Terapias visuales para todas las edades

1. ORIGINAL ARTICLE
Binocular Balance in Normal Vision and Its
Modulation by Mean Luminance
Peng Zhang*, William Bobier†
, Benjamin Thompson‡
, and Robert F. Hess‡
ABSTRACT
Purpose. To better understand the neural basis of sensory dominance in the normal population, we applied a recently
established test designed to quantify the degree of suppression in amblyopia to participants with normal binocular vision.
Methods. This test quantifies the degree of dichoptic imbalance in coherent motion sensitivity by manipulating the
contrast of stimuli seen by the two eyes. The contrast at which balanced dichoptic motion sensitivity occurs is referred
to as the “balance point” and is an estimate of the degree of suppression. We apply the same logic to the measurement
of sensory dominance by measuring the distribution of “balance points” within the normal population.
Results. We show that although most subjects are balanced or only weakly imbalanced, a minority is strongly
imbalanced. To ascertain the site of sensory dominance, we assessed the degree to which normal sensory balance can
be modulated by changing the interocular mean luminance. We found that mismatches in mean luminance between the
two eyes had a pronounced effect on the balance point determination.
Conclusions. Because cells in the lateral geniculate nucleus exhibit a strong modulation to sustained changes in the mean
light level, this may suggests that the inhibitory circuits underlying sensory eye dominance are located at a precortical site.
(Optom Vis Sci 2011;88:1072–1079)
Key Words: eye dominance, binocular balance, mean luminance
T
he current model of binocular vision involves inhibitory as well
as excitatory connections.1
Although the excitatory combina-
tion of the signals from the two eyes is well known and thought
to occur in the early layers of the striate cortex,2
the inhibitory influ-
ences have only recently been appreciated by psychophysicists.3
There
is, however, a long history of inhibitory binocular interactions in the
neurophysiological literature, especially in the lateral geniculate nu-
cleus (LGN)4–6
and cortex.7
For a balanced binocular system, that is
one that does not exhibit a sensory ocular dominance both the excit-
atory and inhibitory interocular influences are in balance. Any imbal-
ance will produce a sensory dominance that has potential clinical
implications.8
For example, the management of monovision associ-
ated with contact lenses, refractive, and cataract surgery benefits
greatly from a knowledge of sensory dominance,9–12
something that
we presently do not assess in the clinic. Motor dominance, which is
measured clinically has been shown to be uncorrelated with sensory
dominance.13
There is a need for an accurate assessment of sensory
dominance in the clinic.
Suppression associated with amblyopia is the end result of a fully
imbalanced binocular system. In this case, there is an attenuation of
the monocular signal, and a resultant strong suppressive influence
from the fixing eye’s input on that of the amblyopic eye. Mansouri et
al.14
proposed a method by which the degree of imbalance in these
signals could be measured psychophysically. They used a global mo-
tion stimulus in which signal elements moving in a coherent direction
were presented to one eye and noise elements moving in a random
direction were presented to the other eye. Performance was quantified
by the signal/noise ratio associated with correct identification of
the signal elements’ direction. Binocular combination was quanti-
fied by the extent to which the threshold varied depending on
which eye saw the signal and which eye saw the noise. By manip-
ulating the interocular contrast, balanced performance (where it
did not matter which eye saw the signal and which saw the noise)
on this motion task could be restored, suggesting that under the
right conditions (appropriate interocular contrast), the two eyes
can work together. They proposed that the contrast ratio deter-
mined in this way would be a good measure of the degree of
imbalance before excitatory summation and advocated its use in
cases of amblyopia to quantify the degree of suppression.
*BSc
†
PhD, FAAO
‡
PhD
School of Optometry, University of Waterloo, Ontario, Canada (PZ, WB),
Department of Optometry and Vision Science, University of Auckland, New Zea-
land (BT), and the Department of Ophthalmology, McGill University, Quebec,
Canada (RFH).
1040-5488/11/8809-1072/0 VOL. 88, NO. 9, PP. 1072–1079
OPTOMETRY AND VISION SCIENCE
Copyright © 2011 American Academy of Optometry
Optometry and Vision Science, Vol. 88, No. 9, September 2011

2. If one views suppression in amblyopes as an extreme version of
normal sensory dominance, then a similar approach could be taken to
quantify the degree of sensory dominance in normals. Li et al.,8
ap-
plied an abbreviated version of this same approach in a group of
normal observers to better understand ocular dominance. Li et al.8
proposed that the degree to which the aforementioned inhibitory in-
teractions are balanced provides an explanation for and a means of
measuring sensory dominance. They measured the degree of sensitiv-
ity imbalance for stimuli of equal contrast (i.e., the extent to which it
matters which eye sees the noise and which eye sees the signal) and
found that this was well correlated with the extent to which an
observer’s performance was constant across a range of clinical eye
dominance measures. Those with a small sensitivity imbalance exhib-
ited variability across a range of clinical eye dominance tests, whereas
those with a stronger imbalance showed a greater consistency across
clinical tests. They also reported that the normal population is com-
posed of two dominance groups, one with mild dominance (the ma-
jority) and one with strong dominance.
The previous application of this approach to the investigation of
eye dominance in the normal population involved the use of stim-
uli of only one contrast level (i.e., a high contrast) that was the same
in both eyes. This was done to make the measurements convenient
and practical in a clinical setting. However, because this previous
study did not measure the interocular contrast at which balanced
dichoptic performance was obtained in the normal visual system
(as originally suggested by Mansouri et al.14
), it relies on the as-
sumption that there is a linear relationship between the extent of
the initial dichoptic sensitivity imbalance (dichoptic sensitivity ra-
tio for stimuli of equal high contrast) and the interocular contrast
ratio required to balance dichoptic performance.
In this study, we set out with two aims. The first was to test this
assumption by quantifying dominance within the normal population
intermsofmeasurementsofafullrangeofinterocularcontrastssothat
the interocular contrast corresponding to balanced dichoptic perfor-
mance (term the balance point) could be obtained. The second was to
assess whether balanced performance is susceptible to changes
in the luminance as well as the contrast between the eyes. The
reason why it is of interest to investigate whether a change in the
interocular luminance alters the balance point measured with
our contrast-varying paradigm is because luminance effects gen-
erally occur at precortical levels and this would then bear on the
possible site of this balancing operation. For example, cells in
the LGN do respond to changes in the mean light level as well
as the contrast15
and even though the input from each eye is
kept separate in the laminar structure of the LGN, inhibitory
interocular interactions have been reported between cells from
different lamina.4,5,16 –18
The balance between these subcortical
signals may underlie sensory dominance.
METHODS
Subjects
Twenty-five naive observers ages ranged from 19 to 36 years,
recruited from the School of Optometry, University of Water-
loo were included in the study. Informed consent was obtained
before the tests, and the study was approved by the Office of
Research Ethics (ORE 15,721) of University of Waterloo.
Before the tasks, each subject underwent a series of clinical tests
to ensure that the inclusion criteria were met. These included
normal vision with 20/20 or better after a subjective refraction; the
absence of any binocular deficits (i.e., amblyopia, strabismus); no
oculomotor abnormalities such as strabismus; no ocular surgery
history; and stereoacuity of Ͻ50 s of arc. Exclusion criteria in-
cluded any history of a binocular vision disorder involving a con-
stant or intermittent tropia.
Unilateral and alternate cover tests were performed on the ob-
servers to ensure the absence of strabismus, and the Modified-
Thorington-test was used to measure their phoria level. Visual
acuity was assessed with computerized Test Chart 2000 Pro on
logarithm of the minimum angle of resolution scale; and stereoa-
cuity was measured with Randot Stereo graded circle test.
Eye Dominance Assessment
Before the observers started the motion coherence tasks, their
motor ocular dominance was assessed. Four tests were performed
in this study to determine motor dominance.
The Hole-in-Card Test (The Dolman Method)
Observers were instructed to hold a card with both hands about
40 cm from their eyes, and align a target at 6 m through the hole in
the card with both eyes open. The experimenter then determined
dominance by asking participants to alternatively close each eye to
determine which eye was aligned with the target.
The Hole-in-Cone Test (A Modification of
Dolman Method)
A cone was made out of a sheet of A4 size paper and observers
were asked to hold it with both hands with the base of the cone in
front of their eyes. Through the hole at the cone’s apex, the ob-
servers aligned a 6 m distant target with both eyes open. The
experimenter then determined the dominance by asking observers
to alternatively close each eye to determine which eye was aligned
with the target.
The Point-a-Finger Test (The Porta Test)
Observers were instructed to extend both arms and put one
thumb over the other. They were then asked to align their
thumbs to a 6 m distant target with both eyes both. Dominance
was determined by alternatively closing each eye to determine
the sighting eye.
The Worth-4-Dot Test
A standard Worth-4-dot test was performed at near (0.4 m) and
far distances (6 m). The observers wore anaglyphic glasses and were
presented with four dots from a flashlight to measure their sup-
pression. Each of the clinical dominance tests provides only a crude
binary estimate of dominance. Each test was done twice and in all
cases, there was agreement between the two measures.
Modulation of Binocular Balance in Normal Vision—Zhang et al. 1073
Optometry and Vision Science, Vol. 88, No. 9, September 2011

3. Dichoptic Motion Thresholds
Apparatus
Stimuli were presented using a MacBook Pro laptop computer
running Matlab (Mathworks, Natick, MA) and Psychophysics
Toolbox, Version 3.19
The stimuli were displayed using a Z800
duel pro headmounted display (eMagin Corporation, Hopewell
Junction, NY). This headmounted display model contains two
OLED screens, one for each eye. The screens have a high lumi-
nance, a linear luminance response profile and refresh simultane-
ously at 60 Hz therefore avoiding motion smear. The device also
allows for different stimuli to be presented to each eye. To achieve
this, each frame of the dichoptic stimulus was computed as a single
image with a resolution of 600 ϫ 1600 pixels. A Matrox
DuelaHead2Go external video board was then used to split each
frame between the two headmounted display screens at a resolu-
tion of 600 ϫ 800 pixels per screen. A photometer (United De-
tector Technology) was used to ensure equal luminance of the two
screens and to perform gamma correction.
Stimuli and Task
Stimuli were random dot kinematograms based on those used
by Mansouri et al.14
(Fig. 1). One hundred dots [with dot lumi-
nance modulation varied according to ͫ(Ldots Ϫ Lbackground)
(Ldots ϩ Lbackground)ͬwere
displayed on a mean luminance background of 50 cd/m2
. Each dot
had a radius of 0.5° and moved at 6°/s. The dots had a limited
lifetime whereby on any single frame, each dot had a 5% chance of
disappearing and being redrawn in a new spatial position. Dots
were presented within a circular display aperture with a radius of
11.1° that was framed by a solid black square outline to aid fusion.
To avoid interaction of the stimulus dots with the central dark
fixation dot (radius 0.35°), stimulus dots did not enter the central
region of the display aperture (radius 2°). Dots that passed through
this central region disappeared and were redrawn on the opposite
side of the central area with the appropriate temporal delay to
maintain a constant speed. When stimulus dots reached the edge of
the display aperture, they were wrapped around. Stimuli were
shown for 1 s.
In each trial, one eye was presented with a population of “signal”
dots that all moved in the same direction (left or right). The other
eye was presented with the noise dots that moved in random di-
rections. The task was to indicate the motion direction of the signal
dots. The total number of dots were fixed (i.e., 100) and the ratio
of the signal dots‫گ‬noise dots was varied with our psychophysical
procedure. To measure the threshold number of signal dots re-
quired for 79% correct performance (the motion coherence
threshold), the number of signal dots was varied on a trial-by-trial
basis using a 3-down 1-up staircase procedure with a proportional
step size of 50% before the first reversal and 25% thereafter. The
starting point for each staircase was 100 signal dots and 0 noise
dots. When dots were removed from the signal population, they
were added to the noise population and vice versa. Each staircase
consisted of six reversals and the last five reversals were averaged to
estimate threshold. During each set of measurements, 10 staircases
were randomly interleaved. This allowed for threshold measure-
ments to be made at five contrast offsets between the two eyes with
signal presented to either the dominant or non-dominant eye
(based on clinical sighting tests). The contrast of the dots in the
non-dominant eye was fixed at a high contrast (80–100%).
The contrast of the dots in the dominant eye was varied across the
following contrast ratios (1, 0.5, 0.25, and 0.125). Each full set of
measurements took ϳ20 min to complete and the measurements
were repeated twice to ensure accurate thresholds were obtained.
FIGURE 1.
The stimuli used for dichoptic motion coherence threshold measurements. In this schematic representation, all the dots in the left eye are moving to
the left and constitute the signal dot population. The dots in the right eye are moving in random directions and constitute the noise population. Arrows
are for illustration purposes and were not presented in the actual stimulus. (Reproduced from Li, et al. (2010). Invest Ophthalmol Vis Sci. 2010
Dec;51(12):6875–81. Copyright Invest Ophthalmol Vis Sci.)
1074 Modulation of Binocular Balance in Normal Vision—Zhang et al.
Optometry and Vision Science, Vol. 88, No. 9, September 2011

4. The data relating to the relationship between interocular contrast
and dichoptic performance was subjected to a linear fit (orthogonal
linear regression) and the balance point derived from the contrast
corresponding to the intersection of the linear fits to the data for
the dominant and non-dominant eyes.20
Each participant was familiarized with the stimuli and task using
a demonstration program where stimuli were presented continu-
ously and the proportion of signal and noise dots could be con-
trolled using the up and down arrow keys on the laptop keyboard.
Once participants were familiar with the task, motion coherence
threshold measurements began. Each set of threshold measure-
ments began with two square stimulus frames presented separately
to each eye with nonius lines next to the fixation marks. Using the
arrow keys on the laptop keyboard, participants could adjust
the position of the stimulus in the non-dominant eye to ensure that
the images in the two eyes were perfectly aligned and fused. The
participant then pressed a key to initiate the threshold measure-
ments. The left and right arrow keys on the laptop keyboard were
used to report the percept of leftward and rightward signal dot
motion, respectively. The testing was self-paced with each stimulus
being shown 250 ms after the response to the preceding trial.
Rationale for the Balance Point Measurement
We wanted to use an approach for which it was clear when and
to what extent information was being combined between the eyes.
We decided to use a signal/noise approach where the signal was
presented to one eye and the noise to the other. If the information
is rigidly combined between the two eyes, then an obvious decre-
ment in performance would result because the noise in one eye
would make it harder to detect the signal in the other eye. In our
case, the signal was a group of randomly placed dots all moving in
the same direction. The subject’s task was to detect this motion
direction. The noise consisted of spatially intermingled dots, each
of which moved at the same speed but in a random direction (Fig.
1). If the threshold for determining the signal direction was limited
by the noise, then information must have been combined binocu-
larly. In a perfectly balanced visual system (i.e., with no sensory
dominance), this argument does not depend on which eye sees the
noise and which eye sees the signal. However, if there is an imbal-
ance, noise in the dominant eye will be more effective than if seen
through the non-dominant eye. Our method of quantifying the
degree of imbalance was to vary the interocular contrast until there
is balanced performance, that is when it no longer matters (i.e.,
equal performance) which eye sees the signal and which eye sees the
noise. Thus, we measured the dichoptic threshold ratio (i.e., the
ratio between the direction detection performance measured in
terms of a signal/noise ratio for when the signal is in one eye and
the noise in the other compared with vice versa) as a function of the
interocular contrast of the stimuli (be they signal in some cases and
noise in other cases) seen by each eye. (The work of Mansouri et
al.14
shows how this approach was developed from research into
suppression.)
Statistical Procedures
Balance points were calculated for each participant by determin-
ing the intersection of linear fits to both the dominant eye and
non-dominant eye motion coherence thresholds as a function of
the contrast shown to the dominant eye. A dominant eye threshold
was measured when the signal dots were seen by the dominant eye
and the noise dots by the non-dominant eye. A non-dominant eye
threshold was measured when the non-dominant eye saw the signal
dots. A description of the rationale for this technique is provided
above and an example dataset is provided in the first paragraph of
the results section. Motion coherence threshold dominance ratios8
were calculated using only the thresholds for which both eyes were
presented with the same contrast and were calculated using the
following formula:
(non-dominant eye threshold Ϫ dominant eye threshold)/
(non-dominant eye threshold ϩ dominant eye threshold)
Therefore, 0 indicates no difference between the eyes, a positive
value indicates a lower threshold for the dominant eye (i.e., better
performance by the dominant eye) and a negative value indicates a
lower threshold for the non-dominant eye.
One-Sample Kolmogorov-Smirnov Tests were used to test that
our independent variables were normally distributed. As both the
balance point data and the motion coherence threshold dominance
ratio data were normally distributed, one sample t-tests were used
to assess whether these measures differed from unity between the
eyes (i.e., a ratio of 1 for the balance point data or a dominance
ratio of 0 for the motion coherence threshold dominance ratio
data). A Pearson product moment correlation coefficient was also
calculated to assess whether the two measures of eye dominance
were correlated. The non-parametric Spearman rho test was used
to assess the relationships between variables that did not meet the
assumptions for parametric tests.
Neutral Density Filters
The mean luminance of one eye was varied by the use of Kodak
Wratten neutral density filters that were placed in front of the
non-dominant eye. The filter values used varied from 1 ND (lu-
minance reduction of a factor of 10) to 3 ND (reduction of a factor
of 1000). The mean luminance was 50 cd/m2
so that in the most
extreme case of 3 ND, the filter restricted vision to the upper
mesopic level. Time was allocated for sufficient dark adaptation
before testing; however, because the subjects previous light expo-
sure was restricted to moderate indoor lighting, this could be rel-
atively short (i.e., 5 min).
RESULTS
Fig. 2 shows an example measurement of the balance point
using the approach outlined above. Here, the dichoptic threshold
for motion direction detection (% signal elements) is plotted
against the interocular contrast ratio for the case where the signal
was presented to the dominant (right) eye (open circles) and noise
to the non-dominant (left) eye as opposed to vice versa (filled
triangles). The contrast presented to the non-dominant eye re-
mains fixed whereas the contrast presented to the dominant eye
varies from being the same as the contrast shown to the non-
dominant eye (interocular contrast ratio of 1) to 20% of the con-
trast shown to the non-dominant eye (ratio of 0.2). Each data set
Modulation of Binocular Balance in Normal Vision—Zhang et al. 1075
Optometry and Vision Science, Vol. 88, No. 9, September 2011

5. has been fit with a linear function using orthogonal linear regres-
sion and the point of intersection represents the contrast imbal-
ance, our measure of the degree of sensory imbalance, at which it
did not matter which eye saw the signal and which eye saw the
noise, thresholds were the same. This is marked with the solid
black arrow in Fig. 2. In this case, as the contrast of the signal is
decreased in the dominant eye, performance deteriorates for the
dominant eye and improves for the non-dominant eye. At the
balance point, which for this participant was an interocular con-
trast ratio of 0.8, the contrast reduction of the dominant eye has
neutralized its initial advantage (i.e., the sensory dominance). This
result suggests that the right eye is the dominant eye and that the
degree of dominance is equivalent to a contrast reduction of 20%.
The first question we address concerns the relationship between
the dichoptic threshold ratio previously measured in normals,8
and the interocular contrast ratio associated with balanced dichop-
tic performance (termed the balance point). The mean contrast
ratio at balance point was 0.88 (SD, 0.18) indicating that, on
average, the dominant eye required 88% contrast when the non-
dominant eye was presented with 100% contrast to achieve di-
choptically matched motion coherence thresholds. This bias in
contrast ratios toward the dominant eye was reliable for the group,
i.e., the ratios were reliably less than one, t(23) ϭ 3.4, p ϭ 0.003.
Consistent with this result is the finding that when both eyes were
shown the same contrast, the average motion coherence threshold
dominance ratio was 0.04 (SD, 0.22) indicating a slight, but in this
case non-significant (p Ͼ 0.05), bias toward the dominant eye.
Therefore, as a group, our sample of observers with normal binoc-
ular vision had well-balanced interocular inhibition for this task.
The distribution of contrast ratios at balance point is shown in Fig.
3. It is clear that most of the balance points are close to unity
indicating well-balanced interocular inhibition. Some values were
greater than unity demonstrating that our dichoptic motion coher-
ence test does not always indicate the same eye dominance as the
sighting tests used to categorize eyes as dominant vs. non-dominant
before running the balance point procedure. This is consistent with
previous findings for individuals with relatively weak eye dominance.8
There were also three participants who demonstrated more
pronounced contrast imbalances between the two eyes with intero-
cular contrast ratios of 0.7 and below, suggesting a stronger imbal-
ance between the eyes in favor of the dominant eye. The strength of
the imbalance is measured in terms of interocular contrast and thus
is an indirect measure. The distribution of the dominance ratios for
motion coherence thresholds when the same contrast was shown to
each eye is shown in Fig. 4. Again, the distribution is bimodal with
most participants showing balanced performance between the eyes
and a minority of participants showing a stronger imbalance in
favor of the dominant eye. There was a significant correlation
between the contrast ratios at balance point and the motion coher-
ence dominance ratios when both eyes saw the same contrast (r ϭ
Ϫ0.79, p Ͻ 0.001; n ϭ 24, Fig. 4), indicating good agreement
between these two measures of interocular suppression. The pat-
tern of eye dominance whereby most observers have weak domi-
nance with a minority exhibiting more pronounced dominance is
consistent with previous reports.8,10,13,21,22
FIGURE 3.
The distribution of contrast ratios that gave matched dichotpic motion
coherence threshold ratios. A value of 1 indicates a perfect balance
between the eyes whereby the same contrast was required by both eyes
for matched dichoptic motion coherence thresholds. A value Ͻ1 indicates
that the dominant eye required less contrast than the non-dominant eye
and a value of Ͼ1 indicates that the dominant eye required more contrast
than the non-dominant eye.
FIGURE 4.
The distribution of motion coherence threshold dominance ratios when
the same contrast was presented to both eyes. A dominance ratio of 0
indicates balanced performance between the two eyes. A positive domi-
nance ratio indicates that the dominant eye thresholds were lower than
the non-dominant eye thresholds (i.e., less signal dots were required when
the noise was presented to the non-dominant eye than when the noise was
presented to the dominant eye). Negative dominance ratios indicate the
opposite relationship.
FIGURE 2.
Example data from a single participant illustrating the technique used to
determine the balance point contrast ratio. A full description of this
procedure is provided in the text.
1076 Modulation of Binocular Balance in Normal Vision—Zhang et al.
Optometry and Vision Science, Vol. 88, No. 9, September 2011

6. The mean motion coherence thresholds, in % signal dots, were
21.1 (SD, 9.6) and 18.6 (SD, 8.1) for the dominant and non-
dominant eyes, respectively, when stimuli were shown to each eye
at the same contrast. The average motion coherence threshold at
the balance point was 20.0 (SD, 7.0). As would be expected for a
populationwithnormalbinocularvisualfunction,wedidnotfindany
correlations between the balance point measure and interocular acuity
difference or the type or magnitude of any phoria.
The second question concerns the possible site (or locus along the
visual pathway) of the suppressive effects measured here. In particular,
we wondered if we could simulate the type of mild suppression one
sees in the normal population and the type of severe suppression one
seesinamblyopiabyreducingthemeanluminancetooneeye.Wedid
this using neutral density filters fitted into light-tight goggles so that
the contrast of stimuli would be unaffected. Because cells in the visual
cortex are relatively unresponsive to sustained changes in mean lumi-
nance compared with their counterparts in the lateral geniculate,6
such a simulation would suggest potential geniculate involvement in
the inhibitory circuit.
In a subset of our observers with normal binocular function, we
were able to replicate the strong imbalance between the two eyes
that has previously been reported for observers with amblyo-
pia14,23,24
using the neutral density filter technique. The results for
one example observer are shown in Fig. 6 where the motion coher-
ence thresholds for each eye are plotted as a function of the intero-
cular contrast ratio. Linear fits using orthogonal linear regression
were then made for each dataset and the intersection of the fits
(indicated by the solid arrows) is the point at which equal perfor-
mance was achieved between the two eyes. Panel A shows the
thresholds for this participant without an ND filter. Under these
conditions, there is a normal balance between the two eyes. Panels
B to D show the results when a 1 log unit, 2 log unit, and 3 log unit
neutral density filter was placed over the non-dominant eye. It is
clear that the balance point is gradually shifted toward lower dom-
inant eye contrasts (smaller interocular contrast ratios), until for
FIGURE 5.
The relationship between the motion coherence threshold dominance
ratio when the same contrast was presented to both eyes (y axis) and the
contrast ratio at the balance point (x axis). The line of best fit found using
orthogonal linear regression is shown by the dashed line.
FIGURE 6.
The measurement of the balance point when neutral density filters are placed before the non-dominant eye. The motion coherence threshold (% signal
dots) is plotted against the contrast presented to the dominant eye (non-dominant eye contrast is fixed at 100%). Results are shown for when the
dominant eye sees the signal and for when the non-dominant eye sees the signal. Each data set is fitted with a linear function. The contrast corresponding
to the intersection of these linear functions represents the balance point measure. As the neutral density filter increases, the balance point is progressively
displaced to lower contrasts. Data were from one representative participant with normal binocular vision.
Modulation of Binocular Balance in Normal Vision—Zhang et al. 1077
Optometry and Vision Science, Vol. 88, No. 9, September 2011

7. the 3 log unit filter the linear fits no longer converge within the
range of interocular contrasts provided. This pattern of results is
indicative of a gradual increase in the imbalance between the two
eyes. Similar results were collected for a group of five normal ob-
servers in which the balance point was derived for a series of neutral
density filters (0, 1, 2, and 3 ND) fitted in front of the non-
dominant eye. These neutral density filter results for a group of
normal participants are shown in Fig. 7.
The data in Fig. 7 for a group of normal observers shows how the
interocular contrast ratio (corresponding to balanced dichoptic
performance) varies with the magnitude of mean luminance reduc-
tion (over a range of 3 log units or ϫ1000) in the non-dominant
eye. There is an orderly reduction in the contrast of the stimuli seen
by the dominant eye required to balance the suppressive effects
induced by the reduced mean luminance in the non-dominant eye.
In other words, a change in interocular mean luminance can cause
strong interocular suppressive effects, similar to that previously
reported in amblyopia.14
DISCUSSION
Ocular dominance is a measure that is clinically useful in
determining the suitability of monovision for contact lens wear,21
cataract surgery,10
and for the correction of presbyopia using re-
fractive surgery.12,25
It is sometimes determined by alternating a
plus 1.5 D lens in front of each eye and determining which eye
tolerates the blur best. That eye is taken as the non-dominant eye.
Other times, a test of motor dominance is used. The basis of these
tests are poorly understood, as sensory dominance correlates with
neither motor dominance13,21,26,27
nor monocular visual sensitiv-
ity.13
Li et al.8
sought an explanation in terms of a recently pro-
posed model of binocular combination,1
which incorporates both
inhibitory and excitatory interactions. In particular, they won-
dered whether ocular dominance is determined by the extent to
which the contralateral inhibitory signals are balanced and they
provided support for the hypothesis in terms of the dichoptic sen-
sitivity ratio using a motion coherence task. They found a strong
correlation between this measure and a more traditional clinical
test for sensory dominance and went on to show that the normal
population is composed of two overlapping dominance groups,
whereby the majority of participants (61%) showed weak domi-
nance but a significant minority (39%) showed strong dominance.
Their conclusion was based only on the measurement of the di-
choptic coherence ratio for stimuli of equal contrast as the measure-
ments were optimized for clinical utility. To provide a more complete
picture of the role that interocular inhibitory interactions may play in
eye dominance, we measured both the balance point, i.e., the contrast
ratio at which the dichoptic coherence ratio is at unity,14
as well as the
threshold ratio at matched high contrast in a group of binocularly
normal individuals. Confirming the results of Li et al.,8
we found a
significant correlation between these two measures and provided fur-
ther support for the existence of two dominance distributions in the
normal population. This was characterized by the majority of subjects
exhibiting balanced or weak dominance but a minority exhibiting
strong dominance. Knowing the strength of sensory dominance has
potentialclinicalvaluethoughatpresentitsmeasurementisnotpartof
standard clinical practice.
Because dominance cannot be predicted solely on the basis of
monocular sensitivity,13
its site along the visual pathway must be at
a stage where neurons receive binocular input. The striate cortex
and in particular layer 4 is where binocular combination first takes
place and it represents the obvious candidate. However, the role of
the LGN cannot be discounted because there are reports of inhib-
itory binocular interactions between cells from right and left eye
laminae4,5,16–18
and also because the feedback from layer 6 of the
striate cortex to the geniculate is known to affect both right and
right eye inputs.28,29
One striking difference between cells in the
LGN and cortex relates to their response to the mean light level.
Geniculate cells having a high resting level are very responsive to
sustained changes in mean luminance whereas cortical cells have
virtually no resting level15
and are not sensitive to changes in mean
luminance (but see ref 30). We wondered whether changes in the
mean interocular light level could affect the dominance when the
interocular contrast was unchanged. If dominance was exclusively
cortical, one would not expect such a stimulus manipulation to
have much effect; however, if dominance also involves the LGN,
mean luminance differences between the eyes could well modulate
dominance. We found that changes in mean luminance (where
stimulus contrast is unaltered) do systematically affect our mea-
surement of the balance point and hence our estimation of domi-
nance; the larger the interocular ratio of mean luminance, the
greater the change in dominance. This is also the case for suppres-
sion in strabismic amblyopia where changes in mean luminance
and contrast have been linked to the suppressed function.31
One
might hypothesize that in normals, although the excitatory com-
bination of left and right eye input takes place in the cortex, the
inhibitory contralateral effects may occur at the level the LGN.
ACKNOWLEDGMENTS
This work was supported by a CIHR (MT53346) grant (to RFH).
Received November 4, 2010; accepted April 14, 2011.
FIGURE 7.
Balance point data as a function of the strength of neutral density filter
placed over the non-dominant eye. The interocular contrast ratio corre-
sponding to the balance point (Fig. 6) is plotted against the value of the
neutral density filter (log units). As shown in the individual example in Fig.
6, results for the group of five subjects show a similar displacement to
lower contrast ratios as the value of the neutral density filter increases. The
dashed line is the best linear fit.
1078 Modulation of Binocular Balance in Normal Vision—Zhang et al.
Optometry and Vision Science, Vol. 88, No. 9, September 2011

8. REFERENCES
1. Meese TS, Georgeson MA, Baker DH. Binocular contrast vision at
and above threshold. J Vis 2006;6:1224–43.
2. Hubel DH, Wiesel TN. Ferrier lecture. Functional architecture of
macaque monkey visual cortex. Proc R Soc Lond B Biol Sci 1977;
198:1–59.
3. Meese TS, Hess RF. Low spatial frequencies are suppressively masked
across spatial scale, orientation, field position, and eye of origin. J Vis
2004;4:843–59.
4. Sanderson KJ, Bishop PO, Darian-Smith I. The properties of the
binocular receptive fields of lateral geniculate neurons. Exp Brain Res
1971;13:178–207.
5. Sanderson KJ, Darian-Smith I, Bishop PO. Binocular corresponding
receptive fields of single units in the cat dorsal lateral geniculate nu-
cleus. Vision Res 1969;9:1297–303.
6. Schroeder CE, Tenke CE, Arezzo JC, Vaughan HG, Jr. Binocularity
in the lateral geniculate nucleus of the alert macaque. Brain Res 1990;
521:303–10.
7. Tsumoto T. Inhibitory and excitatory binocular convergence to vi-
sual cortical neurons of the cat. Brain Res 1978;159:85–97.
8. Li J, Lam CS, Yu M, Hess RF, Chan LY, Maehara G, Woo GC,
Thompson B. Quantifying sensory eye dominance in the normal
visual system: a new technique and insights into variation across tra-
ditional tests. Invest Ophthalmol Vis Sci 2010;51:6875–81.
9. Evans BJ. Monovision: a review. Ophthalmic Physiol Opt 2007;27:
417–39.
10. Handa T, Mukuno K, Uozato H, Niida T, Shoji N, Minei R, Nitta
M, Shimizu K. Ocular dominance and patient satisfaction after
monovision induced by intraocular lens implantation. J Cataract Re-
fract Surg 2004;30:769–74.
11. Schor C, Erickson P. Patterns of binocular suppression and accom-
modation in monovision. Am J Optom Physiol Opt 1988;65:
853–61.
12. Wright KW, Guemes A, Kapadia MS, Wilson SE. Binocular function
and patient satisfaction after monovision induced by myopic photo-
refractive keratectomy. J Cataract Refract Surg 1999;25:177–82.
13. Suttle C, Alexander J, Liu M, Ng S, Poon J, Tran T. Sensory ocular
dominance based on resolution acuity, contrast sensitivity and align-
ment sensitivity. Clin Exp Optom 2009;92:2–8.
14. Mansouri B, Thompson B, Hess RF. Measurement of suprathreshold
binocular interactions in amblyopia. Vision Res 2008;48:2775–84.
15. Troyer TW, Krukowski AE, Miller KD. LGN input to simple cells
and contrast-invariant orientation tuning: an analysis. J Neuro-
physiol 2002;87:2741–52.
16. Xue JT, Ramoa AS, Carney T, Freeman RD. Binocular interaction in
the dorsal lateral geniculate nucleus of the cat. Exp Brain Res 1987;
68:305–10.
17. Pape HC, Eysel UT. Binocular interactions in the lateral geniculate
nucleus of the cat: GABAergic inhibition reduced by dominant affer-
ent activity. Exp Brain Res 1986;61:265–71.
18. Rodieck RW, Dreher B. Visual suppression from nondominant eye
in the lateral geniculate nucleus: a comparison of cat and monkey.
Exp Brain Res 1979;35:465–77.
19. Brainard DH. The Psychophysics Toolbox. Spat Vis 1997;10:433–6.
20. Black JM, Thompson B, Maehara G, Hess RF. A compact clinical
instrument for quantifying suppression. Optom Vis Sci 2011;88:
E334–43.
21. Seijas O, Gomez de Liano P, Gomez de Liano R, Roberts CJ, Piedra-
hita E, Diaz E. Ocular dominance diagnosis and its influence in
monovision. Am J Ophthalmol 2007;144:209–16.
22. Yang E, Blake R, McDonald JE, II. A new interocular suppression
technique for measuring sensory eye dominance. Invest Ophthalmol
Vis Sci 2009;51:588–93.
23. Hess RF, Mansouri B, Thompson B. A new binocular approach to
the treatment of amblyopia in adults well beyond the critical period of
visual development. Restor Neurol Neurosci 2010;28:793–802.
24. Hess RF, Mansouri B, Thompson B. A binocular approach to treat-
ing amblyopia: antisuppression therapy. Optom Vis Sci 2010;87:
697–704.
25. Jain S, Arora I, Azar DT. Success of monovision in presbyopes: review
of the literature and potential applications to refractive surgery. Surv
Ophthalmol 1996;40:491–9.
26. Handa T, Mukuno K, Uozato H, Niida T, Shoji N, Shimizu K.
Effects of dominant and nondominant eyes in binocular rivalry. Op-
tom Vis Sci 2004;81:377–83.
27. Kommerell G, Schmitt C, Kromeier M, Bach M. Ocular prevalence
versus ocular dominance. Vision Res 2003;43:1397–403.
28. Freeman RD, Tsumoto T. An electrophysiological comparison of
convergent and divergent strabismus in the cat: electrical and visual
activation of single cortical cells. J Neurophysiol 1983;49:238–53.
29. Tsumoto T, Freeman RD. Effects of strabismus on development of
cortico-geniculate projections in the kitten. Exp Brain Res 1981;44:
337–9.
30. Geisler WS, Albrecht DG, Crane AM. Responses of neurons in pri-
mary visual cortex to transient changes in local contrast and lumi-
nance. J Neurosci 2007;27:5063–7.
31. Maehara G, Thompson B, Mansouri B, Hess RF. The roles of con-
trast and luminance in amblyopic suppression. J Vision 2009;9:1055.
Robert F. Hess
The Department of Ophthalmology
McGill University
687 Pine Avenue West Rm H4-14
Montreal, Quebec H3A 1A1
Canada
e-mail: robert.hess@mcgill.ca
Modulation of Binocular Balance in Normal Vision—Zhang et al. 1079
Optometry and Vision Science, Vol. 88, No. 9, September 2011