CLINICAL TESTS FOR BINOCULAR VISION

JOHN LEE and ANN McINTYRE

London

As is well known, Claud Worth originally suggested be a face turn to the left or right, head tilt to left or
that there are three grades of binocular function. His right, and chin elevation or depression. All three
most basic grade was simultaneous macular percep­ components may be present in certain cases,
tion, this being followed by fusion; finally, the highest particularly when the deviation is due to a cyclo­
grade of binocular function was claimed to be that of vertical muscle paresis, and the child should be
stereopsis. This 'wedding cake' model of binocular assessed with and without head position to discover
function is probably somewhat misleading and what contribution, if any, this makes to binocular
difficult to relate to the information gained from function. Other, rarer ocular causes of abnormal
the tests of binocular function that are normally used head posture include ptosis, both unilateral and
in orthoptic examination and in screening. The bilateral, congenital motor nystagmus with a null
intention of this paper is to discuss practical clinical zone, and homonymous hemianopia. Non-ocular
points with regard to the detection and assessment of causes include unilateral deafness, and spinal mus­
binocular function, in particular in younger and pre­ cular problems such as kyphoscoliosis and neck
verbal patients. It represents the practical experience dystonia.
and views of the authors and does not attempt to be a The cover uncover test should be done in all
comprehensive review of all methods that have ever patients to a suitable accommodative target; the
been described. demonstration of a manifest ocular deviation, while
not excluding binocularity, as this may well exist in
THE NON·VERBAL YOUNG CHILD
an anomalous form, will certainly prove that normal
bifoveal fixation is not present. In the absence of a
Much may be gained from simple external examina­ manifest deviation, the cover test and the associated
tion of young children. They are usually quite happy alternate cover test will give clear information as to
to look at an interesting fixation light, or indeed at
the presence of binocular function because as the
their mother's face while a fixation light is held at an
eyes are observed following the removal of the cover
appropriate distance from their eyes, and it is usually
it can be seen whether a recovering fusional move­
possible to state whether the corneal reflexes appear
ment is made. If such a movement is made, the
to be symmetrical. The majority of cases of strabis­
alacrity and efficiency with which it is made are
mus in small children tend to have fairly large angles,
useful and direct clues as to the quality of the
so the demonstration of symmetrical corneal reflexes
underlying binocular function and motor fusion.
is already a substantial step towards the assumption
By the same argument the assessment of binocular
of probable binocular function.
convergence to an interesting toy moving in towards
An abnormal head posture is in general a sign that
the face will show whether or not the two eyes
binocularity is being maintained in the presence of an
converge symmetrically to maintain bifoveal fixation
incomitant deviation. It is typical that the head is
on the object of regard. It is possible to have reflex
placed in the position where the eyes are least able to
be used together (e.g. a child with a left sixth nerve convergence of both eyes without binocular function,
palsy will turn the head to the left to avoid left gaze, but usually it will be clear that the patient is following
so as to enjoy good binocular single vision in the the target with only one eye and the other is not
primary position and right gaze). Abnormal head being used. If there is any doubt a cover test during
postures may consist of three components: there may the convergence movement will make matters
clearer.
Finally, in the small child the demonstration that
Correspondence to: Mr J. P. Lee, FRCOphth, Strabismus and
Paediatric Service, Moorfields Eye Hospital, City Road, London
the child overcomes a base-out prism (conventionally
ECIV 2PD, UK. 15-20 prism dioptres) and realigns the visual axes is

Eye (1996) 10, 282-285 . © 1996 Royal College of Ophthalmologists

CLINICAL TESTS FOR BINOCULAR VISION 283

strong corroborative evidence for the presence of DEMONSTRATION OF STEREOPSIS

binocular function. As the prism is removed the eyes Stereopsis is produced by the stimulation of horizon­
are seen to diverge in order to regain binocular tally disparate retinal areas which fall within Panum's
fixation. fusional areas. Should this sensation of horizontal
disparity occur the image will be perceived as one
percept with the illusion of depth. The wider the
inter-pupillary distance, the greater (in theory) the
DEMONSTRATION OF SENSORY AND
discrimination achievable.
MOTOR FUSION
Stereoacuity is the measure of the minimum
In the older child one can also assess the main­ perceivable horizontal disparity expressed in seconds
tenance of single vision on binocl/lar convergence of arc. The lower limit of this in adults is around 40
and in addition the synoptophore may be used to seconds of arc, but it is probable that many normal
assess the presence of fusional ability and is of adult subjects have better stereoacuity than this
particular value where a manifest deviation is when tested with appropriate specialised apparatus.
present. Appropriate targets with details may be Stereopsis can be demonstrated from a few months
used to exclude suppression and the patient may then of age but may continue to develop and increase in
have the full range of binocular fusional vergence its accuracy until about 8 or 9 years old. Tests may be
fully assessed. Children usually enjoy synoptophore qualitative or quantitative and are based on a variety
testing and once they are capable of giving adequate of optical and other principles.
responses they are usually accurate in their
observations. Qualitative Tests
Fusion range can be tested in free space by
observing motor fusion to base-in and base-out In the Lang Two Pencil Test the child is asked to
variable prisms by using a prism bar or a rotary place a pencil on top of one being held vertically by
prism of the Risley variety. Again the demonstration the examiner. It is conventional to do the test with
of smooth convergence or divergence without either eye in turn and then with both eyes together,
awareness of diplopia and without the development the aim being to assess the accuracy or otherwise of
of a manifest deviation is the key observation. Most the positioning shown by the subject. Although the
healthy children will show a convergence range of test is easy to do, and is certainly a game enjoyed by
around 40-50 prism dioptres and a range of children, it will detect only very gross deficiencies of
divergence of between 10 and 15 prism dioptres. stereoacuity and is in no sense an accurate screening
Sensory tests for fusion include Bagolini lenses in
test.
The Synoptophore can again be useful in patients
which the patient views a fixation light through lenses
with striations at 45 and 135 degrees respectively. who have a manifest deviation. It can be used to
The patient fixates a light and gives an appropriate assess the potential for binocular single vision or the
response. If the eyes are straight and there is normal strength of anomalous retinal correspondence. Slides
retinal correspondence the patient should see one are usually used which are labelled as being gross or
light and symmetrical streaks of light forming a cross. detailed, but there are other slides (Braddick) which
If there is a manifest deviation but still anomalous
are capable of quantifying stereopsis between 90 and
720 seconds of arc. These are based on the random
correspondence with binocular function the patient
will still see a cross. In the presence of diplopia and dot principle and can be used in older patients. The
normal correspondence two lights and two lines are synoptophore is enjoyable for children but may give
seen separated by an amount appropriate to the size artefactual information of binocular function demon­
of the manifest deviation. strable with the synoptophore which is not in practice
The Worth Four Dot Test involves viewing four achievable.
lights either at distance or near, of which one is red,
two are green and one is white. If normal fusion is Quantitative Tests
present the patient when viewing the picture through It is our strong clinical impression that the TNO
red and green glasses will report the presence of one Stereo Test is a good test for confirming high-grade
red light, two green lights and one light which can binocular single vision and stereopsis. It consists of
vary between red and green depending on ocular computer-generated red/green random dots viewed
dominance and retinal rivalry. In the event of the at 40 cm using appropriate complementary red and
patient suppressing one or other eye they will see green glasses. It includes non-quantitative screening
either two red lights or three green lights. Diplopic plates consisting of pictures of a butterfly, boxes and
patients will report three green and two red lights, symbols, a suppression test and a series of quantita­
the separation being determined by the angle and tive stereoacuity plates measuring between 480 and
direction of squint. The test is easily done and a 15 seconds of arc. The advantage is there are no
simple response can be obtained from most children. monocular clues. The disadvantages are that it
284 J. LEE AND A. McINTYRE

requires red/green glasses to be worn, which some than the Titmus test in terms of giving fewer
children do not like, and subjects with significant monocular clues, but does not appear to be as good
latent deviations or intermittent tropias may perform as the TNO test, except in so far as the red/green
poorly due to dissociation in the absence of contour glasses of the TNO test tend to produce more
clues and have their stereoacuity thereby under­ disassociation in patients with a poorly controlled
estimated. The test seems robust and stands up well phoria and therefore lead to poorer results. Another
to the wear and tear of a busy orthoptic clinic. It is version of the test consists of a random-dot letter 'E'.
widely regarded as being one of the better clinical It is performed at 50 cm and the child is required to
practical tests for the demonstration of bifoveal identify which of the two plates contains the E. This
binocular vision. test has been used for screening large rural popula­
The Titmus Stereo Test (also known as the Wirt tions for the absence of stereopsis, allowing those
Stereo Test) is a vectograph test in which two targets who fail to be examined in more detail for strabismus
are polarised at 90 degrees relative to one another and amblyopia.
and viewed through polaroid filters which give a The Frisby Stereo Test is a stereo test based on
depth perception illusion. The largest disparity (3000 actual depth of target. It consists of three Perspex
seconds) is given by the picture of a fly, which takes sheets of different thicknesses and the patient is
up half the test and when seen through polaroid asked to identify which figure, of four on each plate,
spectacles can be so dramatic as to cause small and is the one that is either closer or further away, the
susceptible children to burst into tears. This positive fixation target having been printed on one or the
response will confirm the presence of stereopsis but other side of the Perspex sheet. The disparities are
means it is difficult to continue with the test. There between 600 and 15 seconds of arc and it is easy to
are then three pictures of animals which contain use in relatively small children who will either look
many contour clues and which range in stereo towards the disparate target or may well reach out to
disparity from 400 to 100 seconds, and a series of grasp it. As glasses are not required it is an easy test
circles which range in stereo disparity from 800 to 40 to do in a relatively small child. Its disadvantages are
seconds of arc. The test is fairly robust but polaroid that if there is any movement of the plates, or indeed
spectacles have to be worn. These are somewhat of the patient's head, then it is quite easy to pick out
flimsy and are easily broken by the more active child. the disparity even monocularly. Nevertheless, it is a
The advantage of the test is ease of response from relatively easy test to administer.
relatively young children; the disadvantages are that There are now two different Lang Stereo Tests,
it requires spectacles and that it is not difficult to pick both of which are based on the principle of the
out uniocular clues, thereby leading to inappropriate 'three-dimensional postcard'. These are stout plastic
overestimation of stereoacuity. cards with a ribbed surface, one set of images being
The vectograph principle has also been utilised in printed on one slope and another on the opposite
the testing of distance stereopsis using instruments slope. If they are then viewed at an appropriate
such as the American optical projection chart. We working distance one eye will tend to see the left­
have limited experience of these devices and they hand side of the picture the other eye the right. If a
certainly seem to offer some help in the estimation of random-dot image is printed onto the postcard then
stereoacuity at distance. The machine that we have the patient will see something standing out from the
used had two stereoacuity slides: one with different background. On card 1 all the targets have stereo
geometrical shapes and one with circles polarised in disparity and consist of a cat, a car and a star. On the
order to stand out from the background. Little has second test there is a crescent moon, a car and an
been written on the relationship of stereoacuity to elephant, but in addition there is one target which
the management of strabismus, in particular that of does not have stereo disparity which the child can
intermittent exotropia, but systematic studies will identify. This is in order to encourage the child, who
doubtless produce more information on this matter. will pick at least one of the four targets and thereby
Another considerable value of a distance vecto­ does not feel frustrated or disappointed by his or her
graphic test is that it makes a very good test for failure. It is fairly gross test and detects stereo
simulated visual loss, as patients do not realise that if disparities between 1200 and 200 seconds of arc. The
they read along a whole line of letters some of which test is easy to do and does not require any glasses and
are polarised in different phase, they are in fact its inventor claims that small children will respond to
seeing them with the two eyes at the same time. it quite well. The disadvantages are that it is difficult
The Randot Stereo Test is a polaroid test similar to to be sure what the child is grasping or reaching for.
the Titmus Stereo Test but utilising a random-dot It is hard to hold the card parallel to the face without
type of target instead of contour targets. It is head movement, and unless a verbal response is
polarised in the same way and viewed with the obtained it is difficult to be sure whether anything
same kind of glasses. This test appears to be better has been seen.
CLINICAL TESTS FOR BINOCULAR VISION 285

The Awaya Stereo Test appears to be no longer presence of binocularity in subjects whose eyes
widely available and yet is valuable. It is based on appear straight on cover testing.
red/green dissociation, comes with its own set of In the presence of a manifest deviation, assessment
glasses and measures stereoacuities from 4 120 down can establish the state of retinal correspondence, the
to 40 seconds of arc. It appears to lack monocular presence of suppression, or the potential for binocu­
clues and perhaps should be more widely known and lar single vision, and is an essential prerequisite to
distributed. determining the management plan.
Any well-equipped paediatric ophthalmology or
orthoptic department should possess a number of
CONCLUSION different stereo tests, as the perfect stereo test has yet
In the first instance, a carefully taken history will give to be invented. Doubtless, in the future, develop­
many clues to the presence or absence of binocular ments in computer-generated stereograms on VDU
vision. By careful clinical examination using appro­ monitors will lead to more efficient methods of
priate tests, it is relatively easy to confirm the testing for this important visual function.