The Hag on the Chest
For centuries it wore the costume of demons. The mechanism was hiding in the brainstem all along.
You wake in the dark. Your eyes are open, taking in the familiar geometry of the ceiling, the pale rectangle of the window. You are awake. You know this with certainty. And yet when you try to turn over, nothing happens. Your arm will not lift. Your legs are stone. A weight settles on your chest, slow and deliberate, as if something has climbed onto the bed and is now pressing the air out of you. You try to call out. No sound comes. In the corner of the room, half-glimpsed, a figure is watching.
This is sleep paralysis, and roughly eight percent of people will live through it at least once. Some estimates run far higher, depending on how the question is asked. It is one of the most reliably terrifying experiences a healthy human body can produce, and for most of recorded history it was understood not as biology but as visitation. A demon. A witch. A hag who crouched on the sleeper’s ribs in the small hours. The remarkable thing is how little the experience itself has changed across centuries and continents, and how completely the explanation has.
What the body does every night
To understand why you cannot move, you first have to understand what your brain does to you on purpose, every single night, without your consent or awareness.
Sleep is not a uniform state. Across a night you cycle through stages, descending into deep slow-wave sleep and rising again into something far stranger. The strangest stage is REM, named for the rapid eye movements that flicker beneath closed lids. REM is the theatre of vivid dreaming, the stage where the brain becomes almost as electrically active as it is in waking life. And it is during REM that the brain performs one of its quietest and most consequential tricks. It paralyzes you.
The phenomenon has a clinical name: atonia, meaning an absence of muscle tone. During REM sleep, the body below the eyes and the breathing muscles goes limp. This is not an accident or a side effect. It is a defense. Without atonia, you would physically act out your dreams, swinging at imagined attackers, running from imagined pursuers, all while lying in a dark room full of furniture and another sleeping person. The paralysis keeps the dream sealed inside the skull.
The machinery is precise. Deep in the brainstem, a region called the pons issues commands that travel down to the spinal cord, where two inhibitory neurotransmitters, glycine and GABA, switch off the motor neurons that would otherwise drive your muscles 1. The signal is sent, the muscles fall silent, and the body lies still while the mind builds its nightly hallucinations. For the system to work, two things must stay perfectly in step: consciousness must remain offline while atonia is engaged, and atonia must lift before consciousness returns.
Sleep paralysis is what happens when that timing breaks. Awareness arrives early, or the paralysis lingers late, and for a span of seconds to a few minutes the two states overlap. You are conscious. Your muscles are still locked in dream mode. The dreaming brain has leaked into waking life, and you are awake to witness it.
The discovery of the dreaming stage
That we can describe any of this owes a great deal to a quiet night in a Chicago sleep lab in 1953. A graduate student named Eugene Aserinsky, working under the physiologist Nathaniel Kleitman, was monitoring sleepers and noticed that at certain intervals their eyes began to move rapidly beneath the lids, in bursts, as if watching something. He recorded the same pattern in his own young son. When he woke sleepers during these episodes, they reported vivid dreams. The paper Aserinsky and Kleitman published that year announced REM sleep to the world and effectively founded the modern science of sleep 2.
In the decades that followed, researchers worked out what was happening to the body during this newly named stage. The French neuroscientist Michel Jouvet, studying cats in the 1960s, demonstrated that REM sleep came with a near-total loss of muscle tone, and that lesions to a specific part of the brainstem could abolish the paralysis entirely. His animals, deprived of atonia, would rise during REM and stalk, pounce, and fight invisible prey, fast asleep and acting out their dreams. The experiment was a near-perfect inversion of sleep paralysis. Where Jouvet’s cats lost paralysis they should have kept, the human sufferer keeps paralysis they should have lost.
The medical literature had brushed up against the phenomenon even earlier. In 1876 the American neurologist Silas Weir Mitchell, one of the founding figures of his field, described patients gripped by a temporary inability to move on waking, a condition he called night palsy 3. He recorded it decades before anyone understood REM, atonia, or the chemistry of the spinal cord. He simply documented what his patients told him: that they had woken into their own bodies and found them frozen.
Then come the visitors
If sleep paralysis were merely the inability to move, it would be unsettling. What makes it the stuff of folklore is what arrives alongside the stillness.
A large share of sufferers report a presence. Not a vague unease but a specific, felt conviction that someone, or something, is in the room. Often the figure is dark, standing near the bed or in a doorway, radiating menace. Others feel the crushing weight on the chest, a pressure so physical that breathing seems to require effort, accompanied by a sense of choking or suffocation. Still others hear things that are not there: a buzzing or humming, footsteps, the murmur of voices, a name whispered close to the ear.
These are hypnagogic hallucinations, perceptions that surface in the borderland between waking and sleep. They are not signs of madness. They are the dreaming brain’s imagery bleeding into a mind that is now awake enough to interrogate it.
The psychologist Allan Cheyne, at the University of Waterloo, surveyed thousands of accounts and found that the chaos was not chaotic at all. The hallucinations clustered into three recurring types 4. The first he called the Intruder: the sensed presence, the conviction of an unseen watcher, the fear that something is there. The second, the Incubus, named for the medieval demon, covers the chest pressure, the breathing difficulty, the feeling of being crushed or suffocated. The third he termed vestibular-motor: sensations of floating, flying, falling, or being lifted out of the body altogether. The same three notes, struck again and again, in people who had never spoken to one another and often had no idea the experience had a name.
Why these specific terrors and not some random assortment? The answer points back into the brain’s threat-detection circuitry. During REM, the amygdala, the brain’s alarm center, is highly active. In ordinary dreaming this contributes to the emotional intensity of dreams. But in sleep paralysis the alarm is firing in a mind that is awake and aware of its own body, a body that will not respond. The brain registers fear and, finding no obvious cause in the quiet bedroom, does what brains do best. It builds a story. It supplies a cause for the fear it is already feeling. The shadow in the corner, the demon on the chest, the attacker at the bedside: these are the brain’s attempt to explain its own alarm to itself. The terror comes first. The figure is invented to justify it.
The same dream in every language
This is why the experience translates so cleanly across cultures while the costume changes. The mechanism is human and universal. The interpretation is local.
In Newfoundland, the affliction has long been known as the Old Hag, a witch-like figure who sits on the sleeper’s chest and holds them down. To be attacked was to be hagridden, and the folklore was so detailed and so widely shared that it became one of the first cases anthropologists used to connect a cultural belief directly to the underlying physiology 5. In Japan it is kanashibari, a word borrowed from the practice of binding with metal, capturing exactly the sensation of being clamped in place by an invisible force. In parts of the Caribbean and the American South, the figure is a witch or a haunt come to ride the sleeper. In medieval Europe, the nocturnal visitors were demons with names that survive in our vocabulary: the incubus that pressed down on sleeping women, the succubus that visited men.
The figure changes. The chest pressure does not. The frozen limbs do not. The watching presence does not.
No single image has captured the experience more durably than Henry Fuseli’s painting The Nightmare, completed in 1781. A woman lies sprawled across a bed, head and arms hanging off its edge, and squatting on her chest is a small, malevolent imp, a demon of dead weight. A pale horse with blank eyes pushes its head through the curtains behind. The painting predated any scientific account of what it depicted, yet it renders the core of the experience with uncanny accuracy: the helpless body, the crushing weight, the sense of being observed by something that should not be there. Audiences in the eighteenth century recognized it immediately, because so many of them had felt it.
What reliably brings on an episode is now reasonably well mapped. Sleep deprivation is a major trigger, as is psychological stress and an irregular or disrupted sleep schedule, the kind that shift workers and students endure. Sleeping on the back appears to make episodes considerably more likely, possibly because of how the position affects breathing and the brain’s interpretation of it. One reason students report sleep paralysis at higher rates than the general population is almost certainly their chronically fragmented sleep, a perfect environment for the boundary between REM and waking to blur 6. The episodes themselves are mercifully brief, usually a matter of seconds to a couple of minutes, though to the person inside them the time stretches to something close to eternity.
A defense, not a defect
Here the story turns, because it is tempting to file sleep paralysis under malfunction, a glitch, a wiring fault. It is not.
Every component of the experience is a survival system functioning exactly as it evolved to function. The paralysis is protective: it is the very mechanism that keeps you from injuring yourself or others while you dream, doing its job so thoroughly that it occasionally outlasts the dream. The hypervigilant amygdala is a threat-detection system that evolved to keep our ancestors alive, scanning for danger even in sleep. The hallucinated intruder is the product of a brain that would rather generate a false alarm than miss a real predator, a bias that, over evolutionary time, kept more of our forebears breathing than it killed.
Sleep paralysis is what you experience when these three good systems happen to overlap in the wrong window. The paralysis lingers, the alarm is firing, and the threat-detector, given a body that cannot move and a fear it cannot explain, conjures a watcher to account for it. Nothing has broken. The machinery is simply visible for once, caught in the act, running while the lights are on.
Naming the thing in the corner
There is a practical and faintly hopeful conclusion buried in all of this. The demon on the chest was never in the room. The mechanism, however, was always real, and it turns out that understanding the mechanism changes the experience of it.
Researchers who study sleep paralysis consistently find that people who know what is happening to them report less terror than those who believe they are under supernatural attack. The fear feeds on the absence of explanation. Supply the explanation and you starve it. Knowing that the weight on your chest is atonia, that the figure in the corner is your own amygdala writing a story, that the whole episode will pass in under two minutes if you simply breathe and wait, does not make the experience pleasant. But it strips it of its worst dimension, the conviction that something has gone catastrophically and permanently wrong, that the night has produced a genuine monster.
The monster, in the end, was always the same one: a brain doing precisely what it was built to do, keeping the body still so that it could not hurt itself, keeping the alarm lit so that it could not be caught off guard. The two protections, overlapping by accident in the dark, produce a few minutes of pure dread. Then atonia lifts, the limbs return, the figure dissolves, and you are left lying in an ordinary room, finishing a dream you did not know you were still inside.
Sources
- Brooks, P. L. & Peever, J. H., “Identification of the transmitter and receptor mechanisms responsible for REM sleep paralysis,” Journal of Neuroscience, 2012. — https://www.jneurosci.org/content/32/29/9785
- Aserinsky, E. & Kleitman, N., “Regularly Occurring Periods of Eye Motility, and Concomitant Phenomena, During Sleep,” Science, 1953. — https://www.science.org/doi/10.1126/science.118.3062.273
- Mitchell, S. W., “On some of the disorders of sleep,” Virginia Medical Monthly, 1876. — https://en.wikipedia.org/wiki/Silas_Weir_Mitchell_(physician)
- Cheyne, J. A., Rueffer, S. D. & Newby-Clark, I. R., “Hypnagogic and Hypnopompic Hallucinations during Sleep Paralysis,” Consciousness and Cognition, 1999. — https://www.sciencedirect.com/science/article/abs/pii/S1053810099903777
- Hufford, D. J., The Terror That Comes in the Night: An Experience-Centered Study of Supernatural Assault Traditions, University of Pennsylvania Press, 1982. — https://www.upenn.edu/pennpress/book/13591.html
- Sharpless, B. A. & Barber, J. P., “Lifetime prevalence rates of sleep paralysis: A systematic review,” Sleep Medicine Reviews, 2011. — https://www.sciencedirect.com/science/article/abs/pii/S1087079211000050
- Jouvet, M., “Paradoxical sleep: a study of its nature and mechanisms,” Progress in Brain Research, 1965. — https://www.sciencedirect.com/science/article/abs/pii/S0079612308639026
Related reading
