Hypervigilance is a state of heightened, sustained alertness. Feeling on edge, high alert, often paired with a sense of impending doom seemingly for no reason. You feel jittery and jumpy; your nerves feel tuned to whatever threat you’ve noticed or scanning for, strung so tight they could snap.
No real threat needs to be present to be in this state of alertness. Your body’s strategy is “tense first, ask questions later”, ready for whatever you are about to notice, and staying ready even if there’s nothing to find.
Most of hypervigilance is subconscious and automatic, not a conscious decision you make. You don’t decide that “I am going to jump at every squeak this door behind me makes.” You also don’t put it on your to-do list to “make sure to glance over to the window whenever the wind flicks the curtain“. It might even be so automatic you don’t notice ever doing any of it, apart from being inexplicably wrung out at the end of the day.
Hypervigilance is a human response, not a neurodivergent one. Anyone can develop it, and many people do, particularly after trauma or prolonged stress. But it shows up with particular frequency and differences in cost in neurodivergent adults, so let’s see how this applies to us.
Table of Contents[Hide][Show]
The history of studying vigilance
Vigilance exists as an everyday English word, from the Latin vigilantia, meaning “being careful and cautious, watch out for danger”. It only entered neuroscience in recent history.
In 1923, the British neurologist Henry Head borrowed it from everyday use to describe a whole-body state of readiness the nervous system enters after injury, to restore what has been damaged. Head used the word for a purely physiological process, not yet connected to attention or cognition.
Twenty years later, during the Second World War, the psychologist Norman Mackworth was commissioned to figure out why radar operators kept missing incoming signals during long shifts. His experiments asked observers to watch a clock hand for two hours and press a button whenever it made an unusual double-jump. The participant’s detection accuracy dropped rapidly within the first thirty minutes; Mackworth called this “vigilance decrement”. His experiments reframed vigilance as a fluid resource, a cognitive performance you either sustained or lost1.
Mackworth also tested Benzedrine, an amphetamine, on Royal Navy sailors, to see whether it could keep them alert through long watches. It did, though it made no difference to their physical performance. Benzedrine was the first amphetamine drug patented for medical use, sold over the counter with no prescription needed for decades before that changed. Over time, this drug was made safer and regulated, but it was the ancestor of modern ADHD medication.
With vigilance no longer being something our body does automatically and out of self-preservation, suddenly people were there to blame for their failings. As Peter Hancock has argued, this relocated the problem from the design of the environment the person was working in to something wrong with the person.
What does hypervigilance mean?
Vigilance has since taken on different meanings in different fields.
- Cognitive psychologists use it for sustained attention on a task;
- sleep clinicians use it for a level of arousal;
- animal behaviour researchers use it for scanning for predators;
- trauma clinicians use it for scanning for danger2.
None of these definitions means quite the same thing, so it is important to have clarity on what hypervigilance means in a neurodivergent context.
Hypervigilance in the clinical trauma sense sits closest to the animal-behaviour version. It’s also the sense you’re most likely to have seen the word used for. The DSM-5 describes it as a state of heightened alertness and watchfulness, and the classic illustration is someone scanning a room, noting exits, tracking unusual sounds. Hypervigilance refers to both types of alertness: watching for what hasn’t happened yet, and getting stuck on what has3.
Hypervigilance means both scanning for threats and also getting stuck on them
Eye-tracking studies in trauma populations found more consistent evidence for difficulty disengaging from threatening stimuli once noticed than for enhanced detection of threats in the first place.7.
In autistic youth, brain imaging has shown amygdala hyperconnectivity during implicit social threat processing, and the researchers propose that this implicit-level hypervigilance may drive attentional avoidance once a threat becomes explicit. It’s as if the system detects, locks on, and then looks away to limit the overload.8
Threat, in the clinical literature, mostly means physical danger. However, for a lot of neurodivergent adults, that’s not all that we are monitoring.
Social hypervigilance
If you have spent years figuring out unspoken social rules other people seem to know instinctively, tracking whether your tone was understood the way you meant it, watching for the moment someone’s expression shifts from polite to done — that is also hypervigilance. Your nervous system responds to social survival threats using much of the same circuitry it uses for physical ones.
Autistic camouflaging involves continuously tracking social cues, suppressing your own natural responses, and adjusting your behaviour in real time to match what’s expected23. That pattern of constant self-monitoring is, functionally, hypervigilance directed inward rather than outward. And it isn’t unique to autism. Research on racialised midwives describes a similar mechanism: self-checking, code-switching, and constant threat assessment that the researchers call hidden mental work. It is draining energy through the accumulation of microaggressions and social isolation24. Masking, in other words, may be a form of hypervigilance itself.
Masking has been linked to a history of past interpersonal trauma, suggesting that for some people the hypervigilance driving their masking began with a concrete experience, and it is not just an innate feature of being autistic25.
How does hypervigilance work?
The signs of hypervigilance can be different for different people. You might notice you turn your head more often to scan the room, or that you start fidgeting more, you start sweating or getting jumpy at every noise. Or you just feel like you are on high alert for no reason, on edge, and can’t relax even if logically you should feel safe.
In a study comparing autistic adults with non-autistic anxious and non-anxious adults, the autistic group showed larger baseline pupil size before any auditory stimulation had even begun4.
We need to look into the brain to see how this can happen. In a resting-state study of 639 adults, the connection between the amygdala and a brainstem region (called the periaqueductal gray) tracked with how vigilant a person’s behaviour was, even when nothing threatening was happening5. But the connection between the amygdala and the prefrontal cortex, the part of the brain that evaluates whether an alarm is justified, only activated when something anxiety-provoking was actually present. The alarm runs continuously at a subcortical level. The part of the brain that could assess and switch it off only comes online when there’s something concrete to evaluate.
Threat speed and regulation
For slow or ambiguous threats, like a tone of voice shifting or a pause that goes on slightly too long, the thinking circuits (hippocampus, prefrontal cortex, insula) are involved, and trait anxiety shapes how early the response kicks in6.
For fast, immediate threats, the brainstem drives the response directly, and trait anxiety doesn’t factor in at all. Past a certain speed, the evaluating part of the brain is not in the loop. This is part of why knowing you’re safe doesn’t make you feel safe: the circuits that could update your felt sense aren’t always the ones managing the response.
Hypervigilance and attention
When researchers induced a hypervigilant state in neurotypical adults, it directly impaired their ability to shield internal cognitive processes like reflection, planning, and sustained thought by forcing their attention outward9.
Whether this effect is the same or stronger in neurodivergent adults hasn’t been tested yet, but the trade-off will be familiar to many of us. You can scan the room, or you can think through a problem, but doing both at once degrades both. And since the scanning feels more important (nobody wants undetected danger to pounce after all), that tends to win. If you have ever sat through a meeting and realised afterwards that you tracked every expression, every shift in body language, every tonal change, but have no memory of what was actually discussed, now you know why.
The same shifted focus can also be observed in the body. Autistic adults show reduced accuracy in detecting internal signals like heartbeat, but at the same time report higher confidence in their ability to detect them1011. The body’s signals are harder to read, but the subjective sense of monitoring them is inflated, because it feels all-encompassing. This difference (poor signal paired with high confidence) creates its own problems. If you can’t tell the difference between your heart rate being elevated from anxiety and your heart rate being elevated from climbing the stairs, the ambiguity itself can become another source of alarm.
Is sensory hypersensitivity the same as hypervigilance?
Sensory sensitivity is about detection: can you perceive the stimulus?
Sensory responsivity is about reaction: how does your system respond to what it detected?
Autistic adults show elevated self-reported sensitivity and responsivity, but when measured behaviourally and neurally, group differences in actual detection thresholds are inconsistent12. The clinical picture in autism is driven more by how the system reacts to what it perceives than by detecting more in the first place.
Hypervigilance lives on the responsivity side: your senses are not necessarily sharper, but your system treats what they pick up as more threatening.
What makes hypervigilance worse?
Hypervigilance isn’t fixed at one level, and many things can turn its dial.
Unfamiliar places
When you don’t know an environment, you can’t yet tell what’s safe from what isn’t. One way you can gain more clarity is with more scanning. Anxiety-related impairments to threat perception show up under high uncertainty, and largely disappear once uncertainty drops13.
Autistic adults have also been shown to overestimate how volatile a sensory environment actually is, expecting change and unpredictability even when the environment is objectively stable.14 This doesn’t mean there is no actual problem and you are making things up — intolerance of uncertainty is a genuine phenomenon and not a moral failure.
If your system (not you!) already assumes things are more likely to shift than they are, then every new room, every new group of people, every new routine gets treated as a live variable to monitor, on top of whatever else you’re already tracking. This is why information clarity and direct speech are as much an accommodation as a sensory room.
Stacking sensory load
If you are in a loud and bright room in an itchy jumper, your nervous system is burdened more than all three’s individual burden combined. Sensory pressure has a stacking effect.
A recent theoretical account of autistic meltdowns argues that reduced connectivity within the insula (a brain region involved in reading internal body states) leaves the autonomic nervous system running hotter than it should, so the amount of additional stress it takes to tip into overwhelm gets smaller as the load accumulates15. Separately, autistic youth with sensory overresponsivity have been shown to fail to habituate to mildly unpleasant stimuli, especially when sound and sight are present together rather than one at a time16. The same research found that a subset of autistic youth without this overresponsivity could regulate their reactions by recruiting prefrontal control over the amygdala, so this isn’t a universal or fixed feature. But for those who don’t have that buffer, each additional sensory demand doesn’t just add to the load. It brings the threshold for overwhelm closer.
Self-feeding hypervigilance
Once hypervigilance is running, it tends to keep running, because the loop that sustains it works in your favour just often enough to reinforce it.
Scanning for a threat and finding none produces a small drop in anxiety, and that relief reinforces the scanning that produced it, even though the scanning wasn’t what kept you safe17. This loop mostly runs without you noticing it. First-hand accounts from autistic adults describe this as such: heightened sensory reactivity generates stress, the stress amplifies the reactivity further, and the result is overwhelm, disengagement, or shutdown18.
When Dwyer and colleagues modelled these relationships in autistic and ADHD adults, this chain emerged:
- sensory hyperreactivity is associated with anxiety,
- anxiety is associated with hypervigilance,
- and hypervigilance in turn worsens the original hyperreactivity19.
So you don’t need a new trigger to keep going once it’s started.
Is hypervigilance useful?
Not all hypervigilance is a problem. In an unpredictable or unsafe environment, heightened scanning improves your odds of noticing what actually needs noticing. There is nothing wrong with simply the presence of hypervigilance. But it can become exhausting if
- it doesn’t turn off once the environment changes,
- when it spreads to contexts that were never dangerous in the first place,
- or when the relief of finding no threat trains you to keep looking for one17.
Some of what may appear like miscalibrated vigilance in autistic adults isn’t miscalibrated at all. When non-autistic participants were shown workplace vignettes of an autistic employee’s behaviour, they correctly interpreted what was happening only 31.2% of the time, compared to 50.7% accuracy between autistic participants reading each other20. Autistic adults have also been shown to develop stronger learned defensive responses through watching social interactions unfold, compared to non-autistic adults21. If the people around you are, on average, harder for you to read accurately, and you are also harder for them to read accurately, then staying alert to the difference is useful, not paranoia.22
How can you manage hypervigilance?
Remember Mackworth’s reframing from earlier: vigilance went from “the body reorganising itself” to “an operator failing at a task”, and the fault moved from the environment to the person. The same can happen with hypervigilance. It is tempting to treat it as something to be corrected in the individual: “calm down, stop scanning, you’re safe now.“ But if the environment that produced the hypervigilance hasn’t changed, working to eradicate the scanning without addressing what made it necessary puts the entire burden of change on the person who already carries too much.
It is actually more useful to do the opposite: cultivating or seeking out environments that allow safety to register. These can include spaces where unmasking is possible, social rules are explicit rather than implied, because that means you are not constantly bracing for judgement.
When the environment is predictable enough that your nervous system can afford to lower its guard, much of the social hypervigilance (“will I be misread, did my tone land wrong, is this person about to withdraw“) loses its purpose. This doesn’t mean hypervigilance disappears on its own. For many people it has become entrenched through years of the forward feedback loop (relief at finding nothing makes hypervigilance feel rewarding)17, and some active regulation work might be needed on top of changing the environtment. But the starting point is changing the circumstances, not a person’s reaction to them.
Living with hypervigilance
Sustained hypervigilance is exhausting. But hypervigilance is also a coping mechanism, and it has probably served you at some point. Combined with pattern recognition, hypervigilance lets you pick up the early signs of unsafety before they fully develop. It helps you read rooms, anticipate shifts in social dynamics, and protect yourself from situations your experience has taught you to expect and avoid.
So the goal is not to eradicate it, because it is not the enemy. Managing it means you learn to notice when it’s still running after the need has passed, and to build enough capacity to come down from it when it’s no longer serving you.
To explore this rabbit hole more, here are some relevant entries for you next:
- Norepinephrine is the neurotransmitter most directly involved in keeping the hypervigilant alarm switched on.
- The entry on the vagus nerve covers the body’s other route out of a sustained alert state.
- And when you want something more practical, on how to actually come down after a day spent scanning, that’s what decompressing is for.
“Why does coffee make me sleepy when I have ADHD?”