Dynamic Visual Acuity test

Dynamic illegible 'e' test -- DIE test

Timothy C. Hain, MDMarch 7, 2021

The VOR (vestibulo-ocular reflex) gain is most helpful for diagnosing ototoxicity and other bilateral vestibulopathies. There are several methods of quantifying it at the bedside.

The Dynamic illegible 'E' test or DIE is a very helpful procedure (Longridge, 1987). One has a patient view a eye chart similar to that shown below with the head still. This measures visual acuity with the head still. Then one moves the patient's head horizontally at about 2 hz, +- 30 deg excursion and again obtains visual acuity. This measures visual acuity with the head in motion. Eye charts can be downloaded from the web here: https://www.i-see.org/eyecharts.html#download".

Persons with substantial function of their vestibular system have no loss in visual acuity with this procedure. Even persons with unilateral vestibular loss -- i.e. having lost 50% of their vestibular system -- will generally score normally. Rigorous psychophysical techniques can detect and lateralize unilateral loss (Tian et al, 2001), but practically there are easier ways to detect unilateral loss (such as caloric testing).

Patients with bilateral vestibular loss, especially acutely, often lose 6-8 lines of visual acuity. Vertical movement is not as sensitive or specific as the horizontal test (Schubert et al, 2002).

As patients recover from bilateral loss, they perform better on this test. This is probably related to a combination of adaptive changes in the vestibuloocular reflex as well as predictive pre-programming of eye movements (Herdman et al, 2001).

Figure 1: Eye chart made for use at 10 feet, suitable for use in the Dynamic Visual Acuity test in the office. Eye charts similar to this can be downloaded from the web. We also have designed a similar eye chart that we think is superior that can be downloaded (see below).

LogMar DIE testing

The Snellen charts, which are designed to measure visual acuity, are not all that well suited to measuring changes in visual acuity, especially when quantified as "lines" of visual acuity. Suppose someone is using a chart where a few lines are left out ? This would provide a bit less resolution for testing visual acuity, which might not matter, but would destroy the comparability of the DIE test done with a different chart.

The Snellen chart is also designed (roughly) so that 3 lines is equivalent to a factor of 2 loss in visual acuity (i.e. 0.3 LogMar, or equivalently 6 dB). However, this doesn't always work, probably because the numbers of the Snellen chart generally are multiples of 10 feet -- this simplicity would be lost if the numbers scaled logarithmically. The graph below of LogMar vs. Snellen lines shows that the plot is not a straight line, thus the slope is variable.

A more rational approach to DIE testing than using changes in Snellen lines is to use "LogMar" method. LogMar is an acronym for the base-10 logarithm of the minimum angle of resolution. It is simply the log-10 of the inverse visual acuity (e.g. 20/20). Because 20/20=1, normal is a LogMar of 0. 20/200 corresponds to a Logmar of 1 (log of 10), so most of the eye chart can be expressed by LogMar's between from 0-1.

The equivalence table between Snellen lines and LogMar (calculated using a simple Excel spreadsheet) is given below:




Acuity LogMar
8 20/20 0
7 20/25 0.097
6 20/30 0.176
5 20/40 0.301
4 20/50 0.398
3 20/70 0.544
2 20/100 0.699
1 20/200 1

We have designed an eye chart more suitable for DIE testing, shown below. This chart uses Snellen Optotypes, it is sized for 10 ft viewing, and has LogMar values (times 10) shown on the right. This chart is used by simply recording the 10*LogMar value that a person can see with the head still, the value that a person can see while an examiner is oscillating their head. The change in visual acuity can be computed by subtracting the two 10*LogMar values.

The advantage of this chart over a conventional Snellen type chart is that the visual acuity differences are not affected by viewing distance or other conditions that affect visual acuity. Changing viewing distance will of course make the visual acuity measurements wrong, but because it simply adds an offset to the Logmar value, it will not (theoretically anyway), affect the visual acuity difference. This characteristic arises intentionally because the Logmar values are now accurate. We have used 10*LogMar, to make the numbers roughly comparable to those on the conventional Snellen Chart. It should be noted, of course, that the numbers go in the opposite direction -- 1.0 on a conventional Snellen Chart is 20/200 vision, while on this chart, 10 is 10/100 (or 20/200).

For this chart, if the viewing distance is different than it was designed for (lets say, 20 feet instead of 10), the visual acuities will actually be 20/100 rather than 10/100, and simply a factor of 2 different. This means that the LogMar values on this chart should be adjusted by subtracting 10 x log of 2 =(10 * 0.301), or simply 3. In general, the adjustment is:

10*Logmar = Measured 10*LogMar - 10*log10(NewDist/10).



Snellen type eye chart explicitly designed for DIE testing. You can download a higher resolution printable version of this chart here. Be sure not to print this chart in a different scale than it's native scale, as if you do this, the visual acuity values will not be correct.




  1. Herdman SJ and others. Role of central preprogramming in dynamic visual acuity with vestibular loss. Arch Oto HNS 2001:127:1205-1210
  2. Longridge NS, Mallinson AI. The dynamic illegible E (DIE) test: a simple technique for assessing the ability of the vestibulo-ocular reflex to overcome vestibular pathology. J Otolaryngol 1987;16:97-103, Acta Otol (Stockh) 103: 273-279, 1987; Otolarygol HNS 1984:92:671-7
  3. Schubert MC, Herdman SJ, Tusa RJ. Vertical dynamic visual acuity in normal subjects and patients with vestibular hypofunction. Otol and Neurotol 23:373-377, 2002
  4. Tian JR, Shubayev I, Demer JL. Dynamic visual acuity during yaw rotation in normal and unilaterally vestibulopathic humans. Annals NYAS Vol 942, 501-504, 2001
  5. Thomson D. VA testing in optometric practice. Part 2: Newer chart designs. Optometry Today: 22-24, 06/05/05