Visual Development &
Amblyopia
Adler’s Physiology of the Eye 10th Ed.
Chapter 21- Development of Vision in Infancy
Chapter 27 - Activity-Dependent Development of Retinogeniculate Projections
Chapter 31 - Visual Deprivation
Human Amblyopia - Some current issues
Visual Development:
Hierarchical Model of Vision
Visual Development:
Development of Contrast Sensitivity
Peak temporal frequency
(low spatial frequency)
VEP
DEM
FPL
Visual Development:
Development of Contrast Sensitivity
Peak spatial frequency
(low temporal frequency)
Sweep VEP
Grating Acuity
Visual Development:
Temporal Acuity Precedes Spatial
VEP
Temporal
Spatial
Adult
Adult
4 years
psychophysically
6 years
psychophysically
Visual Development:
Response Latency Shows Rapid Change
VEP
125 msec difference at 5 mo
50 msec difference in adults
Visual Development:
OKN Asymmetry
Nasal precedes temporal
Improves rapidly over 6 mo
DEM
VEP
Visual Development:
Vernier Acuity
FPL
FPL
Sweep VEP (filled)
Visual Development:
Binocular Vision
FPL (open)
VEP (solid)
Visual Development:
Binocular Vision
FPL
Global stereopsis
emerges at 3-5 mo
Global stereopsis improves
8 fold in first year
Protracted development
of adult values
Facts and Figures
Brain Weight:
Doubles in 9 mo/90% by 6 yr
Cortical Thickness:
Neuronal Density:
V1 6 mo/parietal 12 yr/temporal 16 yr
V1 5 mo/frontal 7 yr
Synaptic Density:
V1 peaks 4 mo then declines to 11yr
frontal peaks 1 yr then declines to 16yr
Cortical Metabolism: Peaks 4 yr then declines to 15yr
White Matter:
Peaks 2 yr and continues to 30 yrs
Regionally Specific and Non-Linear
Gross Cortical Development
lissencephalic
Regionally Specific Growth
loss
Ages 5-11
gain
Sowell ER, Thompson PM, Leonard CM, Welcome SE, Kan E, Toga AW.
Longitudinal mapping of cortical thickness and brain growth in normal children.
J Neurosci. 2004 Sep 22;24(38):8223-31.
Visual Behaviors Follow
Distinct Time Courses
Critical periods
Visual Cortex Development:
Multiple Stages
Light
Light
First
Binocular
Stage
Visual Cortex Development:
Retinal Waves
Serve to fine tune local specificy
For eye of origin, retinotopy, on/off
Visual Cortex Development:
Retinogeniculate
Prenatal, uses
Spontaneous activity
Visual Cortex Development:
Geniculocortical
Postnatal,
experience
dependent
Visual Cortex Development:
Ocular Dominance
Layer 4c
Visual Cortex Development:
Ocular Dominance Columns
In normal development
each eye acquires an
equal amount of territory
Visual Cortex Development:
Postnatal Development of ODC
Visual Cortex Development:
Competitive Model
Competition, with
‘ a little help from your friends’
Visual Cortex Development:
Competitive Model
Normal Development
Monoc. Deprivation
present at birth
X
X
X
X
Layer 4c
Normally, it is useful to be able to fine tune eye alignment after birth
X
Visual Cortex Development:
Three-Eyed Frog Tectum
Columns seem to be a general
consequence of competition
for connections
Visual Cortex Development:
Spontaneous Activity
Correlated neural activity is important
Visual Cortex Development:
Cooperative Model
Hebb’s Rule
‘winner-take-all’
cooperation between
similar inputs in a
positive feedback cycle
Visual Cortex Development:
Mechanism for Cooperation/Competition
Neurotransmitter
Postsynaptic target cell
Neural growth factor
Developmental Plasticity:
Monocular Deprivation
* Retina and LGN quite normal
* Actually more severe than binocular deprivation
* Minimal effect if done to adults
Developmental Plasticity:
Experimental Strabismus
ODC sharper
than normal
No binocular
integration
Developmental Plasticity:
Cytochrome Oxidase
Weak Fixation Preference
Strong Fixation Preference
Developmental Plasticity:
Summary for Review
This is for
layer 4c
Human Amblyopia
• “Lazy Eye”
• Relatively common developmental visual disorder
• Reduced visual acuity in an otherwise healthy and properly corrected eye
• Associated with interruption of normal early visual experience
• Affects at least 2% of North American population
• Most common cause of vision loss in children
• Well characterized behaviorally, not neurologically
• Treated by patching in children
• Reduced visual acuity - defining feature
– Usually 20/30 - 20/60
–
• Impaired contrast sensitivity
– Prominent at high
– spatial frequencies
– Central visual field is generally most affected
Contrast Sensitivity
Visual Deficits in Amblyopia
Spatial Frequency
• Moderate deficits in object segmentation/recognition and spatial
localization
• Severe deficits in binocular interactions
Subtypes of Amblyopia
• Anisometropic
– Unequal refractive error between the
two eyes
• Strabismic
– Deviated eye that may or may not
have unbalanced refraction
• Deprivation
– Congenital cataract; corneal opacity;
eyelid masses
Mechanisms of Amblyopia
1. Form deprivation

Sharp image is not formed at the retina
2. Abnormal binocular vision

Binocularity is often changed or lost in amblyopia
Suppression may be
necessary to avoid
‘double vision’
Models of Amblyopia
• Competition hypothesis originated with experiments in
kittens in the 1960s by Hubel and Wiesel
• Monocular deprivation of retinal input during ‘critical’
developmental periods leads to striking abnormalities in
the physiology of visual cortical neurons
• Binocular deprivation actually leads to less severe
abnormalities
• Amblyopia may be a form of activity-dependent
deprivation, modulated by competitive interactions
Site of abnormality
Primary visual cortex and beyond
•Loss of disparity sensitivity and
binocular suppression in V1 (primary
visual cortex)
•Although loss in V1 can’t explain the
full abnormality - extrastriate is
implicated.
•Barnes et al. showed with fMRI
abnormalities in many visual areas
beyond V1. Hypothesized that
feedback connections from extrastriate
to V1 may be a primary source of
abnormality.
Current Issues
• Abiding debate about how the strabismic and
anisometropic subtypes differ from each other.
• Chicken and egg situation : Is amblyopia a consequence or
a cause of strabismus/ anisometropia ?
• The relationship between performance on monocular
versus binocular tests has not been well-studied.
Hypothesis
• Impairment in binocular functions may
predict the pattern of monocular deficits,
and thereby help explain the mechanisms
(McKee, Movshon & Levi, 2003).
Subjects
•20 adults (age 19-35)
N
Age
Years of
education
Near acuity
normal or
fellow
Near acuity
amblyopia
Control
7
25.1
13.7
20/18
-
Strabismics
6
26.3
13.5
20/20
61*^
Anisometropes
7
28
14.9
20/23
61*^
Most Subjects have a history of patch treatment in their childhood.
Complete ophthalmologic examination was done to confirm diagnosis
General Methods
•Seven psychophysical Tests
•Monocular Tests
Amblyopic and fellow eye of amblyopic subjects tested separately
Stronger and weaker eye of normal subjects tested separately
•Binocular Tests
Both eyes tested simultaneously - required careful stimulus alignment
It is difficult to achieve precise alignment of stimuli in the two eyes, and we
pioneered new methods for achieving this using methods that are compatible
with fMRI.
Experiments
• Monocular tests
–
–
–
–
Snellen acuity
Grating acuity
Vernier acuity
Contrast sensitivity
• Binocular tests
– Randot stereotest
– Binocular motion integration
– Binocular contrast integration
Summary - Monocular Functions
• Amblyopic eyes showed a deficit for all the
monocular functions tested.
• Strabismic amblyopes are distinguished from
anisometropic amblyopes by their severe loss of
Vernier acuity.
Vernier acuity
•Measures the relative position of
an object
•Much finer than Snellen or grating
acuity (6-10 arc-sec of visual
angle)
• In our normal subjects Vernier is
12 times better than grating acuity
•A type of hyperacuity
Hyperacuity
photoreceptor = •
www.cnl.salk.edu/~thomas/ vernier.html
Binocular Tests - Methods
Dichoptic Stimulation with Avotec
Eye Tracking
with
Avotec/SMI
System
Stimulus Alignment Via Perceptual Report
Stimulus Alignment Via Fovea Reflex
Dual Eye Tracking
Alternate Cover Test
Summary - Binocular Functions
• Stereopsis
– Reduced in amblyopes, especially strabismics
• Binocular motion integration
– Binocular perception impaired in amblyopes,
especially strabismics
Can binocularity predict Vernier acuity?
Re-classification
• We reclassified amblyopes based on binocular properties.
• A simple pass/fail criterion was used to classify. The
subjects who passed both randot stereoacuity test and
binocular motion integration were assigned “binocular”
(33% strabismics and 57% anisometropes passed the
criteria).
• Those who couldn’t pass were assigned “non-binocular”
Result
• Deficits in Vernier acuity are much more
severe in ‘non-binocular’ group as
compared to ‘binocular’.
• Performance in ‘non-binocular’ subgroup
can not be predicted the by snellen/grating
acuities - suggesting additional factors.
Implications
• Vernier performance is better predicted by residual
binocularity than by clinical subtype.
• Interocular suppression may be an important
etiological factor in the development of amblyopia
(e.g., Sireteanu, 1980; Agrawal et al., 2006).
Future Directions
• fMRI experiments that study amblyopic
binocular suppression directly, perhaps in
comparison with binocular rivalry in normal
subjects.