Sleepwalking, Sleep Deprivation and Alcohol

01What happens in the first hours of sleep, and why episodes land there

Sleep is not one flat state. It runs in cycles, and the deepest stage of non-dreaming sleep is loaded into the front of the night. The National Institute of Neurological Disorders and Stroke puts it plainly: stage 3 non-REM sleep is the deep sleep you need to feel refreshed, and it occurs in longer periods during the first half of the night. Heart rate and breathing fall to their lowest, muscles go slack, and it becomes difficult to wake you.

A sleepwalking episode is what happens when something pulls a person part of the way out of that stage and the rest of them stays behind. The movement systems come online. The parts that make memories, weigh consequences and recognize faces do not. In their review of the condition for Lancet Neurology, Zadra and colleagues argue that the old label of a disorder of arousal is too narrow, and that a better description is a simultaneous interplay between states of sleep and wakefulness. That is the version that matches what people describe from the doorway: eyes open, someone moving around the house with purpose, and nobody home behind it.

It also explains the two details that unsettle observers most. The eyes are open because the body is partly awake. There is no memory in the morning because the machinery that writes memories was still asleep, which is a more extreme version of the ordinary reason most dreams are gone within minutes of waking.

That asymmetry is why so much of the reading on this subject is done by somebody other than the sleeper. One woman filming her husband on a landing put the whole experience of watching it into eleven words: after 12 years, i still havent gotten used to this. He will not remember it. She will.

Worth saying once and then leaving alone: this is not a dream being performed. Dream enactment is a REM event, it belongs to the last third of the night, and it is a separate thing with separate implications, which our page on acting out dreams while asleep covers properly. There is a neat symmetry with alcohol here, and it is worth holding on to. The night has two ends and they behave differently. Sleepwalking is a first-half event in deep non-REM sleep. The strange, fragmented, dream-heavy waking at three in the morning is a second-half event, and what alcohol does to the second half of your night is a different mechanism with a different consequence. Same drink, opposite ends of the night. If the mechanics of the timing are what interest you most, why sleepwalking happens in the first half of the night goes further into it.

One number before going on, because it changes how the rest reads. A systematic review pooling 51 studies and 100,490 people, by Stallman and Kohler in PLOS One, estimated a lifetime sleepwalking prevalence of 6.9 percent, with a current rate in the past twelve months of 5.0 percent in children and 1.5 percent in adults. The authors flag a significant risk of bias across the studies they pooled, so treat those as an order of magnitude rather than a precise count. Sleepwalking is common in childhood, much less common in adults, and having it does not put you in rare company.

02What a short week actually does to the following night

This is the half of the question with real experimental evidence behind it, and the experiments are unusually clean.

In 2008, Pilon, Montplaisir and Zadra reported in Neurology what happened when they combined sleep loss with forced arousals. Ten adult sleepwalkers and ten control subjects spent a baseline night in the laboratory, then were played tones at predetermined points either during normal sleep or during the recovery sleep that followed 25 hours awake. During normal deep sleep, the tones triggered an episode in 30 percent of the sleepwalkers. During recovery deep sleep, after the sleep loss, they triggered at least one episode in 100 percent of them. In the control subjects, the same protocol induced no episodes at all.

That last clause is the one to carry away. The same provocation that reliably produced sleepwalking in sleepwalkers produced nothing whatsoever in people without the tendency. Sleep loss did not create sleepwalking. It made existing sleepwalking easier to see.

A much larger study said the same thing at scale. In 2024, Blanchette-Carrière and colleagues compared baseline recordings with post-deprivation recordings in 124 consecutively assessed adult sleepwalkers. After 25 hours awake, nearly twice as many episodes were captured in the laboratory, and the share of patients showing at least one recorded episode rose from 48 percent to 63 percent. Seventeen percent had an event only during the recovery night, while just 2 percent had one only at baseline.

Read the same numbers the other way and you get the honest limit. Even after a full day and night without sleep, more than a third of these diagnosed sleepwalkers still had no recorded episode. Sleep loss is a lever, not a switch, and the same study found it worked much the same regardless of how old someone was when their sleepwalking started or whether anyone else in the family had it. Younger age and more frequent episodes at home were what predicted a strong response.

There is a practical translation of all this that rarely gets stated. The dangerous night in these protocols is not the sleepless one. It is the catch-up sleep afterwards, when the pressure for deep sleep is at its highest and the first hours of the night are at their heaviest. If someone in your house sleepwalks and has just pulled a long shift, a red-eye or an exam week, the night to think about is the one where they finally crash.

03Where alcohol actually shows up in the numbers

The best single picture of what adult sleepwalkers report comes from a case-control study run at the sleep disorders center in Montpellier, France. Between June 2007 and January 2011, Lopez and colleagues assessed 140 consecutive adults and diagnosed primary sleepwalking in 100 of them, median age 30, 55 percent male, each with a night of video polysomnography, compared against 100 age and sex matched controls. Every patient sat through a standardized interview about what set their episodes off.

Just under six in ten, 59.6 percent, named something. Here is what they named.

So, alcohol is on the list and it is at the bottom of it. That already sits oddly beside how confidently the connection gets stated elsewhere. The widely read sleep guides are careful but firm: the Sleep Foundation's page on sleepwalking says that drinking alcohol in the evening can create instability in a person's sleep stages and may heighten the risk of a sleepwalking episode. That is a reasonable description of a plausible mechanism, and it is roughly what most people have absorbed.

The mechanism has been examined, though, and the result is thinner than the confidence suggests. In a 2007 review in the Journal of Sleep Research, Pressman, Mahowald, Schenck and Bornemann went looking for the laboratory evidence and found none of it. There were, at the time of writing, no sleep laboratory studies at all of the effects of alcohol on the sleep of clinically diagnosed sleepwalkers. Not a small number. None. They did find 19 laboratory studies of alcohol's effect on the sleep of healthy non-drinkers and social drinkers, and none of those reported a change in deep sleep as a percentage of total sleep time, though six of the 19 did find a modest but statistically significant increase in deep sleep during the first two to four hours. Their conclusion is quoted more often than it is read in context: there is no direct experimental evidence that alcohol predisposes or triggers sleepwalking or related disorders.

Two things about that review are worth knowing before leaning on it. It was written about criminal cases, where defendants had argued that heavy drinking put them into a sleepwalking state, so its standard of proof is the one a courtroom asks for rather than the one a bedroom needs. And it did not go unanswered. In a 2014 letter to the Journal of Clinical Sleep Medicine, Rumbold, Riha and Morrison pushed back, arguing that the current consensus stating alcohol does not trigger or modify these episodes is artificial, that the small increase in deep sleep is a real finding, and that there is a very urgent need for well-conducted studies in the area. Their own survey of British sleep physicians, which they cite as unpublished data, found most agreed there is a positive association for a minority of patients.

So the two camps in the literature are arguing about a study nobody has run. That is an unsatisfying place to leave a reader who wants to know about their own Saturday nights, and there is one more finding that gets much closer to an answer.

In the same Montpellier study, the authors split their patients by whether their episodes had ever involved violent behavior, meaning physically aggressive or dangerous movement toward themselves or whoever was in the bed. Compared with the 40 patients with no such history, the 55 who had one reported sleep terrors far more often, 81.8 percent against 38.5 percent, and reported triggering factors far more often too, 74.1 percent against 42.5 percent, with an odds ratio of 3.87. And the authors name what stood out inside that difference: particularly alcohol intake.

That is a narrower and more useful statement than either camp usually offers. Alcohol does not look like a thing that turns ordinary sleepers into sleepwalkers, because the one review that went looking for that evidence found the studies had never been done. But among people who already sleepwalk, it concentrates in the group whose episodes are rougher, and in that group it is reported alongside the other triggers rather than instead of them. Both of the things you have read elsewhere are pointing at a real part of this, and they are pointing at different parts.

04The trigger that outranks a short night

Look at that table again, because there is a line in it that tends to get skipped. Strong positive emotions were reported by 41.8 percent of these patients. Sleep deprivation was reported by 26.5 percent. The good day beat the bad night by a wide margin.

The authors were not being vague about what they meant. When they list what they asked about, they gloss strong positive emotions as unusual exciting leisure activity. A concert. A wedding. A first day somewhere. A holiday that started that morning. Something out of the ordinary that left the person wound up rather than worn down.

It is a genuinely cheering fact, and it also quietly reframes the question you probably came in with. A heavy night out is not only alcohol. It is usually a late night, a loud room, a lot of excitement and a short sleep, all arriving together. Three of those four sit at the top of that trigger list on their own merits. The drink is the part that gets blamed because it is the part that is easiest to name in the morning.

Which is also the honest limit of the whole table. These are self-reports collected by interview from people who came to a sleep clinic. People are not good at separating causes that always travel together, and nobody in that study was randomly assigned to a great evening. Take the ordering seriously and the precision loosely.

05Sleepwalking or a blackout, and what each one looks like from outside

This is the question underneath a lot of searches on this subject, and it deserves a straight answer, because the two states can leave the same hole in the morning and are otherwise nothing like each other. The morning itself usually arrives as a joke with a real question inside it, in the manner of the person who woke to the evidence of their own night and asked, was I having a party while sleeping?

In their review of the research for Alcohol, Clinical and Experimental Research, Wetherill and Fromme describe what an alcohol-induced blackout actually is. About half of drinkers report having had one. A person in a blackout is conscious and interacting with their environment, which they distinguish sharply from passing out. Short-term memory keeps working, so the person can hold a detailed conversation and carry out complex actions, including driving. What fails is the transfer of that information into long-term memory, an anterograde failure, so the events happen to a fully awake person and are simply never filed. The authors also make a point worth passing on directly: it can be difficult or impossible to tell from outside whether a drinker is in a blackout, because there is nothing observable to see. They compare it to trying to tell whether someone has a headache.

A sleepwalking episode is the opposite arrangement. The person is asleep, the episode comes out of deep non-REM sleep in the first hours of the night, and the interaction is thin or absent. They do not track a conversation, their responses are off or automatic, and the face is blank rather than drunk.

The practical version is short. Amnesia is common to both and therefore proves nothing. What separates them is what somebody else observed, and the hour. Wetherill and Fromme note that neither type of blackout appears to happen below a breath alcohol concentration of around 0.06, and that most occur far higher, so a modest evening and a 1am wander through the kitchen is a much better fit for an episode than for a blackout.

If nobody saw it and there is no way to know, that is a legitimate answer to sit with rather than a puzzle to solve by reasoning backwards from how you felt.

06The rest of what was true about that night

An episode arrives on a night, and a night contains more than one variable.

Sleep debt is the one that hides best, because it travels with drinking almost every time. A late night out is a short night in bed, and the recovery sleep that follows is the exact condition the laboratory studies used to provoke episodes. Anything else that deepens or fragments the first part of the night belongs on the same list: illness and fever, an unfamiliar bed, noise, a room that is too warm, needing the bathroom, and a stressful stretch at work or at home.

Breathing matters too. Obstructive sleep apnea and periodic limb movements both break up non-REM sleep repeatedly, which is why loud snoring or witnessed pauses in breathing are among the first things a sleep clinic asks about. That is also the most hopeful item here, since it is treatable, and its effects reach into dream life as well, which is the ground our page on sleep apnea and vivid dreams covers.

Medicines are a real factor and the one to be most careful with. A systematic review by Stallman, Kohler and White identified 29 drugs across four classes as possible triggers, with the strongest evidence for zolpidem and sodium oxybate and everything else resting on case reports. Set that against the Montpellier figures for scale: of 100 adult sleepwalkers, exactly one named a medicine, and it was zolpidem. So it happens and it is not common. If episodes began around a new prescription or a dose change, the useful move is to write down the dates and take them to whoever prescribed it. Changing or stopping a medicine to test the theory is not something to do alone, and if your episodes are recent rather than lifelong, what tends to have changed in adult-onset sleepwalking goes through the whole list.

07What you can change, and what a two-week note will tell you

The single change with the best evidence behind it is unglamorous: protect the sleep, especially the night after a short one. Everything the laboratory studies did to provoke episodes involved deepening and then disturbing the first part of the night, and an earlier bedtime after a rough week does more against that than anything else available without a prescription.

The next most useful thing is not about the episode at all, it is about the house. Deal with the stairs, the front door, the car keys, the ground-floor windows and anything breakable near the route someone tends to take. That work is done in the evening rather than in the moment, which matters, because in the moment the sensible thing is to steer somebody gently back to bed rather than to argue with them, and our page on waking a sleepwalker and what to do instead goes through that.

Then, if you want to know whether your own pattern is what you think it is, keep a plain note for two weeks. Date, bedtime, roughly how much sleep you got, when you drank and roughly how much, anything unusually exciting or stressful that day, illness, any medicine and when it was taken, what time the episode happened, what anybody saw, what you remember, and whether anything got broken or bruised. Nothing more elaborate than that.

What a note like that can do is show you a shape: episodes clustering after short nights, or after big days, or after nothing in particular. What it cannot do is prove that one line in it caused the episode on that row, and that limit holds no matter how many weeks you keep it. It is also worth more in a ten-minute appointment than any amount of retelling, because it carries dates and a witness account, which is exactly what a clinician cannot get from the sleeper.

One thing not worth much: a single-night experiment. Skipping the drink for one night and sleeping through proves very little, because the sleep, the stress and the excitement all moved at the same time. If you do end up drinking less over a longer stretch, the sleep changes people notice are usually about dreams rather than about walking, and what happens to your dreams when you stop drinking covers that separately.

08When it is worth raising with a doctor

The Montpellier study is a good place to take the measure of this, as long as you read what it is. Among those 100 adults, 57.9 percent had a history of violent behavior during episodes and 17 percent had suffered an injury needing medical care at least once. Those are high numbers, and they are high because every one of those people was at a sleep clinic. They had been referred precisely because their sleepwalking was bad enough to refer. That is not the base rate for somebody who wanders to the landing twice a year.

Within that, the things a doctor would want to hear about are reasonably specific. Episodes that started new in adulthood rather than continuing from childhood. Anyone getting hurt, including a bed partner. Leaving the house, handling keys, or anything involving a vehicle. Loud snoring, gasping, or breathing that stops and starts. Daytime sleepiness heavy enough to affect driving or work. Episodes several times in one night. And episodes that began around a new medicine.

What the appointment is actually for is less dramatic than it sounds. A history, usually a sleep diary, sometimes an overnight study, and mostly the work of ruling other things out. Sleepwalking is diagnosed largely from the story someone else tells about it, which is why the note and the witness matter more than any wearable's guess at your sleep stages.

If somebody has been hurt tonight, or the situation is unsafe right now, that is an urgent matter rather than one for the next available slot.

09Common questions about sleepwalking, sleep loss and alcohol

Does alcohol cause sleepwalking?

The most-cited review on the question found no sleep laboratory studies at all of alcohol's effects in people diagnosed with sleepwalking, so the direct evidence for causation has never been gathered. What does exist is a clinical association: 12 percent of adult sleepwalkers in the Montpellier study named alcohol as something that made their episodes more frequent or more severe, and it showed up disproportionately among those whose episodes turned violent. The fair summary is that alcohol looks like a contributing factor in some people who already sleepwalk, rather than something that starts it in people who do not.

Can one bad night of sleep set off an episode?

In people who already sleepwalk, yes, and this is the best-evidenced part of the whole subject. A single stretch of 25 hours awake was enough to roughly double the episodes recorded in a laboratory and to raise the share of patients showing an episode from 48 to 63 percent. In people with no history of sleepwalking, the same protocol produced nothing. The episode tends to arrive on the recovery night rather than on the sleepless one.

Was it sleepwalking or a blackout?

The memory gap is identical in both, so it settles nothing on its own. A person in an alcohol blackout is fully awake, can talk fluently and can do complicated things, and simply does not file any of it. A sleepwalker is asleep, usually in the first couple of hours of the night, and does not hold a real conversation. What somebody else saw is the deciding evidence, along with the time on the clock and how much was drunk.

Does an episode mean I was acting out a dream?

Sleepwalking comes out of deep non-REM sleep, where dreaming is at its least vivid, and the movement is not a dream being performed. Dream enactment is a REM event, it belongs to the last third of the night, and the person can usually describe a dream that matched what their body did. Our page on acting out dreams while asleep goes through the difference and why it matters clinically.

Will cutting out alcohol stop it?

Nobody can promise that, because the study that would answer it has not been done. If your episodes have only ever followed nights that involved drinking, drinking less is a reasonable thing to try and a two-week note is how you would find out. Bear in mind that the other things that travel with those nights, the short sleep and the excitement, both rank higher as reported triggers, so a change in the pattern would not by itself point at the drink.

Why does it always happen in the first couple of hours?

Because that is where the deep sleep is. Stage 3 non-REM sleep occurs in its longest stretches during the first half of the night, and an episode is an incomplete arousal out of that stage. The predictability people notice is the timing of their own sleep architecture showing through, and it is one of the most recognizable features of the whole condition.

Should I worry that this only started recently?

New sleepwalking in adulthood is the version most worth mentioning to a doctor, not because it usually means something serious, but because there is often a findable reason and removing it is frequently the whole answer. Sleep debt, a new medicine, untreated breathing trouble, shift work and a stressful stretch are the usual candidates. Most adult episodes that appear out of nowhere have something behind them that a history and a sleep diary can surface.

If you came here to work out whether the pattern you spotted is real, the useful answer is that it probably is, and that the part of it you have been blaming may not be the part doing the most work. DreamTold treats sleep and dreams as a window into your own mind rather than a verdict on your habits, and if the night also left you with an image you cannot shake, look it up in the dream dictionary and think about what it touched. This guide is educational and meant for reflection, not medical advice, and it cannot diagnose anything. If episodes are new in adulthood, if anyone has been hurt, or if there is snoring, gasping or heavy daytime sleepiness alongside them, speak with a qualified clinician.

10Sources

Every claim on this page that comes from somewhere else, with the somewhere else, so you can read the study rather than trust a summary of it.

• National Institute of Neurological Disorders and Stroke: National Institute of Neurological Disorders and Stroke puts it plainly
• PubMed: Blanchette-Carrière and colleagues compared baseline recordings with post-deprivation recordings in 124 consecutively assessed adult sleepwalkers
• PubMed: Pilon, Montplaisir and Zadra reported in Neurology what happened when they combined sleep loss with forced arousals
• PubMed: Pressman, Mahowald, Schenck and Bornemann went looking for the laboratory evidence
• PubMed: systematic review by Stallman, Kohler and White identified 29 drugs across four classes as possible triggers
• PubMed: Zadra and colleagues argue that the old label of a disorder of arousal is too narrow
• PubMed Central: Lopez and colleagues assessed 140 consecutive adults and diagnosed primary sleepwalking in 100 of them
• PubMed Central: Rumbold, Riha and Morrison pushed back
• PubMed Central: systematic review pooling 51 studies and 100,490 people, by Stallman and Kohler in PLOS One
• PubMed Central: Wetherill and Fromme describe what an alcohol-induced blackout actually is
• sleepfoundation.org: page on sleepwalking
• tiktok.com: after 12 years, i still havent gotten used to this
• tiktok.com: was I having a party while sleeping?