The HGN Eye Test and What It Can and Cannot Show

The eye test is the one people remember least and that hurts them most.

You stood in the dark following a pen. You have no idea what the officer saw, because you cannot see your own eyes. There is nothing on the video that a jury can independently evaluate. The officer’s word is the evidence.

That is exactly why the protocol matters so much. Every step of this test is written down: how far the stimulus is held, how fast it moves, how long it is held at the side, what has to be checked first. Departures are documented failures rather than opinions, and Texas DWI lawyers have had considerable success suppressing HGN results for improper administration.

What HGN Is and How It Is Scored

Nystagmus is an involuntary jerking of the eye. The HGN test asks you to follow a stimulus, the officer’s pen, fingertip, or a small light, with your eyes while you hold your head still, and the officer watches for that jerking as your eyes move side to side. Because the movement is involuntary and cannot be consciously controlled, prosecutors often present HGN as the most reliable of the three standardized field sobriety tests.

Before the test starts, the officer is supposed to do three medical checks: equal pupil size, resting nystagmus (jerking while the eyes look straight ahead), and equal tracking (whether both eyes follow the stimulus together). These are screens for medical impairment. Unequal pupils, resting nystagmus, or eyes that do not track together can indicate a head injury, a neurological condition, or a prosthetic eye, and the officer is also trained to ask about head injuries and eye conditions first. If those checks come back abnormal, the officer may choose not to continue, and if the officer continues anyway, the training materials say that is a departure from the standardized protocol that should be noted in the report.

Then the officer looks for three clues in each eye:

• Lack of smooth pursuit. The eye jerks instead of tracking the stimulus smoothly.
• Distinct and sustained nystagmus at maximum deviation. The eye jerks when held as far to the side as it will go.
• Onset of nystagmus before 45 degrees. The jerking begins before the eye reaches a 45-degree angle.

Six clues total, three per eye. Four or more is the threshold for a positive result.

One thing an officer sometimes says on the stand does not come from the training materials at all: that six clues means more intoxicated than four. NHTSA does not claim that more clues means a higher concentration, and an officer who testifies otherwise should be asked for the basis.

The Emerson Limit: What HGN Can Never Prove

The Texas case governing HGN is Emerson v. State, decided by the Court of Criminal Appeals in 1994. It sits on top of Kelly v. State, which two years earlier set the Texas standard for scientific evidence. Under Kelly, scientific evidence is admissible only if three things hold, proven by clear and convincing evidence: the underlying scientific theory is valid, the technique applying that theory is valid, and the technique was properly applied on the occasion in question.

In Emerson the court did its own research into the science and settled the first two prongs once and for all, holding that the theory underlying HGN and the technique as designed and promoted by NHTSA are reliable under Rule 702. The State does not have to re-prove those in every case. What is still fought case by case is the third prong: whether this officer administered this test correctly on this occasion.

And Emerson drew a line the State sometimes blurs. A witness qualified as an expert on the administration and technique of the HGN test may testify about a defendant’s performance on the test, but may not correlate that performance to a precise blood alcohol concentration. The eye test can support an opinion that a driver had been drinking and was impaired. It cannot put a number on it. A prosecutor or an officer who implies that a six-clue HGN shows a particular BAC is claiming something Texas law does not allow.

There is a qualification requirement too, and it is easy to miss because it happens fast. HGN is treated as scientific evidence, not as ordinary lay observation, so before the officer can give even the permitted opinion, the State has to qualify the officer as an HGN expert. Under Emerson, an officer who has received practitioner certification satisfies that requirement without further proof. Proficiency certification is not required. Those certification records are maintained by the Texas Commission on Law Enforcement and can be obtained, which means the claim is checkable rather than something to take on the officer’s word.

Why the Motion in Limine Exists

Emerson is a rule about what a witness may say. Rules about what a witness may say are worth very little once the jury has already heard the thing. That is the entire reason a motion in limine exists on these cases: it gets the ruling before the words come out, rather than after.

Three things it is typically aimed at.

Putting a number on the eye test. This is the Emerson bar itself. Without a ruling in advance, an officer can slide from “I observed six clues” to testimony that implies a concentration, and an objection sustained afterward does not unring it.

Inflated accuracy claims. Officers are frequently prepared to tell a jury that the NHTSA validation studies show the tests are around 90 percent accurate. That framing misstates what those studies measured. One approach is to keep the claim out entirely. The other is to let it in and then take it apart with the underlying data. Which is better is a strategy call that depends on the case and the jury.

Vertical gaze nystagmus. VGN is the up-and-down version of the test, and it stands on much weaker ground than HGN. It was not part of the original research, it has no clue count and no grading criteria, and there is no peer-reviewed literature validating it as a sign of intoxication. It is a standard candidate for exclusion.

Separately from limine practice, where the video or the officer’s own testimony shows the protocol was not followed, the remedy is a hearing on admissibility aimed at the third Kelly prong. Texas courts have suppressed HGN evidence on exactly that basis, including where the State could not prove up the officer’s training and compliance with NHTSA procedure. That work is part of the broader suppression practice on Motion to Suppress.

What Else Causes Nystagmus

This is the heart of an HGN challenge. The jerking is involuntary, which is what makes it persuasive, but the same involuntary jerking has many causes that have nothing to do with drinking. The training materials themselves name several categories, including vestibular and neural forms of nystagmus, and note that pathological conditions such as brain tumors, brain damage, and diseases of the inner ear can produce it.

Documented non-alcohol causes include:

• Inner ear and vestibular disorders such as benign paroxysmal positional vertigo, labyrinthitis, vestibular neuritis, and Meniere’s disease.
• Neurological conditions including multiple sclerosis, brain stem lesions, cerebellar dysfunction, and traumatic brain injury.
• Naturally occurring nystagmus present at baseline in a share of the population.
• Certain prescription medications, including anticonvulsants, sedatives, and some antihistamines.
• Fatigue.
• Gaze-evoked nystagmus that appears at extreme lateral gaze in many people regardless of impairment.
• The conditions of the stop itself. The training materials have warned officers about wind and dust irritating the eyes, and about visual distractions, directing that the subject always be faced away from rotating lights, strobe lights, and traffic passing close by.

Peer-reviewed research has also found nystagmus in a meaningful share of subjects before they were given any alcohol at all, and in a larger share of subjects at very low concentrations during elimination. The medical literature has been notably unimpressed with the roadside version of this test: a standard ophthalmology reference has criticized its use by briefly trained officers and argued it should be discontinued. Where those findings came from, and how the battery was adopted nationally anyway, is on A Critical History of the SFSTs.

An officer who observes nystagmus but does not know, and does not ask, whether you have an inner ear condition, a neurological history, or a relevant medication cannot reliably attribute the jerking to alcohol. Where a diagnosed condition is in play, that is developed with medical records rather than from the video, and it is covered on Medical Conditions Defense. The ordinary characteristics and roadside circumstances that produce clues in sober people are on Who Fails Field Sobriety Tests While Sober.

How Officers Get the Test Wrong

The HGN is the most time-sensitive of the three tests, and its protocol is specific enough that deviations are measurable on video rather than arguable.

The stimulus is held about 12 to 15 inches from your nose and slightly above eye level. For lack of smooth pursuit, it moves smoothly out to the side at a pace of roughly two seconds each way, with at least two complete passes per eye. For nystagmus at maximum deviation, it is held at the far edge for a minimum of four seconds per eye, and the training materials are explicit that four seconds will not itself induce fatigue nystagmus, which needs far longer to appear. For onset before 45 degrees, the stimulus moves slowly, at a pace that would take about four seconds to reach the 45-degree point, and when jerking appears the officer must stop, hold steady, and verify that it continues.

That last instruction contains the most useful sentence in the whole protocol. If the officer has to convince himself or herself that nystagmus is present, it is not there.

Common deviations, all visible on footage:

• Moving the stimulus too fast, which is the single most consequential error and the one the 2007 NHTSA study on protocol variation examined most directly. In that study, varying the stimulus speed from the standard two seconds to one second produced four or more clues in roughly three quarters of subjects who were under 0.08.
• Holding it at the wrong distance, or too high or too low.
• Not holding at maximum deviation long enough to score a clue that requires sustained jerking.
• Making only one pass per eye where the protocol requires two.
• Skipping the pre-test medical checks, or performing them too quickly to mean anything.
• Facing the subject toward the patrol car’s flashing lights or passing traffic.
• Not recording the test on video at all, which removes the only means of reviewing any of the above.

None of that is a general complaint about the science. It is the third Kelly prong, which is the prong Emerson left open, and it is why the administration itself is where the case is usually won or lost.

The Other Two Standardized Tests

The eye test is one of three, and it is the only one of the three that Texas treats as scientific evidence with a qualification requirement attached. The other two are balance and divided attention tasks, scored by counting clues, and they fail in different ways. The walk-and-turn, with its eight clues and its two-clue failing line, is on The Walk-and-Turn Test. How the three fit together, where the battery came from, and whether you had to perform any of them are on Field Sobriety Tests.

Common Questions

Texas HGN Test FAQ

What is HGN?

HGN stands for horizontal gaze nystagmus, an involuntary jerking of the eye that appears as the eye tracks an object toward the side. Alcohol makes it show up earlier and more distinctly. During a Texas DWI stop the officer moves a stimulus across your field of vision and counts clues in each eye.

What does the HGN test prove?

Under Emerson v. State, an officer may testify about how you performed on the test and may give an opinion that you had been drinking and were impaired. The officer may not correlate that performance to a precise blood alcohol concentration. HGN cannot put a number on your BAC, and testimony implying otherwise claims something Texas law does not allow.

Can you fail the HGN test without being drunk?

Yes. The same involuntary jerking has documented causes unrelated to alcohol, including inner-ear and vestibular disorders, multiple sclerosis and other neurological conditions, traumatic brain injury, certain prescription medications, fatigue, and naturally occurring nystagmus present at baseline in part of the population. The conditions of the stop matter too: the training materials direct officers to face the subject away from rotating lights, strobes, and passing traffic.

How many clues does the officer look for on HGN?

Six in total, three in each eye: lack of smooth pursuit, distinct and sustained nystagmus at maximum deviation, and onset of nystagmus before 45 degrees. What is contested case by case is not whether the underlying science is valid, which Emerson settled, but whether this officer administered this test correctly on this occasion.