. That is what assessment means. But the discarding should happen after the examination rather than instead of it — and the concepts the field is currently building on are the reason to take that seriously.
Citations
van Asselt, A., Venema-Avezaat, L., Reekers, D. et al. (2026) — Beyond “Misleading”: Rethinking the Evaluation of Autism and ADHD Social Media Content — Autism in Adulthood
Fricker, M. (2007) — Epistemic Injustice: Power and the Ethics of Knowing — Oxford University Press
Murray, D., Lesser, M. & Lawson, W. (2005) — Attention, monotropism and the diagnostic criteria for autism — Autism
Pearson, A. & Rose, K. (2021) — A conceptual analysis of autistic masking: understanding the narrative of stigma and the illusion of choice — Autism in Adulthood
Yeung, A., Ng, E. & Abi-Jaoude, E. (2022) — TikTok and attention-deficit/hyperactivity disorder: a cross-sectional study of social media content quality — Canadian Journal of Psychiatry
