Download Orientation discrimination in amblyopia.

Orientation Discrimination in Amblyopia
Dernf C. Skorrun, Arthur Bradley, and R. D. Freeman
Using extended sinusoidal gratings to avoid potential problems of eccentric fixation, the authors have
studied orientation discrimination in amblyopia. For all subjects, elevated orientation discrimination
thresholds at high spatial frequencies were found. However, raised thresholds decrease with decreasing
spatial frequency, and can be normal at low frequencies. Orientation discrimination thresholds for both
amblyopic and non-amblyopic eyes are independent of contrast over most of the visible range. Therefore,
amblyopic orientation discrimination thresholds cannot be mimicked in non-amblyopic eyes by reducing
contrast. Control experiments show that the orientation discrimination deficits are not restricted to
vertical stimuli and that they are not a result of exaggerated cyclotorsional eye movements. Invest
Ophthalmol Vis Sci 27:532-537, 1986
Amblyopia is defined as a condition in which visual
acuity is reduced without any apparent cause.1 Recent
studies have shown that amblyopia is associated with
a wide variety of visual anomalies. Thresholds for contrast detection,2 flicker detection,3 motion detection,4
spatial frequency discrimination,5 phase discrimination,6 and vernier acuity7 are all elevated in amblyopia.
Although a diverse array of anomalies exists, the above
studies all show abnormalities with a similar spatial
frequency dependence. Deficits are typically largest at
high frequencies, decrease as frequency decreases, and
are often absent at low spatial frequencies. Because of
this frequency dependence, tests of visual functions in
amblyopia performed with stimuli that contain multiple frequencies or broad-bands of frequencies are difficult to interpret. With such broad band-stimuli, a deficit restricted to a high-frequency range may be missed
if the observer can perform the task effectively with
low-frequency information.
Because of this problem, the recent results of Vogels
et al,8 showing almost normal orientation discrimination thresholds for bar stimuli, which contain a broad
range of spatial frequencies, are hard to interpret. Orientation discrimination thresholds of normal observers
are almost independent of spatial frequency.910 Therefore, the experiment of Vogels et al does not distinguish
between the possibilities of substantial spatial frequency
specific orientation discrimination deficits and rela-
tively normal orientation discrimination performance
at all spatial frequencies.
There are two additional problems associated with
the experimental design of Vogels et al8: (1) A single
bar may not be imaged foveally in eccentrically fixating amblyopes." Because orientation discrimination
thresholds in normal observers may be elevated in the
periphery, the small threshold elevations seen by Vogels
et al may reflect normally elevated thresholds in peripheral vision. (2) Detection of bar stimuli is impaired
in amblyopia.12 Elevated orientation discrimination
thresholds may therefore have resulted from problems
in detecting the stimulus.
We have re-examined the question of orientation
discrimination in amblyopia. Using sinusoidal gratings
as stimuli and by performing control experiments, we
have avoided the problems listed above. We find that
amblyopia is associated with substantial deficiencies in
orientation discrimination that are clearly spatial frequency dependent. The elevated orientation discrimination thresholds are present over a wide range of
stimulus contrasts and cannot easily be attributed to
eccentric fixation or elevated contrast thresholds.
Materials and Methods
Apparatus
A microprocessor-based stimulus generator13 was
used to create gratings with a sinusoidal luminance
profile on the face of a display CRT (Tektronix 606,
Tektronix; Beaverton, OR). Space average luminance
of the screen was 60 cd/m 2 . The screen was masked
down to a 4-degree diameter circle. Grating contrast,
spatial frequency and presentation timing were under
computer control. Grating orientation was changed by
adding to the X-axis half a cycle of a triangular waveform synchronized with the Y-axis sweep. Fine control
From the School of Optometry, University of California, Berkeley,
California.
This work was supported by the National Institutes of Health,
USA, Grant EY01175 to RDF. BCS received support from the Norwegian Research Council for Science and Humanities (NAVF) which
also paid the subjects.
Submitted for publication: April 29, 1985.
Reprint requests: Bernt C. Skottun, Department of Visual Sciences,
School of Optometry, Indiana University, Bloomington, IN 47405.
532
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933128/ on 05/06/2017
533
ORIENTATION DISCRIMINATION IN AMDLYOPIA / Skorrun er ol.
No. 4
Table 1. Clinical data for amblyopic subjects
Subject
Refractive error
(diopters)
Visual acuity
Strabismus
(prism diopters)
Fixation of amblyopic eye
(prism diopters)
2ESO
1 N unsteady
24EXO
3 ST unsteady
8EXO
1/2 T unsteady
RS
RE-8.50-1.00X5
LE-1.00
20/300
20/15
CD
RE +3.00
LE +6.75 -2.00 X 85
20/15
20/175
CW
RE +2.00
LE PLANO
20/40
20/15
EB
RE -5.00
LE+2.00-1.00X50
20/20
20/300
12ESO
2 N unsteady
DM
RE+1.75 -0.25 X 120
LE +2.50
20/15
20/60
10ESO
2 N unsteady
JM
RE+1.50-0.50X88
LE +0.75
20/35
20/15
CS
RE+1.75 -0.75 X89
LE+1.00-0.50X98
20/20+
20/30-
2 ESO alt
21 ESO
1/2 SN steady
na
Clinical examination data for 7 amblyopes.
Abbreviations: N = nasal, T = temporal, S = superior, na = data not available,
ESO = esotropia, EXO = exotropia, alt = alternating, LE = left eye, and RE
= right eye.
of orientation was achieved by attenuating the triangular wave in 0.1 log unit steps. Therefore, for a 1degree orientation difference, the step size (average of
upward and downward steps) was 0.23 degrees, for a
0.5 degree difference step size was 0.12 degrees, etc.
A separate stimulus was created by use of a full cycle
of the triangular wave. This target, a chevron (depicted
schematically in Fig. 6), was also used to measure orientation discrimination.
randomized, and the subject indicated which of the
two presentations contained the tilted grating.
Procedure
Monocular contrast sensitivity measurements were
made using a temporal two-alternative forced choice
(2AFC) staircase technique (3 correct responses—step
down, one incorrect response—step up; this staircase
asymptotes at the 79% correct level14). Subjects pressed
one of two buttons to identify which interval contained
the grating. Contrast was varied in 0.1 log unit steps
and means of the last 5 of 7 staircase reversals were
used as threshold estimates.
Orientation discrimination thresholds for gratings
were determined monocularly using the same staircase
procedure. Two gratings were presented sequentially
for 500 msec. The presentations were separated by a
500-msec period during which the screen was blank
and had the same space average luminance as the gratings. Gratings were matched for contrast and spatial
frequency. One stimulus of each pair was always oriented at a given reference (eg, vertical). The second
stimulus was either (1) a grating tilted clockwise with
respect to the reference, or (2) a chevron with gratings
in the top and bottom halves tilted symmetrically about
that of the reference. The order of presentation was
Subjects
Seven subjects were used who had varying degrees
of amblyopia associated with anisometropia, strabismus, or both. Complete refractive data, determined
from clinical examinations, are shown in Table 1. All
subjects were tested while wearing their appropriate
corrective lenses. Complete sets of data could not be
obtained for all observers: Contrast sensitivity data were
not obtained for JM, and orientation discrimination
versus contrast functions were not obtained for CD.
Control experiments with oblique and chevron stimuli
were performed with only 2 and 3 observers, respectively. Informed consent was obtained from each observer prior to his (or her) participation.
Results
We first measured contrast sensitivity functions of
6 amblyopes (EB, CD, CS, DM, CW, RS). These data
are shown in rows 1 and 3 of Figure 1. A comparison
of data from different observers reveals a common pattern. At high spatial frequencies contrast sensitivity is
lower for amblyopic (open symbols) than for non-amblyopic (filled symbols) eyes. However, as spatial frequency is reduced, the difference between the two eyes
decreases until, at the lowest frequencies (eg, .5-1.0
c/deg), both eyes exhibit similar sensitivities. All six
observers show this type of contrast sensitivity deficit,
which is typical of amblyopes with anisometropia15 and
strabismus.2
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933128/ on 05/06/2017
534
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / April 1986
I000
100
<>
II
0
8*
.0
EB
CS
0L
.5 1 2
4
8 16 .5 I
2
4
8 16 .5 I
2
4
8 16
2
4
8 16
1000
?
100
z<2
85
io
DM
/
CW
RS
cc<75
<E
o
5
1 2
4
8
16 5 I
2
4
8
16 .5 I
Vol. 27
Data for the amblyopic eyes (Fig. 1, open circles)
differ from those for the non-amblyopic eyes. For each
observer, thresholds increase with spatial frequency.
Four subjects (CS, DM, CW, and RS) show a similar
pattern. Orientation discrimination thresholds are
similar for both eyes at the lowest spatial frequencies,
but at higher frequencies, thresholds for the amblyopic
eyes increase up to 3-5 degrees. Two subjects (EB, CD)
show a somewhat different pattern. In both cases orientation discrimination thresholds are elevated at,the
lowest spatial frequency (0.5 c/deg), but these amblyopes also show additional threshold increases at
higher frequencies. The results demonstrate that impairment of orientation discrimination in amblyopia
shows a spatial frequency dependency similar to other
visual deficits. An extreme case of a spatial frequency
dependent orientation discrimination deficit was demonstrated by an additional observer (JM), for whom
contrast sensitivity data were not obtained. This subject
showed equal thresholds of approximately 1 degree for
both eyes between 1 and 10 c/deg, but thresholds for
the amblyopic eye increased abruptly near 14 c/deg
(Fig. 2). Closer examination of the spatial frequency
range from 10 to 17 c/deg (inset) shows that there was
a continuous change in threshold from 1 to 5.5 degrees
over this range of spatial frequencies.
SPATIAL FREQUENCY (c/deg)
Fig. 1. Contrast sensitivity (1/contrast threshold) and orientation
discrimination thresholds (degrees) are plotted as functions of grating
spatial frequency (cycles/degree) for 6 observers (EB, CD, CS, DM,
CW, RS). The contrast sensitivity data (rows 1 and 3) are plotted in
the panel directly above each subject's orientation discrimination
data (rows 2 and 4). Open and filled symbols represent data for amblyopic and non-amblyopic eyes, respectively. Arrows indicate spatial
frequencies used in subsequent experiments (see below).
We determined orientation discrimination thresholds for the same six observers over the same range of
spatial frequencies. These results, obtained with vertical
gratings of 50% contrast, are shown in rows 2 and 4 of
Figure 1. Thresholds for the non-amblyopic eyes (filled
symbols) are similar to those reported for normal observers.10 There are no clear changes in discrimination
threshold with spatial frequency for the non-amblyopic
eyes although thresholds for higher frequencies tend to
be a little lower. Lowest thresholds are generally between 0.5 and 1 degree, which are slightly higher than
thresholds for experienced observers.1617 However, this
may be attributed to using unpracticed observers and
monocular testing; binocular viewing appears to produce lower discrimination thresholds. We conducted
cursory tests using normal subjects and found that
monocular orientation discrimination thresholds were
approximately 25% higher than binocular thresholds.
-a.
10
a>
•a
O
X
C/)
UJ
a:
|
UJ
rr
o
SPATIAL FREQUENCY
(c/deg)
Fig. 2. Orientation discrimination data (degrees) are plotted as a
function of spatial frequency (cycles/degree) for amblyopic observer
(JM). Open and filled symbols represent data for amblyopic and nonamblyopic eyes, respectively. (Two open symbols at 12 c/deg show
replication data for the amblyopic eye.) Thresholds for the amblyopic
eye increased abruptly near 15 c/deg. The inset shows detailed orientation discrimination thresholds over this range of frequencies (1017 c/deg). Arrow indicates the frequency used in a subsequent test
(see below).
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933128/ on 05/06/2017
535
ORIENTATION DISCRIMINATION IN AMDLYOPIA / Skorrun er ol.
No. 4
cs
. 5 1 2 4
16 .5 I
2
4
8
16 .5
I
2
4
8
16
SPATIAL FREQUENCY (c/deg)
Fig. 3. Interocular threshold ratios (amblyopic/non-amblyopic) for
orientation discrimination (open symbols) and contrast sensitivity
(filled symbols) are plotted against spatial frequency (cycles/degree).
Dashed lines indicate no interocular threshold difference.
Although contrast sensitivity and orientation discrimination thresholds are affected quite differently by
spatial frequency in normal observers, the relation between the amblyopic and non-amblyopic eyes shows
certain similarities between the two experiments. In
order to quantitatively compare the effects of spatial
frequency on both contrast sensitivity and orientation
discrimination, we have plotted interocular threshold
ratios (amblyopic eye/non-amblyopic eye) for both sets
of data as a function of spatial frequency (Fig. 3). Both
ratios increase with frequency for all six amblyopes,
and those amblyopes with large contrast sensitivity deficits tend to exhibit the largest orientation discrimination deficits.
The results (Figs. 1-3) show that large spatial frequency specific orientation discrimination deficits exist
in amblyopia. In the next section of this report, we
examine possible sources of these deficits. Our use of
extended gratings makes small eccentric fixation
anomalies an unlikely cause of the elevated discrimination thresholds. However, the similarity between
contrast sensitivity and orientation discrimination deficits suggests a possible connection. The elevated orientation discrimination thresholds may be a direct result of elevated contrast detection thresholds. That is,
orientation discrimination thresholds are elevated because the observers have problems seeing the stimuli.
We investigated this possibility by measuring orientation discrimination thresholds for amblyopic and
non-amblyopic eyes over a wide range of contrasts from
just above detection threshold up to very high contrasts.
Data from six amblyopes are shown in Figure 4. Each
subject was studied at a spatial frequency for which
the amblyopic eye exhibited elevated orientation dis-
crimination thresholds. The frequencies selected for
each observer have been indicated by arrows in Figures
1 and 2. Data for the non-amblyopic eyes (filled symbols) are similar to those observed for normal subjects.18
Over a limited range of contrasts above detection
threshold, orientation discrimination thresholds decrease as contrast is increased. Further increases in
contrast have little or no effect on discrimination
thresholds. This pattern can be seen best for observers
DM and RS, for whom the most complete data are
available. Data for the amblyopic eyes (open circles)
show a similar overall pattern. At all contrast levels,
orientation discrimination thresholds for the amblyopic
eyes are elevated relative to those of the non-amblyopic
eyes.
These results show that, for contrasts well above detection threshold where orientation discrimination is
relatively independent of stimulus contrast, thresholds
for the amblyopic eyes are uniformly elevated. Consequently, the two data sets can not be superimposed
by shifting the amblyopic data along the abscissa to
compensate for the elevated contrast thresholds. That
is, even though the contrast of a suprathreshold grating
may be closer to the amblyopic than the non-amblyopic
contrast threshold, this cannot account for the elevated
orientation discrimination thresholds. This experiment
demonstrates that spatial frequency specific orientation
discrimination deficits in amblyopia cannot be directly
attributed to reduced contrast sensitivity.
50
100
0
100
CONTRAST (%)
Fig. 4. Orientation discrimination thresholds (degrees) for amblyopic (open symbols) and non-amblyopic (filled symbols) eyes are
plotted as a function of grating contrast (%). Each panel shows data
from one amblyopic observer. As indicated, each observer was tested
with a single spatial frequency (shown with arrows in Figs. 1-2). (For
CW and RS orientation discrimination thresholds at some contrasts
were determined twice and lines have been drawn through their averages.)
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933128/ on 05/06/2017
536
RS
3c/deg
-s,
6
DM
8 c/deg
10
y
a.
o
Vol. 27
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / April 1986
6
4
50
CONTRAST
Fig. 5. Orientation discrimination thresholds (degrees) obtained with oblique (45 degree)
reference stimuli are plotted as a
function of contrast (%) for two
observers (RS and DM). Open
and filled symbols represent data
for amblyopic and non-amblyopic eyes, respectively. The
spatial frequencies used are given
in each panel.
I00
problem for orientation discrimination when the task
involves two stimuli separated in time (as in our temporal 2AFC procedure). Orientation differences in the
retinal image between the two presentations could be
due to torsional eye movements that occurred in the
interval separating the presentations of the two stimuli
in addition to orientation differences in the stimulus.
In an attempt to avoid this potential problem we reexamined orientation discrimination using a chevron
stimulus. In this experiment the task was to identify
which interval contained the chevron (see inset of Fig.
6). Because the two orientations to be discriminated
are now presented simultaneously, ie, upper and lower
portions of the chevron, this experiment is unlikely to
be affected by abnormal cyclotortional eye movements
(%)
We considered two other possible explanations for
our results. All orientation discrimination thresholds
described above were determined with vertical reference stimuli. Normal observers show elevated orientation discrimination thresholds with oblique reference
stimuli. 910 These elevated thresholds are present over
the entire range of visible contrasts and increase with
spatial frequency. It is possible, therefore, that the amblyopic elevated thresholds simply reflect normal performance for oblique stimuli even though the gratings
were vertical. One possible explanation for improved
performance with vertical stimuli for normal subjects
is that some extraneous information may be utilized,
eg, an "internal vertical reference" or vestibular cues.19
The elevated amblyopic orientation discrimination
thresholds found with vertical gratings may, therefore,
reflect a failure to integrate this extraneous information
with visual input received through the amblyopic eye.
In order to examine this possibility, we tested orientation discrimination as a function of contrast for
an oblique reference. Results for two observers (RS,
DM) are shown in Figure 5. Except for an overall increase in thresholds for both eyes, the data are similar
to the results obtained with vertical stimuli. Amblyopes
exhibit higher thresholds at all contrast levels. Therefore, as was shown above with vertical stimuli, elevated
thresholds at oblique orientations cannot simply reflect
impaired contrast sensitivity. The results show that reduced orientation discrimination in amblyopia is not
just a selective loss of fine orientation discrimination
for vertical stimuli.
We examined a second possible explanation for our
results. During attempted monocular fixation, amblyopes exhibit abnormally large drifts20 and saccades.21
We considered the possibility that, in addition to their
documented abnormal eye movements, amblyopes
have exaggerated cyclotorsional fluctuations. Such abnormal eye movements could provide a particular
3
a
_i
o
CO
LLJ
en
I
o
S5
o
DM
8 c/deg
50
100
CONTRAST (%)
Fig. 6. Orientation discrimination thresholds obtained with chevrons (see inset) are plotted as a function of contrast (%) for three
amblyopic observers (RS, CW, DM). Orientation discrimination
thresholds are plotted as the angle, in degrees, between the upper and
lower portion of the stimulus (see inset). Open and filled symbols
represent the results for amblyopic and non-amblyopic eyes, respectively. The spatial frequency of the grating is given for each observer.
(Although the inset shows a square-wave luminance profile, the observers were tested with sinusoidal gratings.)
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933128/ on 05/06/2017
No. 4
ORIENTATION DISCRIMINATION IN AMDLYOPIA / Skorrun er ol.
occurring between presentations. Such eye movements
would not affect orientation differences between the
two halves of the chevron. We tested orientation discrimination in this way as a function of contrast for
three amblyopes (RS, CW, and DM). The data, plotted
in Figure 6, show the same qualitative pattern previously found with the standard orientation experiment
(ie, Figs. 4-5). Orientation discrimination thresholds
for amblyopes are elevated at all contrasts. Interestingly,
two of the subjects (RS and CW) exhibited larger interocular differences with the chevron test than with
the original orientation discrimination experiment (Fig.
1). The source of these differences is unknown, but the
elevated orientation discrimination thresholds with
chevrons suggest that abnormal cyclotorsional eye
movements are not the cause of the elevated orientation
discrimination thresholds found in amblyopia. In addition, because orientation discrimination thresholds
obtained with chevron stimuli show the same contrast
independence as seen above for parallel gratings, these
elevated amblyopic thresholds are not due to contrast
sensitivity deficits.
Discussion
We have shown that orientation discrimination
thresholds for amblyopes can be normal or abnormal
depending on the spatial frequency content of the
stimulus. This rinding is consistent with previous suprathreshold studies of contrast,22 phase,6 and spatial
frequency discrimination,23 in that visual performance
for suprathreshold gratings tends to be impaired selectively in amblyopia at middle and higher spatial frequencies. Our data are not attributable to exaggerated
torsional eye movements (Fig. 6), or an absence of the
very fine thresholds for vertical gratings (Fig. 5). Unlike
amblyopic vernier acuity thresholds,7 our results cannot
be attributed to elevated contrast detection thresholds.
In contrast to the present data obtained at medium
and high spatial frequencies, Vogels et al,8 using bar
stimuli, found only small increases in orientation discrimination in amblyopic observers. Because orientation discrimination typically is near normal for low
spatial frequencies, it is possible that the amblyopes in
their study used the low-frequency components of the
bars to perform the orientation discrimination task.
The present finding of spatial frequency specific orientation discrimination deficits serves to emphasize the
importance of using narrow-band stimuli when testing
visual functions in amblyopia.
Key words: amblyopia, orientation discrimination, contrast
sensitivity
537
References
1. Duke-Elder S and Wybar K: System of Ophthalmology, Vol 6,
Ocular Motility and Strabismus. London, Henry Kimpton, 1973.
2. Hess RF and Howell ER: The threshold contrast sensitivity
function in strabismic amblyopia: evidence for a two type classification. Vision Res 17:1049, 1977.
3. Bradley A and Freeman RD: Temporal sensitivity in amblyopia:
an explanation of conflicting reports. Vision Res 25:39, 1985.
4. Rentschler I, Hilz R, and Brettel H: Amblyopic abnormality
involves neural mechanisms concerned with movement processing. Invest Ophthalmol Vis Sci 20:695, 1981.
5. Hess RF and Campbell FW: A preliminary investigation of neural
function and dysfunction in amblyopia—II. Activity within an
amblyopic "channel". Vision Res 20:755, 1980.
6. Lawden MC, Hess RF, and Campbell FW: The discriminability
of spatial phase relationships in amblyopia. Vision Res 22:1005,
1982.
7. Bradley A and Freeman RD: Is reduced vernier acuity in amblyopia due to position, contrast, or fixation deficits? Vision Res
25:55, 1985.
8. Vogels R, Orban GA, and Vandenbussche E: Meridional variations in orientation discrimination in normal and amblyopic
vision. Invest Ophthalmol Vis Sci 25:720, 1984.
9. Caelli T, Brettel H, Rentschler I, and Hilz R: Discrimination
thresholds in the two-dimensional spatial frequency domain. Vision Res 23:129, 1983.
10. Bradley A and Skottun BC: The effects of large orientation and
spatial frequency differences on spatial discriminations. Vision
Res 24:1889, 1984.
11. Brock FW and Givner I: Fixation anomalies in amblyopia. AMA
Arch Ophthalmol 47:775, 1952.
12. Miller EF: Investigation of the nature and cause of impaired
acuity in amblyopia. Am J Optom Physiol Opt 32:10, 1955.
13. Milkman N, Shapley R, and Schick G: Experimental applications.
A microcomputer-based visual stimulator. Behavior Research
Methods and Instrumentation 10:539, 1978.
14. Levitt H: Transformed up-down methods in psychoacoustics. J
Acoust Soc Am 49:467, 1971.
15. Bradley A and Freeman RD: Contrast sensitivity in anisometropic
amblyopia. Invest Ophthalmol Vis Sci 21:467, 1981.
16. Andrews DP: Perception of contour orientation in the central
fovea. Part I. Short lines. Vision Res 7:975, 1967.
17. Westheimer G, Shimamura K, and McK.ee SP: Interference with
line-orientation sensitivity. J Opt Soc Am 66:332, 1976.
18. Skottun BC, Sclar G, Bradley A, and Freeman RD: The interaction of stimulus contrast and orientation selectivity: single cells
and perception. ARVO Abstracts. Invest Ophthalmol Vis Sci
22(Suppl):208, 1982.
19. Bouma H and Andriessen JJ: Perceived orientation of isolated
line segments. Vision Res 8:493, 1968.
20. Schor C and Hallmark W: Slow control of eye position in strabismic amblyopia. Invest Ophthalmol Vis Sci 17:577, 1978.
21. Schor CM and Flom MC: Eye position control and visual acuity
in strabismus amblyopia. In Basic Mechanisms of Ocular Mobility and Their Clinical Implications. Lennerstrand G and Bachy-Rita P editors Oxford, Pergamon Press, 1975, pp. 555.
22. Hess RF, Bradley A, and Piotrowski L: Contrast coding in amblyopia. I. Differences in the neural basis of human amblyopia.
Proc R Soc Lond [Biol] 217:309, 1983.
23. Hess RF, Burr DC, and Campbell FW: A preliminary investigation of neural function and dysfunction in amblyopia—III.
Co-operative activity of amblyopic channels. Vision Res 20:757,
1980.
Downloaded From: http://arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933128/ on 05/06/2017