Positive
emotions
in autism —
autistic joy was
never missing,
the research
just never
measured it

Seventy-two studies, and the emotions science never asked about

Laura Moreno, Andrea Samson and colleagues — a team spanning UniDistance Suisse, the Swiss Center for Affective Sciences, Colorado and the Yale Child Study Center — have published in the Journal of Autism and Developmental Disorders the first scoping review of positive emotions and affect in autism: 72 studies, 1979 to 2024, pre-registered, PRISMA-compliant. A disclosure before anything else: this paper has sat in my stock longer than any source I’ve held, and it’s here because the lead author sent it to me directly after a conversation in these pages’ comment threads. It was worth the wait, and it closes this week the only way this week could close.

Start with what the field has studied, because the inventory is the indictment. Of 72 studies, 36 examined “happy/happiness” — an umbrella term the affective-science literature itself considers too coarse, flattening every distinct positive state into one word. Amusement got nine studies; pride seven; joy, named as such, two. And then the roll call of absences, which deserves printing in full: across forty-five years of research into autistic emotional life, the review found no studies at all on awe, gratitude, hope, love, contentment, relief, serenity, admiration, elation, inspiration or triumph.

Sit with that list. It is not a niche gap; it is most of the positive emotional vocabulary of a human life, unexamined in an entire population — as if the field’s implicit model held that these states either weren’t there or weren’t worth asking about. This week has been about instruments and institutions that cannot see the whole person: the Ombudsman’s system that loses them, the QoL instruments that measure the deficit, the play measures that encode it. Here is the terminus: the half of emotional life the research programme simply never pointed its instruments at. Wednesday’s meta-analysis found autistic emotional quality of life among the lowest-scoring domains. This review adds the necessary asterisk — the instruments behind numbers like those were built from a literature that studied autistic distress exhaustively and autistic gratitude never.

One more structural skew completes the picture: social positive emotions were studied eight times as often as non-social ones (sixteen studies to two). Hold that ratio; the last section turns on it.

No overall difference — the finding underneath the findings

Now the substantive result, and it deserves its exact phrasing: across 72 studies, the findings do not point to overall group differences in positive emotion. Differences appear — but they depend on the specific emotion, the processing stage, the perceptual strategy, cognitive and social capacities, and above all the context. There is no global positive-emotion deficit in autism. There never was one in the data.

The recognition literature shows how the appearance of deficit was manufactured. Where studies found autistic children struggling to recognise complex emotions like pride, controlling for theory-of-mind abilities made the group differences disappear — the difficulty tracked a specific cognitive pathway, not the emotion itself. Recognition of emotion in music — a channel with no social decoding demand — was intact in children and adults alike. Performance correlated with IQ, suggesting deliberate, effortful decoding strategies: autistic recognisers solving analytically what others pattern-match — which this corpus’s readers will recognise as masking’s cognitive signature: competence achieved at higher load. And the review turns the instrument-critique on its own field with unusual candour: the tasks rely on standardised assumptions about how emotions should be expressed and recognised — static faces, forced choices, decontextualised stimuli — norms drawn from neurotypical populations, such that observed differences “may reflect a mismatch between these implicit norms and alternative modes of emotional expression or processing, rather than a deficit per se,” and may therefore overestimate autistic difficulties. The authors invoke both alexithymia — a co-occurring trait, not autism itself, which studies rarely control — and the double empathy problem by name. The Miller communication findings from a month ago said exactly this about conversation; here it is said about joy.

A developmental note belongs here for honesty’s sake: recognition of subtle emotional expression appears to plateau in autistic adolescence rather than continuing to refine — one of the review’s more robust deficit-side findings, and support worth stating for early, well-designed support. The review is not a case that everything is fine. It is a case that the field cannot currently tell what is difference, what is co-occurring alexithymia, what is task artefact, and what is deficit — because it never built the instruments to distinguish them.

Smiling as often, sharing less — experience intact, expression judged

The experience literature contains the review’s emotional centre, and it is a dissociation. Autistic children smile as often as their peers. Their physiological responses to emotional stimuli — the body’s actual reactivity to music, to pleasure, to the world — are typical. What differs is the social layer: they share the joy less often, direct fewer smiles at others, laugh as much at physical comedy but less at social cues, and show avoidant responses to praise. Studies in both children and adults converge on the same formula, which the review states plainly: the evidence points to challenges in emotional expression, not in emotional experience itself.

The joy is there. The broadcasting of it is different. And for forty years, instruments that could only see the broadcast recorded its absence as absence of the feeling — parents observing less expressed positive emotion, raters coding less shared affect, the inner state inferred from the social performance of it. Chevallier’s finding completes the picture: autistic adults showed selective social anhedonia — reduced enjoyment of social reward specifically, with typical responses to non-social pleasures — and autistic laughter appears to express genuine joy without being conscripted into the social-signalling functions laughter usually serves. Read through the double empathy lens, even “shares less” needs its asterisk: sharing is a two-party act, measured here by one party’s norms.

The regulation literature — five studies in forty-five years, itself a finding — adds the strand this corpus would underline: repetitive behaviours functioned as adaptive regulation strategies, deployed for joy as for distress. The behaviour every deficit checklist codes as symptom appears in the data as the tool. And self-compassion — lower on average in autistic adults, unsurprising after the childhoods the shame literature describes — predicted wellbeing and buffered anxiety and depression in autistic and non-autistic people identically. The machinery of flourishing is the same machinery. It has simply been given less to work with.

The joy that lives outside the social frame

Return now to the ratio: sixteen studies on social positive emotions, two on non-social. The review’s discussion names why that inversion matters, and it is the most quietly radical passage in the paper. For many autistic people, the authors write, non-social positive emotions — satisfaction during solitary activities, deep engagement in areas of personal interest, experiences of wonder in natural environments — can be particularly accessible and meaningful sources of wellbeing. The citations behind that sentence are a small canon of the affirming literature: special interests predicting subjective wellbeing; autistic adults’ own accounts of everyday happiness; the new qualitative work on “autistic joy” as its own phenomenon.

This corpus has a name for the architecture underneath: monotropism. The deep tunnel of interest that the hyperfocus piece traced as attention is, from the inside, also an affective structure — absorption as a joy-source, the special interest as a reliable well. Yesterday’s play review found the field coding the four-year-old’s solitary engagement as the deficit while the data showed it doing regulatory work; this review finds the same shape at the scale of a whole research programme: the joy was concentrated exactly where the instruments weren’t pointed, in the solitary, the non-social, the deeply engaged — and a field that studied social emotion eight-to-one recorded, with perfect circularity, that autistic people showed less of what it measured.

To be precise about what this is not: it is not the superpower narrative, and the distinction matters because this corpus dismantled that framing three weeks ago. The superpower story inflates autistic capability into a commercial pitch and makes acceptance conditional on it. This is the opposite operation — no inflation, just an audit of what was never measured: ordinary human joy, present in the physiology, present in the self-reports, present in the solitary hours, unrecorded because the instruments defined joy as its social performance. The review’s agenda follows: study the unstudied emotions, in ecological contexts, distinguishing experience from expression, with autistic people involved in the design.

Wednesday’s meta-analysis measured the gap; today’s review locates the light the gap-measurers never counted. Both are true, and holding both is the whole discipline this corpus practices: the difficulties are real, the environments produce much of them — and running underneath, unmeasured for forty-five years, an emotional life that was never absent. Autistic joy did not need discovering. It needed the field to stop defining it as something else. The people living it — in the deep interest, in the woods, in the corner with the trains — could have said so at any point. Increasingly, in the literature at last, they are.

Citations

Moreno, L., Manfredi, M., Di Poi, G., Gruber, J., McPartland, J. C. & Samson, A. C. (2026) — A Scoping Review on Positive Emotions and Affect in Autism — Journal of Autism and Developmental Disorders

Grove, R., Hoekstra, R. A., Wierda, M. & Begeer, S. (2018) — Special interests and subjective wellbeing in autistic adults — Autism Research

Wassel, E. (2026) — Experiences of autistic joy — Disability & Society

Chevallier, C., Grèzes, J., Molesworth, C., Berthoz, S. & Happé, F. (2012) — Brief report: Selective social anhedonia in high functioning autism — Journal of Autism and Developmental Disorders

Ben Shalom, D., Mostofsky, S. H., Hazlett, R. L. et al. (2006) — Normal physiological emotions but differences in expression of conscious feelings in children with high-functioning autism — Journal of Autism and Developmental Disorders

Kinnaird, E., Stewart, C. & Tchanturia, K. (2019) — Investigating alexithymia in autism: A systematic review and meta-analysis — European Psychiatry

Milton, D. E. M. (2012) — On the ontological status of autism: the ‘double empathy problem’ — Disability & Society