For professionals
When pain does not look like pain
A patient is visibly in pain. Asked how strong it is, he does not answer, his face stays flat, cooperation seems hesitant. In the team the word uncooperative comes up.
This article is not a general emergency department text from the patient side. It looks at the situation from the staff perspective: what can lie behind short answers, a flat expression and sudden silence. The patient perspective is covered in the article Emergency department.
The following answers come from the app 1:1, unabridged and unedited. They were generated in German on 29 September 2026 and translated idiomatically for this English version.
1. A flat expression is not missing pain
The first answer separates perceiving pain from expressing it.
User question
In the emergency department an autistic patient seems uncooperative or emotionless, even though he is visibly in pain. Is that a communication problem?
Autistic Mirror (outside view, hospital staff)
Not primarily a communication problem, but a perception and expression problem at a neurological level that is often misread as a communication problem.
Interoception describes the ability to perceive body signals internally and to classify them. In many autistic people this channel is calibrated atypically: the signal strength with which pain, hunger or exhaustion reaches awareness differs from the neurotypical expected pattern. This does not mean that no pain is present, but that the person may not be able to locate, name or place it in time, while it is physiologically fully active at the same time. Cortisol rises, heart rate rises, the autonomic nervous system reacts, but conscious processing and verbalisation do not reliably couple to it.
Alexithymia follows on from this. It is not an absence of feeling, but reduced access to emotional and bodily states through language and categories. The question "Where does it hurt, on a scale from one to ten?" assumes that pain can be located, graded and translated into numbers. For some autistic people none of these conditions can be met reliably, especially not under acute stress. The result looks from the outside like indifference or refusal to cooperate.
At the same time, an emergency department delivers massive sensory input: light, noise, smells, unfamiliar physical closeness, unpredictable procedures, changing staff. Sensory gating, the largely automatic dampening of background stimuli in the thalamus, often works less efficiently in autistic people than in the neurotypical average. What should arrive filtered arrives unfiltered. The nervous system processes the stimulus environment of the emergency department as such, while it is simultaneously loaded with pain and exhaustion. Under this load, speech and cooperation can block neurologically, not by an act of will, but through overload of the available processing capacity.
What becomes visible then can be a shutdown. A shutdown is a withdrawal of the nervous system under overstimulation, with reduced responsiveness, limited speech or complete speechlessness, and an outwardly flat or absent expression. It is not dissociation and not disinterest, but a protective mechanism of the overwhelmed system. In a shutdown the person is present, may hear everything, but cannot respond as expected.
The question "Is this cooperative or not?" is the wrong axis for the clinical decision. The more productive axis: which information do I need, and which form of gathering it bypasses the barrier? Offering pain scales visually instead of verbally, yes/no questions instead of open questions, allowing time for response latency, reducing stimuli in the room as far as possible. Accompanying persons can often provide behavioural baseline information that complements verbal self-report.
A flat expression in autistic patients does not rule out severe pain. This connection is clinically relevant.
2. When speech drops away mid-treatment
The second answer explains why someone who was just talking can fall silent.
User question
Why can an autistic patient in the emergency department suddenly stop speaking or answering altogether, even though he was talking before?
Autistic Mirror (outside view, hospital staff)
Loss of speech in an emergency department is, in the majority of cases, not a neurological symptom in the classic sense, but the result of a neurological protective mechanism responding to simultaneous overload.
Shutdown describes the state in which the nervous system, under extreme density of stimuli and demands, reduces processing capacity to vital functions. Speech production is not a vital automatism; it is a metabolically expensive, actively controlled achievement that draws on prefrontal coordination, working memory and sequential motor planning at the same time. When the system is overloaded, this channel is given up first, not because the person no longer wants to communicate, but because capacity for this processing pathway is physically no longer available.
What runs in parallel in an emergency department is hard to overestimate from an autistic processing perspective: neon light without breaks, an unpredictable soundscape from many directions, physical contact by unfamiliar people without warning, smells, temperature changes, constantly changing staff with changing demands, pain or physical symptoms, uncertainty about procedures and timing. Autistic sensory gating, the pre-attentive filtering of stimuli that normally decides in the thalamus which signals are passed on for processing at all, works fundamentally differently here than in neurotypical patients. Stimuli that neurotypical people tune out automatically and effortlessly continue unfiltered into cortical processing in many autistic people. This means: each of these emergency department stimuli actively costs processing capacity that would be needed for communication at the same time.
Added to this is the predictive coding component. The brain constantly compares expectations with what actually arrives and has to readjust when they differ. An emergency department is a continuous source of prediction errors, since no procedure is known, no expectation reliably holds and the next step cannot be anticipated. Readjusting these prediction errors is effort of its own, added on top.
The loss of speech often occurs with a delay, because the person could still mobilise reserves at the beginning. Masking, actively adapting one's own behaviour to the expectations of the surroundings, costs extra on top of this. Someone who still "functions" in the first phase may have spent a considerable part of the remaining budget on appearing cooperative from the outside alone.
In terms of differential diagnosis, shutdown in the emergency department is often hard to classify because it resembles a dissociative state. The mechanism, however, is a different one: dissociation is typically a protective response to traumatic content with a splitting-off character, shutdown is a sensory-cognitive exhaustion standstill without a necessary trauma trigger. Both can coexist and reinforce each other, but the intervention strategy differs. In shutdown, reducing incoming stimuli helps; in a dissociative state the same measure can be counterproductive.
In practice this means for the emergency department: the moment speech is lost is not a signal that the underlying problem that brought the patient in has worsened. It is a signal that processing capacity has reached its limit. Reducing stimuli, offering a written route, involving relatives as a communication bridge, announcing procedures before contact happens: these are not comfort measures, but the precondition for the person to become able to give information again.
3. Why the waiting time counts
The third answer maps light, alarm tones and waiting time onto three mechanisms.
User question
Why does the stimulus environment of an emergency department, with light, alarm tones and waiting time, affect autistic patients so strongly?
Autistic Mirror (outside view, hospital staff)
Emergency departments hit autistic people with a combination of three sources of load that act at the same time and reinforce each other.
Sensory gating works pre-consciously: before a stimulus becomes conscious at all, a neurological filter sieves out what is irrelevant. In autistic people this filter is more permeable, so more raw material arrives at the cortical level. The fluorescent light, the beeping of the monitors, the smell of disinfectant, voices from the next room, the fabric of the paper gown, the vibration of the floor when someone walks past: all of this lands in awareness at once and with roughly equal priority, without automatic dampening. Under normal conditions this is neurologically demanding, but manageable. In the emergency department the background of stimuli is so dense that the filter runs under constant load from the start.
Added to this is sensory processing fatigue, a time effect. What was still bearable at the beginning of the wait is qualitatively different after an hour. The filtering work, which does not happen automatically but runs under control through prefrontal capacity, actually consumes metabolic resources. The prefrontal cortex carries this load because the automatic system catches too little. Over time this capacity empties, and what is physically just as loud arrives as subjectively unbearable. Waiting time in an emergency department is therefore not neutral time, but active exhaustion time.
The third layer is predictive coding. The autistic brain weights the incoming sensory value more strongly than its own prediction: it trusts what is coming in right now more than what it expects. In an emergency department there are hardly any predictions that stay stable. Every alarm could mean one's own situation. Every doctor who comes in might address you or might not. The conversation next door could be relevant or irrelevant. Because the system would rather weight too much than too little, the neural comparison between expectation and input stays permanently active and permanently costly. This is not anxiety in the clinical sense, but computational constant work running in the background, regardless of whether the person looks calm from the outside.
These three mechanisms overlap: the filter catches little, the filtering work exhausts itself over the waiting time, and at the same time the prediction system runs at full capacity because the environment is highly unpredictable. Someone who is in a shutdown after two hours in the emergency department is not showing psychiatric decompensation, but neurological exhaustion in a system that was demanded considerably more from the start.
For clinical practice this means: reducing stimuli is not a comfort measure, but a precondition for an autistic person to remain able to give information at all and to report symptoms reliably.
What holds in this account
That sensory overload and long waiting times are among the most common difficulties for autistic people in the emergency department is described, among others in a review in Frontiers in Health Services (2026). Interoception, alexithymia, sensory gating, predictive coding and masking are each studied on their own.
Applying them to exactly the emergency department is interpretation, not a study finding. No study has measured this chain in this situation.
Statements about thalamic filtering, prefrontal capacity, cortisol and metabolic cost are model assumptions, not measurements of the individual person. The description of shutdown draws on polyvagal theory, a widespread and contested framework. Sensory processing fatigue is a descriptive term without its own first publication and is therefore not in the source list.
The second answer separates shutdown from dissociation. This is orientation, not a diagnostic criterion. Because the German texts were taken over unchanged, three spelling mistakes remain in the original; the translation does not reproduce them.
The adjustments mentioned concern environment and conversational form. They are not a treatment recommendation and do not replace the medical decision in the individual case. Pain assessment remains a clinical task.
Autistic Mirror explains why patients in the emergency department can appear different from how they actually are. It does not diagnose and does not replace a medical assessment.
Sources
- Tavassoli, Hoekstra & Baron-Cohen (2014). DOI: 10.1186/2040-2392-5-29
- Garfinkel, Tiley, O'Keeffe, Harrison, Seth & Critchley (2016). DOI: 10.1016/j.biopsycho.2015.12.003
- Bird & Cook (2013). DOI: 10.1038/tp.2013.61
- Porges (1995). DOI: 10.1111/j.1469-8986.1995.tb01213.x
- Pellicano & Burr (2012). DOI: 10.1016/j.tics.2012.08.009
- Hull, Mandy, Lai, Baron-Cohen, Allison, Smith & Petrides (2019). DOI: 10.1007/s10803-018-3792-6