Sleep Paralysis: REM Atonia, Hallucinations, Narcolepsy and What Actually Helps

Bottom line: Sleep paralysis is usually a brief REM-wake transition in which awareness returns before normal REM muscle atonia has fully switched off. It can feel terrifying and can include vivid visual, auditory or sensed-presence hallucinations, but isolated episodes are generally benign. Recurrent episodes deserve a sleep-history review—especially for sleep deprivation, irregular schedules and narcolepsy clues—rather than assuming the experience is psychosis, a supernatural event or a deficiency that needs a supplement.[1-9]

What sleep paralysis actually is

During normal REM sleep, most skeletal muscles are strongly inhibited.

That REM atonia helps keep dream-related motor activity from being acted out.

Sleep paralysis occurs when features of REM and wakefulness overlap: conscious awareness returns while the REM-related muscle inhibition persists.[1,5,8]

The person may be able to:

• see;
• hear;
• breathe;
• move the eyes;
• remember the event clearly;

while being temporarily unable to move the body normally.[5]

Episodes are usually brief and resolve spontaneously.[1,5]

Hallucinations can be part of the REM transition

Sleep paralysis is famous for frightening hallucinations.

People may report:

• a sensed presence in the room;
• shadows or figures;
• voices or other sounds;
• chest pressure;
• a feeling of being watched;
• floating, falling or out-of-body sensations.[2,5]

These experiences can be extremely vivid.

But hallucinations during sleep paralysis occur specifically around sleep-wake transitions and do not automatically imply a psychotic disorder.[5]

The 2024 prevalence meta-analysis estimated combined visual and auditory hallucinations in about 24% of affected participants, while most cases did not include that combined hallucination pattern.[2]

So “sleep paralysis always includes hallucinations” is also false.

Why people feel chest pressure or suffocation

Eye movements and basic respiration remain possible during isolated sleep paralysis, but the experience of breathing can feel abnormal.[5]

That mismatch between waking awareness, REM physiology and fear can produce a powerful sense of pressure or suffocation.

This sensation should not be used to diagnose sleep apnea by itself.

If someone also has habitual snoring, witnessed breathing pauses, gasping or marked daytime sleepiness, that is a separate reason to evaluate for obstructive sleep apnea.

See Snoring vs Sleep Apnea.

How common is sleep paralysis?

There is no single prevalence number that should be repeated without context.

An influential 2011 systematic review pooled 35 studies with 36,533 participants and estimated lifetime prevalence at:[3]

• 7.6% in the general population;
• 28.3% among students;
• 31.9% among psychiatric patients;
• 34.6% among patients with panic disorder.[3]

Those numbers are often quoted.

But a newer 2024 systematic review and meta-analysis included 76 studies from 25 countries with 167,133 participants and produced a pooled estimate near 30%.[2]

That sounds like a major revision—until the heterogeneity is examined.

The newer meta-analysis reported I² values around 96–100% in major pooled analyses and detected publication bias.[2]

That means the included studies were producing extremely different estimates.

The safest conclusion is:

Sleep paralysis is not rare, but exact prevalence depends heavily on population, definition and measurement method.

This is a good example of Why Sleep Studies Disagree.

Recurrent isolated sleep paralysis is different from one lifetime episode

Having one episode at some point is not the same as having recurrent isolated sleep paralysis that causes distress or sleep avoidance.

The 2024 clinical review describes recurrent isolated sleep paralysis as a benign REM parasomnia in which episodes recur independently of narcolepsy or another explaining disorder.[1]

The word isolated matters.

If another sleep disorder better explains the episodes, the diagnostic pathway changes.

Sleep deprivation and irregular schedules are recognized risk factors

The 2024 clinical review lists several common predisposing factors:[1]

• irregular sleep-wake schedules;
• sleep deprivation;
• jet lag.

A broader 2018 systematic review of 42 studies also found associations with sleep disruption, stress and trauma, anxiety symptoms, psychiatric disorders, substance use and other variables.[4]

But these are mostly risk associations.

They do not justify saying that one factor causes every episode.

Stress and PTSD can increase risk without making every episode “psychological”

The 2018 review found sleep paralysis was particularly common in PTSD and, to a lesser degree, panic disorder.[4]

That does not mean the paralysis is imaginary.

The REM-atonia phenomenon is physiological.[1,5,8]

Psychiatric symptoms may influence vulnerability, sleep fragmentation, interpretation and distress without replacing the underlying REM-wake mechanism.

For trauma-related insomnia and nightmares, see PTSD, Nightmares and Sleep.

Does sleeping on your back cause sleep paralysis?

Older observational research found sleep paralysis was reported more often in the supine position than other positions.[7]

In that study, the supine position was reported roughly three to four times more often during sleep paralysis than during ordinary sleep onset.[7]

That is interesting, but it is not strong evidence that back sleeping causes sleep paralysis.

The study relied on reported position and timing rather than randomly assigning sleep position.

So the calibrated claim is:

• supine sleep is associated with more reported episodes in some data;
• positional avoidance may be a reasonable low-risk experiment for an individual who notices a clear pattern;
• there is not enough evidence to call “never sleep on your back” a universal treatment.

Narcolepsy is the most important sleep-disorder differential

Sleep paralysis can occur in narcolepsy.[5,9]

Narcolepsy can also involve:

• excessive daytime sleepiness;
• cataplexy;
• disrupted nighttime sleep;
• hypnagogic or hypnopompic hallucinations.[9]

So a person with recurrent sleep paralysis plus persistent excessive daytime sleepiness or cataplexy deserves a different level of evaluation than someone with a rare episode after several nights of sleep deprivation.[5,9]

The 2016 clinician guide specifically recommends differentiating isolated sleep paralysis from narcolepsy through clinical history and, when uncertainty remains, objective sleep testing such as polysomnography and multiple sleep latency testing.[5]

Sleep paralysis is not REM sleep behavior disorder

These conditions can sound similar because both involve REM sleep.

But their motor physiology points in opposite directions.

In sleep paralysis, REM-related atonia persists into wakefulness.

In REM sleep behavior disorder, normal REM atonia is lost or reduced, allowing dream-enactment behaviors.

They should not be treated as interchangeable REM disorders.[8]

Sleep paralysis is not automatically psychosis

A frightening hallucination during a sleep-wake transition can feel fully real.

The timing is the important clue.

In isolated sleep paralysis, the hallucination occurs as the person is falling asleep or waking and is linked to the paralysis episode.[5]

Hallucinations occurring repeatedly during full daytime wakefulness require a different clinical framework.

That distinction can reduce unnecessary fear without dismissing the intensity of the experience.

What actually helps?

For isolated or occasional episodes, the 2024 clinical review states that no drug treatment is required.[1]

The most defensible first-line approach is:

• explain what the phenomenon is;
• correct sleep deprivation when possible;
• stabilize sleep-wake timing;
• reduce major schedule disruption;
• review whether jet lag or shift-like timing preceded episodes;
• address fear and sleep avoidance if they develop.[1,5]

Reassurance is not “doing nothing”

A person who believes they are being attacked, suffocated or developing psychosis can become afraid to sleep.

Explaining that REM muscle atonia has temporarily persisted into awareness can reduce catastrophic interpretation.[1,5]

That matters because fear itself can worsen bedtime arousal and sleep disruption.

CBT may help distress, but the treatment evidence is thin

The 2024 review notes that cognitive behavioral approaches may help recurrent cases accompanied by anxiety and frightening hallucinations.[1]

The older clinician review also describes psychotherapeutic approaches as promising but emphasizes the need for larger, well-controlled trials.[5]

So CBT for recurrent isolated sleep paralysis should be described as plausible and clinically used, not as a treatment supported by a large RCT evidence base comparable with CBT-I for chronic insomnia.

Supplements have almost no direct evidence here

A case report can be useful for generating hypotheses, but it cannot establish that a vitamin, herb or supplement treats recurrent isolated sleep paralysis.

The current evidence base does not justify claims such as:

• magnesium prevents sleep paralysis;
• vitamin D treats sleep paralysis;
• melatonin cures REM-wake dissociation;
• sedating herbs prevent hallucinations.

The stronger evidence points toward sleep timing, sleep deprivation, education and differential diagnosis—not supplement stacking.[1,4,5]

When recurrent episodes deserve evaluation

Clinical review becomes more important when episodes are:

• frequent;
• severely distressing;
• causing sleep avoidance;
• accompanied by persistent excessive daytime sleepiness;
• accompanied by cataplexy;
• associated with other unusual neurologic episodes;
• difficult to distinguish from seizures, narcolepsy or another parasomnia.[5,9]

What the evidence does not prove

Current research does not justify saying that:

• sleep paralysis is supernatural;
• sleep paralysis means psychosis;
• every hallucination during sleep paralysis is abnormal;
• everyone who has sleep paralysis has narcolepsy;
• back sleeping universally causes it;
• everyone should change sleep position;
• one prevalence estimate applies to every population;
• one supplement treats recurrent isolated sleep paralysis;
• medication is required for ordinary isolated episodes.

Related sleep research

• PTSD, Nightmares and Sleep
• Anxiety and Sleep
• Sleep Apnea vs Insomnia
• Snoring vs Sleep Apnea
• Sleep Deprivation vs Insomnia
• Why Sleep Studies Disagree
• Sleep Regularity

Verdict

Sleep paralysis is usually a benign REM-wake dissociation, not evidence that the person is “losing touch with reality.”

The strongest practical evidence supports education, adequate sleep opportunity and regular sleep timing while checking for narcolepsy or another disorder when the clinical pattern does not fit isolated episodes.

The science is much stronger for what sleep paralysis is than for any specific medication or supplement treatment—and the article should stay honest about that gap.