How your body knows which way is up — three senses cross-check each other, and the trouble starts when they disagree

Sit in the back of a moving car and read a book for ten minutes. Most people start to feel unwell, and nobody has to be told why it might happen — it just does. What’s actually going on is worth understanding, because it exposes how balance works at all.

Balance is not a sense. It is a comparison. Three separate systems measure your motion independently, and a set of nuclei in your brainstem checks their answers against each other [50]. Agreement is silent. Disagreement is the whole story.

Two instruments in the inner ear

Buried in the bone behind each ear is a set of fluid-filled tubes and chambers called the labyrinth. It holds two different measuring devices, not one [50].

The three semicircular canals — loops set at roughly right angles to each other — detect rotational acceleration. Turn your head and the fluid inside lags behind, bending a bundle of hair cells that convert the movement into nerve signals [50]. Three loops at three angles means any rotation, in any plane, registers on at least one of them.

The otolith organs, the utricle and saccule, do a different job. They detect linear acceleration and the pull of gravity — speeding up, slowing down, tilting [50]. They manage this with a small weight: a layer of tiny calcium carbonate crystals sitting on a jelly bed over more hair cells. The crystals have mass, so they lag when you accelerate and slump when you tilt, dragging the hair cells with them [54].

Two instruments, two quantities. Neither can do the other’s job.

And two more witnesses outside it

Your eyes answer the same question by watching the world slide past. Your muscles, joints and skin answer it too — the stretch in your ankles, the pressure under your feet, the tension in your neck. That sense of body position is called proprioception, and it feeds into balance continuously [30].

All three streams converge on the vestibular nuclei in the brainstem, where they’re integrated into one reading of where you are and how you’re moving [50] [34]. From there, signals go out to the eye muscles, to the spinal cord for posture, and up to the cortex [50]. That spread is why a purely mechanical problem in the ear can produce a whole-body sensation.

The fastest of those outputs is the vestibulo-ocular reflex. When your head turns one way, this reflex drives your eyes the same amount the other way, so your gaze stays locked on what you’re looking at [55]. It is why you can read a sign while walking. Shake your head now and the text stays sharp; shake this page instead and it blurs. The difference is that reflex.

What happens when they disagree

Read in a car and the disagreement is stark. Your inner ear reports acceleration, turns and bumps. Your eyes, fixed on a page that moves with the car, report stillness. The brainstem gets two confident, contradictory answers about the same body.

This is the sensory-conflict account of motion sickness, the standard explanation for decades. The leading model treats sickness as an accumulation of conflict rather than a single trigger [7]. It explains the geometry well: reading in a car, sitting below deck on a ship with no horizon, or wearing a headset that shows motion your body never feels.

That last case is now the common one. In one study, around 80% of people who had used a virtual-reality headset before still reported discomfort after just ten minutes in a virtual environment [12]. The symptoms — nausea, dizziness, disorientation — match ordinary motion sickness closely enough that researchers group them together [12].

The account is not complete, though. The newer framing shifts the emphasis from the senses arguing with each other to the brain’s prediction being wrong. Richard Lewis, a neurologist at Harvard-affiliated Massachusetts Eye and Ear, puts it as a mismatch between “what your brain predicts” about your movement and what actually happens [64]. Same family of idea, different accused party. Researchers also point out that neither version explains why susceptibility varies so enormously between people; about a third are highly susceptible, and some are barely affected at all [64] [7].

Once nausea passes a certain point, it can run to vomiting even after the motion stops — clinicians call it the avalanche phenomenon [56].

When one witness starts lying

The most common cause of true spinning vertigo is a purely mechanical fault: some of those calcium crystals break loose from the utricle and drift into a semicircular canal, where they don’t belong [54]. Now, when you roll over in bed, the loose crystals shift and slosh the canal fluid — so the rotation sensor reports a spin that isn’t happening. The eyes and body report no such thing.

This is benign paroxysmal positional vertigo, and it accounts for more than half of all peripheral vertigo cases [65]. Around 20% of people who turn up complaining of vertigo have it, and that figure is probably an undercount, since it’s frequently missed [65]. Dizziness in general drives over three million emergency-department visits a year in the United States [65]. In older adults it can present as falls and unsteadiness without the classic spinning at all [54].

Because the problem is a displaced particle, the standard treatment is mechanical too. A sequence of head positions — the Epley manoeuvre and its relatives — walks the crystals back out of the canal [27] [54].

A different fault is inflammation of the vestibular nerve itself, vestibular neuritis, which cuts the signal from one ear. It affects roughly 4 in 100,000 people [38]. The brain suddenly hears a strong signal from one side and near-silence from the other, and reads that imbalance as violent, continuous motion.

Re-learning, and what the evidence supports

The striking thing is what fixes these states. Not repairing the sensor — recalibrating the comparison.

Sailors are the everyday case. Seasickness affects most naval personnel, with around 20% needing medication and rest in one study. Yet people generally adjust within a few days of continuous exposure [34]. Come back to land and many feel the swaying return for a while, which has a name — mal de débarquement [34]. Astronauts show the same pattern in reverse, with space motion sickness common in the first 72 hours of weightlessness [34]. After six-month missions, cosmonauts show measurable changes in an otolith-driven eye reflex, and first-time flyers change most [70].

Vestibular rehabilitation therapy works on this principle deliberately: repeated, graded head and eye movements that push the brain to re-weight its inputs and compensate for a weak ear [19]. A 2025 meta-analysis of eight trials in people with persistent postural-perceptual dizziness found meaningful improvement on a standard dizziness questionnaire, with customised programmes outperforming virtual-reality-based ones [42]. Importantly, the same clinical guidance is blunt that these exercises are not the answer for BPPV, which needs the repositioning manoeuvre instead [19].

Here is the honest catch about drugs. Antihistamines and antimuscarinics do damp motion sickness, but they cause drowsiness and — per clinical guidance — may impede habituation, the very adaptation that would otherwise solve the problem [56]. The symptom relief and the long-term fix pull in opposite directions.

On the consumer end, expect less. Acupressure wristbands and electrical wrist-stimulation bands sell from under $10 to around $250. They rest on the claim that stimulating a nerve in the wrist dampens nausea signals. The evidence behind them is far weaker than the marketing suggests [64]. Ordinary measures with better support are dull: face forward, watch the horizon, don’t read, keep your head still [64].

None of this is a substitute for assessment. New, severe, or persistent dizziness has causes ranging from a loose crystal to something serious in the brainstem. Telling them apart is a clinician’s job, not a checklist’s [65] [51].