When a Scientist Doesn't Think Like a Scientist: Review of Susan Barry's "Fixing My Gaze"

Susan Barry’s recent book, Fixing my gaze (2009), is an engaging narrative of life with a
small but intrusive disability – strabismus, or “crossed eyes”, with its common effect, poor depth
perception. Barry recounts the events of her infancy, when her strabismus developed; the
surgeries that cosmetically corrected the condition without entirely dealing with problems of
vision; her awkward school days; difficulty with driving and other tasks requiring judgment of
depth; and so on until a behavioral treatment, she says, transformed her life. She describes
her early and recent experiences vividly, even poetically, and intersperses scientific discussions
with anecdotes that serve as the “spoonful of sugar that helps the medicine go down”. But she
seems to forget her training in neuroscience when she should be thinking critically about the
effectiveness of treatments. Regrettably, this rather charming and informative book serves as an
extended advertisement for “developmental optometry” or “orthoptics”, a program that claims to
correct some problems of vision by eye exercises that increase control of convergence and
divergence (co-ordinated eye movements that are needed for ideal visual ability). Barry notes
that she excludes from consideration self-help orthoptic methods like the Bates method, which
were critiqued by Worrall, Nevyas, and Barrett in 2009. However, the method she advocates, the
Brock method, appears to have as little plausibility and as weak an evidence basis as the others
do. In addition, Barry speaks of her improved visual functioning as due to “rewiring” of the
brain, and, like many others who resort to this inapt “wiring” metaphor, suggests that high levels
of juvenile brain plasticity persist throughout life. It’s possible that they do, but Barry’s examples
may be more parsimoniously explained by reference to the less dramatic brain changes we call
“learning”.
Barry’s experience of strabismus and its consequences was not unusual, but her book is
unique in providing an understanding of the subjective experience of a strabismic. She notes her
early problems with reading and later problems with looking into the distance and with driving
confidently-- these in spite of excellent acuity of vision in each eye tested separately.
Nevertheless, she did drive, used a stereomicroscope, played tennis, and did not have any sense
of missing the experience of depth. Indeed, because she could use monocular depth cues
(information that can come from one eye rather than needing both), she did have some ability to
judge depth. As far as is known, though, she could not experience the very clear and accurate
sense of distance that comes from using retinal disparity, or a comparison of images as they
occur simultaneously at the two eyes (the reason for this will be discussed a little later). In
middle age, she began to experience shifts or “jiggling” of vision to such an extent that she
consulted a number of specialists. Following treatment by a developmental optometrist, who
prescribed “orthoptic” or “vision therapy” eye exercises, she reports that she began to have vivid
experiences of depth and improved her confidence and skill in driving and other tasks needing
distance judgments. Barry attributes her improved visual skills to a type of vision therapy, a
treatment category that has been defined as “a proposed optometric treatment for developing
efficient visual skills and processing… as a treatment for accommodative disorders, amblyopia,
binocular disorders (strabismic and nonstrabismic), learning disabilities, and ocular motility
disorders” (CIGNA Medical Coverage Policy, 2008, p. 1).
Although there is much that is valuable in Barry’s description of her subjective visual
experience, it is notable that little in the way of objective measures can be retrieved from her
early life. She provides in endnotes some objective measures made at the beginning and at the
end of her treatment, a period of 7 years. These indicate improvements in co-ordinated control of
eye movements. However, at no time does there seem to have been measurement of actual
judgment of depth, using the simple Howard-Dohlman apparatus so familiar to past generations
of psychology students. This device allows an observer to look through a small window into an
illuminated box in which there are two vertical rods. A system of strings allows the observer to
pull each rod backward or forward until the two appear to be side-by-side, the same distance
away. Normally, people do this poorly when using one eye at a time, and very well when they
use both eyes, and the difference between the monocular and binocular conditions shows how
well the eyes are used in coordination and how well retinal disparity is taken into account.
To examine the claims Barry makes on behalf of developmental optometrists, we need to
consider the plausibility of the claimed “rewiring” mechanism, the possibility of alternative
mechanisms, and the empirical evidence that the treatment Barry received was an effective
treatment for strabismic problems.


Is It Plausible That Orthoptic Treatment Could Cause “Rewiring”?

Barry suggests that she had been unable to use retinal disparity to judge distance, that this
was impossible because of the absence of binocular cells in the visual cortex, and that orthoptic
exercises allowed her to develop binocular cells receiving information from both eyes at once.
To discuss these issues, we need first to consider the question of plasticity. This term refers to
the extent to which development is guided by environmental stimulation. If a characteristic is
pretty well determined by heredity no matter what stimulation occurs (e.g., eye color), plasticity
is low. If it were possible to “rewire” all sorts of brain structures, as Barry implies, very high
plasticity would be present. But some aspects of development show high plasticity only during a
certain period of life (experience-expectant plasticity), while others can be guided by the
environment throughout life (experience-dependent plasticity). Quick language learning in early
life is an example of experience-expectant plasticity, and our slow,plodding acquisition of
vocabulary words later on is an example of experience-dependent plasticity.
The use of retinal disparity is usually considered to be a matter of experience-expectant
plasticity. As Barry points out, babies in the first months do not use their eyes together, but
switch attention from one to the other. By about 6 months, they are moving the eyes together.
This causes the image of an object they are looking at to fall on matching areas (corresponding
points) on the left and on the right retina. Messages are sent to the visual cortex simultaneously
from the two stimulated areas, and they cause activity in a single neuron which “lights up” only
when it gets a message from both eyes. This binocular neuron’s activity indicates that the two
eyes are looking at the same object in the same place, rather than two different objects, one seen
by the right and one by the left eye.
Repeating this experience many times makes the binocular neuron more responsive and
strengthens the connection between retinal areas and their associated neurons. But covering one
of the baby’s eyes for as little as a week (perhaps because of an injury), or a big difference in
the clearness of the images at the two eyes, can prevent the development of the connection with
the binocular neuron, and can even reduce the number of neurons from a single eye to the brain
until that eye is functionally blind. Similarly, if the baby is cross-eyed, and if the eyes often do
not co-ordinate, binocular neurons would not be likely to make normal connections. Studies of
cats and monkeys, by Hubel and Wiesel (1965), showed that animals deprived of co-ordinated
visual experiences in early life would develop abnormal use of their eyes and poor depth
perception. Another relevant point is that human babies go through at about 8 months a period of
exuberant synaptogenesis-- they create many synaptic connections between neurons, and indeed
have more synapses than they will ever have again in their lives. But some months later they
begin a process called “pruning” in which unused synapses are destroyed, and neurons
themselves disappear through programmed cell death ( Blakemore, 1989).
All these points challenge the plausibility of a continuing plasticity of the visual cortex
that would allow for major changes during later adulthood. However, it would be not be wise to
assume that we can completely reject the existence of such plasticity. The animal work by Hubel
and Wiesel is the only experimental evidence we have to depend on, and there are two problems
about generalizing from it. One is that Hubel and Wiesel had to kill the animals in order to
examine most of the brain features that interested them, and this was done fairly soon, so there
was no opportunity to see whether functions would be recovered or whether neural connections
would shift back to what they had been. Generalizing from non-humans to humans does not
necessarily give accurate answers about humans, either; even very similar species may differ in
important respects. A critical period, or time when experience had to occur in order to have an
effect, may have been present for cats, even for monkeys, but may not be the case for human
beings.If human beings are different in this way, it might be plausible that later experience could
affect the use of retinal disparity, although the other points made earlier suggest it is not. The
available information about human beings comes from uncontrolled studies showing the loss of
acuity in an eye following strabismus or patching of an eye in infancy. Rare reports of humans
who have had life-long cataracts removed in adulthood indicate that these people have not
recovered normal vision in spite of therapeutic and educational efforts (Gregory, 1997), but it is
hard to know whether visual problems were caused by the strabismus or cataract experience, or by additional factors that caused the original problem.


Is It Plausible That Other Mechanisms Could Be Affected by Eye Exercises?

Barry stresses the idea that the eye exercises she did might have altered the visual cortex
and other brain areas, perhaps through changes in long-term potentiation and consequent
improvement in the responsiveness of binocular neurons. However, there are other aspects of
vision that relate to retinal disparity; Barry mentions these, but returns to brain “rewiring” as her
preferred explanation. Given the arguments against persistence of early plasticity of the brain,
though, it is important to keep in mind that there are other plausible mechanisms by which
improved control over eye movements could cause better depth perception. These include the use
of the horopter for comparison of distances, constancy mechanisms, and visual adaptation.

Using the horopter.

How do people with a history of normal vision use retinal disparity?
The stimulation of a binocular neuron is not the only factor to be considered when we think
about judgment of depth using both eyes.
Older children or adults who have a normal visual history, and who have useful binocular
neurons, employ retinal disparity as their best way of judging distances. They can fixate (“fix”
the gaze) with both eyes on a single object, so the images at right and left eye fall on
corresponding points and activate the appropriate binocular neuron. This provides a single
“fused” image rather than separate images for each eye (double vision), and the fusion is
accompanied by a sense of depth. When this happens, however, the fixated object is not
necessarily the only one for which a single image is seen. Any object which is at any point on an
imaginary surface, all of whose points are at equal distances from the eyes, also has its images at
right and left eye fall on corresponding points, and is seen as a single object with fused images.
This imaginary surface, the horopter, shifts its position as the individual fixates objects at
different distances. Wherever the horopter may be, though, it determines which objects are seen
singly and which ones have double images. All objects that are off the horopter are experienced
as double images, but those that are very close to the horopter overlap so much that it is almost
as if they were fused. The farther the object from the horopter, the less the images overlap-- no
matter whether the object is on the same side of the horopter as the observer (near her) or on the
opposite side (far away). The overlapping or less-overlapping nature of the double images
provides information about depth.
Why don’t we consciously experience all these double images? We don’t pay attention to
them. We only pay attention to the object we are “looking at”. But we are able to pay attention
to double images by voluntary efforts. For example, when you are driving, you pay attention to
the road, or perhaps to your speedometer or other dashboard instrument. However, your hands on
the steering wheel are probably within your visual field. If you attend to them while still looking
ahead, you may notice that they look large; glance down at a hand, and it seems of normal size
again. Why? When the hand is not at the horopter, you see an overlapping doubled image--
larger than the single image by the amount that does not overlap. Fixating the hand gives you a
single image (assuming that you’re not strabismic) whose size is determined by the distance of
the hand from the eye, not by an extra image.
A person with a history of strabismus-- especially varying amounts of strabismus
produced by repeated surgical interventions-- may have learned that double images have no
reliable relation to the horopter, and learned to exclude them from consideration as depth
judgments are made. But it is not implausible that practice of eye movements, and increased
attentiveness to double images, could enable an adult to develop skill at comparing distances
with information about double or single images. Barry herself refers to a broadening of the part
of the visual field she actually pays attention too, an event that would help the observer pay
attention to images in the periphery.

Constancy and context.

Barry gives only slight attention to an essential aspect of
perception: the capacity for constancy. Constancy is the powerful tendency to experience objects
in the world as remaining the same, in spite of the continual changes in the ways they stimulate
our sense organs. Constancy is most obvious in visual perception. We see a square object as
retaining the same shape and size even though as we move relative to it its image becomes larger
or smaller and the image shape alters through a range of trapezoids. (Indeed, there are probably
few circumstances in which a square object creates a square image on the retina.)
Although the neural foundation of constancy is not well understood, it is clear that this
ability involves context. Shapes, sizes, and other aspects are judged in the context of a complex
surrounding visual field. Look at the square object through a tube that excludes the rest of the
field, and you see it as trapezoidal, not square. Similarly, a person who walks away from you
does not seem to shrink, but look through the tube and you will see that the image is much
smaller than before. Even when there is no movement of the object or the observer, constancy is
needed to overcome the effects of involuntary eye, head, and body movements, which make
images move across the retina.
Barry reports that her visual experience involved jiggling or shifting of the field, that she
had trouble recognizing where she was when driving, and that hawks went too fast for her to
count them on a bird-watching expedition. While other visual abilities are needed for keeping
the field still and so on, constancy also plays a role in these visual tasks. Constancy requires
some attention to the visual field surrounding the object being fixated. It is plausible that practice
and increased control over eye movements would enable an observer to improve constancy and
minimize some of the disturbing apparent movements or other problems with the fixated object. Barry calls attention to sudden changes in three-dimensional vision, as well as in other
aspects of visual experience, and attributes these to a “sudden and global change in brain state, a change in the activity of whole populations of neurons”(Barry, 2009, p. 236). Yet constancy
mechanisms are known for producing “flip-flops” of perceptual change in apparent distance,
size, brightness, and so on. Decades ago, the perception psychologist Adhemar Gelb
demonstrated abrupt changes in the perceived nature of a visual stimulus. He placed a piece of
coal so that it was illuminated by a spotlight that did not light any other part of the room.
Observers described the brightly-lit coal as appearing white. Gelb then introduced into the
spotlight beam a white piece of paper, holding it so that the observer saw coal and paper
simultaneously. Instantaneously, the coal “turned” black and the paper was seen as white.
Although one assumes that changes in neural activity underlie this effect of changed context, it
remains questionable whether there are “populations” involved, and if so, what the size of those
populations may be.

Adaptation.

Visual adaptation—learning by experience to interpret visual stimulation in
different ways-- is a capacity well-documented through experimental work, but also personally
familiar to wearers of corrective lenses. A new lens prescription for eyeglasses usually takes
several days before vision seems completely normal, and the experience can be accompanied by
“swinging of the scene” as movements of images seen through the lens and at the periphery are
compared. Adaptation may be related to cues such as the feeling of the eyeglass frame pressing
on the nose. (After cataract surgery, I wear glasses when driving, to correct the myopia of my
unoperated eye. Initially I saw double when I looked into the side mirror. Now my experience is
normal as long as I sit in the car-- but if I get out while still wearing the glasses, I lose my
balance.)
Experimentally, adaptation can take place in a few hours for subjects wearing prism
lenses that shift all images to one side or make all images tilt by 10 or 15 degrees. Not only do
these prism-wearers report that their visual world recovers its normal appearance, but they show this in objectively measurable ways. Asked to position a dim light straight in front of them, or to
set an illuminated rod to the vertical position, they initially respond in ways that compensate for
the prism displacement, but after a few hours of activity while wearing the prisms, they make
more accurate placements. Taking off the prism lenses, they show temporary after-effects in
which they set straight-ahead or vertical as if objects appear to them to be displaced in the
opposite way from the original prism displacement.
Because visual adaptation occurs quite quickly, and because it can easily be reversed, it
seems to be a matter of ordinary learning, rather than the brain “rewiring” suggested by Barry.
Adaptation may be a plausible alternative explanation for Barry’s much-increased ability to use
information from retinal disparity. Such an explanation might be supported by her anecdotal
reports of people whose new abilities diminished when they stopped doing the exercises, but
recovered when they began again.


Is There Non-Anecdotal Evidence for the Effect of Eye Exercises?

Whether or not there are plausible ways in which eye exercises that improve control of
co-ordinated eye movements might help depth perception and other visual abilities, the most
important way to evaluate “orthoptics” is through systematic tests of perceptual changes
following treatments by developmental optometrists. Barry’s book emphasizes individual
experiences to the almost complete exclusion of the kind of systematic investigation we normally
call “science”. The book’s index does not include the words “research”, “evidence”, or
“experiment”.

CIGNA HealthCare has declined to cover vision therapy treatments on the grounds that
they are considered “experimental, investigational, or unproven for the management of visual
disorders and learning disabilities” (CIGNA Medical Coverage Policy, 2008, p. 1). The company
states that “insufficient evidence exists in the published, peer-reviewed literature to conclude that vision therapy is effective for the treatment of any of the strabismic disorders except preoperative prism adaptation for acquired esotropia” (CIGNA, 2008, p. 3). In its policy, Aetna provides coverage for some uses of vision therapy (Aetna Clinical Policy Bulletins, 2009) but considers it experimental and investigational for anomalous retinal correspondence, one of Susan Barry’s problems.

Most serious research on aspects of vision therapy has concentrated on its role in
treatment of amblyopia (lazy eye). A Cochrane review (Shotton & Elliott, 2008) reported three
randomized controlled trial studies on this subject, but said they showed no clear evidence of an
effect of near visual activity of the kind used in vision therapy protocols. One study which used
retrospective comparisons to case records reported that eye exercises did not appear to reduce
symptoms in patients with esophoria, a mild version of the “in-turned” eye position Susan Barry
had to cope with (Aziz, Cleary, Stewart, & Weir, 2006). A randomized controlled study using
several placebo and other treatments reported significant improvement using specific vision
therapy methods, but had excluded from the study strabismics and patients with a history of
strabismus surgery-- the category into which Susan Barry would fall (Convergence Insufficiency
Treatment Trial (CITT) Study Group, 2008). None of these high-quality investigations appear to have used the Brock protocol which Barry advocates.


In Conclusion

The idea that many aspects of vision can be corrected by training goes back a long way.
Erasmus Darwin, grandfather of Charles, described in the Transactions of the Royal Society a 5-
year-old who used one eye only, turning his head so that images fell on the blind spot of the
other eye. Darwin proposed that the child wear a large false nose that would force use of the
problem eye by occluding the view of the good eye (King-Hele, 1999). As I noted earlier, there
are a number of plausible mechanisms for improvement of visual skills under this type of
regimen. However, Darwin found that the visual ability of his patient got worse in spite of his
efforts, and modern vision therapists have failed to present evidence that their more complicated
methods are any more effective.
One cannot argue with Barry’s subjective experiences and her sense that improved eye
co-ordination opened a new world of visual excitement as well as improved visually-related
skills. Whether others can benefit equally from vision therapy remains questionable, however,
and it is regrettable that Barry’s book, while stressing scientific facts, nevertheless fails to model
scientific thought for its readers. As has occurred before (Linus Pauling and Niko Tinbergen are
unfortunate examples), expertise in one area of science does not seem to guarantee critical
thinking on other topics. However, the general public, and even many more sophisticated
readers, are likely to accept questionable conclusions like Barry’s on the basis of her training in
biology combined with her personal experience. Indeed, Barry’s book was selected as a Library
Journal Best Sci-Tech Book of 2009. Certainly, it is in many ways a “good” book, vividly
written, with a suitable balance between subjective and objective material, and with an
informative discussion of neuroscience issues. But Barry has not shown her readers how to take
an investigative sip at the developmental optometry Koolaid. Instead, she drains the pitcher and offers packets for others to mix up.

References

Aetna Clinical Policy Bulletins: Vision Therapy. (2009). Retrieved on Aug. 31, 2009 from http://www.aetna.com/cpb/medical/data/400-499/0489.html.
Aziz, S., Cleary,M., Stewart, H.K., & Weir, C.R. (2006). Are orthoptic exercises an effective treatment for convergence and fusional deficiencies? Strabismus, 14(4), 183-189.
Barry, S. (2009). Fixing my gaze. New York: Basic.
Blakemore, C. (1989). Principles of development in the nervous system. In C. von Euler, H.Forssberg, & H. Lagerkrantz (Eds.), Neurology of early infant behavior (pp. 7-18). New York: Stockton Press.
CIGNA Medical Coverage Policy (2008). Retrieved on Aug. 30, 2020 from http://www.cigna.com/customer_care_professional/medical/mna_0221_coveragepositioncriteria_vision_therapy_orthoptics.pdf.
Convergency Insufficiency Treatment Trial (CITT) Study Group. (2008). The Convergence Insufficiency Treatment Trial: Design,methods, and baseline data. Ophthalmic Epidemiology, 15, 24-36.
Gregory, R.L. (1997). Eye and brain. Princeton: Princeton University Press.
Hubel, D., & Wiesel, T. (1965). Binocular interaction in striate cortex of kittens reared with artificial squint. Journal of Neurophysiology, 28, 1041-1059.
King-Hele, D.(1999). Erasmus Darwin. London: Giles de la Mare.
Shotton, K., & Elliott, S. (2008, Issue 2). Intervention for strabismic amblyopia. Cochrane Database of Systematic Reviews. Art. No.: CD006461. DOI: 10.1002/14651858.CD006461.pub2.
Worrall, R.S., Nevyas, J., & Barrett, S. (2009). Eye-related quackery. Retrieved Aug. 26, 2009, from www.quackwatch.com/01Quackery/RelatedTopics/eyequack.html.