That Falling Feeling as You Drift Off, Explained

You Were Almost Asleep. Then Your Whole Body Fired.

You know the one. You're sinking, comfortable, right on the edge — and then the floor drops out. Sometimes there's a flash of stepping off a curb or falling off a roof. Your leg kicks. Your arms snap. Your heart is suddenly going and you're wide awake, mildly embarrassed, staring at the ceiling.

It has a name: a hypnic jerk, also called a sleep start or hypnagogic jerk. Up to 70% of adults experience them at some point [^1]. It is, for almost everyone, completely harmless.

It's also one of those things nearly nobody knows anything about, which leaves a vacuum that gets filled with a very popular evolutionary story. That story is almost certainly wrong, and we'll get to it.

Here's what the sleep-medicine literature actually says — and, at the end, the short list of night-time movements that are worth mentioning to a doctor instead.

This is educational, not medical advice.

Key Facts

• Hypnic jerks are experienced by up to 70% of adults at some point in life (Alghamdi, Cureus, 2023)
• They are officially a normal variant, classified under "isolated symptoms and normal variants" in the sleep disorder manual (Stefani & Högl, 2019)
• They're a physiologic form of myoclonus — the same category as a normal sleep-transition twitch, not a seizure (StatPearls, Myoclonus)
• The leading explanation is brainstem instability during the switch from wake to sleep, with a burst of autonomic arousal (Alghamdi, 2023)
• Commonly reported triggers: fatigue, stress, sleep deprivation, vigorous exercise, caffeine and nicotine (Alghamdi, 2023)
• Only "excessive" jerks are a problem — frequent and severe enough to cause insomnia (Stefani & Högl, 2019)

What's Actually Happening

Falling asleep is not a switch. It's a handover, and it's messier than it feels.

The first stage of sleep is called N1 — the lightest stage, lasting roughly 1 to 5 minutes and making up about 5% of a night's sleep ¹. During N1 your brain's electrical pattern shifts from the alert alpha rhythm to slower, lower-voltage theta activity. But here's the part that matters: muscle tone is still present in N1 ¹. Your body has not yet been switched off the way it will be later.

So during that handover you have systems in different states. Some parts of the brain are heading for sleep. The circuits that control movement haven't fully stood down.

The leading explanation for a hypnic jerk is that during this unstable window, a burst of signal comes down from the brainstem reticular formation — a network deep in the brainstem that regulates arousal — and lands on muscles that are still able to respond ². The result is an abrupt contraction involving nearly the whole body, plus a wave of autonomic arousal: fast heart rate, fast breathing, sometimes sweating ².

In plain terms: the brain fumbles the handoff, and the body twitches.

Why It Feels Like Falling

The jerk itself is only half the experience. The falling sensation, the dream flash of missing a step — that's the other half, and it's a different thing.

At sleep onset your brain is already generating imagery. In one overnight study that collected reports across the whole night, dream-like content turned up at sleep onset too, not only later ³. So the most straightforward reading is that your half-asleep brain gets a large, unexplained physical signal — a full-body jolt — and does what brains always do with unexplained input. It builds a story that fits. Falling explains a sudden lurch better than almost anything else.

That's a plausible account, not a proven one. Nobody has demonstrated the direction of causation here, and I'm not going to pretend otherwise.

The Tree Story Is Not Evidence

You've probably heard it: hypnic jerks are a leftover reflex from primate ancestors who slept in trees, a jolt that kept them from falling out of the branches.

It's a great story. It's also just a story.

Look at what the actual sleep-medicine literature says about hypnic jerks. The reviews describe brainstem-level instability, motor pattern analysis, differential diagnosis, and treatment for the rare severe cases ^{2 4}. The arboreal explanation isn't in there, because there's nothing to cite. You cannot test it. There's no fossil record of sleep twitches, no comparative data that would distinguish it from a dozen other explanations, no prediction it makes that could fail.

And it doesn't fit well. If the reflex existed to prevent falls, you'd expect it every night, in everyone, reliably. Instead it's sporadic, worse when you're exhausted or over-caffeinated, and absent most nights for most people. That's the profile of a glitch, not a safety system.

Believe it if you like it. Just don't call it science.

What Makes Them More Frequent

The clinical literature lists the usual suspects: fatigue, stress, sleep deprivation, vigorous exercise, and stimulants including caffeine and nicotine ².

Worth being straight about the quality of that evidence. This list comes from clinical observation and case reports, not randomized trials. Nobody has run a controlled study dosing people with caffeine and counting sleep starts. The list is plausible and consistent across sources, and it's also the softest part of this post.

That said, every item on it has an independent reason to be there.

None of these are cures. Hypnic jerks in a healthy person don't need curing. They're a nuisance, and the honest fix is the boring one: be less exhausted and less wired at bedtime, which happens to be good advice regardless.

When It's Actually Excessive

There's a rare version worth naming. When sleep starts become frequent and severe enough to cause insomnia, sleep specialists call it intensified hypnic jerks ⁴. Some people develop real anxiety about falling asleep, which is its own spiral — jerk, wake, worry about jerking, stay awake.

Sleep doctors also work to distinguish hypnic jerks from a few things that look similar but aren't:

• Propriospinal myoclonus at sleep onset — jerks that start in the abdomen or trunk and spread up and down the body in a consistent pattern. The propagation pattern is how specialists tell the two apart ⁴.
• Periodic limb movements — repetitive, stereotyped leg movements that recur through the night rather than only at sleep onset. The clinical threshold in adults is more than 15 per hour of sleep, and periodic limb movement disorder is estimated at 4% to 11% of adults ⁵.
• Seizures. Hypnic jerks resemble myoclonic seizures, which is why the distinction gets made carefully rather than casually ⁶.

When to Bring It Up With a Doctor

Most night-time movement is nothing. Some of it isn't. Here's the split:

That fifth row on the right deserves a note. Physically acting out dreams during REM sleep is REM sleep behavior disorder, and it is not a hypnic jerk — it happens at least 90 minutes after sleep onset and more often in the second half of the night ⁷. It's uncommon overall (around 0.5% of the general population, rising to 5% to 13% in adults over 60) and it needs a real evaluation, because it's associated with later neurological disease ⁷. Diagnosis requires an in-lab sleep study with video ⁷.

That's not a reason to panic about a twitch as you doze off. It's a reason to know which is which, and to let a clinician sort it out rather than a search bar.

Also: first-episode sleepwalking that begins in adulthood is a prompt to investigate other causes, not something to wait out ⁸.

Where Komori Fits

Honest framing: Komori is not designed to detect or diagnose hypnic jerks, or any other medical condition. It's a general wellness device, not a medical or diagnostic one, and it hasn't shipped yet.

What it's being built to do is narrower. Komori is a contactless system that senses from above the bed — no camera, nothing worn — with a base unit on the nightstand handling power, Wi-Fi, and local processing. When it ships, it's designed to log sleep position, movement and restlessness, bed-exit events, sound events, and room conditions like temperature, humidity, CO2, light, and noise.

The relevant part here is movement and restlessness over many nights, which is context, not diagnosis. A single jolt at sleep onset is not something a wellness device should be interpreting for you. Whether your nights are generally settled or generally restless, and whether that's changed since you moved your last coffee earlier, is the kind of pattern that only shows up across weeks. We've written about what movement actually says about sleep and how often normal people move at night if you want the longer version.

Nothing Komori is designed to produce is a substitute for a sleep specialist's evaluation. If your movements at night worry you, that's the door to knock on.

The Bottom Line

That falling feeling is your brain fumbling the handoff into sleep and your body reacting before the lights are fully out. Up to 70% of adults have felt it. It's officially catalogued as a normal variant, and it means nothing about your health.

Ignore the tree story. Blame the espresso, the stress, and the three short nights that came before.

And if the jerks are frequent enough to make you dread bed, or you're doing something at night that looks like more than a twitch, bring it to a doctor. That's a conversation with a clear answer, and it's a short one.

---

Footnotes

1. Patel AK, Reddy V, Shumway KR, Araujo JF. "Physiology, Sleep Stages." StatPearls Publishing, updated 2024. ↩ ↩²

2. Alghamdi SA. "Hypnic Jerks, Major Depressive Disorder, and Antidepressant Use: A Possible Relationship." Cureus, 2023;15(10). Case report with background review. ↩ ↩² ↩³ ↩⁴

3. Wamsley EJ, Stickgold R. "Dreaming of a learning task is associated with enhanced memory consolidation: Replication in an overnight sleep study." Journal of Sleep Research, 2019;28(1):e12749. ↩

4. Stefani A, Högl B. "Diagnostic Criteria, Differential Diagnosis, and Treatment of Minor Motor Activity and Less Well-Known Movement Disorders of Sleep." Current Treatment Options in Neurology, 2019;21(1):1. ↩ ↩² ↩³

5. Drakatos P, et al. "Periodic limb movements during sleep: a narrative review." Journal of Thoracic Disease, 2021;13(11):6476–6494. ↩

6. Sanders AE, Zafar N, Sharma S. "Myoclonus." StatPearls Publishing, updated 2024. ↩

7. Pham CK, Sankari A, Slowik JM. "Rapid Eye Movement Sleep Behavior Disorder." StatPearls Publishing, updated 2024. ↩ ↩² ↩³

8. Fariba KA, Tadi P. "Parasomnias in Adults." StatPearls Publishing, updated 2023. ↩