How Much Alcohol Causes Nystagmus?

Alcohol can trigger involuntary eye movements, known as nystagmus, at blood alcohol concentrations (BAC) well below the legal driving limit in most countries. Positional nystagmus, the kind that appears when you tilt your head, has been recorded at BAC levels as low as about 0.02%, while gaze-evoked nystagmus, the jerky eye drift that occurs when you look to the side, shows up reliably around 0.06%. Even subtler disruptions to smooth eye tracking have been measured at BAC levels as low as 0.015%. The relationship between alcohol and your eyes is more layered than a single threshold, though, because different types of nystagmus emerge at different points and are driven by different mechanisms.

Two Main Types of Alcohol-Induced Nystagmus

Alcohol does not cause just one form of nystagmus. It causes several, and they show up at different BAC levels and under different conditions. The two most studied are positional alcohol nystagmus (PAN) and horizontal gaze nystagmus (HGN), sometimes called gaze-evoked nystagmus. Understanding the distinction matters because the type you experience depends on what you are doing with your head and eyes at the time.

Positional alcohol nystagmus is triggered by head position relative to gravity. When you lie down or tilt your head to one side after drinking, your eyes begin drifting involuntarily. It happens in two phases: PAN-I occurs during the rising and peak phase of intoxication, and PAN-II appears hours later, after your BAC has started falling. A study measuring this in healthy subjects found the minimum BAC threshold for PAN-I to be about 0.023%, roughly a quarter of the U.S. legal driving limit of 0.08%.¹ PAN-II, the delayed version, can persist long after someone feels sober and is particularly relevant for people who go to bed intoxicated and wake up feeling dizzy.

Gaze-evoked nystagmus, by contrast, does not require any particular head position. It appears when you try to hold your eyes at an eccentric position, like looking far to the left or right. Alcohol impairs the brain’s ability to hold the eyes steady, so they drift back toward center and then snap back to the target in a repetitive jerking pattern. This is the type that law enforcement officers look for during roadside sobriety checks.

How Alcohol Disrupts the Inner Ear and the Brain

The mechanism behind positional nystagmus is surprisingly physical. Inside each semicircular canal of the inner ear, a gel-like structure called the cupula sits in a fluid called endolymph. Normally the cupula and the endolymph have roughly the same density, so gravity alone does not deflect the cupula. Alcohol diffuses into the cupula faster than into the surrounding endolymph, making the cupula lighter. This density mismatch makes the cupula buoyant and sensitive to gravity, so when you tilt your head, the cupula floats upward and sends a false rotation signal to your brain.² That false signal produces both the involuntary eye movements and the spinning sensation many people feel when lying down drunk.

Gaze-evoked nystagmus works through a different pathway entirely. Holding your eyes off-center requires a neural circuit in the brainstem and cerebellum that acts as a kind of integrator, maintaining the signal that keeps your eye muscles contracted at just the right tension. Alcohol depresses this integrator’s function, so the holding signal “leaks,” and the eyes slowly drift back toward the middle. The brain then fires a corrective snap back to the intended position, creating the characteristic sawtooth pattern of nystagmus. A study of 15 healthy participants found that reaching a BAC of 0.06% caused a roughly twofold increase in the velocity of this eye drift at all gaze positions.³

The BAC Levels Where Eye Movements Start to Break Down

Rather than a single tidy number, the answer to “how much alcohol causes nystagmus” is a gradient. Different aspects of eye-movement control fail at different levels of intoxication, and the earliest changes appear at strikingly low BACs.

Smooth pursuit, the ability to track a moving object with your eyes, is one of the first functions to degrade. Research using precise eye-tracking instruments has detected impaired processing of target speed and direction at BAC levels as low as 0.015%, which for many people corresponds to less than one standard drink.⁴ This is not full-blown nystagmus in the clinical sense, but it represents the beginning of the same continuum of oculomotor impairment. At 0.035%, saccades (the rapid darting movements your eyes make when shifting focus) become more sluggish, showing lower peak velocities.⁴

By 0.06% BAC, the impairment is unambiguous. Smooth pursuit accuracy drops significantly, and saccadic velocities are measurably slower.⁵ Gaze-evoked nystagmus is reliably present at this level across subjects.³ And as noted above, positional nystagmus can appear at even lower concentrations when the person changes head orientation.

What catches many people off guard is the gap between when they feel impaired and when their eyes are measurably impaired. At 0.06% BAC, many drinkers report only mild subjective effects, yet laboratory instruments pick up clear oculomotor deficits. Your eyes betray the alcohol before your self-assessment does.

How the Roadside Eye Test Works

The horizontal gaze nystagmus test used by police officers during traffic stops takes advantage of the gaze-evoked nystagmus described above. An officer moves a stimulus, typically a pen or small flashlight, slowly across a person’s field of vision and watches for three indicators in each eye: whether the eye cannot follow the stimulus smoothly, whether nystagmus becomes distinct when the eye is at maximum deviation (looking as far to the side as possible), and whether nystagmus begins before the eye reaches a 45-degree angle from center. That last cue is the most important. In sober individuals, the eyes can usually track out to about 45 degrees without jerking. As BAC rises, nystagmus begins appearing at smaller and smaller angles.

An early study of this approach tested it in two groups: emergency room patients and people suspected of driving while intoxicated. Among the ER patients, 96% of those who were legally intoxicated showed HGN results consistent with impairment. Among the DWI suspects, the correlation between estimated BAC from the HGN angle and the actual breath-alcohol result was 0.878, which is a strong relationship for a field test done in imperfect conditions.⁶

There is a persistent idea that a positive HGN test means someone is above the legal limit. It does not. The test indicates impairment is present; it does not pinpoint a specific BAC. Officers are trained to use it as one piece of evidence alongside other field sobriety tests and, eventually, a chemical test. The correlation is good enough to establish probable cause for further testing, but it was never designed to replace a breathalyzer.

Why the HGN Test Is Less Precise Than It Seems

The roadside HGN test looks objective, and judges have sometimes treated it that way, but the underlying science has real limitations. A review in the forensic science literature found that the test is constrained by large variability in what “normal” eye movements look like across sober individuals. Nystagmus can appear in some completely sober people at or before 45 degrees, which is the very cue officers are watching for. The review also noted that the testing methods and environments used in validation studies are often poorly controlled and that rigorous laboratory validation is lacking.⁷

Several factors affect baseline smooth pursuit and gaze-holding ability in ways that have nothing to do with alcohol. Age is one: older adults tend to have less precise smooth pursuit. Medical conditions, prescription medications, and even psychiatric disorders can alter eye-movement stability.⁷ The conditions of a roadside stop, dim lighting, flashing emergency lights, the stress of the interaction, are also a far cry from a controlled laboratory.

None of this means the HGN test is useless. It means it is a screening tool with a meaningful false-positive rate, and its results should be interpreted alongside other evidence rather than treated as a standalone verdict.

Medical Conditions That Can Mimic Alcohol Nystagmus

One reason the HGN test’s precision is limited is the long list of things besides acute alcohol consumption that can produce nystagmus. An ophthalmologist’s review of the forensic evidence highlights several common confounders:

Vestibular disorders: Survey data from the early 2000s estimated that roughly 35% of U.S. adults aged 40 and older have some degree of objectively measured vestibular dysfunction, corresponding to about 69 million people. Vestibular problems directly alter the signals that control eye-movement stability.
Medications: Benzodiazepines, anticonvulsants, and certain antidepressants can all impair the same brainstem gaze-holding circuits that alcohol disrupts. A person taking prescribed medication could show HGN indistinguishable from alcohol-induced nystagmus.
Fatigue: Sleep deprivation and exhaustion impair oculomotor control and can produce eye-movement instability even in the absence of any substance.
Head trauma: If a traffic stop follows a collision, the trauma itself can cause nystagmus, making it difficult to distinguish injury-related eye movements from intoxication.
Chronic alcoholism: Long-term heavy drinking can cause cerebellar degeneration, which produces a persistent nystagmus that is present even when the person has not had a drink in days.
Caffeine: Even the standard law enforcement training manual acknowledges that caffeine can induce nystagmus in people with underlying vestibular problems.

The ophthalmologist’s review emphasized that distinguishing acute alcohol-related HGN from these other causes is genuinely challenging in the field, where officers do not have access to a person’s medical history or medication list.⁸ For someone who knows they have a vestibular condition or takes a medication that affects eye movement, being aware that these can produce a false positive on a field sobriety test is practically useful information.

How Cannabis and Other Substances Change the Picture

Alcohol rarely arrives alone. When combined with cannabis, the effect on HGN appears to be greater than either substance produces individually. A controlled study that gave participants alcohol, cannabis, and both together found that the combination was significantly associated with increased HGN impairment.⁹ The correlation was modest but consistent, and it reinforces the common clinical observation that polydrug use amplifies impairment beyond what each substance alone would predict.

This is relevant not just for law enforcement encounters but for understanding personal risk. If you have had even a small amount of alcohol and then use cannabis, the oculomotor impairment, and by extension the broader neurological impairment it signals, is greater than the amount of alcohol alone would suggest. The same principle applies to sedating medications like benzodiazepines or antihistamines combined with alcohol, though the nystagmus research is most developed for the alcohol-cannabis combination.

The Correlation Between Nystagmus Intensity and BAC

A natural follow-up question is whether you can estimate how drunk someone is by measuring how severe their nystagmus is. The answer is “somewhat.” In the study measuring positional nystagmus, the intensity of PAN-I correlated with BAC along the time axis, with a correlation coefficient of 0.62 at the one-hour mark after drinking. That is a moderate positive relationship, meaning higher BACs generally produced stronger nystagmus, but with plenty of scatter between individuals.¹

Body sway, another measurable sign of intoxication, showed a similar relationship. At peak BAC about one hour after drinking, the correlation between positional nystagmus intensity and body sway was 0.50, again moderate and positive.¹ The practical takeaway is that nystagmus intensity is a real signal of intoxication level, but it is a noisy signal. Two people at the same BAC can show noticeably different nystagmus severity, and a person with unusually strong nystagmus at 0.05% might look more impaired than someone else at 0.08%.

Why Individual Variation Is So Wide

If you have ever noticed that two people who have had the same number of drinks seem to be affected very differently, the same pattern holds for their eye movements. Several factors drive the variation.

Body composition and drinking history influence how quickly alcohol reaches the brain and how much the brain responds to a given concentration. People with a family history of alcohol dependence may show different oculomotor responses. One study looking at smooth pursuit and saccadic function found that alcohol impaired these at the first testing point, but the effects recovered faster at a second testing point, and risk factors for alcohol dependence influenced the pattern of recovery.¹⁰ In other words, the brain’s tolerance to alcohol’s oculomotor effects is not uniform across people, and it may be partly shaped by genetics and drinking experience.

Age plays a role as well. The neural integrator in the brainstem that holds your eyes in position naturally becomes less efficient with age, which means older adults start from a higher baseline of gaze instability. Add alcohol on top, and the nystagmus may become apparent at a lower BAC than it would in a younger person. Conversely, some younger adults with robust oculomotor circuits might not show obvious nystagmus until their BAC is considerably higher.

Then there is the simple but often overlooked factor of how quickly the alcohol was consumed. BAC is not a static number: it rises, peaks, and falls. Two people with identical peak BACs will have different nystagmus profiles depending on whether they drank steadily over three hours or quickly in 30 minutes, because the rate of alcohol diffusion into the inner-ear structures and the brainstem is not instantaneous, and the body’s compensatory mechanisms engage on a time-delay.

Nystagmus as an Early-Warning Biometric

Emerging research treats alcohol-induced nystagmus not as a binary yes-or-no indicator but as a continuous biomarker for impairment. Because eye-movement disruptions start at BAC levels far lower than those that produce visible behavioral impairment, some researchers see eye tracking as a potentially sensitive tool for detecting incapacitation early, whether in transportation safety, workplace monitoring, or clinical settings.

The finding that pursuit tracking degrades at just 0.015% BAC is particularly striking from this angle.⁴ That level is below what most adults would consider “having had a drink,” and yet the neural circuits governing visual motion processing are already measurably affected. Smartphone-based eye-tracking apps and wearable eye trackers are being explored as tools that could detect this kind of impairment in real time, though none have reached the reliability or regulatory approval needed for widespread use.

The appeal is that eye-movement testing is passive, fast, and does not require a blood or breath sample. But the same confounders that complicate roadside HGN testing, fatigue, medications, vestibular conditions, also apply here. Any practical system would need to establish a personal baseline and account for the many non-alcohol causes of oculomotor instability, which is a nontrivial engineering and medical challenge that the current generation of research is still working through.