Hess Et Al-2014-Clinical and Experimental Optometry PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 10

C L I N I C A L

A N D

E X P E R I M E N T A L

OPTOMETRY
RESEARCH PAPER

The iPod binocular home-based treatment for amblyopia in adults:


efficacy and compliance
Clin Exp Optom 2014; 97: 389398
Robert F Hess* DSc
Raiju Jacob Babu OD
Simon Clavagnier* PhD
Joanna Black OD
William Bobier PhD
Benjamin Thompson PhD
* McGill Vision Research, Department of
Ophthalmology, McGill University, Montreal,
Quebec, Canada

Department of Optometry and Vision Science,


University of Waterloo, Waterloo, Ontario, Canada
and Department of Optometry and Vision Science,
University of Auckland, Auckland, New Zealand

Submitted: 5 April 2014


Revised: 17 May 2014
Accepted for publication: 3 June 2014

DOI:10.1111/cxo.12192
Background: Occlusion therapy for amblyopia is predicated on the idea that amblyopia is
primarily a disorder of monocular vision; however, there is growing evidence that patients
with amblyopia have a structurally intact binocular visual system that is rendered functionally
monocular due to suppression. Furthermore, we have found that a dichoptic treatment
intervention designed to directly target suppression can result in clinically significant
improvement in both binocular and monocular visual function in adult patients with
amblyopia. The fact that monocular improvement occurs in the absence of any fellow eye
occlusion suggests that amblyopia is, in part, due to chronic suppression. Previously the
treatment has been administered as a psychophysical task and more recently as a video game
that can be played on video goggles or an iPod device equipped with a lenticular screen.
The aim of this case-series study of 14 amblyopes (six strabismics, six anisometropes and two
mixed) ages 13 to 50 years was to investigate: 1. whether the portable video game treatment
is suitable for at-home use and 2. whether an anaglyphic version of the iPod-based video
game, which is more convenient for at-home use, has comparable effects to the lenticular
version.
Methods: The dichoptic video game treatment was conducted at home and visual functions
assessed before and after treatment.
Results: We found that at-home use for 10 to 30 hours restored simultaneous binocular
perception in 13 of 14 cases along with significant improvements in acuity (0.11 0.08
logMAR) and stereopsis (0.6 0.5 log units). Furthermore, the anaglyph and lenticular
platforms were equally effective. In addition, the iPod devices were able to record a complete
and accurate picture of treatment compliance.
Conclusion: The home-based dichoptic iPod approach represents a viable treatment for
adults with amblyopia.

Key words: amblyopia, childrens vision, visual acuity


Amblyopia traditionally has been thought of
as a monocular disorder that has a binocular
consequence. According to this view, the
amblyopic visual system is, in some way, lazy
or immature and the logical treatment
approach is to force use of the amblyopic eye
by occluding the fellow sighted eye with a
patch. Previously, the patch was worn all day
for months or in some cases years.1 Now
we know that less patching, even as little as
two hours per day, can be just as effective
as all-day patching and can significantly
improve visual acuity in the amblyopic eye;2
however, not all patients respond to patching and of those who do, many have residual
amblyopia after treatment is terminated
regardless of compliance.3 More importantly, binocular vision is not automatically
restored once the vision in the amblyopic

eye has been improved. In fact, more often


than not, once the patch is removed after
therapy has ended, the amblyopic eye is suppressed by the fellow sighted eye and can,
over time, lose some of the gains achieved as
a result of the therapy.4
There is now evidence to suggest that the
traditional view of amblyogenesis may be
incorrect. Amblyopia may be the consequence of a primary disruption to binocular
vision, in which suppression plays a major
part. This idea is not new,5 it is supported by
the direct relationship between suppression
and amblyopia that has been reported in
animal models,6 by the restoration of vision
in deprived animals7 as well as clinical studies
on adults8 and children9 with amblyopia.
Furthermore, it has been shown recently
that therapy aimed at promoting binocular

2014 The Authors


Clinical and Experimental Optometry 2014 Optometrists Association Australia

vision by strengthening fusion and reducing


suppression, results in improved vision in
the amblyopic eye as well as a recovery of
binocular function and stereopsis.1012 This
treatment was based on psychophysical
measurements, which demonstrated that
patients with amblyopia exhibited binocular
visual function if the image shown to the
amblyopic eye had a higher contrast than
that shown to the fellow eye.13 The treatment incorporates a task that requires information to be combined between the two
eyes and begins with a patient-specific
interocular contrast offset that overcomes
suppression and allows for the task to be
completed. Over time, binocular function
improves and the contrast offset between the
two eyes can be reduced until, in many cases,
no contrast offset is required. Using this

Clinical and Experimental Optometry 97.5 September 2014

389

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

approach, it has been demonstrated recently


that, as well as playing a key role in the
development of amblyopia, suppression may
also actively prevent recovery of visual function by inhibiting visual cortex plasticity.14
Although this binocular treatment was initially developed in the laboratory using cumbersome psychophysical equipment, it has
been translated recently to a more convenient head-mounted display1417 and a handheld iPod device.12,18 These stimulus display
platforms can be used in a clinical setting
and have the potential to be used in the
home. We have also developed a video game
version of the treatment to make it as enjoyable as possible with the goal of improving
compliance and hence treatment outcomes.
The combination of the iPod platform and
the video game version of the treatment is
particularly suitable for use at home. This
would meet with the expectations of clinicians and patients, who are used to amblyopia treatment being administered in the
home rather than the clinic.
The use of our binocular treatment
outside of the clinic setting poses a number
of challenges. These include compliance,
automatic updating of the interocular
contrast difference as binocular function
improves and accurate dichoptic presentation of visual stimuli. In the clinic, compliance problems do not occur as patients
are monitored and interocular contrast can
be adjusted manually, based on session by
session evaluation of task performance. In
addition, dichoptic presentation can be
achieved on the iPod device using a lenticular overlay screen. The advantage of using a
lenticular overlay is that the luminance contrast is preserved. The disadvantage is that
the device must be precisely aligned with the
eyes to reduce crosstalk between the images
presented to each eye. In the clinic, this can
be achieved using a chinrest and a stand
for the iPod; however, this cannot be done
easily in the home. In the present study, we
assessed the practicality of at-home use of
the iPod video game treatment. Patients
were required to comply with a treatment
schedule, the interocular contrast difference was adjusted automatically based on
game performance and patients were
responsible for assuring correct alignment
of the iPod device equipped with a lenticular
screen. We also developed an anaglyph
method of dichoptic stimulation to complement our original lenticular approach.
This has the advantage that head alignment
is no longer necessary; however, red/green

glasses have to be worn. We hoped that this


approach may be better suited to younger
patients. We set out to answer two questions.
1. Is the home-based binocular treatment as
effective as its clinic-based counterpart18
that was conducted under supervision?
2. Is the anaglyphic version as effective as
the lenticular version?
METHODS
Concurrent pilot field tests were run at
the Department of Ophthalmology McGill
University, School of Optometry and Vision
Science at University of Waterloo and the
Department of Optometry and Vision
Science, University of Auckland in New
Zealand. The research was carried out
following clearance from the Institutional
Review boards of each university and
adhered to the tenets of the Declaration of
Helsinki.
Participants (n = 14; aged 13 to 50 years)
who had amblyopia due to anisometropia
(difference in refractive error between the
two eyes), strabismus (misalignment of the
eyes) or both were recruited at the individual testing facilities. The amblyopic participants had a difference of at least two lines
between the eyes on a logMAR visual acuity
chart and had impaired stereo acuity
(greater than 40 arc seconds).
All the participants underwent a standard
clinical protocol in all the three pilot testing
sites. The clinical examination is detailed
below and clinical details provided in
Table 1.

Visual acuity
Visual acuity was obtained using a computerised version of the BaileyLovie logMAR
chart; either the Test Chart 2000 pro and
Khyber Vision iPad application or the
Medmont computerised visual acuity testing
system, model AT20R, (Melbourne, Victoria, Australia). These two tests do not differ
from standard chart-based tests when
correct lighting conditions are employed.19
A letter by letter scoring procedure was
adopted to obtain visual acuity. A termination criterion of five errors on a line was
used.

Stereoacuity
Stereoacuity was measured using the Randot
stereofly test or the Randot Preschool
Stereo Acuity test (Stereo Optical Company,
Chicago, Illinois, USA). Results were

Clinical and Experimental Optometry 97.5 September 2014

390

recorded as log threshold. If stereopsis was


unmeasurable, a log threshold of four was
recorded.

Strabismus
Unilateral and alternate cover tests were
used to determine the presence of a tropia
(manifest deviation) or phoria (latent deviation) and the observed deviation, if any,
was neutralised by the use of a prism of the
required magnitude (prism cover test).
These were worn only during the treatment.
Amblyopic participants were classified as
exotropes or esotropes based on the direction of the deviation.

Worth four dot test


The test was performed at both distance
(1.6 m) and near (33 cm). The distance
measurement was such that the lights subtended one degree of visual angle and the
near measurement was such that a six degree
visual angle was subtended. The placement
of the red-green filters was according to
convention: red filter over the right eye and
green filter over the left eye. The participants were to report whether they saw all the
four lights and report the colour of each. If
the participants reported either only two
reds or three green coloured lights, they
were considered to have complete suppression. If they reported a total of five coloured
light, they were considered to have diplopia
and if they reported four lights with the
bottom light appearing either red or green
then they were considered to have partial
suppression.

Bagolini striated lens test


Suppression by the participants was also
assessed qualitatively using the Bagolini striated lens test, which is considered to have a
less dissociating effect than the Worth four
dot test. Participants viewed a light source at
two test distances (distance 1.0 m and near
33 cm), while wearing the striated glass
lenses oriented at 45 and 135 degrees over
their habitual spectacle or contact lens correction. In the instance of normal binocular
vision, the participants would report an X
corresponding to / seen by one eye and \
seen by the other eye. In the case of complete
suppression, the participants would perceive
only one of the two lines that form an X. In
the case of a central suppression scotoma,
the percept would be of a cross with one line
having a missing region close to the fixation
light. In theory the size of the suppression
2014 The Authors

Clinical and Experimental Optometry 2014 Optometrists Association Australia

2014 The Authors

Clinical and Experimental Optometry 2014 Optometrists Association Australia

13/M

EL

Anisometropic

Anisometrope

Anisometropic

Anisometrope

Mixed
left esotropia 8

Strabismus:
left esotropia 4 PD

Strabismus: Right esotropia 10 PD

Strabismus: Right exotropia 8 PD

Strabismus: Right esotropia 6 PD

No squint
Considered as strabismus on account
of history (30 PD corrected at age 5)

Mixed
Left esotropia 6 PD

Microtrope (4 prism BO test)

Anisometrope

Anisometrope

Type

RE: +1.25/-0.25 x 10
LE: +6.00/-2.00 x 10

RE: +1.00/-0.25 x 105


LE: +3.50 DS

RE: +1.25 DS/-0.25 x 175


LE: +5.00 DS/-0.50 x 180

RE: + 0.25/-0.25 x 175


LE: +1.75/-2.25 x 12

RE: plano
LE: -4.00 /-1.75 x 40

RE: -1.00 DS
LE: -1.00/-0.25 x 160

RE: +6.75/-2.50 x 30
LE: +5.00/-1.75 x 162

RE: -2.75/-0.75 x 25
LE: -3.25/-1.25 x 10

RE: -2.75/-1.00 x 105


LE: -2.75/-1.00 x 80

RE: +1.50/-0.50 X 12
LE: +3.50/-2.00 x 160

RE: -2.50/ -1.25 x 180


LE: + 0.50/-1.50 x 180.

RE: +1.75 DS
LE: +1.75/-1.00 x 165

RE: -1.5 DS
LE: +3.00/-1.5 x 145

RE: +3.00 D/-0.50 x 90


LE: +1.00 DS

Refraction

Stereo (RDS) and suppression


800 arc secs (RDS)
W4D: Partial suppression
Bagolini: Fusion
DOTS: 34.4 FFE contrast, AME/FFE = 3.59
400 arc sec
W4D: Fusion at near, suppression at distance
Bagolini: Fusion with central suppression
DOTS: 12.6 FFE contrast
AME/FFE = 5.95
Stereo: < 800arc sec
W4D: Fusion with partial suppression
Bagolini: Fusion with occasional suppression
DOTS: 31 FFE contrast
AME/FFE = 2.17
Stereo: < 800arc sec
W4D: Fusion (distance and near)
Bagolini: Fusion
DOTS: 28.6 FFE contrast
AME/FFE = 3.14
Stereo: < 800arc sec
W4D: Diplopia (distance)
Fusion with intermittent suppression (near)
Bagolini: central scotoma left eye (distance and near)
DOTS: 31.2 FFE contrast
AME/FFE = 5.46
Stereo: < 800arc sec
W4D: Intermittent suppression (distance)
Fusion (near)
Bagolini: RE central scotoma (distance and near)
DOTS: 38.8 FFE contrast
AME/FFE = 4.42
Stereo: 400 arc sec
W4D: Fusion (distance and near)
Bagolini: Intermittent suppression
DOTS: 15.1 FFE contrast
AME/FFE = 8.51
Stereo: < 800arc sec
W4D: Suppression RE (distance); fusion (near)
Bagolini: RE central suppression (distance and near)
DOTS: 38.4 FFE contrast
AME/FFE = 2.57
400 arc sec
W4D: Fusion
Bagolini: central scotoma at distance; fusion (near)
DOTS: 38.8 FFE contrast
AME/FFE = 2.57
Stereo: 400 arc sec
W4D: Intermittent suppression (distance)
Fusion (near)
Bagolini: LE central scotoma
Stereo: 400 arc sec
W4D: Intermittent suppression (distance)
Fusion (near)
Bagolini: Fusion
No stereo
W4D: left eye suppression at both distance and near
Bagolini: LE central suppression
Stereo: 200 arc sec
W4D: Fusion
Bagolini: Fusion
Stereo: 63 arc sec
W4D: Left eye suppression distance, fusion at near
Bagolini: Intermittant suppression

[20/27+2]

RE: -0.2 [20/12.5] (6/3.8)


LE: 0.3 [20/40) (6/12)

RE: 0.0 [20/20] (6/6)


LE: 0.4 [20/50] (6/15)

RE: 0.2 [20/32] (6/9.5)


LE: 0.53 [20/63] (6/19)

RE:-0.1 [20/16] (6/4.8)


LE: 0.26 [20/32+3] (6/9.5+3)

RE: -0.1 [20/16] (6/4.8)


LE: +0.3 [20/40] (6/12)

RE: -0.1 [20/16] (6/4.8)


LE: +0.3 [20/40] (6/12)

RE: +0.3 [20/40] (6/12)


LE: -0.1 [20/16] (6/4.8-1)

RE: +0.56 [20/80-2] (6/24-2)


LE:-0.1 [20/16] (6/4.8-2)

RE: +0.34 [20/40-2] (6/12-2)


LE: -0.12 [20/16-1] (6/4.8-2)

RE: -0.26 [20/12.5-3] (6/3.8-3)


LE: +0.28 [20/40-1] (6/12-1)

RE: -0.1 [20/16] (6/4.8)


LE: +0.36 [20/40+3] (6/12+3)

RE: +0.5 [20/63] (6/18)


LE: 0.0 [20/20] (6/6)

RE: -0.1 [20/16] (6/4.8)


LE:+0.36 [20/40+3] (6/12+3)

RE:+0.14
LE: -0.1 [20/16] (6/4.8)

(6/7.5+2)

Visual acuity

Detection at age 11
No patching, optical correction only

Detected in childhood (uncertain of age)


No patching
Only wearing near correction habitually
Detected in childhood (uncertain of age)
No patching and a brief period of atropine penalisation

Detected at age 24
No patching
No surgery

Detected at age 27
No patching
No surgery

Detected 12+ years


No patching

Detected at age 4
Patching for 1 year for 8 hours/day. History of strabismus
surgery at age 4 for both eyes.

Detected at 8 years
Patching for 6 hours or more for 1 year.

Detected at age 6
Patching for more than 8 hours a day for more than a year.

History of strabismus surgery at age 5 (left esotropia 30 PD)


Patching for 1 year
2-3 hours/day

Detected at age 5. History of strabismus surgery at age 5.


Patching 1 hour/day for 6 months.

Detected at age 5, Patchingdid not comply.

Detected at age 5.
Patching 2 to 3 hours/day for 1 year.

Detected at age 12 years


No patching

History

Table 1. Clinical details of amblyopic observers participating in the iPod training study

RE: right eye, LE: left eye, RDS: Randotdot stereogram, W4D: Worth 4 Dot test, F: female, M: male, AME: Amblyopic eye, FFE: fellow fixing eye, BO: base out, DS: dioptre sphere, DC: dioptre cylinder, PD: prism dioptres, L: lenticular iPod training used.

32/F

22/M

OT

YZ

39/M

XU

49/F

24/F

ST

NDA

40/M

MT

24/M

28/M

PS

DD

29/M

SA
(L)

46/M

50/M

SJ
(L)

AS

41/F

22/M

Age/sex

MS
(L)

(L)

SB

Observer

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

Clinical and Experimental Optometry 97.5 September 2014

391

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

moved at six degrees per second. The dots


had a limited lifetime whereby on any single
frame, each dot had a five per cent chance of
disappearing and being redrawn in a new
spatial position. To avoid interaction of the
stimulus dots with the central dark fixation
dot (radius 0.35 degrees), stimulus dots
did not enter the central region of the
display aperture (radius 2 degrees). 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 one second.

Refraction
Figure 1. The anaglyphic version of the iPod-based Tetris game. The high-contrast red
blocks were seen by the amblyopic eye. These were the falling blocks. The low-contrast
green blocks were seen by the fellow fixing eye (FFE). These were the superficial ground
plane blocks relevant to the task. Some ground plane blocks were seen by both eyes
(brown/orange). Over time and successful play, the contrast offset between the eyes was
reduced (the fixing eye contrast was increased by 10 per cent of its starting value every 24
hours). We identified two phases of fusional recovery (Figures 7A and B); phase 1 where
the contrast is automatically incrementing in the fixing eye with successful game play and
phase 2 where the contrast in the FFE has reached an asymptote (usually 100 per cent),
which is the same as that of the fellow amblyopic eye.

scotoma determines whether this percept is


seen at both distance and near. In particular,
if the scotoma is seen only at distance then
a smaller scotoma (approximately one
degree) is assumed.

Objective quantification
of suppression
We quantified the amount of suppression
using the dichtopic global motion test.13,15,17
This test involves the presentation of signal
elements to one eye, noise elements to the
other eye and a variable interocular contrast
offset. Suppression is measured by identifying the contrast offset between the two eyes
that is required for normal binocular combination of the signal and noise elements,
whereby lower contrast elements are shown
to the fellow eye. Following previously published protocols, stimuli were presented
using a MacBook Pro laptop computer
running Matlab (Mathworks Ltd, Cambridge UK) and Psychophysics Toolbox,
Version 3.20 The stimuli were displayed using
a Z800 duel pro headmounted display
(eMagin Corporation, New York, NY, USA).
This headmounted display model contains

two OLED screens, one for each eye. The


screens have a high luminance, a linear luminance response profile and refresh simultaneously 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 by 1600 pixels. A
Matrox Duel Head2Go external video board
was then used to split each frame between
the two headmounted display screens at a
resolution of 600 by 800 pixels per screen. A
photometer (United Detector Technology,
San Diego) was used to ensure equal luminance of the two screens and to confirm a
linear luminance response.
Stimuli were random dot kinematograms
based on those used by Mansouri, Thompson and Hess13 and were presented within a
stimulus aperture with a diameter of 22
degrees. One hundred dots (with dot
luminance modulation varied according
(Ldots Lbackground )
to
) were displayed upon a
(Ldots + Lbackground )
mean luminance background of 35 cd/m2.
Each dot had a radius of 0.5 degrees and

Clinical and Experimental Optometry 97.5 September 2014

392

Refraction was performed, if participants


were not habitually wearing any correction
or if they had not visited their eye-care practitioner for more than two years. If a new
prescription was required participants completed a refractive adaptation period that
ended when two consecutive visual acuity
measurements made a minimum of four
weeks apart indicated stable visual acuity
(less than 0.1 logMAR difference in the two
measurements). Participants (strabismic as
well as anisometropic) were asked to wear
their correction full-time during the refractive adaptation period.

Training regimen
The training was completed using an iPod
touch device using the popular Tetris
game.18 The advantages of using the Tetris
game were that most players have played
the game before, it is a very simple game to
learn and the game configuration lends
itself to our dichoptic treatment principle, as
it includes multiple distributed elements.
The players have to align various falling
elementary shapes that appear randomly on
the top of the screen. Players have to interact
continuously with the falling blocks by
changing the position and orientation of the
falling block shapes to form tessellated rows
of blocks at the bottom of the screen.
Dichoptic presentation of the blocks can be
achieved using either a lenticular screen 18
or an anaglyph presentation. Here, we
illustrate the anaglyphic version (Figure 1).
The falling blocks can only be seen by the
amblyopic eye (Figure 1, red blocks). The
blocks forming rows at the bottom of
the screen are seen only by the fellow fixing
eye (Figure 1, green blocks). The ground
2014 The Authors

Clinical and Experimental Optometry 2014 Optometrists Association Australia

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

AS
Play distribution

OT

6:00 am

6:00 am

Midday

Midday

6:00 pm

6:00 pm

Midnight

Midnight

6:00 am

10

15

20

25

6:00 am

30

10

15

20

25

30

35

30

Game performance

100

300

80
200

60
40

100

20
0

10

15

20

25

0
30

FFE Contrast (%)

Total duration (min)

B
80
200

10,000

8,000

8,000

6,000

6,000

4,000

4,000

60
40

100

10,000

2,000

100

300

Prescribed hours
0

10

10

20

15

20

25

30

35

0
30

15

20

25

30

35

30

2,000
0

10

15

20

25

Days of training

30

Days of training

Figure 2. Graphical representation of information contained in the iPods log file after
the home-based treatment for two patients (AS and OT). In row A, the distribution of
game play over a 24-hour period is shown. In row B, the total duration of game play each
day is shown (left, Y-axis) as well as how the contrast changed (right, Y-axis), as a
consequence of game performance (row C). Each data point represents an individual
game. See main text for further details.

plane blocks that are not relevant to the


score are seen by both eyes (brown/orange
blocks). At the start of training, blocks were
presented to the amblyopic eye at a higher
contrast than the blocks presented to the
fellow eye to overcome suppression and
allow for binocular combination.13,15 The
contrast offset was determined separately for
each participant, based on the results of the
suppression measurements made using the
dichtopic global motion test.
RESULTS
The treatment is based on the finding that if
the contrast is reduced in the fellow sighted
eye, depending on the severity of suppression, there will be a value for which the
information is combined by the fellow
sighted and amblyopic eyes.1012 Over time
this reduced contrast can be slowly increased
while binocular combination is maintained.
We use a video game,18 in which different
elements are seen by fellow sighted and
amblyopic eyes and the combination of
these elements is essential to score in the
game, hence it can only be done binocularly.
The elements seen by the fellow sighted eye
are reduced in contrast until the game can
be successfully played (when a player goes

up one level by clearing four lines, the game


gets faster and contrast is incremented but
just once in a 24 hour period) and gradually
increased each and every day that the game
is successfully played. The contrast of elements seen by the fellow sighted eye is automatically increased, if the game is played
successfully and this is an indication that
suppression from the fellow sighted eye is
reducing. Once the game can be successfully
played with the same contrast in both eyes,
suppression has been eliminated and binocular vision in its most rudimentary form,
(fusion) has been re-instated. The game is
played on an iPod device and information
about exactly how the contrast is changed,
how frequently the game is played, when
during the day it was played and how successfully it was played is contained in the iPods
stored log files.
Figure 2 shows two examples of information derived from iPod log files for a 30 to 40
day period of home-based therapy. The
middle graph (B) in each case reflects how
the contrast of stimuli presented to the
fellow fixing eye (FFE) changed as a function of duration of video game play (grey
circles). The contrast of the blocks presented to the amblyopic eye was fixed at 100
per cent. At the start of training, low contrast

2014 The Authors


Clinical and Experimental Optometry 2014 Optometrists Association Australia

stimuli had to be presented to the fellow eye


to allow for binocular combination (verified
by successful game play). As training progressed, higher contrasts could be tolerated
in the fellow eye until, by the end of training, no interocular contrast difference was
needed. This progressive change in the
interocular contrast required for binocular
combination indicates a weakening of the
suppressive influence of the fellow eye over
the amblyopic eye. Once the fellow eye can
be given stimuli of the same contrast as that
seen by the amblyopic eye, suppression has
been eliminated. Typically, this is what
happens over the treatment period. The rate
at which contrast changes is determined by
the algorithm we used (10 per cent change
per 24 hours, if the game is played successfully) and also by individual variation,
depending on the severity of suppression.
The top graph (A) in Figure 2 shows how the
treatment was distributed over a 24-hour
period. In these two cases, the compliance
for patient AS is excellent but initially (up till
day 15) patient OT exhibits poor compliance. The rate of improvement in the contrast tolerated by the fixing eye (grey circles
in B) reflects this; in OTs case the contrast
improvement is delayed by 15 days. The
bottom graph (C) shows how the game
performance varied across the treatment
period. Game performance needs to be consistent and above a threshold level for the
contrast to be automatically changed every
24 hours. In these two cases, game performance meets these criteria, although for OT
game performance is poor in the first 15 days
most likely due to inadequate playtime.
Figure 3 shows that the contrast tolerated
by the fellow sighted eye at the end of the
treatment period (the contrast asymptote)
increased in all patients and that this was as
true for both anaglyphic (unfilled symbols)
and lenticular (filled symbols) platforms.
The averaged pre-treatment contrast was 29
per cent and the average post-treatment contrast was 97 per cent (indicated by the solid
grey diamond in Figure 3). Contrast has to
be reduced for the fellow eye because of
suppression and therefore, an increase in
the contrast tolerated by the fellow eye
indicates the extent to which dichoptic
game play was successful in reducing suppression and re-instating binocular combination in its simplest form. The form of
this increased tolerance to the contrast of
stimuli shown to the fellow eye is shown in
the data depicted in Figure 2 (grey dotted
curve in B). The results in Figure 3 show that

Clinical and Experimental Optometry 97.5 September 2014

393

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

NDA

EL

6:00 am

6:00 am

Midday

Midday

6:00 pm

6:00 pm

Midnight

Midnight
6:00 am

80
60

Total duration (min)

40
20
0

10

15

20

25

30

35

40

10

15

20

25

30

150
100

100
80

100

60
40

50

20
0

10

15

20

25

30

35

0
40

Prescribed hours

8,000

6,000

6,000

4,000

4,000

2,000

2,000
10

15

20

25

30

35

40

Days of training

40
20

8,000

60

50

10,000

80

100

10,000

0
20
40
60
80
100
Post-treatment fellow eye contrast (%)

150

Game performance

Pre-treatment fellow eye contrast (%)

6:00 am

100

FFE contrast (%)

Play distribution

10

15

20

25

0
30

10

15

20

25

30

Days of training

Figure 3. Comparison of the contrast that


the fellow eye could tolerate while still
maintaining dichoptic game play before
and after the iPod home-based treatment.
Results falling on the unity line indicate no
change, results falling below the unity line
indicate an increase in contrast which
signifies a decrease in suppression. Results
obtained for the lenticular (filled symbols)
screen and the anaglyphic (unfilled
symbols) screen are displayed separately.
The average pre- and post-treatment contrast is indicated by the solid grey diamond
with its associated 95 per cent confidence
intervals.

Figure 4. Graphical representation of information contained in the iPods log files after
the home-based treatment for two patients, whose contrast results in Figure 3 represent
outliers (NDA and EL). In A, the distribution of game play over a 24-hour period is shown.
In B, the total duration played each day is shown (left, Y-axis) as well as how the contrast
of elements presented to the fellow eye changed (right, Y-axis) as a consequence of
performance (C). Each data point represents an individual game.

all but two patients had their suppression


eliminated and their binocular combination
re-instated.

reset to 70 per cent during a clinical visit and


then, through successful game play, gradually, manually increased again to 90 per cent
but manually reset to 82 per cent because of
continuing asthenopia. This subject could
achieve 100 per cent visually but it was associated with discomfort. On the basis of
the log file alone, this patient benefited
from training in that suppression has been
reduced to a very low level. Stereoscopic
vision improved from no stereopsis pretreatment to coarse stereopsis (800 arc
seconds) post-treatment. Acuity improved
from 0.53 logMAR pre-treatment to 0.36
logMAR post-treatment. Furthermore, as
NDA wore a habitual near correction for
playing the iPod device, near visual acuity
was also measured pre- and post-training

The two outliers


The log files for these two patients whose
contrast did not reach 100 per cent are displayed in Figure 4.
Patient NDAs log file shows an initial
strong increase in contrast reaching 100 per
cent (B), associated with good game performance (C) and excellent compliance (A). At
the early stage, when this patient was treated,
the automatic contrast adjustment was not in
operation and the contrast was increased
manually on daily clinic visits, if the performance was good. On day 15, NDA complained of asthenopia and the contrast was

Clinical and Experimental Optometry 97.5 September 2014

394

for this patient using the BaileyLovie near


word chart. Near visual acuity improved
from N15 to N5 for the amblyopic eye, while
near visual acuity remained stable at N3 for
the fellow eye. Patient ELs log file shows that
there was a great degree of variability in
performance (game scores) oscillating from
successful play to failure (C). One possible
reason for this is seen in the results in B; the
game was never played for very long (B), well
below that prescribed (horizontal dotted
line in the top graph). EL appears to peak at
55 per cent contrast at a number of points
during the training period (days 15 to 18, 28
to 29) suggesting that, for the limited time
the game was played, the anti-suppression
therapy was unable to reduce the depth of
his suppression beyond this point in the
2014 The Authors

Clinical and Experimental Optometry 2014 Optometrists Association Australia

0.7

3.5

0.6

Pre-treatment VA (logMAR)

four-week period. This patient had good


stereopsis to begin with (63 arc seconds) and
after treatment, it improved marginally to 40
arc seconds. Acuity improved from 0.44
logMAR, pre-treatment to only 0.32 logMAR
after treatment. Even though EL was the
youngest patient, the age per se probably was
not the important factor, as it has been
shown that if compliance is good, significant
gains in contrast can be achieved in the
paediatric population.21,22
The improvements in stereopsis resulting
from the home-based treatment are shown
in Figure 5, where stereoscopic performance before and after treatment is plotted
such that results falling below the diagonal
line indicate decrements in performance,
those on the diagonal line, no change in
performance and those above the diagonal
line, improvements in performance. The
mean values for pre- and post-training across
all participants are indicated by the solid
black symbol with associated 95 per cent
confidence intervals. Results for lenticular
(filled symbols) and anaglyphic (unfilled
symbols) platforms are displayed separately.
On average, stereoacuity improved by 0.61
log units in the present study with a mean
before treatment of 3.14 (1388 seconds) and
after treatment of 2.54 (344 seconds), t(13)
= 5.0, p < 0.001. In some cases, stereopsis was
not measurable before treatment and was
re-established as a result of treatment to fine
or coarse levels. The unfilled diamond represents the corresponding pre- and posttreatment average for the same treatment
principle (including both Tetris and global
motion stimuli) supervised within the clinic
(Figure 3). The improvement in stereoscopic acuity from at-home treatment was
slightly less than that found previously for
the in-clinic treatment. There were no
obvious differences in outcome for the lenticular (filled symbols) and anaglyphic
(unfilled symbols) platforms for the at-home
treatment.
The visual acuity of the amblyopic eye
before and after at-home treatment is presented in Figure 6. Results falling above the
diagonal unity line indicate improved
amblyopic eye acuity. The average pre- and
post-treatment acuity is indicated by the
solid grey diamond with its associated 95
per cent confidence intervals. The unfilled
diamond represents the average pre- and
post-treatment acuity of the same treatment
supervised within the clinic. On average, the
change in logMAR acuity was significant
(mean pre-treatment = 0.36, post-treatment

Pre-treatment stereopsis
(log threshold)

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

3
2.5
2
1.5
1
1

1.5
2
2.5
3
3.5
Post-treatment stereopsis
(log threshold)

0.5
0.4
0.3
0.2
0.1
0
0

0.2
0.4
0.6
Post-treatment VA (logMAR)

Figure 5. Comparison of stereoacuity


before and after the iPod home-based
treatment. Results falling on the sloping
diagonal line indicate no change, results
falling above the unity line indicate an
improvement. Results are displayed separately for the lenticular screen (filled
symbols) and the anaglyphic (unfilled
symbols) screen. The average pre- and
post-treatment stereopsis is indicated by
the solid grey diamond with its associated
95 per cent confidence intervals. The
unfilled diamond refers to the average preand post-treatment acuity of the same treatment supervised within the clinic.23 Three
patients (two lenticular, one anaglyphic)
had no measurable stereopsis before and
after treatment (the data points overlap in
the top right of the figure). A number of
data points overlap due to the categorical
nature of the stereotest. Two participants
improved from no measurable stereopsis
to 800 arc seconds (4.0 to 2.9 log units),
three improved from 400 to 100 arc
seconds (2.6 to 2.0 log units) and two
improved from 400 to 40 arc seconds (2.6
to 1.6 log units).

Figure 6. Comparison of visual acuity


(logMAR) before and after the iPod homebased treatment. Results falling on the
diagonal unity line indicate no change,
results falling above the unity line indicate
an improvement. Results obtained for the
lenticular (filled symbols) screen and
anaglyphic (unfilled symbols) screens are
displayed separately. The average pre- and
post-treatment acuity is indicated by the
solid grey diamond with its associated 95
per cent confidence intervals. The unfilled
diamond refers to the average pre- and
post-treatment acuity of the same treatment supervised within the clinic.23

0.25, t[13] = 5.2, p < 0.001), with no obvious


difference for lenticular (filled symbols) and
anaglyphic (unfilled symbols) platforms.
Two patients achieved visual acuity improvements of 0.2 logMAR or better (patients AS
and PS). The mean improvement was less
than that found previously for the in-clinic
treatment. Note that the severity range of
this at-home sample (Figure 6) was much
less than that of our previous in-clinic
sample, (0.6 logMAR compared with 1.2
logMAR).2

The comparisons with previous work


described above include results achieved
using a range of different types of displays.
To test for any differences in treatment
outcome for the iPod device, when used
at home versus in the clinic, we compared
the results of this study to those reported
by Hess and colleagues,23 who treated
patients in the clinic using the lenticular
version of the iPod. There was no significant
difference between the two datasets. The
average improvement in stereopsis across a

2014 The Authors


Clinical and Experimental Optometry 2014 Optometrists Association Australia

Clinical and Experimental Optometry 97.5 September 2014

395

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

C
70

Play duration (min)

100
80
60

20
0

-60

-30

Phase 1
Phase 2
Days of training

40
30
20
10

Phase 1

Phase 2

D
150

100
80
60
40
20
0

50

30

Play duration (min)

FFE contrast (%)

40

Prescribed hours
60

-60

-30

100
Prescribed hours
50

30

Phase 1 Phase 2
Days of game play

10

15

20

25

30

35

40

Phase 1 duration (days)

Figure 7. Compliance data for the at-home iPod study. In A, contrast improvements as a
function of consecutive days during the treatment period with all the functions aligned to
the time corresponding to 100 per cent contrast. In B, same as in A except plotted as a
function of the days of consecutive game play (removal of days where the game was not
played). In C, the averaged compliance in terms of play duration in minutes for phase 1
and phase 2 (see A and B). The unfilled data points correspond to the individual results
of the two outliers discussed in Figure 4 (large symbolADL; smaller symbolEL). In D,
individual compliance data show the range of individual variation. The filled symbols are
for the lenticular platform and the open symbols for the anaglyphic platform.

number of studies using our previous clinicbased approach23 was 0.78 0.74 log units of
seconds of arc, which was not significantly
different from the 0.6 0.5 log units
improvement found in the current study (p
= 0.36). Similarly, for monocular acuity, the
previous clinic-based protocol23 had resulted
in improvements of 0.19 0.17 logMAR,
which was not statistically different from the
0.11 0.08 logMAR improvement found in
the present study (p = 0.67). The contrast
improvements were also similar between
studies, with six of 10 (60 per cent) reaching
100 per cent in the fellow eye as compared
with 12 of 14 (86 per cent) in the present
study. Comparison of the lenticular and
anaglyphic platforms in the present study
also indicated no significant differences for
acuity (0.11 0.05 logMAR versus 0.10 0.09
logMAR), stereopsis (0.56 0.50 log units
versus 0.56 0.45 log units) or contrast (75
per cent success versus 90 per cent success),
respectively. These findings suggest that
home-based outcomes are as good as previ-

ously reported clinic-based outcomes23 and


that the anaglyphic platform was just as effective as the previous lenticular platform.
An analysis of the iPod log files also
allowed at-home treatment compliance to
be assessed. Participants were asked to play
for one hour per day for periods of time that
ranged between 22 and 108 days. Figure 7
shows a summary of the log file data for all
participants who reached 100 per cent contrast in their fixing eye as a result of treatment (13 of 14). In A, contrast improvement
as a function of the number of consecutive
days within the training period is shown.
With the exception of the initial plateau in
the results for patient SJ (due to insufficient
play time), all the subjects show a similar
change in contrast, suggesting an effective
treatment duration of 30 days (phase 1). The
results shown in Figure 7B are for consecutive days in which the game is actually played
(with the days in which there was no play
removed). The results are similar in A and B
apart from a delay that is evident in A, sug-

Clinical and Experimental Optometry 97.5 September 2014

396

gesting that consecutive training days are


not a requirement, if days are missed the
only consequence is the need for a longer
treatment duration. In C, the compliance is
shown in terms of play duration/day in
minutes. The horizontal dotted line is the
prescribed one hour per day and the averaged results in terms of phase 1 training
(Figure 7A and B) are not statistically
different from that. The two data points
superimposed on this bar figure represent
the individual results for the two outliers
(NDAsmall symbol and ELlarger
symbol) previously described in Figure 4.
Note that once the contrast asymptote is
obtained (phase 2), play time significantly
decreases. The individual results for compliance are seen in D and there is considerable
variability with nine of 14 achieving close
to the expected levels or above expected
levels and four of 14 achieving lower than
expected levels (less than 50 minutes
per day).
DISCUSSION
We set out to answer two questions.
1. Is the home-based binocular treatment
as effective as the supervised clinic-based
treatment?23
2. Is the anaglyphic platform for dichoptic
stimulation as successful as the previous
lenticular platform? The findings of this
study suggest that home-based outcomes
are as good as previously reported clinicbased outcomes23 and that the anaglyphic
platform is just as effective as the previous
lenticular platform.
There are two important differences
between treatment that is supervised in the
clinic compared with that done at home.
First, the viewing conditions, in particular
the alignment for the lenticular screen,
may not be optimal during treatment and
second, the degree of compliance may be
reduced at home. Both of these may lead to
poorer outcomes for any home-based treatment. Two findings argue that the alignment
of the lenticular display is well maintained
for the home-based treatment. First, similar
results were found between the clinic23 and
home-based protocols and we had gone to
some trouble (iPod fixed, remote key controller and chin and forehead rest) to ensure
optimal alignment (important to ensure
independent images to each eye) was maintained in our previous clinic-based protocol.23 Second, the anaglyphic version does
not require a fixed head alignment and we
2014 The Authors

Clinical and Experimental Optometry 2014 Optometrists Association Australia

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

found comparable results in the present


study between lenticular and anaglyphic
platforms.
Compliance with the treatment of amblyopia has always been an important issue24 and
we wanted to assess the degree of compliance for our home-based treatment. Information obtained from the log files similar
to that illustrated in Figure 2 revealed the
following information. In phase 1 of the
contrast recovery (Figure 7A and B) there
was no statistical difference between treatment duration prescribed and achieved
(Figure 7C). In phase 2 of the contrast recovery (Figure 7A and B), compliance was
reduced for the group as a whole to about 60
per cent (Figure 7C). There is a considerable degree of variability in compliance
within the group with a small subset of
patients (four of 14) only playing the
game for approximately half that prescribed
(Figure 7D). We found no correlation
between the visual outcome and the way in
which the game time was distributed so long
as 30 minutes to one hour per day of game
play was achieved.
This new approach to treatment aims to
restore binocular vision as a first step, something that is often not achieved after the
conclusion of the conventional occlusion
therapy, even if the degree of amblyopia has
been reduced.4 Furthermore, we are doing
this in adults for whom there is no current
therapy. We achieved restoration of simultaneous binocular perception in 12 out of
the 14 patients studied (Figure 3). There
were no reports of diplopia consistent with
our previous studies1012,18 and those of
others.16,25 There were significant gains in
stereopsis (Figure 5) and amblyopic eye
acuity (Figure 6). We conclude that this
binocular approach, which targets reestablishing binocular function and improving visual acuity in the amblyopic eye of adult
amblyopes can be successfully implemented
using either a lenticular or anaglyphic
version. This facilitates its application to the
paediatric population as the latter approach
does not require precise head-to-iPod alignment.21,22 We also show comparable results
for at-home compared with in-clinic use,
making it a more convenient treatment
option. Finally, the associated log files
provide a complete record of compliance in
terms of not only how many hours the game
was played each day but also how this game
play was distributed throughout the day and
how successful the patient was at playing the
game. We did not find any statistical differ-

ence between the training prescribed and


that actually carried out at home for the
group as a whole and we found no strong
relationship between how the game play was
distributed throughout the day and the
visual outcome, making the treatment tolerant to individual differences in life style. For
example, many patients distributed their
game play over periods shorter than one
hour and this seems not to have affected
their treatment outcome, so long as a one
hour daily average was maintained. In some
cases, treatment was not done on consecutive days and this resulted in a delayed rather
than a reduced outcome. The choice of the
10 per cent step in the automatic contrast
adjustment was not seen to limit the speed of
recovery and we conclude it was conservatively set.

Comparison with
alternate methods
At present, there is no generally accepted
treatment for amblyopia in adults, as patching has been shown to be less effective for
patients above 13 years26 and would have
significant compliance issues. Perceptual
learning approaches have been applied and
have shown promising results that, like our
binocular treatment approach, are independent of age and type of amblyopia.27,28
Perceptual learning studies have focused
on monocular function with training conducted during periods of patching. It is also
notable that the vast majority of previously
published scientific studies in this area,
including our own, have treated participants
in the laboratory or clinic setting. The use of
a home-based approach, as described here,
is an important step forward as it not only
aligns the binocular treatment approach
with current treatments for amblyopia, such
as patching and refractive correction, which
all occur in the home, but also allows, for the
first time, remote internet monitoring of
treatment between office visits.
ACKNOWLEDGEMENTS

This work was supported by a CIHR grant to


Professor Hess (#53346) and a University of
Auckland Faculty Development Research
Fund grant and an HRC grant to Benjamin
Thompson.
The binocular treatment described is
patented by McGill University and licensed
to Amblyotech (www.amblyotech.com).
Robert F Hess and Benjamin Thompson are
named inventors.

2014 The Authors


Clinical and Experimental Optometry 2014 Optometrists Association Australia

REFERENCES
1. Loudon SE, Simonsz HJ. The history of the treatment of amblyopia. Strabismus 2005; 13: 93106.
2. Repka MX, Beck RW, Holmes JM, Birch EE,
Chandler DL, Cotter SA, Hertle RW et al. A
randomized trial of patching regimens for treatment of moderate amblyopia in children. Arch
Ophthalmol 2003; 121: 603611.
3. Repka MX, Wallace DK, Beck RW, Kraker RT,
Birch EE, Cotter SA, Donahue S et al. Two-year
follow-up of a 6-month randomized trial of
atropine vs patching for treatment of moderate
amblyopia in children. Arch Ophthalmol 2005; 123:
149157.
4. Birch EE. Amblyopia and binocular vision. Prog
Retin Eye Res 2012; 33: 6784.
5. Wiesel TN, Hubel DH. Comparison of the effects of
unilateral and bilateral eye closure on cortical unit
responses in kittens. J Neurophysiol 1965; 28: 1029
1040.
6. Bi H, Zhang B, Tao X, Harwerth RS, Smith EL 3rd,
Chino YM. Neuronal responses in visual area V2
(V2) of macaque monkeys with strabismic amblyopia. Cerebral Cortex 2011; 21: 20332045.
7. Mitchell DE, Duffy KR. The case from animal
studies for balanced binocular treatment strategies
for human amblyopia. Ophthalmic Physiol Opt 2014;
34: 129145.
8. Li J, Thompson B, Lam CSY, Deng D, Chan LY,
Machara G, Woo GC et al. The role of suppression
in amblyopia. Invest Ophthalmol Vis Sci 2011; 52:
41694176.
9. Narasimhan S, Harrison ER, Giaschi DE. Quantitative measurement of interocular suppression in
children with amblyopia. Vision Res 2012; 66: 110.
10. 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. Restorat Neurol Neurosci 2010; 28: 793
802.
11. Hess RF, Mansouri B, Thompson B. A binocular
approach to treating amblyopia: anti-suppression
therapy. Optom Vis Sci 2010; 87: 697704.
12. Hess RF, Mansouri B, Thompson B. Restoration of
binocular vision in amblyopia. Strabismus 2011; 19:
110118.
13. Mansouri B, Thompson B, Hess RF. Measurement
of suprathreshold binocular interactions in
amblyopia. Vision Res 2008; 48: 27752784.
14. Li J, Thompson B, Deng D, Chan LY, Yu M, Hess
RF. Dichoptic training enables the adult amblyopic
brain to learn. Curr Biol 2013; 23: R308R309.
15. Black J, Maehara G, Thompson B, Hess RF. A
compact clinical instrument for quantifying suppression. Optom Vis Sci 2011; 88: 334342.
16. Knox PJ, Simmers AJ, Gray LS, Cleary M. An
exploratory study: prolonged periods of binocular
stimulation can provide an effective treatment for
childhood amblyopia. Invest Ophthamol Vis Sci 2012;
53: 817824.
17. Black JM, Hess RF, Cooperstock JR, To L,
Thompson B. The measurement and treatment of
suppression in amblyopia. J Vis Exp 2012; 70: e3927.
18. To L, Thompson B, Blum J, Maehara G, Hess RF,
Cooperstock J. A game platform for treatment of
amblyopia. IEEE Trans Neural Syst Rehabil Eng 2011;
19: 280289.
19. Black JM, Jacobs RJ, Phillips G, Chen L, Tan E, Tran
A, Thompson B. An assessment of the iPad as a
testing platform for distance visual acuity in adults.
BMJ Open 2013; 3: e002730.

Clinical and Experimental Optometry 97.5 September 2014

397

iPod binocular treatment for amblyopia in adults Hess, Babu, Clavagnier, Black, Bobier and Thompson

20. Brainard DH. The psychophysics toolbox. Spat Vis


1997; 10: 433436.
21. Li S, Subramanian V, To L, Jost RM, Jost S, Stager D
Jr, Dao L et al. Binocular iPad treatment for
amblyopia. Invest Ophthalmol Vis Sci 2013; 54:
E-Abstract 4981.
22. Birch, EE. Binocular iPad treatment for amblyopia.
Child Vision Research Society 2013; http://
cvrsoc.org/docs/CVRS2013-web.pdf; E-Abstract
p. 41.
23. Hess RF, Thompson B, Black JM, Machara G,
Zhang P, Bobier WR, Cooperstock J. An iPod
treatment for amblyopia: An updated binocular
approach. Optometry 2012; 83: 8894.
24. Searle A, Norman P, Harrad R, Vedhara K. psychosocial and clinical determinants of compliance with
occlusion therapy for amblyopic children. Eye 2002;
16: 150155.
25. Mansouri B, Singh P, Globa A, Pearson P. Binocular training reduces amblyopic visual acuity impairment. Strabismus 2014; 22:16.
26. Scheiman MM, Hertle RW, Beck RW, Edwards AR,
Birch E, Cotter SA, Crouch ER et al. Randomized
trial of treatment of amblyopia in children aged 7
to 17 years. Arch Ophthalmol 2005; 123: 437447.
27. Polat U, Ma-Naim T, Belkint M, Sagi D. Improving
vision in adult amblyopia by perceptual learning.
P Natl Acad Sci USA 2004; 101: 66926697.
28. Levi DM, Li RW. Perceptual learning as a potential
treatment for amblyopia: a mini-review. Vision Res
2009; 49: 25352549.

Clinical and Experimental Optometry 97.5 September 2014

398

2014 The Authors


Clinical and Experimental Optometry 2014 Optometrists Association Australia

You might also like