Why Some People Sleepwalk (and Never Remember It)

That simultaneity is the correction worth making. Sleepwalking is usually described as incomplete waking, and a recent comprehensive review argues that framing it purely as a disorder of arousal is too restrictive — the better picture is genuine coexistence of sleep and wakefulness in one brain.

When it happens

Sleepwalking emerges from N3 slow-wave sleep, the deepest stage of non-REM sleep, and therefore clusters in the first third of the night when slow-wave sleep is most abundant. This timing is diagnostically useful: complex behaviour late in the night, during the REM-rich final hours, points to something else entirely.

It belongs to a family called disorders of arousal, alongside confusional arousals and sleep terrors, all of which share the same origin in incomplete emergence from deep sleep. Sleep talking frequently accompanies it — the same partial arousal reaching a different output channel, which is why somniloquy and sleepwalking so often occur in the same people.

How common

A systematic review and meta-analysis covering 51 studies and more than 100,000 people put lifetime prevalence at around 6.9 percent. Current prevalence — within the past twelve months — was about 5 percent in children and 1.5 percent in adults.

The authors flagged significant risk of bias across the studies and advised using the figures cautiously, which is worth repeating. Sleepwalking is difficult to measure because it depends on someone else witnessing it, and an adult sleeping alone may have no idea they do it at all.

There is a strong familial component. Having an affected parent substantially raises the odds, and specific genetic associations have been reported, though no single inheritance pattern accounts for it.

What the brain is doing

The decisive evidence came from patients undergoing intracranial monitoring for other reasons, where electrodes recorded directly from cortex during spontaneous episodes.

The recordings showed dissociation. Motor and cingulate cortices displayed activity patterns characteristic of wakefulness, while frontoparietal associative cortices continued producing slow waves typical of deep sleep. An earlier imaging case study had pointed the same way, finding activation in posterior cingulate and cerebellar regions alongside deactivation of frontoparietal association cortex.

This explains the specific profile of a sleepwalker's behaviour with unusual precision. Motor control is available, so walking, opening doors and manipulating objects are all possible. Executive function, judgement and self-awareness depend on the regions still asleep, which is why behaviour is inappropriate to context and why sleepwalkers do not recognise the strangeness of what they are doing.

Memory is the clearest case. Encoding an episodic memory requires structures that remain in slow-wave sleep throughout, so the behaviour happens with nothing recording it. The absence of memory is not repression or dissociation in the psychological sense — the machinery that writes memories was simply not running, in the same way an errand disappears when the event model holding it closes.

Three things that are not true

The 2025 review noted that several widely held views about sleepwalking are characterised by key misconceptions. Three are worth naming.

• That waking a sleepwalker is dangerous. It is not. Waking someone mid-episode may leave them confused, disoriented or briefly agitated, but there is no physical harm in it. The genuine danger is the episode itself — falls, stairs, doors, kitchens — so guiding someone back to bed, waking them if necessary, is the sensible response.
• That sleepwalkers are acting out dreams. They are not. Sleepwalking arises from non-REM sleep, when dreaming is far less vivid and narrative. Acting out dream content is a different phenomenon occurring in REM, with a different mechanism and different clinical significance.
• That sleepwalking reflects psychological disturbance. There is no established link between sleepwalking and personality or psychiatric disorder in the way folklore implies. It is a heritable neurological phenomenon of sleep-state regulation.

What raises the odds

The reliable precipitants share a common feature: they either deepen slow-wave sleep or fragment it, both of which increase the chance of a partial arousal from it.

Sleep deprivation is the most consistently reported, and can be used experimentally to provoke episodes in known sleepwalkers. Fever, stress, an unfamiliar environment, a full bladder, noise and other sleep disorders that fragment the night all appear as triggers. A systematic review of medication-induced sleepwalking identified a range of drugs associated with it, sedative-hypnotics prominent among them.

Alcohol is frequently listed as a trigger and the evidence is weaker than that suggests. It is a common assumption rather than a well-supported finding, and worth treating with more caution than it usually receives.

The courtroom problem

Sleepwalking is one of the very few phenomena on this site with a body of legal literature attached, because the dissociation it produces raises a genuine question about responsibility. If judgement, insight and memory-forming systems are demonstrably asleep while motor systems are awake, the standard assumption that a person intends their actions does not straightforwardly apply.

Sleep-related violence has been raised as a defence in criminal cases, and reviews of the field note that accurate diagnosis has become increasingly important as the number of medicolegal cases grows. The clinical difficulty is real: episodes cannot be reproduced on demand, the person has no memory to report, and the diagnosis rests substantially on history and on witness accounts.

This article makes no claim about how any such case should be decided. It is worth knowing that the neurology is taken seriously enough by courts to require expert testimony, and that the difficulty of proving what a sleeping brain was doing is exactly the difficulty the intracranial recordings above were addressing.

When it needs attention

Childhood sleepwalking is common and usually resolves without intervention. Episodes that begin in adulthood, occur frequently, involve leaving the house or driving, or result in injury are a different matter and warrant proper assessment by a qualified clinician — as does anything involving apparent enactment of dream content, which points somewhere else entirely.

The practical measures are unglamorous: securing doors and windows, clearing obstacles, avoiding upper bunks, and addressing sleep deprivation, which is the one trigger most within anyone's control.