Learn › Head impulse test explained
The head impulse test explained: testing each semicircular canal
This guide explains how the head impulse test probes each of the six semicircular canals, what a corrective saccade tells you, and where the video version (vHIT) helps and misleads. It finishes with a hands-on exercise to fix the canal planes in your mind.
- Identify the three push-pull canal pairs and the head movement that tests each canal
- Interpret overt and covert corrective saccades and a reduced gain
- Recognise the limits of the head impulse test, including in central and Ménière's disease presentations
What the head impulse test measures
The vestibulo-ocular reflex (VOR) turns the eyes opposite to head movement so that gaze stays on a target. The head impulse test (HIT) challenges this reflex with a small, brisk, unpredictable head rotation and checks whether the eyes stay on target. Because the movement is fast, it tests the canals at the high frequencies typical of walking and turning. Caloric testing, by contrast, stimulates at very low frequency.
Impulsive testing of single canals was first done in 1988 with scleral search coils. That method stayed in research laboratories. Video methods that test each of the six canals individually followed in 2009 and 2013, and many dizzy clinics now use them. In some, video testing has replaced caloric testing as the first-line test.1
At the bedside, the patient fixes on the examiner’s nose. The examiner holds the head and delivers a rapid rotation of about 10 to 20 degrees, in a direction the patient cannot predict. The examiner watches the eyes. In a healthy person they stay on the nose.
Push-pull pairs: which canal is being tested
The six canals form three coplanar pairs, one canal from each ear. A head rotation in the plane of a pair excites one canal and inhibits its partner. This is the push-pull arrangement described in Ewald’s laws.
| Pair | Canals | Impulse delivered |
|---|---|---|
| Horizontal | Right and left horizontal | Yaw (turn) towards the canal being tested |
| LARP | Left anterior, right posterior | Pitch in the LARP plane: forward tests anterior, backward tests posterior |
| RALP | Right anterior, left posterior | Pitch in the RALP plane: forward tests anterior, backward tests posterior |
Ewald’s second law explains why the test works. Excitation of a canal produces a stronger response than inhibition of its partner. A high-acceleration impulse towards a damaged ear therefore exposes the deficit, because the healthy partner cannot fully compensate when it is inhibited. A slow rotation does not expose it.
The canal planes dictate the head position. The horizontal canals tilt up at the front by roughly 30 degrees in the upright head, so the head is flexed forward by about that amount to bring them into the plane of rotation. The vertical canals lie about 45 degrees from the sagittal plane. For vertical testing the head is therefore turned (commonly about 35 to 45 degrees), so that the plane of one pair lines up with a simple forward-and-back pitch. See semicircular canal anatomy for the orientations.
Corrective saccades: what they mean
A saccade is a rapid eye movement that shifts gaze. In a normal HIT, the eyes rotate opposite to the head at nearly the same speed, so no saccade is needed. If the excited canal is weak, eye velocity lags behind head velocity and gaze is carried off the target. The brain then generates a corrective (catch-up) saccade to bring the eyes back.
- Overt saccade: occurs after the head movement has finished and can be seen at the bedside.
- Covert saccade: occurs during the head movement and is usually too quick to see. Video recording can detect it.
Reviews report that vHIT is more sensitive than the clinical HIT, particularly when only covert saccades are present.2 A reduced gain (eye velocity divided by head velocity) with catch-up saccades supports a peripheral weakness of the canal being tested. Strictly, it localises the problem to the canal or its afferent pathway, not to a specific disease.
Saccades do not always mean a peripheral lesion. Catch-up saccades usually indicate peripheral hypofunction. In acute vestibular syndrome, however, a normal bedside HIT should prompt a search for a central cause. Quantitative studies also describe abnormal gain patterns in some brainstem and cerebellar lesions.3 Always interpret the HIT alongside nystagmus, skew deviation and gait.
Video head impulse test (vHIT) basics and limits
vHIT uses lightweight goggles with a camera that tracks the eye, and a motion sensor that records head velocity. The software reports gain for each canal and displays any saccades. Normal values and cut-offs differ between devices and canals, so use the manufacturer’s or your department’s norms.
Technique matters. The interpretation depends on factors such as goggle orientation on the head, head pitch, and the contribution of the vertical canals to the horizontal response.4 A poorly fitted goggle or a slipping strap can create an apparent deficit.
A normal vHIT does not mean normal vestibular function. In a meta-analysis of patients with Ménière’s disease, the caloric test was abnormal in 64% but vHIT in only 28%.5 The two tests sample different frequency ranges, and a dissociation between them is not a contradiction.
vHIT is also being studied in emergency departments. Several studies suggest that VOR gain can help to separate posterior circulation stroke from vestibular neuritis. The authors of one review state that further research is needed before it is used to rule out stroke.6 Treat it as one finding within a structured examination, not as a stand-alone decision tool.
Try it with your model
Use Vestibular3D (available here) to link each impulse direction to a canal. A 9 cm or 14 cm model is easiest to demonstrate to a group.
- Hold the model facing you and identify the two horizontal canals. Flex the head forward about 30 degrees and confirm that both horizontal canals now lie in the plane of a yaw rotation.
- Turn the model’s head briskly to the right. Use the + and − markings on the ampullae to state which horizontal canal is excited and which is inhibited. Repeat to the left.
- Return the head to neutral, then turn it about 35 to 45 degrees. Trace the plane of one anterior canal and the opposite posterior canal and confirm that they lie in the same plane. Name the pair (LARP or RALP).
- Pitch the head forward, then backward, in that plane. For each direction, say which canal is excited and which partner is inhibited. Then say which direction you would use to test the left posterior canal.
When to escalate
- Acute continuous vertigo with a normal head impulse test, especially with direction-changing nystagmus, skew deviation or inability to walk unaided: treat as possible stroke and arrange urgent assessment.
- New neurological signs (diplopia, dysarthria, limb weakness, facial numbness) or a new severe headache with dizziness.
- Recent neck trauma, severe neck pain or suspected vertebral or carotid dissection: do not perform vigorous head impulses; seek urgent medical review.
- Sudden hearing loss with vertigo: refer urgently for audiometry and specialist assessment.
- The six canals form three push-pull pairs (horizontal, LARP, RALP); an impulse excites one canal and inhibits its partner.
- The head impulse test is a high-frequency test of the VOR for individual canals, relying on stronger excitatory than inhibitory responses (Ewald's second law).
- A corrective saccade means eye velocity fell short of head velocity. Overt saccades are visible at the bedside; covert saccades need video recording.
- A normal bedside HIT in acute vestibular syndrome should prompt a search for a central cause.
- A normal vHIT does not exclude vestibular disease, as caloric and vHIT results often differ in Ménière's disease.
- Technique, goggle fit and head position affect vHIT results, so check them before trusting a gain value.
Check yourself
A 62-year-old man has had continuous vertigo and vomiting for 6 hours. He cannot stand unaided and has spontaneous nystagmus. The bedside head impulse test is normal in both directions. What does this suggest?
A central cause, such as posterior circulation stroke, should be suspected. In acute vestibular syndrome with spontaneous nystagmus, a normal HIT (no catch-up saccade) argues against vestibular neuritis and calls for urgent assessment and imaging as local pathways dictate.
A vHIT shows reduced gain and catch-up saccades for the left posterior canal. Which canal pair is involved, which canal is its partner, and what impulse tests it?
The left posterior canal belongs to the RALP pair, with the right anterior canal as its partner. With the head turned so the RALP plane is aligned with the pitch movement, a backward pitch excites the left posterior canal.
A patient with probable Ménière's disease has a reduced caloric response on the right but a normal vHIT in all six canals. Is this result contradictory?
No. Caloric testing assesses the horizontal canal at very low frequency, while vHIT assesses high-frequency function. In Ménière's disease the caloric test is abnormal more often than vHIT (64% versus 28% in one meta-analysis), so a dissociation of this kind is common.
- Halmagyi GM, Chen L, MacDougall HG, et al. The Video Head Impulse Test. Front Neurol 2017;8:258. doi:10.3389/fneur.2017.00258
- Alhabib SF, Saliba I. Video head impulse test: a review of the literature. Eur Arch Otorhinolaryngol 2017;274(3):1215-1222. doi:10.1007/s00405-016-4157-4
- Choi JY, Kim HJ, Kim JS. Recent advances in head impulse test findings in central vestibular disorders. Neurology 2018;90(13):602-612. doi:10.1212/WNL.0000000000005206
- Curthoys IS, McGarvie LA, MacDougall HG, et al. A review of the geometrical basis and the principles underlying the use and interpretation of the video head impulse test (vHIT) in clinical vestibular testing. Front Neurol 2023;14:1147253. doi:10.3389/fneur.2023.1147253
- Tamanini JB, Mezzalira R, Vallim MGB, et al. Dissociation between video head impulse test and caloric test: a marker of menière's disease? – A systematic review and meta-analysis. Braz J Otorhinolaryngol 2023;89(4):101279. doi:10.1016/j.bjorl.2023.101279
- Jaganathan N, Mohamed MH, Md Pauzi AL, et al. Video head impulse test in stroke: a review of published studies. Front Neurol 2024;15:1339039. doi:10.3389/fneur.2024.1339039
Written by the Vestibular3D Clinical Team. For education only, not medical advice. Reviewed by a Consultant Neurologist and Neuro‑Otologist.