A depletion that accumulates from the unremarkable, rather than the dramatic
Nicholas Norman Adams, an academic psychologist at Robert Gordon University in Aberdeen, published an autoethnographic account of autistic burnout in Disability & Society at the end of July 2026. It is a single case — his own — and he is transparent about his position: he holds an ADHD diagnosis made in childhood and confirmed in adulthood, and is currently progressing through the formal autism diagnostic pathway, following an outdated Asperger’s diagnosis raised repeatedly during his childhood and teens. He writes, as he puts it, from a dual position: someone who has experienced burnout periodically, and someone who researches wellbeing and neurodivergence professionally.
What makes the account useful is its precision about causes. Burnout, in his description, does not arrive after a crisis. It accumulates from things nobody else would register as events at all — a stranger striking up conversation on public transport, being required to contribute in a meeting, a friend ringing unexpectedly. Each demands social and emotional labour he may not have available at that moment, so the energy is borrowed against a deficit. He calls these invisible stressors, and the term earns its place: they are invisible precisely because they are ordinary, and their ordinariness is why the resulting collapse looks unaccountable to everyone watching.
Underneath them sits the constant cost of masking — the continuous adjustment of how you move, speak and respond to align with what is expected. Adams describes the particular loneliness of it, a functional solitude in which he can feel alone while surrounded by people, because the bulk of his mental energy is going into projecting a calm and unremarkable surface. This is the load-bearing point of the whole model. The masking is not incidental to the burnout. It is the expenditure that empties the account, and it runs continuously, in every ordinary interaction, whether or not anything notable happens.
Five stages measured in performance no longer affordable, not function lost
From this he builds the Social Presentation and Emotional Load model, five stages describing the progression. Stage one is outward success: the person presents as competent, productive and sociable, at the cost of intensive masking — reading expressions, comparing behaviour against norms, continually evaluating how they are landing. Stress and anxiety are present, alongside some satisfaction when the performance seems to work.
Stage two is the beginning of the strain. The outward presentation still holds, but sustaining it becomes harder as the load compounds. Masking becomes partial; the social mimicry lapses intermittently; the satisfaction thins as the performance stops reliably passing.
By stage three the performance fragments. Here the model makes a distinction worth noticing: practical and occupational functioning may be preserved while the social components of masking become minimal or absent. Interactions turn blunter, less fluid, closer to baseline. Internally, Adams describes anxiety that has become unbearable, low mood, and persistent worry about how others are reading him — the self-esteem cost of no longer being able to sustain the version of himself he believes is required.
Stage four is withdrawal, and this is where the model becomes genuinely informative. The need to mask is not overwhelmed; it is intentionally removed. Social interaction is avoided deliberately, as a protective measure to preserve what remains. Task-based functioning may continue in isolation. And crucially, mood begins to stabilise. Stage five is complete retreat — work, social contact and communication all paused, a full shutdown, with Adams describing occasions when he turned off or discarded a phone simply to make the noise stop. At this stage emotional responses gradually return to baseline as the stressors are removed.
Read as a sequence, the five stages do not describe a person progressively losing the ability to function. They describe a person progressively losing the ability to perform — and then, at the point of retreat, recovering.
A recovery that identifies its own cause by what it takes away
That last detail is the finding, and it deserves to be stated plainly. Mood stabilises at stage four, when social contact is removed. It returns to baseline at stage five, when the demand is removed entirely. Nothing is administered. No deficit is remediated. The person is not treated. The environment is subtracted, and the person recovers.
If removing the demand is what restores the person, then the demand was the cause. That is not a rhetorical flourish; it is the ordinary logic of identifying an agent by withdrawal. Autistic burnout, on this account, is not a feature of being autistic. It is the arithmetic of an unsustainable performance running on a finite budget in an environment that never priced it — which is why Adams insists it is a socially mediated state of distress rather than a disorder, and why he says his burnout is not the result of being autistic but of navigating environments that routinely disregard how he works.
This also explains a persistent clinical problem the literature has flagged: burnout is frequently misattributed to depression or anxiety. Look at the stages and the reason is obvious. Nobody observes stage one — that is the version of the person everyone knows and considers fine. The observer arrives around stage three or four, when the withdrawal has become visible, and finds someone socially avoidant, flat, exhausted and low. That presentation maps neatly onto depression. The difference is that the cause is upstream and environmental, and the response that helps is removal of demand rather than treatment of an internal state.
One tension is worth naming, because it runs against an argument I’ve made recently. Adams cites work framing immersive activities such as video gaming as legitimate self-regulation rather than a symptom — and I’ve argued that retreat into frictionless digital environments spends the very desire it appears to restore. His own account resolves this more cleanly than either position alone. His stage five recovery involves cutting off technology — discarding the phone, seeking silence and isolation. What restores him is the removal of input, not the substitution of an easier input. Withdrawal into quiet lets the system reset; withdrawal into a stream engineered to capture attention swaps one demand for a frictionless one, and the account does not run down.
Rest as infrastructure rather than a reward to be earned
The practical implication follows directly from the mechanism, and it is unglamorous. If burnout accumulates from ordinary demands and resolves when they are removed, then the useful intervention is periodic, protected, unapologetic removal — built in before the collapse, not administered after it.
Adams is sharp about why this rarely happens: meetings go ahead, social events are fixed, obligations cannot be shifted, and the structure of modern life makes no room for the kind of retreat he needs. So the need is met only at the point where it has become an emergency, which is stage five — full withdrawal from everything, at maximum cost to the person and to whatever they were meant to be doing. Provided early, the same rest is cheap. Provided late, it costs weeks.
The frame he lands on is that rest should not be something to apologise for, and masking should not be the price of inclusion — and this is where a single autoethnography does what a larger study often cannot. It is one person’s account, offered as a conceptual tool rather than a validated instrument, and it should be read as that. But it names a mechanism with unusual clarity: the burnout is the bill for a performance that was demanded, and the recovery arrives when the demand stops. Any environment serious about not burning out its neurodivergent people can act on that without waiting for the model to be validated. Stop requiring the performance, and there is nothing to recover from.
Citations
Adams, N. N. (2026) — A lived-experience perspective of dealing with autistic burnout: introducing the Social Presentation and Emotional Load model — Disability & Society
Raymaker, D. M., Teo, A. R., Steckler, N. A. et al. (2020) — Having All of Your Internal Resources Exhausted Beyond Measure: Defining Autistic Burnout — Autism in Adulthood
Arnold, S. R., Higgins, J. M., Weise, J. et al. (2023) — Confirming the nature of autistic burnout — Autism
Hull, L., Levy, L., Lai, M.-C. et al. (2021) — Is social camouflaging associated with anxiety and depression in autistic adults? — Molecular Autism
