(Investigative Ophthalmology and Visual Science. 1999;40:3031-3036.)
© 1999
by The Association for Research in Vision and Ophthalmology, Inc.
Binocular Depth-from-Motion in Infantile and Late-Onset Esotropia Patients with Poor Stereopsis
Manami Maeda1,
Miho Sato1,
Tomohisa Ohmura1,
Yoji Miyazaki1,
Ai–Hou Wang2 and
Shinobu Awaya1
1 From the Department of Ophthalmology, Nagoya University School of Medicine, Nagoya, Japan; and the
2 Department of Ophthalmology, College of Medicine, National Taiwan University, Taipei, Taiwan.
 |
Abstract
|
|---|
PURPOSE. There are at least two possible ways to detect motion-in-depth binocular
without monocular cues: the binocular disparities at different times
and a mechanism that detects interocular velocity differences. The
perception of interocular velocity differences (Binocular
depth-from-motion [BDFM]) depends on the relative velocity of the
images on the retina of the left and right eyes, and this information
can be experienced by normal and some strabismic patients. The purpose
of this study was to determine the characteristics of esotropic
patients who have BDFM but have poor stereopsis.
METHODS. Forty-one infantile and 28 late-onset esotropia patients with poor
stereopsis were studied. Dynamic stereopsis and BDFM were tested with
computer-generated random dot stereograms and kinematograms. The
correlations between BDFM and other binocular functional tests were
determined.
RESULTS. A total of 31 (44.9%) patients, 15 (36.5%) of the infantile and 16
(57.1%) of the late-onset esotropia group, passed the BDFM test. None
of these patients passed the random dot stereo test under static or
dynamic conditions. Fusion of the Worth four dot test at near
0.3 m was correlated with the presence of BDFM. Three of the 15
infantile and 10 of the 16 late-onset esotropic patients with positive
BDFM showed gross stereopsis as measured by the Titmus Fly. The angle
of strabismus was significantly smaller in the patients with positive
BDFM for the infantile and the late-onset esotropia groups.
CONCLUSIONS. BDFM was present in about half of the esotropic patients who do not
have fine stereopsis. Ocular alignment within 10 to 15 prism diopters
is an important factor in obtaining BDFM. Strabismus surgery still
provides some binocular benefit for infantile esotropia patients who
were bypassed for early surgery. Separate mechanisms may underlie
static stereopsis and BDFM.
|
Introduction
|
|---|
Static stereopsis, which arises from the disparity of the
retinal images on the two retinas, gives rise to the finest depth
perception. To obtain a high level of stereopsis, good ocular alignment
is essential,; therefore, obtaining or regaining good ocular alignment
is one of the important goals for strabismus treatment.
In strabismus clinics, conventional static stereopsis is routinely
examined, but dynamic stereopsis, which requires expensive and
space-occupying equipment, has not been routinely used. Fusion and
stereopsis are considered to be static depth perception processes,
which comes from the retinal disparity of the binocular images.
However, in a dynamic world, the retinal images are always in motion,
and there are both monocular and binocular depth cues from the motion.
Monocularly, objects moving in depth result in changing retinal image
size and motion parallax. There are at least two possible ways to
detect motion-in-depth binoculary without monocular cues: the binocular
disparities at different times and a mechanism that detects
inter-ocular velocity differences. The perception of interocular
velocity differences (binocular depth-from-motion [BDFM]) depends on
the relative velocity of the images on the retina of the left and right
eyes, and this information can be experienced by normal and some
strabismic patients1
(Fig. 1)
. For objects moving in the real world, interocular velocity
differences occur at the same time as changes in binocular disparity.
Kitaoji and Toyama2
reported that motion stereopsis can be
preserved in patients with small angle strabismus who do not have
static stereopsis. Thus, testing for the presence of binocular
motion-in-depth in a clinical setting and learning the prerequisites
for BDFM perception can provide important information to assess the
visual capabilities of strabismic patients.
View larger version (11K):
[in this window]
[in a new window]
|
Figure 1. Diagram of the movement of retinal images with the movement of a
target. (A) The retinal images move in opposite directions
to give a depth sensation. (B) The retinal images move in
the same direction and give no cue for depth sensation.
|
|
One of the authors (AHW) has developed a computer program3
that generates dynamic random-dot stereograms. In this program,
disparity and motion cues can be included or omitted independently. We
have tested patients with either early- or late-onset esotropia with
this program to see whether BDFM was present despite the absence of
static stereopsis.
|
Methods
|
|---|
This study was performed in the Department of Ophthalmology of
Nagoya University between December 1996 and December 1997.
Controls
We examined 12 normal subjects (age, 26–30 years, 6 men and 6
women) without any ocular abnormality other than refractive errors
(visual acuity; better than 20/20) to determine the normal responses
and to select the optimal testing conditions.
Esotropia Patients
Informed consent was obtained from the subjects or their
guardians, and all procedures were conducted in accordance with the
principles embodied in the Declaration of Helsinki.
Forty-one patients with infantile esotropia and 28 with
late-onset esotropia were examined. The patients in both groups showed
worse than 3000 seconds arc. stereopsis as determined by the large Fly
(+) in the Titmus Stereo Test. Patients with infantile esotropia were
defined as those: whose esotropia was diagnosed before 6 months of age
by an ophthalmologist or was confirmed by photographs; whose deviation
was not abolished by a hyperopic correction; who had no central nervous
system disorders or developmental delay; and who were born after 37
weeks of gestation. Late-onset esotropia was defined as an
accommodative or partially accommodative esotropia with hyperopia,
which became manifest after 18 months of age. Patients with paralytic
esotropia, esotropia from organic disorders, premature birth (gestation
before 37 weeks), and developmentally delayed children were excluded.
The mean age at the time of examination was 8.8 years (age range, 4–19
years) for the infantile esotropic group and 13.1 years (age range,
5–31 years) for the late-onset esotropic group.
Visual Stimulator
The computer program was run on a FMV-5100D4 computer (FUJITSU,
Tokyo, Japan) and displayed on a 17-inch CRT in a dimly lit room.
Anaglyphic random-dot stereograms and kinematograms were made up of red
and green random-dots. The size of the screen was 31.0 x 22.0 cm
with 320 x 200 pixels. When viewed at a 55.5-cm distance, each
pixel subtended 360 second-arc. Four rectangles of 60 x 50 pixels
(6° x 5° arc) were displayed on the monitor, and one was
programmed to provide a depth cue. Throughout the testing session, the
background was made up of a stable random-dot pattern with the same
density as the 4 rectangles.
The pair of rectangles seen by right and left eyes, t1R and t1L, was
called a stereogram pair if they had the same random-dot pattern and
included a disparity cue. If the rectangles were different and
therefore had no motion cue, they were called a temporal correlogram.
The rectangles seen by the same eye, t1R and t2R, were called
kinematograms if they had the same random-dot texture and had cues for
apparent motion. If the rectangles were different and therefore had no
motion cues, they were called a temporal correlogram. In the
kinematogram pair, if the rectangles moved in opposite directions, the
fused target invoked a depth sensation (i.e., a movement in depth) in
normal subjects.
Test 1 was designed to test binocular motion-in-depth elicited by
interocular disparity cues and/or BDFM elicited by movement cues. For
this, t1L, t1R, t2L, and t2R were made up of the same random-dot
pattern. Test 2 was designed to test only stereopsis, and t1R =
t1L, but t1R and t1L were different from t2R and t2L. When the program
is stopped, only one rectangle with a disparity of 360 second–arc is
seen in depth by normal subjects, but no depth is seen by stereo blind
subjects. Test 3 tested only BDFM without disparity cues, and t1R was
different from t1L; however, t1R and t1L had the same pattern as t2R
and t2L, respectively. The other three rectangles were seen by both
eyes and were designed to move in the same directions. When the program
is paused, four indistinguishable rectangles are seen by normal
subjects, and no rectangles can be seen in depth by stereo blind
subjects (Fig. 2)
.
View larger version (30K):
[in this window]
[in a new window]
|
Figure 2. (A) Test 1 dynamic stereopsis and BDFM. In one
rectangle, the same random dots include disparity cues moving in
opposite directions for the two eyes. (B) Test 2 dynamic
stereopsis. In one rectangle, the same random dots include disparity
cues moving in the same direction and no motion cue is included.
(C) Test 3 BDFM. In one rectangle, the random dots move in
opposite directions for the two eyes.
|
|
Each patient sat facing the screen at 55.5-cm distance with a green
filter on the right eye and red filter in front of the left eye. The
computer randomly determined which one of the four rectangles would
provide a depth cue. The subjects were asked to select the figure that
appeared to move back and forth in the "Z" direction. For control,
we occluded one eye or paused the program (Test 3) during the course of
the tests to be certain that answers were based on disparity or motion
cues. The correct answer was given to the patients immediately after
each test. The test was repeated 10 times in a forced-choice manner,
and a passing score was set at eight correct answers. After the subject
passed Test 1, Tests 2 and 3 were conducted. If a subject passed both
Tests 1 and 2, he/she was designated as having stereopsis from
disparity. If the subject passed both Tests 1 and 3, but not Test 2, he
or she was defined as having BDFM-positive without stereopsis.
Testing Methods
Controls.
All three tests were first performed on the normal subjects. Test
1 and Test 2 were performed on control subjects. Test 3 was performed
with various range of movement and with different velocities of the
random-dot pattern to determine the optimal stimulus conditions. The
range and speed that allowed the control subjects to detect the depth
most easily were selected as the testing condition for the esotropic
patients. To determine the minimum visual acuity necessary to pass the
BDFM test, we blurred the vision in one or both eyes with Einschleich
occlusion partielle filters (Ryser Optik, Basel, Germany) and
conducted Tests 1, 2, and 3 on 5 normal patients. In addition, to
verify that this test can be passed by horizontal disparity but not by
flickering or odd sensation, we tested with 3 normal subjects by
rotating the monitor 90°.
Esotropic Subjects.
We performed complete ophthalmic examinations including visual acuity,
Titmus Stereo tests, TNO stereo test, Bagolini striated lenses
test, and Worth four-dot test at near (0.3 m) and distance (5 m) on all
the subjects. The angle of strabismus was measured by simultaneous
prism cover test at near and far.
Data Analysis
The differences between the patients having BDFM (BDFM+) and those
lacking BDFM (BDFM-) were analyzed statistically with either the
chi-square test or Mann–Whitney test, and P < 0.05 was
accepted as statistically significant.
|
Results
|
|---|
Normal Subjects
The results obtained from the normal subjects showed that the
optimal repetition rate and range of movement of the two targets was
3.5 Hz and 1 pixel, respectively. All normal subjects perceived a depth
sensation on all tests. When the visual acuity was artificially reduced
to 16/20 or better, no control subjects failed Test 3. However, they
also reported that they felt depth sensation more strongly on Tests 1
and 2 than Test 3. No one passed any tests when the monitor was rotated
90°.
Patients with Esotropia
The relationship between BDFM and the sensory tests is shown in
Table 1
. None of the patients passed Test 2 regardless of the time of onset of
the ocular deviation (i.e., none had stereopsis). Fifteen (36.5%)
patients in the infantile group and 16 (57.1%) in the late-onset group
(total = 31, 44.9%) passed Tests 1 and 3 but not 2 and were
classified as having positive BDFM without stereopsis. Of the other 38
patients, 26 in the infantile group and 12 in the late-onset group did
not pass any test. There was no statistical difference between
early-onset esotropia and late-onset esotropia in the incidence of
positive BDFM.
The distribution of the angle of deviation at near for the infantile
esotropia and the late-onset esotropia groups is shown in Figures 3
A and 3B, respectively. The patients with positive BDFM (black squares)
had significantly smaller angles of deviation (infantile onset, 4.33
SD, 5.68 prism diopters [pd]; late-onset, 8.00 SD, 6.65 pd) than the
patients lacking BDFM (infantile-onset, 12.2 SD, 2.46 pd; late-onset,
18.4 SD, 9.00 pd; P = 0.0016; Mann–Whitney test).
There was no statistical difference between the age at the time of
examination and the presence of BDFM in both groups (infantile-onset,
P = 0.205; late-onset, P = 0.0885
Mann–Whitney test). Of the 41 infantile esotropia patients, 36 had
undergone strabismus surgery at a mean age of 49.8 months with a
(range, 25–120 months). Of the 28 late-onset esotropia patients, 10
patients had undergone strabismus surgery at a mean age of 166 months
(range, 24–356 months). There was no statistical difference between
the presence of BDFM and whether the patient had undergone surgery. In
addition, there was no statistical difference between a positive BDFM
and the age at the time of surgery in both groups (infantile-onset,
P = 0.796; late-onset, P = 0.317
Mann–Whitney test).
View larger version (15K):
[in this window]
[in a new window]
|
Figure 3. (A) Present angle of strabismus and BDFM (infantile
esotropia). (B) Present angle of strabismus and BDFM
(late-onset esotropia).
|
|
There was a high correlation between a positive Titmus Fly test and a
positive BDFM (P = 0.0009, chi-square test). None of
the patients passed Plate I of the TNO stereo test. Fusion of the Worth
four-dot test at 0.3 m and positive BDFM were highly correlated
(P < 0.0001, chi-square test). There was a
significantly higher number of patients with BDFM who were able to fuse
the Worth four-dot test at 0.3 m. None of the patients fused the
Worth four-dot test at 5 meters. There was no correlation between
suppression under Bagolini striated lenses test and BDFM and also no
correlation between the visual acuity and the presence of BDFM.
|
Discussion
|
|---|
In this study, we tested whether binocular detection of
motion-in-depth, which is evoked by the simultaneous nasal and temporal
shift of the retinal images in the two eyes, was present in patients
with infantile and late-onset esotropia. We found that 18 of the 31
subjects who did not have stereopsis as determined by the Titmus fly
test, showed binocular detection of motion-in-depth sensation. Removing
the disparity cues from the visual stimulus did not interfere with
their depth perception but removing motion cues from the visual
stimulus completely blocked their depth judgment. This latter fact was
verified by the absence of disparity sensors as tested by static
stereopsis test. When one eye was occluded, none of the subjects could
pass the BDFM test. We also rotated the monitor screen 90° to give a
vertical directional movement of the targets and no depth sensation was
evoked. These findings indicate that the binocular depth perception
came solely from inter-ocular velocity differences in the horizontal
direction.
In contrast, control subjects answered that they felt stronger depth
sensation with Tests 1 and 2 than Test 3, which means they rely on
interocular disparity cue rather than interocular velocity cue. This
finding is supported by the study of Halpen,4
who examined
the quality of motion-in-depth from interocularly uncorrelated
motion-defined forms and concluded that the perceived magnitude of
depth is less than that seen with interocularly correlated targets.
Cumming and Parker5
reported that motion-in-depth is
primarily detected by means of temporal changes in binocular disparity
and that interocular velocity differences play a minor role, if any, in
normal subjects. Our finding suggests that the subjects had
grown up without the experience of interocular disparity perception,
interocular velocity differences could play some role for
motion-in-depth sensation.
The angle of the strabismus measured by simultaneous prism cover test,
fusion of the Worth four-dot test at near, and the existence of gross
stereopsis were factors that were correlated with the presence of
positive BDFM. The possible reason for the strong correlation between
the Worth four-dot test at near and BDFM is related to the size of the
stimulus and the technique used in the computer program for binocular
separation. The subtense of the Worth four-dot test at 5 m is
0.5° and that at 0.3 m is 6°, which is the same as the
horizontal size of each rectangle of the BDFM test. Therefore, patients
with peripheral fusion are probably good candidates for positive BDFM.
The Bagolini striated test, on the other hand, was not correlated with
BDFM. If the BDFM had been tested under different conditions of
binocular separation, the results might have been different.
Kitaoji and Toyama2
reported earlier that motion
stereopsis can be preserved in strabismic patients and that the
existing strabismus angle was an important factor related to motion
perception. In our computer-generated patterns, the movement of the
images was not as smooth as that shown by a galvanometer because of the
nature of computer graphic generation. When the image movement is too
large, or too slow, or too fast, the sensation of continuous movement
is not elicited. Because of the faster movement of the images, 3.5 Hz,
compared with the 1 Hz used by Kitaoji and Toyama,2
it was
not possible to verify whether the approaching and receding phases were
correctly identified. However, the subjects with BDFM clearly stated
that they perceived a back and forth movement. Some of the esotropic
patients pointed out that the range of movement varies on each
individual.
This study verified that the patients without good binocularity under
static conditions, such as infantile esotropia patients, can obtain or
regain a different kind of binocularity under dynamic conditions.
Interestingly, infantile-onset esotropia patients in this study were
all aligned after age two, and the age at surgery did not interfere
with the success rate of the BDFM test. This finding is supported by
the fact that the plasticity of the motion pathway remains
"soft-wired" longer than the critical period for fine stereopsis in
humans.6
It has been established that the peripheral retina is more sensitive to
motion and that motion is transmitted to the middle temporal area (MT),
where almost all neurons are directionally selective. Direction and
orientation selectivity of neurons in visual are in the MT of the
macaque.7
The detection and analysis of motion may also be
required in conjunction with the depth perception. Bradley et al.
reported the existence of an important link between disparity and
transparent motion detection in MT and suggested that binocular
disparity in MT may facilitate velocity processing.8
Recently, MT is reported to be important for the perception of
structure-from-motion.9
Thus, BDFM that comes from
binocular detection of velocity could be helpful in the perception of
motion-in-depth and structure-from-motion with strabismic patients who
lack disparity perception.
|
Conclusions
|
|---|
BDFM was present in more than half of the esotropic patients who
do not have fine stereopsis. Ocular alignment within 10 to 15 prism
diopters is an important factor in obtaining BDFM. Strabismus surgery
still provides some binocular benefit for infantile esotropia patients
who were bypassed for early surgery. Separate mechanisms may underlie
static stereopsis and BDFM.
|
Acknowledgements
|
|---|
The authors thank Professor Yozo Miyake for his continuous
encouragement during the course of this study.
|
Footnotes
|
|---|
Submitted for publication March 18, 1999; accepted April 29, 1999.
Commercial relationships policy: N.
Corresponding author: Miho Sato, Department of Ophthalmology, Nagoya University School of Medicine, 65 Tsuruma–Cho, Showa–Ku, Nagoya
466-0065, Japan. E-mail: misato{at}med.nagoya-u.ac.jp
|
References
|
|---|
-
Beverley, KI, Regan, D. (1973) Evidence for the existence of neural mechanisms selectively sensitive to the direction of movement in depth J Physiol 235,17-29[Abstract/Free Full Text]
-
Kitaoji, H, Toyama, K. (1987) Preservation of position and motion stereopsis in strabismus Invest Ophthalmol Vis Sci 28,1260-1267[Abstract/Free Full Text]
-
Wang, A–H. (1996) Binocular depth-from-motion versus depth from disparity of strabismic patients Invest Ophthalmol Vis Sci Suppl 37((3)),S1290Abstract nr B193
-
Halpen, DL (1991) Stereopsis from motion defined contours Vision Res 31,1611-1617[Medline][Order article via Infotrieve]
-
Cumming, BJ, Parker, AJ (1994) Binocular mechanisms for detecting motion-in-depth Vision Res 34,483-495[Medline][Order article via Infotrieve]
-
Norcia, AM, Hamer, RD, Jampolsky, A, Orel–Bixler, D. (1995) Plasticity of human motion processing mechanisms following surgery for infantile esotropia Vision Res 35,3279-3296[Medline][Order article via Infotrieve]
-
Albright, TD (1984) Direction and orientation selectivity of neurons in visual are MT of the macaque J Neurophysiol 52,1106-1130[Abstract/Free Full Text]
-
Bradley, DC, Quian, N, Anderson, AA (1995) Integration of motion and stereopsis in middle temporal cortical area of macaques Nature 373,609-611[Medline][Order article via Infotrieve]
-
Gregory, C, Cumming, BG, Newsome, WT (1998) Cortical area MT and the perception of stereoscopic depth Nature 394,677-680[Medline][Order article via Infotrieve]
Top
Abstract
Introduction
