Optokinetic nystagmus, or OKN for short, is the eye movement elicited by the tracking of a moving field. It differs from smooth pursuit which is the eye movement elicited by tracking of a single distinct target. As moving fields contain within them distinct targets, OKN generally contains within it smooth pursuit. As a consequence, usually OKN performance (gain -- ratio of eye tracking velocity to target velocity), exceeds that of smooth pursuit.
Practically, OKN is somewhat difficult to elicit now due to the adoption of video frenzel goggle systems for eye movement monitoring. Most of these allow viewing of the optokinetic stimulus through a small aperature in the goggle, reducing the potential full field stimulus to a small aperature. One would think that a better methodology would involve using a VR system, equipped with an eye movement monitor. More easily said than done.
|Example of OKN being elicited within a rotatory chair enclosure.|
An optokinetic stimulus should present moving targets in both the central and peripheral visual fields. There are several methods of doing this:
- light bar
- stripe projector
- LCD projector or large screen array
- physical drum
The stimulus that most closely matches the definition of OKN is to use a full field surround such as is shown above and below. Even these are compromises as individual bars can be tracked, and there are usually distinct objects in the visual field that can be fixated. Accordingly, the "best" OKN stimulus is real, physical surround that rotates, containing a pattern which has no distinguishing features such as a random dot pattern.
|Optokinetic stripe projector (image courtesy of Neurokinetics.||Projection of stripes on wall suitable for evoking OKN and OKAN|
Stripe projector -- the slotted can shown in the figure above is a good method of producing OKN. While the projector shown above is rather expensive, cheap devices such as used for "disco" parties may be usable. The main problem with it, and other projected systems, is that people can fixate on still objects in the room. Still, we think that this is a reasonable method.
LCD projector -- a convenient but rather ineffective method of eliciting OKN is to use the LCD projector. The LCD projector has the same problem as the stripe projector -- people can fixate. Also, because it is not anywhere near a full field, it does not stimulate the entire peripheral field. It is marginally better than a light bar. We recommend against using this method.
A variant of the LCD projector method is to use an array of large televisions arranged in a half circle. Above is an example of this from our practice in Chicago. Patients in our practice can run the OKN stimulus with login credentials can run the software using this link. We think this is better than any of the projection systems, but not as good as the physical drum (see below).
Physical drum -- the best method of eliciting OKN is to actually rotate a physical drum around the person. A light-weight drum can be constructed out of cloth or foam-core, and be driven with a servo motor. If you have the space to do this (and also a lot of interest in OKN), this is the best way to do it.
|Light bar. An inferior method of eliciting smooth pursuit and OKN.|
It is not easy to build a good OKN full field stimulator, and clinicians as well as commercial vendors have attempted to use other devices.
A recent problem with any full field stimulator is finding a method of monitoring the eyes that does not constrict the viewing field. Most video Frenzel goggles have patients view through a hole, and turn a full field into a restricted field. This is certainly a problem, that to the authors knowledge, has simply been ignored. In theory, one might build a hybrid device with a camera over one eye and a full field over the other (haven't seen one yet), or might use a remote eye movement monitoring system (these do exist).
ENG vendors often sell a "light bar", consisting of an array of LED lights that can be controlled in such a way to produce moving bars of light. An array of LED's is illuminated in a way such that a bar of light moves to the left or right. While technically this stimulus does stimulate the peripheral visual field, it only stimulates a small strip of it. We suggest not using light-bars for optokinetic stimulation. This stimulus is clearly not a full field, contains easily tracked lights, and mainly is an alternative method of eliciting smooth pursuit. In the author's opinion, this type of OKN has no diagnostic value over simple pursuit.
Another commercial variant is to use a video-projector to display a moving field on a portion of a screen. This method is a little better, as an appropriate pattern may have no individually trackable portions, but still it does not produce a full field and because of this is vulnerable to persons fixating on the edges of the projected image. In the author's opinion, this method of eliciting OKN is of unproven value but worth considering.
Clinicians sometimes may try to elicit OKN using simpler devices such as the drum below. This device, not even pretending to be a full field, is even more of a "pursuit" stimulus than the larger drum methods shown above. Still, it can be useful in evoking convergence retraction nystagmus in persons with dorsal midbrain lesions, as well as asymmetrical tracking in persons with latent nystagmus.
"OKN" can also be elicited by strips that contain squares of alternating colors, tape measures, and even busy ties. The usefulness of these devices, in the author's opinion, is questionable.
Both types of OKN stimulators can be obtained from "good-lite"
Normal values for OKN gain are similar to those given for pursuit gain, or slightly greater, but OKN gain is less strongly reduced at high frequencies (27). While normal values are available for OKN phase, it is uncertain whether or not phase is affected by disease. Practically, OKN is best evaluated by comparing it to smooth pursuit, using the normal values developed for pursuit.
Things to investigate related to OKN:
(Normal shift of "beating field" when the drum changes direction. )
Normally, people undergoing OKN testing adopt an eye position that is opposite in direction to the velocity of the OKN drum -- this is called shift of the "beating field". This is presumably adaptive as it allows people to "ride" their gaze evoked nystagmus as well as potentially have a longer period to track. When there are diseases of the fast phases, often beginning with slow saccades, the eye doesn't shift into the "beating field", but rather gets "hung up" in the orbit. This is seen in disorders such as PSP and some of the brainstem related cerebellar degenerations.
Curiously, Seshagiri et al (2018) stated that OKN testing using a "tape" outperformed ataxia score (ICARS) in patients known to have SCA, in "quantifying oculomotor dysfunction in SCA". As OKN testing with a tape is intrinsically subjective, and this was an unblinded study, the utility of this finding needs confirmation.
References: Seshagiri et al, Optokinetic nystagmus in patients with SCA. A bedside test for oculmotor dysfunction grading. Neurology 2018;91.
Pitfalls having to do with OKN:
There are several pitfalls unique to optokinetic testing. While less sensitive to attention than pursuit, because OKN is disturbing to some patients, there may be an active attempt made to suppress it by fixating upon an object in the room which is not moving. This pattern is easily recognized because these persons are generally otherwise normal individuals, and because their initial responses are robust. Also, many commercial "optokinetic simulators", are actually devices which elicit smooth pursuit. If one is using such a device, the diagnostic points listed below which depend on noticing differences between pursuit and optokinetic responses do not apply.
References related to normal values for OKN/OKAN.
- Peterka RJ, Black FO, Schoenhoff MB. (1990) Age-related changes in human vestibulo-ocular and optokinetic reflexes: pseudorandom rotation tests. J. Vestibular Res. 1, 61-71.
|Calibration error is easily seen in this OKN recording.See the pursuit recording above (same patient). Here the clue is that this elderly patients is apparently tracking faster than the target. This is impossible without a technical error.|
Optokinetic nystagmus (OKN), like pursuit, has only minor diagnostic utility. Although OKN is more specific than pursuit, as it is not as affected by inattention and medication as is pursuit, it is also less sensitive. Presumably the relative lack of sensitivity of OKN to ocular and central disorders is due to the fact that OKN is the sum of two tracking mechanisms, namely the smooth pursuit system, which uses foveal vision, and a separate tracking system, which uses both foveal and extrafoveal vision.
Causes of OKN disorders
Visual disorders (such as retinitis pigmentosa or just blindness)
Pursuit system disorder (i.e. floccular lesion, medication)
Fast phase disorder (i.e. PSP)
Congenital nystagmus : CN can cause gigantic asymmetries of OKN. The most common CN is latent nystagmus. Here there are typically huge asymmetries of OKN depending on the viewing eye.
|Very poor OKN in patient with retinitis pigmentosa. This man had "bony spicules" in his periphery, and very little peripheral vision. See other images of his pursuit and saccades.|
There are several causes of OKN abnormalities. There are three specific patterns of abnormal OKN, the first of which is symmetrically reduced OKN gain. Reduced OKN occurs in visual disorders, in pursuit system disorders, and in disorders of fast phases. While smooth pursuit is most affected by visual acuity, which represents foveal vision, OKN is produced both by foveal and extrafoveal vision, and thus may persist even when visual acuity is poor. In disorders which selectively affect foveal vision, a slow build-up of OKN may occur to a constant velocity stimulus (7). In disorders which spare foveal vision but abolish peripheral vision, such as extremely severe retinal pigmentary degenerations, no buildup of OKN is seen. Another context in which pursuit is normal but OKN is symmetrically reduced are patients with fast phase disorders. The most common clinical disorder of this type is a degenerative disorder of the brainstem called progressive supranuclear palsy (PSP), in which saccades are slowed and difficult to initiate. Accordingly, patients with PSP may have normal pursuit to a sinusoid or triangular wave target, but poor OKN to a drum moving at constant velocity because their OKN "hangs up" in the orbit. In other words, the eyes deviate out to the orbital edge and just stay there, instead of undergoing periodic resetting quick phases which bring the eye back to the center. These patients show a similar disorder of vestibular fast-phases, and get hung up when rotated at constant velocity. In the later stages of PSP, both pursuit and OKN are lost.
Zee DS, Yee RD, Robinson DA. (1976a) Optokinetic responses in labyrinthine-defective human beings. Brain Res; 113:423-28.
There are a small number of patients with unsteadiness that have difficulty making rapid eye movements. For the most part, these patients have "PSP", or progressive supranuclear palsy, but there are also a few that have other neurological degenerations. During OKN, these patients don't reset their eyes efficiently, and theire eyes get "pulled" towards the slow-phase direction. This is seen in the graphic above.
This does not apply to persons who have slowing of all eye movements (such as oculomotor palsy) -- this is a problem with brainstem circuitry.
"Hanging up" in the orbit, also occurs during rotatory chair testing, as long as the chair goes fast/far enough to drive the eye "into the orbit". Normal persons reset to the other side of midline, but these people's eyes "pin" to the edge. The plot above is from a patient with presumed PSP.
Nearly all of these people will also have slow saccadic eye movements, as shown on the right hand side above. This can sometimes be difficult to detect at bedside however.
These people will "pass with flying colors", vestibular tests that do not exceed the orbital boundary, such as the VHIT test, but they will have difficult to interpret rotatory chair tests, with greater problems (i.e. lower gain) at low frequencies. One might encounter someone with normal phase but very low gain due to this interaction with fast phases.
Asymmetrical OKN is not as helpful for diagnosis of central nervous system disorders as asymmetrical pursuit, mainly because it occurs so infrequently. Presumably asymmetrical OKN is uncommon because it requires lesions in two tracking systems -- foveal and extrafoveal. Only a minor asymmetry of OKN appears following complete unilateral peripheral vestibular lesions. Asymmetrical OKN is present in patients with maldeveloped foveas and also appears briefly following unilateral parieto-occipital lesions.
Asymmetrical OKN is very common in persons with a type of congenital nystagmus called "Latent Nystagmus". The asymmetry depends on the viewing eye. If the person is amblyopic, this may mean that they are always viewing out of one eye.
Baloh RW, Yee RD, Honrubia V. (1980) Optokinetic asymmetry in patients with maldeveloped foveas. Brain Res 186: 211-216.
Reversed or inverted OKN occurs in patients with congenital nystagmus, which is discussed under the heading of fixation. In these patients, the nystagmus beats in the direction of stripe movement. However, the slow-phase velocity of the nystagmus does not scale with the stimulus speed.
- Abadi RV, Dickinson CM. (1985) The influence of preexisting oscillations on the binocular optokinetic responses. Ann Neurol 17: 578-586
OKN is much less useful than is rotatory chair testing as it is rarely affected substantially by disease. This is because it has redundant drivers -- the pursuit system and a lower level optokinetic following system that goes through the vesibular nucleus.
OKN may be useful for detection of malingering or lack of cooperation because it takes quite a bit of effort to block OKN.
In VVI, a person is rotated with a visual surround or target also present.
The most useful variant of this is to have a person look at (fixate) a laser that is fixed to the rotatory chair. This type of VVI is a good index of ones CNS's ability to suppress nystagmus, and thus it is a measure of cerebellar and brainstem function. If patients is not told to keep their eye on the chair-fixed target, the test is meaningless. This is a very common error.
Another type of VVI is done by simply rotating the person in the light. Here the person has both OKN and vestibular input. For this this type of VVI gain should be close to 1.0. If it is more than this, you have a technical problem. If it is less than this, at low-frequency, you either have a calibration problem or an uncooperative patient. Sometimes patients are poorly instructed and don't know that they are supposed to keep their eyes open.
Practically, the VVI test using the micromedical technology rotatory chair should not be trusted, as it almost never produces abnormal results. We suspect that there is a software issue in the analysis routines.