The Evolutionary Basis of Strabismus and Nystagmus in Children

The
Evolutionary Basis
of Strabismus and
Nystagmus in Children
Michael C. Brodsky
123
The Evolutionary Basis of Strabismus
and Nystagmus in Children
Michael C. Brodsky
The Evolutionary Basis

of Strabismus and
Nystagmus in Children
Michael C. Brodsky
Professor of Ophthalmology and Neurology
Knights Templar Research Professor of Ophthalmology
Mayo Clinic
Rochester, MN
USA
ISBN 978-3-030-62719-5 ISBN 978-3-030-62720-1 (eBook)

https://doi.org/10.1007/978-3-030-62720-1
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2021
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and trans-
mission or information storage and retrieval, electronic adaptation, computer software, or by similar or
dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publica-
tion does not imply, even in the absence of a specific statement, that such names are exempt from the
relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To the winds of fate that charted my course
on this unexpected scientific journey through
our collective past.
Foreword
Let me congratulate you for picking up this book. Stay with it. You will be in for a
fantastic journey.
In ancient times, physicians, scientists, and philosophers were asking the same
questions: What is the human experience; why are we here? Recall that Socrates is
considered by some to be the father of Western medicine. As well as a great philoso-
pher. Think also of Galileo Galilei, Sir Isaac Newton, or Charles Darwin. But some-
where along the path of the human experience, perhaps culminating in Descarte’s
famous “I think, therefore I am,” physicians became increasingly focused on smaller
and smaller parts of the whole. This is not all bad. Think of how a powerful tele-
scope will allow study and observation of the smallest and most distant star. But
when viewed through that lens, one loses perspective of the great surround. Literally,
one cannot see the galaxy through all the stars. But if one looks through the “wrong
end” of the telescope, although the view is minified, the surrounding celestial bodies
can be seen and put in perspective. The greatest thinkers over time had the ability
and desire to focus on both views, the detailed and the wide. It is exactly this type
of original, broad-brush, and breakthrough thinking that characterizes Dr. Brodsky’s
writings. They cross-correlate studies from a number of different disciplines. His
ideas are bold and initially may seem iconoclastic, but they gradually weave a web
that consolidates much of what we see in our strabismic patients.
If one looks at strabismus through a telescope, we perhaps see crossed eyes,
DVD, or an overacting inferior oblique muscle. But if we step back as Dr Brodsky
has and look at the big picture through the other end of the telescope, through the
wide angle lens of evolutionary biology, we see that many strabismic disorders are
often the result of exaptations (a term describing a trait that has been co-opted for a
use other than the one for which natural selection has built it), and atavisms (a reap-
pearance of an ancestral trait that has been lost through evolutionary change over
successive generations). Evolution is not the underpinning of perfect design. As Dr.
Brodsky quotes R.W. Rodieck, “Evolution is not engineering-it is more like tinker-
ing-it is not design. The visual system never had an opportunity to be designed-or
redesigned from scratch; instead, both form and function reflect its long, particular,
and capricious history.”
This is book truly sui generus—unlike any I have read. It is unique in format. The
book consists of a collection of 20 articles (editorials, opinion pieces, and studies)
that the author has previously published over the arc of his career. Each chapter is
vii
viii Foreword
followed by a postscript written in 2020. These postscripts give insight into how the
author’s views may have changed since the articles’ publication. They view each
article through an ongoing evolution of understanding. As Brodsky states, “Anyone
who tells you that they have the final answer is not a scientist. In science, models
evolve with new and better data.”
Although I had read each of these articles when initially published, I see the util-
ity this book provides for reading them as parts of a piece. Each movement in a
symphony may seem complete in and of itself. But when experienced in sequence,
the full artistic message of the symphony can be more fully enjoyed. Similarly, each
of these 20 chapters can stand alone shedding light on its given subject. But when
read as parts of a whole thesis, we come to realize the author’s underlying message,
“Our past is always with us.”
Burton J. Kushner, MD
Professor Emeritus, Department of Ophthalmology and
Visual Sciences, University of Wisconsin,
Wisconsin, MD, USA
Preface
Pediatric ophthalmologists manage a unique collection of ocular motility distur-

bances that are rarely encountered in neurology and neuro-ophthalmology clinics.
Conditions that abound include essential infantile esotropia, infantile nystagmus,
latent nystagmus, dissociated vertical divergence, primary oblique muscle overac-
tion, dissociated horizontal deviation, and intermittent exotropia. How do these
developmental binocular disorders arise? Are they neurologic in origin? If so, what
specific neurologic pathways are involved, and why are other signs of neurologic
dysfunction conspicuous by their absence in so many affected patients?
Science tends to ignore its own history. As it turns out, we can begin to under-
stand this complex topic only by zooming out and trying to place these disorders in
a broader physiological context. I have found that the fundamental explanations for
the pediatric forms of early-onset strabismus and nystagmus lie buried away within
the field of evolutionary biology. Although basic physiologic studies in animals
identified primitive visual reflexes long ago, their significance to human visual
development somehow eluded us. These ancestral visual reflexes can now be cross-
linked with the many forms of pediatric strabismus and nystagmus, thus enabling us
to understand the necessary origin of these ocular motor aberrations, and to eluci-
date the neurological pathways that mediate them. By examining the evolutionary
underpinnings of these disorders, we can unlock the many hidden secrets that
underlie and unify these perplexing disorders.
Evolution (from the Latin evolutio), means an unfolding or an opening out of a
curve, emergence or release from an envelop or enclosing structure, or a transforma-
tion. By understanding the evolutionary origins of the panoply of eye movement
disturbances in infancy, we can grasp the deeper meaning of the complex pyrotech-
nic show of binocular eye movements that accompany infantile strabismus, and
appreciate the necessity of each eye movement as a balancing act in one plane of
visual space. In short, we can lift the veil to get a fleeting glimpse of why these
complex movements occur. You simply cannot understand infantile strabismus
unless you see it as a fingerprint of evolution.
Much of the unique phenomenology we encounter in pediatric ophthalmology
clinics can be explained by the fact that our two eyes once functioned together as
balance organs. In infantile strabismus, the visual cortex is remodeled to the subcor-
tical template, so that spontaneous cortical suppression of one eye can release these
primitive subcortical reflexes in the same way that covering an eye does.
ix
x Preface
The intrusion of these subcortical eye movements reflects a primitive binocular

tonus imbalance that disrupts binocular alignment and stable vision. Alas, many of
these visually driven preternatural eye movements are simply the flip-side of ves-
tibular eye movements in different planes. In neurology clinics, primitive reflexes or
“release phenomena” are known to resurface when higher cortical functions become
impaired, with different primitive reflexes emerging in early infancy than in adult-
hood. These disorders can be likened to “subconscious disorders” in Freudian psy-
chiatry that intrude to wreak havoc on our psyches. This is true everywhere in life
and in medicine, from neurology to psychiatry to culture to strabismus. Our past is
always with us.
This compilation of articles seeks to explain and to unify the common conditions
that define pediatric ophthalmology as holdovers of primitive evolutionary reflexes
that were not only useful but necessary in lateral-eyed animals. These reflexes linger
within the substratum of the ocular motor system. They originally functioned to
maintain vertical orientation during body rotations and to generate optokinetic
responses to stabilize vision during body movements. These visuo-vestibular eye
movements transmit motion input from the eyes to the vestibular nucleus, which
provides the central integration to distinguish world movement from body move-
ment. These primitive binocular reflexes lost their utility in frontal-eyed animals
with binocular vision and stereopsis, but were not erased from our neurological
scaffolding. They are suppressed by cortical vision in humans, but get reawakened
when cortical binocular vision fails to develop in early infancy. They reside in the
subcortical system, meaning the midbrain, brainstem, and cerebellum, which origi-
nally subserved luminance and motion detection in lateral-eyed afoveate animals.
Without first seeing them in their proper evolutionary context, we are groping in
the dark.
So are these binocular visual disorders neurologic in origin? Well, yes and no.
They are atavisms (from the Latin atavus, or “ancestor,” signifying a reappearance
of an ancestral trait that has been lost through evolutionary change over successive
generations). In humans, these atavistic binocular movements reflect persistent acti-
vation of subcortical visual pathways that are normally extinguished early in life.
They arise from a finely orchestrated system of primitive visual pathways that are
unfortunately disregarded in neuro-ophthalmology textbooks. As they are confined
within the visual system and its connections to the vestibular system, they are not a
“tip of the iceberg” of generalized neurological disease. But any preexisting neuro-
logic disorder can destabilize early binocular alignment and secondarily precipitate
these conditions. The following chapters will weave some context into what we are
seeing and hopefully provide the reader with a conceptual synthesis to codify the
stereotypical forms of early-onset strabismus and nystagmus that fill our pediatric
ophthalmology clinics.
Michael C. Brodsky
Rochester, MN, USA
Acknowledgement
I am grateful to the Mayo Clinic for providing me with the research time to envi-
sion, assemble, and curate this unique collection. I am also indebted to the Knights
Templar Eye Foundation, which generously provided research funding for the time
and resources needed to prepare this book. Finally, I wish to thank Asja Rehse, my
editor extraordinaire at Springer, who made it easy.
xi
Contents
1 Dissociated Vertical Divergence�� 1

What Is DVD?�� 1
What Is a Dorsal Light Reflex?�� 3
Is DVD a Dorsal Light Reflex?�� 7
Is DVD a Skew Deviation?�� 9
Conclusions�� 11
Postscript�� 14
2 DVD Remains a Moving Target!�� 17
Variable Torsional Component�� 18
Pseudoinferior Oblique Muscle Overaction�� 18
Additivity with Oblique Muscle Overaction�� 18
Association with Torticollis �� 19
Relationship to latent nystagmus �� 20
Postscript�� 22
3 Primary Oblique Muscle Overaction �� 25
Central Tonus Mechanisms for Primary Oblique
Muscle Overaction �� 26
Vestibular Interactions with the Ocular Motor System�� 26
Superior Oblique Muscle Overaction and A-Pattern Strabismus
in Neurologic Disorders�� 31
Inferior Oblique Muscle Overaction and V-Pattern Strabismus
in Congenital Esotropia�� 34
Primary Oblique Muscle Overaction and Hering’s Law�� 35
Oblique Muscle Overaction and Dissociated Vertical Divergence�� 37
Nonneurologic Causes of Oblique Muscle Overaction�� 37
Postscript�� 42
4 Do You Really Need Your Oblique Muscles?�� 45
Primary Adaptations in Oblique Muscle Function�� 45
Oblique Muscle Exaptations�� 47
From Visual Panorama to Frontal Binocular Vision�� 47
Cyclovergence, Stereoscopic Perception, and the Pitch Plane�� 47
xiii
xiv Contents
Cycloversion, Nonstereoscopic Perception, and the Roll Plane�� 54

Reversion From Exaptation to Adaptation�� 56
Postscript�� 62
5 Latent Nystagmus�� 65
What Is Latent Nystagmus? �� 65
Nasotemporal Asymmetry and Latent Nystagmus�� 66
Is Latent Nystagmus a Vestibular Nystagmus?�� 67
Neuroanatomy of Latent Nystagmus �� 68
Clinical Signs of Vestibular Origin�� 70
Evolutionary Underpinnings of Latent Nystagmus �� 73
Experimental Evidence That Latent Nystagmus Is Vestibular
in Origin �� 74
Postscript�� 80
6 Dissociated Vertical Divergence�� 83
Patients and Methods �� 85
Results�� 85
Comment�� 86
Postscript�� 93
7 The Reversed Fixation Test �� 95
Reversed Fixation Test �� 96
Scenario 1�� 97
Scenario 2�� 97
Scenario 3�� 100
Scenario 4�� 100
Comment�� 100
Postscript�� 104
8 Does Infantile Esotropia Arise From a Dissociated Deviation? �� 107
9 The Accessory Optic System �� 115
What Is the AOS?�� 115
10 Visuo-Vestibular Eye Movements �� 125
The Problem Is Gravity�� 125
Bilaterally Symmetrical Organs Function as Balance Organs�� 126
Lateral Eyes are Sensory Balance Organs �� 126
Primitive Reflexes are Resurrected When Normal
Neurodevelopment Fails to Occur �� 126
Ocular Motor Incursions Operate as Visual Balancing
Reflexes in Lateral-Eyed Animals�� 127
Contents xv
Primitive Visual Reflexes are Evoked by a Physiologic

Imbalance in Binocular Visual Input�� 129
Latent Nystagmus, Primary Oblique Overaction, and Dissociated
Vertical Divergence are Visuo-Vestibular Eye Movements �� 129
Visuo-Vestibular Eye Movements are Generated by Subcortical
Central Vestibular Pathways�� 130
Visuo-Vestibular Eye Movements Arise from a Central Vestibular
Imbalance that Dissociates Clinically into 3 Distinct Planes�� 130
Visual Reflexes are Stereoisomers of Vestibular Reflexes �� 132
11 An Expanded View of Infantile Esotropia �� 137
12 The Lizard’s Tail: An Ocular Allegory�� 147
13 The Optokinetic Uncover Test�� 151
Methods�� 152
Results�� 152
Comment�� 154
14 A Unifying Neurologic Mechanism for Infantile Nystagmus�� 165
What Is the AOS?�� 166
Role of the AOS in Infantile Nystagmus�� 167
Pendularity �� 167
Directionality �� 169
Foveation Periods�� 169
Absence of Oscillopsia�� 169
Null Position�� 170
Time of Onset�� 170
Association with Seesaw Nystagmus�� 171
Reversed Optokinetic Nystagmus�� 171
Augmentation by Fixation and Pursuit�� 172
Association with Congenital Visual Loss�� 172
Superimposition of Latent Nystagmus�� 172
Involvement of Cerebellar Pathways �� 173
Theories of Causation�� 175
15 Intermittent Exotropia and Accommodative Esotropia:
Distinct Disorders or Two Ends of a Spectrum? �� 183
xvi Contents
16 An Optokinetic Clue to the Pathogenesis of Crossed

Fixation in Infantile Esotropia�� 191
17 Is Infantile Esotropia Subcortical in Origin?�� 195
Introduction�� 195
Evolutionary Basis of Essential Infantile Esotropia (EIE)�� 196
Neuroanatomy of Monocular Nasotemporal Asymmetry (MNTA)�� 197
Evidence for a Subcortical Pathophysiology in Essential
Infantile Esotropia (EIE)�� 197
Potential Role of Prolonged Subcortical Neuroplasticity�� 199
An Optokinetic Etiology for Essential Infantile Esotropia (EIE)?�� 200
Conclusion �� 200
18 Phoria Adaptation: The Ghost in the Machine �� 207
Introduction�� 207
What Is Phoria Adaptation? �� 208
Stimulus for Phoria Adaptation�� 209
Measurement of Phoria Adaptation �� 209
Resilience of Phoria Adaptation�� 210
Time Course of Phoria Adaptation�� 210
Protean Clinical Manifestations�� 211
Neural Substrate�� 215
Future Questions�� 217
19 Monocular Nasotemporal Optokinetic Asymmetry—Unraveling
the Mystery �� 225
20 Infantile Nystagmus—Following the Trail of Evidence�� 231
Glossary of Terms�� 235
Bibliography �� 239
Index�� 241
Dissociated Vertical Divergence
A Human Dorsal Light Reflex
1
In the course of evolution, primitive responses to external stimuli are suppressed by

newer neurologic reflexes. When newer systems fail to function properly, primitive
reflexes that have been phylogenetically retained may reappear [1]. The emergence
of dissociated vertical divergence (DVD) in children with early-onset strabismus
may signal this type of atavistic response. It is my hypothesis that DVD is a human
dorsal light reflex that produces a visually mediated modulation of central vestibular
tone. This visuo-vestibular reflex, which is activated by fluctuations in binocular
visual input, manifests in humans when binocular control mechanisms fail to
develop in infancy.
What Is DVD?
Dissociated vertical divergence is an enigmatic disorder characterized by a slow

ascent of 1 eye that is followed, after a variable interval, by a slow descent of the
higher eye back to the neutral position [2–4]. The deviating eye frequently extorts
during its ascent, then intorts as it descends to resume fixation. Dissociated verti-
cal divergence manifests when binocular visual input is mechanically, optically,
or sensorially preempted (Fig. 1.1) [5, 6]. During the period of vertical misalign-
ment, visual input from the hyperdeviated eye is usually suppressed by the brain
so that affected individuals do not experience diplopia [2–4]. Since the intermit-
tent hyperdeviation of 1 eye is unassociated with a corresponding hypotropia in
the nondeviated eye on alternate cover testing, DVD is said to ignore Hering’s
law and seems to defy explanation according to current concepts of neuro-
anatomy [4].
Dissociated vertical divergence is a postscript to any early disruption of normal
binocular interactions [6]. It is seen most commonly in the setting of congenital
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 1

M. C. Brodsky, The Evolutionary Basis of Strabismus and Nystagmus in
Children, https://doi.org/10.1007/978-3-030-62720-1_1
2 1 Dissociated Vertical Divergence
Fig. 1.1 Dissociated vertical divergence. (a) Absence of hyperdeviation with binocular fixation.
(b) Right hyperdeviation and extorsion evoked by occlusion. (c) Left hyperdeviation and extorsion
evoked by occlusion. (d) Absence of hyperdeviation during binocular occlusion (as viewed through
translucent occluders)
esotropia but is also observed in association with congenital exotropia and after
surgical realignment of the hypotropic eye in congenital “double elevator palsy” [6].
As succinctly stated by Helveston [3], “… DVD is a reflex type of event that is pro-
grammed to occur if the appropriate mechanisms for nullifying its expression are
not functional.”
What Is a Dorsal Light Reflex? 3
A cogent neurophysiological explanation of DVD must account for the follow-

ing observations:
1. Both eyes never drift up simultaneously (Fig. 1.1) [7].

2. An inverse form of DVD, in which 1 eye drifts downward below horizon-
tal position, has only rarely been observed [8]. Why should this intermit-
tent vertical deviation manifest only as a hypertropia, and why should this
hypertropia alternate between the 2 eyes?
3. DVD develops as a delayed phenomenon in children with infantile stra-
bismus. It is usually first noted between 2 and 5 years of age [4].
4. The amplitude of the hyperdeviation is often asymmetrical in the 2 eyes,
and DVD may be unilateral in amblyopic eyes [2–4].
5. The vertical amplitude of DVD is variable, making accurate measurement
difficult [2, 3].
6. The slow velocity of the upward and downward drift of the deviating eye
does not resemble a saccade or pursuit movement but rather a slow diver-
gence in which fixation is constantly maintained by the nondeviating
eye [2, 3].
7. Placement of a vertical prism before 1 eye will induce a corresponding
vertical divergence [2].
8. The amplitude of the vertical deviation is incrementally related to the
asymmetry of visual input in the 2 eyes [2–4]. This effect is most clearly
shown by the Bielschowsky phenomenon, in which filters of increasing
density placed before the fixating eye cause the hypertropic eye to descend
incrementally, sometimes into a hypotropic position [2].
9. In some individuals, the horizontal component of DVD pre-dominates [9].
10. DVD is accompanied by a manifest head tilt in approximately 35% of
cases [10–12].
What Is a Dorsal Light Reflex?
Dissociated vertical divergence recapitulates a primitive visuo-vestibular righting

response in lateral-eyed animals (termed the dorsal light reflex or Lichtruckenreflex)
[13]. To determine the role of DVD in humans, one must first understand the nature
and function of the dorsal light reflex in lower animals. Animals need to know which
way is up, since maintaining vertical orientation is important for balance, naviga-
tion, and survival. Primates rely predominantly on graviceptive input to the vestibu-
lar system (ie, gravity receptors within the 2 labyrinths) to maintain vertical
orientation. However, many lower animals combine graviceptive input with
weighted visual information from the 2 eyes to establish and maintain verticality.
(The need to use visual input is evident when one considers that a fish swimming in
turbulent waters is subjected to mechanical forces that produce constant fluctuations
in vestibular input [14].)
In 1935, von Holst [15, 16] reported that a fish tends to orient its dorsal surface
toward the direction of maximal light intensity (Fig. 1.2). A fish restrained in an
upright position and illuminated from one side will move the eye that is ipsilateral
to the light source downward and the contralateral eye upward (Fig. 1.2) [15–17].
These postural and ocular responses to asymmetrical illumination are righting
reflexes that function to reorient the body and eyes with respect to the apparent
visual vertical, as judged by the direction of maximal light. Either reflex is only
partly compensatory, with its gain determined by the interplay of visual and vestibu-
lar drives [17].
von Holst [15, 16] discovered that visual and otolithic signals are yoked within
the central vestibular system to establish postural orientation in the roll plane. In the
restrained, labyrinthectomized fish, labyrinthine input can no longer curb this visu-
ally induced postural reflex, and the vertical divergence response to a lateral light
stimulus is approximately doubled [17]. The enhancement of this visual righting
reflex in the absence of vestibular input demonstrates that the dorsal light reflex is a
visually mediated ocular tilt reaction that is counterbalanced by the otoliths [14]. If
the dorsal light reflex in fish resulted from otolithic imbalance, ablation of the
a
13• 22• 27•
30• 60•
Direction
of Light
90•
b
30• 60•
Direction
of Light
5• 4• 3•
6• 10• 12• 90•
Fig. 1.2 Dorsal light response in goldfish. (a) In the unrestrained goldfish, the degree of body tilt
in the roll plane increases with increasing angles of illumination. (b) In the restrained goldfish,
vertical divergence of the eyes increases with increasing angles of illumination. (Reprinted with
permission from reference [18])
What Is a Dorsal Light Reflex? 5
otoliths would abolish it rather than increase it. When the right labyrinth is removed
and the left eye is blinded, visual and utricular tone summate and the fish will com-
mence permanent rolling to the right (Fig. 1.3) [14, 16]. When the left labyrinth and
left eye are left intact, utricular and visual innervation oppose each other and normal
postural responses are again observed [14, 16]. Thus, left utricular activation has an
effect similar to that of unilateral illumination of the right eye, and it counteracts the
effect of unilateral illumination of the left eye (Fig. 1.3). Visual information pertain-
ing to light asymmetry converges with utricular input pertaining to gravitational
asymmetry to modulate postural reflexes at the level of the vestibular nuclei [18].
The dorsal light reflex bears a profound resemblance to DVD (Table 1.1). (For
purposes of comparison, dorsal can be conceptualized as the direction from which
vertical sunlight illuminates the labyrinths in upright fish and humans.) In fish, the
amplitude of the dorsal light response increases with the intensity of illumination.
In DVD, the finding of variable vertical amplitudes in the 2 eyes must also be a
Fig. 1.3 In fish, visual and

utricular input is summated
to establish postural
Right Utricular Left Utricular
orientation in the roll Activation Activation
(frontal) plane. Curved
arrows denote direction of
body tilt induced by
unilateral visual or
utricular stimulation
Right eye Left eye
illumination illumination
Table 1.1 Similarities Between Dorsal Light Reflex in Goldfish and Dissociated Vertical
Divergence in Humans
Dorsal Light Reflex Dissociated Vertical Divergence
Evoked by asymmetrical light input to the eyes Evoked by asymmetrical visual input to the eyes
Visually deprived eye shifts dorsally Visually deprived eye shifts dorsally
Tropotactic response* Probably tropotactic
Magnitude of response dependent on strength Magnitude of response dependent on degree
of light stimulus of binocular visual disparity
Magnitude of response dependent on “mood” Vertical amplitude variable
and external factors
Long latency of vertical reequilibration after Slow reversal of hyperdeviation after
prolonged lateral illumination cessation of monocular occlusion
Decreasing exponential waveform Decreasing exponential waveform
Body tilts in roll plane toward light (ie, toward Presence and direction of head tilt variable
the side of the preferred eye)
Right utricular output counteracts right eye Undetermined
illumination
*Functions to reestablish binocular equilibrium rather than to directionally orient the eyes toward
incoming light
function of the degree of visual input asymmetry, as evidenced by the Bielschowsky

phenomenon in which the amplitude of the DVD can be titrated by placing filters of
varying density before the fixating eye [2]. von Holst devoted considerable discus-
sion to the phenomenon of Umstimmung (change of bias), which is the variability of
response from one trial to the next depending on internal and external factors (length
of time in darkness, mood, hunger, immediate visibility of prey) [16, 19]. This vari-
ability is also a prominent feature of DVD. In DVD, the observed asymmetry in the
hyperdeviation of the 2 eyes may reflect the momentary visual advantage of 1 eye,
as determined by the degree of amblyopia or by fluctuations in the level of suppres-
sion [8]. In the dorsal light response and in DVD, the eyes diverge slowly and the
divergence persists for a variable period after the inciting stimulus is removed [14].
The rate of each reaction shows a decreasing exponential waveform, suggesting that
the driving force for both reactions is proportional to the deviation from an altered
postural equilibrium [14, 20].
That DVD is a dorsal light reflex should not be taken to imply that it is also
dependent on the direction of incoming light. von Holst observed that a fish that has
had bilateral labyrinthectomy and 1 eye removed will commence permanent rolling
rather than orient the remaining eye with reference to the light source [16]. This
experiment demonstrates that the dorsal light response is a tropotactic response (ie,
one that functions to reestablish binocular equilibrium rather than to directionally
orient an eye toward incoming light) rather than a telotactic response (ie, a direct
orientation toward or away from light without the necessity of maintaining bilateral
balance) [13]. In humans with DVD, diffusion of light into 1 eye can produce a
hyperdeviation similar to that which occurs during occlusion of that eye [3, 8,
21–23]. That DVD persists in the supine position [24] suggests that the dorsal light
reflex in humans and lower animals functions as a binocular-disparity signal that
produces a preprogrammed neural output to the extraocular muscles. This neural
output retains its innervational characteristics regardless of light direction or body
position.
The role of the otoliths in inhibiting this visual postural reflex in humans is
unknown. The reversal of vertical amplitude asymmetry induced by reclining in a
head-hanging position supports the notion that otolithic input can modulate DVD to
some degree [24]. Utricular counterbalancing might explain why the higher eye, in
which vision is suppressed, eventually descends to its neutral position.
Neuroanatomical studies have shown that direct retinofugal projections to the
pretectal accessory optic nuclei and the lateral valvula cerebelli control the dorsal
light reflex in goldfish [25–28]. Unilateral lesions of the ipsilateral pretectal
nucleus or lateral valvuli cerebelli selectively abolish responses to light stimula-
tion of the contralateral eye [27]. Bilateral lesions completely abolish this visually
guided response, while lesions of the optic tectum have no such effect. In goldfish,
only the caudal portion of the lateral valvuli cerebelli receives visual input from
the contralateral retina via the ipsilateral pretectal nucleus. The rostral portion
receives sensory vestibular (ie, utricular) input, which is integrated with visual
input from the caudal portion to maintain optimal roll orientation. The valvula
cerebelli has no analogous structure in the mammalian cerebellum, but it is likely
Is DVD a Dorsal Light Reflex? 7
that visuo-vestibular information contributes to postural adjustment through the

cerebellum [25].
Is DVD a Dorsal Light Reflex?
How would a dorsal light reflex manifest in humans, who have frontally placed
eyes, predominantly binocular visual fields, and stereopsis? Binocular vision must
function to suppress this reflex, since a vertical divergence of the eyes would effec-
tively abolish binocularity and stereopsis. When binocularity is poorly developed,
however, this primitive reflex can reemerge if binocular visual input fluctuates. If
reduced visual input to 1 eye is interpreted by the brain as visual tilt, then the dorsal
light reflex should serve to equilibrate visual input between the 2 eyes. For this to be
the case, the central vestibular system must interpret decreased visual input to 1 eye
as being equivalent to a hypotropia, and respond with a vertical vergence signal. In
humans, the oblique muscles have been found to play the predominant role in pro-
ducing visual vertical divergence movements (Fig. 1.4), as elegantly demonstrated
by Enright [29] and elaborated on by others [30, 31]. Clinical observation [2, 8] and
scleral search coil recordings [30–32] in individuals with DVD have confirmed that
extorsion and elevation of the deviating eye are accompanied by a reciprocal intor-
sion and depression of the fixating eye, which conforms to the visual vertical diver-
gence response in humans [20]. The prominent role of the oblique muscles in
restoring vertical fusion is evidenced by the dynamic extorsion of the rising eye that
accompanies DVD, as well as the horizontal divergence of the non-fixating eye,
which reflects the tertiary abducting function of these muscles [31].
In the case of DVD, reduced visual input in the left eye would activate the left
inferior oblique and right superior oblique muscles to produce a cyclovertical diver-
gence movement (Fig. 1.5) [31]. Contraction of these muscles produces a clockwise
torsional movement of both eyes, together with infraduction of the right eye and
supraduction of the left eye. Maintenance of fixation with the right eye would
require simultaneous innervation to the elevators (superior rectus and inferior
Fig. 1.4 Vertical
divergence of the eyes in a
DVD. Vertical divergence
produced by decreased
visual input to the right eye
(a) and left eye (b) show
opposite vertical and
torsional components.
These divergence
b
movements enable normal
individuals to fuse small
vertical image disparities
a Visuo-vestibular Innervation
b Visuo-vestibular + Fixational Innervation
Fig. 1.5 Visuo-vestibular and fixational innervation in dissociated vertical divergence with hyper-
deviation of the left eye. In accord with the dorsal light reflex, decreased visual input in the left eye
is equivalent to a hypotropia, which activates a vertical vergence mechanism to equilibrate the 2
eyes. (Arrow size denotes relative magnitudes of innervation.) (a) In humans, vertical vergence is
mediated primarily by the oblique muscles, causing contraction of the inferior oblique muscle on
the left and the superior oblique muscle on the right. Visuo-vestibular innervation, if unaffected by
fixational innervation, would cause simultaneous elevation and extorsion of the left eye and depres-
sion and intorsion of the right eye. (b) Compensatory fixational innervation to the elevators is
necessary to maintain a steady vertical position of the right eye. By Hering law, fixational innerva-
tion to the right eye will augment the effects of vestibular innervation to the inferior oblique and
superior rectus muscles of the left eye, actively driving the hyperdeviation. Compensatory fixa-
tional innervation negates infraduction of the right eye but leaves it with a small intorsional
predominance
oblique muscles) of the right eye to counteract the infraducting action of the supe-
rior oblique muscle. By Hering law, compensatory fixational innervation to the
elevators of the fixating eye recruits the same muscles contralaterally, which would
augment vestibular innervation to the superior rectus and inferior oblique muscles
of the higher eye and actively drive the vertical component of the deviation (Fig. 1.5).
Thus, the observed hyperdeviation in DVD is the composite of 2 visual righting
reflexes: a visuo-vestibular reflex that activates a torsional divergence of the eyes
and a compensatory fixational reflex that maintains fixation with the visually advan-
taged eye and produces an upward movement of the other eye. The pivotal role of
the monocular fixation in DVD explains the nonexistence of simultaneous bilateral
DVD and dictates that an upward movement of 1 eye never begins while the other
eye is higher. It is only after the higher eye has completed its descent to the midposi-
tion that the other eye may start to ascend.
The interplay between visuo-vestibular and fixational innervation explains the
perplexing observation that, after removal of a cover from an eye of a patient with
DVD, the higher eye will sometimes descend below the neutral position before
resuming fixation [2–4]. A similar phenomenon is observed when occlusion of the
Is DVD a Skew Deviation? 9
fixating eye induces a downward refixation movement in the hyperdeviated eye, and
the covered eye makes a simultaneous downward movement below midposition
before ascending and extorting [2–4, 33, 34]. In both instances, a fixation shift pro-
vides a momentary glimpse into an underlying bias in central vestibular tone.
Is DVD a Skew Deviation?
Skew deviation is a descriptive term used to denote an acquired, supranuclear, verti-

cal misalignment of the eyes that fails to conform to known innervational patterns
of the extraocular muscles [35]. It is seen primarily in patients with unilateral brain-
stem lesions, particularly those involving the brainstem tegmentum in the mesodi-
encephalon or the medulla [36–38], although injury to the peripheral vestibular
system or the cerebellum can also cause it [39, 40]. It is now accepted that such
lesions inhibit or activate unilateral graviceptive output in the roll plane, allowing
utricular innervation from one side to predominate, which evokes an ocular tilt
reaction [41, 42]. This utricular ocular tilt reaction produces a tonic or paroxysmal
vertical divergence of the eyes (ie, skew deviation), which differs from DVD in that
the higher eye intorts, the lower eye extorts, and the head tilts toward the lowermost
eye [41–44] (Fig. 1.6).
Brandt and Dieterich [40, 41] have determined that skew deviation is usually
accompanied by binocular torsion, that both findings are components of the ocular
tilt reaction, and that the only difference between skew deviation and an ocular tilt
reaction is the presence of a head tilt. This utricular ocular tilt reaction is associated
with a pathologic shift in the internal representation of the gravitational vector, and
functions as a righting reflex to adjust the eyes, head, and body to a position that the
a b c
Fig. 1.6 Utricular ocular tilt reaction in humans. When the eyes are frontally placed, a torsional
component becomes necessary to neutralize a tilt in the roll plane. (a) Physiologic ocular tilt reac-
tion. A rightward body tilt activates right utricular and inhibits left utricular pathways subserving
graviceptive tone in the roll plane, resulting in vertical divergence with conjugate torsion of the
eyes and head tilt toward the lowermost eye. (b) Normal eye position with head upright. (c)
Pathologic ocular tilt reaction. A leftward ocular tilt reaction can be caused by an inhibitory lesion
of the left utricular pathways or an excitatory lesion of the right utricular pathways
central nervous system erroneously computes as being vertical [40–45]. The head
tilt is not compensatory for the ocular misalignment or torsion. Rather, the vertical
divergence, ocular torsion, and head tilt are all compensatory adjustments to the
new subjective vertical [41]. Under physiological conditions, this ocular tilt reaction
functions as a righting reflex that enables an organism to maintain gravitational
orientation during tilt in the roll plane (as occurs when skiing or motorcycle riding).
Tilting in the roll plane evokes a physiological skew deviation in lower animals and,
to a lesser degree, in humans (Fig. 1.6) [42]. However, the purely vertical skew
deviation in lower animals is supplanted by a torsional divergence movement in
humans, since the eyes are frontally placed.
No physiological position of body tilt evokes DVD, and no neurologic lesion has
produced it [46]. The intermittent forms of acquired skew deviation that have been
documented in patients with midbrain lesions have not been associated with extor-
sion of the ascending eye or intorsion of the descending eye, as occurs in DVD [36,
47–49]. Vertical divergence of the eyes with the “inverse” torsional characteristics
of DVD is a signature of abnormal binocular vision. Evolutionary conservation of
the dorsal light reflex as the visual counterpart of the utricular ocular tilt reaction
(Fig. 1.3) would explain the acquisition of these inverse torsional movements in
frontal-eyed animals.
In humans, both ocular tilt reactions are conserved as complementary “mirror-
image” righting reflexes, as evidenced by the observed extorsion of the rising eye in
DVD and the intorsion of the rising eye in the utricular ocular tilt reaction (Table 1.2).
In this context, DVD can be conceptualized as an inverse skew deviation. It is the
visual counterpart of neurologic skew deviation; the former is a visuo-vestibular
ocular tilt reaction, while the latter is a utricular ocular tilt reaction. An imbalance
of visual input from the 2 eyes modulates the afferent limb of DVD, while an imbal-
ance in graviceptive input from the utricles modulates the afferent limb of neuro-
logic skew deviation. If the utricular ocular tilt reaction and the dorsal light reflex
Table 1.2 Complementary Ocular Tilt Reactions in Humans

Dissociated Vertical Divergence Acquired Skew Deviation
Age at onset 2–4 y Any age
Tempo of onset Gradual Acute
Variability Intermittent Usually constant
Ocular torsion Extorsion of higher eye, Intorsion of higher eye, extorsion
intorsion of lower eye of lower eye
Pathophysiology Binocular visual imbalance Utricular imbalance (central or
peripheral)
Neurologic lesion None Unilateral utricular, brainstem,
or cerebellar lesion
Subjective awareness of Undetermined Variable
visual tilt
Head tilt Variable Toward side of lower eye
Perceived visual vertical Undetermined Rotated in direction of ocular
torsion
Diplopia None Vertical
Nystagmus Latent nystagmus Seesaw or hemi-seesaw
Conclusions 11
function together as complementary righting responses in lower lateral-eyed ani-

mals (Fig. 1.3), it is difficult to imagine that one mechanism (the utricular ocular tilt
reaction) would be retained to such a degree in frontal-eyed animals and the other
(the dorsal light reflex) completely discarded. Indeed, small degrees of DVD can be
evoked by monocular occlusion in some normal individuals [2, 50].
Conclusions
Dissociated vertical divergence is a dorsal light reflex that utilizes binocular visual
input to calibrate central vestibular tone. In lower animals, this human dorsal light
reflex functions to equilibrate visual input by simultaneously increasing dorsal light
input to one eye and decreasing it to the other. When binocular control mechanisms
are poorly developed in humans, this dorsal light reflex evokes a phylogenetically
newer vertical vergence movement of the eyes to produce an inappropriate vertical
divergence with torsional characteristics opposite to those required to neutralize a
body tilt (a righting reflex gone wrong).
In the upright human, the observed ocular movement is upward because the eye
with greater visual input is generally used for fixation and the movement of the visu-
ally disadvantaged eye is dorsally directed. The absence of a corresponding hypode-
viation on alternate cover testing reflects the instantaneous shift in visual advantage
to the uncovered eye that occurs with monocular occlusion. This hypothesis explains
the reciprocal nature of the observed hyperdeviation, the dynamic torsion that dis-
tinguishes DVD from other forms of skew deviation, and the tight link between
DVD and visual fixation.
The existence of DVD provides testimony to the duality of the ocular tilt mecha-
nism in humans, to the interplay between visual feedback and vestibular modulation
of extraocular muscle tone, and to the evolutionary role of binocular vision in the
suppression of this visuo-vestibular response.
References
1. Nathan P. The nervous system. 2nd ed. Oxford, UK: Oxford University Press;
1982. p. 84–103.
2. Bielschowsky A. Disturbances of the vertical motor muscles of the eyes. Arch
Ophthalmol. 1938;20:190–6.
3. Helveston EM. Dissociated vertical deviation: a clinical and laboratory study.
Trans Am Ophthalmol Soc. 1980;78:734–79.
4. von Noorden GK. Binocular vision and ocular motility. In: Theory and manage-
ment of strabismus. 4th ed. St Louis: CV Mosby; 1990. p. 341–5.
5. Flynn JT. Strabismus: a neurodevelopmental approach. New York: Springer-
Verlag NY Inc; 1991. p. 71–4.
6. Olsen RJ, Scott WE. Dissociative phenomena in congenital monocular eleva-
tion deficiency. J AAPOS. 1998;2:72–8.
7. Spielmann A. Les divergences verticales dissociées: excès desursum version lié

à la fixation: ou les divergences verticales dissociées à travers un écran translu-
cide. Ophtalmologie. 1987;1:457–60.
8. Guyton JS, Kirkman N. Ocular movement, I: mechanics, pathogenesis, and sur-
gical treatment of alternating hypertropia (dissociated vertical divergence, dou-
ble hypertropia) some related phenomena. Am J Ophthalmol. 1956;41:438–75.
9. Wilson ME, McClatchey SK. Dissociated horizontal deviation. J Pediatr
Ophthalmol Strabismus. 1991;28:90–5.
10. Anderson JR. Latent nystagmus and occlusion hyperphoria. Am J Ophthalmol.
1954;38:217–31.
11. Betchel RT, Kushner BJ, Morton GV. The relationship between dissociated ver-
tical divergence (DVD) and head tilts. J Pediatr Ophthalmol Strabismus.
1996;33:303–6.
12. Santiago AP, Rosenbaum AL. Dissociated vertical deviation and head tilts. J
AAPOS. 1998;2:5–13.
13. Duke-Elder S. The effect of light on movement. In: Duke-Elder S, editor. System
of ophthalmology: the eye in evolution. London, UK: Henry Kimpton; 1958.
p. 27–81.
14. Pfeiffer W. Equilibrium orientation in fish. Int Rev Gen Exp Zool.

1964;1:77–111.
15. von Holst E. Über den Lichtrückenreflex bei Fische. Pubbl Stn Zool Napoli
II. 1935;15:143–8.
16. von Holst E. Die Gleichgewichtssinne der Fische. Verh Dtsch Zool Ges.

1935;37:109–14.
17. Graf W, Meyer DL. Central mechanisms counteract visually induced tonus
asymmetries: a study of ocular responses to unilateral illumination in goldfish.
J Comp Physiol. 1983;150:473–81.
18. Meyer DL, Bullock TH. The hypothesis of sense-organ-dependent tonus mech-
anisms: history of a concept. Ann N Y Acad Sci. 1977;290:3–17.
19. von Holst E. Quantitative Untersuchungen über Umstimmungsvorgänge im
Zentralnervensystem, I: der Einfluss des Appetits auf das Gleichge
wichtsverhalten von Pterophyllum. Z Vergl Physiol. 1948;31:134–48.
20. van Rijn LJ, Simonsz HJ, ten Tusscher MPM. Dissociated vertical deviation
and eye torsion: relation to disparity-induced vertical vergence. Strabismus.
1997;5:13–20.
21. Verhoeff FH. Occlusion hypertropia. Arch Ophthalmol. 1941;25:780–95.
22. Crone RA. Alternating hyperphoria. Br J Ophthalmol. 1954;38:591–604.
23. Jampolsky A. Unequal visual inputs and strabismus management. In: Strabismus
symposium: transactions of the New Orleans Academy of Ophthalmology. St
Louis: CV Mosby Co; 1978. p. 358–492.
24. Goltz HC, Irving EL, Hill JA. Dissociated vertical deviation: head and body
orientation affect the amplitude and velocity of the vertical drift. J Pediatr
25. Watanabe S, Takabayashi A, Takagi S, et al. Dorsal light response and changes
of its responses under varying acceleration conditions. Adv Space Res.
1989;9:231–40.
Conclusions 13
26. Yanagihara D, Watanabe S, Mitarai G. Neuroanatomical substrate for the dorsal

light response, I: differential afferent connections of the lateral lobe of the val-
vula cerebelli in goldfish (Carassius auratus). Neurosci Res. 1993;16:25–32.
27. Yanagihara D, Watanabe S, Takagi S, Mitarai G. Neuroanatomical substrate for
the dorsal light response, II: effects of kainic acid–induced lesions of the val-
vula cerebelli on the goldfish dorsal light response. Neurosci Res. 1993;16:33–6.
28. Mori S. Localization of extratectally evoked visual response in the corpus and
valvula cerebelli in carp, and cerebellar contribution to “dorsal light reaction”
behavior. Behav Brain Res. 1993;59:33–40.
29. Enright JT. Unexpected role of the oblique muscles in the human vertical
fusional reflex. J Physiol Lond. 1992;451:279–93.
30. van Rijn LJ, Collewijn H. Eye torsion associated with disparity-induced verti-
cal vergence in humans. Vis Res. 1994;17:2307–16.
31. Guyton DL, Cheeseman EW, Ellis FJ, et al. Dissociated vertical deviation: an
exaggerated normal eye movement used to damp cyclovertical latent nystag-
mus. Trans Am Ophthalmol Soc. 1998;96:389–429.
32. Inoue M, Kita Y. Eye movements in dissociated vertical deviation. Nippon
Ganka Gakkai Zasshi. 1993;97:1312–9.
33. Anderson JR. Latent nystagmus and occlusion hyperphoria. Br J Ophthalmol.
1954;38:217–31.
34. Caldwell E. The significance of alternating sursumduction. Am Orthopt

J. 1967;17:39–43.
35. Brodsky MC, Baker RS, Hamed LM. Pediatric neuro-ophthalmology.

New York: Springer Verlag NY Inc; 1996. p. 252–5.
36. Keane JR. Alternating skew deviation: 47 patients. Neurology. 1985;35:725–8.
37. Brandt T, Dieterich M. Different types of skew deviation. J Neurol Neurosurg
Psychiatry. 1991;54:549–50.
38. Galetta SL, Liu GT, Raps EC, et al. Cyclodeviation in skew deviation. Am J
Ophthalmol. 1994;118:509–14.
39. Mossman S, Halmagyi GM. Partial ocular tilt reaction due to unilateral cerebel-
lar lesion. Neurology. 1997;49:491–3.
40. Brandt T, Dieterich M. Skew deviation with ocular torsion: a vestibular brain-
stem sign of topographic diagnostic value. Ann Neurol. 1993;33:528–34.
41. Brandt T, Dieterich M. Pathological eye-head coordination in roll: tonic ocular
tilt reaction in mesencephalic and medullary lesions. Brain. 1987;110:649–66.
42. Jáuregui-Remaud K, Feldon M, Clarke A, et al. Skew deviation of the eyes in
normal human subjects induced by semicircular canal stimulation. Neurosci
Lett. 1996;205:135–7.
43. Halmagyi GM, Brandt T, Dieterich M, et al. Tonic contraversive ocular tilt reac-
tion due to unilateral meso-diencephalic lesion. Neurology. 1990;40:1503–9.
44. Hedges TR, Hoyt WF. Ocular tilt reaction due to an upper brainstem lesion:
paroxysmal skew deviation, torsion, and oscillation of the eyes and head tilt.
Ann Neurol. 1982;11:537–40.
45. Brandt T, Dieterich M. Vestibular syndromes in the roll plane: topographic
diagnosis from brainstem to cortex. Ann Neurol. 1994;36:337–447.
46. Brandt T, Dieterich M. Central vestibular syndromes in roll, pitch, and yaw
planes. J Neuroophthalmol. 1995;15:291–303.
47. Mitchell JM, Smith JL, Quencer RM. Periodic alternating skew deviation. J
Clin Neuroophthalmol. 1981;1:5–8.
48. Greenberg HS, DeWitt LD. Periodic nonalternating ocular skew deviation
accompanied by head tilt and pathologic lid retraction. J Clin Neuroophthalmol.
1983;3:181–4.
49. Corbett JJ, Schatz NJ, Shults WT, et al. Slowly alternating skew deviation:
description of a pretectal syndrome in three patients. Ann Neurol. 1981;10:540–6.
50. van Rijn LJ, ten Tusscher MPM, de Jong I, Hendrikse F. Asymmetrical vertical
phorias indicating dissociated vertical deviation in subjects with normal bin-
ocular vision. Vis Res. 1998;38:2973–8.
Postscript
Although this article was dedicated to dissociated vertical divergence (DVD), which
is considered to be a tertiary phenomenon in infantile strabismus, it was revelatory
for me in that it “cracked the code” of infantile strabismus by showing DVD to be
an atavistic visual reflex. Accordingly, it provided compelling evidence that titrat-
able binocular deviations can arise from unbalanced binocular input through the
subcortical visuo-vestibular pathways.
Many who disagreed with this analysis fundamentally misunderstood it. Some
pointed out that there is no vestibular imbalance in infantile strabismus. This is, of
course, true, as it is the visual rather than the otolithic input that is fluctuating in
DVD. More adamantly, they noted that the visual cortex is integral to DVD, as corti-
cal suppression of one eye can activate it, suggesting to them that DVD must be a
cortical phenomenon. However, our cortical visual motion pathways are reconfig-
ured to the older subcortical pathways in infantile strabismus, so that cortical sup-
pression can produce the same outcome as unequal luminance input in lateral eyed
animals. This also explains how DVD can occur spontaneously when both eyes are
open (and therefore receiving equal luminance input). As fixation has a stronger
effect on DVD than luminance disparity, it is likely that conscious fixation evokes
cortical suppression of one eye in patients with strabismus. So DVD can be acti-
vated both by an exteroceptive stimulus (covering one eye) or an interoceptive stim-
ulus (cortical suppression). The unanswered question is whether covering one eye
evokes DVD through subcortical pathways, whether it evokes cortical suppression
and thereby evokes DVD, or whether both mechanisms can be operative to varying
degrees.
Some have questioned the existence of this subcortical mechanism by pointing
out that there no subcortical photometer in humans. This is untrue. The melanopsin
pathways that project from intrinsically sensitive retinal ganglion cells to the supra-
chiasmatic nucleus to entrain circadian rhythms. I once considered the melanopsin
to be a good candidate system for DVD. This evolutionarily ancient opsin is selec-
tively sensitive to blue light, which comes from the sky. However, preliminary lumi-
nance studies with my colleagues, Drs. Scott Larson and Randy Kardon at the
Postscript 15
University of Iowa, have not shown a selective effect of blue light binocular dispar-
ity in generating DVD. In retrospect, this could have been predicted since the
Bagolini bar, a red filter bar, is often placed over one eye to titrate DVD and thereby
confirm the diagnosis.
Whereas volitional body movements are controlled by the cerebral cortex, indi-
vidual muscles are innervated by subcortical centers. Only within the vestibular
system, where visual motion is encoded in canal referenced planes, can torsional
eye movements be generated. To my mind, the fact that DVD is linked to cortical
suppression (through fixation, occlusion, diffusion, or suffusion with light), is what
distinguishes DVD as a human dorsal light reflex.
DVD Remains a Moving Target!
2
Dissociated vertical divergence (DVD) is an ocularmotor phenomenon that is

observed in early-onset strabismus and that, until recently, has remained inexplica-
ble. I have proposed that DVD is a dorsal light reflex that emerges in humans when
early-onset strabismus precludes the development of normal binocular vision [1].
This hypothesis is based on physiologic studies that show that unequal light input to
the 2 eyes of a vertically restrained fish produces a vertical divergence of the eyes,
with depression of the eye that has greater visual input and elevation of the eye that
has lesser visual input [2]. In unrestrained fish, unequal visual input induces a body
tilt in the roll (frontal) plane toward the side with greater input (ie, a dorsal light
reflex) [3]. Land-based animals cannot tilt in space without falling; therefore, a
body tilt in aquatic animals is supplanted by a head tilt in humans.
Because humans have frontally placed eyes, ocular rotation in the roll plane cor-
responds to a cyclovertical movement in which the eye with lesser visual input ele-
vates and the eye with greater visual input is used for fixation [1]. Three-dimensional
scleral search coil recordings have shown that DVD consists of a cyclovertical ver-
gence movement of both eyes, followed by a vertical movement of both eyes that
reestablishes fixation with the lowest eye [4, 5]. These prominent torsional move-
ments implicate the oblique muscles as the primary activators of DVD [4, 5], as
predicted long ago by Guyton and Kirkham [6]. For example, in DVD with hyper-
deviation of the left eye, simultaneous innervation to the right superior oblique and
the left inferior oblique muscles create a cyclovertical divergence of the eyes. When
this occurs, vertical stabilization of the lower right eye necessitates upward fixa-
tional innervation to counteract the infraducting action of the superior oblique mus-
cle. This conjugate fixational innervation to the superior rectus and inferior oblique
muscles of both eyes stabilizes the right eye and actively drives the hyperdeviation
in the left eye. Because both eyes are innervated simultaneously to produce this dis-
sociated movement, one should not refer to the elevating eye as the “affected eye”
or “the eye with DVD.”

18 2 DVD Remains a Moving Target!
The superimposition of DVD on the static and dynamic innervational derange-

ments associated with congenital esotropia expands its clinical expression to pro-
duce the following signs:
Variable Torsional Component
The variable torsional component of DVD can be explained by considering the base-
line torsional position of the eyes. von Noorden [7] has observed that when DVD is
accompanied by “true” inferior oblique muscle overaction, the dynamic intorsion of
the deviated eye is absent as it returns to primary position. Conversely, I have
observed that patients with DVD and superior oblique muscle overaction often have
a prominent torsional component that may exceed the vertical component of the
deviation [8]. These observations suggest that the torsional amplitude of DVD is
defined by the disparity between the extorsional position of the hyperdeviated eye
during inferior oblique muscle activation, and the baseline torsional position of the
globe. When the baseline position is one of extorsion (as occurs with inferior oblique
muscle overaction) [9], this torsional disparity is minimal and the corresponding
torsional movement is small as the extorted eye ascends and descends to an extorted
position. When the baseline position of the globes is one of intorsion (as occurs with
superior oblique muscle overaction), there is a large torsional disparity that is mani-
fested as a prominent torsional movement as the eye ascends and descends.
Pseudoinferior Oblique Muscle Overaction
DVD can manifest in adduction and thereby simulate inferior oblique muscle overac-
tion. This finding is classically attributed to occlusion of the adducting eye by the nose
[7, 10]. However, the notion that the child’s nose acts as an occluder is dubious,
because young children have a poorly formed nasal bridge. Furthermore, alternate
cover testing in lateral gaze (which can only be obtained when no occlusion is present)
confirms that the adducting eye is elevated when the abducting eye is fixating in
DVD. It has been my observation that spontaneous hyperdeviation of the adducting
eye in DVD begins as the eye rotates into the vertical field of action of the inferior
oblique muscle, suggesting that paroxysmal activation of the inferior oblique muscle
can excessively elevate the adducted eye. As the eye moves into adduction, the increas-
ing vertical action of the inferior oblique muscle explains how DVD can manifest
primarily in adduction in the absence of true inferior oblique muscle overaction (ie,
without hypotropia of the abducting eye, a V pattern, or fundus extorsion) [10, 11].
Additivity with Oblique Muscle Overaction
In children with early-onset esotropia, DVD often coexists with oblique muscle
overaction [7, 10–12]. Whereas DVD is caused by paroxysmal activation of the
oblique muscles (with secondary fixational activation of the vertical rectus
Association with Torticollis 19
muscles), oblique muscle overaction is generally caused by sensory torsion of the

globes that leads to tight oblique muscles [5, 9]. When both conditions are present,
their additive effects determine the observed vertical deviation in lateral gaze [10,
11]. Although some believe that DVD is neither created nor abated by surgical
weakening of the oblique muscles, the relative effects of oblique muscle surgery on
the torsional and vertical components of DVD in different positions of gaze have not
been measured.
Association with Torticollis
DVD is often associated with an anomalous head tilt that may be directed toward or
away from the side of the hyperdeviated eye [13–18]. In some patients, head tilting
may be a compensatory means of controlling the hyperdeviation. Jampolsky [15,
16] has characterized the head tilt pattern in DVD in which a hyperdeviation of
either eye increases or becomes manifest when the head is tilted to the opposite side.
This observation is now understandable (and indeed predictable) on the basis of the
torsional kinematics of DVD. For example, DVD with hyperdeviation of the left eye
occurs when the right superior oblique and left inferior oblique muscles receive
simultaneous innervation and a secondary fixational innervation is recruited to
maintain monocular fixation with the lower (and visually preferred) right eye [5]. A
head tilt to the right activates otolithic innervation to the right superior oblique and
the left inferior oblique muscles and thereby increases the left hyperdeviation,
whereas a head tilt to the left would recruit otolithic innervation to neutralize this
cyclovertical divergence [19]. This reasoning suggests that a head tilt to the side of
the hyperdeviating eye can serve as a compensatory means of recruiting otolithic
innervation to control the hyperdeviation.
Other patients with DVD have a head tilt pattern characterized by a hyperde-
viation that becomes manifest or increases when the head is tilted to the same side
[15–18]. According to Jampolsky [16], when fixation with 1 eye predominates in
DVD, tonic fixational innervation to the contralateral superior rectus muscle can
eventually lead to a contracture. If this contracture occurs, it can alter the head tilt
response to one characteristic of superior rectus muscle contracture, with an
increase in the vertical deviation on head tilt to the side of the hyperdeviating eye
[16–18]. When superior rectus muscle contracture develops, a compensatory head
tilt to the opposite side is presumably used to minimize otolithic innervation to the
tight superior rectus muscle, and a head tilt to the side of the higher eye will make
manifest or increase a hyperdeviation [16–18]. In this setting, surgical recession
of a tight superior rectus muscle can reduce or eliminate the compensatory head
tilt [18].
Some head tilts in children with DVD are accompanied by a manifest hyperde-
viation [13, 14]. A dorsal light reflex in humans would produce a head tilt away
from the side of the hyperdeviating eye. Such a tilt would reflect a central vision-
dependent tonus asymmetry [19] and would potentiate the DVD. This head tilt does
not exist to realign the eyes, and surgical treatment of the associated hyperdeviation
should not eliminate it.
Relationship to Latent Nystagmus
Most children with congenital esotropia and DVD also have latent nystagmus [5, 7,
12, 14, 20]. Guyton et al5 have proposed that DVD is a compensatory movement
that damps the cyclovertical component of latent nystagmus. von Noorden has
argued that this variable damping of latent nystagmus could be an effect rather than
the cause of DVD [5]. Pritchard [21] recently found DVD in 47% of patients with
intermittent exotropia, most of whom had no latent nystagmus.
Latent nystagmus, like DVD, is linked to unequal visual input to the extent that
occlusion, suppression, or amblyopia makes it manifest and alternate occlusion
reverses its direction. In describing DVD, Bielschowsky [22] noted that “. . . while this
eye maintains constant fixation, the other eye, which is deviated upward behind the
screen, will make at irregular intervals vertical movements of different extent. . . .”
This observation correlates with eye movement recordings that show that DVD can be
associated with a vertical latent nystagmus in the hyperdeviating eye [23, 24]. If latent
nystagmus occurs because asymmetric visual input resets central vestibular tone in the
yaw plane (ie, plane of rotation about an earth-vertical axis), then DVD might incor-
porate a similar oscillation. Alternatively, the cyclovertical divergence that occurs in
DVD could translate a component of the latent nystagmus into the vertical plane,
which could reduce the horizontal component and cause it to appear on eye movement
recording as a vertical latent nystagmus.
Ironically, these complex ocular kinematics ultimately raise basic diagnostic
questions. Although the torsional component of DVD is not cosmetically problem-
atic, what is its significance for the patient? When does it begin? Is it synonymous
with the torsional nystagmus seen in uncorrected congenital esotropia? Is it associ-
ated with a shift in the subjective visual vertical? When DVD improves spontane-
ously, does it merely revert to a torsional component that is subclinical?
Our fundamental treatment goals for DVD are also called into question. Is the
goal of strabismus surgery simply to eliminate the vertical component of DVD?
Does oblique muscle weakening selectively alter the torsional component? Can ver-
tical rectus muscle surgery selectively reduce the vertical component of DVD at the
expense of worsening the torsional component? Answers to these questions will
inevitably redirect our approach to surgical treatment of DVD.
In its essence, DVD is a balancing movement that uses binocular visual input to
calibrate ocular motor and postural tone in the roll plane. As we undertake to recog-
nize its subjective correlates and define its adaptational functions, we are reminded
that DVD remains a moving target.
References
1. Brodsky MC. Dissociated vertical divergence: a righting reflex gone wrong.
Arch Ophthalmol. 1999;117:1216–22.
2. von Holst E. Über den Lichtrückenreflex bei Fische. Pubbl Staz Zool (Napoli).
1935;15:143–8.
Relationship to latent nystagmus 21

J Comp Physiol. 1983;150:473–81.
4. Van Rijn LJ, Collewijn H. Eye torsion associated with disparity-induced verti-
5. Guyton DL, Cheeseman EW, Ellis FJ, Straumann D, Zee DS. Dissociated verti-
cal deviation: an exaggerated normal eye movement used to damp cyclovertical
latent nystagmus. Trans Am Ophthalmol Soc. 1998;96:389–429.
6. Guyton JS, Kirkham N. Ocular movement, I: mechanics, pathogenesis, and sur-
gical treatment of alternating hypertropia (dissociated vertical divergence, dou-
ble hypertropia) and some related phenomena. Am J Ophthalmol.
1956;41:438–75.
7. von Noorden GK. Binocular vision and ocular motility: theory and man-
agement of strabismus. 4th ed. St Louis: Mosby–Year Book; 1990.
p. 341–5.
8. Brodsky MC. Dissociated torsional deviation. Binoc Vis Q. 1999;14:6.
9. Guyton DL, Weingarten PE. Sensory torsion as the cause of primary oblique
muscle overaction/underaction and A- and V-pattern strabismus. Binoc Vis
Q. 1994;9:209–36.
10. Wilson ME, Parks MM. Primary inferior oblique overaction in congenital eso-
tropia, accomodative esotropia, and intermittent exotropia. Ophthalmology.
1989;96:950–7.
11. McCall LC, Rosenbaum AL. Incomitant dissociated vertical divergence and
superior oblique overaction. Ophthalmology. 1991;98:911–8.
12. Helveston EM. Dissociated vertical deviation: a clinical and laboratory study.
Trans Am Ophthalmol Soc. 1980;78:734–79.
13. Lang J. Congenital convergent strabismus. Int Ophthalmol Clin. 1972;4:
88–92.
15. Jampolsky A. Management of vertical strabismus. In: Pediatric ophthalmology
and strabismus, transactions of the New Orleans Academy of Ophthalmology.
New York: Raven Press; 1986. p. 157–64.
16. Jampolsky A. A new look at the head tilt test. In: Fuchs AF, Brandt TH, Büttner
U, Zee DS, editors. Contemporary ocular motor and vestibular research: a trib-
ute to David A. Robinson. Stuttgart: Springer Verlag; 1994. p. 432–9.
17. Betchel RT, Kushner BJ, Morton GV. The relationship between dissociated ver-
tical divergence (DVD) and head tilts. J Pediatr Ophthalmol Strabismus.
1996;33:303–6.
18. Santiago AP, Rosenbaum AL. Dissociated vertical deviation and head tilts. J
AAPOS. 1998;2:5–11.
19. Brodsky MC. Vision-dependent tonus mechanisms of torticollis: an evolution-
ary perspective. Am Orthopt J. 1999;49:156–60.
20. Anderson JR. Latent nystagmus and occlusion hyperphoria. Am J Ophthalmol.
1954;38:217–31.
21. Pritchard C. Incidence of dissociated vertical divergence in intermittent exotro-

pia. Am Orthopt J. 1998;48:90–3.
Ophthalmol. 1938;20:175–200.
23. Irving EL, Goltz HC, Steinbach MJ, Kraft SP. Objective video eye movement
recording: a useful tool in diagnosis of dissociated vertical deviation. Binoc Vis
Q. 1997;12:181–90.
24. Irving EL, Goltz HC, Steinbach MJ, Kraft SP. Vertical latent nystagmus com-
ponent and vertical saccadic asymmetry in subjects with dissociated vertical
deviation. J AAPOS. 1998;2:344–50.
Postscript
In this extension of my original treatise, I address the clinical phenomenology of

DVD. This analysis has stood the test of time. Occasionally, DVD can manifest as
an enigmatic head tilt following strabismus surgery for infantile esotropia. The
known mechanisms by which DVD can be accompanied by a head tilt are summa-
rized in the following figure. Although the diagram depicts clear dichotomies, it
often remains difficult if not impossible to categorically assign a distinct cause to
the torticollis in any given patient (Fig. 2.1). The tug-of-war between the need for
binocular fusion and the need for vertical orientation may explain the absence of
torticollis in most patients with DVD.
Congenital Esotropia
Horizontal Realignment of the Eyes
Binocutarity
with Fixation
Equal Preference or Amblyopia
Vision Superimposed
Hypertropia
Symmetrical DVD Asymmetrical DVD Unilateral DVD

– Superior Rectus + Superior Rectus
Contracture Contracture
No head tilt Compensatory Compensatory Noncompensatory

head tilt toward head tilt away head tilt away
side of higher eye from side of from higher eye
higher eye (dorsal light reflex)
Fig. 2.1 Clinical algorithm for the differential diagnosis of dissociated vertical divergence
(DVD)-associated head tilts. (Used with permission from Brodsky MC, Jenkins R, Nucci
P. Unexplained head tilt following surgical treatment of congenital esotropia: a postural manifesta-
tion of dissociated vertical divergence. Br J Ophthalmol. 2004;88:264-272, British Medical
Association)
Postscript 23
Binocular Occlusion OS Occlusion OD
Light
Fixation
(Normal light)
Fig. 2.2 Video-oculography in a patient with bilateral DVD and latent nystagmus (upper panel)
showing a brief true vertical divergence of the eyes, causing them to initially rotate in opposite
directions when either eye is occluded (lower panel). Note that the latent nystagmus intensifies as
the DVD develops. OD, right eye; OS, left eye. (With permission from Christoff A, Raab EL,
Guyton DL, et al: DVD-A conceptual, clinical, and surgical overview. J AAPOS 2014;18:378-384.
Elsevier Press)
For any number of reasons, I consider it implausible that DVD occurs to damp
latent nystagmus. To accept this notion one would have to ignore all of the evolu-
tionary evidence that both DVD and latent nystagmus are subcortical reflexes aris-
ing from unequal visual input from the two eyes. Many patients without latent
nystagmus get DVD and, in most patients with both conditions, DVD does not
appear to reduce the amplitude of latent nystagmus. In fact, most patients with DVD
show an intensification of latent nystagmus when either eye is covered (Fig. 2.2).
The presence of DVD in some patients with intermittent exotropia and no latent
nystagmus dispels the notion that latent nystagmus must be the driving force for
DVD to develop. No controlled studies have examined patients with latent nystag-
mus without DVD to see if these patients also show nystagmus damping. Importantly,
latent nystagmus is of greater intensity when the poorer-seeing eye is fixating. If
DVD were occurring to damp latent nystagmus, one would necessarily expect to see
a greater DVD amplitude when the patient was attempting to fixate with the poorer
eye. However, the opposite is often the case. Latent nystagmus and DVD occur
together but neither arises from the other. It is easy to confuse links and correlations
with cause and effect.
Primary Oblique Muscle Overaction
The Brain Throws a Wild Pitch
3
Primary oblique muscle overaction is a common ocular motility disorder character-

ized by vertical incomitance of the eyes in lateral gaze [1]. In primary inferior
oblique muscle overaction, an upshoot of the adducting eye occurs when gaze is
directed into the field of action of the inferior oblique muscle, producing a greater
upward excursion of the adducted eye than of the abducted eye [1]. The opposite
occurs in primary superior oblique muscle overaction. Although ductions appear to
be normal and there is no evidence of yoke muscle paresis, alternate cover testing
discloses a vertical tropia of similar magnitude in the abducting eye. Primary infe-
rior oblique muscle overaction is usually associated with ocular extorsion and
V-pattern strabismus, whereas primary superior oblique muscle overaction is usu-
ally associated with ocular intorsion and A-pattern strabismus [2–5]. Superior
oblique muscle overaction is often accompanied by other neurologic disease,
whereas inferior oblique muscle overaction generally occurs in children who have
congenital esotropia but no other overt neurologic abnormalities. Surgical weaken-
ing of the overacting oblique muscles improves versions, eliminates the associated
A or V pattern, and reduces torsion.
In 1916, Ohm [6–8] postulated that pattern strabismus and oblique muscle over-
action may be due to abnormal vestibular innervation. Almost a century later, a uni-
fying neurologic mechanism to explain primary oblique muscle overaction remains
elusive. This ocular motor phenomenon seems to defy fundamental principles of
physiology since nowhere else in the body do individual muscles bilaterally overact.
The primary function of the oblique muscles in lower vertebrates such as fish is
to counterrotate the eyes torsionally in response to pitch (fore-and-aft) movements
of the body [9, 10]. As a fish pitches its body to swim upward or downward, a com-
pensatory “wheel” rotation of the eyes is produced by the oblique muscles in
response to vestibular stimulation [9]. The existence of this physiologic oblique
muscle overaction in lower animals led us to question whether a central vestibular
imbalance in the pitch plane might offer an explanation for the occurrence of

26 3 Primary Oblique Muscle Overaction
primary oblique muscle overaction in humans. Clinical observations suggest that an

imbalance in central vestibular premotor output to the extraocular muscle subnuclei
can produce the primary oblique muscle overaction that accompanies congenital
strabismus. This central vestibular imbalance develops when early loss of binocular
vision or neurologic disease alters central vestibular output in the pitch plane to
produce excessive tonus of the extraocular muscles that elevate the eyes (in the case
of congenital esotropia and inferior oblique muscle overaction) or depress the eyes
(in the case of neurologic disease and superior oblique muscle overaction).
entral Tonus Mechanisms for Primary Oblique

C
Muscle Overaction
The term tonus was originally coined by Ewald [11] to describe the state of excita-
tion of a living muscle during rest. In 1977, Meyer and Bullock [12] advanced their
tonus hypothesis, which states that neuronal tonus pools within the central nervous
system receive multisensory input and that tonus asymmetries between antagonistic
pools can produce tonic motor responses. According to this hypothesis, the eyes are
not merely sensory organs but components of a multimodally driven tonus pool that
calibrates baseline muscle tone (ie, tonus-inducing organs) [12, 13]. This hypothesis
explains how sensory information collected by the eyes can help to govern extra-
ocular muscle tonus. The bilateral positioning of the eyes and ears permits them to
function as balance organs. Visual and graviceptive input are yoked together within
the central vestibular system to determine optimal postural orientation.
In his early pioneering studies of vision-dependent tonus responses in fish, von
Holst [14] found that a posterior shift of a dorsal light source induces a pitch-up move-
ment of the body, whereas an anterior shift induces a pitch-down movement, as if the
animal is programmed to position the body so that the light source retains a dorsal
orientation (Fig. 3.1). With the body stabilized in the upright position, an overhead light
moving fore-and-aft evokes a wheel-turning movement of both eyes, which rotate to
maintain torsional alignment with the light source (Fig. 3.1) [14–16]. Since light nor-
mally comes from overhead when a fish is upright, a posterior movement of the light is
registered as a pitch forward movement of the body (ie, a movement of the body away
from the light). This change in visual input evokes increased tonus to the inferior
oblique muscles, which extort the eyes (Fig. 3.1). The observation that visual and ves-
tibular input can alter postural and extraocular muscle tonus to produce a physiologic
bilateral oblique muscle overaction in lower animals suggests that similar excitatory
stimuli may be operative in strabismic humans with primary oblique muscle overaction.
Vestibular Interactions with the Ocular Motor System
To understand why primary oblique muscle overaction so often accompanies early-

onset strabismus, it is instructive to examine the components of central vestibular
tone that influence eye position. The primary function of the vestibuloocular system
Vestibular Interactions with the Ocular Motor System 27
a b
c d
e f
Fig. 3.1 Physiologic effects of gravistatic (postural) and visual input to the oblique muscle tonus
in fish. These bilateral torsional eye movements function to align the eyes with the perceived visual
vertical by modulating oblique muscle tonus. (a) A pitch-down body movement evokes increased
inferior oblique muscle tonus and extorsion of the eyes. (b) A pitch-up movement evokes increased
superior oblique muscle tonus and intorsion of the eyes. (c) In the unrestrained fish, an anterior
light source evokes a pitch-down body movement. (d) In the unrestrained fish, a posterior light
source evokes a pitch-up body movement. (e) In the restrained fish, anterior movement of overhead
light evokes increased superior oblique muscle tonus and intorsion of both eyes. (f) In the restrained
fish, posterior movement of overhead light evokes increased inferior oblique muscle tonus and
extorsion of both eyes
is to maintain eye position and stabilize fixation during head movements [16].
Vestibuloocular movements are the most primitive of all extraocular movements. As
expounded by Walls,
. . . the primitive function of the eye muscles was not to aim the eyes at all. Their original
actions were all reflex and involuntary, and were designed to give the eyeball the attributes
of a gyroscopically-stabilized ship, for the purpose of maintaining a constancy of the visual
field despite chance buffetings and twistings of the animals body by water currents.
[9] (p303)
In the rabbit, for example, a rightward body tilt along its long axis causes the
right eye to be lower in space than the left eye. This tilt elicits a compensatory verti-
cal divergence of the eyes to elevate the right eye and depress the left eye, thereby
stabilizing the eyes in space [17–19]. A pitch forward of the body would produce a
compensatory extorsional movement of both eyes [14, 15, 20].
Now consider the same pitch-down body movement in a rabbit that is fixating
with the right eye maximally abducted and the left eye maximally adducted
(Fig. 3.2). Since the eyes are laterally placed in the rabbit, this position of gaze
Fig. 3.2 Overhead view of a rabbit fixating an object in the right posterior visual field. Solid lines
correspond to the visual axis of the abducted right eye and the adducted left eye. When the rabbit
pitches forward (as when starting to run down a hill), the head rotates downward and the tail rotates
upward. Although both eyes move downward in space, the left visual axis (which is directed
toward the nose) rotates downward, while the right visual axis (which is directed toward the tail)
rotates upward (curved arrows). This divergence of the visual axes corresponds to a right hypertro-
pia that must be neutralized by vestibular innervation to elevate the lower left eye and depress the
higher right eye. The compensatory vertical divergence for a pitch-forward position corresponds to
primary inferior oblique muscle overaction
Vestibular Interactions with the Ocular Motor System 29
would direct the left visual axis anterior to its neutral position and the right visual
axis posterior to its neutral position. A forward pitch in the body plane with the eyes
in this position would tilt the left visual axis to a lower position in space than the
right visual axis (Fig. 3.2). This tilt would necessitate compensatory vestibuloocular
innervation to increase upward tonus in the left eye and increase downward tonus in
the right eye, while extorting both eyes in response to the body pitch. Conversely, if
the body were pitched back during dextroversion, the higher visual axis of the
adducted left eye would necessitate increased downward tonus in the left eye and
increased upward tonus in the right eye to stabilize the position of the eyes in space.
The necessary vestibulo-ocular movements, which correspond to the vertical diver-
gence in lateral gaze seen in humans with primary oblique muscle overaction, are
programmed at an early evolutionary stage to assure stability of the visual field in
all fields of gaze. In 1996, Zee [18] formulated this hypothesis to explain how the
alternating skew deviation in lateral gaze that occurs in humans could be a reversion
to a phylogenically old otolith-mediated righting reflex in lateral-eyed animals.
Zee’s hypothesis also explains how superior oblique muscle overaction with alter-
nating hypotropia of the adducting eye reflects a pitch-up otolithic bias and a down-
ward tonus bias to the extraocular muscle plant, whereas inferior oblique muscle
overaction with alternating hypertropia of the adducting eye would result from a
pitch-down otolithic bias and an upward tonus bias to the extraocular muscle plant.
In lateral-eyed animals and in humans, the semicircular canals are roughly
aligned with the extraocular muscles (Fig. 3.3) [17]. When the head is rotated in a
particular plane, a semicircular canal within the labyrinth detects acceleration and
sends excitatory innervation to the extraocular muscle(s). Within the brainstem and
Fig. 3.3 The close Midline

anatomical relationship of Optic Axis
the semicircular canals and Superior

Oblique
the extraocular muscles in
humans is shown. (Figure 51°
modified with permission

from Simpson and
Graf [17])
Superior Rectus
23°
Anterior
41°
Semicircular Canals
56°
Posterior
Top View
cerebellum, peripheral vestibular input is summated to produce appropriate innerva-

tion to the extraocular muscle subnuclei and maintain the position of the eyes in
space (Fig. 3.4) [18–22]. Each anterior semicircular canal provides excitatory inner-
vation to the ipsilateral superior rectus and the contralateral inferior oblique muscles
while inhibiting the yoked ipsilateral inferior rectus and contralateral superior
oblique muscles (Fig. 3.4). Likewise, each posterior semicircular canal system pro-
vides excitatory innervation to the ipsilateral superior oblique and the contralateral
inferior rectus muscles while inhibiting the ipsilateral inferior oblique and the con-
tralateral superior rectus muscles. In humans, a pitch-up movement of the head (as
occurs when raising the chin) activates both posterior semicircular canals, which
send excitatory innervation to both depressors in both eyes. Like their target extra-
ocular muscles, the semicircular canal pathways have a push-pull (yoke) relation-
ship, so that activation of one canal inhibits the antagonist canal [19]. Thus, the
SR
LR
MR
IO
III
IV
MLF
AC
PC
HC
VI
LVN
MVN
Fig. 3.4 Neuroanatomical projections from the labyrinths to the extraocular muscles. The orienta-
tion of the anterior semicircular canal corresponds to that of the ipsilateral superior rectus and
contralateral inferior oblique muscles. The orientation of the posterior semicircular canals corre-
sponds to that of the ipsilateral superior oblique and contralateral inferior rectus muscles. The
orientation of each horizontal canal corresponds to that of the horizontal rectus muscles. Turning
the head to the right stimulates the right horizontal canal to increase excitatory innervation to the
right medial rectus muscle and left lateral rectus muscle so that the eyes rotate equally and opposite
to the direction of head rotation. HC indicates horizontal canal; AC, anterior canal; PC, posterior
canal; LVN, lateral vestibular nucleus; MVN, medial vestibular nucleus; VI, abducens nucleus;
MLF, medial longitudinal fasciculus; IV, trochlear nucleus; III, oculomotor nucleus; SR, superior
rectus muscle; MR, medial rectus muscle: LR, lateral rectus muscle; and IO, inferior oblique mus-
cle. (Data modified with permission from Tusa [21])
Superior Oblique Muscle Overaction and A-Pattern Strabismus in Neurologic Disorders 31
pitch-up movement that excites both posterior canals also inhibits both anterior
semicircular canals, which send inhibitory innervation to the ocular elevators. The
result is an equal contraversive movement of both eyes to adjust for the pitch-up
head movement. Injury to or inhibition of anterior canal pathways subserving
upward eye movements causes a functional activation of the posterior canal down-
gaze pathways and produces downward eye movements [22].
In addition to the semicircular canals, each labyrinth contains otolithic sensors
consisting of the utricle and the saccule [19]. While the semicircular canals respond
to angular acceleration and produce dynamic vestibuloocular movements, the paral-
lel otolithic system responds to linear acceleration and is sensitive to changes in
static head position [19]. Damage to the semicircular canal pathways produces pha-
sic ocular deviations and nystagmus, while damage to the otolithic projections cor-
responding to the semicircular canal pathways causes tonic ocular deviations
(strabismus) [19, 23–25]. The otolithic pathways are not as well studied, but are
believed to have similar projections to the corresponding canal pathways [19]. For
the sake of simplicity, we refer to the otolithic pathways corresponding to a particu-
lar canal pathway simply as the anterior canal or posterior canal system, recogniz-
ing the similarity in projections between the otoliths and semicircular canals.
Likewise, we also use the term anterior (or posterior) canal predominance to mean
“predominance of the otolithic pathways corresponding to those of the anterior (or
posterior) semicircular canals.”
The otolithic pathways corresponding to the anterior and posterior semicircular
canals are under different inhibitory control (Fig. 3.5) [19, 21, 22, 26, 27]. The ante-
rior canals receive inhibitory connections from the cerebellar flocculi, while the
posterior canals do not. Thus, a structural lesion or metabolic abnormality that
inhibits output from the cerebellar flocculi can also disinhibit the anterior canals,
resulting in an upward deviation of the eyes [21]. Conversely, bilateral lesions of the
ventral tegmental tract or brachium conjunctivum can injure central pathways from
the anterior semicircular canals and produce a posterior canal predominance, result-
ing in tonic downgaze. Maturation of cerebellar floccular inhibition to anterior canal
pathways may be dependent on normal visual experience early in life. Ocular stabi-
lization is normally modulated by visual and vestibular input. When binocular
visual input is pre-empted, this multisensory mechanism may fall under greater
weight of labyrinthine control, allowing excitatory anterior canal output to predomi-
nate [28].
uperior Oblique Muscle Overaction and A-Pattern Strabismus

S
in Neurologic Disorders
A bilateral lesion that injures both anterior canal pathways or disinhibits both pos-
terior canal pathways will increase prenuclear innervation to the superior oblique
and inferior rectus subnuclei, resulting in a posterior canal predominance and
increased downward tonus to both eyes. This downward tonus must be overcome by
fixational innervation (Fig. 3.6). Since the inferior rectus muscles retain their
SR
IR
III N
I S
O
MLF
BC BC
BC VTT
MLF
SVN
SVN
MVN
VTT
AC AC
HC HC
Cerebellum PC
PC
FLO NOD FLO
Fig. 3.5 Segregation of pathways controlling anterior and posterior canal tone. Only the anterior
canal pathways receive inhibitory innervation by the cerebellar flocculus. A loss of modulation
from the cerebellar flocculi could disinhibit the anterior canals and produce an upward tonus
imbalance, leading to bilateral inferior oblique muscle overaction, bilateral extorsion, and V-pattern
strabismus. FLO indicates flocculus; NOD, nodulus; AC, anterior canal; PC, posterior canal; HC,
horizontal canal; SVN, superior vestibular nucleus; VTT, ventral tegmental tract; MVN, medial
vestibular nucleus; MLF, medial longitudinal fasciculus; BC, brachium conjunctivum; III N, ocu-
lomotor nucleus (S, I, O, and M represent the oculomotor subnuclei); SR, superior rectus muscle;
and IR, inferior rectus muscle. (Data modified with permission from Tusa [21])
vertical field of action in adduction while the superior oblique muscles have mini-
mal vertical action in abduction, this downgaze predominance would produce a
relative overdepression of the adducting eye in lateral gaze (Fig. 3.7). Activation of
both superior oblique muscles produces bilateral intorsion in the primary position
and an A pattern due to the tertiary abducting action of the superior oblique muscles
in downgaze. In addition, binocular intorsion rotates the inferior rectus insertions
laterally and reduces the adducting action of the inferior rectus muscles in downgaze.
The vestibuloocular pathways pass through the posterior fossa and are suscepti-
ble to injury when structural abnormalities involve the brainstem or the cerebellum.
In children with hydrocephalus and myelomeningocele, the constellation of
Superior Oblique Muscle Overaction and A-Pattern Strabismus in Neurologic Disorders 33
Right Labyrlnth Vestibular Innervation Left Labyrlnth

PC PC
a
HC HC
AC AC
Vestibular and Fixational Innervation

PC PC
b
HC HC
AC AC
Fig. 3.6 Superior oblique muscle overaction. (a) Vestibular innervation. A central vestibular tonus
imbalance corresponding to bilateral posterior canal predominance would produce tonic down-
gaze, divergence, and intorsion of the eyes if unopposed by fixational innervation. (b) Vestibular
plus fixational innervation. Fixational innervation, which conforms to the Hering law, recruits
bilateral innervation to the superior rectus and inferior oblique muscles to negate the vertical com-
ponent of the downward tonus bias. Fixational innervation allows a disconjugate intorsional bias
to persist. PC indicates posterior canal; HC, horizontal canal; and AC, anterior canal
Fig. 3.7 Superior oblique muscle overaction. The observed eye movements in different fields of
gaze are a summation of fixational innervation that conforms to Hering’s law, and an underlying
central vestibular imbalance that does not. All 4 depressors are receiving excessive vestibular
innervation. Since the vertical action of the superior oblique muscles is maximal in adduction, the
adducting eye exhibits a downshoot in adduction relative to the abducting eye. The tertiary abduct-
ing effects of the overacting superior oblique muscles are maximized by vestibular innervation in
downgaze and minimized by fixational intervation in upgaze, producing an A pattern
A-pattern strabismus, bilateral superior oblique muscle overaction, and bilateral

intorsion is often associated with tonic downgaze early in life [29–38]. Children
with myelomeningocele not only have hydrocephalus but also frequently have an
associated Chiari II malformation [35, 36]. Since prenuclear input to the vestibular
system from the vestibulocerebellum is primarily inhibitory, bilateral compression
of or injury to those vestibulocerebellar pathways activating the anterior canals

would disinhibit the posterior canals and increase extraocular muscle tonus in their
target muscles.
Previous investigators [35–40] have speculated that bilateral superior oblique
muscle overaction may be supranuclear or prenuclear in nature, citing the frequency
with which it accompanies defective upgaze. Biglan [37, 38] attributed the overact-
ing superior oblique muscles, A pattern, and chronic downward deviations of the
eyes in children with myelomeningocele to defects in the vertical gaze pathways
producing either a failure to inhibit the downgaze pathways or excessive stimulation
of downward gaze. Acute comitant esotropia caused by neurologic disease such as
hydrocephalus or Chiari malformation is often associated with bilateral superior
oblique muscle overaction [39]. Although orbital anatomical factors have also been
implicated as a cause of superior oblique muscle overaction in hydrocephalus [1,
29], the high frequency of structural abnormalities within the posterior fossa led
Hamed [35, 36] and colleagues to propose that superior oblique muscle overaction
and alternating skew deviation in lateral gaze may share a common neuroanatomi-
cal substrate. Recently, Hoyt [41] has observed that premature infants with periven-
tricular leukomalacia or intraventricular hemorrhage may initially manifest a tonic
downgaze that evolves into an A-pattern esotropia and bilateral superior oblique
muscle overaction.
I nferior Oblique Muscle Overaction and V-Pattern Strabismus

in Congenital Esotropia
Clinical observations and eye movement recordings have documented abnormal

ocular responses to vestibular stimulation in children with strabismus [42–45]. Gait
and postural control have also been studied in children with different kinds of stra-
bismus, and a defect of postural stability has been demonstrated in esotropic but not
exotropic children [46, 47]. The high prevalence of incoordination and balance dis-
orders in children with “isolated” congenital esotropia also supports the notion that
early loss of single binocular vision associated with congenital esotropia may affect
central vestibular tone [47].
Humans display an inherent upward tonus predominance of the eyes, which cor-
relates with anatomical differences in the orientation of the anterior and posterior
canals [28]. This inherent up-down asymmetry in central pathways may explain
why vertical vestibular optokinetic responses normally favor upward rather than
downward slow phases [48–51]. It may also explain why a slight downbeat nystag-
mus may be seen in individuals attempting to fixate an imaginary target in darkness
[28, 52] and why a hyperphoria of the adducting eye can often be elicited in indi-
viduals fixating in lateral upgaze with a Maddox rod covering one eye [53]. The
development of single binocular vision serves to increase downward tonus to the
extraocular muscles and hold this inherent upward bias in check. Conversely, the
disruption of single binocular vision associated with congenital esotropia reduces
downward tonus to the extraocular muscles, perhaps by disrupting maturation of
Primary Oblique Muscle Overaction and Hering’s Law 35
inhibitory pathways from the cerebellar flocculi to the anterior canals. The resulting
anterior canal predominance would increase upward tonus in the extraocular mus-
cles and predispose to bilateral inferior oblique muscle overaction (Fig. 3.8).
Prolonged occlusion of one eye can also induce inferior oblique muscle overac-
tion in nonstrabismic humans with normal stereopsis [54, 55], suggesting that either
prolonged interruption or early loss of single binocular vision may also be regis-
tered as forward pitch (ie, away from the light). The retention of this primitive
vision-dependent tonus mechanism in humans would explain why poor sensory
fusion leads to inferior oblique muscle overaction rather than superior oblique mus-
cle overaction. The “neurologic lesion” that induces this central vestibular imbal-
ance is loss of binocular visual input.
Primary Oblique Muscle Overaction and Hering’s Law
Primary oblique muscle overaction appears to defy Hering’s law [56], which dic-
tates that, in any volitional conjugate movement, both eyes receive equal innerva-
tion. As summarized by Bielschowsky, “all of the muscles of both eyes always
participate in each movement; one half experiences an increase in tonus and the
other half a decrease” [57] (p178). This control system optimizes binocular vision
in all positions of gaze [56–58]. Although Hering’s law requires that the ocular
Right Labyrinth Vestibular Innervation Left Labyrinth
PC A PC
HC HC
AC AC
Vestibular and Fixational Innervation

PC PC
B
HC HC
AC AC
Fig. 3.8 Primary inferior oblique muscle overaction. (a) Visuovestibular innervation. Failure to
develop normal binocular vision is associated with increased upward tonus to the eyes, perhaps
through reduced anterior canal inhibition from the cerebellar flocculi. A central vestibular tonus
imbalance corresponding to bilateral anterior canal predominance would produce tonic upgaze,
horizontal divergence, and extorsion of the eyes if unopposed by fixational innervation. (b)
Visuovestibular plus fixational innervation. Fixational innervation recruits equal innervation from
the inferior rectus and superior oblique muscles to negate the vertical component of the upgaze
bias, and allows the disconjugate extorsional bias to persist. PC indicates posterior canal; HC, hori-
zontal canal; and AC, anterior canal
motor system synthesize a conjugate signal to the motor neurons involved in the
execution of any ocular movement, it should be evident from the previous discus-
sion that equal innervation to any set of vertical yoke muscles would produce dis-
sociated movements of the 2 eyes. To execute conjugate vertical eye movements, the
extraocular muscles of both eyes must receive appropriate innervation to move the
eyes equally rather than receiving equal innervation.
Hering [56] made reference only to voluntary eye movements as conforming to
his law of equal innervation. Since the semicircular canals and their corresponding
otolithic pathways segregate innervation to each set of yoke muscles, it is not sur-
prising that dissociated eye movements of central origin are generally associated
with vestibular disease. Paradoxically, these dissociated movements may reflect the
fact that the vertical yoked muscles receive roughly equal innervation rather than
the necessary innervation to rotate the eyes equally in one plane.
Our model of primary oblique muscle overaction as a pitch plane imbalance
predicts that oblique muscles overact bilaterally in conjunction with rather than
relative to their yoke vertical rectus muscles. In primary gaze, the torsional action
of the overacting oblique muscles predominates in both eyes, producing the bilat-
eral extorsion observed in primary inferior oblique muscle overaction and the bilat-
eral intorsion observed in primary superior oblique muscle overaction. When both
sets of elevators or depressors receive excessive central vestibular innervation,
adduction of either eye produces excessive vertical excursion of the adducting eye
as it moves into the vertical field of action of the overacting oblique muscle
(Fig. 3.7). In this context, an upward tonus imbalance to both eyes manifests as
bilateral overelevation of the adducting eye, and a downward tonus imbalance mani-
fests as bilateral overdepression of the adducting eye. Volitional gaze out of the
vertical field of action of the overacting yoke muscles recruits physiologic innerva-
tion to counterbalance the vertical tonus imbalance, while gaze into the vertical field
of action of the overacting yoke muscles allows this underlying tonus imbalance to
predominate, producing the A and V patterns observed clinically (Fig. 3.7). The
ocular torsion produced by primary oblique muscle overaction also initiates a cas-
cade of secondary mechanical events, including rotational displacement of the rec-
tus muscle insertions, oblique muscle length adaptation, and mechanical tightening
of the oblique muscles, as detailed elegantly by Guyton and Weingarten [5]. These
peripheral responses augment the overelevation or overdepression of the adducting
eye and the corresponding A and V pattern observed clinically.
This neurologic model would also explain why primary oblique muscle overac-
tion is usually associated with a negative Bielschowsky head-tilt test [1, 59]. A head
tilt to either side recruits ipsilateral otolithic innervation to stimulate 1 of the 2
overacting vertical muscles in each eye while inhibiting the other. The net result for
each eye is a minimal change in vertical tonus in the primary position. However, this
model would predict that pitching the head forward and backward (ie, a vertical
head-tilt test) would superimpose a physiologic tonus imbalance on the underlying
central vestibular tonus imbalance in the pitch plane and thereby alter the ampli-
tudes of an existing A or V pattern and the amplitudes of the associated hyperdevia-
tions in lateral gaze. Accordingly, the clinical practice of pitching the patient’s head
Nonneurologic Causes of Oblique Muscle Overaction 37
forward and backward to obtain strabismus field measurements in upgaze and

downgaze would augment an existing A or V pattern.
blique Muscle Overaction and Dissociated

O
Vertical Divergence
Dissociated vertical divergence may coexist with primary oblique muscle overac-
tion [60]. Dissociated vertical divergence has been attributed to a central vestibular
tonus imbalance in the roll plane induced by fluctuations of binocular visual input
[60, 61]. This hypothesis is based on physiologic studies [60, 61] in fish that show
that unequal visual input to the 2 eyes induces a reflex body tilt in the roll (frontal)
plane toward the side with greater visual input. This dorsal light reflex is a balancing
movement that uses light from the sky as a visual reference to maintain vertical
orientation by equalizing luminance input to the 2 laterally placed eyes. In a verti-
cally restrained fish, unequal visual input induces a vertical divergence of the eyes,
with depression of the eye that has greater visual input and elevation of the eye that
has lesser visual input. This vertical divergence of the eyes corresponds to the dis-
sociated vertical divergence seen in humans who fail to develop single binocular
vision secondary to early-onset strabismus.
In humans with dissociated vertical divergence, suppression or mechanical
occlusion of one eye increases upward tonus to the extraocular muscles of that eye
and downward tonus to the extraocular muscles of the opposite eye [60, 61].
Simultaneous recruitment of central vestibular innervation to both elevators in the
visually deprived eye has been invoked to explain the spontaneous overelevation in
adduction that can be observed with dissociated vertical divergence, when no V pat-
tern or baseline extorsion is present [60]. The observation that decreased visual
input increases upward tonus to one eye (in the case of dissociated vertical diver-
gence) and to both eyes (in the case of inferior oblique muscle overaction) attests to
the retention of primitive vision-induced tonus mechanisms [62, 63] in humans, and
to the atavistic resurgence of these primitive subcortical reflexes when strabismus
precludes the development of binocular vision. Our neurologic model of primary
oblique muscle overaction as a central vestibular tonus imbalance in the pitch plane
complements the recently proposed theory of dissociated vertical divergence as a
central vestibular tonus imbalance in the roll (frontal) plane, and begs the question
of whether latent nystagmus might be similarly driven by a central vestibular tonus
imbalance in the yaw (axial) plane.
Nonneurologic Causes of Oblique Muscle Overaction
Most of the mechanisms invoked to explain the existence of A and V patterns with
oblique muscle overaction have described orbital anatomical abnormalities that
could account for the abnormal movements on a biomechanical basis [64–75]. It is
beyond the scope of this article to review and critique all of them. In some patients,
neurologic and anatomical causes of oblique muscle overaction may coexist.

Children with hydrocephalus and tonic downgaze, for example, may also have fron-
tal bossing with anterior displacement of the trochlea, which can increase tension on
the superior oblique muscles and produce a mechanical superior oblique muscle
overaction [1, 29]. Recently, Clark [76] and Demer [77] and their colleagues have
used magnetic resonance imaging to demonstrate that heterotopia of extraocular
muscle pulleys within the orbits can also produce overelevation or overdepression
of the adducting eye and simulate oblique muscle overaction. Orbital pulley malpo-
sition may account for some children who have superior oblique muscle overaction
and A-pattern strabismus in the absence of neurologic disease. Since orbital ana-
tomical abnormalities can produce excessive vertical excursion of one or both eyes
in the field of action of the oblique muscles, many authorities advocate use of the
descriptive terms overelevation and overdepression of the adducting eye rather than
the diagnostic term overaction of the oblique muscles to characterize these move-
ments [1, 76].
Conclusions
Lower lateral-eyed animals use light from the sky above and gravity from the earth
below as major sources of sensory input to neuronal tonus pools within the central
vestibular system. These neuronal tonus pools calibrate extraocular muscle and pos-
tural tonus to maintain vertical orientation. In lower animals, oblique muscle tonus
is determined by luminance and gravitational input in the pitch plane. In humans,
the brain leverages visual and gravistatic sensory input to calibrate extraocular mus-
cle tonus in the pitch plane. Early loss of single binocular vision is treated by the
central vestibular system as forward pitch, necessitating increased upward tonus to
the extraocular muscles and manifesting as primary oblique muscle overaction.
Neurologic lesions within the posterior fossa can produce the opposite central ves-
tibular imbalance, in which a backward pitch evokes increased downward tonus to
the extraocular muscles and produces primary superior oblique muscle overaction.
This duality reflects an ancestral bimodal tuning of central vestibular output to the
extraocular muscles that is subordinate to binocular vision in humans.
References
1. von Noorden GK. Binocular vision and ocular motility: theory and manage-
ment of strabismus. 5th ed. St Louis: Mosby-Year Book Inc; 1996. p. 367–91.
2. Piper HF. Über die Bedeutung des V- and A-Phänomens beim Schielen. In:
Sitzungsbericht 107 Versammlung Rheinland Westfalen Augenärzte. 1963:63.
3. Piper HF. Verlagerte Muskelansätze als eine Urwsache des Schrägschielens
(und ihre operative Korrektur). In: Sitzungsbericht 109 Versammlung Rheinland
Westfalen Augenärzte. 1964:86.
4. Weiss JB. Ectopies et pseudoectopies maculaires par rotation. Bull Mem Soc Fr
Ophtalmol. 1966;79:329–49.
Conclusions 39

muscle overaction/underaction and A- and V-pattern strabismus. Binocul Vis
Eye Muscle Surg Q. 1994;9(suppl):209–35.
6. Ohm J. Das Ohrlabyrinth als Erzeuger des Schielens. Z Augenheilkd.
1916;36:253–73.
7. Ohm J. Einige Abildungen von vestibulärem Schielen. Z Augenheilkd.
1918;39:204–7.
8. Ohm J. Schrägschielen. Arch Augenheilkd. 1928;39:619–43.
9. Walls GL. The Vertebrate Eye and Its Adaptive Radiation. Bloomfield Hills:
Cranbrook Institute of Science; 1942. p. 303.
10. Traill AB, Mark RF. Optic and static contributions to ocular counter-rotation in
carp. Exp Biol. 1970;52:109–24.
11. Ewald JR. Physiologische Untersuchungen über das Endorgan des Nervus
Oktavus. Bergmann: Wiesbaden; 1892.
14. von Holst E. Über den Lichtrückenreflex bei der Fische. Pubbl Stn Zool Napoli
II. 1935;15:143–58.
15. von Holst E. Die Gleichgewichtssine der Fische. Verh Dtsch Zool Ges.

1935;37:109–14.
16. Walls GL. The evolutionary history of eye movements. Vis Res. 1962;2:69–80.
17. Simpson JI, Graf WG. Eye-muscle geometry and compensatory eye move-
ments in lateral-eyed and frontal-eyed animals. Ann N Y Acad Sci.
1981;374:20–30.
18. Zee DS. Considerations on the mechanisms of alternating skew deviation in
patients with cerebellar lesions. J Vestib Res. 1996;6:395–401.
19. Leigh RJ, Zee DS. The neurology of eye movements. 3rd ed. New York: Oxford
University Press Inc; 1999. p. 19–89.
20. Graf W, Meyer DL. Eye position in fishes suggest different modes of interaction
between commands and reflexes. J Comp Physiol. 1978;128:241–50.
21. Tusa RJ. Nystagmus: diagnostic and therapeutic strategies. Semin Ophthalmol.
1999;14:65–73.
22. Brandt T, Dieterich M. Central vestibular syndromes in roll, pitch, and yaw
planes. Neuro-Ophthalmology. 1995;15:291–303.
23. Brandt T, Dieterich M. Vestibular syndromes in the roll plane: topographic
diagnosis from brainstem to cortex. Ann Neurol. 1994;36:337–47.
24. Dieterich M, Brandt T. Wallenberg’s syndrome: lateropulsion, cyclorotation,
and subjective visual vertical in thirty-six patients. Ann Neurol.
1992;31:399–408.
25. Glasauer S, Dieterich M, Brandt T. Simulation of pathological ocular counter-
roll and skew torsion by a 3-D mathematical model. NeuroReport.
1999;10:1843–8.
26. Baloh RW, Spooner JW. Downbeat nystagmus: a type of central cerebellar nys-
tagmus. Neurology. 1981;31:304–10.
27. Straumann D, Zee DS, Solomon D. Three-dimensional kinematics of ocular

drift in humans with cerebellar atrophy. J Neurophysiol. 2000;83:1125–40.
28. Böhmer A, Straumann D. Pathomechanism of mammalian downbeat nystag-
mus due to cerebellar lesion: a simple hypothesis. Neurosci Lett.
1998;250:127–30.
29. France TD. Strabismus in hydrocephalus. Am Orthopt J. 1975;25:101–5.
30. France TD. The association of “A” pattern strabismus with hydrocephalus. In:
Moore S, Mein J, Stockbridge L, editors. Orthoptics: past, present, future:
transactions of the Third International Orthoptic Congress, Boston, July 1–3,
1975. New York: Stratton; 1976. p. 287–92.
31. Rabinowicz IM, Walker JW. Disorders of ocular motility in children with
hydrocephalus. In: Moore S, Mein J, Stockbridge L, editors. Orthoptics: past,
present, future: transactions of the Third International Orthoptic Congress,
Boston, July 1–3, 1975. New York: Stratton; 1976. p. 279–86.
32. Maloley A, Weber S, Smith DR. A and V patterns of strabismus in meningomy-
elocele. Am Orthopt J. 1977;27:115–8.
33. Gaston H. Does the spina bifida clinic need an ophthalmologist? Z Kinderchir.
1985;40(suppl 1):46–50.
34. Lennerstrand G, Gallo JE. Neuro-ophthalmological evaluation of patients with
myelomeninocele and Chiari malformations. Dev Med Child Neurol.
1990;32:415–22.
35. Hamed LM. Overaction of the superior oblique muscle: some nosologic consid-
erations. Am Orthopt J. 1993;43:82–6.
36. Hamed LM, Maria BL, Quisling RG, Mickle JP. Alternating skew on lateral
gaze: neuroanatomic pathway and relationship to superior oblique overaction.
Ophthalmology. 1993;100:281–6.
37. Biglan AW. Ophthalmological complications of meningomyelocoele: a longitu-
dinal study. Trans Am Ophthalmol Soc. 1990;88:389–462.
38. Biglan AW. Strabismus associated with meningomyelocele. J Pediatr

39. Hoyt CS, Fredrick DR. Serious neurologic disease presenting as comitant eso-
tropia. In: Rosenbaum AL, Santiago AP, editors. Clinical strabismus manage-
ment: principles and surgical techniques. Philadelphia: WB Saunders Co; 1999.
p. 152–62.
40. Keane JR. Alternating skew deviation: 47 patients. Neurology. 1985;35:
725–8.
41. Hoyt CS. Ocular motor consequences of cortical visual impairment. Paper pre-
sented at: Jampolsky Festschrift; April 10, 2000; San Francisco, CA.
42. Hoyt CS. Abnormalities of the vestibular response in congenital esotropia. Am
J Ophthalmol. 1982;93:704–8.
43. Hoyt CS, Mousel DK, Weber AA. Transient supranuclear disturbances of gaze
in healthy neonates. Am J Ophthalmol. 1980;89:708–13.
44. Doden W, Adams A. Elektronystagmographische Ergebnisse der Prüfung des
optischvestibulären Systems bei Schielenden. Ber Dtsch Ophthalmol Ges.
1957;60:316–7.
Conclusions 41
45. Salman SD, von Noorden GK. Induced vestibular nystagmus in strabismic
patients. Ann Otol Rhinol Laryngol. 1970;79:352–7.
46. Sandstedt P, Odenrick P, Lennerstrand G. Gait and postural control in children
with divergent strabismus. Binocul Vis Eye Muscle Surg Q. 1986;1:141–6.
47. Lennerstrand G. Central motor control in concomitant strabismus. Graefes
Arch Clin Exp Ophthalmol. 1988;226:172–4.
48. Baloh RW, Demer JL. Optokinetic-vestibular interaction in patients with

increased gain in the vestibulo-ocular reflex. Exp Brain Res. 1991;83:427–33.
49. Baloh RW, Richman L, Yee RD, Honrubia V. The dynamics of vertical eye
movements in normal human subjects. Aviat Space Environ Med. 1983;54:32–8.
50. Böhmer A, Baloh RWL. Vertical optokinetic nystagmus and optokinetic after
nnystagmus in humans. J Vestib Res. 1990;1:309–15.
51. Matsuo V, Cohen B. Vertical optokinetic nystagmus and vestibular nystagmus
in the monkey: up-down asymmetry and effects of gravity. Exp Brain Res.
1984;53:197–216.
52. Goltz JC, Irving EL, Steinbach MJ, Eizenman M. Vertical eye position control
in darkness: orbital position and body orientation interact to modulate drift
velocity. Vis Res. 1997;37:789–98.
53. Slavin ML, Potash SD, Rubin SE. Asymptomatic physiologic hyperdeviation in
peripheral gaze. Ophthalmology. 1988;95:778–81.
54. Liesch A, Simonsz HJ. Up- and downshoot in adduction after monocular patch-
ing in normal volunteers. Strabismus. 1993;1:25–36.
55. Neikter B. Effects of diagnostic occlusion on ocular alignment in normal sub-
jects. Strabismus. 1994;2:67–77.
56. Hering E. The theory of binocular vision (trans: Bridgeman B, Stark L).
New York: Plenum Press; 1868.
Ophthalmol. 1938;20:175–200.
58. Mays L. Has Hering been hooked? Nat Med. 1998;4:889–90.
59. Parks MM, Mitchell PR. Oblique muscle dysfunction. In: Tasman W, Jaeger
EA, editors. Duane’s clinical ophthalmology. Philadelphia: JB Lippincott;
1991. p. 1–9.
60. Brodsky MC. DVD remains a moving target! J AAPOS. 1999;3:325–7.
J Comp Physiol. 1983;150:473–81.
63. Pfeiffer W. Equilibrium orientation in goldfish. Int Rev Gen Exp Zool.

1964;1:77–111.
64. Urist MJ. The etiology of so-called A and V syndromes. Am J Ophthalmol.
1958;46:835–44.
65. Vallesca A. The A and V syndromes. Am J Ophthalmol. 1961;52:172–95.
66. Breinen GM. Vertically incomitant horizontal strabismus: the A-V syndromes.
N Y State J Med. 1961;61:2243–9.
67. Brown HW. Vertical deviations [In Symposium, Strabismus]. Trans Am Acad

Ophthalmol Otolaryngol. 1953;57:157–62.
68. Urrets-Zavalia A, Solares-Zamora J, Olmos HR. Anthropological studies on the
nature of cyclovertical squint. Br J Ophthalmol. 1961;45:578–96.
69. Fink W. The role of developmental anomalies in vertical muscle deficits. Am J
Ophthalmol. 1955;40:529–53.
70. Gobin MH. Sagittalization of the oblique muscles as possible cause for the “A,”
“V,” and “X” phenomena. Br J Ophthalmol. 1968;52:13–8.
71. Nakamura T, Awaya S, Miyake SL. Insertion anomalies of the horizontal mus-
cles and dysfunctions of the oblique muscles in the A-V patterns [in Japanese].
Nippon Ganka Gakkai Zasshi. 1991;95:698–703.
72. Postic G. Etiopathogénie des syndromes A et V. Bull Mem Soc Fr Ophtalmol.
1965;78:240–52.
73. Limón de Brown E, Monasterio FO, de Saint Martine R, Feldman MS. Estrabismo
en el sindrome de Treacher-Collins-Franceschetti. Cir Cir. 1993;60:210.
74. Locke JC. Heterotopia of the blind spot in ocular vertical muscle imbalance.
Am J Ophthalmol. 1968;65:362–74.
75. Saunders RA, Holgate RC. Rectus muscle position in V-pattern strabismus: a
study with coronal computed tomography scanning. Graefes Arch Clin Exp
Ophthalmol. 1988;226:183–6.
76. Clark RA, Miller JM, Rosenbaum AL, Demer JL. Heterotopic muscle pulleys
or oblique muscle dysfunction? J AAPOS. 1998;2:17–25.
77. Demer JL, Oh SY, Poukens V. Evidence for active control of rectus extraocular
muscle pulleys. Invest Ophthalmol Vis Sci. 2000;41:1280–90.
Postscript
The phenomenon of primary oblique muscle overaction is misunderstood by many.

Some believe that it can be ascribed to anatomic derangements of the extraocular
muscle positions and their connective tissue constraints within the orbit. These
effects are indisputable in the small subset of children with craniosynostosis and
other orbital anomalies. Others argue that “sensory” disconjugate torsion can gives
rise to apparent oblique muscle overaction with the eventual development of sec-
ondary oblique muscle contracture. However, essential infantile esotropia and pri-
mary inferior oblique overaction each arise from increased bilateral tonus to the
corresponding extraocular muscles. While secondary oblique muscle strengthening
and contracture can intensify this torsion, there is no reason to attribute primary
oblique muscle overaction to the torsion it produces.
In pediatric ophthalmology, we routinely examine versions to look for extraocu-
lar muscle overaction whereas, in neuro-ophthalmology we often limitation our
examination to ductions to look for rotational limitations. This neurodiagnostic
dichotomy reflects the reality that prenuclear activations within subcortical centers
produce innervational overaction of extraocular muscles, whereas nuclear or infra-
nuclear lesions produce underaction. Isolated overaction of the medial and oblique
Postscript 43
muscles in infantile esotropia must therefore reflect prenuclear disinhibition of cor-

responding subcortical nuclei that control these specific extraocular muscles. These
subcortical nuclei lie within the vestibular system, which receives afferent input not
only from the two labyrinths but from the two eyes as well.
In this article, Sean Donahue and I examined an early binocular visual distur-
bance (effecting visuo-vestibular system) or a posterior fossa injury (affecting the
central vestibular system) can produce the necessary prenuclear disinhibition to
explain primary oblique muscle overaction. A key concept is that bilateral oblique
muscle overaction corresponds to a visual disturbance in the pitch (fore-aft) plane.
This understanding was heavily influenced by the work of Brandt and Dieterich
showing that the bilateral correlate of tilt is pitch (elaborated in Chap. 4), as well as
by an evolutionary hypothesis by David Zee showing that pitch rotation during lat-
eral gaze would produce vestibular findings corresponding to primary oblique mus-
cle overaction (Zee DS: Considerations on the mechanisms of alternating skew
deviations in patients with cerebellar lesions. J Vestib Res 1996;6:395–401). We
used Brandt and Dieterich’s terminology anterior or posterior canal imbalance to
signify bilateral neurologic injury to central otolithic pathways corresponding to
these vertical semicircular canals. (Brandt T, Dieterich M: Central vestibular syn-
dromes in roll, pitch, and yaw planes. J Neuro-Ophthalmol 1995;15:291–303).
At the time, we did not understand that there existed a subcortical neurologic
substrate for visual input to produce an extraocular muscle tonus imbalance in
canal-based planes. Now, with the understanding of the accessory optic system as
an optokinetic system encoded in canal based planes, this subcortical mechanism
has come into focus. However, an optokinetic system would produce a dynamic
imbalance, whereas primary oblique muscle overaction stands out as a tonic imbal-
ance. It remains unclear whether altered luminance input is registered or modulated
by the human accessory optic system.
Do You Really Need Your Oblique
Muscles? 4
Adaptations and Exaptations
The human extraocular muscles have evolved to meet the needs of a dynamic,
3-dimensional visual world. Under normal conditions, the extraocular muscles are
choreographed to an ensemble of visual tracking, refixation movements, and ver-
gence modulation that assures stable binocular fixation [1]. But a fundamental
dichotomy defines the central programming of the human ocular motor plant. While
the rectus muscles produce large ocular rotations into secondary and tertiary posi-
tions of gaze, the oblique muscles evoke very limited torsional excursions of the
eyes [1]. With rare exceptions [2], large torsional eye movements cannot be gener-
ated by normal individuals in the absence of a head movement [3–7]. This disparity
is also seen with vestibular eye movements in which a horizontal or vertical head
rotation induces an ocular counterrotation that effectively stabilizes the position of
the eyes in space, but a head tilt in the roll plane evokes a static ocular counterroll
of only 10% [8]. This negligible static counterroll led Jampel [9] to conclude that
the primary role of the oblique muscles in humans is to prevent torsion. So the ques-
tion is whether the human oblique muscles retain only a vestigial function in which
they are consigned to make a nominal contribution to vertical gaze, or whether the
primary function of the human oblique muscles is to modulate torsional eye position
and to maintain perceptual stability of the visual world.
Primary Adaptations in Oblique Muscle Function
To address this basic question, one must first examine the role of the oblique mus-
cles in lower animals. The extraocular muscles originally functioned to stabilize the
eyes in space during body movements and corresponding rotations of the visual
environment. In lateral-eyed vertebrates such as fish and rabbits, the oblique mus-
cles produce torsional movements of the eyes in response to pitch movements of the
body [10, 11]. When the animal pitches forward or backward, the oblique muscles

46 4 Do You Really Need Your Oblique Muscles?
produce a partial wheel-like counterrotation of both eyes that helps to stabilize the
torsional position of the eyes in space [10, 11]. In fish, a directional shift in overhead
luminance in the sagittal plane also produces an ipsidirectional pitch movement of
the body (ie, a dorsal light reflex in the pitch plane) [11–13]. When the animal’s
body is restrained during this stimulus, this dorsal light reflex causes both eyes to
rotate torsionally so that their upper poles move in the same direction as the light
source [12, 13]. Torsional optokinetic nystagmus has also been recorded in the rab-
bit, indicating that environmental rotation in the pitch plane can directly activate the
oblique muscles [14].
The oblique muscles also contribute to ocular movements during roll (ie, rota-
tions about the head-tail axis of the animal) [15]. A body tilt evokes utricular inner-
vation to the ipsilateral superior rectus and superior oblique muscles (which are
elevators in fish and rabbits) and the contralateral inferior rectus and inferior
oblique muscles (which are depressors in fish and rabbits) [16]. The resulting
supraduction of the lower eye and infraduction of the higher eye helps to stabilize
the vertical position of the eyes during body roll. The magnitude of the ocular
counterroll relative to a body roll is only approximately 50% in lateral-eyed ani-
mals such as rabbits [17]. A similar vertical divergence can also be induced by a
rotating optokinetic cylinder rotating around the long axis of the fish [10] or by
providing unequal visual input to the 2 eyes [11, 14]. For example, increasing
visual input to the left eye of a fish by shining a light at an angle onto the top of a
fish tank produces a body tilt toward the left in the freely swimming fish (a dorsal
light reflex in the roll plane). When body roll is restrained, the same stimulus
evokes a vertical divergence of the eyes (supraduction of the right eye and infra-
duction of the left) that tends to equalize visual input to the 2 eyes [18]. These
primitive adaptations use visual and graviceptive input to set postural and extra-
ocular muscle tonus during pitch and roll [19].
Human ocular torsion can be subdivided into cyclovergence (a disconjugate
torsional rotation of the globes producing extorsion or intorsion of both eyes)
and cycloversion [1, 7] (a conjugate torsional rotation of both globes producing
intorsion of one eye and extorsion of the other eye). These 2 torsional eye move-
ments in humans correspond to the torsional eye movements in lower animals
induced by pitch and roll. Since pitch evokes a disconjugate torsional rotation
(ie, either intorsion or extorsion of both eyes) in lateral-eyed animals, phyloge-
netic retention of this primitive adaptation in humans would mean that a pitch
stimulus (a slant of the visual environment around the interaural axis) would
evoke a cyclovergence response (a disconjugate torsional movement of both
eyes) in humans, whereas a roll stimulus (a tilt of the head or the visual environ-
ment around the nasooccipital axis) would evoke a cycloversion response in
humans. These primitive adaptations are indeed measurable in the laboratory as
the small cyclovergence movements that are induced artificially by haploscopy or
optically induced cyclodisparity [6, 20–24] and in the small cycloversion move-
ments that are evoked by head tilt (ie, the human ocular counterroll to a gravicep-
tive stimulus) [16], by torsional optokinetic stimuli [25–27], or by static-tilted
visual stimuli [28, 29].
Oblique Muscle Exaptations 47
Oblique Muscle Exaptations
From Visual Panorama to Frontal Binocular Vision
Although we retain our primitive adaptations, the function of the human oblique
muscles has evolved to meet the needs of single binocular vision. In the course of
evolution, primitive adaptations give way to exaptations. An adaptation is some-
thing fit (aptus) by construction for (ad) its usage [30]. Exaptation is a relatively
new evolutionary concept advanced by Gould [30] to describe a feature, now useful
to an organism, that did not arise as an adaptation for its present role, but that was
subsequently co-opted for its current function. Such structures are fit (aptus) not by
explicit molding for (ad) current use, but as a consequence of (ex) properties built
for other reasons [30]. According to this definition, a mechanism must have a func-
tion and must enhance the fitness of its bearer to qualify as an exaptation [30, 31].
For example, the feathers of birds may have originally evolved for thermal insula-
tion (an adaptation), only to be subsequently co-opted for flight (an exaptation)
[31, 32].
Cyclovergence, Stereoscopic Perception, and the Pitch Plane
According to Blakemore et al [32], binocular animals have abandoned the enor-
mous biologic advantage of panoramic vision in order to have their eyes pointing
forward, the most obvious advantage of which is stereopsis. Frontal repositioning of
the eyes seems to have exapted the oblique muscles to subserve stereopsis. Evolution
has grafted a new torsional control system that is subordinate to binocular vision on
top of the “primitive” dynamic torsional programming of the oblique muscles.
Although the brain programs eye torsional position by regulating the tonus of all
extraocular muscles, the oblique muscles have the predominant effect on ocular tor-
sion. It is therefore instructive to examine torsional eye position as a function of
oblique muscle innervation.
How do the human oblique muscles subserve stereopsis? Under conditions of
binocular fixation, an object closer in space than the fixation point will produce an
image on the temporal retinas, while an object farther in space than the fixation
point will produce an image on the nasal retinas [6]. This horizontal disparity forms
the basis for stereoscopic perception. If one examines the circles that appear ele-
vated on a Titmus stereoacuity test under binocular conditions, examination with
each eye will show a nasal displacement of the circle in space, indicating that the
image falls on the temporal retina in each eye when the circle is viewed binocularly.
When the Titmus test is turned upside down so that the monocular image falls on the
nasal retinas of each eye, the circles appear to lie behind the plane of the page.
Now consider a binocular individual with normal stereopsis who is fixating on
the center of a vertical object that is slanted so that its inferior aspect is closer than
the superior aspect (Fig. 4.1). As the individual fixates the center of the slanted
object, the visual image of the upper pole is postfixational, which means that it falls
onto the nasal retina of each eye, whereas the visual image of the lower pole is pre-
fixational, falling onto the temporal retina of each eye. The reader can appreciate
this slant illusion by holding a pencil in the midsagittal plane with the upper pole
slanted away from the body and the lower pole tilted toward the body. On occlusion
of either eye, the upper pole of the pencil will appear to be tilted, with the upper pole
leaning toward the side of the uncovered eye (Fig. 4.1) [33]. So under monocular
conditions, the person perceives a disconjugate image torsion that is analogous to
how the image would be seen if there were intorsion of each eye.
If a vertical cyclodisparity in the 2 eyes is translated by the visual cortex into a
binocular sensation of depth in the pitch plane (ie, slant), can retinal image torsion
cause a vertical binocular image to be perceived as slanted in the pitch plane? The
Left Eye Right Eye
Fig. 4.1 Tilt is the monocular correlate of stereoscopic slant. (a) An individual binocularly view-
ing a vertical object that is slanted in the pitch plane. (b) The monocular images corresponding to
the object are extorted when viewed with each eye
answer is yes. The reader can appreciate this phenomenon by placing a white
Maddox Rods over each of the 2 eyes in a trial frame, and looking toward a bright
focal light source with the grids oriented horizontally to produce a vertical line
(Fig. 4.2). Now counterrotate the lenses so that their upper pole of each line moves
nasally and their lower pole moves temporally until the image of the binocular verti-
cal line breaks into 2 tilted lines (Fig. 4.2). If the rotation is stopped at the break
point, the torsional diplopia can be overcome and the 2 lines can be fused. When
cyclofusion occurs, which is almost purely on a sensory (as opposed to a motor)
Right Eye Left Eye
Fig. 4.2 Binocular cyclodisparity of vertical lines is perceived stereoscopically as slant. (a)
Rotation of horizontal double Maddox Rods to produce binocular image intorsion (as would be
seen if both eyes were extorted). (b) Sensory cyclofusion causes the patient to stereoscopically
perceive a vertical line (solid line) as slanted in the pitch (ie, sagittal) plane (dashed line)
basis, the single line will suddenly appear to be stereoscopically slanted in the pitch
plane, with the upper pole inclined toward the observer and the lower pole inclined
away from the observer. The opposite inclination is seen when the upper poles of the
Maddox Rods are rotated temporally. The original treatise by Wheatstone [34]
describing his invention of the stereoscope in 1838 provided the first example of
pitch stereopsis produced by dichoptic lines that are tilted in different directions.
This stereoscopic effect shows us something remarkable about the normal bin-
ocular visual system. It tells us that when an isolated vertical image cyclodisparity
falls within the physiologic range of sensory fusion, it is misregistered stereoscopi-
cally as a slant of the vertical object in the pitch plane [7, 21, 22, 32–38]. In a real
world setting, however, perceived pitch is not solely a function of retinal cyclodis-
parity, but it depends both on the brain’s computation of registered eye rotation and
on retinal cyclodisparity. So to create an accurate stereoscopic representation of
vertical objects in the pitch plane, cyclovergence should not occur and the eyes must
be locked into a well-defined static orientation relative to a given gaze position (ie,
conforming to Donder’s law) [1, 7, 21, 22]. The innervational patterns of oblique
muscle recruitment, which counteract the torsional actions of the rectus muscles in
different positions of gaze, must also be subordinate to this goal.
That binocular torsional control represents an active function of the human
oblique muscles rather than an evolutionary loss of contractile function is seen in
the kinematics of human convergence [39]. It has long been recognized that both
eyes extort during convergence and that this extorsion increases in downgaze and
decreases in upgaze [40, 41]. (Extorsion is even considered by some to be a compo-
nent of the synkinetic near reflex [42].) While the existence of these cyclovergence
movements were once thought to constitute a violation of Listing’s law, they can be
reconciled with Listing’s law if it is assumed that convergence is associated with a
temporal rotation of Listing’s plane in each eye (Fig. 4.3) [43]. Vertical rotation of
the eyes around these temporally rotated axes produces incyclovergence of the eyes
in upgaze and excyclovergence in downgaze [44–46]. Several lines of evidence sug-
gest a neural and biomechanical basis for these cyclovergence movements. In mon-
keys, Mays et al [47] measured single cell recordings within the trochlear nucleus
and found decreased unit activity during convergence. This decrease in firing rate
was greater when the monkey converged in downgaze than in upgaze, a finding that
corresponds to the observed convergence-associated torsional movements in
humans. More recently, dynamic magnetic resonance imaging by Demer et al [48,
49] have found that downward rotation of the lateral rectus muscle pulley and
medial rotation of the inferior rectus pulley during convergence, indicating that the
inferior oblique muscle may also play a role in convergence-associated torsion, pre-
sumably via its collagenous attachments to the lateral rectus and inferior rectus
muscles.
From an evolutionary perspective, it is worth examining whether these torsional
movements during convergence simply represent primitive adaptations that have
been phylogenetically retained, like the small ocular counterroll which has no
known function in humans [17]. In the lateral-eyed animal, upgaze corresponds to
an intorsional movement of both eyes when the rotation is viewed from the frontal
D D
LP LP
U U
D D
U U
D D
LP LP
U U
Fig. 4.3 View of both eyes from above demonstrating orientation of Listing’s plane (LP) during
distance fixation (a) and convergence (b). Curved arrows denote cyclovergence movements of the
eyes associated with vertical rotation about horizontal visual axes in Listing’s plane. A “saloon
door” rotation of Listing’s plane, which is opposite in direction to the ocular rotation, can be used
to reconcile the convergence-associated extorsion of the eyes in downgaze (D) and intorsion of the
eyes in upgaze (U) with Listing’s law
perspective, while downgaze corresponds to an extorsional rotation of both eyes.

One could argue that convergence in humans may simply stress the system to a
degree that prevents binocular suppression of these primitive rotations. However,
recent studies suggest that the frontal binocular visual system has latched on to this
primitive torsional bias and exapted it to subserve stereoscopic perception in the
pitch plane [50, 51]. To subserve stereopsis, the oblique muscles have been exapted
to torsionally align the eyes with their corresponding visual images in a way that
preserves the binocular horizontal disparities that produce stereoscopic perception
in different positions of vertical gaze [51]. This exaptation serves to minimize the
brain’s computational load for stereoscopic perception [51].
Human cyclovergence is most robust at near fixation, where it plays an active
role in stereoscopic vision. If convergence were not linked to cyclovergence, sym-
metrical convergence on a frontoparallel plane would induce incyclodisparity of the
horizontal images in upgaze and excyclodisparity of the horizontal images in down-
gaze for each eye, solely on the basis of the geometric angle from which each eye
views a planar surface. (The opposite cyclodisparity bias occurs for vertical visual
landmarks due to projection geometry; however, it is reduced or reversed by the
horizontal retinal shear that was described by Helmholtz [40].) In convergence, the
increased intorsion of the eyes in upgaze and extorsion in downgaze helps to tor-
sionally align the horizontal meridians of the eyes with their respective horizontal
visual landmarks, thereby facilitating stereopsis. Since convergence is generally
used for downgaze, where near objects are situated, the innervational link between
convergence and extorsion presumably serves to set the operational position for
stereopsis as slightly in downgaze, where near objects can be held by the arms and
illuminated by overhead light [40, 41, 52]. Although the torsional movements asso-
ciated with convergence are preprogrammed [17], they exhibit remarkable plasticity
[53] and are enhanced by the depth perception of stereograms [54], demonstrating
that they also rely on visual input to more accurately subserve the needs of binocular
vision and depth perception. Without these torsional movements, convergence dur-
ing vertical gaze would limit optimal stereoscopic perception to 1 gaze elevation,
requiring repositioning of the head to optimize depth perception of targets at differ-
ent earth elevations. Thus, this oblique muscle exaptation provides the luxury-
optimizing stereopsis for targets at different eye elevations without the necessity of
head movements in the pitch plane.
Almost 30 years ago, Blakemore et al [32] recorded action potentials from bin-
ocular neurons in the cat’s visual cortex and measured orientation selectivity during
simultaneous binocular stimulation. Certain binocular cells responded specifically
to objects tilted in 3-dimensional space toward the cat or away from it [32]. Such
binocular cells may form at least part of the substrate for sensory cyclofusion in
humans. The survival value of stereoscopic spatial orientation explains why cyclo-
fusional movements are so limited in primary gaze and why sensory cyclofusion is
so well developed [3–7, 55, 56]. Sensory cyclofusion without motor cyclofusion is
a prerequisite for pitch stereopsis (ie, slant perception). Motor cyclofusion of these
torsionally disparate vertical landmarks would induce a misperception of stereo-
scopic slant for vertical lines. For any position of gaze, the oblique muscles must
torsionally anchor the eyes to produce a stable motor substrate for slant perception
of vertical objects.
Psychophysical experiments have shown that horizontal visual landmarks are
selectively used in humans to lock in torsional eye position, although the neural
feedback loops for this process are unknown. As early as 1861, Nagel [35] observed
that the rotation of horizontal fusion contours produced cyclovergence, while the
rotation of vertical contours produced only a stereoscopic effect [28]. Ogle and
Ellerbrock [37] noted that cyclofusion of torsionally disparate horizontal lines that
were presented dichoptically with no visual background caused a previously fused
vertical line to pitch in the sagittal plane. According to Bradshaw and Rogers [57],
cyclovergence is not well driven by disparities along vertical meridians even when
these are created by a real inclined surface. By inducing cyclodisparity of horizon-
tal lines to the 2 eyes dichoptically, small cyclofusional eye movements can be
elicited in humans under experimental conditions [7, 20, 23, 24, 37, 38, 56]. The
small size of these cyclofusional movements suggests that the human cyclover-
gence system is equipped to provide a fine-motor modulation to a system that is
designed primarily for stability rather than movement [7]. The greater stereoscopic
value of vertically compared than horizontally tilted images explains why verti-
cally oriented gratings evoke smaller cyclovergence movements than horizontally
oriented gratings [23].
Humans inhabit a terrestrial environment composed of primarily vertical and
horizontal landmarks that serve as reference points for vertical orientation [21, 22].
In a terrestrial setting, the most prominent horizontal contour is the horizon, which
may be the main visual reference for stabilizing the eyes relative to the outside
world [21]. While cyclodisparities of vertical contours may be caused by slant of the
observed objects, cyclodisparities of horizontal contours indicate cyclovergence
errors that need correction [20, 21]. As summarized by Howard and Rogers [33]:
An orientation disparity between the images of lines in the horizontal plane of regard can
be due only to eye misalignment, whereas an orientation disparity from a vertical line may
be due to inclination of the line in depth. It would therefore be adaptive if cyclovergence
were evoked only by disparities in horizontal elements, leaving residual disparities in verti-
cal elements intact as clues for inclination.
If cyclodisparities in horizontal visual landmarks of the visual scene that occur at

low stimulus frequencies and low amplitudes serve as the physiologic stimulus for
cyclovergence as proposed by Howard and Zacher [58], then horizontal cyclofusion
must serve to torsionally anchor the eyes, allowing the visual cortex to extract and
construct a stable and reproducible representation for pitch stereopsis [7]. By using
cyclodisparity of horizontally oriented lines as the feedback signal for torsional
misalignment of the eyes and allowing cyclodisparity of vertically oriented lines to
signal depth (ie, slant), the brain can recruit the oblique muscles to control cyclover-
gence and thereby assure accurate stereoscopic perception of vertical objects in the
pitch plane. While a cyclodisparity of horizontal lines are the driving force for this
cyclofusional reflex, stereopsis and vertical visual orientation are its emergent
functions.
If a binocular vertical cyclodisparity produces a stereoscopic sensation of pitch,

then ocular torsion produced by cyclovertical muscle palsy should produce an
abnormal sensation of pitch stereopsis. Not surprisingly, abnormal pitch stereopsis
is a common symptom of acquired superior oblique palsy. Lindblom et al [59] found
that adults with acquired unilateral or bilateral superior oblique palsy perceived the
upper pole of a vertical rod as being tilted toward them in the sagittal plane under
binocular conditions. This stereoscopic illusion corresponded to the associated
extorsion of the paretic eye (Fig. 4.4). Subjects also perceived the unfused image
corresponding to the eye with the superior oblique palsy as intorted (ie, tilted in the
roll plane) relative to the other image.
As seen in superior oblique palsy, the concept of Panum’s space can be extrapo-
lated to torsional eye position. Ocular torsion within the realm of fusion induces an
illusory pitch stereopsis of isolated vertical lines, while torsion outside the realm of
fusion induces torsional diplopia in the roll plane [5, 60, 61]. Ocular torsion of 5°,
as generally occurs with unilateral superior oblique palsy, is not an impediment to
fusion, whereas ocular torsion of more than 10°, which accompanies bilateral supe-
rior oblique palsy, precludes fusion [6]. In unilateral superior oblique palsy, it is
often stated that strabismus surgery or prismatic correction to vertically realign the
eyes is sufficient to restore cyclofusion, even when the extorsion persists in the pal-
sied eye. However, the persistence of extorsion in one eye is not without perceptual
consequence, and it should be remembered that sensory fusion of torsional images
can cause vertical objects to be perceived as slanted. Psychophysical experiments
by Howard and Kaneko [62] have shown that an isolated shear disparity of vertical
lines will induce a stereoscopic slant, whereas a cyclodisparity that twists both ver-
tical and horizontal lines will not induce a perceived slant of the visual environment.
These experiments would predict that patients with unilateral superior oblique palsy
and extorsion of 1 eye would stereoscopically perceive isolated visual landmarks in
the sagittal plane as slanted toward them.
Cycloversion, Nonstereoscopic Perception, and the Roll Plane
Unlike cyclovergence, which is remarkably stable and seems to depend primarily on

where the eyes are looking, human cycloversion shows both intrasubject and inter-
subject variation [39]. These findings implicate different neural control strategies
for cycloversion and cyclovergence [39]. One explanation for this disparity is that
cycloversion is probably not as important to stereoscopic vision as cyclovergence,
which determines stereoscopic volume at any given pitch plane, alters slant percep-
tion of vertical objects, and is necessary for stereo constancy [39].
Nevertheless, large cycloversional movements of the eyes create a problem for
stereoscopic perception. Misslisch et al [17] have argued that the superimposition of
a primitive cycloversion movement of the eyes (such as an ocular counterroll evoked
by a head tilt) on convergence would induce a cyclodisparity and disrupt stereoacu-
ity. The brain strikes a balance between gyroscopic and stereoptic mechanisms by
damping the ocular counterroll by approximately 70% in convergence [18]. In this
Right Eye Left Eye
Fig. 4.4 Right superior oblique palsy. (a) Examiner’s view of patient’s retinas showing extorsion
of the right eye. (b) Under binocular conditions, the patient perceives a vertical object (solid line)
as stereoscopically slanted in the pitch plane (dashed line)
way, exaptations of the neural circuitry that steers our cyclovergence and cyclover-
sion movements seem to override our primitive adaptations to promote stereopsis
[39, 50].
While the human oblique muscles function under static conditions to constrain
torsional rotation of the eyes, it is not the physiologic role of any muscle to simply
constrain movement (check ligaments and muscle pulleys are better suited to this
function). To the surprise of many, Tweed et al [63] have recently found that the
human oblique muscles execute large cycloversional saccades immediately
preceding head movements in the roll plane. The 3-dimensional scleral search coil
recordings were performed as normal study participants observed a laser spot while
their eyes were directed 20° downward. The subjects then made combined eye-head
movements to refixate the laser spot as it jumped 20° (from right head tilt and right
gaze to left head tilt and left gaze). Eye movement recordings showed that these
subjects generated ipsiversive torsional saccades that ranged in size from 11° to 17°
and averaged 14.5°. These cycloversional eye movements preceded the head move-
ments by 20 to 60 milliseconds, indicating that these movements were not vestibular
in origin. The eyes arrived at the target first and locked on, hanging in space as the
head rotated around them (Fig. 4.5). When the head came to a halt, the ocular tor-
sion relative to the head had stabilized near zero, and the eyes were poised for the
swiftest possible response to further movement of the target. These torsional eye
movements, which occur at the initiation of a head tilt and are not visible on gross
inspection, may reveal another exaptation of our torsional control system. The
human oblique muscles may have been exapted to generate saccadic torsional eye
movements to reestablish roll plane orientation in anticipation of a postural rotation
in the roll plane. These anticipatory saccades instantaneously recalibrate torsional
eye position to provide a stable visual representation of tilt in the roll (frontal) plane.
The evolution of frontally positioned eyes for stereopsis may have created a survival
advantage for the grafting of this new torsional control system on top of the ances-
tral control system that produces the ocular counterroll (Table 4.1). Similar tor-
sional saccades have not been observed in lateraleyed animals (although studies
involving eye-head or eye-body coordination in animals are extremely difficult to
perform).
Reversion From Exaptation to Adaptation
Since ocular torsion within the realm of fusion can produce a stereoscopic tilt in
the pitch plane, it seems reasonable to ask whether strabismus or other neuro-
logic disease, which can be associated with a pathologic tilt in the internal
1 2 3 4 1
Fig. 4.5 Depiction of torsional eye position during head tilt from side to side. (1) Initial torsional
position during right head tilt. (2) During head tilt to the left, the eyes lead the head and quickly
assume their final torsional position corresponding to the left head tilt. (3) The eyes “hang in
space” until the head catches up. (4) Head tilt to the right produces the reverse sequence of tor-
sional eye movements. (Reprinted with permission from The American Association for the
Advancement of Science, copyright 1999 [63].)
Table 4.1 Oblique Muscle Adaptations and Exaptations

Adaptations Adaptations Exaptations
(Lateral-Eyed Animals) (Humans) (Humans)
Cyclovergence
Pitch plane
Dorsal light reflex Primary oblique muscle overaction Fusional cyclovergence
(for pitch stereopsis)
Ocular counterroll Alternating skew deviation
(torsional)
Cycloversion
Roll plane
Dorsal light reflex Dissociated vertical divergence Dissociated vertical
(vertical component) divergence (torsional
Cycloversion to torsional component)
optokinetic targets, static visual tilt,
binocular vertical disparity
Ocular counterroll (vertical) Ocular counterroll (torsional), skew ? Anticipatory torsional
deviation saccades
representation of the gravitational vertical, could recalibrate prenuclear innerva-

tion to the extraocular muscles and produce a torsional deviation of the eyes that
conforms to this internal shift. Again, the answer is yes. One of the primitive
functions of the human oblique muscles is to rotate the eyes toward the subjec-
tive visual vertical; when this internal representation is altered under pathologic
conditions, ocular torsion is the inevitable result. In humans, as in lower animals,
the central vestibular system uses weighted input from the 2 labyrinths and
weighted visual input from the 2 eyes to establish subjective vertical orientation
in pitch and roll [64, 65]. In humans, cycloversion is evoked by visual or gravi-
ceptive imbalance in the roll plane [64] whereas cyclovergence is evoked by a
visual or graviceptive imbalance in the pitch plane [65]. In the roll plane, unilat-
eral loss of otolithic tone secondary to brainstem, cerebellar, or utricular injury
causes skew deviation, whereas asymmetrical visual input in humans with con-
genital strabismus evokes dissociated vertical divergence (Table) [65]. In skew
deviation, unequal graviceptive tone from the otoliths produces a pathologic tilt
in the internal representation of the visual vertical in the roll plane, which is
associated with a corresponding torsional repositioning of both eyes and a verti-
cal divergence of the eyes (Table 4.1) [66]. In dissociated vertical divergence, a
dorsal light reflex in the roll plane is also associated with visually induced tilt in
the subjective visual vertical, and the cycloversional component is ipsidirectional
to the patient’s perceived visual tilt [67]. Since the cycloversional component of
dissociated vertical divergence does not accompany the dorsal light reflex in
lateral-eyed animals, and it cannot be accounted for by anatomical repositioning
of the extraocular muscles, this component of the human dorsal light reflex seems
to represent an exaptation to restore vertical orientation during monocular view-
ing [67]. The associated cycloversion movement that occurs when humans fuse
vertically disparate images [68, 69] may indicate that binocular vertical disparity
in humans is similarly misregistered by the brain as tilt [70].
In the pitch plane, these same primitive adaptations are operative. Donahue and
I [65] have proposed that primary oblique muscle overaction is associated with a
slant of the internal representation of the visual vertical in the pitch plane. A sub-
jective inclination of the superior portion of the visual environment toward the
individual would produce a corresponding extorsion of both globes and lead to
inferior oblique muscle overaction [65]. Conversely, structural neurologic disease
within the brain stem or cerebellum would produce the intorsion and superior
oblique muscle overaction so commonly seen in children with Chiari malforma-
tions, meningomyelocele, or hydrocephalus [65]. The alternating skew deviation
on lateral gaze with bilateral abducting hypertropia that is associated with cranio-
cervical disease may represent another central vestibular disturbance in the pitch
plane [71, 72].
Under pathological conditions, the oblique muscles still function to keep the
eyes in binocular torsional register with the perceived visual environment, and an
altered torsional position of the eyes constitutes an ocular motor recalibration to the
tilted or slanted internal representation of the visual world that characterizes central
vestibular disease. By recognizing that a subjective tilt of the visual environment
evokes a corrective torsional repositioning of both eyes, we can begin to place the
horse before the cart in understanding congenital strabismus.
Conclusions
To understand why you really need your oblique muscles, it is necessary to distin-
guish primitive adaptations, which originally evolved to stabilize laterally placed
eyes during body pitch and roll, from exaptations, which subsequently evolved to
meet the needs of frontal binocular vision. The human oblique muscles have been
exapted to override primitive torsional adaptations with newer mechanisms that
subserve stereopsis. These exaptations govern the relative torsional alignment of the
eyes in different positions of gaze. Since perception of stereoscopic slant is a func-
tion of torsional eye position, the human oblique muscles modulate cyclovergence
to establish a stable stereoscopic pitch representation of the visual world. In the roll
plane, the human oblique muscles generate a cycloversion movement of the eyes
just prior to volitional head movement to lock in a stable visual perception of tilt.
These exaptations provide spatial accuracy and temporal continuity to our stereo-
scopic visual perception of slant and our nonstereoscopic visual perception of tilt.
In a larger sense, they furnish us with a multifaceted torsional control system that
provides 3-dimensional stability to the visual world and thereby improves fitness.
Exaptations in human oblique muscle function do not completely erase more
primitive adaptations. These primitive adaptations produce the small vestigial tor-
sional eye movements that can be measured experimentally by inducing pitch or tilt
of the external visual environment. They manifest clinically when congenital stra-
bismus or other central vestibular disease alters our internal representation of the
visual vertical. One may conclude that only our primitive adaptations are vestigial;
our oblique muscles are not.
Conclusions 59
References
1. Van Rijn LJ. Torsional eye movements in humans [master’s thesis]. Rotterdam:
Universiteitsdrukkerij Erasmus Universiteit Rotterdam; 1994. p. 1–9.
2. Balliet R, Nakayama K. Training of voluntary torsion. Invest Ophthalmol Vis
Sci. 1978;17:303–14.
3. von Noorden GK. Clinical observations in cyclodeviations. Ophthalmology.
1979;86:1451–60.
4. Guyton DL, von Noorden GK. Sensory adaptations to cyclodeviations. In:
Proceedings of the third meeting of the International Strabismological
Association, May 10–12. New York: Grune & Stratton; 1978. p. 1978.
5. Guyton DL. Ocular torsion: sensorimotor principles. Graefes Arch Clin Exp
Ophthalmol. 1988;226:241–5.
ment of strabismus. 5th ed. St Louis: CV Mosby; 1996. p. 8–84.
7. Van Rijn LJ, Van der Steen J, Collewijn H. Instability of ocular torsion during
fixation: cyclovergence is more stable than cycloversion. Vis Res.
1994;34:1077–87.
8. Collewijn H, Van der Steen J, Ferman L, Jansen TC. Human ocular counterroll:
assessment of static and dynamic properties from electromagnetic scleral coil
recordings. Exp Brain Res. 1985;59:185–96.
9. Jampel RS. Ocular torsion and the primary retinal meridian. Am J Ophthalmol.
1981;91:14–24.
10. Walls GL. The vertebrate eye and its adaptive radiation. Bloomfield Hills:
Cranbrook Institute of Science; 1942. p. 301–4.
11. Graf W, Meyer DL. Eye positions in fishes suggest different modes of interac-
tion between commands and reflexes. J Comp Physiol. 1978;128:241–50.
12. von Holst E. Über den Lichtrückenreflex bei der Fische. Pubbl Stn Zool Napoli
II. 1935;15:143–58.

1935;37:109–14.
14. Collewijn H, Noorduin H. Vertical and torsional optokinetic eye movements in
the rabbit. Pflügers Archiv. 1972;335:87–95.
15. Leigh RJ, Zee DS. The neurology of eye movements. 3rd ed. New York: Oxford
University Press; 1999. p. 22.
16. Simpson JT, Graf W. Eye-muscle geometry and compensatory eye movements
in lateral-eyed and frontal-eyed animals. Ann N Y Acad Sci. 1981;374:20–30.
17. Misslisch H, Tweed D, Hess BJM. Stereopsis outweighs gravity in the control
of the eyes. J Neurosci. 2001;21:RC126, 1–5.
18. Graf W, Meyer DL. Central mechanisms counteract visually-induced tonus
J Comp Physiol. 1983;150:473–81.
20. Crone RA, Everhard-Halm Y. Optically induced eye torsion, I: fusional cyclo-
vergence. Albrecht Von Graefes Arch Klin Exp Ophthalmol. 1975;195:231–9.
21. Van Rijn LJ, van der Steen J, Collewijn H. Eye torsion elicited by oscillating
gratings: effects of orientation, wavelength, and stationary contours. Vis Res.
1994;34:533–40.
22. Collewijn H, Van der Steen J, Van Rijn LJ. Binocular movements and depth
perception. In: Gorea A, Fregnac Y, Kopoula Z, Findlay J, editors.
Representations of vision. Trends and tacit assumptions in vision research.
Cambridge, UK: Cambridge University Press; 1991. p. 165–83.
23. Howard IP, Sun L, Shen X. Cycloversion and cyclovergence: the effects of the
area and position of the visual display. Exp Brain Res. 1994;100:509–14.
muscle overaction/underaction and A-and V-pattern strabismus. Binocul Vis
Eye Muscle Surg Q. 1994;9:209–36.
25. Brecher GA. Die optokinetische Auslösung von Augenrollung und rotato-
rischem Nystagmus. Pflügers Arch Gesamte Physiol Menschen Tiere.
1934;234:13–28.
26. Cheung BS, Howard IP. Optokinetic torsion: dynamics and relation to circular-
vection. Vis Res. 1991;31:1327–35.
27. Wade NJ, Swanston MT, Howard IP, Ono H, Shen X. Induced rotary motion
and ocular torsion. Vis Res. 1991;31:1979–83.
28. Crone RA. Optically induced eye torsion, II: optostatic and optokinetic cyclo-
version. Albrecht Von Graefes Arch Klin Exp Ophthalmol. 1975;196:1–7.
29. Goodenough DR, Sigman E, Oltman PK, Rosso J, Mertz H. Eye torsion in
response to a tilted visual stimulus. Vis Res. 1979;19:1177–9.
30. Gould SJ. Exaptation: a crucial tool for evolutionary psychology. J Soc Issues.
1991;47:43–65.
31.
Buss DM. Adaptations, exaptations, and spandrels. Am Psychol.
1998;53:533–48.
32. Blakemore C, Fiorentini A, Maffei L. A second neural mechanism of binocular
discrimination. J Physiol. 1972;226:725–49.
33. Howard IP, Rogers BJ. Binocular vision and stereopsis. New York: Oxford
University Press; 1995. p. 381–426.
34. Wheatstone C. Contributions to the physiology of vision. Philos Trans R Soc
Lond. 1838;128:371–94.
35. Nagel A. Das Sehen mit zwei Augen: Leipzig und Heidelberg 1861: Ref: Nagel
A: Über das Vorkommen von wahren Rollungen des Auges um die Gesichtalinie.
Albrecht Von Graefes Arch Ophthalmol. 1868;14:228–46.
36. Enright JT. Stereopsis, cyclotorsional “noise” and the apparent vertical. Vis
Res. 1990;30:1487–97.
37. Ogle KN, Ellerbrock VJ. Cyclofusional movements. Arch Ophthalmol.

1946;36:700–35.
38. Sullivan MJ, Kertesz AE. Peripheral stimulation and human cyclofusional
response. Invest Ophthalmol Vis Sci. 1979;18:1287–90.
39. Porrill J, Ivins JP, Frisby JP. The variation of torsion with vergence and eleva-
tion. Vis Res. 1999;39:3934–50.
Conclusions 61
40. von Helmholtz H. Handbuch fur der Physiologischen Optik. Hamburg:

Voss; 1867.
41. Hering E. The theory of binocular vision (trans: Bridgeman B, Stark L).
New York: Plenum Press; 1977. p. 129–45.
42. Allen MJ, Carter JH. The torsion component of the near reflex. Am J Optom.
1967;44:343–9.
43. Mok D, Ro A, Cadera W, Crawford JD, Vilis T. Rotation of Listing’s plane dur-
ing vergence. Vis Res. 1992;32:2055–64.
44. Vilis T. Physiology of three-dimensional eye movements: saccades and ver-
gence. In: Fetter M, Misslisch H, Tweed D, editors. Three-dimensional kine-
matics of eye, head, and limb movements. Amsterdam: Harwood Academic
Press; 1997. p. 56–72.
45. Mikhael S, Nicolle D, Vilis T. Rotation of Listing’s plane by horizontal, verti-
cal, and oblique prism-induced vergence. Vis Res. 1995;35:3243–54.
46. Van den Berg AV, Van Rijn LJ, de Faber J-THN. Excess cyclovergence in
patients with intermittent exotropia. Vis Res. 1995;35:3265–78.
47. Mays LE, Zhang Y, Thorstad MH, Gamlin PDR. Trochlear unit activity during
ocular convergence. J Neurophysiol. 1991;65:1484–91.
48. Demer JL, Oh SY, Poukens V. The orbital layers of human and monkey oblique
extraocular muscles (EOMs) insert on the orbital connective tissue system
[ARVO abstract]. Invest Ophthalmol Vis Sci. 2001;42:S517.
49. Demer JL. Mechanical interactions of oblique extraocular muscles (EOMs)
with actively controlled rectus pulleys maintain kinematics of linear oculomo-
tor plant. Soc Neurosci. In press.
50. Tweed D. Visual-motor optimization in binocular control. Vis Res.

1997;37:1939–51.
51. Schreiber K, Crawford JD, Fetter M, Tweed D. The motor side of depth vision.
Nature. 2001;410:819–22.
52. Nakayama K. Kinematics of normal and strabismic eyes. In: Schor CM,

Ciuffreda KJ, editors. Vergence eye movements: basic and clinical aspects.
Boston: Butterworths; 1983. p. 199–295.
53. Schor CM, Maxwell JS, Graf EW. Plasticity of convergence-dependent varia-
tions of cyclovergence with vertical gaze. Vis Res. 2001;41:3353–69.
54. Kapoula Z, Bernotas M, Haslwanter T. Listing’s plane rotation with conver-
gence: role of disparity, accommodation, and depth perception. Exp Brain Res.
1999;126:175–86.
55. Jampel RS, Stearns AB, Bugola J. Cyclophoria or cyclovergence: illusion or
reality? In: Moore S, Mein J, Stockbridge L, editors. Orthoptics: past, present,
future. New York: Stratton Intercontinental Medical Book Corp; 1976. p. 403–8.
56. Crone RA. Human cyclofusional response. Vis Res. 1971;11:1357–8.
57. Bradshaw MF, Rogers BJ. Is cyclovergence state affected by the inclination of
stereoscopic surfaces [ARVO abstract]? Invest Ophthalmol Vis Sci.
1994;35:1316.
58. Howard IP, Zacher JE. Human cyclovergence as a function of stimulus fre-
quency and amplitude. Exp Brain Res. 1991;85:445–50.
59. Lindblom B, Westheimer G, Hoyt WF. Torsional diplopia and its perceptual
consequences. Neuroophthalmology. 1997;18:105–10.
60. Kertesz AE, Jones RW. Human cyclofusional response. Vis Res. 1970;10:891–6.
61. Angio G. A vertical horopter. Opt Acta (London). 1974;21:277–92.
62. Howard IP, Kaneko H. Relative shear disparities and the perception of surface
inclination. Vis Res. 1994;34:2505–17.
63. Tweed D, Haslwanter T, Fetter M. Optimizing gaze control in three dimensions.
Science. 1998;281:1363–6.
65. Brodsky MC, Donahue SP. Primary oblique muscle overaction: the brain throws
a wild pitch. Arch Ophthalmol. 2001;119:1307–14.
66. Brandt T, Dieterich M. Skew deviation with ocular torsion: a vestibular brain-
stem sign of topographic diagnostic value. Ann Neurol. 1993;33:528–33.
67. Brodsky MC. Subjective correlates of dissociated vertical divergence. Paper
presented at: 27th Annual Meeting of the European Strabismologic Association;
June 8, 2001; Florence, Italy.
68. Van Rijn LJ, Simonsz HJ, ten Tusscher MPM. Dissociated vertical deviation
and eye torsion: relation to disparity-induced vertical divergence. Strabismus.
1997;5:13–20.
69. Cheeseman EW, Guyton DL. Vertical fusional vergence: the key to dissociated
vertical deviation. Arch Ophthalmol. 1999;117:1188–91.
70. Brodsky MC. Vertical visual disparity and the human oblique muscles. Binocul
Vis Strabismus Q. 2001;16:251–2.
71. Moster ML, Schatz NJ, Savino PJ, Benes S, Bosley TM, Sergott RC. Alternating
skew on lateral gaze (bilateral abducting hypertropia). Ann Neurol.
1988;23:190–2.
72. Donahue SP, Brodsky MC. Posterior canal predominance in bilateral skew
deviation [letter]. Br J Ophthalmol. 2001;85:1395.
Postscript
There continues to be tenacious resistance to the notion that human extraocular

muscles can primarily overact due to excessive innervation. Yet we know that while
cortical centers control body movements, subcortical centers control individual
muscles. In infantile strabismus, this phenomenon results from persistent activation
of visuo-vestibular pathways at the subcortical level. Its resurgence in humans with
defective binocular vision represents a physiologic activation rather a neurological
deficit. As disconjugate binocular torsion induces a relative retinal disparity that is
perceived stereoscopically, these subcortical torsional pathways are necessarily
extinguished early in infancy in the service of binocular vision and stereopsis.
However, they continue to function in patients with infantile strabismus.
The evolving human binocular visual system never knew that stereopsis was the
goal. According to R.W. Rodieck, “Evolution is not engineering-it is more like
Postscript 63
tinkering-it is not design. The visual system never had an opportunity to be

designed-or redesigned from scratch; instead, both form and function reflect its
long, particular, and capricious history.” (from The First Steps in Seeing. 1998,
page 4, Sinauer Publications). What seems clear is that stereopsis necessitated exa-
ptation of human oblique muscle function to constrain static torsion. Torsional
control must be critical to binocular vision for evolution to have reconfigured this
system. To this end, it is noteworthy that the superior rectus and superior oblique
muscles uniquely receive crossed innervation from their brainstem nuclei. Why?
To segregate the intorters on one side of the brainstem and the extorters on the
other side. As discussed in Chap. 18, phoria adaptation also contributes to the sta-
bilization of ocular torsion. Although ocular torsion is tightly constrained under
normal binocular conditions, the human ocular counterroll of approximately 5%
demonstrates that it is not absolute.
A core concept here is that tilt is the monocular correlate of pitch. If both retinas
are rotated in the same torsional direction, we perceive a stereoscopic tilt of the
visual world, but if both are rotated in opposite directions (permitting fusion), we
perceive a stereoscopic slant of the visual world, and our plane of optimal stereo-
scopic perception also becomes slanted away from us accordingly.
Near is the ground, far is the sky. As we ambulate from an upright position, the
terrestrial plane of the ground is pitched away from the visual axis as viewed from
our upright position. Evolutionarily, this may explain a number of observations
including our physiologic V pattern (to promote convergence in downgaze), and the
binocular extorsion that is measurable during convergence (to actively rotate the
vertical retinal meridians to maximize stereopsis. Everything fits together, although
not by design.
Interestingly, it was Darwin who first articulated the essential role of evolution-
ary reconfiguration. In his 1862 discussion of orchid evolution, he wrote that “When
this or that part has been spoken of as contrived for some special purpose, it must
not be supposed that it was originally always formed for that purpose. The regular
course of events seems to be, that a part which originally served for one purpose, by
slow changes became adapted for wholly different purposes.”
Darwin C: On the Various Contrivances by which British Foreign Orchids are
Fertilized by Insects, and on the Good Effects of Intercrossing. London, England:
John Murray;1862, 346.
Latent Nystagmus
Vestibular Nystagmus with a Twist
5
Vestibular disease holds little interest for the ophthalmologist. Although patients
with vestibular disease can develop nystagmus, diplopia, and oscillopsia, these
symptoms can be treated empirically. Peripheral vestibular disease is caused by
injury to the labyrinth rather than to the eye, whereas central vestibular disease is
caused by brainstem or cerebellar disorders involving the central vestibular path-
ways along their course to the ocular motor nuclei [1, 2]. But the ophthalmologist
encounters a unique form of vestibular nystagmus that is caused by unbalanced
input from the two eyes rather than from the two labyrinths. This visuo-vestibular
nystagmus is known as “latent nystagmus.”
Congenital esotropia is associated with a clinical triad of latent nystagmus, infe-
rior oblique muscle overaction, and dissociated vertical divergence [3]. These
unique eye movements conform to primitive vision-dependent tonus mechanisms
that are reactivated by congenital strabismus or early abnormal visual experience
[4–7]. Evolutionary analogues of primary oblique muscle overaction and dissoci-
ated vertical divergence have been identified in lower vertebrates [5, 6]. In fish,
these are physiologic extraocular movements that use weighted binocular visual
input to modulate extraocular muscle tonus and to maintain visual orientation dur-
ing body movements [5, 6]. The stimulus for bilateral inferior oblique muscle over-
action corresponds to a visuo-vestibular imbalance in the sagittal (pitch) plane,
while dissociated vertical divergence corresponds to a similar imbalance in the
coronal (roll) plane [5–7]. We propose that latent nystagmus results from a similar
visuo-vestibular tonus imbalance in the horizontal turning (yaw) plane.
What Is Latent Nystagmus?
Latent nystagmus is a binocular horizontal oscillation that becomes apparent when

1 eye is covered. First described by Faucon in 1872 [8], latent nystagmus develops
when congenital esotropia precludes frontal binocular vision early in infancy

66 5 Latent Nystagmus
[9–13]. In this setting, a conjugate horizontal jerk nystagmus can be induced by

covering 1 eye, blurring 1 eye, or reducing image brightness in 1 eye [1, 10, 14]. In
latent nystagmus, the slow-phase rotation of the fixating eye is directed toward the
nose and the fast-phase rotation of the fixating eye is directed toward the ear [1, 2,
14]. As such, fixation with the right eye generates a right-beating nystagmus, while
fixation with the left eye produces a left-beating nystagmus [10]. In children with
congenital esotropia and alternating fixation, the direction of nystagmus will spon-
taneously reverse when fixation is switched from one eye to the other [10–15]. Even
after the eyes have been surgically realigned, occlusion of either eye will continue
to induce latent nystagmus. The intensity of latent nystagmus is maximal in abduc-
tion and minimal in adduction, causing some patients to maintain a head turn to
place the fixating eye in an adducted position. The intensity of latent nystagmus
decreases when visual attention declines and increases during attempted fixation
[13–22]. In fact, some patients can reverse the direction of their latent nystagmus by
looking at an imagined target and mentally switching fixation from one eye to the
other [16, 22].
In children with latent nystagmus, the development of amblyopia or the recur-
rence of ocular misalignment can disrupt binocular vision and make a latent nystag-
mus become manifest [23]. The magnitude of the resulting manifest latent nystagmus
is proportional to the degree of the interocular visual disparity [1]. Most patients
with clinical latent nystagmus actually have a small spontaneous jerk nystagmus
that can be measured with both eyes open using eye movement recording [14].
However, successful treatment of amblyopia or strabismus can convert a manifest
latent nystagmus to a clinical latent nystagmus [23]. Manifest latent nystagmus has
also been reported in children with unilaterally reduced vision and sensory esotro-
pia resulting from congenital disorders such as cataract or optic nerve hypoplasia
[9–11]. In this setting, a child will often maintain a head turn to position the fixating
eye in adduction [9–11].
Various theories have been advanced to explain latent nystagmus [24]. These
include a primitive tonus imbalance [1], an egocentric disorder [14], a disorder of
the subcortical optokinetic system [21], a subcortical maldevelopment of retinal slip
control [22], abnormal cortical motion processing [25, 26], a disorder of proprio-
ception [27], and an evolutionary preponderance of the nasal half of the retina [3,
28]. These disparate theories can be reconciled by considering the critical evolution-
ary function of the eyes as sensory balance organs.
Nasotemporal Asymmetry and Latent Nystagmus
Latent nystagmus is associated with nasotemporal asymmetry of the horizontal

optokinetic response during monocular viewing [25, 26]. However, not all patients
with nasotemporal asymmetry have latent nystagmus [29]. In patients with naso-
temporal asymmetry, the monocular optokinetic responses to nasally moving tar-
gets are brisk, while those to temporally moving targets are poor in each eye. This
“nasalward” movement bias under monocular viewing conditions corresponds both
Is Latent Nystagmus a Vestibular Nystagmus? 67
in direction and in waveform to the nasalward slow-phase drift of the fixating eye in
latent nystagmus [21, 29]. To our knowledge, Roelofs [30] first observed horizontal
optokinetic asymmetry in patients with latent nystagmus in 1928. Fifty years later,
experiments by van Hof–van Duin [31] and Wood et al [32] suggested that reduced
binocularity in strabismus can lead to nasotemporal asymmetry. In 1977, Kommerell
[33] suggested that latent nystagmus could be regarded as the consequence of hori-
zontal optokinetic asymmetry. In 1982, Hoffman [34] developed a model to explain
nasotemporal asymmetry based on combined cortical and subcortical input to the
nucleus of the optic tract in the cat. In 1983, Schor [21] proposed that latent nystag-
mus and nasotemporal optokinetic asymmetry are mediated by the nucleus of the
optic tract.
Human nasotemporal asymmetry has received considerable attention because it
persists throughout life in humans with congenital strabismus [15, 21, 25, 34–36].
Even after surgical realignment, nasotemporal asymmetry remains as a “footprint in
the snow” of abnormal visual development [36]. Nasotemporal asymmetry is seen
in rabbits, kittens, monkey infants, and human infants within the first 6 months of
life [36]. The evolutionary retention of this primitive nasotemporal asymmetry in
human infancy shows how ontogeny recapitulates phylogeny during human visual
development [37, 38].
In ordinary life, large parts of the visual field move together during self-motion
[39]. Optic flow occurs during translation (which is signaled by the otoliths and
linear optic flow) and rotation (which is signaled by the semicircular canals and
rotational optic flow) [39]. The low sensitivity to nasal to temporal optic flow in
afoveate, lateraleyed animals is commonly assigned the function of preventing the
locomoting animal from responding to the image motion of stationary contours dur-
ing forward motion, while permitting full compensation for rotational input during
turning movements [39–41]. The absence of nasotemporal optokinetic responses in
lateral-eyed animals assures that during forward movements, ineffective temporal-
ward eye movements do not destabilize images of objects that are directly ahead of
the animal [39–41]. The optokinetic responses of both eyes are controlled by which-
ever eye is stimulated by temporal-to-nasal movement of the visual world [40].
Latent nystagmus recapitulates this monocularly driven horizontal optokinetic
movement.
Is Latent Nystagmus a Vestibular Nystagmus?
Vestibular eye movements are reflex contraversive rotations of the eyes that occur
during involuntary head movements, acting to stabilize the position of the eyes in
space and thereby maintain visual orientation [1, 2, 42, 43]. According to Walls,
“ … vestibularly-controlled reflex eye movements are historically the oldest of all,
with all other kinds of eye-muscle controls and operations accreted to them above
the primitive fish level of evolution” [44] (p71). During head movements, input to
the semicircular canals within the 2 labyrinths provides the afferent stimulus for the
vestibulo-ocular reflex [1, 2, 43, 45]. The semicircular canals respond to angular
acceleration and produce dynamic vestibulo-ocular eye movements. Damage to a

horizontal semicircular canal pathway produces a nystagmus in the plane of the
injured canal [43, 45]. In lateral- and frontaleyed animals, the geometry of the semi-
circular canals conforms closely with the orientation of the extraocular muscles
[46]. When the head is rotated in a particular plane, a semicircular canal within the
labyrinth detects acceleration and sends excitatory innervation to the corresponding
extraocular muscles. Within the brainstem and cerebellum, peripheral vestibular
input is summated to produce innervation to the appropriate extraocular muscle
subnuclei and to maintain the position of the eyes in space. Each horizontal semicir-
cular canal provides excitatory input to the ipsilateral medial rectus muscle and the
contralateral lateral rectus muscle [1, 2, 43, 45].
Visual stabilization mechanisms act in concert with labyrinthine reflexes [43]. In
normal life, optokinetic responses are elicited mainly by head movements, which
also stimulate the vestibular system [39]. Because vestibular neurons receive such
prominent visual and vestibular inputs, disrupting either input reduces the tonic
activation of these neurons, with the effect of disturbing the responses to the other
sensory modality [39]. Thus, labyrinthectomy eliminates optokinetic nystagmus in
rabbits [47, 48], whereas blocking optic nerve activity with tetrodotoxin reduces the
gain of the vestibulo-ocular reflex [39, 49]. According to Miles, each sensory modal-
ity “has played such a major role in the evolution of the other that it is impossible to
understand the operation of either one in isolation” [41] (p393). A confluence of
neuroanatomical, clinical, evolutionary, and experimental evidence has led us to
conclude that latent nystagmus is a vestibular nystagmus that is brought about by
unequal visual input from the 2 eyes rather than from the 2 ears (ie, a visuo-vestibular
nystagmus). The evidence that latent nystagmus arises when the 2 eyes revert to
their primitive function as balance organs can be summarized as follows.
Neuroanatomy of Latent Nystagmus
Studies in subhuman primates have shown that latent nystagmus arises as a result of
incomplete development of visual input from occipitotemporal cortex to subcortical
vestibular pathways [50, 51]. In monkeys with latent nystagmus, there is a loss of
binocularity in the nucleus of the optic tract (NOT), the subcortical structure that
feeds into the vestibular system, with most cells driven by the contralateral eye [50,
51]. The areas that normally provide binocular input to the NOT are the middle tem-
poral (MT) visual area and the medial superior temporal (MST) visual area in occipi-
totemporal cortex. When strabismus is surgically induced in infant monkeys during
the first 2 weeks of life, these monkeys also develop latent nystagmus and visual area
MT/MST loses binocularity. If either eye is covered during infancy, visual area MT/
MST and NOT develop normal binocularity, but the striate cortex still shows loss of
binocularity and these monkeys do not develop latent nystagmus [52]. This finding
suggests that the initial cause of latent nystagmus is loss of binocularity in visual area
MT/MST from the misaligned eye during the first few weeks of life [52].
Neuroanatomical experiments have confirmed the Schor hypothesis [21] that the
NOT is the generator of latent nystagmus [21, 34, 50–52]. A latent nystagmus
occurs in monkeys following artificial induction of esotropia within the first 2 weeks
of life [53]. Unilateral electrical stimulation of the NOT in binocularly deprived
monkeys induces a conjugate nystagmus with the slow phases directed toward the
side of stimulation [54, 55]. Latent nystagmus can be abolished by direct injection
of muscimol, a potent γ-aminobutyric acid A agonist into the NOT in monkeys [50].
Simultaneous bilateral blockage of the NOT virtually abolishes latent nystagmus for
the duration of the blockade [50].
Subcortical optokinetic responses are also mediated by the pretectal NOT [15,
21, 34–36]. The monocular pathways subserving nasotemporal asymmetry and its
neutralization by binocularly driven pathways from the visual cortex were first elu-
cidated by Hoffman in the cat [34, 35]. The cat NOT is a diffuse cell aggregation in
the pretectum that is optimally located to integrate direct retinal and diffuse cortical
projections [34]. These nuclei have high levels of spontaneous activity and operate
in a push-pull fashion such that the sum of their opponent innervation determines
the optokinetic response [21, 34].
The NOT contains neurons that are sensitive to visual motion [54]. Many units in
the primate NOT have large receptive fields that are appropriate for encoding full-
field visual motion to support optokinetic eye movements [54]. Stimulation of the
right and left NOT results in optokinetic nystagmus with slow phases to the right
and left, respectively [21]. Output from the NOT is maximal for horizontal move-
ments but 0 for vertical movements [34].
This phylogenetically ancient subcortical system is depicted in Fig. 5.1. Crossed
connections from each eye to the contralateral NOT transmit horizontal visual
motion information to the vestibular nucleus before impinging on the ocular motor
nuclei [40, 56]. Pretectal neurons in the left NOT receive only crossed input from
the right eye and respond only to leftward motion, while those in the right NOT
receive only crossed input from the left eye and respond only to rightward motion
[15, 21, 36]. In the first 6 months of infancy, this subcortical system predominates
in humans, so that temporally directed monocular optokinetic responses are poor in
early infancy compared with nasally directed optokinetic responses [36]. By 6
months of age, cortical binocular pathways, which are responsive to temporally
directed motion, provide a route whereby the NOT, with its specialized directional
responses, can be accessed from either eye [37, 38]. In animals with well-developed
foveae and frontal, stereoscopic vision, the visual inputs feeding directly to the pre-
tectum are supplemented by inputs routed through the visual cortex that selectively
respond to moving images with no positional disparity in the 2 eyes [57]. This cou-
pling between optokinetic nystagmus and stereopsis allows frontal-eyed animals to
selectively stabilize the moving images of those parts of the scene within a selected
depth plane, while disregarding induced image motion of the visual world at other
distances [40, 57]. In humans with congenital strabismus, binocularly driven corti-
copretectal pathways never become established, allowing the primitive monocular
nasotemporal asymmetry to predominate.
LR MR MR LR
NOT NOT
III III
LGN MLF LGN
AC AC
PC VI VI
PC
HC HC
VN VN
Labyrinth Labyrinth
R L R L
R L CC R L
Cortex Cortex
Fig. 5.1 Schematic diagram depicting cortical and optokinetic pathways. Cortical input to tempo-
rally directed movement, which is present only in frontal-eyed animals, requires the establishment
of normal binocular cortical connections. This input is absent in humans with congenital strabis-
mus. Direct crossed pathways from the eye to the nucleus of the optic tract provide nasalward
subcortical optokinetic responses even when binocular cortical connections are absent (R and L
represent monocular cortical cells corresponding to the right and left eyes, respectively). Note that
the nucleus of the optic tract (NOT) relays horizontal visuo-vestibular information to the vestibular
nucleus (VN), where it is integrated with horizontal vestibular input from the labyrinths to estab-
lish horizontal extraocular muscle tonus. LGN indicates lateral geniculate nucleus; CC, corpus
callosum; V1, abducens nucleus; III, oculomotor nucleus; LR, lateral rectus muscle; MR, medial
rectus muscle; AC, anterior canal; PC, posterior canal; and HC, horizontal canal
Clinical Signs of Vestibular Origin
Bilateral positioning of the eyes and ears promotes survival by enabling the organ-
ism to crosslink input from different sense organs to impart balance. Each eye and
its ipsilateral semicircular canals share the same directional bias to movement. For
example, the right horizontal semicircular canal is activated by head rotation to the
right (which induces a rotation of the visual world to the left) and inhibited by head
rotation to the left (which induces a rotation of the visual world to the right) [2, 43,
45, 58, 59]. The monaural and monocular directional biases summate, so that activa-
tion of the right horizontal semicircular canal during rightward head rotation is rein-
forced by the physiologic activation of the right eye by the induced nasal rotation of
the visual world. The close geometrical relationship between the semicircular canals
and the extraocular muscles presumably facilitates the integration of head motion
and visual movement and their orderly summation to produce transformation to an
appropriate ocular motor response [46, 60].
Latent nystagmus usually conforms to Alexander’s law, which states that the
intensity of a peripheral vestibular nystagmus increases when the eyes are moved in
the direction of the fast phase and decreases when the eyes are moved in the direction
of the slow phase [2, 21, 60–63]. Latent nystagmus damps when the fixating eye is
turned toward the nose (which is also the direction of the slow phase) and increases
in intensity when the fixating eye is turned toward the ipsilateral ear (which is in the
direction of the fast phase) [14, 15, 62, 63]. A similar damping of horizontal nystag-
mus is seen in peripheral horizontal vestibular nystagmus after disease or injury to 1
horizontal semicircular canal. By contrast, Alexander’s law does not apply to con-
genital nystagmus, which reverses direction in different positions of gaze. The con-
traversive head turn in latent nystagmus (ie, a head turn opposite in direction to the
deviation of the fixating eye) also characterizes vestibular eye movements [2].
Additional evidence for the duality of optic and vestibular innervation can be elic-
ited by occluding 1 eye in the patient with latent nystagmus, spinning the patient,
suddenly stopping the spin, then immediately observing the effect of the postrota-
tional nystagmus on the latent nystagmus when either eye is occluded. A horizontal
nystagmus induced by body spinning nullifies or accentuates latent nystagmus
depending on the direction of spin relative to the fixating eye (Fig. 5.2). For example,
spinning the patient to the right excites the right horizontal canal and inhibits the left
horizontal semicircular canal to induce a nystagmus with a slow-phase rotation to the
left and a fast-phase rotation to the right. If the spin is suddenly stopped (after approxi-
mately 10 rotations), a shift in endolymph deflects the cupula in the opposite direc-
tion, causing transient excitation of the left horizontal semicircular canal and transient
inhibition of the right horizontal semicircular canal and inducing a left-beating nys-
tagmus (termed “postrotational nystagmus”). If the left eye is occluded to induce
latent nystagmus prior to this maneuver, the latent nystagmus will diminish or disap-
pear immediately following cessation of the spin. If the occluder is quickly moved to
cover the right eye, the intensity of the latent nystagmus with the left eye viewing will
be correspondingly increased relative to that observed with the left eye fixating before
the spin. In this way, the clinician can observe how visual input is summated with
vestibular input to establish central vestibular tone in the horizontal plane.
The more visual input is dominated by 1 eye in latent nystagmus, the higher the
velocity of the slow-phase rotations in the direction toward the opposite eye [22].
Simonsz and Kommerell [63] performed eye movement recordings before and after
occlusion therapy for amblyopia in patients with latent nystagmus. After prolonged
occlusion, the slow-phase velocity of the nystagmus in the amblyopic eye decreased
to the same extent that the slow-phase velocity of the nystagmus in the preferred eye
increased. The sum of the 2 slow-phase velocities remained the same in
a b
V V
O O O O
++ ––
c d
V V V V
O O O O
–– ++ –– ++
Fig. 5.2 Visual and vestibular interaction in latent nystagmus. Latent nystagmus decreases with
spinning toward the fixating eye and increases with spinning toward the occluded eye. O represents
direction of ocular (visuo-vestibular) tonus; V, direction of horizontal vestibular tonus. Both O and
V correspond to the slow phase of the induced nystagmus. ++ Indicates stimulated horizontal
semicircular canal; --, inhibited horizontal semicircular canal; (a) Occlusion of the left eye
increases visuo-vestibular tonus to the left. (b) The patient with latent nystagmus is spun to the
right to stimulate the right horizontal semicircular canal, which increases leftward horizontal ves-
tibular tonus and causes a slow conjugate drift of both eyes to the left. At this point, the latent
nystagmus would be enhanced by vestibular input (if the examiner could observe it). (c) When the
spinning is suddenly stopped, the opposite vestibular stimulus is exerted, causing the left semicir-
cular canal to drive the eyes to the right. This rightward vestibular tonus imbalance nullifies the
leftward visual tonus imbalance induced by monocular fixation with the right eye, thereby reduc-
ing the intensity of the latent nystagmus. (d) When the occluder is quickly switched to the right
eye, the visual tonus imbalance is augmented by an ipsidirectional visual tonus imbalance, increas-
ing the intensity of the latent nystagmus
straight-ahead gaze, demonstrating that visual input to the 2 eyes (just like rota-
tional input to the 2 horizontal canals) maintains a push-pull relationship [21, 63].
This observation lends further support to a vestibular underpinning for latent nys-
tagmus. The clinical similarities between latent nystagmus and peripheral vestibular
nystagmus are summarized in Table 5.1.
Evolutionary Underpinnings of Latent Nystagmus
The notion of latent nystagmus as a horizontal visuo-vestibular tonus imbalance

provides conceptual unification with its associated inferior oblique overaction and
dissociated vertical divergence in patients with congenital esotropia. The evolution-
ary progenitors of all of these visuo-vestibular movements use binocular input to
establish physical orientation in space. These primitive reflexes rely on a dissociated
form of binocular vision between the 2 laterally placed eyes, which has been super-
seded by normal cortical binocular vision in humans [5]. In congenital esotropia,
however, these primitive subcortical reflexes are not erased by binocular cortical
input. Eye movement recordings have demonstrated that dissociated vertical diver-
gence incorporates a vertical latent nystagmus, suggesting a shared common origin
for these movements [64].
Given that visual and labyrinthine input are pooled together within the central
vestibular system of lower animals [65–67], a visual counterpart to peripheral ves-
tibular nystagmus would seem necessary on evolutionary grounds. Many authors
have attributed latent nystagmus as a tonus imbalance of the horizontal extraocular
muscles [1, 4, 7, 14, 19, 68]. Latent nystagmus corresponds to a tropotactic vision-
induced tonus imbalance (ie, one that functions to reestablish binocular equilibrium
rather than to directionally orient an eye toward incoming light) [69]. Ohm recog-
nized the physiologic coaptation of visual and vestibular innervation and its role in
the generation of latent nystagmus long before others did (as was also the case with
dissociated vertical divergence and primary oblique muscle overaction) [70–73]. In
a monograph written near the end of his life, he stated, “The impulses that originate
from both eyes keep both vestibular nuclei in equilibrium. The equilibrium becomes
unbalanced when one eye is being occluded. Then, a nystagmus beating towards the
Table 5.1 Peripheral Vestibular Nystagmus vs Latent Nystagmus

Peripheral Vestibular Nystagmus Latent Nystagmus
Induced by unequal bilateral labyrinthine input Induced by unequal binocularvisual input
Stimulation of right horizontal semicircular canal Stimulation of right eye evokes a conjugate
evokes a conjugate right-beating nystagmus right-beating nystagmus
Nystagmus intensity proportional to the degree of Nystagmus intensity proportional to the
horizontal canal imbalance degree of binocularvisual imbalance
Damps during gaze away from the side of the Damps during gaze away from the side of
normal canal the fixating eye
Conforms to Alexander’s law Conforms to Alexander’s law
Modulated by subcortical neural pathways Modulated by cortical and subcortical
neural pathways
side of the open eye appears” [72]. Kestenbaum [10] emphasized that latent nystag-
mus could not be attributed to luminance per se, since shining a bright light in the
right eye worked like occlusion of the right eye and caused a left-beating nystag-
mus. He noted that the presence of a sharper visual image on the retina of one eye
than the other appeared to be the decisive stimulus for inducing latent nystagmus.
Predominance of a primitive visuo-vestibular imbalance provides an evolution-
ary basis for the shift in egocenter that has been invoked to explain latent nystagmus
[14]. According to this hypothesis, the egocenter is localized to the median body
plane under normal binocular conditions, but shifts to the side of the fixating eye
under monocular conditions. Dell’Osso et al [14] hypothesized that humans with
latent nystagmus retain an abnormal egocenter in the median plane even under mon-
ocular conditions, causing the fixating eye to drift toward midline. In the lateral-
eyed animal, fixation with the right eye would instantaneously shift the egocenter to
the left of the object of regard, necessitating a body turn to frontalize the object and
a contraversive eye rotation to maintain fixation [14]. As neatly summarized by
Dichgans and Brandt: … the results of visual and vestibular stimulation on egocen-
tric localization indicate the close similarity in the perceptual consequences of stim-
ulation of the two organs. The assumption of a unitary central representation of
egocentric space, based on visual and vestibular (as well as acoustic and somatosen-
sory) afferents is perceptually obvious [42] (pp763–764).
It remains to be determined whether a higher order egocentric shift could cause
the visuo-vestibular imbalance which generates the linear slow phase of latent
nystagmus.
xperimental Evidence That Latent Nystagmus Is Vestibular

E
in Origin
Optokinetic responses are fundamentally intertwined with vestibular responses, and

a major site of this commingling is the vestibular complex [39]. Waespe and Henn
[58] and Henn et al [59] performed single-cell recordings from the medial vestibular
nucleus in monkeys and found that single neurons can be activated either by body
rotation or optokinetic stimulation. Units that were excited by head acceleration to
the left were also exited by motion of optokinetic stripes to the right. Most cells
responded to both the whole-field visual motion, as well as to the vestibular indica-
tions of head rotation, and the responses of vestibular neurons followed approxi-
mately the same time course as the delayed component of optokinetic nystagmus
[58, 59]. As summarized by Dichgans and Brandt:
All of the recent studies performed in awake animals show a tonic modulation of resting
discharge of vestibular units in response to exclusive constant velocity motion of the visual
surround. The modulation, although to a variable degree, seems to occur in the great major-
ity of horizontal semicircular canal-dependent units of all vertebrate species tested. A unit
which is excited by a head acceleration, say, to the left is also excited by motion of the sur-
round to the right, which represents the optokinetic stimulus that in man would cause the
sensation of turning to the left [42] (pp781,783).
This underlying vestibular response to both visual motion and body rotational stim-
uli may explain the overlapping nystagmus response that characterizes latent, opto-
kinetic, and peripheral vestibular nystagmus [2, 19, 21, 62, 71]. This overlap may
reflect the fact that all 3 movements subserve a similar physiologic function (ie,
detection of rotation of the body and visual environment).
Latent nystagmus, optokinetic nystagmus, and the vestibuloocular reflex also
show velocity storage, a phenomenon in which constant vestibular input or visual
flow in the same direction is stored for up to 20 seconds in the brainstem, even when
the stimulus is terminated (Table 5.2) [50, 51]. The presence of velocity storage
serves to enhance the slow-tracking eye movements to vestibular stimulation and
optic flow response at low frequencies of rotation [2, 74]. Although latent nystag-
mus has variously been attributed to anomalous cortical motion processing [25, 26],
or a cortical pursuit asymmetry [53, 75], the absence of velocity storage mechanism
within the pursuit system implicates the vestibular system as the generator of latent
nystagmus.
Table 5.2 Experimental Evidence That Latent Nystagmus Is Vestibular in Origin

Nucleus of the Optic Tract (NOT) Latent Nystagmus in
in Monkeys Monkeys Latent Nystagmus in Humans
Electrical stimulation of NOT on Covering one eye causes Same
one side causes conjugate conjugate nystagmus in
nystagmus in both eyes with slow both eyes with slow
phases directed to the same side phases directed to the
as the stimulated NOT [54, 55]. same side as the
activated NOT
(contralateral to the
viewing eye) [50, 51].
The velocity of the slow phases of When one eye is Slow rise and slow decay is
nystagmus from electrical covered, the velocity of usually not seen because the eye
stimulation of NOT slowly the slow phases of velocity in humans is much
increases. When stimulation is nystagmus slowly slower (1–5 dynes [d/s]) than
stopped, the slow-phase increases. When the eye that found in monkeys (20–90
functioning eye velocity slowly is uncovered, the d/s). In some humans, latent
decays. This slow increase and slow-phase eye velocity nystagmus eye velocity can be
slow decay in eye velocity is due slowly decays [51]. 20 d/s and in those cases a slow
to the charging and discharge of rise and slow decay is found
the vestibular velocity storage (R.J.T., unpublished data).
system [54, 55].
NOT projects to the vestibular Chemical suppression Unknown
velocity storage system and NOT of NOT blocks OKAN
is responsible for eliciting and suppresses latent
optokinetic after nystagmus nystagmus [50].
(OKAN), which is the portion of
optokinetic nystagmus that is
mediated by the velocity storage
system [54, 55].
Conclusions
Latent nystagmus is a unique form of vestibular nystagmus that is evoked by unbal-

anced visual input from the 2 eyes rather than unequal rotational input from the 2
labyrinths. The neurophysiological substrate for latent nystagmus is operative in
lateral-eyed animals and in human infants with undeveloped binocular corticopre-
tectal pathways. When congenital esotropia disrupts the establishment of these bin-
ocular visual connections, visual input from the fixating eye to the contralateral
NOT evokes a visuo-vestibular counterrotation of the eyes that corresponds to a
turning or twisting movement of the body toward the object of regard (“vestibular
nystagmus with a twist”). In this setting, unbalanced binocular visual input can
induce a motion bias in the vestibular nucleus to generate the visual counterpart of
horizontal labyrinthine nystagmus, namely, latent nystagmus. As the eyes rotate
frontally during evolution, this visuo-vestibular function is sacrificed, but the cen-
tral nervous system retains these latent subcortical visual pathways. Latent nystag-
mus is nature’s proclamation that our 2 eyes, when dissociated from birth, can revert
to their ancestral function as sensory balance organs.
References
1. Cogan DG. Nystagumus. In: Neurology of the ocular muscles. 2nd ed.
Springfield: Charles C Thomas Publishers; 1956. p. 226–7.
2. Leigh RJ, Zee DS. Diagnosis of central disorders of ocular motility. In: The
neurology of eye movements. 3rd ed. New York: Oxford University Press;
1999. p. 445–7.
3. Lang J. The congenital strabismus syndrome. Strabismus. 2000;8:195–9.
8. Faucon A. Nystagmus par insuffisance des Droits externes. J Ophtal Paris.
1872;1:233–41.
9. Harcourt B. Manifest latent nystagmus affecting patients with uniocular con-
genital blindness. In: Gregersen E, editor. Transactions of the European
Strabismological Associations. Copenhagen: European Strabisomological
Associations; 1984. p. 259–64.
10. Kestenbaum A. Nystagmus. In: Clinical methods of neuro-ophthalmologic
examination. New York: Grune & Stratton Inc; 1947. p. 234–5.
11. Kushner BJ. Infantile uniocular blindness with bilateral nystagmus: a syn-
drome. Arch Ophthalmol. 1995;113:1298–300.
12. Shawkat FS, Harris CM, Taylor DS. Spontaneous reversal of nystagmus in the
dark. Br J Ophthalmol. 2001;85:428–31.
Conclusions 77
13. Dell’Osso LF, Traccis S, Abel LA. Strabismus—a necessary condition for
latent and manifest latent nystagmus. Neuro-ophthalmology. 1983;3:247–57.
14. Dell’Osso LF, Schmidt D, Daroff RB. Latent, manifest latent, and congenital
nystagmus. Arch Ophthalmol. 1979;97:1877–85.
15. Kommerell G. The relationship between infantile strabismus and latent nystag-
mus. Eye. 1996;10:274–81.
16. Dell’Osso LF, Abel LA, Daroff RB. Latent/manifest latent nystagmus reversal
using an ocular prosthesis: implications for vision and ocular dominance. Invest
Ophthalmol Vis Sci. 1987;28:1873–6.
17. Ciancia AO. On infantile esotropia with nystagmus in abduction. J Pediatr
18. Melek N. El nistagmus en la esotopia con limitacion bilateral de la abduccion.
Paper presented at: Anales del V o Congreso del CLADE; Guaruja, Brazil;
1978: 27–31.
19. Gresty MA, Metcalfe T, Timms C, et al. Neurology of latent nystagmus. Brain.
1992;115:1303–21.
20. Abadi RV, Scallan C. Manifest latent and congenital nystagmus waveforms in
the same subject: a need to reconsider the underlying mechanism of nystagmus.
Neuro-ophthalmology. 1999;21:211–21.
21. Schor CM. Subcortical binocular suppression affects the development of latent
and optokinetic nystagmus. Am J Optom Physiol Optic. 1983;60:481–502.
22. Kommerell G, Zee DS. Latent nystagmus: release and suppression at will.
Invest Ophthalmol Vis Sci. 1993;34:1785–92.
23. Zubcov AA, Reinecke RD, Gottlob I, et al. Treatment of manifest latent nystag-
mus. Am J Ophthalmol. 1990;110:160–7.
24. Sekiya H, Hasegawa S, Mukuno K, Ishikawa S. Sensitivity of nasal and tempo-
ral hemiretinas in latent nystagmus and strabismus evaluated using the light
reflex. Br J Ophthalmol. 1994;78:327–31.
25. Tychsen L, Lisberger SG. Maldevelopment of visual motion processing in
humans who had strabismus with onset in infancy. J Neurosci. 1986;6:2495–508.
26. Norcia AM, Garcia H, Humphry R, et al. Anomalous motion VEPs in infants
and in infantile esotropia. Invest Ophthalmol Vis Sci. 1991;32:436–9.
27. Ishikawa S. Latent nystagmus and its etiology. In: Reinecke RD, edi-

tor. Strabismus: proceedings of the International Strabismological Association,
Kyoto, Japan. New York: Grune & Stratton Inc; 1978. p. 203–14.
28. Lange J. A new hypothesis on latent nystagmus and on the congenital squint
syndrome. In: van Dalen AT, Houtman WA, editors. Strabismus symposium,
Amsterdam 1981, Documenta ophthalmologic proceedings series, vol. 32.
Norwell: Kluwer Academic Publishers; 1982. p. 83–8.
29. Shallo-Hoffman J, Faldon M, Hague S, et al. Motion detection deficits in infan-
tile esotropia without nystagmus. Invest Ophthalmol Vis Sci. 1997;38:219–26.
30. Roelofs CO. Nystagmus latens. Arch Augenheilkd. 1928;98:401–47.
31. van Hof-van Duin J. Early and permanent effects of monocular deprivation on
pattern discrimination and visuomotor behaviour in cats. Brain Res.
1976;111:261–76.
32. Wood CC, Spear PD, Braun JJ. Direction specific deficits in horizontal optoki-
netic nystagmus following removal of the visual cortex in the cat. Brain Res.
1977;60:231–7.
33. Kommerell G. Beziehungen zwischen Strabismus un Nystagmus. In: Kommerell
G, editor. Augenbewegungsstörungen, Neurophysiologie und Klinik:
Symposion der Deutschen Ophthalmologischen Gesellschaft, Freiburg, 1977.
Munich: Bergmann; 1978. p. 367–73.
34. Hoffman K-P. Cortical versus subcortical contributions to the optokinetic reflex
in the cat. In: Lennerstrand G, Keller EL, editors. Basis of ocular motility: pro-
ceedings of a Wenner-Gren Center and Smith-Kettlewell Eye Research
Foundation I International Symposium. New York: Pergamon Press; 1982.
p. 303–11.
35. Hoffman KP. Neural basis for optokinetic defects in experimental animals with
strabismus. In: Kaufman H, editor. Transactions of the 16th Meeting of the
European Strabismological Association. Giessen: European Strabismological
Association; 1987. p. 35–46.
36. Braddick O. Where is the nasotemporal asymmetry? Curr Biol. 1996;6:250–3.
37. Atkinson J. Development of optokinetic nystagmus in the human infant and
monkey infant: an analogue to development in kittens. In: Freeman RD, editor.
Developmental neurobiology of vision. New York: Plenum Publishing Corp;
1979. p. 277–87.
38. Naegele JR, Held R. The postnatal development of monocular optokinetic nys-
tagmus. Vis Res. 1982;22:341–6.
39. Wallman J. Subcortical optokinetic mechanisms. In: Miles FA, Wallman J, edi-
tors. Visual motion and its role in the stabilization of gaze. Amsterdam: Elsevier
Science; 1993. p. 321–42.
40. Ohmi M, Howard IP, Eveleigh B. Directional preponderance in human optoki-
netic nystagmus. Exp Brain Res. 1986;63:387–94.
41. Miles FA. The sensing of rotational and translational optic flow by the primate
optokinetic system. In: Miles FA, Wallman J, editors. Visual motion and its role
in the stabilization of gaze. Amsterdam: Elsevier Science; 1993. p. 393–403.
42. Dichgans J, Brandt TH. Visual-vestibular interaction: effects on self-motion
perception postural control. In: Held R, Leibowitz HW, Teuber HL, editors.
Handbook of sensory physiology, vol. 8. Berlin: Springer; 1978. p. 755–804.
43. Markham CH. How does the brain generate horizontal vestibular nystagmus?
In: Baloh RW, Halmagyi GM, editors. Basic vestibular mechanisms: part 1.
New York: Oxford University Press; 1996. p. 48–61.
44. Walls GL. The evolutionary history of eye movements. Vis Res. 1962;2:69–80.
45. Cohen BL. Eye movements from semicircular canal nerve stimulation in the
cat. Ann Otol Rhinol Laryngol. 1964;73:153–70.
ments in lateral-eyed animals and frontal-eyed animals. Ann N Y Acad Sci.
1981;374:20–30.
47. Baarsma EA, Collewijn H. Changes in compensatory eye movements after uni-
lateral labyrinthectomy in the rabbit. Arch Otorhinolaryngol. 1975;211:219–30.
Conclusions 79
48. Barmack NH, Pettorossi VE, Erickson RG. The influence of bilateral labyrin-
thectomy on horizontal and vertical optokinetic reflexes in the rabbit. Brain
Res. 1980;196:520–4.
49. Collewijn H, Van der Steen J. Visual control of the vestibuloocular reflex in the
rabbit: a multi-level interaction. In: Glickstein M, Yeo C, Stein J, editors.
Cerebellum and neuronal plasticity. New York: Plenum Publishing Corp; 1987.
p. 277–91.
50. Mustari MJ, Tusa RJ, Burrows AF, Fuchs AF, Livingston CA. Gaze-stabilizing
deficits and latent nystagmus in monkeys with early-onset visual deprivation:
role of the pretectal NOT. J Neurophysiol. 2001;86:662–75.
51. Tusa RJ, Mustari MJ, Burrows AF, Fuchs AF. Gaze-stabilizing deficits and
latent nystagmus in monkeys with brief, early-onset visual deprivation: eye
movement recordings. J Neurophysiol. 2001;86:651–61.
52. Tusa RJ, Mustari MJ, Das VE, Boothe RG. Animal models for visual

deprivation- induced strabismus and nystagmus. Ann N Y Acad Sci.
2002;956:346–60.
53. Kiorpes L, Walton PJ, O’Keefe LP, Movshon JA, Lisberger SG. Effect of early-
onset artificial strabismus on pursuit eye movements and on neuronal responses
in area MT of macaque monkeys. J Neurosci. 1996;16:6537–53.
54. Mustari MJ, Fuchs AF. Discharge patterns of neurons in the pretectal nucleus of
the optic tract in the monkey. J Neurophysiol. 1990;64:77–90.
55. Schiff D, Cohen B, Raphan T. Nystagmus induced by stimulation of the nucleus
of the optic tract in the monkey. Exp Brain Res. 1988;70:1–40.
56. Cazin L, Precht W, Lannou K. Optokinetic responses of vestibular nucleus neu-
rons in the rat. Pflugers Arch. 1980;38:31–8.
57. Howard IP, Simpson WA. Human optokinetic nystagmus is linked to the stereo-
scopic system. Exp Brain Res. 1989;78:309–14.
58. Waespe W, Henn V. Neuronal activity in the vestibular nuclei of the alert mon-
key during vestibular and optokinetic stimulation. Exp Brain Res.
1977;27:523–38.
59. Henn V, Cohen B, Young LR. Visual-vestibular interaction in motion percep-
tion and the generation of nystagmus. Neurosci Res Program Bull.
1980;18:457–651.
60. Ezure K, Graf W. A quantitative analysis of the spatial organization of the ves-
tibuloocular reflexes in lateral-and frontal-eyed animals, I: orientation of semi-
circular canals and extraocular muscles. Neuroscience. 1984;12:85–93.
61. Robinson DA, Zee DS, Hain TC, Holmes A, Rosenberg LF. Alexander’s law:
its behavior and origin in the human vestibulo-ocular reflex. Ann Neurol.
1984;16:714–22.
62. Dell’Osso LF, Jacobs JB. A normal ocular motor system model that simulates
the dual mode fast-phases of latent/manifest nystagmus. Biol Cybern.
2001;85:459–71.
63. Simonsz HJ, Kommerell G. Effect of prolonged monocular occlusion on latent
nystagmus. Neuro-ophthalmology. 1992;12:185–92.
65. Pfeiffer W. Equilibrium orientation in fish. Int Rev Gen Exp Zool. 1964;1:
77–111.
J Comp Physiol. 1983;150:473–81.
68. Keiner GB. Physiology and pathology of optomotor reflexes. Am J Ophthalmol.
1956;42:233–49.
69. Duke-Elder S. The effect of light on movement. In: System of ophthalmology:
the eye in evolution. London: Henry Klimpton; 1958. p. 27–81.
70. Ohm J. Der latente Nystagmus bei angeborener einseitiger Blindheit. Graefes
Archiv Ophthalmol Archiv Augenheilkd. 1948;148:318.
71. Ohm J. Der latent Nystagmus im Stockdunkeln. Arch Augenheilkd.

1928;99:417–38.
72. Ohm J. Nystagmus und Schielen bei Sehschwachen und Blinden. Stuttgart:
Ferdinand Enke; 1958. p. 81.
73. Fetter M, Zee DS. Recovery from unilateral labyrinthectomy in rhesus monkey.
J Neurophysiol. 1998;59:370–93.
74. Cohen B, Henn V, Raphan T, Dennett D. Velocity storage, nystagmus, and
visual-vestibular interactions in humans. Ann N Y Acad Sci. 1981;374:421–33.
75. Tychsen L, Hurtig RR, Scott WE. Pursuit is impaired but the vestibulo-ocular
reflex is normal in infantile strabismus. Arch Ophthalmol. 1985;103:536–59.
Postscript
This treatise proposes that latent nystagmus is driven subcortically by binocular

imbalance from the two eyes. The diagram in Fig. 5.1 shows a simplified depiction
of this concept. As summarized in Chap. 17, the subcortical visuo-vestibular cir-
cuitry is more complex, with motion pathways from the NOT-DTN projecting to the
dorsal cap of the inferior olive, then down to the cerebellar flocculus, before con-
necting with the vestibular nucleus.
Some investigators believe that latent nystagmus is cortical in origin. As with
DVD, cortical suppression can generate latent nystagmus. However, cortical sup-
pression activates the subcortical circuitry that generates latent nystagmus. So des-
ignating latent nystagmus as cortical or subcortical misses the point. Chris Harris
nicely summarized the evidence for and against a strictly cortical and subcortical
position (Harris C: Latent nystagmus. Continuing Education Training, Course code
C-31801, www.optometry.co.uk/clinical). Both templates are integral to latent nys-
tagmus, but this oscillation clearly arises from a primitive subcortical visual motion
bias that was present evolutionarily before the cortex ever developed. The
Postscript 81
prominent torsional ocular rotations that accompany latent nystagmus further attest
to their visuo-vestibular origin. Again, the unanswered question is whether covering
one eye evokes DVD directly via the subcortical circuitry, indirectly via the cortex,
or both.
What accounts for the jump from optokinesis as the stimulus for monocular
nasotemporal asymmetry (MNTA) to fixation as the stimulus for latent nystagmus?
Evolutionarily, motion gives way to fixation. Things have to move for fixation to
occur. The modulation of latent nystagmus by fixation occurs because cortical
reflexes are retrofitted to subcortical ones. One concern with the subcortical hypoth-
esis for MNTA and latent nystagmus has been that directionally-selective horizontal
retinal ganglion cells have been found in rats, rabbits, and cats, but not in primates.
Some believe that the sweep of nasalward motion across consecutive ganglion cells
could encode the equivalent directional signal. Alternatively, it could be that these
directional ganglion cells have not yet been found in primates, as they are probably
relatively rare so it would not be an easy task to find them in tissue as densely
packed as the primate retina.
This analysis calls into question the misguided classification of latent nystagmus
as “fusion maldevelopment nystagmus.” This designation, proposed by a self-
appointed NIH-sponsored committee, implies that the primary cause of essential
infantile esotropia and latent nystagmus is a primary defect in cortical fusional
development. There is no evidence that this is the case. It is more plausible that
fusion maldevelopment and latent nystagmus are both inevitable outcomes of an
early binocular misalignment that eventuates in infantile esotropia. Chap. 17 dem-
onstrates how prolonged subcortical neuroplasticity could allow subcortical MNTA
to generate both essential infantile esotropia and latent nystagmus, with the observed
reconfiguration of cortical motion pathways occurring as a secondary phenomenon.
Dissociated Vertical Divergence
Perceptual Correlates of the Human Dorsal Light Reflex
6
In living organisms, light from the sky above and gravity from the earth below have
led to the evolution of sensory organs for vision and balance. The bright sky serves
as a hemispheric light source that provides a stable visual reference for which way
is up. In lower animals, the central vestibular system integrates visual input from the
2 eyes and graviceptive output from the 2 labyrinths to modulate postural and extra-
ocular muscle tonus and maintain vertical orientation [1–3]. Although visual input
is usually subordinate to vestibular input in establishing postural orientation, some
lateral-eyed animals also maintain vertical orientation by equalizing visual input to
the 2 eyes [4]. Many fish and insects exhibit a dorsal light reflex in which illumina-
tion from one side evokes a reflex body tilt toward the light [4–7]. When a light is
shined down from the right side, for example, the right eye receives greater visual
input than the left eye (Fig. 6.1). This binocular disparity would only exist in nature
if the animal were tilted with its right side toward the sky. This visual imbalance
causes the central vestibular system to register a leftward body tilt relative to the
body position that would be necessary for the 2 eyes to receive equal binocular
visual input, and to reflexively alter postural tonus to correct the tilt [1–4] (Fig. 6.1).
In a vertically stabilized fish, the same stimulus evokes a vertical divergence of the
eyes to reorient the interpupillary axis relative to the new light source, causing the
eye with lesser visual input to shift dorsally, and the eye with greater visual input to
shift ventrally [3] (Fig. 6.1).
The term dorsal pertains to the back or upper aspect of an animal. The dorsal
aspect of the head in fish, quadrupeds, and bipeds corresponds to the top of the head.
In bipeds such as humans, the back retains its phylogenetic dorsal orientation,
although it is no longer the upper aspect in the upright position. Humans with con-
genital strabismus who never develop single binocular vision exhibit an atavistic
resurgence of the dorsal light reflex in the form of dissociated vertical divergence
(DVD) [8]. When these strabismic humans fixate monocularly, the nonfixating eye
exhibits a slow dorsal rotation termed dissociated vertical deviation. This dorsal

a b
Fig. 6.1 In the fish, increasing luminance to the right eye by shining a light from the 1-o’clock
position (as seen by the animal) shifts subjective vertical clockwise toward the 1-o’clock position
(as seen by the animal). Visual objects aligned with the gravitational vertical are therefore misreg-
istered as being tilted counterclockwise toward the 11-o’clock position relative to the animal’s
subjective vertical. (a) Vertical objects and the animal’s body are subjectively misregistered as
being tilted counterclockwise, necessitating a clockwise body tilt toward the body position that is
necessary for equal binocularvisual input. (b) When a corrective body tilt is prevented, the ocular
component of this righting reflex evokes a vertical divergence of the eyes to rotate the interpupil-
lary axis toward a position that is perpendicular to the altered subjective vertical
rotation can be elicited by optically or mechanically reducing visual input to either

eye. It can also occur spontaneously when there is a fluctuating degree of sensory
suppression in one or both eyes [8]. Eye movement recordings have confirmed that
DVD comprises a cyclovertical divergence in which the fixating eye depresses and
intorts while the nonfixating eye elevates and extorts [9–11]. Thus, the alternative
term, dissociated vertical divergence, provides a more accurate mechanistic descrip-
tion of this movement [8].
In fish, abrupt fluctuations in binocular visual input cause the central vestibular
system to register a tilt, and the ensuing dorsal light reflex serves to annul this tilt by
realigning the eyes and body to new a orientation that the brain interprets as vertical.
If DVD is a human dorsal light reflex, then humans with DVD might be expected to
experience a perceived visual tilt (ie, a sensation that the visual environment is
tilted) when the gradient of visual input to the 2 eyes is abruptly altered. In this case,
the cyclovertical divergence associated with DVD should serve to reduce or elimi-
nate the perception of tilt. Alternatively, if DVD is not a human dorsal light reflex,
then the torsional component of DVD might be expected to induce a perception of
visual tilt. In this case, patients should perceive no visual tilt prior to the onset of
Results 85
DVD, but a perceived visual tilt should be present after the DVD has occurred. To
elucidate the perceptual correlates of DVD, I examined visual tilt perception in 9
patients with DVD and 9 normal control subjects.
Patients and Methods
Nine patients with bilateral DVD were examined prospectively and consecutively.
Children who had been treated with previous cyclovertical muscle surgery were
excluded from the study since the resulting fundus torsion could influence the per-
ception of tilt. Dissociated vertical divergence was diagnosed when occlusion of
either eye evoked a dorsal rotation of the occluded eye. The amplitude of the DVD
was measured in both eyes using the vertical prism under cover test. Visual acuity
was measured in each eye, versions were examined, field measurements were
obtained, and a dilated retinal examination was performed to look for static torsion.
To determine whether patients experienced a perceived visual tilt (ie, a perceived
tilt of the visual environment without any sensation of body tilt) or subjective tilt (ie,
a perceived tilt of the body) under monocular conditions, each patient was instructed
to view a pencil held vertically in the sagittal plane midway between the 2 eyes dur-
ing occlusion of each eye and during alternate occlusion of the eyes. The patient’s
head was maintained in the upright position before and during testing. Care was
taken to assure that the pencil was not slanted forward or backward in the sagittal
plane, as a slant would optically induce a monocular image tilt. Each patient was
instructed to hold up his or her index finger and to move this finger to demonstrate
any perceived movement of the pencil as each eye was occluded. The cover test and
alternate cover test were repeated 3 to 4 times in each patient to assure that percep-
tual responses were consistent from one trial to the next. When a perceived motion
was noted, the patient was specifically asked whether the pencil appeared to move
sideways or to tilt. The patient was then asked, “does it look like the pencil is mov-
ing, or does it feel like you are moving?” A labyrinthine imbalance such as that
brought on by spinning or by disease of the semicircular canals, is generally per-
ceived as a sensation of body movement in space. Thus, a perceived sensation of
body movement in space would suggest that unequal visual input is inducing a laby-
rinthine imbalance, while perceived visual tilt with no sensation of body movement
would suggest that the neural pathways activated by unequal visual input do not
directly alter the relative output of the 2 labyrinths. This test was also performed in
9 control subjects who had normal stereopsis and no history of strabismus.
Results
Patient ages ranged from 4½ to 51 years. All but 1 patient was younger than 15
years. Seven patients had a history of congenital strabismus that had been treated
with only horizontal muscle surgery. One child with prominent DVD had a history
of perinatal bruising of both eyelids, but had never developed congenital esotropia,
latent nystagmus, or nasotemporal asymmetry. (The term nasotemporal asymmetry

refers to a disparity in monocular nasally directed and temporally directed optoki-
netic responses, with normal nasal optokinetic responses, and impaired temporal
optokinetic responses. This asymmetry is normal within the first 6 months of life
and is retained throughout life in patients with congenital esotropia.) One adult with
no history of horizontal strabismus also had bilateral DVD with no latent nystagmus
or nasotemporal asymmetry. All patients were neurologically normal except for one
prematurely born child with congenital esotropia and delayed walking who proba-
bly had periventricular leukomalacia.
The amplitude of the DVD in the 2 eyes was symmetrical in 5 patients and asym-
metrical in 4 patients. Six patients had grossly visible latent nystagmus. Three
patients had bilateral inferior oblique muscle overaction with a “V” pattern and
bilateral static extorsion of the globes. The one patient with a history of prematurity
had bilateral superior oblique muscle overaction with an “A” pattern and intorsion
of the globes.
On monocular occlusion, 8 patients reported an instantaneous tilt of the pencil
with its top tipped toward the side of the covered eye (Fig. 6.2). This immediate
perception of a tilt seemed to precede any motor movement of the uncovered eye
and was probably caused by a change in perception only. This visual tilt was fol-
lowed quickly by a perceived rotation of the pencil back to vertical, which coincided
with a dorsally directed drift of the covered eye. (Note that if the fixating eye is
making a torsional movement, the subject will perceive the pencil as moving in the
opposite direction that the eye is rotating. Thus, if the subject sees the rotation of the
pencil such that the top goes from being nasally tipped to straight up, this means that
the 12-o’clock meridian of the cornea is rotating nasally. So an apparent extorsional
movement of the pencil would correspond to an intorsional movement of the globe.
This intorsional movement in the fixating eye begins the phenomenon known as
DVD [9–11].) The one patient who did not report a perceived tilt was the adult who
had idiopathic DVD and no other history of strabismus.
In patients who had markedly asymmetrical DVD, the visual tilt was usually
observed only when the dominant eye was covered. All 8 patients who reported a
sensation of visual tilt denied any sensation of head or body tilt. No spontaneous
head tilting was observed during alternate cover testing. Three patients also per-
ceived a sideways movement of the pencil during alternate occlusion. One adult
with DVD perceived no visual tilt or sideways movement on occlusion of either eye.
All 9 control subjects reported a horizontal movement of the pencil without tilt on
alternate occlusion.
Comment
In DVD, monocular occlusion evokes a subjective tilt of the visual environment,

which is followed by a cyclovertical divergence of the eyes and a perceived rotation
of the tilted visual environment back to the vertical. This sequence of perceptual
changes suggests that the cycloversional component of the DVD functions to
Comment 87
Fig. 6.2 Visual tilt evoked by monocular occlusion in the strabismic human with dissociated verti-
cal divergence. Gray image of the pencil denotes perceived visual tilt immediately following
occlusion of one eye. Curved arrows denote the perceived rotation of the tilted visual image back
to vertical, coinciding with the appearance of dissociated vertical divergence. On occlusion of
either eye, the vertical pencil positioned in the sagittal plane is perceived as instantaneously tilted,
with its upper pole tipped toward the side of the covered eye. This subjective visual tilt is quickly
followed by a perceived rotation back to vertical that coincides with the cyclovertical divergence
of the eyes
correct a perceived tilt and to thereby reestablish vertical orientation under condi-
tions of monocular fixation. In this discussion, I adhere to the convention in vestibu-
lar research of describing subjective visual tilt from the point of view of the subject
rather than the examiner, although figures are shown from the perspective of the
examiner to facilitate clinical application. According to this convention, when a
patient looks “to the right,” it is to the patient’s right rather than the examiner’s right,
and so a clockwise rotation of visual environment or a torsional rotation of the
patient’s eyes must also be defined as clockwise from the patient’s perspective.
True vertical corresponds to the gravitational vertical. Our perception of true verti-
cal is influenced by graviceptive input to the 2 labyrinths and visual input to the 2
eyes. Subjective vertical applies to an individual’s internal vertical orientation relative
to true earth coordinates. When the subjective vertical is altered by neurologic disease
or abnormal binocular vision input, the patient will experience a subjective visual tilt,
which is a percieved tilt of the visual environment relative to the subjective vertical
[8]. As shown in Fig. 6.1, increased luminance input to the right eye of a fish tilts the
subjective vertical clockwise (as viewed by the animal), so that a vertical visual stimu-
lus would then appear to be tilted counterclockwise relative to the subjective vertical.
The direction of perceived visual tilt in DVD corresponds to the postural responses of
lateral-eyed animals that exhibit a dorsal light reflex. The same perceptual shift is
reported by strabismic humans with DVD when a binocular visual imbalance is
induced by occlusion of one eye (Fig. 6.2). Thus, with occlusion of the left eye, a right
eye predominance would shift the subjective vertical clockwise (as seen by the
patient), so that true vertical then appears to be rotated counterclockwise relative to
the patient’s altered subjective vertical (Fig. 6.3). This counterclockwise tilt of the
SV SV
Fig. 6.3 Depiction of perceived visual tilt following monocular occlusion in patients with dissoci-
ated vertical divergence. SV indicates the subjective vertical (the patient’s internal representation
of vertical). A perceived tilt of the visual environment (a tilt of the subjective visual vertical) is
determined by the position of vertical objects in the visual world relative to the internal representa-
tion of vertical (ie, relative to the subjective vertical). Left, Occlusion of the left eye evokes a
monocular tilt in the subjective vertical. Since the visual environment is perceived in relation to the
tilted subjective vertical, which the patient perceives as vertical, the monocular visual environ-
ment, as viewed with the right eye, is now perceived as tilted counterclockwise relative to the
patient’s subjective vertical. Right, The human dorsal light reflex is a twofold movement consisting
of a primitive vertical divergence, which realigns the interpupillary axis with the tilted subjective
vertical (as in fish), and a newer cycloversional movement that rotates both eyes torsionally in the
direction of the tilted visual environment. This counterclockwise cycloversional movement (ie,
intorsion of the right eye and extorsion of the left eye) produces a clockwise rotation in the sub-
ject’s tilted visual environment to realign it with the tilted subjective vertical (which the subject
perceives as vertical), thereby annulling the subjective visual tilt
Comment 89
visual world, which is seen monocularly with the uncovered right eye, evokes a cyclo-
vertical divergence movement of the eyes to erase the perceived tilt (Fig. 6.3).
I have proposed that DVD is an atavistic resurgence of the dorsal light reflex that
is evoked by a binocular visual disparity in humans with early-onset strabismus [8].
Since the eyes retain some of their primitive function as balance organs in humans,
unequal visual input induces a central vestibular imbalance in which the internal
sense of vertical no longer corresponds to the gravitational vertical (Fig. 6.3) [8].
Some patients with DVD also have a head tilt away from the side of the hyperdevi-
ated eye [12, 13], which corresponds to the postural component of the dorsal light
reflex in fish [14]. This head tilt, which is not compensatory for binocular vision
[12, 13], may serve to align the head to the tilted internal vertical representation (ie,
the subjective vertical) [14].
Under monocular viewing conditions, patients with DVD may experience a
“schizophrenic” perceptual situation in which the eyes tell the brain that the external
world and the body are tilted relative to the altered internal representation of verti-
cal, while the otoliths tell the brain that the head is upright [8]. A similar sensory
conflict is induced when humans view a tilted visual world under experimental con-
ditions [15]. Under the latter circumstances, the brain strikes a compromise, and the
subjective visual vertical is tilted to an intermediate position between what the eyes
and labyrinths are telling the brain [15].
In DVD, the dorsal rotation of the visually deprived eye corresponds to the ver-
tical divergence induced by the primitive dorsal light reflex. Although the deviating
eye is said to drift “upward” in humans with DVD, the direction of rotation is not
necessarily upward in space but always dorsal relative to the head, regardless of
whether the patient is positioned in the upright, supine, or head-hanging position
[16]. Humans with DVD also display a phylogenetically newer cycloversional
movement that corrects the perceived tilt of the visual environment [17]. This ipsi-
directional cycloversional movement can also be evoked by viewing torsional
optokinetic stimuli or by inducing a static visual tilt of the visual environment
[18–22]. All of these visual stimuli evoke a reflex cycloversional movement, which
serves to align the tilted visual environment with the tilted internal representation
of vertical [22].
Thus, the human dorsal light reflex comprises a twofold movement—a vertical
divergence to realign the interpupillary axis of the eyes relative to the altered inter-
nal representation of the vertical, and a cycloversional movement that torsionally
rotates the eyes in the direction of the tilted visual world to correct the perceived
visual tilt (Fig. 6.3). The vertical component of the human dorsal light reflex is a
primitive adaptation that corresponds to the purely vertical divergence in fish. The
phylogenetically newer cycloversional component conforms to Stephen Jay Gould’s
definition of an exaptation, which is a feature that did not arise as a primary adapta-
tion, but one that was subsequently co-opted or grafted on to meet the newer
demands of evolution (in this case, frontally placed eyes) [23]. According to Gould,
exaptations are features that were not originally built by natural selection for their
current role, but that now enhance fitness (in this case, by restoring vertical orienta-
tion under monocular viewing conditions in the frontal-eyed human) [23].
The human dorsal light reflex demonstrates how the central vestibular system
can uncouple the vertical and torsional components of a cycloversional movement
to allow each component to subserve its corrective function relative to specific con-
ditions of a perceived tilt. During head or body tilt in humans, altered otolithic tone
(ie, unbalanced input from the 2 labyrinths rather than from the 2 eyes) stimulates
elevation and intorsion of the lower eye in space, and depression and extorsion of
the higher eye in space [24]. This stimulus forms the basis of the Bielschowsky
Head Tilt Test [25]. In DVD, however, the elevating eye extorts, and the depressing
eye intorts, producing a cyclovertical divergence in which the torsional eye move-
ments are opposite to those evoked by head tilt [8]. This dissociation could occur
only if the central vestibular system separately processes visual disparity input and
graviceptive input from the trunk and otoliths, then integrates them to establish
subjective vertical orientation as it does in fish [1, 8]. The central ocular motor com-
mand centers must be at liberty to implement separate commands for vertical diver-
gence and cycloversion in humans, although the 2 signals produce a single integrated
extraocular movement.
This study needs to be viewed in light of its inherent limitations. First, it is a
qualitative and subjective study that was performed in a clinic setting. The advan-
tage of this simple method is that it enables the practicing ophthalmologist to
confirm or refute these results without the need for special instrumentation. This
disadvantage is that, because eye movement recordings were not obtained, a pre-
cise and quantitative temporal relationship between the appearance of DVD and
the resolution of the subjective visual tilt could not be confirmed. Second, it is
well recognized that both eyes normally extort in convergence [26]. One could
therefore question whether the perceived image intorsion when the dominant eye
is uncovered might result from each eye being in an extorted convergent position
relative to the vertical object. If this were the case, however, one would expect
the control group to have experienced the same perceptual changes since their
eyes would also extort during convergence. However, this did not occur. Third,
von Helmholtz [27] determined that the subjective vertical retinal meridian is
tilted approximately 1° in each eye, with its top tipped temporally [27]. This
subjective tilt could cause the top of a vertical line to appear to be tipped nasally
(ie, tilted toward the side of the covered eye, as shown in Fig. 6.2) [28]. As such,
it could be argued that this small degree of extorsional tilt in the subjective verti-
cal retinal meridians of each eye could potentially contribute to the perception
that the monocular image appears instanteously intorted when either eye is
uncovered. If this were the case; however, one would again expect the control
group to have perceived a similar visual tilt. Yet, only patients with DVD per-
ceived a visual tilt under monocular conditions, suggesting that this effect did not
influence the results of this study. Furthermore, the original von Helmholtz mea-
surements of the subjective retinal vertical, like standard horopter measurements,
were obtained using isolated visual stimuli in the absence of any surrounding
contextual cues. In this study, normal background contextual cues were present,
so one would not expect patients in either group to detect this small vertical bias.
Comment 91
Fourth, it is not clear why the group of patients with DVD were less likely than
the control group to report a perceived lateral movement of the pencil when the
cover was switched from one eye to the other. It could be that the perception of
tilt overrides the perception of lateral movement in these patients; that surgically
corrected congenital esotropia is associated with an altered perception of visual
space under monocular conditions; or that I was more attuned to elucidating the
direction and sequential changes of the perceived tilt in patients with DVD, and
that verbal or nonverbal cues could have influenced patient responses. Last, one
could argue that if DVD is not a human dorsal light reflex, this cyclovertical
divergence could be the cause rather than the result of the observed perceptual
changes. In other words, extorsion of the covered eye in DVD could cause the
vertical pencil to appear momentarily intorted when the either eye is uncovered.
As the eye intorts to fixate, the image would extort, producing a similar sequence
of perceptions as observed. If this were the case, one would expect patients with
DVD to report that the image of the pencil shifted from an initial position of
intorsion to a final position of extorsion. However, all patients with DVD stated
that the tilted image of the pencil appeared vertical following its rotation, which
would only occur if this reflex cycloversional movement served to correct the
monocular visual tilt. This perceptual response suggests that this complex cyclo-
vertical movement must function to reestablish vertical visual orientation when
binocular vision is preempted.
In conclusion, DVD is associated with a subjective tilt of the visual environment
and a reflex cyclovertical divergence of the eyes. This subjective visual tilt appears
to drive both components of the resulting cyclovertical divergence. These percep-
tual correlates add to the accumulating body of evidence that DVD is a human
dorsal light reflex, which serves to restore vertical visual orientation under mon-
ocular conditions. In DVD, a subjective sensation of visual tilt under monocular
viewing conditions evokes 2 compensatory eye movements—a phylogenetically
older vertical divergence movement (ie, a primitive adaptation) to realign the eyes
relative to the altered internal representation of vertical, and an exaptive cyclover-
sion movement that torsionally rotates the eyes in the direction of the tilted visual
environment to restore vertical visual orientation by neutralizing the perceived
visual tilt. The vertical component of this movement corresponds to the ancestral
dorsal light reflex in fish, while the cycloversional component of the human dorsal
light reflex appears to be an exaptation that functions to annul the subjective visual
tilt under monocular conditions when the eyes are frontally placed. This twofold
reflex movement corresponds both in theory and in actuality to the ocular motor
response that would result from a tilted internal representation of the visual verti-
cal. In the human dorsal light reflex, the direction of the cycloversion movement
relative to the vertical divergence is opposite to that observed during a head tilt in
space, demonstrating that the vertical divergence and the cycloversional compo-
nent of visual tilt are independently programmed, and that these 2 extraocular
movements can be dissociated by unequal visual input to the 2 eyes in humans with
congenital strabismus.
References
1964;1:77–111.
J Comp Physiol. 1983;150:473–81.
4. Duke-Elder S. The effect of light on movement. In: Duke-Elder S, editor.
System of ophthalmology: the eye in evolution. London: Henry Klimpton;
1958. p. 27–81.
5. von Holst E. Über den Lichtrückenreflex bei Fische. Pubbl Staz Zool Napoli.
1935;II(15):143–8.
1935;37:109–14.
7. Mittelstaedt H. Interaction of eye-, head-, and trunk-bound information in spa-
tial perception and control. J Vestib Res. 1997;7:283–302.
9. Van Rijn LJ, Collewijn H. Eye torsion associated with disparity-induced verti-
10. Inoue M, Kita Y. Eye movements in dissociated vertical deviation. Nippon
Ganka Gakkai Zasshi. 1993;97:1312–9.
11. Guyton DL, Cheeseman EW, Ellis FJ, et al. Dissociated vertical deviation: an
exaggerated normal eye movement used to to damp cyclovertical latent nystag-
mus. Trans Am Ophthalmol Soc. 1998;96:389–429.
13. Lang J. Squint dating from birth or with early onset. In: Transactions of the
First International Congress of Orthoptists. London: Henry Klimpton; 1968.
p. 231–7.
15. Mittelstaedt H. The subjective visual vertical as a function of visual and extra-
retinal cues. Acta Psychol. 1986;63:63–85.
16. Goltz HC, Irving EL, Hill JA. Dissociated vertical deviation: head and body
orientation affect the amplitude and velocity of the vertical drift. J Pediatr
17. van Rijn LJ, Simonsz HJ, ten Tusscher MPM. Dissociated vertical deviation
and eye torsion: relation to disparity-induced vertical vergence. Strabismus.
1997;5:13–20.
18. Brecher GA. Die optokinetische Auslösung von Augenrollung und rotato-
rischem Nystagmus. Pfugers Arch Gesamte Physiol Menschen Tiere.
1934;234:13–28.
19. Crone RA. Optically-induced eye torsion, II: optostatic and optokinetic cyclo-
version. Albrecht Von Graefes Arch Klin Exp Ophthalmol. 1975;196:1–7.
Postscript 93
20. Crone RA, Everhard-Halm Y. Optically-induced eye torsion, I: fusional cyclo-

vergence. Albrecht Von Graefes Arch Klin Exp Ophthalmol. 1975;195:231–9.
21. Goodenough DR, Sigman E, Oltman PK, et al. Eye torsion in response to a
tilted visual stimulus. Vis Res. 1979;19:1177–9.
22. Brodsky MC. Do you really need your oblique muscles? adaptations and exap-
tations. Arch Ophthalmol. 2002;120:820–8.
23. Gould SJ. Exaptation: a crucial tool for evolutionary psychology. J Soc Issues.
1991;47:43–65.
24. Kori AA, Schmid-Priscoveanu A, Straumann D. Vertical divergence and coun-
terroll movements evoked by whole-body position steps about the roll axis of
the head in humans. J Neurophysiol. 2001;85:671–8.
25. Hofmann FB, Bielschowsky A. Die Verwertung der Kopfneigung zur Diagnose
der Augenmuskellähmungen aus der Heber und Senkergruppe. Graefes Arch
Ophthalmol. 1900;51:174.
26. Allen MJ, Carter JH. The torsion component of the near reflex. Am J Optom.
1967;44:343–9.
27. von Helmholtz H. Handbuch der Physiologischen Optik. Vol 3. Treatise on
physiological optics (trans: Southall JPC). Rochester: Optical Society of
America; 1925.
28. von Tschermak-Seysenegg A. Introduction to physiological optics (trans:

Boeder P). Springfield: Charles C Thomas; 1952. p. 140–2.
Postscript
This article deals with the complex issue of why the torsional eye movements in
DVD are opposite in direction to those produced by the vestibular pathways. In my
original analysis of DVD, I had explained this phenomenon teleologically by stating
that the torsional directions between vestibular and visual pathways had to be oppo-
site in direction to summate to zero under conditions of normal functioning. In this
analysis, I demonstrate that monocular occlusion is associated with a subjective
sensation of tilt, which is rapidly corrected by the DVD. The absence of any subjec-
tive tilt after a DVD torsional movement occurs suggests that these movements are
corrective in nature (i.e. that DVD is indeed a righting reflex).
From there, I use the direction of the perceived visual tilt to deconstruct the
vertical component of DVD, which rotates the eyes back to the altered internal
orientation for true vertical (i.e. toward the subjective vertical) from torsional
component (which is directed toward and thereby corrects the perceived tilt of the
visual world (termed the subjective visual vertical). Whereas the perceived true
vertical and the position of the world rotate in the same direction during body tilt,
in visual tilts they scissor apart, necessitating torsional movements that are oppo-
site in direction to those produced by body tilt. The neurologic pathways by which
these visual torsional movements are separately mediated in humans remain
unknown.
In its essence, DVD is a tilt rather than a hyperdeviation. If the reader tilts his or
her head to the left and looks in the mirror, they will find that the right eye is higher
than the left in space. Do they suddenly have a large right hypertropia? No, they
have a tilt with no true hypertropia. The true hypertropia is actually the magnitude
of the total vertical deviation in space minus the tilt. That is why you can have a true
hypertropia superimposed upon a DVD. The vertical component of DVD is there-
fore a tilt in physical space to realign the eyes with internal orientation of the indi-
vidual. On top of this, the gyroscopic torsional component of DVD creates a roll
movement to restore restore vertical orientation of the outside world. Despite its
atavistic origin, DVD reflects complex computational processing at many levels
within the visual system.
The Reversed Fixation Test
A Diagnostic Test for Dissociated Horizontal Deviation
7
The adjective “Dissociated” was first used to describe binocular eye movements by
Bielschowsky; he applied the term “dissociated vertical divergence” to the alternat-
ing hyperdeviation that accompanies congenital strabismus [1–3]. In 1976, Raab [4]
described the slow unilateral abduction of the deviating eye as a horizontal variant
of dissociated vertical divergence. In 1990, Spielmann [5] assigned the name “dis-
sociated horizontal deviation” to patients with infantile strabismus with intermittent
esodeviation of either eye. Spielmann noted that some esodeviations are smaller
when the child is visually inattentive than when the child is fixating or during cover
testing. This finding suggested that monocular fixation can increase esotonus in
some patients with infantile strabismus [5].
Beginning also in 1990, several clinical reports applied the term “dissociated
horizontal deviation” to intermittent exodeviations that were larger in 1 eye or
confined to 1 eye on alternate cover testing, ie, dissociated [6–11]. Affected
patients manifested an exodeviation of variable amplitude [6–11]. Associated
findings such as dissociated vertical divergence, latent nystagmus, and sensorial
suppression of 1 eye (even when the eyes were aligned) also distinguished this
form of dissociated horizontal deviation from intermittent exotropia [6–8]. The
exodeviation was noted to be larger during periods of visual inattention [9]. The
authors of these reports advocated limiting surgery to a single lateral rectus mus-
cle in the exodeviating eye (recession with or without a posterior fixation suture)
for unilateral cases of dissociated horizontal deviation, while reserving bilateral
lateral rectus muscle recession for cases of so-called bilateral dissociated horizon-
tal deviation or unilateral dissociated horizontal deviation combined with exotro-
pia [6–11].

96 7 The Reversed Fixation Test
Dissociated horizontal deviation has to be defined as the horizontal vergence that

is brought about by a change in the balance of visual inputs derived from the right
and left eyes. The etiology of dissociated horizontal deviation is open to conjecture.
Zubcov et al12 proposed that greater degrees of convergence might be used to damp
larger degrees of latent nystagmus that develop when the eye with poorer vision is
fixating. However, it is equally possible that the latent nystagmus is not causal and
that the change in vergence results from the fixation shift from one eye to the other,
with the unequal damping of latent nystagmus occurring as a side effect [13].
Depending on the baseline horizontal deviation of the eyes under binocular condi-
tions, the same fixation-induced vergence mechanism could manifest as a dissoci-
ated esotropia or dissociated exotropia.
Many cases of asymmetrical esodeviation and exodeviation are attributable to
conditions that simulate dissociated horizontal deviation [13, 14]. For example,
many patients with asymmetrical exodeviations have a history of strabismus sur-
gery for infantile esotropia [11, 15]. In such cases, a slipped, overrecessed, or weak
medial rectus muscle or a tight lateral rectus muscle can produce an incomitance
with an increase of the squint angle when the eye with impaired motility is forced to
fixate. Dissociated horizontal deviation must also be considered in the postoperative
patient with congenital strabismus who develops an intermittent exodeviation that
spontaneously changes to esotropia [16, 17]. Before considering this diagnosis, the
examiner must also exclude a convergence substitution movement in patients with
impaired volitional gaze [13, 14].
The examiner must also place the patient in the full cycloplegic refraction to
exclude unequal accommodative convergence caused by uncorrected anisometro-
pia. For example, a patient with anisohyperopia (plano OD; + 5.00 OS) may show
15 prism diopters of exotropia when fixating with the right eye, and 10 prism diop-
ters of esotropia when fixating with the left eye.
Thus, to confirm the diagnosis of dissociated horizontal deviation, the head
posture, the direction of gaze, the fixation distance, and the degree of accom-
modation must remain unchanged as fixation switches from one eye to the other
[13, 14]. The reversed fixation test, as described by Mattheus and Kommerell
[15–17], was developed to measure dissociated deviations. This test makes it
possible to rule out simulating conditions. The performance of this test with
prisms and its outcomes in different situations of horizontal deviation are
described below.
Reversed Fixation Test
Consider a patient who has had surgery for infantile esotropia and subsequently
developed an intermittent exodeviation of the left eye. Alternate cover testing shows
that only the left eye drifts out under the cover. The reversed fixation test is now
necessary to confirm or rule out the presence of a dissociated deviation.
Scenario 2 97
Scenario 1
Step 1: A base-in prism is placed before the left eye to neutralize the exodevia-
tion. The left eye is covered for about 5 seconds. The cover is then switched briefly
to the right eye then immediately back to the left eye to confirm that the exodevia-
tion is neutralized. If not, the prism is changed and the procedure repeated until the
exodeviation of the left eye is neutralized, eg, by 25 prism diopters. With the prism
held before the left eye, the occluder is moved to cover the right eye. No movement
of the left eye is seen (Fig. 7.1, top).
Step 2: (Reversed fixation test): After about 5 seconds, the occluder again is
briefly switched from the right eye to the left eye (with the prism still held in place
in front of the left eye). No movement of the right eye is observed (Fig. 7.1, bot-
tom). This means that the 25 prism diopters are also corrective for left eye fixation.
Interpretation: The absence of right eye movement in the reversed fixation test
demonstrates that the exotropia is not dissociated. Rather, this patient may have an
adduction deficit of the right eye due to postoperative slippage of the right medial
rectus muscle.
Scenario 2
Step 1: A 25 prism diopter base-in prism placed before the left eye is found to
neutralize the exodeviation. With the 25 prism diopter base-in prism still in front of
the left eye, the right eye is covered for several seconds (Fig. 7.2, top). No move-
ment of the left eye is seen.
Step 2: (Reversed fixation test): With the prism still in front of the left eye, after
several seconds the occluder is switched from the right eye to the left eye. An abduc-
tion saccade of the right eye is observed, corresponding to an adducted position of
the right eye under the cover (Fig. 7.2, bottom). This adducted position can be mea-
sured by having a second observer (or the patient) place a base-out prism in front of
the right eye while keeping the right eye occluded. Then the cover is for a short
moment switched to the left eye and immediately back to the right eye to observe
whether the adducted position of the right eye is neutralized. If not, the prism is
changed and the procedure repeated until the adducted position is neutralized, eg,
by a 25 prism diopter base-out prism.
Interpretation: In this situation, there is less esotonus when the right eye is fixat-
ing than when the left eye is fixating. Thus, when the left eye is made to fixate in its
unchanged, abducted position (through the prism), the right eye assumes an adducted
position of the same size behind the cover, ie, there is no squint angle. This patient
has a dissociated component of 25 prism diopters which equals the entire exodevia-
tion of the left eye (provided that unequal accommodative convergence is excluded
by dynamic retinoscopy or by maximum visual acuity at distance with the patient in
his or her corrected cycloplegic refraction). If there is no horizontal incomitance,
the patient will also be orthotropic in straight gaze as long as the left eye is fixating.
25∆
25∆
Fig. 7.1 Intermittent exotropia. Top: Following neutralization with a 25 prism diopter base-in
prism, no change in esotonus occurs when the occluder is shifted to the right eye. Bottom:
Therefore, no movement of the right eye is seen when the occluder is again shifted to the left eye,
demonstrating that the exotropia is not dissociated
Scenario 2 99
25∆
25∆ 25∆
Fig. 7.2 Dissociated horizontal deviation. Top: Following neutralization with a 25 prism diopter
base-in prism, the esotonus increases when the occluder is shifted to the right eye. Bottom:
Reversed fixation test. When the occluder is again shifted to the left eye, a 25 prism diopter abduc-
tion saccade of the right eye is seen, demonstrating the dissociated nature of the exodeviation
Scenario 3
Step 1: A 25 prism diopter base-in prism is placed before the left eye to neutral-
ize the exodeviation. With the prism held before the left eye, the occluder is moved
to cover the right eye. No movement of the left eye is seen (Fig. 7.3, top).
Step 2: (Reversed fixation test): With the prism still in front of the left eye, the
occluder is briefly switched from the right eye to the left eye, then immediately back
to the right eye. An abduction saccade of the right eye is observed, corresponding to
an adducted position of the right eye under the cover (Fig. 7.3, bottom). When mea-
sured with a second observer (or the patient) placing a base-out prism in front of the
right eye, the size of the abduction saccade is 10 prism diopters.
Interpretation: Assuming that unequal accommodative convergence has been
ruled out, this patient has 10 prism diopters of dissociated horizontal component
superimposed on 25 prism diopters of underlying exotropia. The patient who spon-
taneously alternates fixation, will alternately manifest an exodeviation of 15 prism
diopters in the right eye and an exodeviation of 25 prism diopters in the left eye.
Scenario 4
Consider the patient with the same history who manifests a large exodeviation dur-
ing periods of visual inattention. Alternate cover testing shows a large exodeviation
of the left eye and a small esodeviation of the right eye.
Step 1: A 25 prism diopter base-in prism is placed before the left eye to neutral-
ize the exodeviation. With the prism held before the left eye, the occluder is moved
to cover the right eye. No movement of the left eye is seen (Fig. 7.4, top).
Step 2: (Reversed fixation test): With the prism still in front of the left eye, the
occluder is switched from the right eye to the left eye. An abduction saccade of the
right eye is observed, corresponding to an adducted position of the right eye under
the cover. When measured with a second observer (or the patient) placing a base-out
prism in front of the right eye, the size of the abduction saccade is 30 prism diopters
(Fig. 7.4, bottom).
Interpretation: Assuming that unequal accommodative convergence has been
ruled out, this patient alternately manifests an exotropia of 25 prism diopters in the
left eye and an esotropia of 5 prism diopters in the right eye, attributable to a dis-
sociated component of 30 prism diopters.
Comment
The dissociated component of congenital strabismus may be inapparent on initial

examination and variable under binocular conditions. The reversed fixation test was
devised by Mattheus and Kommerell [15–17] as a technique to visualize the dissoci-
ated component in patients with dissociated vertical divergence. The reversed fixa-
tion test is particularly useful for distinguishing dissociated vertical divergence
Comment 101
25∆
25∆ 10∆
Fig. 7.3 Dissociated horizontal deviation. Top: Following neutralization with a 25 prism diopter
base-in prism, the esotonus increases when the occluder is shifted to the right eye. Bottom:
Reversed fixation test. When the occluder is again shifted to the left eye, a 10 prism diopter abduc-
tion saccade of the right eye is seen, demonstrating the partially dissociated nature of the
exodeviation
25∆
25∆ 30∆
Fig. 7.4 Dissociated horizontal deviation with coexistent exodeviation and esodeviation. Top:
Following neutralization with a 25 prism diopter base-in prism, the esotonus significantly increases
when the occluder is shifted to the right eye. Bottom: Reversed fixation test. When the occluder is
again shifted to the left eye, a 30 prism diopter abduction saccade of the right eye is seen, demon-
strating that this patient has 25 prism diopters of exotropia (causing the exodeviation of the left
eye) with an additional dissociated component of 30 prism diopters (causing the 5 prism diopter
esodeviation of the right eye during left eye fixation as shown in the top figure)
Comment 103
from the nondissociated vertical divergence caused, for example, by primary oblique
muscle overaction [17].
In the horizontal plane, a major advantage of the reversed fixation test is that an
incomitant horizontal deviation will not be misdiagnosed as “dissociated.” This test
allows the examiner to measure a dissociated component as the difference between
the squint angle when the right eye is fixating and the angle when the left eye is
fixating, without any change of the horizontal gaze direction. If the reversed fixation
test shows that the angle is the same on right and left eye fixation, a dissociated
component of the deviation has been ruled out. Since fixation per se (with either
eye) can evoke differing degrees of esotonus, the degree of exodeviation that is
apparent with visual inattention (eg, when the patient is asked to remember an event
24 hours ago or to solve a mathematical task) can be larger than that measured when
either eye is used to fixate [11, 14].
The reversed fixation test shows that dissociated horizontal deviation exists in a
minority of patients with unilateral exodeviation [13, 14]. While this disorder is
referred to as dissociated horizontal deviation in the left or the right eye, it is well to
remember that the dissociated component results from a change in vergence tonus
on a supranuclear level when the fixating eye is switched, and that it actually
involves both eyes.
In conclusion, dissociated horizontal deviation is a unique clinical disorder that
can be diagnosed only after a variety of other simulating conditions have been
excluded. The reversed fixation test is a decisive test for the diagnosis of dissociated
horizontal deviation. With the reversed fixation test, keeping the neutralizing prism
in its position in front of the same eye, a change in the fixation from one eye to the
other normally does not alter the strabismus angle. With corrected cycloplegic
refraction, when prism alternate cover testing discloses a movement of the eye not
viewing through the prism, a dissociated component of the strabismus is established.
References
1. Bielschowsky A. Über die Genese einseitiger Vertikalbewegungen der Augen.
Z Augenheilkd. 1904;12:545–57.
2. Bielschowsky A. Die einseitigen und gegensinnigen (“dissoziierten”)
Vertikalbewegungen der Augen. Albrecht Von Graefes Arch Ophthalmol.
1930;125:493–553.
3. Bielschowsky A. Disturbances of the vertical motor muscles of the eye. Arch
Ophthalmol. 1938;20:175–200.
4. Raab EL. Dissociated vertical deviation in combined motor anomalies. In: Fells
P, editor. Proceedings of the Second Congress of the International
Strabismological Association. Marseilles: Diffusion Générale de Libraire;
1976. p. 189–93.
5. Spielmann A. Déséquilibres verticaux et torsionnels dans le strabisme précoce.
Bull Soc Ophtalmol Fr. 1990;4:373–84.
6. Romero-Apis D, Castellanos-Bracamontes A. Desviacion horizontal disociada
(DHD). Rev Mex Oftalmol. 1990;64:169–73.
7. Quintana-Pali L. Desviacion horizontal disociada. Bol Hosp Oftalmol.

1990;42:91–4.
9. Romero-Apis D, Castellanos-Bracamontes A. Dissociated horizontal deviation:
clinical findings & surgical results in 20 patients. Binocul Vis Strabismus
Q. 1992;7:173–8.
10. Wilson ME. The dissociated strabismus complex. Binocul Vis Strabismus

Q. 1993;8:45–6.
11. Wilson ME, Hutchinson AK, Saunders RA. Outcomes from surgical treatment
for dissociated horizontal deviation. J AAPOS. 2000;4:94–101.
12. Zubcov AA, Reinecke RD, Calhoun JH. Asymmetric horizontal tropias, DVD,
and manifest latent nystagmus: an explanation of dissociated horizontal devia-
tion. J Pediatr Ophthalmol Strabismus. 1990;27:59–64.
13. Gräf M. Dissociated horizontal deviations (DHD): terminology, diagnosis and
etiology. In: Faber TJ, editor. Transactions of the 26th Meeting of the European
Strabismological Association, Florence, June 2001. Tokyo: Swets & Zeitlinger;
2001. p. 105–8.
14. Gräf M. Dissociated horizontal deviations (DHD): nomenclature and etiology:
an orientating attempt in terminological confusion. Klin Monatsbl Augenheilkd.
2001;218:401–5.
15. Mattheus S, Deberitz I, Kommerell G. Differentialdiagnose zwischen inkomi-
tierendem und dissoziiertem Schielen. Arbeitskreis Schielbehandlung
Berufsverband Augenärzte Deutschlands. 1978;10:135–7.
16. Kommerell G, Mattheus S. Reversed fixation test (RFT), a new tool for the
diagnosis of dissociated vertical deviation. In: Reinecke RD, editor. Strabismus.
Orlando: Grune & Stratton; 1984. p. 721–8.
17. Mattheus S, Kommerell G. Reversed fixation test as a means to differentiate
between dissociated and nondissociated strabismus. Strabismus. 1996;4:3–9.
Postscript
This paper documented the reversed fixation test (RFT) as a determinative test for
dissociated horizontal deviation (DHD), as first articulated by Michael Gräf in
2001. Note that an equal amount of esotonus exerted by fixation with each of the
two eyes would yield a negative reversed fixation test in the presence with
DHD. Such cases would have equal exodeviations with each eye fixing. Thus, some
cases of DHD will inevitably fly under our radar.
After further study, it became clear to me that the diagrams in this article incor-
rectly depict large shifts in eye position corresponding to the reversed fixation test
(RFT). In point of fact, they are tiny, on the order of 2–8 PD. This tiny shift occurs
because the reversed fixation test measures only the difference in dissociated esoto-
nus that is exerted when each of the two eyes are fixating. Since fixation with one
eye is already exerting considerable esotonus, and the deviation has been corrected
Postscript 105
by a base in prism, the switch to fixation with the eye that exerts greater or lesser
esotonus would show a movement that reflects the difference. Thus, even a small
shift is still significant for DHD.
By inference, when intermittent exotropia is accompanied by a positive reversed
fixation test, one can probably conclude that its onset was within the first six months
of life. However, other cases of intermittent exotropia are accompanied by DVD
(either manifest or measurable on prism and alternate cover testing). This associa-
tion may reflect that fact that DVD is a primitive visual reflex that is present all of
us, so if intermittent exotropia begins early in life, DVD can be coexpressed as well.
Does Infantile Esotropia Arise
From a Dissociated Deviation? 8
Tonus refers to the effects of baseline innervation on musculature in the awake, alert
state [1]. Since the normal anatomical resting position of the eyes is one of exode-
viation, extraocular muscle tonus plays a vital physiologic role in establishing ocu-
lar alignment. Under normal conditions, binocular esotonus is superimposed on the
baseline anatomical position of rest to maintain approximate ocular alignment, save
for a minimal exophoria that is easily overcome by active convergence. When bin-
ocular visual input is preempted early in life, monocular fixation may give rise to a
larger dissociated esotonus that gradually drives the 2 eyes into a “convergent” posi-
tion, resulting in infantile esotropia [2].
In our companion article [2], we examine clinical and evolutionary evidence for
the proposition that dissociated horizontal deviation is a clinical expression of dis-
sociated esotonus. When superimposed on a baseline orthoposition, dissociated
esotonus manifests as an intermittent esotropia that is asymmetrical or unilateral
[3]. More commonly, dissociated esotonus is superimposed on a baseline exodevia-
tion, producing an intermittent exodeviation that is asymmetrical, unilateral, or
associated with a paradoxical esodeviation when the nonpreferred eye is used for
fixation [4–11].
Although the term dissociated has historically been restricted to the description
of vergence eye movements [12–14], in a more general sense it describes any ocular
movements that result from a change in the relative balance of visual input from the
2 eyes [15]. These movements arise almost exclusively in the setting of infantile
strabismus [16], which has a strong predilection for esotropia over exotropia. It is
held that infantile esotropia disrupts binocular control mechanisms and thereby
engenders these dissociated eye movements [16]. This time-honored notion assumes
a distinct and unrelated pathogenesis for infantile esotropia.
The purpose of this analysis is to raise an unexamined question regarding the
pathogenesis of infantile esotropia. Since dissociated deviations almost uniquely
accompany infantile strabismus, could infantile esotropia arise from a dissociated
deviation? Our findings raise the possibility that dissociated esotonus could be the
proximate cause of infantile esotropia.
108 8 Does Infantile Esotropia Arise From a Dissociated Deviation?
Contrary to the stereotype of “congenital” esotropia as a large-angle deviation

that is present at birth, most cases are acquired (ie, “infantile” in origin) [17, 18].
Furthermore, the eyes do not simply snap in to their final esotropic position. Before
12 weeks of age, nascent infantile esotropia is an intermittent, variable esodeviation
that gradually becomes constant after building in intensity to a large fixed angle of
horizontal misalignment [17, 18]. Ing [19] has noted that 50% of patients with
infantile esotropia show an increase in the measured angle between the time of first
examination and the date of surgery. Clearly, unequal visual input in infancy must
produce a gradual and progressive increase in the angle of esotropia. That this
esodeviation appears during the early period when stereopsis is developing, but
before macular anatomy has matured sufficiently to provide high-resolution acuity
[20], suggests that it is actively driven primarily by an imbalance in peripheral
visual input.
In a recent hypothesis, Guyton [21] has invoked vergence adaptation and muscle
length adaptation to explain how a small innervational bias (such as the convergence
produced by increased accommodative effort in the presbyope) can build slowly
over time into a large constant deviation. Vergence adaptation refers to the tonus
levels that normally operate to maintain a baseline ocular alignment and thereby
minimize retinal image disparity. According to Guyton, vergence adaptation can
allow primitive ocular motor biases to gradually amplify and create strabismic devi-
ations under pathological conditions [21]. Muscle length adaptation refers to the
change in extraocular muscle length due to gain or loss of sarcomeres. Muscle
length adaption is driven in part by the physiologic effects of vergence adaptation.
Our results suggest that dissociated esotonus could provide the sensorimotor
substrate for vergence adaptation when binocular cortical control mechanisms fail
to take hold. The finding of a positive Bielschowsky phenomenon in dissociated
horizontal deviation [5, 8] shows that peripheral luminance reflexes are retained, as
in dissociated vertical divergence [22]. In this setting, both peripheral (luminance
and optokinetic) and central (fixational) reflexes augment dissociated esotonus and
lead over time to infantile esotropia. Subcortical visual reflexes would provide the
default system through which dissociated esotonus operates to reestablish the base-
line horizontal eye position. This process can ultimately lead to loss of sarcomeres
and secondary shortening of the medial rectus muscles. The fact that the eyes
straighten to an almost normal baseline position under general anesthesia [23–27],
however, suggests that esotonus is the driving force for infantile esotropia and that
mechanical effects play a secondary role in its pathogenesis. It is therefore possible
that the stable large-angle esodeviation that we recognize as infantile esotropia sim-
ply represents the final stage of dissociated esotonus. As with many other forms of
ocular misalignment, the constant esodeviation that develops over time may eventu-
ally obscure the pathogenesis.
Early monocular visual loss is known to generate esotonus and reproduce the
same constellation of dissociated eye movements that accompany infantile esotro-
pia [25]. Patients with unilateral congenital cataract often develop large-angle eso-
tropia, latent nystagmus, dissociated vertical divergence, and a head turn to fixate in
adduction with the preferred eye [25]. By contrast, early infantile esotropia is often
Does Infantile Esotropia Arise From a Dissociated Deviation? 109
characterized by similar visual acuity in the 2 eyes, with alternating suppression of

the nonfixating eye. So perhaps dissociated horizontal deviation is not an epiphe-
nomenon of infantile esotropia but a “footprint in the snow” of the dissociated esot-
onus that is responsible for its inception.
There remains the unfortunate tendency in the strabismus literature to conflate
esotonus of the eyes as a baseline innervation with convergence of the eyes as an
active function. Jampolsky [28, 29] has emphasized the mechanistic importance of
distinguishing between convergence as an active binocular function and esotonus as
a baseline innervational state that is centrally driven by unequal visual input to the
2 eyes. The importance of this distinction lies in understanding that convergence
implies a deviation from baseline under normal conditions of sensory input, whereas
tonus implies a return to baseline under altered conditions of sensory input. The
distinction between convergence (the effect) and monocular esotonus (the cause)
lies at the heart of understanding infantile esotropia. Horwood and colleagues have
recently shown that normal infants display fleeting large-angle convergence eye
movements during the first 2 months of life and that these spontaneous convergence
movements are ultimately predictive of normal binocular alignment [30]. By con-
trast, infantile esotropia tends to increase over the period when this excessive con-
vergence is disappearing in normal infants [31]. This time course challenges the
dubious assumption that infantile esotropia arises from excessive convergence out-
put. The evidence for dissociated esotonus suggests that we retain a primitive tonus
system, independent of convergence output, that can operate under conditions of
unequal visual input to reset eye position to a new baseline “convergent” position.
This mechanism would explain why infantile esotropia is so much more common
than infantile exotropia.
If the dissociated esotonus that manifests as dissociated horizontal deviation
gives rise to infantile esotropia, why does dissociated horizontal deviation manifest
as an intermittent exotropia? Although we use the term intermittent exotropia diag-
nostically, it is ultimately a descriptive term comprising a variety of conditions with
different diagnostic implications. The intermittent exodeviation caused by dissoci-
ated horizontal deviation simply constitutes one distinct form of intermittent exotro-
pia with its own unique pathophysiology.
Many clinicians apply the hybrid term intermittent exotropia/dissociated hori-
zontal deviation, implying that the 2 conditions often coexist, and perhaps acknowl-
edging some diagnostic ambiguity [5–10]. So what are the innervational substrates
for these distinct but overlapping categories of intermittent exotropia? Although
Burian [32] believed intermittent exotropia to be caused by an active divergence
mechanism, independent studies have found that these patients are approximately
30 prism diopters more exotropic when deeply anesthetized than in the awake state
[26, 27], suggesting that intermittent exotropia actually results from intermittent
fusional control of a large baseline exodeviation [33, 34].
When intermittent exotropia is associated with dissociated horizontal deviation,
fixation with either eye superimposes dissociated esotonus on the baseline exode-
viation to produce a variable intermittent exodeviation [2]. The distinction between
nondissociated intermittent exotropia and dissociated horizontal deviation lies
primarily in the relative activation of binocular fusion (which behaves as an all-or-

nothing phenomenon in most forms of intermittent exotropia) vs dissociated esoto-
nus (which functions as an open-loop process without reference to ultimate binocular
alignment in dissociated horizontal deviation). Because fixation with the nonpre-
ferred eye exerts greater esotonus [2], the baseline exodeviation can be unilateral,
asymmetrical, or associated with a paradoxical esotropia when the nonpreferred eye
is used for fixation.
Infantile esotropia and intermittent exotropia are universally regarded as distinct
forms of strabismus that occupy opposite points on a clinical spectrum. In contrast
to infantile esotropia, intermittent exotropia usually has a later onset and is rarely
associated with prominent dissociated eye movements (although small degrees of
dissociated vertical divergence can be detected) [35]. At first glance, it is difficult to
imagine how these diametrical forms of horizontal misalignment are not mutually
exclusive.
The beauty of dissociated horizontal deviation is that it allows us to recast hori-
zontal strabismus as the relative balance of mechanical and innervational forces,
without regard to final eye position. Dissociated esotonus can still be expressed
from an exodeviated position, because it is generated by unbalanced binocular input
that exerts its influence on any baseline deviation. Consequently, intermittent exo-
tropia is a common clinical manifestation of dissociated esotonus. Mechanistically,
there is nothing sacred about orthotropia as a clinical demarcation and nothing sig-
natory about the direction of horizontal misalignment.
In this light, dissociated horizontal deviation is transformed from a clinical curi-
osity to a fundamental piece of the puzzle for understanding horizontal strabismus.
The exotropic form of dissociated horizontal deviation uniquely embodies the coex-
istence of the mechanical exodeviating forces that give rise to intermittent exotropia
and the dissociated esotonus that may give rise to infantile esotropia. For example,
infantile exotropia is often accompanied by dissociated eye movements such as
latent nystagmus and dissociated vertical divergence [36, 37]. Some infants exhibit
an intermittent form of exotropia with other dissociated eye movements [38], sug-
gesting a component of dissociated horizontal deviation. Patients with primary dis-
sociated horizontal deviation also display an intermittent exodeviation of one or
both eyes with other signs of dissociation [6].
All of these conditions share a common pathophysiology wherein dissociated
esotonus is superimposed on a baseline exodeviation to produce an intermittent
exodeviation that varies in size depending on which eye is used for fixation. In
patients without binocular fusion, dissociated esotonus can cause a constant exode-
viation to appear intermittent. In patients who retain binocular fusion, it can pro-
duce a combined clinical picture of intermittent exotropia (with intermittent fusion),
an asymmetrical exodeviation of the 2 eyes, or an exodeviation of the nonpreferred
eye with a paradoxical esodeviation of the preferred eye. In classifying these disor-
ders pathogenetically, it is critically important to distinguish sensorimotor factors
from the different forms of ocular misalignment that they ultimately produce.
Dissociated horizontal deviation shows us how it is only the resultant horizontal
deviations, and not the underlying conditions, that are diametrically opposed.
Does Infantile Esotropia Arise From a Dissociated Deviation? 111
In conclusion, our findings raise the intriguing possibility that dissociated esoto-
nus, an unrecognized dissociated eye movement, may be the cause, rather than the
effect, of infantile esotropia. If this proves to be the case, then the prevailing concept
of infantile esotropia as the proximate cause of dissociated deviations may need to
be revised.
References
1. To cross or not to cross: the proceedings of the Ocular Motor Tonus Symposium
sponsored by the Smith-Kettlewell Eye Research Institute, Tiberon, CA, June
2–4, 2006. p. 2–4. http://www.ski.org/Tonus/index.html.
2. Brodsky MC, Fray KJ. Dissociated horizontal deviation after surgery for infan-
tile esotropia: clinical characteristics and proposed pathophysiologic mecha-
nisms. Arch Ophthalmol. 2007;125(12):1683–92.
3. Spielmann A. Déséquilibres verticauxet torsionnels dans le strabisme précoce.
Bull Soc Ophtalmol Fr. 1990;4:373–84.
4. Romero-Apis D, Castellanos-Bracamontes A. Desviacion horizontal disociada
(DHD). Rev Mex Oftalmol. 1990;64:169–73.
Ophthalmol Strabismus. 1991;28(2):90–5.
6. Romero-Apis D, Castellanos-Bracamontes A. Dissociated horizontal deviation:
clinical findings and surgical results in 20 patients. Binocul Vis Strabismus
Q. 1992;7:173–8.
7. Quintana-Pali L. Desviacion horizontal disociada. Bol Hosp Oftalmol.
1990;42:91–4.
8. Zabalo S, Girett C, Domínguez D, Ciancia A. Exotropia intermitente con des-
viación vertical discodiada. Arch Oftalmol B Aires. 1993;68:11–20.
9. Wilson ME. The dissociated strabismus complex. Binocul Vis Strabismus
Q. 1993;8:45–6.
10. Wilson ME, Hutchinson AK, Saunders RA. Outcomes from surgical treatment
for dissociated horizontal deviation. J AAPOS. 2000;4(2):94–101.
11. Spielmann AC, Spielmann A. Antinomic deviations: esodeviation associated
with exodeviation. In: Faber TJ, editor. Transactions 28th Meeting European
Strabismological Association, Bergen, Norway, June, vol. 2003. London:
Taylor & Francis; 2004. p. 173–6.
12. Bielschowsky A. Über die Genese einseitiger Vertikalbewegungen der Augen.
Z Augenheilkd. 1904;12:545–57.
13. Bielschowsky A. Die einseitigen und gegensinnigen (“dissoziierten”)

Vertikalbewegungen der Augen. Albrecht Von Graefes Arch Ophthalmol.
1930;125:493–553.
14. Bielschowsky A. Disturbances of the vertical motor muscles of the eye. Arch
Ophthalmol. 1938;20:175–200.
15. Lyle TK. Worth and Chavasse’s Squint. The binocular reflexes and treatment of
strabismus. Philadelphia: Blakiston; 1950. p. 40–1.
16. Brodsky MC. Visuo-vestibular eye movements: infantile strabismus in three
dimensions. Arch Ophthalmol. 2005;123(6):837–42.
17. Pediatric Eye Disease Investigator Group. The clinical spectrum of early-onset
esotropia: experience of the congenital esotropia observational study. Am J
Ophthalmol. 2002;133(1):102–8.
18. Pediatric Eye Disease Investigator Group. Spontaneous resolution of early-
onset esotropia: experience of the congenital esotropia observational study. Am
J Ophthalmol. 2002;133(1):109–18.
19. Ing MR. Progressive increase in the quantity of deviation in congenital esotro-
pia. Trans Am Ophthalmol Soc. 1994;92:117–31.
20. Fawcett SL, Wang YZ, Birch EE. The critical period for susceptibility of human
stereopsis. Invest Ophthalmol Vis Sci. 2005;46(2):521–5.
21. Guyton DL. Changes in strabismus over time: the roles of vergence tonus and
muscle length adaptation. Binocul Vis Strabismus Q. 2006;21(2):81–92.
Arch Ophthalmol. 1999;117(9):1216–22.
23. Roth A, Speeg-Schatz C. Eye muscle surgery. Basic data, operative techniques,
surgical strategy. Paris: Swets & Zeitlinger; 1995. p. 283–324.
24. Jampolsky A. Strabismus and its management? In: Taylor DS, Hoyt CS, editors.
Pediatric ophthalmology and strabismus. 3rd ed. London: Elsevier Saunders;
2005. p. 1001–10.
25. Thouvenin D, Nogue S, Fontes L, Norbert O. Strabismus after treatment of
unilateral congenital cataracts: a clinical model for strabismus physiopathogen-
esis? In: Faber d, editor. Transactions 28th European Strabismological
Association Meeting, Bergen, Norway, vol. 2003. London: Taylor & Francis;
2004. p. 147–52.
26. Apt L, Isenberg S. Eye position of strabismic patients under general anesthesia.
Am J Ophthalmol. 1977;84(4):574–9.
27. Romano PE, Gabriel L, Bennett W, et al. Stage I intraoperative adjustment of
eye muscle surgery under general anesthesia: consideration of graduated adjust-
ment. Graefes Arch Clin Exp Ophthalmol. 1988;226(3):235–40.
28. Jampolsky A. Ocular divergence mechanisms. Trans Am Ophthalmol Soc.
1970;68:730–808.
29. Jampolsky A. Unequal visual inputs in strabismus management: a comparison
of human and animal strabismus. In: Symposium on strabismus: transactions of
the New Orleans Academy of Ophthalmology. St Louis: CV Mosby; 1978.
p. 422–5.
30. Horwood A. Too much or too little: neonatal ocular misalignment frequency
can predict lateral abnormality. Br J Ophthalmol. 2003;87(9):1142–5.
31. Horwood AM, Riddell PM. Can misalignments in typical infants be used as a
model for infantile esotropia? Invest Ophthalmol Vis Sci. 2004;45(2):714–20.
32. Burian HM. Pathophysiology of exodeviations. In: Manley DR, editor.

Symposium on horizontal ocular deviations. St Louis: CV Mosby; 1971.
p. 119–27.
33. Kushner BK. Exotropic deviations: a functional classification and approach to
treatment. Am Orthopt J. 1988;38:81–93.
Postscript 113
34. Kushner BJ, Morton GV. Distance/near differences in intermittent exotropia.

Arch Ophthalmol. 1998;116(4):478–86.
35. Pritchard C. Incidence of dissociated vertical deviation in intermittent exotro-
pia. Am Orthopt J. 1998;48:90–3.
36. Moore S, Cohen RL. Congenital exotropia. Am Orthopt J. 1985;35:68–70.
37. Rubin SE, Nelson LB, Wagner RS, et al. Infantile exotropia in healthy infants.
Ophthalmic Surg. 1988;19(11):792–4.
38. Hunter DG, Kelly JB, Ellis FJ. Long-term outcome of uncomplicated infantile
exotropia. J AAPOS. 2001;5(6):352–6.
Postscript
This editorial elaborates on the previous chapter and further expands on from my
AOS thesis on dissociated horizontal deviation (DHD), which proposed that we
have been thinking about DHD the wrong way. Rather than being a dissociated
exodeviation, DHD actually arises when monocular fixation causes dissociated
esotonus to be superimposed on baseline exodeviation (usually following strabis-
mus surgery for infantile esotropia). In this setting, fixation with the poorer eye
generates a greater degree of dissociated esotonus, sometimes causing the baseline
esotropia to convert to an esotropia. Therefore, when patients become inattentive, a
much larger exodeviation becomes evident.
It was Michael Gräf at the University of Giessen who first recognized that mon-
ocular fixation evokes convergence in DHD. He has since devised the “17 x 13” test
to elicit this phenomenon. Just ask the patient to multiply 17 x 13 and you will see
the large baseline exodeviation appear. I referred to this particular quality of DHD
as the “Heisenberg Uncertainty Principle” of clinical strabismus measurement,
wherein the act of measuring a horizontal deviation by occluding either eye inevita-
bly changes the measurement you obtain (in this case by evoking monocular fixa-
tion with its commensurate esotonus). Strabismus surgery based on the measured
deviation (using prism and alternate cover testing), will correct the measurements
but leave a large residual exodeviation during periods of inattention.
My early interest in DHD was directed at the establishing utility of the reversed
fixation test in identifying dissociated esotonus (Chap 7). This quick clinical test
works well except when the dissociated esotonus happens to be equal in the two
eyes. In this essay, I propose that the dissociated esotonus that characterizes DHD
could be the underlying cause of essential infantile esotropia. Accordingly, what we
recognize as essential infantile esotropia may be the end-stage phenotype of disso-
ciated esotonus.
In patients with infantile or consecutive exotropia, the innervational effects of
dissociated esotonus are easily visualized. In infantile esotropia, however, the dis-
sociated esotonus is fully expressed, rendering it impossible to visualize DHD dur-
ing monocular fixation. This mechanism also explains why we occasionally see
DHD manifesting as an intermittent esotropia of varying amplitude depending on
which eye is fixating. (Fig. 8.1).
Fig. 8.1 Preoperative photos of DHD showing eye position during (a) period of inattention, (b)
left eye fixation at near, (c) right eye fixation at near. (Reprinted with permission from Brodsky
MC: Intermittent esotropia in an adult. JAMA Ophthalmol 2021, with permission from the
American Medical Association)
The Accessory Optic System
The Fugitive Visual Control System in Infantile Strabismus
9
Infantile strabismus is characterized by dissociated binocular vision, which is the

normal condition in lateral-eyed animals [1, 2]. Early binocular misalignment gives
rise to dissociated eye movements (changes in eye position evoked by unequal
visual input to the 2 eyes) [3]. These include latent nystagmus, dissociated vertical
divergence, and dissociated horizontal deviation [1–3], all of which have a promi-
nent torsional component. Primary oblique muscle overaction, which accompanies
infantile strabismus but is not dissociated in nature, is also characterized by a tor-
sional misalignment of the eyes [4].
These binocular deviations all correspond to normal visuovestibular reflexes that
are operative in lateral-eyed animals [1–4]. Evolutionarily, these visual reflexes
antedate development of the visual cortex, which does not generate torsional eye
movements in humans [5]. Therefore, any attempt to anatomize infantile strabismus
must explain the reemergence of these atavistic reflexes, as well as their prominent
torsional components. I propose that the accessory optic system (AOS), an atavistic
subcortical visual motion detection system, could generate the dissociated and non-
dissociated torsional eye movements that accompany human infantile strabismus.
What Is the AOS?
The AOS consists of 3 nuclei at the mesodiencephalic border that receive direct reti-
nal input from the accessory optic tract (AOT) [6–9] (Fig. 9.1). The AOT comprises
an inferior and a superior fasciculus, with its superior fasciculus divided into a pos-
terior branch, a middle branch, and an anterior branch that is identical to the original
transpeduncular tract (tractus peduncularis transversus) discovered in 1870 by
Gudden [10, 11]. The number of accessory optic fibers is small [7]. In almost all
mammalian species, most optic fibers reach the accessory optic nuclei via the
transpeduncular tract, which is visible as it courses over the brachium of the supe-
rior colliculus [12].
116 9 The Accessory Optic System
Fig. 9.1 Neuroanatomical connections of the accessory optic system. The brainstem is depicted
from the front (with the left-hand side of the animal on the right-hand side of the drawing). Accessory
terminal nuclei include the dorsoterminal nucleus (DTN), which lies adjacent to the nucleus of the
optic tract (NOT); medial terminal nucleus (MTN); lateral terminal nucleus (LTN); and principal
part of the inferior olive (IOp). Optokinetic input from the right retina crosses to the left accessory
optic nuclei (depicted), which send ipsilateral projections to the left dorsal cap (DC) of the inferior
olive and then back to the right flocculus (not shown), resulting in a double decussation of motion
pathways from each eye. (Adapted with permission from Simpson et al. [12]) CP indicates posterior
commissure; D, nucleus of Darkschewitsch; DMNm, deep mesencephalic nucleus, pars medialis;
EW, nucleus of Edinger-Westphal; INC, interstitial nucleus of Cajal; inSFp, intersitial nucleus of the
superior fasciculus, posterior fibers; MAO, medial accessory nucleus, inferior olivary complex; ML,
medial lemniscus; MLF, medial longitudinal fasciculus; PAGm, periaqueductal gray, medial part;
pdl, dorsolateral division, basal pontine complex; pm, medial division, basal pontine complex; pv,
ventral division, basal pontine complex; PVG, periventricular gray; RN, red nucleus; rpc, pontine
reticular nucleus, pars caudalis; rpo, pontine reticular nucleus, pars oralis; vl, lateral vestibular
nucleus; VLO, ventrolateral outgrowth, inferior olivary complex; vm, medial vestibular nucleus; vs,
superior vestibular nucleus; vsp, spinal vestibular nucleus; VTRZ, visual tegmental relay zone; β,
nucleus β of the inferior olive; 3n, oculomotor nerve; 4n, trochlear nerve; and 6n, abducens nerve
What Is the AOS? 117
In most mammalian species, the AOS is composed of 3 paired terminal nuclei,

namely, the dorsoterminal nucleus (DTN), the lateroterminal nucleus (LTN), and
the medioterminal nucleus (MTN), which receive innervation from primary optic
fibers [7–9]. Input to these 3 accessory optic terminal nuclei is predominantly from
the contralateral eye [7–9, 11, 12]. Along with the nucleus of the optic tract (NOT),
these 3 terminal nuclei project differentially to the dorsal cap of the inferior olive
[13–16], which provides the only source of climbing fibers to the flocculonodular
lobe of the cerebellum [7–9, 13–17]. In this way, cells of the AOS converge with
those of the vestibular system in the vestibulocerebellum [7–9].
Despite its name, the AOS is a primary visual system receiving direct visual
information from the retina via 1 or more AOTs [13] that are responsible for
visuovestibular interaction in afoveate animals [7, 16, 17]. Its retinal input is derived
from ON–type direction-sensitive ganglion cells. The AOS neurons have large
receptive fields (averaging about 40° vertically and 60° horizontally), are direction
selective, and have a preference for slow-moving stimuli [7–9, 12, 13]. The AOS
processes information about the speed and direction of movement of large textured
parts of the visual world [7–9]. The AOS signals self-motion as a function of slip of
the visual world over the retinal surface and generates corrective eye movements to
stabilize the retinal image [7–9]. As an analyzer of self-motion, the AOS subserves
visual proprioception in the afoveate animal [7–9].
The AOS is a visual system that is organized in vestibular coordinates [7–9].
According to results of experimental studies by Simpson and colleagues, visual and
vestibular signals that produce compensatory eye movements are organized about a
common set of axes derived from the orientation of the semicircular canals (Fig. 9.2)
[7, 9, 12, 16, 17]. Because the AOS is directionally sensitive to low-velocity move-
ments while the vestibular system typically responds to movements of higher veloc-
ity, the AOS and vestibular labyrinths form 2 complementary systems to detect
self-motion and promote image stabilization so that objects in the visual world can
be quickly and accurately analyzed [7, 8, 12, 13].
The AOS exists in all vertebrate classes [6, 7, 18, 19], including humans [20], but
it has been studied most extensively in the rabbit. The 3 preferred directions for cells
in the accessory optic terminal nuclei define 3 directions in visual space, namely,
horizontally from posterior to anterior for the DTN, vertically up and down for the
MTN, and vertically down for the LTN [7–9, 11–14, 21]. Its 3 pretectal accessory
optic nuclei are closely related to the NOT and receive input predominantly from
the contralateral eye [7–9, 12, 13]. Direction-sensitive ON-type retinal ganglion
cells encode retinal image slip [22, 23] and transmit this information to the AOS,
inferior olive [24] floccular climbing fibers [25], and floccular Purkinje cells [26].
These 3 pairs of channels remain anatomically distinguishable within the AOS,
inferior olive, and floccular zones, which (when stimulated) elicit eye movements
organized in a canal-like coordinate system [18, 27–29]. Each pair conveys signals
about flow of the visual surround about 1 of 3 rotation axes, which are approxi-
mately collinear with the best-response axes of the semicircular canals and the rota-
tion axes of the extraocular muscles [28].
Rotation Axes Visual Climbing Fibers

(Left Flocculus of Dominant Eye)
VA
Paired
vertical recti
148°
Paired
obliques
43°
∗ ∗
∗ 143°
47° 90°
0°
Paired 45° Paired
vertical canals vertical canals
41° 139°
Fig. 9.2 Spatial orientation of preferred axes of 3-dimensional rotation for dorsal cap neurons in
the right inferior olive recorded during optokinetic stimulation in a spherical enclosure. (Adapted
with permission from Van der Steen et al. [27] VA indicates vertical axis)
The rabbit flocculus ipsilateral to the seeing eye is optimally sensitive to optoki-
netic stimulation about a 135° axis, while the flocculus contralateral to the seeing
eye is optimally sensitive to optokinetic stimulation around a horizontal 45° axis
(Fig. 9.3) [26–29]. For horizontal stimulation, the DTN and its adjacent NOT are
selectively sensitive to nasally directed optokinetic stimulation presented to the con-
tralateral eye [7, 8, 12, 13]. Conversely, electrical microstimulation in the alert rab-
bit’s flocculus produces abduction of the ipsilateral eye [27, 29, 30] or dissociated
torsional and vertical rotations of the 2 eyes, corresponding to the plane of 1 semi-
circular canal [26–30]. Because floccular motion detection for each eye is not fully
represented on its own side of the body, monocular optokinetic responses must be
derived from the synthesis of bilateral floccular representations [28]. Therefore, the
flocculus provides a subcortical binocular visual system that generates asymmetri-
cal torsional eye movements under dissociated conditions of optokinetic stimula-
tion [28].
Studies using decortication have revealed contributions from the visual cortex to
the AOS. Disruption of contributions from the visual cortex to the AOS by strabis-
mus may alter the inherent biases of the accessory optic nuclei [31–33]. The
What Is the AOS? 119
a Monocular viewing Posterior (135°) Axis Type CCW

Ipsilateral Spontaneous activity
CW
2.5
Mean Firing Frequency, Spikes/s

0°
2.0
45°
1.5
90°
1.0
135° 0.5
Top view
0.0
Spikes/s
0.6°/s 0 45 90 135 180

5.3
2.6 CW CCW Planetarium Axis Angle, °
15 Repetitions
Spontaneous 5s 100 ms Bin width
activity
b Monocular viewing Anterior (45°) Axis Type

Contralateral
0° 0.5
Mean Firing Frequency, Spikes/s
22.5°
0.0
45°
67.5°
0.5
90°
112.5° 0.0
135°
Top view157.5°
0.5
0.6°/s
Spikes/s
5.3 0.0
2.6 CW CCW 15 Repetitions 0 45 90 135 180
Spontaneous 5s 100 ms Bin width
activity Planetarium Axis Angle, °
Fig. 9.3 Sensitivity to monocular optokinetic stimulation in a spherical enclosure. (a) Responses
of posterior axis climbing fiber Purkinje cells to stimulation presented to the ipsilateral left eye. (b)
Responses of anterior axis climbing fiber Purkinje cells to stimulation presented to the contralat-
eral eye. CCW indicates counterclockwise optokinetic rotation; CW, clockwise optokinetic rota-
tion. (Adapted with permission from Van der Steen et al. [27])
ipsilateral visual cortex is necessary for several response properties that distinguish
DTN and LTN neurons in the cat from those in the rabbit. Following decortication,
cat DTN and LTN neurons lose their binocularity and become almost totally domi-
nated by the contralateral eye [33]. For example, LTN neurons excited by upward
movement, which in the cat are equal in number to those excited by downward
movement, become less numerous so that the cat LTN becomes like that of the rab-
bit, consisting of neurons excited by slow downward movements to the contralateral
eye [33]. Unlike the LTN and DTN, neurons in the cat MTN are largely monocular
and similar to those in the rabbit [12]. The monocular nasotemporal optokinetic
asymmetry that characterizes infantile strabismus is known to result from monocu-
lar cortical input to the NOT and DTN [34], unmasking a subcortical visuovestibu-
lar bias that generates latent nystagmus [35]. The AOS provides a neuroanatomical
substrate whereby vertical monocular subcortical motion biases could generate the
canal-based torsional eye movements that characterize primary oblique muscle
overaction and dissociated vertical divergence [2, 4]. Although we observe and ana-
lyze these eye movements in yaw, pitch, and roll [2], they are encoded in a canal-
oriented push-pull bilateral coordinate system that detects optokinetic flow in every
direction [36].
Photic stimulation can activate the AOT in the rabbit [4, 37]. The AOS neurons
show the same responses to retinal illumination as ON-type direction-sensitive reti-
nal ganglion cells, being excited only at the onset of retinal stimulation [23], and
generate a firing response that is related to light intensity [32]. In this way, the AOS
may implement the visuovestibular reflexes that characterize infantile strabismus
[1, 2]. However, because the AOS is primarily a motion detector, central modulation
of the primitive luminance reflexes that characterize infantile strabismus may
require input from additional subcortical visual pathways. It is possible that other
primitive luminance pathways may provide parallel subcortical luminance input to
the visuovestibular system [38]. Like the AOS, luminance input that modulates the
dorsal light reflex in fish (which corresponds to dissociated vertical divergence and
primary oblique muscle overaction in humans with infantile strabismus) [1, 2] is
transmitted to the central pretectal nucleus in the contralateral midbrain and then
down to the vestibulocerebellum, which integrates visual and vestibular input [39].
These luminance and motion pathways may constitute the subcortical equivalents of
the “what” and “where” visual streams within the association visual cortex. How
these subcortical visual streams intercommunicate to consolidate spatial and tempo-
ral summation of visual information at the subcortical levels remains a mystery. But
the likelihood that they provide the innervational substrate for the atavistic eye
movements that characterize infantile strabismus should not be ignored.
Conclusions
The AOS provides a critical piece of the puzzle for infantile strabismus by serving
as a neuroanatomic substrate for visuovestibular eye movements. The AOS is atavis-
tic, present in humans, subcortical, crossed, and sensitive to optokinetic motion. It
Conclusions 121
operates in a canal-based coordinate system and generates dissociated torsional eye

movements. For these reasons, the AOS is uniquely suited to generate the dissoci-
ated eye movements that characterize infantile strabismus. The fact that its retinal
fibers terminate in the 3 nuclei of the AOT along with the adjacent NOT (a part of
the pretectal nuclear complex that generates latent nystagmus) lends further cre-
dence to this hypothesis. Dissociated binocular vision in infancy may unlock this
atavistic visual system, generating canal-based ocular rotations that we anthropo-
morphize to diagnose “torsion” in the frontal plane.
This analysis implies that mutations involving the AOS or its target zones within
the cerebellar flocculus could provide a potential template for infantile strabismus.
If so, then the age-old dichotomy postulated by Worth (congenital defect in cortical
fusion) and Chavesse (early binocular misalignment) [40] could be explained by
binocular subcortical dysfunction intrinsic to the visuovestibular system.
References
Arch Ophthalmol. 1999;117(9):1216–22.
2. Brodsky MC. Visuo-vestibular eye movements: infantile strabismus in 3 dimen-
sions. Arch Ophthalmol. 2005;123(6):837–42.
3. Brodsky MC. Dissociated horizontal deviation: clinical spectrum, pathogene-
sis, evolutionary underpinnings, diagnosis, treatment, and potential role in the
development of infantile esotropia (an American Ophthalmological Society
thesis). Trans Am Ophthalmol Soc. 2007;105:272–93.
a wild pitch. Arch Ophthalmol. 2001;119(9):1307–14.
5. Brodsky MC. Dissociated vertical divergence: cortical or subcortical in origin?
Strabismus. 2011;19(2):67–70.
6. Marg E. The accessory optic system. Ann N Y Acad Sci. 1964;117:35–52.
7. Simpson JI, Soodak RE, Hess R. The accessory optic system and its relation to
the vestibulocerebellum. Prog Brain Res. 1979;50:715–24.
8. Simpson JI, Leonard CS, Soodak RE. The accessory optic system: analyzer of
self-motion. Ann N Y Acad Sci. 1988;545:170–9.
9. Simpson JI. The accessory optic system. Annu Rev Neurosci. 1984;7:13–41.
10. Gudden B. Ueber einen bisher nicht eschriebenen Nervenfasernstrang im

Gehirne der Säugethiere und des Menschen. Arch Psychiatry. 1870;2:364–6.
11. Gudden B. Ueber den Tractus peduncularis transversus. Arch Psychiatry.

1881;11:415–23.
12. Simpson JI, Giolli RA, Blanks RH. The pretectal nuclear complex and the
accessory optic system. Rev Oculomot Res. 1988;2:335–64.
13. Giolli RA, Blanks RHI, Lui F. The accessory optic system: basic organization
with an update on connectivity, neurochemistry, and function. Prog Brain Res.
2006;151:407–40.
14. Takeda T, Maekawa K. The origin of the pretectoolivary tract: a study using the
horseradish peroxidase method. Brain Res. 1976;117(2):319–25.
15. Maekawa K, Takeda T. Afferent pathways from the visual system to the cere-
bellar flocculus of the rabbit. In: Baker R, Berthoz A, editors. Control of gaze
by brain stem neurons: developments in neuroscience. Amsterdam: Elsevier/
North-Holland Biomed; 1977. p. 187–95.
16. Maekawa K, Simpson JI. Climbing fiber activation of Purkinje cells in the floc-
culus by impulses transferred through the visual pathway. Brain Res.
1972;39(1):245–51.
17. Maekawa K, Simpson JI. Climbing fiber responses evoked in vestibulocerebel-
lum of rabbit from visual system. J Neurophysiol. 1973;36(4):649–66.
18. Ebbesson SO. On the organization of central visual pathways in vertebrates.
Brain Behav Evol. 1970;3(1):178–94.
19. Cooper HM, Magnin M. A common mammalian plan of accessory optic system
organization revealed in all primates. Nature. 1986;324(6096):457–9.
20. Fredericks CA, Giolli RA, Blanks RH, Sadun AA. The human accessory optic
system. Brain Res. 1988;454(1–2):116–22.
21. Simpson JI, Leonard CS, Soodak RE. The accessory optic system of rabbit.
II. Spatial organization of direction selectivity. J Neurophysiol.
1988;60(6):2055–72.
22. Oyster CW. The analysis of image motion by the rabbit retina. J Physiol.
1968;199(3):613–35.
23. Soodak RE, Simpson JI. The accessory optic system of rabbit. I. Basic visual
response properties. J Neurophysiol. 1988;60(6):2037–54.
24. Leonard CS, Simpson JI, Graf W. Spatial organization of visual messages of the
rabbit’s cerebellar flocculus. I. Typology of inferior olive neurons of the dorsal
cap of Kooy. J Neurophysiol. 1988;60(6):2073–90.
25. Simpson JI, Graf W, Leonard CS. Three-dimensional representation of retinal
image movement by climbing fiber activity. In: Strata P, editor. The olivocere-
bellar system in motor control. Berlin: Springer Verlag; 1989. p. 323–37.
Experimental Brain Research Series 17.
26. Graf W, Simpson JI, Leonard CS. Spatial organization of visual messages of the
rabbit’s cerebellar flocculus. II. Complex and simple spike responses of Purkinje
cells. J Neurophysiol. 1988;60(6):2091–121.
27. Van der Steen J, Simpson JI, Tan J. Representation of three-dimensional eye
movements in the cerebellar flocculus of the rabbit. In: Schmid R, Zambarbieri
D, editors. Oculomotor control and cognitive processes. Amsterdam: Elsevier;
1991. p. 63–77.
28. Tan HS, van der Steen J, Simpson JI, Collewijn H. Three-dimensional organiza-
tion of optokinetic responses in the rabbit. J Neurophysiol. 1993;69(2):303–17.
29. Simpson JI, Van der Steen J, Tan J, Graf W, Leonard CS. Representations of
ocular rotations in the cerebellar flocculus of the rabbit. Prog Brain Res.
1989;80:213–23.
30. Ito M, Nisimaru N, Yamamoto M. Specific patterns of neuronal connexions
involved in the control of the rabbit’s vestibulo-ocular reflexes by the cerebellar
flocculus. J Physiol. 1977;265(3):833–54.
Postscript 123
31. Grasse KL, Cynader MS. Response properties of single units in the accessory
optic system of the dark-reared cat. Brain Res. 1986;392(1–2):199–210.
32. Grasse KL, Cynader MS. The accessory optic system of the monocularly
deprived cat. Brain Res. 1987;428(2):229–41.
33. Grasse KL, Cynader MS, Douglas RM. Alterations in response properties in the
lateral and dorsal terminal nuclei of the cat accessory optic system following
visual cortex lesions. Exp Brain Res. 1984;55(1):69–80.
34. Hoffmann KP. Cortical vs subcortical contributions to the optokinetic reflex in
the cat. In: Lennerstrand G, Zee DS, Keller EL, editors. Functional basis of
ocular motility disorders. Oxford, UK: Pergamon; 1982. p. 303–10.
35. Brodsky MC, Tusa RJ. Latent nystagmus: vestibular nystagmus with a twist.
Arch Ophthalmol. 2004;122(2):202–9.
36. Simpson JI, Graf W. The selection of reference frames by nature and its inves-
tigators. Rev Oculomot Res. 1985;1:3–16.
37. Hamasaki D, Marg E. Microelectrode study of accessory optic tract in the rab-
bit. Am J Phys. 1962;202:480–6.
38. Schiller PH. Parallel information processing channels created in the retina. Proc
Natl Acad Sci U S A. 2010;107(40):17087–94.
39. Yanagihara D, Watanabe S, Takagi S, Mitarai G. Neuroanatomical substrate for
the dorsal light response. II. Effects of kainic acid-induced lesions of the val-
vula cerebelli on the goldfish dorsal light response. Neurosci Res.
1993;16(1):33–7.
40. Chavesse FB, Worth CA, Lyle TK. The binocular reflexes and treatment of
strabismus. Philadelphia: Blakeston; 1950.
Postscript
The accessory optic system (AOS) provides a critical piece of the puzzle in infantile
strabismus by explaining how visual motion can be processed in a vestibular coor-
dinate system. The existence of this subcortical system explains how torsional eye
movements can be generated in response to unequal binocular visual input, and how
individual extraocular muscles can be driven to overact when binocular visual
development is comprised. The clear implication is that this atavistic system some-
how remains functional in infantile strabismus and infantile nystagmus. This propo-
sition is not so bold when one considers that our extraocular muscles are oriented in
a semicircular canal-based planar coordinate system. The AOS subserves optokine-
sis, however, so it remains unclear whether primitive luminance pathways are
encoded in a similar coordinate system, or whether luminance disparity generates
correlative input the AOS.
The reader would be remiss in not asking why humans retain primitive neuro-
logic systems such as the AOS after newer cortical systems develop. A satisfying
answer can be found in John Allman’s 1999 book Evolving Brains. In one section,
during a visit to a power generation plant in the 1970s, he noted that several
different control systems, some ancient, others new, were situated adjacent to each
other. Some contained pneumatic controls and a system of controls based on vac-
uum tube technology, while the newer control systems relied upon computer-driven
control systems. Yet, these systems were all used together to control the generation
of electricity at the plant. When he asked why the archaic systems had not been
superseded by the new ones, he was told that the demand for power had always been
too great for the plant to ever be shut down. He concluded that “The brain has
evolved in the same manner as the control systems in this power plant. The brain,
like the power plant, can never be shut down and fundamentally reconfigured, even
between generations. All the old control systems must remain in place, and new
ones with additional capacities are added and integrated in such a way as to enhance
survival.”
The subcortical visual pathways are old school. They are concerned with side
vision, while the visual cortex is more concerned with central vision (as evidenced
by its generous macular representation). Yet the human AOS remains functional
within the first 2–3 months of life of infancy. The implication of this article is that
direct retinotectal input to the AOS may perpetuate its function and thereby generate
the unique eye movements that characterize (and possibly generate) infantile stra-
bismus. Accordingly, infantile strabismus may be generated primarily by the AOS,
or be perpetuated secondarily by the AOS only when cortical binocular vision fails
to develop. It may even be that the secondary reconfiguration of the motion visual
cortex, as first proposed by Hoffman, stimulates the AOS to retain the necessary
neuroplasticity to subserve this function. Long ago, Clifton Schor proposed that
cortical suppression of one eye could momentarily disinhibit the subcortical path-
ways to allow their retinotectal input to predominate. Any combination of these
mechanisms could be operative. Anyone who tells you that they have the final
answer is not a scientist. In science, models evolve with new and better data. Science
is a series of leaps, with new hypotheses leading to ongoing falsifications. So keep
an open mind.
Visuo-Vestibular Eye Movements
Infantile Strabismus in 3 Dimensions
10
Infantile strabismus inaugurates a triad of stereotypical eye movements comprising

latent nystagmus, primary inferior oblique muscle overaction, and dissociated verti-
cal divergence [1]. These unique ocular movements have long been a medical curi-
osity because they are not encountered in other forms of neurological disease. Their
coexistence in patients with infantile strabismus implicates a common pathophysi-
ology. Until we understand how these eye movements cohere into a larger whole,
“they remain intellectual orphans” [2]. The purpose of this article is to examine the
physical laws that govern these ocular movements, to integrate these movements
into “an organized whole in a way that snaps existing and heretofore patternless data
into a highly symmetrical pattern” [3], and to examine the evolutionary forces that
underlie their clinical expression in infantile strabismus.
An evolutionary undergird for these ocular movements can be constructed from
the following physiologic principles.
The Problem Is Gravity
Why is symmetry ubiquitous in the animal kingdom? More specifically, what is the
evolutionary origin of the bilateral symmetry that dictates that we have 2 eyes, ears,
arms, and legs situated equidistant from the midline? The problem of symmetry in
biological systems is a complex subject of ongoing study [4–6]. One of the evolu-
tionary functions of biological symmetry is balance. Because of the earth’s gravita-
tional field, survival requires that animals maintain vertical orientation. An animal
that falls tends to get eaten. For living organisms, gravity necessitates balance.
Because physical symmetry promotes balance, biological symmetry is an important
part of nature’s solution to the problem of gravity.
126 10 Visuo-Vestibular Eye Movements
Bilaterally Symmetrical Organs Function as Balance Organs
The evolution of bilaterally symmetrical organs situated equidistant from the sagit-
tal plane promotes balance by virtue of physical weight alone. Our 2 lungs, when
inflated with air, become ballast in water. But many bilateral structures also function
as physiologic balance organs. While it is axiomatic that our 2 labyrinths superin-
tend balance, other bilateral organs modulate balance in different ways. Our 2 cere-
bral hemispheres each control motor tone in the contralateral limbs. Our 2 cerebellar
hemispheres clearly control balance. Even the 2 kidneys have been shown to func-
tion in part as balance organs [7].
Lateral Eyes are Sensory Balance Organs
Our ancestral environment is characterized by 2 physical constants: light from the

sky above and gravity from the earth below [8]. Although the sun is the source of
light, it adds little to the overall brightness of the sky. Throughout evolution, the
bright sky has served as a space-stable luminance hemisphere [8]. Because light
comes from above, when the eyes are laterally positioned, vertical alignment is
signaled by equal luminance to the 2 eyes. Evolution has programmed this visual
cue into our balance system to optimize survival.
While it is axiomatic that our 2 labyrinths keep us informed of our relation to our
gravitational position (a static otolithic system) and movement (a dynamic semicir-
cular canal system), our 2 eyes subserve the same function because an animal can-
not rely solely on vestibular input during momentary physical perturbations by wind
and waves [9]. The striking correspondence of eye-muscle position to semicircular
canal and extraocular muscle orientation permits the eyes and ears to subserve bal-
ance in lockstep. As summarized by Duke-Elder [10], … the control of the move-
ments of living animals, both plants and animals, by light is a fundamental function
of great phylogenetic age, preceding the acquirement of vision and, indeed, leading
directly to its development. The association of the functions of equilibration and
orientation with the visual system of higher animals is in every sense basic.
rimitive Reflexes are Resurrected When Normal

P
Neurodevelopment Fails to Occur
Many primitive reflexes that promoted survival in the ancestral setting lose their
beneficial function as evolution proceeds [11, 12]. With progressive encephaliza-
tion, newer cortical reflexes are grafted onto older subcortical reflexes, which persist
in latent form even when they no longer serve a useful function [13, 14]. Abnormal
neurodevelopment in infancy is associated with a persistence of numerous subcorti-
cal reflexes (Table 10.1) [15]. In pediatric neurology, these primitive reflexes are
among the clinical signposts of abnormal neurodevelopment [15]. In pediatric
Ocular Motor Incursions Operate as Visual Balancing Reflexes in Lateral-Eyed Animals 127
Table 10.1 Normal Developmental Reflexes (Modified From Swaiman) [15]

Reflex Appearance Age Disappearance Age, mo
Adductor spread of knee jerk Birth 7–8
Moro Birth 5–6
Palmar grasp Birth 6
Planter grasp Birth 9–10
Routing Birth 3
Tonic neck response Birth 5–6
ophthalmology, these echoes from our visual ancestry signal maldevelopment of

cortical binocular vision.
In the developing visual system, monocular nasotemporal asymmetry is perhaps
the quintessential example of a visual reflex in which ontogeny recapitulates phy-
logeny. Nasotemporal asymmetry refers to a monocular horizontal optokinetic
response that is brisk in the nasal direction and poor or absent in the temporal direc-
tion [16]. This directional optokinetic bias is normal in lateral-eyed animals and in
human infants within the first 6 months of life. Humans with infantile strabismus
retain a nasotemporal asymmetry that provides the monocularly driven horizontal
movement bias for latent nystagmus [16].
cular Motor Incursions Operate as Visual Balancing Reflexes

O
in Lateral-Eyed Animals
Primitive visual reflexes rely on a dissociated form of binocular vision between the
2 laterally placed eyes, which has been superseded by cortical binocular vision in
humans [9]. Humans experience frontal binocular vision with forfeiture of periph-
eral vision in exchange for cortical fusion and stereopsis. Infantile strabismus recre-
ates the dissociated binocular condition that allows lateraleyed animals to process
dissociated luminance and visual input from each eye with little or no binocular
overlap. Infantile strabismus effectively disables the newer cortical binocular sys-
tem and unveils the primitive visual reflexes that have been inscribed into our primi-
tive ocular motor control system [14, 16].
The resurrection of these primitive visual reflexes generates a triad of ocular
movements that have come to define infantile strabismus. This symphony of eye
movements reveals each of the primitive visual reflexes that use dissociated binocu-
lar visual input to maintain physical orientation in 3-dimensional space.
Latent nystagmus corresponds to the optokinetic component of ocular rotation
that is driven monocularly by nasal optic flow during a turning movement of the
body in lateral-eyed animals [16]. When infantile esotropia disrupts the establish-
ment of binocular visual connections, visual input from the fixating eye to the con-
tralateral nucleus of the optic tract evokes a counterrotation of the eyes that
corresponds to a turning movement of the body toward the object of regard. The
clinical expression of this visual reflex is also evident in the monocular
nasotemporal asymmetry to horizontal optokinetic stimulation that characterizes

infantile strabismus [16].
Dissociated vertical divergence corresponds to the dorsal light reflex that has
been observed in fish and other lateral-eyed animals when unequal luminance to the
2 eyes evokes a body tilt or vertical divergence of the eyes toward the side with
greater luminance [8, 9]. In humans, dissociated vertical divergence is a visual bal-
ancing reflex that uses weighted binocular visual input to orient eye position to the
perceived vertical (Fig. 10.1) [8, 9]. The exaptation of a cycloversional movement
into the human dorsal light reflex permits active modulation of perceived visual tilt
when the eyes are frontally positioned [8].
Primary inferior oblique muscle overaction corresponds to a similar dorsal light
reflex that is induced in fish when a forward or backward shift in overhead lumi-
nance evokes an ipsidirectional body pitch or torsional rotation of the eyes back-
ward to reorient the body with respect to the light [19]. These binocular torsional
rotations of the eyes constitute a physiologic form of primary oblique muscle
Fixation + Light
Fixation
Fixation + Light
Fig. 10.1 Ohm’s visual balancing metaphor for dissociated vertical divergence [17] as depicted
by Mattheus and Kommerell [18]. Position of the scales corresponds to the relative vertical posi-
tions of the eyes in dissociated vertical divergence as determined by binocular visual input. White
retina indicates light input, dark retina indicates no light input, and x corresponds to the presence
of fixation in light or in darkness. (Reprinted with permission from Aeolus Press, Buren, the
Netherlands)
Latent Nystagmus, Primary Oblique Overaction, and Dissociated Vertical Divergence… 129
overaction that can be induced by altering binocular visual input. In humans, a for-
ward or backward rotation relative to overhead light sends excitatory innervation to
each of the elevators or each of the depressors. Because the vestibular system segre-
gates innervation to its target extraocular muscles, the vertical actions of the human
oblique muscles summate in adduction with those of the rectus muscles to produce
the innervational overelevation of the adducting eye that defines primary inferior
oblique muscle overaction. The fundamental association of primary oblique muscle
overaction with early loss of binocular vision in humans suggests that the brain
registers abnormal binocular visual input as forward rotation [19].
rimitive Visual Reflexes are Evoked by a Physiologic

P
Imbalance in Binocular Visual Input
Most patients with infantile esotropia have no neurologic disease. From where do
these involuntary ocular rotations arise? Stereotypical eye movements that do not fit
any paradigm must come from somewhere. As stated by Keiner, “Nothing comes
from nothing”[14] (p151). The source of these dynamic intrusions has eluded us for
the simple reason that they arise from a physiologic imbalance rather than from a
neurologic lesion. They are harmonics of our earlier orchestration that bubble to the
surface when they are not superseded by cortical binocular reflexes.
Just as physiologic vestibular movements reflect the degree of unbalanced laby-
rinthine input, the size of these eye movements are proportional to the degree of
binocular visual imbalance, which can fluctuate depending on momentary cortical
suppression of either eye. Because the eye movements of infantile strabismus are
calibrated by weighted binocular visual input, they increase in proportion to the
disparity of visual input from the 2 eyes [9].
atent Nystagmus, Primary Oblique Overaction,

L
and Dissociated Vertical Divergence are Visuo-Vestibular
Eye Movements
In lateral-eyed animals, visual and labyrinthine input are pooled together within the
central vestibular system to establish central vestibular tone [17–19]. Based on
momentary fluctuations in bilateral input, the central vestibular system constantly
modulates eye position and body position to maintain physical orientation in
3-dimensional space [8]. Vestibular input predominates over visual input, but both
systems are integrated at the level of the vestibular nucleus to maintain balance
[20–22]. In infantile strabismus, binocular visual imbalance alone can alter central
vestibular tone [9]. Because input to the vestibular nucleus can be directly driven by
the balance of visual input from the 2 eyes, this unique form of central vestibular
imbalance does not require that the central nervous system receive an imbalance of
labyrinthine input [9]. It is as if infantile strabismus transforms the 2 eyes into phys-
iologic vestibules. These unique eye movements are visuo-vestibular in origin.
isuo-Vestibular Eye Movements are Generated by Subcortical

V
Central Vestibular Pathways
In the absence of binocular cortical development, the system defaults to a subcorti-

cal control system in which the central vestibular system is the gyroscope that sets
postural tonus to the extraocular muscles. These subcortical motor pathways receive
both afferent input from the optic nerves and efferent input from the visual cortex.
As with the pupillary light reflex, however, cortical input can modulate these sub-
cortical reflexes [23]. Thus, voluntary or involuntary suppression of 1 eye can acti-
vate latent nystagmus and dissociated vertical divergence [24, 25].
The clinical expression of these subcortical reflexes in the setting of infantile
strabismus reflects the evolution of a hierarchical visual system in which cortical
binocular vision can hold our subcortical reflexes in check [26]. In infantile strabis-
mus, subcortical visual reflexes are reactivated because the system reverts to a dis-
sociated binocular system [27].
isuo-Vestibular Eye Movements Arise from a Central

V
Vestibular Imbalance that Dissociates Clinically into 3
Distinct Planes
Vestibular eye movements are governed by the anatomical orientation of the laby-
rinths and occur in 3 major head-referenced planes [28]. Yaw rotation occurs around
the z or vertical axis, pitch rotation occurs around the y or interaural axis, and roll
rotation occurs around the x or nasal-occipital axis (Fig. 10.2). While vestibular eye
movements are driven by sensory input from the 2 labyrinths, visuo-vestibular eye
movements are driven by visual input from the 2 eyes. In infantile strabismus, the 2
Fig. 10.2 Cartesian Z
coordinate system for yaw,
pitch, and roll movements
of the head. (Axes differ
from those of the Fick
coordinate system for eye
rotation.) Yaw
Roll Pitch
X Y
Visuo-Vestibular Eye Movements Arise from a Central Vestibular Imbalance th… 131
eyes function as physiologic vestibules that can directly modulate central vestibular
tone in a complementary way to a labyrinthine imbalance. These visuo-vestibular
eye movements generate latent nystagmus in the yaw plane, primary oblique muscle
overaction in the pitch plane, and dissociated vertical divergence in the roll plane
(Table 10.2) [16].
Ultimately, all 3 visuo-vestibular eye movements are clinical expressions of a
visually induced imbalance in central vestibular tone, as evidenced by the fact that
they manifest as rotations in the same head-referenced planes as vestibular eye
movements, which are governed by the orientation of the anatomical components of
the 2 labyrinths [9, 16, 19]. One could geometrically display any variable combina-
tion of these 3 movements as a single vector in 3-dimensional space based on the
size and direction of each movement in its corresponding plane.
Because these visuo-vestibular movements correspond to vector components of
a single central vestibular imbalance in 3-dimensional space, it is not surprising that
their clinical features can overlap. Thus, dissociated vertical divergence incorpo-
rates a vertical latent nystagmus, revealing the shared common origin of these
movements [29]. Clinically, primary oblique muscle overaction and dissociated ver-
tical divergence often summate to produce an admixture of overelevation in adduc-
tion [30]. Similarly, latent nystagmus incorporates vertical and torsional movements
that overlap those of dissociated vertical divergence [31].
Each plane of rotation imparts unique clinical features to its visuo-vestibular
response. The fact that latent nystagmus is a dynamic imbalance while primary
oblique muscle overaction and dissociated vertical divergence are tonic imbalances
probably reflects the natural head and body movements that occur in the 3 orthogo-
nal planes of physical space. During navigation, an animal may turn its head and
body back and forth rapidly, but pitch or tilt movements tend to be slow and sus-
tained. Our evolutionary programming has devised its compensations accordingly.
Thus, the nucleus of the optic tract responds to optokinetic motion in the yaw plane
while subcortical neural pathways that modulate the dorsal light reflex produce a
tonic postural readjustment to a perceived vertical orientation [16]. Accordingly,
horizontal optokinetic input is yoked to horizontal canal input in the yaw plane, but
pitch and roll visual stimuli are yoked to otolithic input. The cerebellar flocculus
may modulate spatially oriented sensorimotor transformations for visuo-vestibular
eye movements [32].
Table 10.2 Planar Components of Central Vestibular Disease

Vestibular Counterpart (Peripheral Vestibular Ocular
Plane Visuo-Vestibular Imbalance vs Central) Reflex
Yaw Latent nystagmus Horizontal vestibular nystagmus Dynamic
(peripheral)
Pitch Primary inferior oblique Bilateral alternating of skew Tonic
overaction deviation (central)
Roll Dissociated vertical Skew deviation (central) Tonic
divergence
Visual Reflexes are Stereoisomers of Vestibular Reflexes
For each plane of head rotation, visuo-vestibular eye movements generate a

3-dimensional rotation of the eyes that is a mirror image of its corresponding
vestibulo-ocular imbalance. Thus, latent nystagmus is the visual counterpart to hori-
zontal vestibular nystagmus, primary oblique muscle overaction is the mirror-image
visual counterpart to bilateral alternating skew deviation, and dissociated vertical
divergence is the visual counterpart to the ocular tilt reaction of which skew devia-
tion is a component. For unity to exist under physiologic conditions, visuo-vestibular
eye movements must complement vestibulo-ocular movements so that their additive
functions summate to 0 in each plane. This central yoking of binocular and bilaby-
rinthine sensory input within the central vestibular system underscores the role of
symmetry in our evolutionary design and explains the close geometric correspon-
dence between the semicircular canals and the extraocular muscles [33]. The system
fits together nicely.
Conclusions
The eye movements of infantile strabismus are harmonics of our earlier orchestra-
tion; echoes of our visual ancestry. Visual analogues of latent nystagmus, dissoci-
ated vertical divergence, and primary oblique muscle overaction are found in the
normal visuo-vestibular eye movements of lateral-eyed animals. These visual
reflexes subserve balance by enabling the animal to leverage fluctuations in binocu-
lar visual input to maintain postural orientation.
By disrupting the development of frontal binocular vision, infantile strabismus
permits our primitive visual reflexes to “bubble to the surface.” The resulting ocu-
lar intrusion movements reflect an imbalance of binocular visual input in 3 planes
of physical space. These overlapping visuo-vestibular eye movements can be
deconstructed clinically into their 3 dimensions, corresponding to yaw, pitch, and
roll. The symphony of binocular movements that accompanies infantile strabis-
mus is a necessary expression of our primitive heritage that impelled them into
being. At a higher level, these movements reveal the wisdom of antiquity by
reminding us how survival in a gravitational world once necessitated the evolution
of a physical symmetry and its co-orchestration at both anatomical and physio-
logic levels.
Infantile strabismus provides a natural experiment to uncover the primitive
visual reflexes that lie buried within us. These visual reflexes are humanity’s
umbilical cord reaching back to a time when the 2 eyes functioned together as
sensory balance organs. In infantile strabismus, dissociated binocular vision
remains the fulcrum on which the visuo-vestibular system teeters. The beauty of
infantile strabismus is that each piece of our visual ancestry is laid out in bold
relief. We see unraveled before us a blazing display of the primitive visual
reflexes that were once necessary to maintain physical orientation in 3- dimen-
sional space.
Conclusions 133
More importantly, these preternatural eye movements provide a unique glimpse

into a larger universal system of physical order. As eloquently stated by the 19th-
century English author Thomas de Quincey:
Even the articulate or brutal sounds of the globe must be all so many languages
and ciphers that somewhere have their corresponding keys—have their own gram-
mar and syntax; and thus the least things in the universe must be secret mirrors to
the greatest [34].
References
1. Lang J. The infantile strabismus syndrome. Strabismus. 2000;8:195–9.
2. Pylyshyn ZW. Computation and cognition. Cambridge, MA: MIT Press; 1986.
3. Miller AI. Insights of genius, vol. 333. Cambridge, MA: MIT Press; 1996.
4. Gardner M. Fearful symmetry. In: The night is large: collected essays,
1938-1995. New York: St Martin’s Griffen; 1996. p. 3–12.
5. Gardner M. The Ambidextrous Universe. WH Freeman & Company:
New York; 1990.
6. McManus C. Right hand, left hand: the origins of asymmetry in brains, bodies,
atoms, and cultures. Cambridge, MA: Harvard University Press; 2002. p. 1–412.
7. Mittelstaedt H. Interaction of eye-, head-, and trunk-bound information in spa-
tial perception and control. J Vestib Res. 1997;7:283–302.
8. Brodsky MC. Dissociated vertical divergence: perceptual correlates of the
human dorsal light reflex. Arch Ophthalmol. 2002;120:1174–8.
10. Duke-Elder S. The effect of light on movement. In: System of ophthalmology:
the eye in evolution. London, UK: Henry Klimpton; 1958. p. 27–81.
11. Wolpert L. Principles of development. Oxford, UK: Oxford University Press;
1998. p. 443.
12. Schott JM, Rosser MN. The grasp and other primitive reflexes. J Neurol
Neurosurg Psychiatry. 2003;74:558–60.
13.
Zeeman WPC. Biologisches zum horopterproblem. Ophthalmologica.
1949;117:254–75.
14. Keiner GBJ. New viewpoints on the origin of squint. Martinus Nijhoff: Hague;
1951. p. 1–221.
15. Swaiman KF. Pediatric neurology: principles and practice. CV Mosby: St
Louis; 1999.
17. Ohm J. Nystagmus und Schielen bei Sehschwachen und Blinden. Stuttgart:
Enke; 1958. p. 22.
18. Mattheus S, Kommerell G. Reversed fixation test as a means to differentiate
between dissociated and nondissociated strabismus. Strabismus. 1996;4:3–9.

1964;1:77–111.

J Comp Physiol. 1983;150:473–81.
23. Kardon R. The pupil. In: Adler’s physiology of the eye: clinical application. St
Louis: CV Mosby; 2002. p. 714–33.
24. Dell’Osso LF, Abel LA, Daroff RB. Latent/ manifest latent nystagmus reversal
using an ocular prosthesis: implications for vision and ocular dominance. Invest
26. Lyle TK. Worth and Chavesse’s Squint: the binocular reflexes and the treatment
of strabismus. 8th ed. Philadelphia: Blakiston Co; 1950. p. 15–20.
27. Schor CM. Neural control of eye movements. In: Adler’s physiology of the eye:
clinical application. St Louis: CV Mosby; 2002. p. 834–9.
28. Leigh RJ, Zee DS. The neurology of eye movements. 3rd ed. Oxford, UK:
Oxford University Press; 1999. p. 263.
ment of strabismus. 4th ed. St Louis: Mosby-Year Book; 1990. p. 341–5.
31. Guyton DL. Dissociated vertical deviation: etiology, mechanisms, and associ-
ated phenomena. J AAPOS. 2000;4:131–44.
32. Simpson JI, Van der Steen J, Tan J, et al. Representations of ocular rotations in
the cerebellar flocculus of the rabbit. In: JHJ A, Hulliger M, editors. Progress in
brain research, vol. 80; 1989. p. 213–23.
ments in lateral-eyed animals and frontal-eyed animals. Ann N Y Acad Sci.
1981;374:20–30.
34. De Quincey T. The Collected Writings of Thomas De Quincey, vol. 1. London:
A & C Black; 1896-1897. p. 129.
Postscript
This teleological analysis attempts to reconstruct infantile strabismus based on fun-

damental physical principles involving the basic orientation of the vestibular canal-
based system and the corollary orientation of the extraocular muscles. The goal is to
try to not only show how dissociated movements occur, but to provide a raison
d’etre for these movements. In this article, I propose that the two eyes revert to their
secondary function as balance organs in the setting of infantile strabismus, and that
the unique dissociated eye movements of infantile strabismus are visuo-vestibular
in origin. This conclusion is based on the substantial experimental evidence that the
Postscript 135
subcortical visuo-vestibular pathways mediate similar movements in lower animals,

where binocular from the two eyes gets transmitted via the cerebellum to the ves-
tibular nucleus, where it is yoked together with binaural input from the two laby-
rinths. The article then explains how each of the three recognized dissociated eye
movements can be understood as arising from a binocular visual imbalance in one
of three orthogonal planes of visual space. Accordingly, a new classification of
visuo-vestibular eye movements becomes necessary to define and describe these
dissociated eye movements.
At the time this article was published, I did not realize that the accessory optic
system (AOS) provided the requisite subcortical neurologic substrate to naturally
encode visual motion in a semicircular canal coordinate system. As discussed later
in Chap. 9, this atavistic system inevitably generates torsional eye movements as
viewed from the frontal-plane. Torsional eye movements can only be generated at
the level of the vestibular system (from either labyrinthine or early binocular visual
imbalance). It is not known whether the AOS is reconfigured to produce the inverse
torsional ocular rotations to those arising from vestibular stimulation in frontal-eyed
animals. While these subterranean visual pathways are strictly subcortical in lateral-
eyed animals, the visual cortex becomes secondarily reconfigured (or retrofitted) to
match the subcortical template, enabling these visuo-vestibular movements to be
secondarily driven through the visual cortex in primates. This is why differing evo-
lutionary perspectives can easily give rise to opposing scientific positions in the
field of strabismus.
This analysis codifies the different dissociated eye movements in infantile stra-
bismus into a single unified construct. Despite the usefulness of this Cartesian coor-
dinate system (yaw, pitch, and roll) in classifying dissociated eye movements,
however, Simpson and Graf have emphasized that this semicircular canal-based
coordinate system does not naturally encode the extrinsic rectangular Cartesian ref-
erence frame of orthogonal movements specific to yaw, pitch, and roll that we inter-
pret in this context. Indeed, this canal-based geometric template requires a
sensorimotor transformation to generate the variable ocular kinematics that charac-
terize different vertebrate species. (for detailed analysis see Simpson JI, Graf W:
The selection of reference frames by nature and its investigators. In Berthoz &
Melvll Jones (eds): Adaptive mechanisms in gaze control: Facts and Theories, 1995,
Elsevier Science, pp 3–16).
It is surprising to many that the human cerebellum is also organized in a canal-
based coordinate system that bilaterally summates to generate pursuit movements in
two dimensions. (Fitzgibbon EG et al: Torsional nystagmus during vertical pursuit.
J Neuro-Ophthalmol 1996;16:79–90). The normal cerebellum functions as a senso-
rimotor integrator to minimize torsion and effectively compress this three-
dimensional visuo-vestibular template into a two-dimensional plane conducive to
foveal pursuit and stereopsis. The ultimate irony is that our ocular motor system
must be compressed into two dimensions for our visual cortex to process the visual
world in three dimensions.
An Expanded View of Infantile Esotropia
Bottoms Up!
11
In the August issue of the Archives, Dr Lawrence Tychsen wrote an accompanying

editorial [1] to my article on the accessory optic system (AOS) and its potential role
in infantile strabismus [2]. I thank him for sharing his thoughts and taking an inter-
est in this work and am pleased to have the opportunity to comment further on this
subject.
Some believe that infantile esotropia is caused solely by abnormal binocular con-
nections within the visual cortex. These connections include the primary visual cor-
tex as well as higher cortical centers within the middle temporal area/medial superior
temporal area that are involved in visual motion processing [2–4]. According to this
view, latent nystagmus is attributed to a cortical pursuit imbalance in the middle
temporal area/medial superior temporal area with disinhibition of convergence lead-
ing to infantile esotropia [5]. The finding of a normal vestibular ocular response is
thought to rule out involvement of the vestibular system in infantile esotropia [6].
The other dissociated movements that accompany infantile esotropia are then attrib-
uted to active vergence damping at the cortical level [2]. This model requires that
visually driven torsional eye movements be generated by the visual cortex and
appears to be supported by the failure of decortication to reproduce dissociated eye
movements [7]. Accordingly, periventricular leukomalacia is considered a neuro-
logic model for infantile esotropia [2].
Because the visual cortex is readily accessible for study in monkeys, most of the
investigational study of infantile esotropia has been focused on that area. Therefore,
it is understandable that this general perspective is held by some investigators. What
are the shortcomings of this perspective?
First, it attributes infantile esotropia to binocular alterations within the cerebral
cortex that are secondary to binocular misalignment rather than causal. Hubel and
Wiesel [8] first showed that cutting extraocular muscles in kittens induces alternat-
ing strabismus and corollary changes within ocular dominance columns in the pri-
mary visual cortex. These changes were clearly the effect, rather than the cause, of
infantile strabismus. In humans, no better evidence for this phenomenon can be
found than in the work of Tychsen and coworkers [3, 9], who have induced infantile
138 11 An Expanded View of Infantile Esotropia
esotropia in monkeys using base-out prisms and concluded that abnormal cortical
binocular vision is the cause of infantile esotropia. Such experiments confirm that
cortical binocular alterations following abnormal binocular visual experience can
lead to infantile esotropia.
Second, the conclusion that latent nystagmus arises from a cortical smooth pur-
suit defect [5, 6] ignores the evolutionary record. Latent nystagmus is driven by a
normal monocular nasotemporal optokinetic asymmetry that fails to resolve in chil-
dren with infantile esotropia [5, 10]. This nasal optokinetic preponderance mani-
fests as latent nystagmus when fixation with either eye induces a tonic deviation of
both eyes with nasalward drift of the fixating eye [11–15]. The question is whether
this optokinetic asymmetry is generated at the cortical or the subcortical level, and
there is experimental evidence that can be taken to support either conclusion
[13–15].
What is clear is that the same monocular nasotemporal optokinetic asymmetry is
present in lower vertebrates that do not possess a visual cortex [16, 17]. Hoffmann
[18] has shown in cats that development of binocular corticotectal pathways to the
nucleus of the optic tract (NOT) and the dorsal terminal nucleus is necessary to
cancel this asymmetry within the first year of life. So phylogenetically and ontoge-
netically, the monocular nasotemporal optokinetic asymmetry that drives latent nys-
tagmus antedates development of the binocular visual cortex. The neuroanatomical
substrate of this optokinetic asymmetry has been mapped to normal subcortical
visual pathways that travel first from the retina to the contralateral NOT and dorsal
terminal nucleus (a part of the AOS) in the pretectum and then via the cerebellar
flocculus to the vestibular nuclei [19–22]. These subcortical, visuovestibular path-
ways modulate optokinetic responses to whole-field movement (optic flow) in afo-
veate animals as opposed to cortical pursuit (a foveal function) [21]. Given that the
monocular nasotemporal optokinetic asymmetry that generates latent nystagmus
derives from a subcortical, afoveate, optokinetic pathway, a binocular defect in cor-
tical (foveal) smooth pursuit cannot be the primary cause of latent nystagmus.
The proximate cause of latent nystagmus is that these subcortical optokinetic
pathways remain operational. Since cortical pursuit pathways do not generate tor-
sional eye movements, one simply cannot explain the large torsional movements
that accompany latent nystagmus without invoking activation of these subcortical
visual pathways that project to the vestibular nuclei. In this context, latent nystag-
mus conforms to a visuovestibular nystagmus that is ultimately driven by subcorti-
cal binocular visual input to the vestibular nuclei [10, 23]. Latent nystagmus also
shows the phenomenon of velocity storage [24, 25], which is unique to vestibular
eye movements [26]. As expected, the vestibular ocular reflex remains normal [6]
because labyrinthine input to the vestibular system is unaffected.
Optokinetic eye movements provide the pursuit system for lateral-eyed afoveate
animals [17]. In primates, the subcortical NOT has commissioned the cortex to pro-
vide integrated visual motion information from the 2 hemifields through the middle
temporal area and medial superior temporal area [27]. Efferent signals from the NOT
and dorsal terminal nucleus may transmit velocity error information to the pursuit
system [28]. Pursuit and visuovestibular pathways remain functionally intermingled
within the medial superior temporal area of the primate motion visual cortex [29,
An Expanded View of Infantile Esotropia 139
30]. Therefore, it is not surprising that this visual motion asymmetry is represented
at the cortical level, causing smooth pursuit to be coordinately impaired [31]. So if
one looks for cortical visual motion asymmetry in primates with latent nystagmus,
one will find it and conclude that latent nystagmus must be cortical in origin.
Third, the notion that binocular cortical alterations unmask an innate convergence
bias to produce infantile esotropia is problematic at several levels [32]. Horwood [33,
34] has shown that normal infants display large, fleeting, convergence movements of
the eyes during the first 4 months of life, which are indicative of an emerging ver-
gence system. A greater frequency of these convergence movements between ages 2
and 4 months is predictive of normal binocular development and a lower likelihood
of developing infantile esotropia [33]. If infantile esotropia represents excessive con-
vergence, why would normal infants display more excessive convergence move-
ments than infants who are destined to develop infantile esotropia? It is also
noteworthy that the prevalence of infantile esotropia increases during a period in
which these convergence movements are disappearing in normal infants [34].
Jampolsky [35, 36] has emphasized the importance of distinguishing between
convergence as an active binocular function and esotonus as a passive innervational
output that is centrally driven by unequal visual input to the two eyes. Patients with
accommodative esotropia show excessive convergence, while those with infantile
esotropia do not. Misinterpreting infantile esotropia as a motor outcome of exces-
sive convergence also supports the misconception that dissociated vertical diver-
gence and dissociated horizontal deviation can be dismissed as excessive vergence
damping of latent nystagmus [37], despite the fact that latent nystagmus is not pres-
ent in some patients who have these other dissociated eye movements [32].
Fourth, the claim that early neurologic perturbation of cortical visual input is a high
risk factor for the motor components of infantile strabismus is certainly correct, but
the motor effects are due to a secondary involvement of lower motor circuits. Children
with neurologic disease constitute the minority in those with findings that define
infantile esotropia. There is no substantial evidence that injury to the developing cere-
bral cortex is necessary for infantile esotropia to develop [38]. Indeed, the absence of
neurologic disease is definitional for infantile esotropia. So while neurologic injury is
a risk factor for the development of esotropia in infancy, it should not be inferred from
this association that infantile esotropia is a “soft sign” of neurologic disease.
This leads to a concern about the use of periventricular leukomalacia as a neuro-
logic model for infantile esotropia. It is true that this condition produces preterm
injury to binocular visual input as evidenced by the bilateral focal white matter
injury to the optic radiations that are visible on magnetic resonance imaging [39]. It
is also the case that periventricular leukomalacia is a neurologic disease that can
cause elements of the infantile esotropia syndrome to be expressed. However, chil-
dren with periventricular leukomalacia have a different constellation of clinical
findings (including tonic downgaze, primary superior oblique overaction, optic
nerve hypoplasia, spontaneous conversion of esotropia to exotropia) from those
seen in infantile esotropia [39]. Furthermore, periventricular leukomalacia is now
recognized as a global injury to both cortical and subcortical structures (including
the thalamus, basal ganglia, and cerebellum) [40], making it a poor model for infan-
tile esotropia caused by isolated injury to binocular cortical visual inputs.
Although dissociated eye movements are an expression of subcortical visual

reflexes in lower animals, there is no question that the human visual cortex is altered
by infantile esotropia [3–5, 8, 41] and that cortical suppression of one eye (brought
about by fixation, blurring, occlusion, or volition) is a necessary physiological con-
dition to generate dissociated eye movements in humans [38, 42, 43]. After normal
corticopretectal connections become established in primates, subcortical retinopre-
tectal projections may lose their influence and be replaced by cortical afferents [44],
which would explain the absence of dissociated eye movements in some [7, 13, 45]
but not all [46] patients following “decortication.”
There is ample evidence that once cortical binocular vision is disrupted early in
infancy, infantile esotropia and its dissociated eye movements are generated and
operationalized down-stream at the subcortical level [23]. Because studies of infan-
tile esotropia have been largely limited to the cortex, it is on the subcortical limb of
the operation that analysis should now be focused. What is the line of evidence
showing that strabismus is generated in the basement of the brain? The evidence can
be summarized as follows [23]:
1. Lateral-eyed animals have dissociated binocular vision.

2. Lateral-eyed animals display a constellation of subcortical visual reflexes
that are driven by unbalanced visual input to the two eyes.
3. These subcortical reflexes are visuovestibular in origin (meaning that dissoci-
ated binocular visual input is conveyed to the vestibular nucleus and to the
vestibulocerebellum, where luminance and optokinetic input are modulated).
4. Developmental disorders that impair cortical development can allow
primitive subcortical reflexes to be expressed in humans [23].
5. Humans have evolved from lower animals and retain a vestibulocerebel-
lum and an atavistic AOS.
6. Humans who have dissociated binocular vision early in development
(most commonly, but not exclusively, infantile esotropia) display a unique
constellation of dissociated eye movements.
7. Each of these dissociated eye movements in humans conforms to subcor-
tical visual reflexes that are operative in lateral-eyed animals, are evoked
by fluctuating binocular visual input, operate in one plane of visual space,
and are driven by optokinetic input or luminance disparity.
8. These dissociated eye movements (latent nystagmus, primary oblique
muscle overaction, dissociated vertical divergence, dissociated horizontal
deviation) therefore seem to arise from subcortical visual reflexes in lat-
eral-eyed animals and are triggered in humans by cortical suppression of
one eye.
9. All of these dissociated eye movements have a prominent torsional
component.
10. Human torsional eye movements and other cyclovertical deviations such
as skew deviation are a signature of vestibular disease and are conspicu-
ously absent with lesions at the level of the cerebral cortex [47].
This evidence supports the view that dissociated eye movements are generated in
the basement of the brain by primitive visual pathways that are modulated by the
binocular visual cortex. As the AOS modulates full-field optokinetic movements
similar to the NOT and operates in the canal-based coordinate system, it emerges as
an attractive candidate for generating these atavistic torsional movements [1]. In the
evolutionary sense, these are not abnormal eye movements but are normal eye
movements that are resurgent. A comprehensive neuroanatomical model for infan-
tile esotropia must therefore incorporate the cortico-mesencephalic-cerebellar path-
ways (not brainstem) [1], which are known to provide a sensorimotor circuit for the
control and modulation of eye movements in human and nonhuman primates [22].
For these reasons, we should consider that the “cause” of infantile esotropia lies
not only upstairs in the cerebral cortex but also downstairs in the cerebellar cortex.
In humans, neurologic lesions such as midline cerebellar tumors [48] and Chiari
malformations [49] can also cause acquired comitant esotropia. In the rabbit, stimu-
lation of the dorsal cap of the inferior olive (which provides visual climbing fibers
to the cerebellar flocculus) inhibits reflex contractions to the contralateral medial
rectus muscle and the ipsilateral inferior oblique muscle [50–52]. In this context,
bilateral floccular inhibition would provide one mechanism whereby infantile eso-
tropia and bilateral inferior oblique overaction could be generated without
“convergence.”
This body of evidence leads to the conclusion that infantile esotropia may also
incorporate subcortical optokinetic pathways such as the AOS. As genetic studies
provide molecular localization, we could find that lesions anywhere along this
expanded roadmap may provide the ocular motor derangements to derail develop-
ment of cortical binocularity and give rise to infantile strabismus. Depending on
their localization (visual cortex, AOS, cerebellar flocculus), targeted neuropharma-
cologic treatments may become available.
References
1. Tychsen L. The cause of infantile strabismus lies upstairs in the cerebral cortex,
not downstairs in the brainstem. Arch Ophthalmol. 2012;130(8):1060–1.
2. Brodsky MC. The accessory optic system: the fugitive visual control system in
infantile strabismus. Arch Ophthalmol. 2012;130(8):1055–9.
3. Tychsen L. Causing and curing infantile esotropia in primates: the role of decor-
related binocular input (an American Ophthalmological Society thesis). Trans
Am Ophthalmol Soc. 2007;105:564–93.
4. Tychsen L, Richards M, Wong A, et al. Spectrum of infantile esotropia in pri-
mates: behavior, brains, and orbits. J AAPOS. 2008;12(4):375–80.
5. Tychsen L, Richards M, Wong A, Foeller P, Bradley D, Burkhalter A. The neu-
ral mechanism for latent (fusion maldevelopment) nystagmus. J
Neuroophthalmol. 2010;30(3):276–83.
6. Tychsen L, Hurtig RR, Scott WE. Pursuit is impaired but the vestibule-ocular
reflex is normal in infantile strabismus. Arch Ophthalmol. 1985;103(4):536–9.
7. Tychsen L. Absence of subcortical pathway optokinetic eye movements in an
infant with cortical blindness. Strabismus. 1996;4(1):11–4.
8. Hubel DH, Wiesel TN. Binocular interaction in striate cortex of kittens reared

with artificial squint. J Neurophysiol. 1965;28(6):1041–59.
9. Tychsen L, Richards M, Wong AM, et al. Decorrelation of cerebral visual inputs
as the sufficient cause of infantile esotropia. Am Orthopt J. 2008;58:60–9.
Symposium der Deutschen Ophthalmologischen Gesellschaft, Freiburg, 1977.
and optokinetic nystagmus. Am J Optom Physiol Optic. 1983;60(6):481–502.
13. Kommerell G. Ocular motor phenomena in infantile strabismus: asymmetry in
optokinetic nystagmus and pursuit, latent nystagmus, and dissociated vertical
divergence. In: Lennerstrand G, von Noorden GK, Campos EC, editors.
Strabismus and amblyopia: experimental basis for advances in clinical manage-
ment. London, UK: Macmillan; 1988. p. 99–109.
mus. Eye. 1996;10(pt 2):274–81.
15. Braddick O. Where is the naso-temporal asymmetry? motion processing. Curr
Biol. 1996;6(3):250–3.
16. Tauber ES, Atkin A. Optomotor responses to monocular stimulation: relation to
visual system organization. Science. 1968;160(3834):1365–7.
17. Easter SS Jr. Pursuit eye movements in goldfish (Carassius auratus). Vis Res.
1972;12(4):673–88.
18. Hoffmann KP. Cortical vs subcortical contribution to the optokinetic reflex in
the cat. In: Lennerstrand G, Zee DS, Keller EL, editors. Basis of ocular motil-
ity: proceedings of a Wenner-Gren Center and Smith-Kettlewell Eye Research
Foundation International Symposium. New York: Pergamon; 1982. p. 303–11.
19. Langer T, Fuchs AF, Chubb MC, Scudder CA, Lisberger SG. Floccular effer-
ents in the rhesus macaque as revealed by autoradiography and horseradish
peroxidase. J Comp Neurol. 1985;235(1):26–37.
20. Büttner U, Boyle R, Markert G, et al. Cerebellar control of eye movements. In:
Freund HJ, Büttner U, Cohen B, Noth J, editors. Progress in brain research.
Amsterdam: Elsevier; 1996. p. 225–33.
21. Simpson JI, Giolli RA, Blanks RHI. The pretectal nuclear complex and acces-
sory optic system. In: Buettner-Ennever JA, editor. Neuroanatomy of the oculo-
motor system. Amsterdam: Elsevier; 1988. p. 335–64.
22. Voogd J, Schraa-Tam CKL, van der Geest JN, De Zeeuw CI. Visuomotor cer-
ebellum in human and nonhuman primates. Cerebellum. 2012;11(2):392–410.
24. Tusa RJ, Mustari MJ, Burrows AF, Fuchs AF. Gaze-stabilizing deficits and
latent nystagmus in monkeys with brief, early-onset visual deprivation: eye
movement recordings. J Neurophysiol. 2001;86(2):651–61.
role of the pretectal NOT. J Neurophysiol. 2001;86(2):662–75.
26. Laurens J, Angelaki DE. The functional significance of velocity storage and its
dependence on gravity. Exp Brain Res. 2011;210(3–4):407–22.
27. Ilg UJ, Hoffmann KP. Functional grouping of the cortico-pretectal projection. J
Neurophysiol. 1993;70(2):867–9.
28. Hoffmann KP, Ilg U. Role of the pretectum and accessory optic system in pur-
suit eye movements of the monkey. In: Berthoz A, editor. Multisensory control
of movement. Oxford, UK: Oxford University Press; 1993. p. 93–111.
29. Fetsch CR, Rajguru SM, Karunaratne A, Gu Y, Angelaki DE, Deangelis

GC. Spatiotemporal properties of vestibular responses in area MSTd. J
Neurophysiol. 2010;104(3):1506–22.
30. Chen A, DeAngelis GC, Angelaki DE. Convergence of vestibular and visual
self-motion signals in an area of the posterior sylvian fissure. J Neurosci.
2011;31(32):11617–27.
31. Distler C, Hoffmann KP. Private lines of cortical visual information to the nucleus
of the optic tract and dorsolateral pontine nucleus. Prog Brain Res. 2008;171:363–8.
32. Brodsky MC. Dissociated horizontal deviation: clinical spectrum, pathogene-
sis, evolutionary underpinnings, diagnosis, treatment, and potential role in the
development of infantile esotropia (an American Ophthalmological Society
thesis). Trans Am Ophthalmol Soc. 2007;105:272–93.
33. Horwood A. Neonatal ocular misalignments reflect vergence development but
rarely become esotropia. Br J Ophthalmol. 2003;87(9):1146–50.
34. Horwood A. Too much or too little: neonatal ocular misalignment frequency
can predict later abnormality. Br J Ophthalmol. 2003;87(9):1142–5.
1970;68:730–822.
36. Jampolsky A. Unequal visual inputs in strabismus management: a comparison
of human and animal strabismus. In: Burian HM, editor. Symposium on strabis-
mus: transactions of the New Orleans Academy of Ophthalmology. St Louis:
Mosby; 1978. p. 422–5.
37. Guyton DL, Cheeseman EWJ Jr, Ellis FJ, Straumann D, Zee DS. Dissociated
vertical deviation: an exaggerated normal eye movement used to damp cyclo-
vertical latent nystagmus. Trans Am Ophthalmol Soc. 1998;96:389–429.
38. Brodsky MC. The role of cortical alterations in infantile strabismus. Strabismus.
2012;20(1):35–6. https://doi.org/10.3109/09273972.2011.650817.
39. Brodsky MC, Fray KJ, Glasier CM. Perinatal cortical and subcortical visual
loss: mechanisms of injury and associated ophthalmologic signs. Ophthalmology.
2002;109(1):85–94.
40. Volpe JJ. Brain injury in premature infants: a complex amalgam of destructive
and developmental disturbances. Lancet Neurol. 2009;8(1):110–24.
41. Kiorpes L, Walton PJ, O’Keefe LP, Movshon JA, Lisberger SG. Effects of
early-onset artificial strabismus on pursuit eye movements and on neuronal
responses in area MT of macaque monkeys. J Neurosci. 1996;16(20):6537–53.
Invest Ophthalmol Vis Sci. 1993;34(5):1785–92.
Arch Ophthalmol. 1999;117(9):1216–22.
44. Hoffmann KP. Neural basis for optokinetic defects in experimental models with
strabismus. In: Kaufmann H, editor. Transactions of the 16th Meeting of the
European Strabismological Association, Giessen, September 1987. Giessen:
Gahmig Druck Giessen; 1987. p. 35–6.
45. Jung R, Kornhuber HH. Results of electronystagmography in man: the value of
optokinetic, vestibular, and spontaneous nystagmus for neurologic diagnosis
and research. In: Bender MB, editor. The ocular motor system. New York:
Harper & Row; 1964. p. 428–88.
46. Ter Braak JWG, Schenk VWD, Van Vliet AGM. Visual reactions in a case of
long-standing cortical blindness. J Neurol Neurosurg Psychiatry. 1971;34:140–7.
47. Brodsky MC, Donahue SP, Vaphiades M, Brandt T. Skew deviation revisited.
Surv Ophthalmol. 2006;51(2):105–28.
48. Williams AS, Hoyt CS. Acute comitant esotropia in children with brain tumors.
49. Hentschel SJ, Yen KG, Lang FF. Chiari I malformation and acute acquired com-
itant esotropia: case report and review of the literature. J Neurosurg. 2005;102(4
suppl):407–12.
50. Ito M, Nisimaru N, Yamamoto M. Specific patterns of neuronal connexions
involved in the control of the rabbit’s vestibulo-ocular reflexes by the cerebellar
flocculus. J Physiol. 1977;265(3):833–54.
51. Ito M, Miyashita Y, Ueki A. Functional localization in the rabbits inferior olive
determined in connection with the vestibulo-ocular reflex. Neurosci Lett.
1978;8(4):283–7.
52. Takeda T, Maekawa K. Bilateral visual inputs to the dorsal cap of inferior olive:
differential localization and inhibitory interactions. Exp Brain Res.
1980;39(4):461–71.
Postscript
This article challenges the dogmatic position that essential infantile esotropia is
caused by a primary abnormality within the visual cortex. This Worthian concept of
essential infantile esotropia (EIE) arises a priori from the fact that the primary
visual cortex provides a necessary substrate for single binocular vision and stereop-
sis, and empirically from the finding that cortical motion pathways and correlative
pursuit functions are measurably abnormal. The intuitive reasoning has long been
that if the visual cortex subserves binocular vision and stereopsis, and these vital
Postscript 145
functions are disrupted in humans with EIE, then a primary aberration within the
visual cortex must causes the eyes to cross. There is no evidence that this is the case.
This commentary argues that the observed cortical aberrations could be the effect
rather than the cause of EIE. According to this Chavessian hypothesis, the early
binocular misalignment that disrupts cortical visual development can be generated
in the basement of the brain (i.e. within the cerebellar cortex).
As elaborated on in Chap. 17, there is a considerable body of evidence that this
is likely to be the case. Based on the studies of Anna Horwood, the long-held belief
that primary visual cortical aberrations disinhibit convergence is untenable, given
that normal neonates demonstrate large convergence movements, which are predic-
tive of normal binocular alignment. Increased medial rectus and inferior oblique
muscle tonus (the cause) is different from convergence (the effect).
So what is the source of this altered extraocular muscle tonus? The subcortical
accessory optic system, which remains operational in the first 2–3 months of human
infancy, is the only neurologic system that is equipped to cause individual extraocu-
lar muscles to overact and generate the dissociated torsional eye movements that
accompany EIE. As detailed in Chap. 17, extended subcortical neuroplasticity could
generate EIE in the absence of neurologic disease. Alternatively, early binocular
misalignment could perpetuate the function of these subcortical visual motion path-
ways and cause cortical motion pathways to secondarily remodel.
Whether the subcortical/cortical debate represents a dichotomy or a continuum
will soon be solved at the level of genetics. Does a genetic mutation or epigenetic
signal within the subcortical visual pathways turn off a promotor region for normal
inactivation of the AOS, leading to a cascade of secondary cortical changes, or does
failure to produce a functional protein within the binocular visual cortex generate
reversion to an older evolutionary program that secondarily reactivates that atavistic
subcortical system? Or are the subcortical and cortical visual systems both repro-
grammed by a combination of genetic and epigenetic factors? Time will tell.
Ironically, those who advocate for early strabismus surgery in EIE are often
those who attribute it to a primary fusion maldevelopment within the visual cortex
(in which case timing of surgery shouldn’t matter). The occasional finding that very
early strabismus surgery (at 3–4 months of age) can occasionally restore full stere-
opsis further militates against a primary cortical defect in binocular vision as the
underlying cause. Similarly, the association of monofixation syndrome (the inevi-
table outcome of infantile esotropia) with a stable sensory motor situation makes it
difficult to invoke the visual cortex as the underlying cause. You can’t have it both
ways. For how could same visual cortical aberration cause EIE preoperatively only
to fortify binocular vision postoperatively?
The Lizard’s Tail: An Ocular Allegory
12
Charles Darwin first articulated the essential role of evolutionary reconfiguration. In

his 1862 discussion of orchid evolution, he wrote that “When this or that part has
been spoken of as contrived for some special purpose, it must not be supposed that
it was originally always formed for this sole purpose. The regular course of events
seems to be, that a part which originally served for one purpose by slow changes
becomes adapted for different purposes” [1]. In 1982, Gould and Vrba applied the
term exaptation to define “features that now enhance fitness but were not designed
by natural selection for their current role” [2]. For example, birds’ feathers origi-
nally designed for thermal regulation are co-opted for flying [2, 3]. The universal
role of “redesign” in the evolution of complex biological and behavioral systems is
now recognized at genetic and cellular levels. As pointed out by Zetterberg and
Zetterberg [4], the resourcefulness of evolution is also conspicuous in the evolution-
ary history of the eye. In the zebrafish embryo, for example, the not really finished
(nrf) gene is essential for development of photoreceptors. However, its primary role
in chickens is to promote erythropoiesis and in humans to facilitate nuclear–mito-
chondrial interactions in a variety of cell types [5].
Which brings us to the question of why lizards have tails. The remarkable ability
of Agama lizards to land safely after jumping was the theme of a recent study in
Nature [6]. The investigators filmed Agama lizards as they jumped from a horizon-
tal platform onto a vertical wall. They found that these lizards utilize flexion and
extension movements of their tails to position their trunks at the optimal pitch angle
for landing. As discussed in an accompanying editorial by Alexander [7], this mid-
air balancing act relies on the principle of conservation of angular momentum,
which dictates that the angular momentum of a system remains constant unless an
external torque acts upon it: When the lizard’s trunk starts to rotate forward, it bends
its tail upward to pitch its trunk backward and maintain aerial stability (Fig. 12.1a)
[8]. A body swinging a flexible tail experiences less rotation in the pitch plane (the
sagittal plane of forward or backward rotation around a horizontal axis) than a body
with no tail or a rigid tail [6]. So the lizard’s tail serves as a balance organ to main-
tain midair inertial control of the pitch position (and the tilt position by swinging
148 12 The Lizard’s Tail: An Ocular Allegory
a b
Fig. 12.1 Conservation of angular momentum. (Modified from Alexander [7], with permission)
from side to side) [7], locking the body into its necessary spatial coordinates. A
similar effect is used by tightrope walkers who carry a long balancing pole (really a
counterbalancing pole). When the tightrope walker starts to tip to the left, he or she
raises the right side of the pole to restore vertical orientation (Fig. 12.1b) [7].
So what does this observation have to do with ophthalmology? The human
oblique muscles have evolved to become the “lizard’s tail” of the oculomotor sys-
tem. The ancestral function of these ocular appendages in lateral-eyed animals was
to gyroscopically stabilize the eyes during pitch rotation of the body [9–11], which
requires generating a movement of sufficient speed and magnitude to minimize full-
field retinal slip and maintain stability of the visual percept. These torsional eye
movements are sensorially driven by a coordinated input to the central vestibular
system from the eyes (luminance and optokinesis) and the labyrinths [12]. In infan-
tile strabismus, the absence of binocular vision recalls this mechanism and unleashes
a primitive torsional bias, leading to primary oblique muscle overaction and correla-
tive torsion [13]. What we call primary oblique muscle overaction may be driven by
a central vestibular imbalance involving the normal gyroscopic orientation of the
pitch plane that results from dissociated binocular visual input [13, 14]. Atavism is
the appearance in an individual of characteristics presumed to have been present in
some remote ancestor, or a reversion to an earlier biological type [15]. One clinical
hallmark of an atavism is that it tends to reappear when higher developmental func-
tions become disabled. In this setting, the primary oblique muscle overaction of
infantile esotropia may be a declaration of atavistic reversion from the cortical bin-
ocular system to its subcortical progenitor that uses the perceived vertical to estab-
lish the torsional coordinates of the eyes [13].
With rare exceptions [16], binocular humans do not have tails and some consider
the function of the human ocular oblique muscles to be largely vestigial. Ask a
patient to twist his or her eyes in one direction or the other, and you will be met with
The Lizard’s Tail: An Ocular Allegory 149
a blank stare. So what good are muscles that do not produce movements and are
inaccessible to volitional drive (except in trained subjects) [17]. Jampel [18, 19]
postulated that the human ocular oblique muscles function to prevent torsion by
counteracting the torsional actions of the vertical rectus muscles when the head or
body inclines. Because conjugate torsion of the eyes creates the perception of tilt
and disconjugate torsion is interpreted as stereoscopic pitch of the visual world [8],
torsional stability simplifies a variety of central computations, equipping the visual
cortex to construct a stable internal representation of visual space as it navigates the
3-dimensional world. The human oblique muscles are the only extraocular muscles
to have contrived functional origins within anterior orbits, directionalizing their
posterior insertions to anchor the primary retinal meridians as the rectus muscles
contract to rotate the eyes. Like the lizard’s swinging tail, the human oblique mus-
cles have been evolutionarily positioned and acrobatically reprogrammed to impose
rotational stability during combined head and eye movements as they coordinately
orchestrate binocular alignment in all fields of gaze. This is a wondrous bal-
ancing act!
References
1. Darwin C. On the various contrivances by which British and foreign orchids are
fertilized by insects, and on the good effects of intercrosssing. London, UK:
John Murray; 1862. p. 346.
2. Gould SJ, Vrba ES. Exaptation: a missing term in the science of form. J Soc
Issues. 1991;47:43–65.
3. Buss DM, Haselton MG, Bleske AL, Wakefield JC. Adaptations, exaptations,
and spandrels. Am Psychol. 1998;53:533–48.
4. Zetterberg H, Zetterberg M. Evolution, exaptation, and stereopsis. Arch
Ophthalmol. 2005;123:1281.
5. Becker TS, Burgess SM, Amsterdam AH, et al. not really finished is crucial for
development of the zebrafish outer retina and encodes a transcription factor
highly homologous to human nuclear respiratory factor-1 and avian initiation
binding repressor. Development. 1998;125:4369–78.
6. Libby T, Moore TY, Chang-Sui E, et al. Tail-assisted pitch control in lizards,
robots, and dinosaurs. Nature. 2012;481:181–4.
7. Alexander RM. Leaping lizards and dinosaurs. Nature. 2012;481:148–9.
8. Jusufi A, Kawano DT, Libby T, Full RJ. Righting and turning in mid-air using
appendage inertia: reptile tails, analytical models and bio-inspired robots.
Bioinsp Biomim. 2010;5(045001):1–12.
9. Von Holst E. Über den Lichtrückenreflex bei der Fische. Pubbl Stn Zool Napoli
II. 1935;15:143–58.
10. Walls GL. The vertebrate eye and its adaptive radiation. Bloomfield Hills:
Cranbrook Institute of Science; 1942. p. 303.
11. Collewijn H, Noorduin H. Vertical and torsional optokinetic eye movements in
the rabbit. Pflügers Arch. 1972;335:87–95.
12. Graf W, Meyer DL. Eye positions in fishes suggest different modes of interac-
tion between commands and reflexes. J Comp Physiol. 1978;128:241–50.
150 12 The Lizard’s Tail: An Ocular Allegory
14. Brodsky MC. Do you really need your oblique muscles? Adaptations and exa-
ptations. Arch Ophthalmol. 2002;120:820–8.
15. Stedman’s medical dictionary. 26th ed. Baltimore: Williams & Wilkins;

1995. p. 161.
16. Ledely FO. Evolution and the human tail: a case report. N Engl J Med.
1982;306:1212–5.
17. Balliet R, Nakayama K. Training of voluntary torsion. Invest Ophthalmol Vis
Sci. 1978;17:303–14.
18. Jampel RS. Ocular torsion and the function of the vertical extraocular muscles.
Am J Ophthalmol. 1975;79:292–304.
19. Jampel RS. Ocular torsion and the primary retinal meridian. Am J Ophthalmol.
1981;91:14–24.
Postscript
This metaphorical article examines the inherent complexities of compensatory bal-

ancing movements that are critical in our daily functioning, and demonstrates how
our ocular oblique muscles have been reassigned to execute this function. The
principle of conservation of angular momentum is operationalized at an uncon-
scious level. When initiating a leftward turn on a bicycle, for example, I notice that
my handle bars move instantaneously to the right just before the turn. Obviously, I
am actively turning the handle bars, but this movement is completely unconscious.
This paradoxical movement also involves conservation of angular momentum. In
order to turn left, I have to first get my body leaning to the left. But how can I do
this? By quickly rotating the handlebars to the right, my body is shifted to the left
to conserve angular momentum. Just like the tightrope walker.
The Optokinetic Uncover Test
Subcortical Optokinesis in Infantile Esotropia
13
Infantile esotropia is characterized by the idiopathic onset of crossed eyes within the
first 6 months of life [1]. It is often accompanied by crossed fixation, primary
oblique muscle overaction, latent nystagmus, and dissociated vertical divergence
[2]. This constellation of findings is also seen in the setting of prematurity and other
neurological disorders [3].
Patients with infantile esotropia retain a monocular nasotemporal optokinetic
asymmetry (MNTA) characterized by brisk monocular optokinetic responses to
nasally moving optokinetic targets and poor monocular optokinetic responses to
temporally moving optokinetic targets [4–8]. The phenomenon of MNTA is
believed to underlie latent nystagmus [4, 7–10]. In primates and humans, MNTA is
normally seen during the first several months of life [10–22]. Its spontaneous reso-
lution coincides with maturation of binocular cortical pursuit pathways that pro-
vide the temporal component for the optokinetic reflex from each eye [10–21].
Accordingly, the persistence of MNTA in infantile strabismus has been attributed
to a cortical pursuit deficit caused by early failure of cortical binocular vision to
develop [22–24].
Neonates show poor pursuit responses to focal moving stimuli but demonstrate
strong optokinetic responses to large full-field rotating stimuli [19–21]. These full-
field responses are attributed to the activation of subcortical optokinetic pathways
that modulate full-field rotational optokinetic responses and remain active until
maturation of binocular cortical pursuit pathways within the first 6 months of life
[19–21]. This MNTA is seen in lateraleyed afoveate animals during turning move-
ments, in which the full-field velocity of the nasalward optokinetic stimulus to one
eye determines optokinetic rotation of both eyes. Because MNTA antedates devel-
opment of the visual cortex both phylogenetically and ontogenetically, debate has
152 13 The Optokinetic Uncover Test
been waged as to whether the persistent MNTA in humans with strabismus is caused
by a cortical pursuit defect, a persistent activation of subcortical optokinetic path-
ways, or a combination of both [3, 8, 10, 19, 25].
Cortical pursuit movements in foveate animals require foveal (or perifoveal)
stimulation, while subcortical optokinetic movements in afoveate animals are gen-
erated by full-field optic flow detected by peripheral retina [26]. If the persistence of
MNTA in humans with infantile strabismus is caused solely by defective cortical
pursuit, its activation should necessitate foveal (or parafoveal) fixation. We devised
the optokinetic uncover test to examine the role of peripheral retinal motion input in
generating horizontal optokinetic responses in patients with infantile strabismus.
Methods
Ten patients had a history of infantile esotropia confirmed by MNTA on optokinetic

testing using a handheld rotating drum (Fig. 13.1). Patients ranged in age from 1 to
38 years, and all underwent full ophthalmological examinations with special atten-
tion to signs of crossed fixation, latent nystagmus, primary oblique muscle overac-
tion, and amblyopia assessed by the inability to maintain fixation with one eye
(Table 13.1). Seven of 10 patients had no history of neurological disease, develop-
mental delay, or prematurity. Two of these patients had undergone previous strabis-
mus surgery and had residual esotropia. Three of 10 patients had mild neurological
disease. Two patients with mild prematurity and speech and fine-motor delays, as
well as 1 patient with Down syndrome, were included because their clinical findings
otherwise conformed to those of infantile esotropia.
Results
On optokinetic testing, all patients showed brisk responses to nasally directed mon-
ocular optokinetic targets and poor responses to temporally directed optokinetic tar-
gets with each eye. During attempted pursuit of temporally directed optokinetic
targets, removal of the occluder from the contralateral eye produced an immediate
improvement in optokinetic responses. In 3 patients with alternating fixation, this
improved optokinetic response produced a fixation shift to the contralateral eye,
allowing the optokinetic response to be foveally driven by the eye receiving the
nasally directed stimulus. In 7 patients with a fixation preference for one eye, this
improved optokinetic response was accompanied by an immediate or delayed fixa-
tion shift when the preferred eye was uncovered and by maintenance of fixation
when the nonpreferred eye was uncovered. In 3 patients, 3-dimensional video-ocu-
lography (Sensorimotoric Instruments) was performed using a full-field optokinetic
stimulus projected on a flat surface (stripe width of 2.2° and velocity of 15° per sec-
ond with the patient viewing at 3 m). These recordings confirmed that uncovering the
contralateral eye produced improvement in the optokinetic waveform regardless of
whether a change in fixation occurred (Fig. 13.1). In 3 patients with mild
Results 153
Rightward CKN Stimulus Leftward CKN Stimulus

Cover Reverse OKN
Right eye
left eye direction Uncover
Left eye
right eye Cover
Uncover Cover left eye
left eye Right eye
25
20
15
Right
10
5
0
–5
–10
Left
–15
–20
–25
–30
Right eye fixing Left eye fixing Left eye fixing Right eye fixing
Fig. 13.1 Video-oculography tracing of horizontal optokinetic responses to full-field stripes

(patient 4 in the video; http://www.jamaophth.com). Red line corresponds to the right eye horizon-
tal position. Blue line corresponds to the left eye horizontal position. The first 3 beats show a poor
optokinetic response when the right (amblyopic) eye is fixing a rightward (temporalward) optoki-
netic target. When the left eye is uncovered, the tracing improves, and fixation shifts to the left
(nonamblyopic) eye. The right eye is then covered, while maintaining a rightward (nasalward)
monocular stimulus to the left eye, resulting in a strong optokinetic response. The drum direction
is then reversed to leftward, creating a monocular temporalward stimulus to the left eye, which
produces a poor optokinetic response. When the right eye is uncovered, allowing it to receive a
nasalward stimulus, the optokinetic response improves even though the left eye maintains fixation.
Covering the left eye to induce a rightward (nasalward) monocular optokinetic stimulus leads to
further improvement in the optokinetic waveform. OKN indicates optokinetic nystagmus
Table 13.1 Profiles of 10 Patients Manifesting Monocular Nasotemporal Optokinetic Asymmetry

Patient Dissociated Angle Systemic or
No./Sex/ Latent Vertical Fixation at Neurological
Age, y Nystagmus Divergence Preference Amblyopia Neara Disease
1/M/4 No No Right eye Yes 30 No
2/M/4 Yes Yes Left eye Yes 20 Speech and
fine-motor delay
3/M/2 No No Left eye Yes 25 No
4/F/27 Yes Yes Left eye Yes 10 No
5/F/5 No No None No 40 Mild speech and
motor delay
6/F/38 Yes Yes Left eye No 18 No
7/M/6 Yes Yes Left eye Yes 20 No
8/M/2 No No None No 50 No
9/F/1 No No None No 45 No
10/M/4 No No None No 45 Trisomy 21
a
Prism diopter esotropia unless otherwise indicated
neurological disease or prematurity, the optokinetic uncover test produced results

identical to those in 7 patients with idiopathic infantile esotropia. Six of the untreated
patients subsequently underwent strabismus surgery to correct their infantile esotro-
pia. Other clinical findings are summarized in the Table.
Comment
In all patients with infantile esotropia, defective temporalward monocular optoki-

netic responses in the fixating eye improved when the esotropic nonfixating eye was
uncovered, allowing it to receive nasalward optokinetic input under binocular con-
ditions. From a practical viewpoint, this observation shows that the examiner should
not test for MNTA when the patient has both eyes open in the patient and assume
that the suppressed eye will not contribute to the optokinetic response in a patient
with infantile esotropia. At a mechanistic level, this observation challenges the
long-standing assumption that MNTA in humans with infantile esotropia can be
attributed solely to defective cortical pursuit. In our patients, the modulation of hori-
zontal optokinetic responses by the nonfoveal retina of the esotropic eye suggests
that subcortical optokinetic pathways must continue to modulate peripheral optic
flow in humans with infantile strabismus. This improvement of optokinetic wave-
form is consistent with the clinical observation that patients with manifest latent
nystagmus show a visible worsening of their nystagmus when the deviated eye is
covered [27].
In a 1936 publication, Ter Braak [28] showed that afoveate animals, such as the
rabbit, generate optokinetic nystagmus in response to movement of large objects,
despite the fact that they did not track small objects. In the monkey, these full-field
optokinetic responses persisted after decortication, suggesting that they were medi-
ated by a subcortical optokinetic system [29, 30]. In the rabbit, each eye is driven by
mainly forward (nasalward) optokinetic motion, which provides the stimulus for the
optokinetic responses of both eyes [31]. These asymmetrical subcortical optokinetic
responses are now known to be modulated by the contralateral nucleus of the optic
tract (NOT) and dorsal terminal nucleus (DTN) of the accessory optic system within
the mesencephalon, which respond only to ipsilateral optokinetic stimulation (nasal-
ward for the viewing eye) [32]. In rabbits and in monkeys, these pathways project
down to the inferior olive and contralateral flocculus and then on to the vestibular
nuclei to modulate visuovestibular responses to full-field optokinetic rotation dur-
ing turning movements (Fig. 13.2) [33–35].
In primates, development of binocular corticopretectal pathways to the NOT
and the DTN of the accessory optic system, which provide ipsilateral pursuit
responses (temporalward for the viewing eye), are necessary to cancel this opto-
kinetic asymmetry within the first year of life (Fig. 13.3) [11–18, 36]. These cor-
ticopretectal projections to the NOT-DTN come predominantly from the middle
temporal area and the medial superior temporal area, as well as from V1 and V2
[16], while those to accessory optic nuclei (lateral terminal nucleus and medial
terminal nucleus) come exclusively from the middle temporal area and the medial
superior temporal area [37]. Within the superior temporal sulcus, the middle tem-
poral area is involved in motion detection and is responsible for pursuit move-
ments, while the medial superior temporal area modulates pursuit and
visuovestibular detection of optic flow [38, 39]. As normal binocular cortical pur-
suit pathways become established within the first 6 months of human life,
visuovestibular responses to full-field optokinetic stimulation become
Comment 155
“encephalized” as they are incorporated into the cortical pursuit system (Fig. 13.3)
[18, 19].
In humans, these same subcortical optokinetic pathways retain their nasalward
directional predominance and are believed to mediate the full-field horizontal opto-
kinetic responses that are observed in early infancy until they are rendered inactive
by the establishment of binocular cortical pursuit pathways (Fig. 13.3) [20, 21].
However, the general notion that they can retain function in the absence of cortical
input has been controversial [25, 40]. For example, it has been suggested that the
relative preservation of responses to nasally directed stimuli in patients with incom-
plete bilateral occipital lobe destruction could be owing to remnants of the subcorti-
cal projection to the NOT-DTN that may have been released from cortical control
[25, 40, 41]. Ter Braak and Schenk [42] described a patient with acquired cortical
blindness who retained some preservation of full-field optokinetic responses, but
subsequent studies [43, 44] have found no evidence of this effect. These findings
suggest that subcortical optokinetic pathways, once “shut off” after the first few
months of life, are incapable of reactivating [21, 22]. However, they do not address
the question of whether specific derangements in binocular cortical development
can act to preserve their function.
Our finding that peripheral retinal optokinetic input can override defective foveal
pursuit suggests that infantile strabismus allows these subcortical optokinetic path-
ways (which rely on peripheral retinal input) to remain operational. Unlike “decor-
tication,” infantile strabismus may maintain the function of the subcortical visual
pathways in a way that prevents them from being shut off. In 1983, Schor [8] pro-
posed that selective maldevelopment of cortical binocular vision could provide a
competitive advantage to reinforce the activation of direct subcortical projections
from the nasal retina of each eye to the contralateral NOT-DTN and thereby potenti-
ate their function. According to Hoffmann and colleagues [11, 17], the Hebbian
mechanism of synaptic formation predicts that only neurons firing in a correlated
manner (in this case, sharing the same direction sensitivity) become consolidated
during development. Through this activity-dependent mechanism, the ipsilateral
direction sensitivity of the NOT-DTN would preferentially allow crossed nasal ret-
inogeniculate pathways that connect monocularly through the visual cortex to
establish corticofugal connections to the ipsilateral NOT-DTN, enabling these latent
subcortical pathways to remain functional in infantile esotropia (Fig. 13.4) [11, 17].
In this way, infantile esotropia could remodel the cortical motion pathways to selec-
tively maintain the nasally biased subcortical gateway through which unbalanced
binocular visual input can turn the gears on these evolutionarily ancient eye move-
ment control systems (Fig. 13.4) [45]. The fact that cortical suppression can elicit
latent nystagmus [46], and that peripheral retinal input can also drive this response,
suggests that selective preservation of crossed corticopretectal connections may
allow these latent subcortical mechanisms to be expressed in infantile esotropia.
This explanation has several implications for the MNTA that characterizes infan-
tile esotropia. First, a selective preservation of crossed monocular nasal retinal pro-
jections from the contralateral eye would eliminate the temporal retinal projections
subserving cortical pursuit from the ipsilateral eye, explaining why the cortical
EOM
NOT-DTN LGN
NPH
V1
O
M NRTP
N STS
(MT/MST)
d.c.
P 10
Cer
P
R
F
VN
Fig. 13.2 Neuroanatomical pathways involved in the optokinetic reflex and optokinetic nystag-
mus system of the right half of the brain in the squirrel monkey. The connections of the nasal retina
of the left eye and the temporal retina of the right eye are shown. Cer indicates cerebellum; d.c. IO,
dorsal cap of inferior olive; EOM, extraocular muscles; GS, Scarpa ganglion of the vestibular
nerve; LGN, lateral geniculate nucleus; NOT-DTN, nucleus of the optic tract–dorsal terminal
nucleus; NPH, nucleus prepositus hypoglossus; NRTP, nucleus reticularis tegmenti pontis; OMN,
ocular motor nuclei; PPRF, paramedian pontine reticular formation; STS, movement sensitive
areas of the middle temporal area (MT) and the medial superior temporal area (MST) in the cortex
around the superotemporal sulcus; V1, primary visual cortex; and VN, complex of the brainstem
vestibular nuclei. (Modified from the chapter by Behrens et al [35] with permission)
Comment 157
Early Infancy Late Infancy

a Binocular stimulus
b Binocular stimulus
R L In V1-only monocular cells
R+L Binocular cells
NOT-DTN NOT-DTN
LGN LGN
SCC SCC
MST MST MST MST

MT MT MT MT
L R+L R+L L R+L R+L L
L
R R R R
V1 V1 V1 V1
Fig. 13.3 Normal cortical and subcortical projections during early human development. (Based
on a model proposed by Hoffmann [36]). The brain is viewed from the top of the head, so the left
eye is on the left. (a) In early infancy, a leftward optokinetic stimulus is transmitted contralaterally
via a subcortical pathway from the nasal retina of the right eye to the left nucleus of the optic
tract–dorsal terminal nucleus (NOT-DTN) (solid red arrow), which is sensitive to leftward motion.
Ipsilateral corticofugal input from binocular cells in the left hemisphere to the NOT-DTN (inter-
rupted green arrow) has not yet developed. (b) Later in infancy, horizontal optokinetic responses
become encephalized by late infancy as binocular cortical pursuit pathways become fully opera-
tional (solid green arrow) and subcortical optokinetic pathways regress (interrupted red arrow). At
this stage, a leftward optokinetic stimulus to both eyes stimulates corticofugal pathways projecting
from binocular cells in V1 to the middle temporal area (MT) and the medial superior temporal area
(MST) and on to the ipsilateral NOT-DTN. L indicates left eye monocular cells; LGN, lateral
geniculate nucleus; R, right eye monocular cells; R + L, cortical binocular cells (that are absent in
early infancy); and SCC, semicircular canals
component of the optokinetic asymmetry can be monocularly driven in infantile

esotropia and binocularly driven with a hemispheric lesion involving the cortical
pursuit pathways. Second, this neurodevelopmental derangement would help to
explain how the cortical motion asymmetry in infantile esotropia is dictated the
directionality of tne NOT-DTN [47–49]. Third, it suggests that infantile esotropia
can arrest development of the nascent optokinetic system at a stage wherein MNTA
can be generated from the visual cortex (top down) or from subcortical pathways
(bottom up). Fourth, as divined by Schor [8] 30 years ago, the associated monocular
corticofugal projections from the motion centers in the middle temporal area and
the medial superior temporal area to the other ipsilateral accessory optic nuclei that
control cyclovertical rotations of the eyes would (by a similar Hebbian mechanism)
explain the torsional optokinetic biases [6] and complex cyclovertical movements
that accompany latent nystagmus but remain conspicuously absent in a hemispheric
(binocular) pursuit defect. Fifth, the ability of subcortical visual pathways (Fig. 13.2)
to be activated bidirectionally via subcortical and cortical visual pathways would
render obsolete the debate about neuroanatomical localization in infantile eso-
tropia [3].
This study has several inherent limitations. First, we did not use a circular full-
field optokinetic apparatus, which produces the sensation of circularvection (the
false sensation of physical rotation) and is believed to be necessary to directly stimu-
late the visuovestibular system [50]. Although this testing paradigm would have for-
tified our conclusions, we were only able to confirm our observation using a flat
full-field optokinetic stimulus, which (like the optokinetic drum) probably elicits
pursuit responses in healthy individuals [50]. However, the enhancement of tempo-
ralward foveal optokinetic responses when peripheral retina of the nonfixating eye
was exposed to nasalward optokinetic motion demonstrates that cortical pursuit can-
not be the only system involved in the generation of MNTA (and latent nystagmus,
by inference). A selective preservation of crossed cortical projections to the NOT-
DTN from the nasal retina of the contralateral eye (Fig. 13.4) would allow nasalward
cortical pursuit pathways to generate subcortical visuovestibular eye movements
(with their associated torsional components), rendering these 2 classes of eye move-
ment indistinguishable in the setting of infantile esotropia. Second, some of the
patients with infantile esotropia had undergone previous strabismus surgery.
Consequently, the angle of the esotropic deviation was smaller than that seen in
infantile esotropia. Therefore, it cannot be assumed that the optokinetic uncover test
would have necessarily shown a positive response in these patients with infantile
esotropia prior to surgery. Nevertheless, all of our patients with unoperated infantile
esotropia showed this effect, and all of our patients had an angle of deviation large
enough to preclude perifoveal fixation, confirming that this binocular optokinetic
mechanism arises from the synthesis of foveal and peripheral retinal input. Third, the
inclusion of some patients with mild neurological disease could potentially challenge
the results of our study. However, given that all patients showed the same pattern of
responses on the optokinetic uncover test, our results suggest that these patients share
a common pathogenesis for their infantile esotropia. Fourth, because optokinetic
motion is known to displace the eyes in the direction of the fast phase [51], it might
be argued that this displacement could have led to the false conclusion that the
uncovered esodeviated eye was not fixating when the eye receiving temporal optoki-
netic stimulation was uncovered. As shown in the video, however, the fixating eye
was clearly evident in all patients, and videooculography confirmed that the nasal
retina of the esotropic eye could drive the optokinetic response.
In conclusion, the results of this study provide evidence that human subcorti-
cal optokinetic pathways may remain active in the presence of infantile
Comment 159
Infantile Esotropia Optokinetic Uncover Test
a b
Occluder Occluder
removed
R L In V1-only monocular cells
R+L Binocular cells
NOT-DTN NOT-DTN
LGN LGN
Transmission only from R

Not from L
SCC SCC
MST MST MST MST

MT MT MT MT
L L L L
R R R R
V1 V1 V1 V1
Fig. 13.4 Optokinetic uncover test in infantile esotropia. (a) When the right eye is covered and the
left eye receives a leftward (temporal) optokinetic stimulus, the temporal retina is ineffective in
generating a response. The absence of binocular cells within the visual cortex means that tempo-
rally directed stimuli cannot be transmitted to the ipsilateral nucleus of the optic tract–dorsal ter-
minal nucleus (NOT-DTN) because maturation of this function requires an early directional match
between action potentials originating from binocular cortical cells in the middle temporal area
(MT) and the medial superior temporal area (MST) with those projecting from the right eye via the
ipsidirectional subcortical pathway to the NOT-DTN. (b) When the right eye is uncovered, the
nasalward optokinetic stimulus to the right retina generates a leftward optokinetic response that
could be mediated by crossed subcortical connections to the NOT-DTN (small red arrow) or
through the retinogeniculate pathway to the monocular right eye cells within V1 of the left hemi-
sphere to MT-MST, where corticofugal projections can establish connections with cells in the left
NOT-DTN that share the same leftward direction preference (large red arrow). The fact that this
nasalward response can be driven by peripheral retina suggests that subcortical (bottom up) opto-
kinetic pathways rather than cortical foveal pursuit (top down) pathways have a predominant role
in this response. L indicates left eye monocular cells; LGN, lateral geniculate nucleus; R, right eye
monocular cells; R + L, cortical binocular cells (that are absent in infantile esotropia); and SCC,
semicircular canals
esotropia. Selective preservation of corticotectal projections from the nasal ret-

ina of the contralateral eye would enable these subcortical visual pathways to
retain their original function in the setting of infantile esotropia. The optokinetic
uncover test provides a unique insight into the role of peripheral motion detec-
tion under binocular conditions, allowing us to deconstruct the optokinetic sys-
tem into its cortical and subcortical components. The intrinsic monocular
optokinetic biases that define these subcortical pathways may be the proximate
cause of MNTA, while failure of the binocular cortical pursuit pathways to
develop may provide the permissive cause that allows these subcortical pathways
to remain functional.
References
1. Campos EC. Why do the eyes cross? a review and discussion of the nature and
origin of essential infantile esotropia, microstrabismus, accommodative esotro-
pia, and acute comitant esotropia. J AAPOS. 2008;12(4):326–31.
3. Brodsky MC. An expanded view of infantile esotropia: bottoms up! Arch
Ophthalmol. 2012;130(9):1199–202.
ment. New York: Macmillan Press; 1988. p. 99–109.
5. Wright KW. Clinical optokinetic nystagmus asymmetry in treated esotropes. J
Pediatr Ophthalmol Strabismus. 1996;33(3):153–5.
6. Valmaggia C, Proudlock F, Gottlob I. Optokinetic nystagmus in strabismus: are
asymmetries related to binocularity? Invest Ophthalmol Vis Sci.
2003;44(12):5142–50.
Symposium der Deutschen Ophthalmologischen Gesellschaft, Freiburg, 1977.
and optokinetic nystagmus. Am J Optom Physiol Optic. 1983;60(6):481–502.
9. Kiorpes L, Walton PJ, O’Keefe LP, Movshon JA, Lisberger SG. Effects of
early-onset artificial strabismus on pursuit eye movements and on neuronal
responses in area MT of macaque monkeys. J Neurosci. 1996;16(20):6537–53.
11. Hoffmann KP. Neural basis for optokinetic defects in animals with strabismus.
In: Kaufmann H, editor. Transactions of the European Strabismological
Association: 16th Meeting. 1987. p. 35–46.
12. Dürsteler MR, Wurtz RH. Pursuit and optokinetic deficits following chemical
lesions of cortical areas MT and MST. J Neurophysiol. 1988;60(3):940–65.
13. Hoffmann KP, Ilg U. Role of the pretectum and accessory optic system in pur-
suit eye movements of the monkey. In: Berthoz A, editor. Multisensory control
of movement. Oxford, UK: Oxford University Press; 1993. p. 93–111.
14. Movshon JA, Lisberger SG, Kiorpes L, et al. Effects of artificial strabismus on
pursuit eye movements and MT neurons in macaque monkeys. Perception.
1995;24:6–7.
15. Distler C, Hoffmann KP. Cortical input to the nucleus of the optic tract and
dorsal terminal nucleus (NOT-DTN) in macaques: a retrograde tracing study.
Cereb Cortex. 2001;11(6):572–80.
16. Distler C, Mustari MJ, Hoffmann KP. Cortical projections to the nucleus of the
optic tract and dorsal terminal nucleus and to the dorsolateral pontine nucleus
in macaques: a dual retrograde tracing study. J Comp Neurol.
2002;444(2):144–58.
Comment 161
17. Hoffmann KP, Bremmer F, Thiele A, Distler C. Directional asymmetry of neu-

rons in cortical areas MT and MST projecting to the NOT-DTN in macaques. J
Neurophysiol. 2002;87(4):2113–23.
18. Distler C, Hoffmann KP. Visual pathway for the optokinetic reflex in infant
macaque monkeys. J Neurosci. 2011;31(48):17659–68.
19. Braddick O. Where is the naso-temporal asymmetry? motion processing. Curr
Biol. 1996;6(3):250–3.
20. Braddick O, Atkinson J, Wattam-Bell J. Normal and anomalous development of
visual motion processing: motion coherence and “dorsal-stream vulnerability”.
Neuropsychologia. 2003;41(13):1769–84.
21. Braddick O, Atkinson J. Development of human visual function. Vis Res.
2011;51(13):1588–609.
22. Tychsen L. Causing and curing infantile esotropia in primates: the role of decor-
related binocular input (an American Ophthalmological Society thesis). Trans
Am Ophthalmol Soc. 2007;105:564–93.
23. Tychsen L, Richards M, Wong A, Foeller P, Bradley D, Burkhalter A. The neu-
ral mechanism for latent (fusion maldevelopment) nystagmus. J
Neuroophthalmol. 2010;30(3):276–83.
reflex is normal in infantile strabismus. Arch Ophthalmol. 1985;103(4):536–9.
mus. Eye (Lond). 1996;10(pt 2):274–81.
26. Winterson BJ, Steinman RM. The effect of luminance on human smooth pur-
suit of perifoveal and foveal targets. Vis Res. 1978;18(9):1165–72.
27. Dell’Osso LF, Schmidt D, Daroff RB. Latent, manifest latent, and congenital
nystagmus. Arch Ophthalmol. 1979;97(10):1877–85.
28. Ter Braak JWG. Untersuchen uber optokinetischen nystagmus. Arch

Neerlandaises Physiol. 1936;21:309–76.
29. Yee RD, Baloh RW, Honrubia V. Study of congenital nystagmus: optokinetic
nystagmus. Br J Ophthalmol. 1980;64(12):926–32.
30. Ter Braak JWG, Van Vliet AGM. Subcortical optokinetic nystagmus in the
monkey. Psychiatr Neurol Neurochir. 1963;66:277–83.
31. Collewijn H, Noorduin H. Conjugate and disjunctive optokinetic eye move-
ments in the rabbit, evoked by rotatory and translatory motion. Pflugers Arch.
1972;335(3):173–85.
the vestibulocerebellum. In: Granit R, Pampeiano O, editors. Reflex control of
posture and movement. Amsterdam: Elsevier; 1979. p. 715–24.
33. Langer T, Fuchs AF, Chubb MC, et al. Floccular efferents in the rhesus macaque
as revealed by autoradiography and horseradish peroxidase. J Comp Neurol.
1985;235(1):26–37.
34. Buttner U, Boyle R, Markert G, et al. Cerebellar control of eye movements. In:
Freund HJ, Buttner U, Cohen B, Noth J, editors. Progress in brain research.
35. Behrens F, Grusser OJ, Roggenkämper P. Open-loop and closed-loop optoki-
netic nystagmus in squirrel monkeys (Saimiri sciureus) and in man. In: Allum
JHJ, Hulliger M, editors. Progress in brain research. Amsterdam: Elsevier;

1989. p. 183–96.
36. Hoffmann KP. Cortical versus subcortical contribution to the optokinetic reflex
in the cat. In: Lennerstrand G, Zee DS, Keller EL, editors. Basis of ocular motil-
ity: proceedings of a Wenner-Gren Center and Smith-Kettlewell Eye Research
Foundation I International Symposium. New York: Pergamon Press; 1982.
p. 303–11.
37. Lui F, Gregory KM, Blanks RHI, Giolli RA. Projections from visual areas of
the cerebral cortex to pretectal nuclear complex, terminal accessory optic
nuclei, and superior colliculus in macaque monkey. J Comp Neurol.
1995;363(3):439–60.

Neurophysiol. 2010;104(3):1506–22.
selfmotion signals in an area of the posterior sylvian fissure. J Neurosci.
2011;31(32):11617–27.
40. Mehdorn E. Nasal-temporal OKN-asymmetries after bilateral occipital infarc-
tion in man. In: Lennerstrand G, Zee DS, Keller EL, editors. Functional basis of
ocular motility disorders. Oxford, UK: Pergamon Press; 1982. p. 321–4.
41. Zee DS, Tusa RJ, Herdman SJ, Butler PH, Gücer G. Effects of occipital lobec-
tomy upon eye movements in primate. J Neurophysiol. 1987;58(4):883–907.
42. Ter Braak JWG, Schenk VWD. Visual reactions in a case of long-standing cor-
tical blindness. J Neurol Neurosurg Psychiatry. 1971;34:140–7.
43. Jung R, Kornhuber HH. Results of electronystagmography in man: the value of
optokinetic, vestibular, and spontaneous nystagmus for neurologic diagnosis
and research. In: Bender MB, editor. The ocular motor system. New York:
Harper & Row; 1964. p. 428–88.
44. Tychsen L. Absence of subcortical pathway optokinetic eye movements in an
infant with cortical blindness. Strabismus. 1996;4(1):11–4.
Invest Ophthalmol Vis Sci. 1993;34(5):1785–92.
47. Norcia AM, Garcia H, Humphry R, Holmes A, Hamer RD, Orel-Bixler

D. Anomalous motion VEPs in infants and in infantile esotropia. Invest
Ophthalmol Vis Sci. 1991;32(2):436–9.
humans who had strabismus with onset in infancy. J Neurosci.
1986;6(9):2495–508.
49. Shallo-Hoffmann J, Faldon M, Hague S, Riordan-Eva P, Fells P, Gresty

M. Motion detection deficits in infantile esotropia without nystagmus. Invest
Postscript 163
50. Tian J, Zee DS, Walker MF. Rotational and translational optokinetic nystagmus
have different kinematics. Vis Res. 2007;47(7):1003–10.
51. Garbutt S, Harwood MR, Harris CM. Anticompensatory eye position (“contra-
version”) in optokinetic nystagmus. Ann N Y Acad Sci. 2002;956:445–8.
Postscript
Our consistent observation that the nasalward optokinetic input to the cortically-
suppressed, esodeviated eye can override foveal input to the fixating eye demon-
strates that subcortical optokinesis remains operative in patients with EIE. This
clinical observation has potential implications for children with other visual disor-
ders as well. For example, it has been noted that children with cortical blindness do
not have optokinetic responses, and therefore inferred that the subcortical accessory
optic system (AOS) cannot remain operative. This conclusion should be taken with
a grain of salt for several reasons. First, this reasoning is tautological, as the ones
who do show optokinetic responses are deemed to have incomplete cortical blind-
ness. Second, although the normal accessory optic system remains functional in the
first 2–3 months of life, optokinetic responses have never been tested during this
early time period to see if they are initially present in children with cortical blind-
ness. Finally, we have reported that many infants with seemingly isolated cortical
visual injury have absent vestibulo-ocular reflexes, suggesting that coexistent sub-
cortical injury may not always be evident on neuroimaging studies. To my knowl-
edge, the potential role of the accessory optic system in motion detection in humans
with “blindsight” has not been investigated.
A Unifying Neurologic Mechanism
for Infantile Nystagmus 14
Infantile nystagmus is conjugate oscillation of the eyes that may appear in isolation
but more often appears in conjunction with afferent visual deficits, such as aniridia,
albinism, optic nerve hypoplasia, congenital retinal dystrophies, isolated foveal
hypoplasia, congenital cataracts, and achiasma. Infantile nystagmus is defined by
the following general features [1–3]:
1. A predominantly horizontal pendular or jerk nystagmus

2. Onset occasionally at birth but usually at age 2 to 3 months
3. Unlike vestibular nystagmus, infantile nystagmus stays horizontal
in upgaze
4. Nystagmus often damps during near fixation
5. Absence of oscillopsia, vertigo, or imbalance
6. Central or eccentric null position
7. Apparent inversion or “reversal” of optokinetic nystagmus in approxi-
mately two-thirds of cases
8. Increase in the oscillation during attempted fixation or pursuit
9. Presence of foveation periods on eye movement recordings and
10. Absence of neurologic abnormalities outside of the visual system
At least 50% of patients with infantile nystagmus have associated maldevelop-

ment of the retinas or optic nerves [4–6]. Patients without detectable visual system
deficits often show X-linked or autosomal dominant inheritance of infantile nystag-
mus [7, 8]. A gene (FRMD7; OMIM: 300628) with strong expression in the devel-
oping neural retina has been identified as a cause of X-linked infantile nystagmus
[9–11]. Eye movement recordings show a number of waveforms that have in com-
mon short foveation periods that follow each refixation movement [1]. Although
neuronal misdirection of the anterior pathways through the chiasm can facilitate the
166 14 A Unifying Neurologic Mechanism for Infantile Nystagmus
development of infantile nystagmus in patients with albinism and achiasma, most

patients with infantile nystagmus have normal hemispheric visual-evoked poten-
tials, indicatingthat chiasmal misdirection is not necessary to generate this oscilla-
tion [12, 13].
We have come to see infantile and latent nystagmus as existing on a clinical and
neuroanatomic continuum. In humans, infantile strabismus creates a dissociated
form of binocular vision in which the 2 eyes are subjected to unequal visual input.
When cortical binocular vision fails to develop in infancy, we hypothesize that the
ocular motor disturbances of infantile strabismus correspond to subcortical binocu-
lar visual reflexes that are operative in lateral-eyed animals [14]. Dissociated eye
movements, such as latent nystagmus, manifest when the nascent binocular cortical
pursuit pathways fail to suppress monocularly driven subcortical optokinetic path-
ways that are modulated by the nucleus of the optic tract and the dorsal terminal
nucleus (NOT-DTN) of the accessory optic system (AOS) [15, 16]. It therefore
seems reasonable to question whether infantile nystagmus, a binocular oscillation
often associated with reduced central vision in both eyes, could similarly arise from
an atavistic resurgence of this subcortical visual system.
We propose that abnormal development of binocular foveal vision allows the
AOS, a primitive subcortical “pursuit” system, to remain expressed in humans and
that the inherent antagonism between the cortical and subcortical pursuit systems is
manifested in the form of infantile nystagmus. We use the term isolated instead of
idiopathic (ie, unknown cause) infantile nystagmus because the latter is incorrect as
well as misleading (ie, nystagmus in patients with associated sensory deficits is not
caused by those deficits; correlation—even time-locked correlation—does not equal
causality). The direct cause of infantile nystagmus has been attributed [17, 18] to the
normally oscillating part of the smooth pursuit system. Ocular motility studies [1–3]
have shown the nystagmus exhibited by patients with and without associated visual
sensory deficits to be identical. Because infantile nystagmus is a single entity with
a known cause, there is no foundation for arbitrarily describing the nystagmus of
one subset of patients as idiopathic.
What Is the AOS?
As detailed in a recent publication [16], the AOS is a primary visual system receiv-
ing direct visual information from the retina via one or more accessory optic tracts
[19]. The AOS is responsible for visuovestibular interaction in afoveate animals
[20–22]. Its retinal input is derived from ON-type direction-sensitive ganglion
cells. The AOS neurons have large receptive fields (averaging approximately 40°
vertically and 60° horizontally), are direction selective, and have a preference for
slow-moving stimuli [19, 22–25]. The AOS processes information about the speed
and direction of movement of large textured parts of the visual world [20, 23, 24].
After accounting for an eye movement-generated slip (efference copy), the AOS
signals motion as a function of slip of the visual world over the retinal surface and
generates corrective eye movements to stabilize the retinal image [20, 23, 24]. As
Role of the AOS in Infantile Nystagmus 167
an analyzer of world motion, the AOS subserves visual proprioception in the afove-
ate animal [20, 23, 24]. Ontogenetically, the function of this subcortical optoki-
netic system antedates full maturation of the vestibuloocular reflex in lower
vertebrates [26, 27].
The AOS is involved in determining self-motion and is uniquely organized in a
vestibular coordinate system [20, 23, 24]. Experimental studies by Simpson and
colleagues [20, 23, 25] indicate that visual and vestibular signals that produce com-
pensatory eye movements are organized about a common set of axes derived from
the orientation of the semicircular canals. Because the AOS is directionally sensitive
to low-velocity movements, whereas the vestibular system typically responds to
movements of higher velocity, the AOS and vestibular labyrinths form 2 comple-
mentary systems to detect self-motion and promote image stabilization so that
objects in the visual world can be quickly and accurately analyzed [19, 20, 24, 25].
The AOS exists in all vertebrate classes [20, 28–30], including humans [31]. In
human infancy, the AOS is believed to generate full-field subcortical optokinetic
responses within the first 2 months of life before normal binocular cortical pursuit
pathways supersede this phylogenetically ancient subcortical system (Fig. 14.1)
[32–36].
Role of the AOS in Infantile Nystagmus
Full-field optokinetic nystagmus and foveal pursuit are diametrically opposed ocu-
lar motor functions—you have to turn one off to have the other [36]. Failure to do
so produces a discordance wherein activation of one system provides visual feed-
back that reactivates the other system in a positive feedback manner. The following
clinical and experimental evidence support this mechanism as the substrate for the
horizontal oscillations of infantile nystagmus.
Pendularity
The pendular nature of infantile nystagmus may reflect the balanced antagonism of
these hierarchical optokinetic systems. Pendular pursuit oscillations result from
negative feedback and excessive time delay in the internal pursuit feedback loop.
Most of the control systems in nature (as well as those constructed by man) use
negative feedback (part of the output is fed back to subtract from the input). This
system produces a stable, time-invariant control structure. However, when there is
an increased delay in that feedback pathway, the resulting phase shift in the feed-
back signal causes an effective reversal in its sign. For example, a 180° phase delay
in a sine wave results in the positive portion coinciding with the original negative
portion. Because the feedback is negative by design (ie, the feedback signal is mul-
tiplied by −1), that delayed positive portion becomes negative and it sums with the
undelayed negative portion of the original signal. This signal continues to grow with
each cycle. The result is an oscillation similar to that produced in auditorium
a Early Infancy b Late Infancy

Binocular stimulus Binocular stimulus
1 1
2 2
2 1
1 2
NOT-OTN NOT-DEN NOT-DTN NOT-DTN

LGN LGN LGN LGN
1 2
SCC SCC
MST MST MST
MT MST
MT MT MT
V1 V1 V1
V1
V1 monocular cells
Binocular cells
Fig. 14.1 Depiction of the Brain, as Viewed From Above, Showing Normal Cortical and
Subcortical Projections During Early Human Development. (a) Early in infancy, horizontal opto-
kinetic stimuli (shown as leftward or rightward motion) from each nasal retina are transmitted via
a subcortical pathway to the contralateral nucleus of the optic tract-dorsal terminal nucleus (NOT-
DTN) of the accessory optic system (solid red arrow), which is directionally sensitive to ipsiversive
motion (ie, nasalward for the contralateral eye). During this early stage of development, the corti-
cal pursuit pathways (shown as corticofugal projection from the middle temporal area-medial
superior temporal area [MT-MST] to the ipsilateral NOT-DTN) have not yet become functional
(interrupted green arrows). (b) Later in infancy, horizontal optokinetic responses become encepha-
lized, binocular cortical pursuit pathways become fully operational (solid green arrows), and sub-
cortical optokinetic pathways regress (interrupted red arrows). L indicates left eye monocular
cells; LGN, lateral geniculate nucleus; R, right eye monocular cells; R+L, cortical binocular cells;
SCC, splenium of the corpus callosum; and VI, primary visual cortex. Based on the model pro-
posed by Hoffmann [32]
microphone and speaker systems when the volume (gain) is too high for the short,
built-in time delay from microphone to speaker In infantile nystagmus, the antago-
nistic pursuit signals produced by coexpression of the cortical pursuit system and
the subcortical AOS (which functions as an afoveate full-field pursuit system) may
provide the substrate for the underlying pendular oscillation. Saccadic refixation
movements get secondarily superimposed to provide the variable waveforms that
have been identified within the oscillation [1].
Directionality
Infantile nystagmus manifests clinically as a predominantly horizontal conjugate

nystagmus. This trajectory may reflect the antagonism between the subcortical
NOT-DTN (which generates monocularly driven optokinetic responses) and the
middle temporal area and medial superior temporal area (MT-MST) within the
visual cortex, which generates binocular horizontal pursuit movements. In mon-
keys, MST generates optokinetic responses that are largely visuovestibular in origin
[37, 38] and maximally responsive to side-to-side full-field optic flow [39, 40]. This
intrinsic directional sensitivity could explain why infantile nystagmus is usually
maximal in the horizontal plane, reflecting delayed (in isolated cases) or deranged
(in the cases with structural abnormalities with in the eye or neuronal misdirection)
maturation of the horizontal pursuit pathways involving MT-MST in the visual cor-
tex. This predominant mismatch between horizontal cortical and subcortical motion
pathways could also explain how the oscillation of infantile nystagmus maintains its
horizontal trajectory in upgaze, unlike in vestibular nystagmus. The rare patients
with vertical or oblique nystagmus [41, 42] need further study to better understand
how our hypothesis may apply to them.
Foveation Periods
Unique to infantile nystagmus is its association with brief foveation periods follow-
ing saccadic refixation [43–45]. These foveation periods utilize the preserved ability
of the fixation and cortical pursuit systems to stabilize the eyes on target until an
incursion of the subcortical optokinetic signal modulated by the AOS succeeds in
derailing these functions. During foveation periods, ocular motor functions, such as
fixation, pursuit, and the optokinetic and vestibuloocular reflexes, continue to func-
tion robustly in infantile nystagmus despite the superimposition of a pendular oscil-
lation [43–45].
Absence of Oscillopsia
The absence of oscillopsia during normal eye movements arises from an effer-
ence copy that reconstructs target and world motion so that the perception of
either is not influenced by the sweeping of the visual world across the retinas
[46–48]. The absence of oscillopsia in infantile nystagmus requires an online
efference copy to correct for a constantly changing (in amplitude and waveform)
oscillation [46–48]. There is some evidence [24] that efference copy may be
embedded within the functioning AOS. This would explain why rare patients
who first express infantile nystagmus in their adolescence do not invariably expe-
rience oscillopsia [49].
Null Position
Infantile nystagmus is often associated with an eccentric horizontal null position

leading to a gaze preference. In infantile nystagmus, this null position is the binocu-
lar equivalent of the monocular optokinetic bias that defines latent nystagmus. In the
presence of infantile esotropia, the visuovestibular-optokinetic imbalance that
drives latent nystagmus is modulated by the differential activity of each NOT-DTN,
which is sensitive to nasally directed input from the contralateral eye [50]. In the
setting of infantile nystagmus, this model would predict that any superimposed bin-
ocular visual imbalance could generate a horizontal null position without strabis-
mus via asymmetrical activation of the NOT-DTN.
Time of Onset
Infantile nystagmus is usually noted between the second and third months of life
[4]. The developmental time course of infantile nystagmus overlaps that of infantile
esotropia in that both conditions first appear during early infancy when binocular
cortical pursuit pathways normally mature sufficiently to overtake the subcortical
optokinetic system. During the first 2 months of human life, cortical pursuit path-
ways have not fully matured, and full-field optokinetic eye movements are attrib-
uted to subcortical visual systems, such as the NOT and the AOS [34, 35]. Ontogeny
recapitulates phylogeny as the emergence of infantile nystagmus coincides with the
developmental maturation of the cortical pursuit pathways, which fail to exert their
predominance over the subcortical optokinetic pathways in the first 2 months of life.
When maturation of these cortical pathways is delayed or deranged, the subcortical
pathways continue to function, with the persistent antagonism between these 2 sys-
tems expressed in the form of infantile nystagmus. The fact that antagonism cannot
be established in children with cortical visual loss demonstrates how an intact visual
cortex is necessary for infantile nystagmus to develop [51–53]. Once the visual
cortex has developed, the subcortical optokinetic pathways normally cease to func-
tion, which explains why infantile nystagmus does not arise when central binocular
visual loss develops after early infancy [53].
This explanation does not account for cases of infantile nystagmus that have
been noted to appear at birth because the cortical pursuit system is not yet opera-
tional [33–35]. Animal models [54–57] suggest that genetic mutations can invert
the directional sensitivity of the subcortical AOS to allow this oscillation to be
expressed. Inversion of the retinal slip, the error signal of the subcortical optoki-
netic feedback loop, has been demonstrated in an albino rabbit [54] and in achias-
matic zebrafish belladonna mutant [55–57], which shows nystagmus in light with
waveform characteristics superficially resembling infantile nystagmus. Inversion
of the subcortical error signal within the AOS could produce an unstable feedback
mechanism that would tend to increase retinal slip [57]. That result would allow
infantile nystagmus to manifest as the cortical pursuit system matures and would
preclude accurate calibration of pursuit damping. When mutational effects are
localized to the subcortical optokinetic system, electrophysiologic symmetry may
be preserved within the visual cortex. Conversely, some infants with normal vision
exhibit a transient nystagmus that disappears as the visual system matures in the
postnatal period [58], perhaps reflecting an atavistic expression of the normal AOS
before it is inactivated as normal connections from cortical pursuit pathways
become established. There is always the alternative that, in cases of isolated infan-
tile nystagmus appearing at birth, pursuit-system damping calibration is precluded
by some other factor (eg, genetic) independent of sensory loss or AOS directional
inversion.
Association with Seesaw Nystagmus
Eye movement recordings have shown that infantile nystagmus is accompanied by

a subclinical seesaw nystagmus [59]. Seesaw nystagmus is a cyclovertical oscilla-
tion of the eyes that has been attributed to reactivation of the lateral and medial
terminal nuclei within the AOS [60]. Because torsional eye movements are not gen-
erated within the visual cortex [50], reactivation of the AOS seems necessary to
explain the coexistent cyclovertical component of this oscillation. In rare infantile
nystagmus cases with achiasma or hypochiasma, a clinically visible seesaw nystag-
mus is also present [61].
Reversed Optokinetic Nystagmus
Despite their apparent reversal, the optokinetic responses in infantile nystagmus are
not reversed; affected patients still generate the correct optokinetic responses during
each foveation period. These responses are driven by the (still intact) cortical pur-
suit system [44]. So-called reversed horizontal optokinetic responses in infantile
nystagmus are attributable to a shift in the null zone brought about by viewing hori-
zontal optokinetic targets [62, 63]. Underlying this dynamic null shift may be the
breakthrough responses of the AOS to full-field optokinetic stimuli that are reacti-
vated by the fully functioning pursuit system.
Optokinetic motion is known to displace the resting position of eyes in the direc-
tion of the fast phase [64]. For example, rightward-moving optokinetic targets
would displace the eyes to the right, simulating right gaze and generatinga right-
beating nystagmus. This binocular displacement would produce the appearance of a
reversed optokinetic nystagmus because the shift in gaze position to the right is
equivalent to a displacement of the null position to the left [53]. In this scenario, the
optokinetic shift in gaze position to the right tells the (still active) AOS that the
world has rotated to the left, reinforcing a right-beating nystagmus.
Augmentation by Fixation and Pursuit
In contradistinction to peripheral vestibular nystagmus, the intensity of infantile

nystagmus increases during visual activity involving fixation or pursuit [65–67].
Thus, daydreaming in the light will diminish infantile nystagmus, whereas any
active gaze effort in the dark will evoke it [68]. The fixation system works synergis-
tically with the pursuit system and helps it to function smoothly by correcting for
small position errors. When fixation and pursuit are activated by attention to the
external visual world, infantile nystagmus intensifies because of increased antago-
nism between the cortical and subcortical optokinetic inputs.
Association with Congenital Visual Loss
Approximately half of the cases of infantile nystagmus are accompanied by struc-

tural or electrophysiologic derangements involving central foveal vision in both
eyes [4–6]. The common denominator in these disorders is binocular maldevelop-
ment of central foveal vision subserving high-frequency contrast sensitivity. When
present from birth, these derangements preclude the normal development of cortical
foveal pursuit and promote persistent expression of the subcortical optokinetic path-
ways driven by the NOT-DTN. Animal models [54–57] demonstrate that albinism
and achiasma can also alter the trajectories of the subcortical optokinetic pathways.
Hereditary forms of infantile nystagmus are commonly associated with no other
detectable visual system abnormalities. However, a study by Weiss and Kelly [69]
found that patients with isolated infantile nystagmus showed delayed development
of normal grating acuity responses. This finding suggests that patients with appar-
ently isolated infantile nystagmus may have a genetically determined maturational
delay in binocular development of cortical vision. Developmental retardation in the
maturation of cortical pursuit during the necessary time window could lead to per-
sistent activation of the subcortical optokinetic system and promote the expression
of infantile nystagmus. If a perturbation in developmental timing alone is sufficient
to permit the AOS to generate subcortical optokinetic responses, this temporal delay
could explain the occurrence of infantile nystagmus in patients who ultimately
develop otherwise normal visual systems. In these infants, genetic mutations or
developmental perturbations that delay the normal bilateral maturation of cortical
(foveal) pursuit pathways could be sufficient to allow the subcortical optokinetic
system to remain functional.
Superimposition of Latent Nystagmus
Infantile nystagmus syndrome and latent nystagmus (also termed fusion maldevel-
opment nystagmus syndrome) [70] exist on a continuum wherein early loss of vision
in both eyes (eg, from bilateral congenital cataracts or bilateral optic nerve hypopla-
sia) precipitates infantile nystagmus, whereas early visual loss in one eye (eg, from
a unilateral congenital cataract or unilateral optic nerve hypoplasia) gives rise to a

manifest latent nystagmus [71]. Furthermore, infantile nystagmus is often accompa-
nied by a superimposed latent nystagmus [72] causing visual acuity to be further
degraded under conditions of monocular testing [73]. Like infantile nystagmus,
latent nystagmus has a torsional component and requires persistent activation of
subcortical optokinetic pathways for its expression [50]. Latent nystagmus is gener-
ated by nasalward motion input to the NOT-DTN of the AOS from the contralateral
eye [50]. Latent nystagmus reflects a monocular optokinetic predominance in the
nasal direction that is mediated subcortically by the NOT-DTN (a part of the AOS)
within the mesencephalon [15, 50, 74]. This monocular, subcortical, nasotemporal
optokinetic asymmetry is normal in lateral-eyed animals but suppressed in humans
with cortical binocular vision [75].
Whereas latent nystagmus represents a unilateral activation of nasally directed
subcortical optokinetic pathways, infantile nystagmus signals a bilateral activation
of subcortical optokinetic pathways. This interrelationship explains how binocular
central visual loss from birth gives rise to infantile nystagmus, whereas monocular
visual loss from birth gives rise to esotropia with latent nystagmus. It also predicts
that latent nystagmus should be superimposed on infantile nystagmus when strabis-
mus coexists. This framework allows us to cross-correlate these 2 oscillations and
conceptualize them as involving the same neurologic pathways along a continuum
of top-down (cortical-subcortical) optokinetic imbalance in infantile nystagmus and
side-to-side (binocular) optokinetic imbalance in latent nystagmus. Infantile nystag-
mus could be considered the binocular form of latent nystagmus. The waveform
differences arise from the functional sites affected by these imbalances (a tonic
imbalance in the visuovestibular system resulting in latent nystagmus and an inter-
ference with the calibration of the pursuit-system damping causing infantile
nystagmus).
Involvement of Cerebellar Pathways
The responses of the AOS are modulated by the flocculus within the vestibulocere-
bellum. Experimental findings from numerous studies [76] implicated an integral
role for the cerebellum in the modulation of infantile nystagmus. High-resolution
magnetic resonance imaging using voxel-based morphometry has found that gray
matter volume increases within cortical area V5/MT, the fusiform gyrus, and the
vestibulocerebellum. These changes may relate to the antagonism between cortical
pursuit and the subcortical AOS, the effort involved in maintaining fixation, or both.
In situ hybridization studies [9, 77, 78] on the embryonic human brain have
shown that the FRMD7 mutation that characterizes X-linked infantile nystagmus is
expressed in the developing forebrain, midbrain, cerebellum, primordium, spinal
cord, and neural retina. Barreiro et al [79] modeled the ocular motor system and
concluded that infantile nystagmus could result from adaptive cerebellar mecha-
nisms. Yoshida et al [80] found that mutant mice deficient in the glutamate receptor
δ2 subunit, which have limited plasticity of synapses onto Purkinje cells, also have
Left eye Right eye
Front Right eye

Sag
Front Front
Sag Inf 0 Sag
MT
Sup A Lat R
Mod R
V1,V2 MST
Hor
Hor
Retinal slip around
Hor canal/muscle axes
Optic tract
nucleus
ADS
Vestibular nuclei and group Y NPT Flocculus
NATP
C2 F1 F2 F3 F4
Front
Sag Dentate DC
VLO
Ant C Prep Hyp
Hor C
Hor Inferior olive

Left semicircular canals
Fig. 14.2 Diagram of Cortical and Subcortical Optokinetic Pathways Mediating Infantile
Nystagmus. We propose that cortico-mesencephalic-cerebellar pathways involving the middle
temporal area–medial superior temporal area (MT-MST), the accessory optic system (AOS), the
nucleus of the optic tract, and the flocculus of the cerebellum generate this horizontal oscillation.
Pathways for horizontal eye movements are shaded blue, pathways for movements around an
oblique horizontal axis are shaded in red, and mossy fiber pathways (unrelated to the climbing fiber
pathways mediating the AOS) are shaded in green. Ant C indicates anterior canal; C2, FI, F2, F3,
and F4, layers within the cerebellar flocculus; DC, dorsal cap; Front, frontal plane; Hor, horizontal
plane; Hor C, horizontal canal; Inf O, inferior oblique muscle; Lat R, lateral rectus muscle; Med
R, medial rectus muscle; MT, middle temporal area; MST, middle superior temporal area; NPT,
nuclei of the paramedian tracts; NRTP, nucleus reticularis tegmenti pontis; Prep Hyp, nucleus
prepositus hypoglossi; Sag, sagittal plane; Sup R, superior rectus muscle; VI, V2, cortical areas;
and VLO, ventrolateral outgrowth. Question marks indicate disputed projection of the nucleus of
the optic tract to the vestibular nuclei and the NRTP. (Modified from Voogd et al. [83])
oscillations of approximately 10 Hz in Purkinje cell responses and eye movements

that resemble pendular infantile nystagmus.
Finally, periodic alternating nystagmus is modulated by the nodulus of the ves-
tibulocerebellum and requires the coexistence of a visual deficit and a vestibular
malfunction [81, 82]. Thomas et al [10] showed that periodic alternating nystagmus
occurs predominantly with missense mutations in FRMD7 and concluded that this
is most likely related to instability of the optokinetic-vestibular systems. Perhaps
efferent activity from the cerebellar flocculus modulates the slow eye movements of
Theories of Causation 175
infantile nystagmus while the cerebellar nodulus and uvula give rise to periodic
alternating nystagmus. These findings do not imply that infantile nystagmus is ana-
tomically localized within the cerebellum but that the vestibulocerebellum actively
participates in generating this oscillation (Fig. 14.2) [83].
Theories of Causation
Other causational theories for infantile nystagmus have invoked atavistic evolu-
tionary mechanisms to explain the oscillation. Yee et al [84] examined the role of
the subcortical AOS in humans and concluded from their optokinetic measure-
ments that infantile nystagmus must be due to an inherent defect in the subcortical
optokinetic system. Dell’Osso [85] rebutted this conclusion and pointed out that
the apparent full-field optokinetic defects could be attributed to the nystagmus
itself rather than to an inherent defect in the subcortical optokinetic system.
Kommerell and Mehdorn [86] ascribed infantile nystagmus to a primary defect in
the optokinetic system, which can no longer maintain retinal slip control.
Kommerell [87] and Kommerell et al [88] inferred that, because patients with
infantile nystagmus can differentiate velocities of optokinetic stimuli, the defect
cannot be in the retinocortical pathway and must be between the cortex and the
ocular motor centers in the brainstem. Optican and Zee [89] reproduced some of
the waveforms of infantile nystagmus by using a computer model to create a rever-
sal in the sign of the velocity pathway to make the neural integrator unstable.
However, the demonstration of simultaneous decelerating (due to a leaky neural
integrator) and accelerating (due to infantile nystagmus) slow eye movements in a
family who had both deficits disproved the hypothetical explanation [90]. Jacobs
and Dell’Osso [17] modeled infantile nystagmus as attributable to an intrinsic
instability in the foveal pursuit system. According to this model, evolution has
necessarily designed the cortical pursuit system to operate on the threshold of
oscillation to provide minimal response time without instability [18]. Given that
the normal horizontal pursuit system is intrinsically underdamped, it is almost
guaranteed that some individuals will fail to achieve the delicate balance between
quasiinstability (underdamped) and total instability (undamped), manifesting as
infantile nystagmus.
Harris and Berry [91, 92] have proposed that contrast sensitivity to low spatial
frequencies is enhanced by motion of an image across the retina and that the best
compromise between moving the image and maintaining the image on the fovea (or
its remnant) is to oscillate the eyes with jerk nystagmus with increasing velocity
waveforms, as is seen empirically in infantile nystagmus. During infancy, “evolu-
tion would need to tread a fine line by programming the development of ocular
motor control in tandem with foveal maturation to maximize visual contrast without
causing nystagmus” [92] (p68). Accordingly, loss of high spatial-frequency infor-
mation (whether caused by foveal, optic nerve, or optical aberrations) could lead to
an atavistic resurgence of low-frequency peripheral retinal visual mechanisms that
enhance their function during oscillation of the eyes [92].
Our hypothesis reconciles these aforementioned theories by proposing that the

absence of timely maturation of high-frequency (foveal) motion sensitivity, by
whatever cause, must secondarily delay development of high-frequency cortical
pursuit pathways, lifting the developmental brakes on the peripheral retinal motion
input to the AOS to precipitate infantile nystagmus. In this setting, persistent activa-
tion of a normal AOS could be responsible for the abnormal retinal slip control
postulated by Kommerell and Mehdorn [86]. Although an abnormal AOS may con-
tribute to infantile nystagmus in patients with albinism or achiasma (as well as in
those infants that have nystagmus at birth), the preponderance of evidence points to
a persistent activation of a normal AOS that disrupts retinal slip control during
attempted fixation and foveal pursuit. This hypothesis explains how the saccadic,
pursuit, and optokinetic systems can continue to function normally in infantile nys-
tagmus despite the superimposed aberrations that this oscillation generates. It also
assigns a specific neuroanatomic correlate to the ocular motor control systems that
have long been suspected to generate infantile nystagmus. By virtue of this analysis,
it becomes clear that infantile nystagmus arises from a neuroanatomically dispersed
system of cortico-mesencephalic-cerebellar feedback loops (Fig. 14.2) [83].
Conclusions
We hypothesize that infantile nystagmus signals a breakdown in the armistice

between subcortical and cortical optokinetic pathways during early development.
Neuroanatomically, infantile nystagmus arises from a discordance between the
foveate smooth pursuit modulated by the visual cortex and the afoveate full-field
optokinetic system modulated by the AOS. Under conditions of normal visual
development, cortical smooth pursuit overrides and suppresses the contradirectional
optokinetic input that is generated by the AOS after the first few months of life.
However, delayed or diminished development of central foveal vision may unleash
the fundamental antagonism between subcortical optokinetic pathways and cortical
pursuit, causing infantile nystagmus to be expressed. Because the AOS functions as
a subcortical pursuit system in afoveate animals, its reactivation in humans explains
the paradox that infantile nystagmus is inherent to the pursuit system, yet pursuit
gain remains normal.
Through the prism of evolution, infantile nystagmus can be seen as a clinical
expression of the fundamental antagonism between the subcortical optokinetic system
and the cortical foveal pursuit system. Because these visual motion detection systems
need to be coordinately expressed for infantile nystagmus to arise, this palindromic
oscillation does not develop in afoveate lateral-eyed animals or in humans with corti-
cal visual loss. In isolated cases, infantile nystagmus may reflect slow development of
central foveal vision, which impedes maturation of cortical pursuit and allows the
AOS to establish a foothold and exert its influence. Based on this model, it seems
plausible that isolated infantile nystagmus could also arise from a genetic failure to
inactivate the AOS even in the absence of abnormal foveal visual development.
Conclusions 177
Finally, we have proposed a hypothetical framework wherein 2 distinctly differ-

ent types of oscillations (infantile nystagmus and latent nystagmus) may be linked.
Although each has its own direct motor cause (uncalibrated pursuit damping or
tonic visuovestibular imbalance, respectively), they share a common root. That pos-
sibility is supported by the following observations: infantile nystagmus may exhibit
a latent component; infantile nystagmus may coexist with latent nystagmus; the
“null” that is characteristic of infantile nystagmus is hypothesized to arise from the
same Alexander law variation of slow-phase velocity that governs latent nystagmus;
and the low-amplitude, high-frequency pendular oscillation of NOT-DTN nystag-
mus is often found superimposed on both infantile and latent nystagmus waveforms.
As with all hypotheses, novel experiments in animals and humans will be required
to either disprove or support our proposed framework.
References
1. Dell’Osso LF, Daroff RB. Congenital nystagmus waveforms and foveation
strategy. Doc Ophthalmol. 1975;39(1):155–82.
2. Brodsky MC. Pediatric neuro-ophthalmology. 2nd ed. New York: Springer;
2010. p. 383–94.
3. Hertle RW, Dell’Osso LF. Nystagmus in infancy and childhood: current con-
cepts in mechanisms, diagnoses, and management. New York: Oxford
4. Gelbart SS, Hoyt CS. Congenital nystagmus: a clinical perspective in infancy.
Graefes Arch Clin Exp Ophthalmol. 1988;226(2):178–80.
5. Sarvananthan N, Surendran M, Roberts EO, et al. The prevalence of nystagmus:
the Leicestershire nystagmus surgery. Invest Ophthalmol Vis Sci.
2009;50(11):5201–6.
6. Hertle RW. Examination and refractive management of patients with nystag-
mus. Surv Ophthalmol. 2000;45(3):215–22.
7. Self JE, Ennis S, Collins A, et al. Fine mapping of the X-linked recessive con-
genital idiopathic nystagmus locus at Xq24-q26.3. Mol Vis. 2006;12:1211–6.
8. Self J, Lotery A. The molecular genetics of congenital idiopathic nystagmus.
Semin Ophthalmol. 2006;21(2):87–90.
9. Tarpey P, Thomas S, Sarvananthan N, et al. Mutations in FRMD7, a newly
identified member of the FERM family, cause X-linked idiopathic congenital
nystagmus. Nat Genet. 2006;38(11):1242–4.
10. Thomas S, Proudlock FA, Sarvananthan N, et al. Phenotypical characteristics of
idiopathic infantile nystagmus with and without mutations in FRMD7. Brain.
2008;131(pt 5):1259–67.
11. Watkins RJ, Patil R, Goult BT, Thomas MG, Gottlob I, Shackleton S. A novel
interaction between FRMD7 and CASK: evidence for a causal role in idio-
pathic infantile nystagmus. Hum Mol Genet. 2013;22(10):2105–18.
12. Guo SQ, Reinecke RD, Fendick M, Calhoun JH. Visual pathway abnormalities
in albinism and infantile nystagmus: VECPs and stereoacuity measurements. J
Pediatr Ophthalmol Strabismus. 1989;26(2):97–104.
13. Apkarian P, Shallo-Hoffmann J. VEP projections in congenital nystagmus;

VEP asymmetry in albinism: a comparison study. Invest Ophthalmol Vis Sci.
1991;32(9):2653–61.
17. Jacobs JB, Dell’Osso LF. Congenital nystagmus: hypotheses for its genesis and
complex waveforms within a behavioral ocular motor system model. J Vis.
2004;4(7):604–25.
18. Dell’Osso LF. Biologically relevant models of infantile nystagmus syndrome:
the requirement for behavioral ocular motor system models. Semin Ophthalmol.
2006;21(2):71–7.
19. Giolli RA, Blanks RHI, Lui F. The accessory optic system: basic organization
with an update on connectivity, neurochemistry, and function. Prog Brain Res.
2006;151:407–40.
the vestibulocerebellum. Prog Brain Res. 1979;50:715–24.
21. Maekawa K, Simpson JI. Climbing fiber activation of Purkinje cells in the floc-
culus by impulses transferred through the visual pathway. Brain Res.
1972;39(1):245–51.
22. Maekawa K, Simpson JI. Climbing fiber responses evoked in vestibulocerebel-
lum of rabbit from visual system. J Neurophysiol. 1973;36(4):649–66.
23. Simpson JI. The accessory optic system. Annu Rev Neurosci. 1984;7:13–41.
24. Simpson JI, Leonard CS, Soodak RE. The accessory optic system: analyzer of
self-motion. Ann N Y Acad Sci. 1988;545:170–9.
25. Simpson JI, Giolli RA, Blanks RHI. The pretectal nuclear complex and the
accessory optic system. In: Büttner-Ennever JA, editor. Neuroanatomy of the
oculomotor system. Amsterdam: Elsevier Science Publishers BV; 1988.
p. 335–64.
26. Beck JC, Gilland E, Tank DW, Baker R. Quantifying the ontogeny of optoki-
netic and vestibuloocular behaviors in zebrafish, medaka, and goldfish. J
Neurophysiol. 2004;92(6):3546–61.
27. Straka H. Ontogenetic rules and constraints of vestibulo-ocular reflex develop-
ment. Curr Opin Neurobiol. 2010;20(6):689–95.
28. Ebbesson SO. On the organization of central visual pathways in vertebrates.
Brain Behav Evol. 1970;3(1):178–94.
29. Marg E. The accessory optic system. Ann N Y Acad Sci. 1964;117:35–52.
30. Cooper HM, Magnin M. A common mammalian plan of accessory optic system
organization revealed in all primates. Nature. 1986;324(6096):457–9.
31. Fredericks CA, Giolli RA, Blanks RH, Sadun AA. The human accessory optic
system. Brain Res. 1988;454(1–2):116–22.
Conclusions 179
32. Hoffmann K-P. Neural basis for optokinetic defects in experimental models
with strabismus. In: Kaufmann H, editor. Transactions of the 16th Meeting of
the European Strabismological Association, September 1987. Giessen: Gahmig
Druck Giessen; 1987. p. 35–6.
33. Distler C, Hoffmann KP. Development of the optokinetic response in macaques:
a comparison with cat and man. Ann N Y Acad Sci. 2003;1004:10–8.
visual motion processing: motion coherence and “dorsal-stream vulnerability”.
Neuropsychologia. 2003;41(13):1769–84.
2011;51(13):1588–609.
36. van Die G, Collewijn H. Optokinetic nystagmus in man. Hum Neurobiol.
1982;1:111–9.

Neurophysiol. 2010;104(3):1506–22.
self-motion signals in an area of the posterior sylvian fissure. J Neurosci.
2011;31(32):11617–27.
39. Gu Y, Watkins PV, Angelaki DE, DeAngelis GC. Visual and nonvisual contribu-
tions to three-dimensional heading selectivity in the medial superior temporal
area. J Neurosci. 2006;26(1):73–85.
40. Gu Y, Fetsch CR, Adeyemo B, Deangelis GC, Angelaki DE. Decoding of MSTd
population activity accounts for variations in the precision of heading percep-
tion. Neuron. 2010;66(4):596–609.
41. Good WV, Brodsky MC, Hoyt CS, Ahn JC. Upbeating nystagmus in infants: a
sign of anterior visual pathway disease. Binoc Vis Q. 1990;5:13–8.
42. Pieh C, Simonsz-Toth B, Gottlob I. Nystagmus characteristics in congenital
stationary night blindness (CSNB). Br J Ophthalmol. 2008;92(2):236–40.
43. Dell’Osso LF, van der Steen J, Steinman RM, Collewijn H. Foveation dynamics
in congenital nystagmus, I: fixation. Doc Ophthalmol. 1992;79(1):1–23.
in congenital nystagmus, II: smooth pursuit. Doc Ophthalmol. 1992;79(1):25–49.
in congenital nystagmus, III: vestibulo-ocular reflex. Doc Ophthalmol.
1992;79(1):51–70.
46. Leigh RJ, Dell’Osso LF, Yaniglos SS, Thurston SE. Oscillopsia, retinal image
stabilization and congenital nystagmus. Invest Ophthalmol Vis Sci.
1988;29(2):279–82.
47. Dell’Osso LF, Leigh RJ. Ocular motor stability of foveation periods: required
conditions for suppression of oscillopsia. Neuro-ophthalmology.
1992;12:303–26.
48. Dell’Osso LF, Averbuch-Heller L, Leigh RJ. Oscillopsia suppression and fove-
ation period variation in congenital, latent, and acquired nystagmus. Neuro-
ophthalmology. 1997;18:163–83.
49. Gresty MA, Bronstein AM, Page NG, Rudge P. Congenital-type nystagmus
emerging in later life. Neurology. 1991;41(5):653–6.
Ophthalmol. 2012;130(9):1199–202.
51. Fielder AR, Evans NM. Is the geniculostriate system a prerequisite for nystag-
mus? Eye. 1988;2(pt 6):628–35.
52. Brodsky MC, Fray KJ, Glasier CM. Perinatal cortical and subcortical visual
loss: mechanisms of injury and associated ophthalmologic signs. Ophthalmology.
2002;109(1):85–94.
53. Cogan DC. The neurology of eye movements. Springfield: Charles C Thomas;
1948. p. 189.
54. Winterson BJ, Collewijn H. Inversion of direction-selectivity to anterior fields
in neurons of nucleus of the optictract in rabbits with ocular albinism. Brain
Res. 1981;220(1):31–49.
55. Huang Y-Y, Rinner O, Hedinger P, Liu SC, Neuhauss SC. Oculomotor instabili-
ties in zebrafish mutant belladonna: a behavioral model for congenital nystag-
mus caused by axonal misrouting. J Neurosci. 2006;26(39):9873–80.
56. Huang MY, Chen CC, Huber-Reggi SP, Neuhauss SC, Straumann D. Comparison
of infantile nystagmus syndrome in achiasmatic zebrafish and humans. Ann N
Y Acad Sci. 2011;1233:285–91.
57. Huber-Reggi SP, Chen C-C, Grimm L, Straumann D, Neuhauss SC, Huang
MY. Severity of infantile nystagmus syndrome-like ocular motor phenotype is
linked to the extent of the underlying optic nerve projection defect in zebrafish
belladonna mutant. J Neurosci. 2012;32(50):18079–86.
58. Good WV, Hou C, Carden SM. Transient, idiopathic nystagmus in infants. Dev
Med Child Neurol. 2003;45(5):304–7.
59. Dell’Osso LF, Jacobs JB, Serra A. The sub-clinical see-saw nystagmus embed-
ded in infantile nystagmus. Vis Res. 2007;47(3):393–401.
60. Nakada T, Kwee IL. Seesaw nystagmus: role of visuovestibular interaction in
its pathogenesis. J Clin Neuro-ophthalmol. 1988;8(3):171–7.
61. Dell’Osso LF, Daroff RB. Two additional scenarios for see-saw nystagmus:
achiasma and hemichiasma. J Neuro-ophthalmol. 1998;18(2):112–3.
62. LeLiever WC, Barber HO. Observations on optokinetic nystagmus in patients
with congenital nystagmus. Otolaryngol Head Neck Surg. 1981;89(1):110–6.
63. Halmagyi GM, Gresty MA, Leech J. Reversed optokinetic nystagmus (OKN):
mechanism and clinical significance. Ann Neurol. 1980;7(5):429–35.
64. Garbutt S, Harwood MR, Harris CM. Anticompensatory eye position (“contra-
version”) in optokinetic nystagmus. Ann N Y Acad Sci. 2002;956:445–8.
65. Abel LA, Talcevik LA. The effects of visual task demand on foveation in con-
genital nystagmus. Vis Res. 2005;45(9):1139–46.
66. Wiggins D, Woodhouse JM, Margrain TH, Harris CM, Erichsen JT. Infantile nys-
tagmus adapts to visual demand. Invest Ophthalmol Vis Sci. 2007;48(5):2089–94.
67. Cham KM, Anderson AJ, Abel LA. Task-induced stress and motivation decrease
foveation-period durations in infantile nystagmus syndrome. Invest Ophthalmol
Vis Sci. 2008;49(7):2977–84.
Conclusions 181
68. Dell’Osso LF. Fixation characteristics in hereditary congenital nystagmus. Am

J Optom Arch Am Acad Optom. 1973;50(2):85–90.
69. Weiss AH, Kelly JP. Acuity development in infantile nystagmus. Invest

70. Hertle RW. National Eye Institute Sponsored Classification of Eye Movement
Abnormalities and Strabismus Working Group. A next step in naming and clas-
sification of eye movement disorders and strabismus. J AAPOS. 2002;6(4):201–2.
71. Kushner BJ. Infantile uniocular blindness with bilateral nystagmus: a syn-
drome. Arch Ophthalmol. 1995;113(10):1298–300.
72. Dell’Osso LF. Congenital, latent and manifest latent nystagmus—similarities,
differences and relation to strabismus. Jpn J Ophthalmol. 1985;29(4):351–68.
73. Evans DE, Biglan AW, Troost BT. Measurement of visual acuity in latent nys-
tagmus. Ophthalmology. 1981;88(2):134–8.
role of the pretectal NOT. J Neurophysiol. 2001;86(2):662–75.
75. Ohmi M, Howard IP, Eveleigh B. Directional preponderance in human optoki-
netic nystagmus. Exp Brain Res. 1986;63(2):387–94.
76. Hüfner K, Stephan T, Flanagin VL, et al. Cerebellar and visual gray matter
brain volume increases in congenital nystagmus. Front Neurol. 2011;2:60. https://
doi.org/10.3389/fneur.2011.00060.
77. Betts-Henderson J, Bartesaghi S, Crosier M, et al. The nystagmus-

associated FRMD7 gene regulates neuronal outgrowth and development. Hum
Mol Genet. 2010;19(2):342–51. https://doi.org/10.1093/hmg/ddp500.
78. Watkins RJ, Thomas MG, Talbot CJ, et al. The role of FRMD7 in idiopathic
infantile nystagmus. J Ophthalmol. 2012;2012:460956. https://doi.
org/10.1155/2012/460956.
79. Barreiro AK, Bronski JC, Anastasio TJ. Bifurcation theory explains waveform
variability in a congenital eye movement disorder. J Comput Neurosci.
2009;26(2):321–9.
80. Yoshida T, Katoh A, Ohtsuki G, Mishina M, Hirano T. Oscillating Purkinje
neuron activity causing involuntary eye movement in a mutant mouse deficient
in the glutamate receptor δ2 subunit. J Neurosci. 2004;24(10):2440–8.
81. Furman JM, Wall C III, Pang DL. Vestibular function in periodic alternating
nystagmus. Brain. 1990;113(pt 5):1425–39.
82. Leigh RJ, Robinson DA, Zee DS. A hypothetical explanation for periodic alter-
nating nystagmus: instability in the optokinetic-vestibular system. Ann N Y
Acad Sci. 1981;374:619–35.
83. Voogd J, Schraa-Tam CKL, van der Geest JN, De Zeeuw CI. Visuomotor cer-
ebellum in human and nonhuman primates. Cerebellum. 2012;11(2):392–410.
84. Yee RD, Baloh RW, Honrubia V. Study of congenital nystagmus: optokinetic
nystagmus. Br J Ophthalmol. 1980;64(12):926–32.
85. Dell’Osso LF. Nystagmus and other ocular oscillations. In: Lessell S, van Dalen
JTW, editors. Neuro-ophthalmology, vol. 2. Amsterdam: Excerpta Medica;
1982. p. 148–71.
86. Kommerell G, Mehdorn E. Is an optokinetic defect the cause of congenital

and latent nystagmus? In: Lennerstrand G, Zee DS, Keller EL, editors.
Functional basis of ocular motility disorder. Oxford, UK: Pergamon Press;
1982. p. 159–67.
87. Kommerell G. Congenital nystagmus: control of slow tracking movements by
target offset from the fovea. Graefes Arch Clin Exp Ophthalmol.
1986;224(3):295–8.
88. Kommerell G, Horn R, Bach M. Motion perception in congenital nystagmus.
In: Keller EL, Zee DS, editors. Adaptive processes in visual and oculomotor
systems: proceedings of a Conference Held in Asolimar, California, USA,
16-20 June 1985. Oxford, UK: Pergamon; 1986. p. 485–91.
89. Optican LM, Zee DS. A hypothetical explanation of congenital nystagmus.
Biol Cybern. 1984;50(2):119–34.
90. Dell’Osso LF, Weissman BM, Leigh RJ, Abel LA, Sheth NV. Hereditary con-
genital nystagmus and gaze-holding failure: the role of the neural integrator.
Neurology. 1993;43(9):1741–9.
91. Harris C, Berry C. A distal model of congenital nystagmus as non-linear adap-
tive oscillations. Nonlinear Dyn. 2006;44:367–80.
92. Harris C, Berry D. A developmental model of infantile nystagmus. Semin
Ophthalmol. 2006;21(2):63–9.
Postscript
The strength of this analysis is that it explains virtually all components of infantile
nystagmus and affirms that this neurologic oscillation is confined within the visual
system. The recent finding that babies of “methadone moms” have delayed cortical
visual maturation leading to persistent infantile nystagmus also supports this
hypothesis. As summarized in Chap. 20, lines of evidence from several different
studies support an central role for the accessory optic system in that pathogenesis of
infantile nystagmus. How often the genetic mutations leading to idiopathic infantile
nystagmus exert their cellular effects via retinal electrochemical aberrations,
extended subcortical neuroplasticity, delayed visual cortical maturation, or by
enhancing an intrinsic ocular motor instability remains to be determined.
Intermittent Exotropia
and Accommodative Esotropia: Distinct 15
Disorders or Two Ends of a Spectrum?
In clinical diagnosis, we often imbue medical disorders with distinct personalities

based on a constellation of findings with little understanding of their underlying
causes. Our purpose here is to juxtapose intermittent exotropia and accommodative
esotropia and to consider the possibility that they represent a continuum of horizon-
tal deviation, rather than distinct mechanisms of disease.
Over years of observation, we have come to view intermittent exotropia and
accommodative esotropia as diametrical disorders of horizontal eye position that
use common binocular control mechanisms. Intermittent exotropia is characterized
by a gradual, progressive exodeviation of either eye that is present mainly during
distance fixation or when the patient is inattentive [1–3].
The measured deviation usually is greater at distance and, except in a small,
definable group of patients with a high accommodative convergence-to-
accommodation ratio [4], strabismus surgery targeted for the distance deviation pro-
duces surprisingly little risk for surgical overcorrection at near [5]. This phenomenon
is attributed to the observation that the resting position of the eyes (as determined
under nondepolarizing paralyzing anesthesia) is abnormally divergent in patients
with intermittent exotropia [2, 6–9], so that extra fusional convergence effort is
necessary to view near objects binocularly [10, 11]. In this condition, the measured
near deviation is reduced artificially by a phenomenon that Kushner termed tena-
cious proximal fusion, meaning that fusional convergence initially masks the true
exophoria at near until prolonged patching causes it to become manifest [10, 11].
We believe tenacious proximal fusion is a manifestation of phoria adaptation, a
compensatory mechanism that resets binocular alignment toward orthophoria,
thereby compensating for developmental, environmental, or pathologic alterations
in the binocular mechanism to restore the dynamic range (or fusion reserve) in
which fast fusional vergence can function [12]. Phoria adaptation operates indepen-
dently of the immediate vergence system and thereby minimizes the need for ver-
gence control. Although phoria adaptation is elicited classically by introducing
prisms before the eyes to induce a vergence error, this compensatory mechanism
also can function to reduce or eliminate the measured phoria. Patients with
184 15 Intermittent Exotropia and Accommodative Esotropia…
intermittent forms of strabismus, such as intermittent exotropia and accommodative

esotropia, retain the ability to respond to vergence perturbations using this fusional
mechanism. Phoria adaptation has been considered by some to be a cerebellar-
dependent response [13–15], but midbrain vergence-related neurons also may play
a role [16].
Conversely, accommodative esotropia is characterized by a gradually progres-
sive esodeviation that is often greater with near fixation [17, 18]. Accommodative
esotropia has long been attributed to uncorrected hyperopia that increases accom-
modative effort with corollary accommodative convergence. Compared with other
forms of strabismus, accommodative esotropia more often is associated with a high
accommodative convergence to accommodation (AC/A) ratio, which generates a
larger esodeviation during near fixation. In the absence of accommodative effort,
patients with accommodative esotropia are conceptualized as having an esodevia-
tion that is produced solely by their increased accommodative demand in the alert
state, which is necessarily greater for near.
In patients with a greater near esodeviation, however, surgical correction targeted
for the near deviation generally does not produce overcorrection of the distance
deviation, suggesting that the true distance esodeviation must be greater than the
measured distance esodeviation [19–23]. Accordingly, patients with larger near
deviations may have a large distance esophoria, but be exhibiting a “tenacious”
distance fusion, which again may be a healthy sign of phoria adaptation. That
patients with accommodative esotropia can “eat up prisms,” allowing the examiner
to build the measured esodeviation slowly with prism adaptation [ 24–26], suggests
that this is the case. The need to augment standard surgical doses for accommoda-
tive esotropia also suggests that these patients may be concealing a larger esodevia-
tion than can be measured with prism and alternate cover testing (as opposed to
controlling one, which would necessitate a corresponding phoria). Thus, the effects
of phoria adaptation become admixed with those of disparity-driven nonaccommo-
dative convergence (in intermittent exotropia) or blur-driven accommodative con-
vergence (in accommodative esotropia) to produce some of the defining clinical
signs of each condition.
Recent discoveries by Horwood et al [27] and Horwood and Riddell [28–30] at
the University of Reading have cast new light on the pathogenesis of intermittent
exotropia and its juxtaposition to accommodative esotropia. These examiners used
a specially designed remote haploscopic video-refractor to manipulate blur, dispar-
ity, and proximal looming cues separately while simultaneously monitoring accom-
modation and vergence angles [27–30]. To the surprise of many, these investigators
found that disparity cues provide the primary drive for both convergence and accom-
modation in normal subjects [27, 30]. This observation may explain why monocular
accommodation responses (which eliminate disparity cues) so often are worse than
binocular ones [27]. These investigators have found that patients with intermittent
exotropia underaccommodate in the exodeviated state during distance fixation and
overaccommodate in the orthotropic state during near fixation [28].
This unexpected finding of underaccommodation at distance does not arise from
an inherent defect in accommodation, but presumably reflects the fact that the
Intermittent Exotropia and Accommodative Esotropia… 185
control of intermittent exotropia is driven primarily by retinal disparity (which is

less at distance than at near) [28]. Patients with intermittent exotropia seem to use
disparity-induced vergence cues to restore binocular alignment, with greater con-
vergence stress demand during near fixation triggering overaccommodation, and
possibly contributing to the development of myopia over time [31]. Other investiga-
tors have used different methodologies to draw similar conclusions regarding the
primary role of disparity-induced vergence in controlling intermittent exodeviations
[32–34]. That binocular disparity seems to provide a stronger cue than blur for both
vergence and accommodation suggests that assessment of the convergence accom-
modation to convergence (CA/C) ratio may be more mechanistically accurate in
normal subjects as well as in those with intermittent exotropia [28, 35]. Horwood
and Riddell [29, 30] found patients with accommodative esotropia to be more
responsive to blur than to disparity, causing their excessive accommodative demand
to miscalibrate vergence position to produce an open-loop esodeviation that often is
greater for near fixation. To the extent that they are driven by blur, assessment of the
accommodative convergence-to-accommodation ratio provides a more accurate
mechanistic assessment of the resulting esotropia at any given distance [29].
The confluence of these findings raises the question of whether intermittent exo-
tropia and accommodative esotropia exist on a spectrum from disparity-driven con-
vergence accommodation to blur-driven accommodation convergence (Fig. 15.1). If
so, there is no evidence that this duality results from inherent innervational differ-
ences. Rather, they seem to reflect only the physiologic demands placed on the
Accomodative Phoria
exotropia – adaptation
+
Accomodation Distance
Diopters Blur +
+ Realignment
CA/C
+ AC/A
Angle Disparity +
Convergence Near
+ Realignment
–
Intermittent Phoria
exotropia adaptation
Fig. 15.1 Systemic model representations of synkinetic interaction between accommodation and
vergence highlights that those two motor systems are cross-linked and produce blur-driven accom-
modation or convergence and disparity-driven convergence or accommodation [40]. Phoria adap-
tation then changes the motor output of the vergence system to adjust binocular alignment at
fixation distances that readily permit binocular fusion. Black lines signify major operational path-
ways, whereas dashed lines signify functional crosslinkages between these pathways. AC/A =
accommodative convergence-to-accommodation ratio; CA/C = convergence accommodation-to-
convergence ratio; + − positive feedback loop; − = negative feedback loop
binocular visual system by the nature of the deviation and refractive error. Long ago,
Jampolsky [36] first presented evidence and maintained that there is no active cen-
trally mediated divergence and that apparent horizontal divergence movements of the
eyes are effectuated by deconvergence. In this way, both the central convergence and
accommodation centers function conjointly as an on-off system rather than a push-
pull system. It may reflect the existence of a central convergence center alone that is,
at its core, the neurologic substrate for intermittent exotropia and accommodative
esotropia, which both involve the modulation of central convergence effort.
But why should intermittent exotropia be more sensitive to disparity and accom-
modative esotropia be more sensitive to blur? Evolutionarily, the crossed nasal ret-
ina is more attuned to blur because there is minimal disparity when each eye sees a
different side, as is the case in fish and rabbits. However, the uncrossed temporal
retina is attuned to binocular disparity in the midline, which may explain in part the
expression of these compensatory responses under pathologic conditions in humans.
These mechanisms fit well with the original evolutionary functions of each hemiret-
ina, which must be retained in humans even after the evolution of crossed chiasm
segregates them into 2 hemispheres. The observation that both conditions are
anchored to the accommodative system (indirectly or directly) may help to explain
their gradual onset, their progressive nature, their occasional clinical overlap
[37–39], and their clinical manifestations.
Does the presence of tenacious proximal fusion in intermittent exotropia and
tenacious distance fusion in some cases of accommodative esotropia reflect an inde-
pendent compensating role of phoria adaptation in resetting horizontal vergence
tone? Although phoria adaptation usually is elicited by introducing prisms before
the eyes to induce a vergence error, this neural adaptation also can function to reset
the measured phoria and thereby negate strabismus. Strabismic patients may retain
the ability to respond to perturbations using this mechanism, perhaps even in
enhanced form. Furthermore, it makes physiologic sense that phoria adaptation
must be confined to the specific range of fixation distance that constrains the devia-
tion to fall within the limits of its compensatory fusional range. For this to be the
case, both intermittent exotropia and accommodative esotropia must invoke a bin-
ocular stabilization mechanism (i.e., phoria adaptation) that conceals the basic pho-
ria measurements that we extract after prolonged occlusion. Perhaps at a sensorimotor
level, near convergence is more robust (because of proximal or other factors), so
that the demands on phoria adaptation are actually less at near than at distance in
intermittent exotropia.
Phoria adaptation would explain how a distance-near disparity in accommoda-
tive esotropia can be managed effectively with medial rectus recessions to target the
near esodeviation without overcorrecting a much smaller distance esodeviation
[23]. This hidden mechanism would signify that some children with distance-near
disparity have a distance esodeviation that “flies under the radar” of our routine
clinical examination by prism and alternate cover testing. One test for this hypoth-
esis would be to repeat prism and alternate cover testing after prolonged occlusion
of one eye in patients with accommodative esotropia who have greater esodevia-
tions at near than distance. If phoria adaptation is operative, one would expect
Intermittent Exotropia and Accommodative Esotropia… 187
prolonged occlusion of one eye to increase the distance esodeviation to approximate

the near deviation, as suggested by results of a recent study [41]. If phoria adapta-
tion is not operative, one would expect prolonged occlusion of one eye to induce a
large postoperative exophoria at distance when the near esodeviation is eliminated.
We have not found this to be the case. As new physiologic discoveries emerge, we
soon may come to see intermittent exotropia and accommodative esotropia as two
sides of the same coin.
References
1. Jampolsky A. Differential diagnostic characteristics of intermittent exotropia
and true exophoria. Am Orthopt J. 1954;4:48–55.
2. Knapp P. Divergent deviations. In: Allen JH, editor. Strabismus ophthalmic
symposium II. St. Louis: Mosby; 1958. p. 354–63.
3. Parks MM. Concomitant exodeviations. In: Tasma W, Jaeger EA, editors.
Duane’s clinical ophthalmology. Philadelphia: Lippincott; 1991. p. 1–14.
4. Kushner BJ. Diagnosis and treatment of exotropia with a high accommodation
convergence-accommodation ratio. Arch Ophthalmol. 1999;117:221–4.
5. Kushner BJ. The distance angle to target in surgery for intermittent exotropia.
6. Jampolsky A. Treatment of exodeviations. In: Pediatric ophthalmology and
strabismus: transactions of the New Orleans Academy of Ophthalmology.
New York: Raven Press; 1986. p. 201–34.
7. Guyton DL. The 10th Bielschowsky Lecture. Changes in strabismus over time:
the roles of vergence tonus and muscle length adaptation. Binocul Vis
Strabismus Q. 2006;21:81–92.
8. Breinin GM. The position of rest during anesthesia and sleep; electromyo-
graphic observations. AMA Arch Ophthalmol. 1957;57:323–6.
9. Lancaster WB. Terminology with extended comments of the position of rest
and on fixation. In: Allen JH, editor. Strabismus ophthalmic symposium II. St.
Louis: Mosby; 1958. p. 503–22.
10. Kushner BJ. Exotropic deviations: a functional classification and approach to
treatment. Am Orthopt J. 1988;38:81–93.
11. Kushner BJ, Morton GV. Distance/near differences in intermittent exotropia.
12. Leigh J, Zee DS. The neurology of eye movements. 4th ed. New York: Oxford
13. Milder DG, Reinecke RD. Phoria adaptation to prisms. A cerebellar-dependent
response. Arch Neurol. 1983;40:339–42.
14. Kono R, Hasebe S, Ohtsuki H, et al. Impaired vertical phoria adaptation in
patients with cerebellar dysfunction. Invest Ophthalmol Vis Sci. 2002;43:673–8.
15. Takagi M, Tamargo R, Zee DS. Effects of lesions of the cerebellar oculomotor
vermis on eye movements in primate: binocular control. Prog Brain Res.
2003;142:19–33.
16. Morley JW, Judge SJ, Lindsey JW. Role of monkey midbrain near-response
neurons in phoria adaptation. J Neurophysiol. 1992;67:1475–92.
17. Costenbader F. Clinical course and management of esotropia. In: Allen JH, edi-
tor. Strabismus ophthalmic symposium II. St. Louis: Mosby; 1958. p. 325–53.
18. Parks MM. Concomitant esodeviations. In: Ocular motility and strabismus.
Hagerstown: Harper & Row; 1975. p. 99–111.
19. Parks MM. Accommodative esotropia: cause and treatment. Am Orthopt

J. 1955;5:74–9.
20. Wright KW, Bruce-Lyle L. Augmented surgery for esotropia associated with
high hypermetropia. J Pediatr Ophthalmol Strabismus. 1993;30:167–70.
21. West CE, Repka MX. A comparison of surgical techniques for the treatment of
acquired esotropia with increased accommodative convergence/accommoda-
tion ratio. J Pediatr Ophthalmol Strabismus. 1994;31:232–7.
22. Arnoldi KA, Tychsen L. Surgery for esotropia with a high accommodative con-
vergence/accommodation ratio: effects on accommodative vergence and bin-
ocularity. Ophthalmic Surg Lasers. 1996;27:342–8.
23. Kushner BJ. Surgical treatment of teenagers with high AC/A ratios. Am Orthopt
J. 2014;64:37–42.
24. Cipolli C, Bolzani R, Corazza R, et al. Vergence movements in comitant stra-
bismus. Percept Mot Skills. 1990;71:1259–64.
25. Kutschke PJ, Scott WE, Stewart SA. Prism adaptation for esotropia with a
distance-near disparity. J Pediatr Ophthalmol Strabismus. 1979;29:12–5.
26. Repka MX, Connett JE, Baker JD, Rosenbaum AL. Surgery in the prism adap-
tation study: accuracy and dose response. Prism Adaptation Study Research
Group. J Pediatr Ophthalmol Strabismus. 1992;29:150–6.
27. Horwood AM, Riddell PM, Toor SS. Convergence-accommodation beats

accommodative convergence. In: Haugen OH, editor. 35th Meeting of the
European Strabismological Association. Romania September: Bucharest;
2012. p. 2–5.
28. Horwood AM, Riddell PM. Evidence that convergence rather than accommo-
dation controls intermittent distance exotropia. Acta Ophthalmol.
2012;90:e109–17.
29. Horwood AM, Riddell PM. Accommodation and vergence response gains to
different near cues characterize specific esotropias. Strabismus. 2013;21:155–64.
30. Horwood AM, Riddell PM. Disparity-driven vs blur-driven models of accom-
modation and convergence in binocular vision and intermittent strabismus. J
AAPOS. 2014;18:576–83.
31. Ekdawi NS, Nusz KJ, Diehl NN, Mohney BG. The development of myopia
among children with intermittent exotropia. Am J Ophthalmol. 2010;149:503–7.
32. Ahn SJ, Yang HK, Hwang J-M. Binocular visual acuity in intermittent exotro-
pia: role of accommodative convergence. Am J Ophthalmol. 2012;154:981–6.
33. Laird PW, Hatt SR, Leske DA, Holmes JM. Distance stereoacuity in prism-
induced convergence stress. J AAPOS. 2008;12:370–4.
34. Laird PW, Hatt SR, Leske DA, Holmes JM. Stereoacuity and binocular visual
acuity in prism-induced exodeviation. J AAPOS. 2007;11:362–6.
35. McLin L, Schor C. Voluntary effort as a stimulus to accommodation and ver-
gence. Invest Ophthalmol Vis Sci. 1988;29:1739–46.
Postscript 189

1970;68:730–822.
37. Beneish R, Williams F, Polomeno RC, Little JM. Consecutive exotropia after
correction of hyperopia. Can J Ophthalmol. 1981;16:16–8.
38. Raab EL. Consecutive accommodative esotropia. J Pediatr Ophthalmol

Strabismus. 1985;22:58–9.
39. Weir CR, Cleary M, Dutton GN. Spontaneous consecutive extropia in children
with motor fusion. Br J Ophthalmol. 2001;85:242–3.
40. Schor C. Influence of accommodative and vergence adaptation on binocular
motor disorders. Am J Optom Physiol Optic. 1988;65:464–75.
41. Garretty T. Convergence excess esotropia: a proposed new classification and
the effect of monocular occlusion on the Ac/A ratio. J Pediatr Ophthalmol
Strabismus. 2010;47:308–12.
Postscript
In this piece, we invoke phoria adaptation to explain why strabismus surgery tar-
geted for the greater distance exodeviation does not overcorrect the smaller near
exodeviation in intermittent exotropia, and why strabismus surgery targeted for the
greater near esodeviation does not overcorrect the smaller distance esodeviation in
accommodative esotropia with high AC/A ratio. The selective effects of phoria
adaptation as a function of distance may reflect that fact that near fixation (where
convergence is superimposed upon a baseline exodeviation) brings the eyes closest
to orthotropia in intermittent exotropia whereas distance fixation (i.e. nonaccom-
modation) may bring the eyes close to orthotropia in accommodative esotropia.
Phoria adaptation will selectively compensate where it can.
This analysis led to my subsequent publication invoking phoria adaptation as the
major source of normal binocular tonus and an integral component of these inter-
mittent forms of horizontal strabismus. Thus, while visuo-vestibular tonus predomi-
nates under the dissociated binocular conditions that characterize infantile
strabismus, phoria adaptation provides the major extraocular muscle tonus pool
under most other binocular conditions. The myriad effects of phoria adaptation are
addressed more comprehensively in Chap. 18.
An Optokinetic Clue to the Pathogenesis
of Crossed Fixation in Infantile Esotropia 16
Infantile esotropia is characterized by crossed fixation, wherein each eye fixates

objects within the contralateral side of visual space [1]. Crossed fixation is attrib-
uted classically to foveal proximity to the position of the object of regard on the
nonsuppressed retina [2]. However, we report a clinical observation that calls this
classic interpretation into question. We describe 3 children with infantile esotropia
who exhibited a spontaneous fixation switch when the horizontal direction of a
rotating optokinetic drum was reversed.
Case Description All 3 children had isolated infantile esotropia. When viewing a
horizontally rotating optokinetic drum under monocular viewing conditions, all dis-
played brisk optokinetic responses to nasalward optokinetic targets and poor opto-
kinetic responses to temporalward optokinetic targets. When viewing a horizontally
rotating optokinetic drum under binocular viewing conditions, all spontaneously
fixated with the eye exposed to nasalward optokinetic motion. When the horizontal
direction of optokinetic rotation was reversed so that the contralateral eye was
receiving nasalward optokinetic input, a spontaneous fixation shift occurred after 5
to 10 seconds.
Discussion In infantile esotropia, we found that a fixation switch can be induced

solely by a reversal of viewed horizontal motion direction in the frontal plane.
Because the eyes remain esotropic under conditions of nonfixation, it is generally
assumed that having crossed eyes naturally predisposes to crossed fixation because
the visual axis of the opposite eye most closely approximates an object of regard.
Dickey et al [3] have noted that, as the object of regard is moved from one side to
the other, a midline fixation switch to the other eye spontaneously occurs in the
absence of amblyopia. The image of the fixation target falls on the cortically sup-
pressed nasal retina area of the esodeviating eye, making this fixation switch hard to
reconcile with the mechanism of foveal proximity to a nonsuppressed area of
peripheral retina. Given that the fixation target is moving nasally with respect to the
eye destined to take up fixation, this fixation switch could also be triggered by the
192 16 An Optokinetic Clue to the Pathogenesis of Crossed Fixation in Infantile Esotropia
nasal direction of the movement (relative to the nonfixating eye) rather than the
midline position of the fixated object.
During optokinetic stimulation, the spontaneous fixation shift in our 3 patients
was always from the eye with temporally directed optokinetic input to the eye with
nasally directed optokinetic input. Because infantile esotropia necessitates a large
refixation saccade to shift fixation in primary position, this observation challenges
the notion that foveal proximity to a stimulated area of nonsuppressed retina leads
to the phenomenon of crossed fixation. Our findings suggest that this peripheral
monocular nasotemporal asymmetry provides a stronger fixation stimulus than
foveal proximity.
Monocular nasotemporal optokinetic asymmetry arises from a subcortical opto-
kinetic system that is present in lateral-eyed afoveate animals [4], and is detectable
in normal binocular humans within the first several months of life [5]. It is medi-
ated by crossed (nasal) retinal pathways to the accessory optic system and nucleus
of the optic tract within the midbrain [4, 5]. In infantile esotropia, it persists
throughout life and is attributed to a failure of early cortical binocular vision to
develop [5].
The persistence of monocular nasotemporal optokinetic asymmetry in humans
creates the equivalent of a temporalward motion hemianopia during fixation with
either eye. In infantile esotropia, however, both nasal retinas become frontally
directed, raising the intriguing possibility that infantile esotropia provides a posi-
tional reconfiguration to restore bidirectional motion detection under binocular con-
ditions in humans with early binocular misalignment. This mechanism could explain
why the eyes remain crossed even during nonfixation, and why directional optoki-
netic reversal can elicit a fixation switch in humans with infantile esotropia. This
subcortical optokinetic response would remain operative despite cortical suppres-
sion of the esodeviating eye.
References
1. Costenbader F. Clinical course and management of esotropia. In: Allen JH, edi-
tor. Strabismus ophthalmic symposium II. Trans New Orleans Acad Ophthalmol.
St. Louis: Mosby; 1958. p. 325–53.
2. Agaoglu MN, LeSage SK, Joshi AC, et al. Spatial patterns of fixation-
switch behavior in strabismic monkeys. Invest Ophthalmol Vis Sci.
2014;55:1259–68.
3. Dickey CF, Metz HS, Stewart SA, et al. The diagnosis of amblyopia in crossed
fixation. J Pediatr Ophthalmol Strabismus. 1991;28:171–5.
4. Wallman J. Subcortical optokinetic mechanisms. In: Miles FA, Wallman J, edi-
tors. Visual motion and its role in stabilization of gaze. Amsterdam: Elsevier
Science; 1993. p. 321–42.
5. Braddick O. Where is the naso-temporal asymmetry? Motion processing. Curr
Biol. 1996;6:250–3.
Postscript 193
Postscript
This report demonstrates that patients with essential infantile esotropia (EIE) and
crossed fixation tend to switch fixation to the eye receiving nasalward optokinetic
input, suggesting that monocular nasotemporal optokinetic asymmetry (MNTA)
could underlie the phenomenon of crossed fixation. As MNTA derives from the
nasal retina, EIE can be conceptualized as a condition wherein the nasal retinas are
repositioned to lie within the frontal plane, thereby permitting bidirectional hori-
zontal motion detection (nasalward for each fixing eye). This explanation should not
imply that EIE is generated for the purpose of restoring horizontal motion bidirec-
tionality, but that the preexisting nasalward deviation of both eyes can be exapted
under binocular conditions to promote this critical function. This mechanism dove-
tails with the subcortical hypothesis in the following chapter to suggest that a
dynamic (optokinetic) imbalance could underlie a static deviation (essential infan-
tile esotropia).
Is Infantile Esotropia Subcortical
in Origin? 17
Introduction
Essential infantile esotropia (EIE) is characterized by a progressive nasalward devi-

ation of the eyes that develops within the first 6 months of life [10, 14, 32]. It pres-
ents in the absence of additional neurological disease and is therefore isolated to the
sensorimotor visual system within the brain [9]. It is accompanied by a unique con-
stellation of dissociated eye movements (such as dissociated vertical divergence,
latent nystagmus, and dissociated horizontal deviation), which are driven by unbal-
anced binocular visual input and incorporate prominent torsional movements of the
eyes [9]. Most cases are nonhereditary and have no defined genetic etiology [26].
In EIE, both eyes are maintained in an adducted position necessitating a head turn
to fixate with either eye (termed crossed fixation) (Fig. 17.1). However, the child can
volitionally abduct either eye fully, demonstrating that this condition is prenuclear in
origin. When the child is sedated using nondepolarizing paralyzing general anesthe-
sia, the eyes assume a fairly normal position [37]. Thus, EIE has to be caused by
increased prenuclear innervation (or tonus) to the medial rectus and inferior oblique
muscles in the awake state [9, 10]. The fact that normal neonates exhibit large con-
vergence movements, and that these excessive convergence movements predict
development of normal binocular alignment [20], suggests that EIE does not arise
from a cortical disinhibition of convergence centers within the midbrain [9].
At the outset, several indicators of a subcortical origin for EIE are evident [10].
Neuroanatomy dictates that volitional body movements are innervated by the cere-
bral cortex, while individual muscles are innervated by subcortical centers [10]. The
fact that individual extraocular muscles become hyperinnervated in EIE therefore
suggests a subcortical origin for this prenuclear innervation. A defining feature of
EIE is the persistence of subcortical visual reflexes with torsional rotations of the
eyes [10]. Dynamic torsional movements of the eyes are generated by vestibular
control centers, suggesting that EIE may also arise subcortically within prenuclear
visual pathways that directly modulate by visual motion input [10].
196 17 Is Infantile Esotropia Subcortical in Origin?
Fig. 17.1 Essential infantile

esotropia. Note tonic
nasalward deviation of
both eyes
The dissociated binocular rotations that accompany EIE are directionally driven
by unequal binocular visual input to the two eyes [7]. Each of these movements
have precise analogs in lower lateral-eyed animals, and the subcortical neuroana-
tomical pathways for all of these movements are well-established [7]. For example,
MNTA underlies the clinical phenomenon of latent nystagmus in humans [6, 22, 35]
while dissociated vertical divergence conforms to a human dorsal light reflex [7].
The expression of these atavisms suggests that this subcortical visual motion cir-
cuitry is integral to the pathogenesis of EIE [9, 10].
Evolutionary Basis of Essential Infantile Esotropia (EIE)
The fundamental evolutionary clue to the pathogenesis of EIE lies in the ubiquitous
finding of monocular nasotemporal asymmetry (MNTA) to horizontal optokinetic
stimuli. Patients with infantile esotropia show brisk nasalward, and absent tempo-
ralward optokinetic responses [6, 22, 35, 40, 42]. This MNTA persists throughout
life after surgical realignment of the eyes. Although it can be measured electro-
physiologically within the visual cortex this horizontal optokinetic asymmetry is
found in most lateral-eyed afoveate animals and used for the detection of full-
field optokinetic flow [27, 28, 36, 41]. It is modulated by the nucleus of the optic
tract and the dorsal terminal nucleus of the accessory optic system (NOT-DTN)
within the midbrain [28, 41]. In lateral-eyed animals, these centers generate visuo-
vestibular eye movements by sending visual motion signals through the vestibu-
locerebellum to the vestibular nucleus, which rotates the eyes at the appropriate
speed and direction to minimize retinal slip [1, 2, 13, 24]. These visuo-vestibular
pathways may continue to remain operational in humans with EIE.
Phylogenetically and ontogenetically, MNTA antedates development of the
visual cortex [9, 27, 36]. In lateral-eyed animals, MNTA may help to prevent tem-
poralward optokinetic stimuli from pinning the eyes back as the animal is moving
forward [28, 31, 41]. Another evolutionary advantage conferred by this reflex may
derive from the fact that the nose and mouth are positioned “nasally” to the eyes,
necessitating accurate nasalward optokinetic tracking as the animal turns toward
Evidence for a Subcortical Pathophysiology in Essential Infantile Esotropia (EIE) 197
potential food substances seen by either eye. During this turning movement, the
opposite eye is subjected to a temporalward full-field optokinetic stimulus that may
be of different velocity (when objects in its field of vision are situated at a different
distance). Having the eye with nasalward optokinetic input dictate the response for
both eyes allows the animal to accurately track the world it is turning toward and
disregard the world from which it is turning away. In normal human infants, this
subcortical full-field optokinetic imbalance is evolutionarily retained and expressed
within the first few months of life. Phylogenetically, this persistence of MNTA in
EIE represents another atavism. Ontogenetically, EIE often develops during the
time period when the subcortical motion pathways are normally turned off (i.e., 3–6
months of age). This time course is often taken as evidence that EIE must be cortical
in origin, but it could also signify that EIE represents a state of extended subcortical
neuroplasticity [7].
euroanatomy of Monocular Nasotemporal

N
Asymmetry (MNTA)
The diagram (Fig. 17.2) shows the neuroanatomy of MNTA in lateral-eyed ani-

mals [1, 2]. The subcortical motion pathways are normally operative in the early
months of human infancy [4, 5]. In EIE, investigators have studied and character-
ized the secondary aberrations within the motion pathways of the visual cortex.
However, the innervational changes that generate esotonus in early infancy prob-
ably arise from within subcortical motion pathways that remain functional in
humans that develop EIE [10]. In Fig. 17.2, the left eye is receiving nasalward
optokinetic input, which crosses through the chiasm to the right NOT and DTN
of the AOS. These structures respond to rightward optokinetic input (nasalward
for the left eye), and project their output signal to the dorsal cap of the inferior
olive, which acts as a comparator of motor commands from the cerebral cortex
and brainstem nuclei and feedback from receptors via the spinal cord, visual
system, and vestibular organs. The inferior olive relays gated information to the
cerebellar flocculus, which processes visuo-vestibular input [1, 2, 13, 24]. The
cerebellar flocculus sends its output signal to the vestibular nucleus, which inte-
grates visual motion input from the eyes with head motion input from the laby-
rinths. The vestibular nucleus provides a prenuclear signal to the ocular motor
nuclei, which innervate the appropriate extraocular muscles to minimize reti-
nal slip.
vidence for a Subcortical Pathophysiology in Essential

E
Infantile Esotropia (EIE)
How could persistence of subcortical MNTA lead to the development of EIE?

Beginning at 2–3 months of age, binocular cells from the visual cortex, which are
bidirectionally sensitive to horizontal optokinetic motion, begin to establish connec-
tions to the NOT-DTN, allowing foveal pursuit to override the monocular
EOM
NOT-DTN LGN
NPH
V1
O
M NRTP
N STS
(MT/MST)
d.c.
P 10
P
Cer Cortical
R
F
VN
Subcortical
Fig. 17.2 Schematic diagram depicting the optokinetic system in the right half of the brain during
leftward rotation of the visual world (i.e., nasalward for the left eye). Bold lines depict nasalward
subcortical optokinetic pathways mediated by the nasal retina of each eye. Dashed lines depict
temporalward optokinetic pathways mediated by temporal retina of each eye. Cer, cerebellum;
dclO, dorsal cap of the inferior olive; GS, ganglion Scarpae; LGN, lateral geniculate nucleus, NOT,
nucleus of the optic tract; NPH; nucleus prepositus hypoglossus; NRTP, nucleus reticularis teg-
menti pontis; OMN, ocular motor nuclei; PPRF, paramedian pontine reticular formation; STS,
cortical pursuit areas MT and MST; around the superotemporal sulcus; V1, primary visual cortex;
VN, comlex of the brainstem vestibular nuclei. (Reproduced with permission from Behrens et al.
[2], Springer-Verlag, 1989)
subcortical optokinetic bias. By 6 months of age, monocular horizontal responses to

optokinetic stimuli become symmetrical. According to the Hoffmann hypothesis
[17–19], when the development of cortical binocular vision is preempted, no bin-
ocular corticotectal connections are established because there are no binocular cells
in V1. And because “neurons that fire together wire together” [16], only the crossed
nasal fibers from the left eye can stream through the right visual motion cortex (MT/
MST) to hook up to the right NOT-DTN, because these are the pathways that have
the same directional sensitivity. So the visual motion cortex becomes an extension
of the subcortical optokinetic system. Thus, while the visuo-vestibular pathways are
normally subsumed by the cortical pursuit system in humans, EIE effectively causes
Potential Role of Prolonged Subcortical Neuroplasticity 199
the opposite to occur, resulting in a motion visual cortex that is reconfigured to

match the directional template of the subcortical visual pathways.
The electrophysiologic and perceptual detection of cortical motion asymmetries
within the visual cortex of patients with EIE [23, 30, 39], has unfortunately led to
the dogmatic conclusion that EIE must be caused by a primary problem within the
visual cortex [38]. However, these findings are exactly what you would see if the
cortical foveal pursuit pathways secondarily reconfigured themselves to match the
preexisting subcortical template for full-field OKNs. This ontogenetic retrofitting of
cortical horizontal motion pathways establishes a modified circuit wherein the
human subcortical motion pathways can be driven “top-down” by nasalward foveal
pursuit pathways within the visual cortex (mainly MT but also MST) [19]. When
this occurs, MNTA can be secondarily driven through the visual cortex. Consequently,
the cortical suppression of one eye that accompanies EIE can trigger subcortical
visual reflexes such as latent nystagmus and dissociated vertical divergence [7, 21].
So although you can detect evidence of MNTA within the motion pathways of the
visual cortex, this is likely the effect, rather than the cause, of the problem. The
primary cause lies within the subcortical visual motion circuitry. This transposition
of cause and effect is the major conceptual oversight in the cortical model of
EIE. Infantile esotropia does not require a primary defect in binocular vision.
Indeed, there is no evidence that one exists.
Potential Role of Prolonged Subcortical Neuroplasticity
The human subcortical full-field optokinetic system remains operational within the
first few months of human life [4, 5], (when MNTA is normally detectable in
infancy). As cortical binocular vision matures, symmetrical horizontal optokinetic
responses develop [3, 22]. Since subcortical motion pathways normally shut down
as cortical motion pathways develop, one could question whether the persistence of
subcortical reflexes could be a secondary consequence of a primary dysgenesis
within the visual cortex. The fact that the retinotectal pathways normally shut down
at approximately the same time as the binocular cortical pursuit pathways become
established certainly makes it plausible that their corticotectal connections with
NOT-DTN provide a contributory signal to inactivate them.
However, there is evidence that the subcortical pathways do not remain active in
the absence of cortical binocular vision [29]. Infants with homonymous hemianopia
due to early unilateral hemispherectomy, for example, do not continue to express
MNTA to full-field optokinetic input beyond first few months of life [29]. This
observation suggests that subcortical pathways may have a predefined period of
plasticity that causes them to shut down after several months of infant life, regard-
less of whether functioning binocular cortical pursuit pathways can establish corti-
cotectal connections with the NOT-DTN. There is something unique in children
with EIE that permits subcortical motion pathways within the human accessory
optic system to remain operational [8]. A mutation or metabolic perturbation within
the subcortical pathways could also enhance their neuroplasticity and preserve their
function throughout life, increasing baseline esotonus and derailing binocular align-
ment to necessitate a secondary remodeling of the visual cortex, which could then
serve to further perpetuate their function. Although corticofugal output can influ-
ence subcortical plasticity in the mature mammalian visual motor system [25], any
direct role of normal corticotectal connections in extinguishing the subcortical
visual motion pathways in human infancy remains to be established.
An Optokinetic Etiology for Essential Infantile Esotropia (EIE)?
If persistent MNTA represents a state of enhanced subcortical neuroplasticity, can

simultaneous subcortical optokinetic input from both nasal retinas drive the eyes
into an esotropic position? In lateral-eyed animals, the answer is yes. Under real
world conditions, a lateral-eyed animal only rarely receives simultaneous nasalward
optokinetic flow when it moves backward in space. But in the laboratory, disconju-
gate full-field optokinetic input to both eyes can generate esotropia (nasalward opto-
kinesis) or exotropia (temporalward optokinesis). This phenomenon has been
demonstrated in rabbits [15], and more recently in zebrafish (Video 1 in the online
version at https://doi.org/10.1016/bs.pbr.2019.04.001) [33]. So binocular nasalward
optokinetic input has the potential to drive the eyes into an esotropic position.
Although binocular MNTA has been speculated to cause EIE at the level of the
visual cortex [38], the nasal retina of the esodeviated eye is cortically suppressed, so
the two eyes cannot generate simultaneous cortical responses. But at the subcortical
level, there is no interocular suppression [12]. So both eyes can be pulled in simul-
taneously, just like in fish and in rabbits. Although monocular fixation with either
eye may be integral to the pathogenesis of EIE [11], a stable esotropia soon persists
during nonfixation, demonstrating how this subcortical binocular nasalward drive
can be maintained by simultaneous full-field viewing with both eyes. This mecha-
nism would require that the subcortical optokinetic system remain operational
despite cortical suppression of the esodeviating eye. Figure 17.3 juxtaposes the pro-
posed subcortical and cortical mechanisms of pathogenesis for EIE.
Conclusion
EIE conforms to a state of extended subcortical neuroplasticity in which subcortical

visual pathways fail to shut down within the third month of life, allowing an early
subcortical nasalward optokinetic bias to simultaneously drive the eyes into an
esodeviated position. When this occurs, the cortical motion pathways may second-
arily reconfigure themselves to this monocular subcortical template. Some forms of
EIE (those secondary to a motor impedance of binocular fusion) may actually cor-
respond to a genetic mutation or perturbation that extends subcortical neuroplasti-
city. If so, then EIE would simply represent a default to an older ocular motor
Conclusion 201
control system that retains its neuroplasticity and continues to be expressed.

Accordingly, as the visual cortex becomes monocular, the subcortical visual motion
system may become binocular [7, 10, 34], meaning that both nasal retinas are simul-
taneously activated at the subcortical level to drive the eyes inward to generate
EIE. Neural integration or set-point adaptation [43] may then reestablish a stable
esodeviation of the eyes. Based upon this analysis, we should consider the possibil-
ity that EIE is subcortical in origin, and that its associated cortical motion asym-
metries may be the inevitable effect, rather than the underlying cause, of early
binocular misalignment.
Accordingly, EIE can be conceptualized as a kind of evolutionary footprint,
wherein the two nasal retinas are repositioned into the frontal plane, but they continue
to do the same thing as in lateral-eyed animals. In humans, this positional reconfigura-
tion solves a practical problem by restoring horizontal optokinetic bidirectionality.
Whichever direction a target an object is moving, the child can effectively follow it
with the eye that is receiving nasalward optokinetic input. This explains why a spon-
taneous fixation switch can be elicited by directional reversal of horizontal optokinetic
motion in children with EIE [12]. As we learn more about mechanisms of subcortical
neuroplasticity, we may come to understand EIE as another inevitable expression of
our ancestral visual pathways within basement of the brain.
Pathogenesis of Infantile Esotropia
SUBCORTICAL CORTICAL
Visuovestibular System Nascent Pursuit System
(Full-Field, Afoveate) (Cortical, Foveate)
NOT-DTN MT/MST
AOS
MNTA MNTA
Cortical
MT/MST EIE/LN Disinhibition
of Convergence
Maturation of Horizontal Maturation of

Cortical Pursuit Cortical Pursuit
Cortical Inhibition
of Convergence
Horizontal Alignment
Single Binocular Vision
Fusion
Stereopsis
Fig. 17.3 Flow chart juxtaposing primary subcortical and cortical mechanisms for development
of EIE. MNTA, monocular nasotemporal asymmetry; AOS, accessory optic system; DTN, dorsal
terminal nucleus of the accessory optic system; EIE, essential infantile esotropia; LN, latent nys-
tagmus; MST, medial superior temporal area; MT, middle temporal area; NOT, nucleus of the
optic tract
References
1. Behrens F, Grusser O-J. The effect of monocular pattern deprivation and open-
loop stimulation on optokinetic nystagmus in squirrel monkeys (Saimiri sciu-
reus). In: Flohr H, editor. Post-Lesion Plasticity. Berlin: Springer; 1988.
p. 453–72.
2. Behrens F, Grusser O-J, Roggenkämper P. Open and closed-loop optokinetic
nystagmus in squirrel monkeys (Saimiri sciureus) and in man. In: Allum JHJ,
Hulliger M, editors. Progress in Brain Research, vol. 89. Amsterdam: Elsevier;
1989. p. 183–6.
3. Braddick O. Where is the naso-temporal asymmetry? Curr Biol. 1996;6:274–81.
2011;51:1588–609.
visual motion processing motion coherence and ‘dorsal-stream’ vulnerability.
Neurophsychologia. 2003;41:1769–84.
6. Brodsky MC. Latent nystagmus: vestibular nystagmus with a twist. Arch
Ophthalmol. 2004;122:202–9.
sions. Arch Ophthalmol. 2005;123:837–42.
infantile strabismus. Arch Ophthalmol. 2012a;130:1055–9.
Ophthalmol. 2012b;130:1199–202.
10. Brodsky MC. Essential infantile esotropia: potential pathogenetic role of

extended subcortical neuroplasticity. Invest Ophthalmol Vis Sci. 2018;59:1964–8.
11. Brodsky MC, Fray KF. Does infantile esotropia arise from a dissociated devia-
tion? Arch Ophthalmol. 2007;125:1703–6.
12. Brodsky MC, Klaehn LD. An optokinetic clue to the pathogenesis of crossed
fixation in infantile esotropia. Ophthalmology. 2017;124:272–3.
13. Buttner U, Boyle R, Markert G, et al. Cerebellar control of eye movements. In:
Freund HJ, Buttner U, Cohen B, Noth J, editors. Progress in Brain Research.
14. Campos EC. Why do the eyes cross? A review and discussion of the nature and
origin of essential infantile esotropia, microstrabismus, accommodative esotro-
pia, and acute comitant esotropia. J AAPOS. 2008;12:326–31.
15. Collewijn H. The oculomotor system of the Rabbit and its plasticity. Berlin/
Heidelberg/New York: Springer-Verlag; 1981. p. 67–8.
16. Hebb DO. The organization of behavior. New York: Wiley and Sons; 1949.
17. Hoffman KP. Neural basis for optokinetic defects in experimental models with
strabismus. In: Kaufmann H, editor. Transactions of the 16th Meeting of the
European Strabismological Association. Giessen: Gahmig Druck Giessen;
1987. p. 35–6.
18. Hoffmann KP. Cortical vs subcortical contribution to the optokinetic reflex in
the cat. In: Lennerstrand G, Zee DS, Keller EL, editors. Basis of ocular
Conclusion 203
motility: proceedings of a Wenner-Gren Center and Smith-Kettlewell Eye

Research Foundation International Symposium. New York: Pergamon; 1982.
p. 303–11.
19. Hoffmann KP, Bremmer F, Thiele A, Distler C. Directional asymmetry of neu-
rons in cortical areas MT and MST projecting to the NOT-DTN in macaques. J
Neurophysiol. 2002;87:113–2123.
20. Horwood A. Neonatal ocular misalignments reflect vergence development but
rarely become esotropia. Br J Ophthalmol. 2003;87:1146–50.
mus. Eye. 1996;10:274–81.
22. Kommerell G, von Noorden GK. Ocular motor phenomena in infantile strabis-
mus: asymmetry in optokinetic nystagmus and pursuit, latent nystagmus, and
dissociated vertical divergence. In: Lennerstrand G, Campos EC, editors.
Strabismus and Amblyopia: Experimental Basis for Advances in Clinical
Management. London: Macmillan; 1988. p. 99–109.
23. Kommerell G, Ullrich D, Gilles U, Bach M. Asymmetry of motion VEP in infan-
tile strabismus and central vestibular nystagmus. Doc Ophthalmol. 1995;89:373–81.
24. Langer T, Fuchs AF, Chubb MC, et al. Floccular afferents in the rhesus macaque
as revealed by autoradiography and horseradish peroxidase. J Comp Neurol.
1985;235:26–37.
25. Liu B, Hubermann AD, Scanziani M. Corticofugal output from visual cortex
promotes plasticity of innate motor behavior. Nature. 2016;538:383–7.
26. Maconachie GD, Gottlob I, McLean RJ. Risk factors and genetics in common
comitant strabismus: a systematic review of the literature. JAMA Ophthalmol.
2013;131:1179–86.
27. Masseck OA, Hoffmann KP. Comparative neurobiology of the optokinetic
reflex. Ann N Y Acad Sci. 2009;1164:430–9.
optokinetic system. In: Miles FA, Wallman J, editors. Visual Motion and its Role
in the Stabilization of Gaze. Amsterdam: Elsevier Science; 1993. p. 393–403.
29. Morrone MC, Atkinson J, Cioni G, et al. Developmental changes in optokinetic
mechanisms in the absence of unilateral cortical control. NeuroReport.
1999;10:1–7.
30. Norcia AM. Abnormal motion processing and binocularity: infantile esotropia
as a model system for effects of early interruptions of binocularity. Eye.
1996;10:259–65.
31. Ohmi M, Howard IP, Everleigh B. Directional preponderance in human optoki-
32. Pediatric Eye Disease Investigator Group. The clinical spectrum of early-onset
esotropia. Experience of the Congenital Esotropia Observational Study. Am J
Ophthalmol. 2002;33:102–8.
33. Portugues R, Feierstein CE, Engert F, Orger MB. Whole brain activity maps
reveal stereotyped, distributed networks for visuomotor behavior. Neuron.
2014;81:1328–43.
34. Ramdya P, Engert F. Emergence of binocular functional properties in a mon-

ocular neural circuit. Nat Neurosci. 2008;11:1083–90.
36. Tauber ES, Atkin A. Optomotor responses to monocular stimulation. Relation
to visual system organization. Science. 1968;160:1365–7.
37. Thouvenin D, Sotiropoulos MC, Arné J, Fournié PR. Esotropias that totally
resolve under general anesthesia treated exclusively with bilateral fadenopera-
tion. Strabismus. 2008;16:131–8.
38. Tychsen L. The cause of infantile strabismus lies upstairs in the cerebral cortex,
not downstairs in the brainstem. Arch Ophthalmol. 2012;130:1060–1.
humans who had strabismus with onset in infancy. J Neurosci. 1986;6:2495–508.
2003;44:5142–50.
41. Wallmann J. Subcortical optokinetic mechanisms. In: Miles FA, Wallman J,
editors. Visual motion and its role in the stabilization of gaze. Amsterdam:
Elsevier Science; 1993. p. 321–42.
42. Wright K. Clinical optokinetic nystagmus asymmetry in treated esotropes. J
Pediatr Ophthalmol Strabismus. 1996;33:99–109.
43. Zee DS, Leigh RJ. Ocular stability and set-point adaptation. Philos Trans R Soc
Lond Ser B Biol Sci. 2017;372(1718):20160199. https://doi.org/10.1098/
rstb.2016.0199.
Postscript
Single binocular vision and stereopsis are known to reside within the visual cortex.
By extension, essential infantile esotropia (EIE) has long been considered to be
cortical in origin. This a priori assumption is seemingly corroborated by the finding
of optokinetic asymmetry within cortical motion pathways. This analysis examines
the alternative possibility that extended subcortical neuroplasticity can lead to
EIE. It invokes the argument that persistent binocular optokinetic input at the sub-
cortical level (i.e. monocular nasotemporal asymmetry) can drive the eyes inexora-
bly into an esotropic position despite the presence of alternating cortical suppression.
Accordingly, EIE can be genetic or epigenetic (i.e. arising experientially from early
binocular misalignment) in origin.
This model further invokes the mechanism proposed by Hoffmann hypothesis
wherein the cortical visual motion pathways become secondarily reconfigured to
match the subcortical template. Once this occurs, then cortical suppression of either
eye can actively drive the dissociated eye movements that characterize EIE. Whether
cortical suppression allows the subcortical motion pathways to generate these
movements (Schor C: Subcortical binocular suppression affects the development of
latent and optokinetic nystagmus. Am J Optom Physiol Opt 1983;60:481–502), or
Postscript 205
whether the remodeled cortical pursuit pathways drive the movements through the
NOT-DTN has yet to be determined. Either way, the system eventually reverts to a
subcortical blueprint that can now be funneled in through the visual cortex. And
since the dissociated eye movements conform to the original function of the subcor-
tical pathways, we can begin to understand how and why this cortical reconfigura-
tion must have happened. This mechanism is likely age dependent, with subcortical
pathways known to predominate within the first few months of human life, and the
altered cortical motion pathways predominating later. The dichotomy between cen-
tral and peripheral retina, as described by Marshall Parks (see Binocular Vision in
Duane’s Clinical Ophthalmology, Lippincott, 1991) may have had its evolutionary
origin in the subcortical (extrafoveal) versus cortical (foveal) visual systems. The
primate visual cortex recapitulates this dichotomy in its dorsal (visual motion) and
ventral (visual detail) pathways.
It is critical to distinguish cause and effect in disease pathogenesis. This analysis
raises the possibility that EIE may be subcortical in origin, with the cortical motion
pathways becoming secondarily reconfigured to match their subcortical template.
Accordingly, decreased binocularity and stereopsis that accompany EIE may be
secondary rather than primary. Nevertheless, they may give rise to cortical suppres-
sion which can perpetuate EIE and its dissociated deviations. At the very least, it
seems that aberrations within the subcortical and cortical visual motion pathways
are both integral to the pathogenesis of EIE. Accordingly, the misguided term fusion
maldevelopment syndrome should be discarded, as it connotes a chain of causation
that necessarily originates within the visual cortex, and thereby converts a termino-
logical confusion into a conceptual one.
Phoria Adaptation: The Ghost
in the Machine 18
Introduction
I esteem it a great honor to be invited to give the 50th Scobee Lecture here in my
hometown of San Francisco. Dr. Richard Scobee was revered as a consummate
teacher, clinician, and diagnostician. His teachings permeate our field to such a
degree that we don’t always recognize them in the fabric of what we do. Dr. Scobee
served as a founding member of the American Orthoptic Council and founded the
American Orthoptic Journal, which was first published in 1951, just a year before
his untimely death. His definitive textbook, The Oculorotary Muscles [2], provided
the foundation for the field of orthoptics, and created a cogent framework for what
was to become an emerging field of study. So it was nice that, while preparing this
lecture, so many of Dr. Scobee’s seminal observations seemed to resurface within a
new context. I hope that my choice of topics helps to carry the torch and honor
his memory.
A phoria refers to any deviation of the eyes from one of alignment (orthophoria)
when binocular visual input is preempted [3]. Strabismus occurs when a phoria
persists under binocular conditions, resulting in a manifest deviation (a tropia).
Strabismus surgery is predicated on the notion that alternate occlusion unmasks the
underlying phoria. By targeting surgical dosing to the measured phoria, the resting
position of binocular alignment can be reset to near zero, allowing fusion to actively
maintain binocular alignment. As pointed out by Lancaster, it is tacitly assumed that
the resting deviation when the fusion reflex is preempted (i.e. under conditions of
monocular fixation) is identical to the resting deviation when the fixation reflex is
prevented in both eyes. A recent study has shown this to be the case for humans with
normal binocular vision (Brodsky) [4].
Despite the fact that our eyes stay straight, binocular vision is not a static func-
tion [5]. Maintaining binocular vision is a dynamic force, controlled by both
momentary peripheral fusion (termed motor fusion), tonic vergence, and vergence
adaptation (termed phoria adaptation) that provides the necessary neural integra-
tion to minimize the baseline phoria and thereby stabilize binocular vision. Changes
208 18 Phoria Adaptation: The Ghost in the Machine
in vergence consist of a fast phasic response followed by a slow tonic response.

Alternate occlusion represents the elimination of fast fusional vergences, which
must then be followed by a longer decay of the slow fusional vergence response [6].
A prism bar measures step-like changes which is more reflective of the rapid phasic
response while synoptophore testing measures the slow tonic response.
What Is Phoria Adaptation?
Phoria adaptation is the slow buildup of tonic vergence innervation to establish and
sustain a new vergence set point following placement of a prism over one or both
eyes [7–10]. Phoria adaptation recalibrates extraocular muscle tonus to realign the
visual axes and thereby meet the demands of binocular alignment [11], and restore
the dynamic range (or fusional reserve) in which fast fusional vergence can function
[12]. Phoria adaptation resets binocular alignment toward orthophoria; thereby
compensating for developmental, environmental, or pathologic alterations in the
binocular mechanism [12]. Because phoria adaptation is stimulated by the effort put
forth by the fast fusional vergence system, repeated tests of fusional amplitude will
increase the limits of fusional movement [13, 14]. After removal of a prism, the rate
of disappearance of this change in tonus is inversely proportional to the length of
time this tonicity is maintained [6, 8, 13, 14].
Clinical investigation into phoria adaptation as a malleable but invisible force
encompasses the history of pediatric ophthalmology, capturing the interest of many
progenitors of our field such as Hering [15], Maddox [16], Hofmann and
Bielschowsky [1, 17], Ogle [18, 19], Reinecke [10], and Guyton [20]. Its slow time
decay was characterized by Hofmann and Bielschowsky over a century ago [1, 6, 9,
20]. Phoria adaptation provides an innervational repository to modulate vergence
tonus to stabilize horizontal, vertical, and torsional binocular alignment [21].
Phoria adaptation serves to establish a new equilibrium as the starting position
for further vergence eye movements and thus decreases the effort for either conver-
gence or divergence eye movements [22]. This resiliency and tenacity means that a
measured phoria can change transiently following active vergence effort [5, 13, 14].
Because of its lingering effects, however, phoria adaptation can alter measurements
of both vergence ranges (which reflect activity of both the fast and the slow ver-
gence system) and also of vergence facility (which examine the patient’s ability to
shift rapidly between divergence and convergence as a function of the fast fusional
vergence system) [7, 23]. For this reason, phorias are routinely measured before
vergence amplitudes are measured [6]. Although phoria adaptation is classically
elicited by introducing prisms before one eye to induce a vergence error, this com-
pensatory mechanism can insinuate itself into many forms of intermittent strabis-
mus. Because phoria adaptation has a memory, its slow dissipation serves to mask
or conceal a large phoria on clinical examination [17].
Phoria adaptation operates independently of the fast vergence system and thereby
minimizes the fixation disparity that results in the need for vergence control [9, 18].
Ogle and Prangen found that an increase in slow fusional vergence was
Measurement of Phoria Adaptation 209
accompanied by a reciprocal reduction in fast fusional vergence [19]. However,

these two complementary systems may exist on a spectrum wherein some individu-
als who use disparity as a primary driver of fusional vergence show larger phorias
and less inherent capacity for phoria adaptation, while others show strong phoria
adaptation and minimal disparity-driven vergence (written communication, Anna
Horwood, July 20, 2017).
Stimulus for Phoria Adaptation
Binocular disparity does not generate phoria adaptation. There is always binocular
disparity in a complex world (in different planes even when the eyes are aligned),
and this binocular disparity is a necessary stimulus for stereopsis. While binocular
retinal image disparity provides the stimulus for the fast fusional vergence system,
it is the motor signal from the fast fusional vergence system that serves as the error
signal for phoria adaptation [5]. The input to the slow neural integrator for phoria
adaptation is generated by the output or effort of fast fusional vergence. Because
there has to be an effort to fuse for phoria adaptation to be recruited, fixation dispar-
ity induced by a conflict in vergence and accommodation may provide a driving
stimulus for prism-induced phoria adaptation [9] (Fig. 18.1). The fast vergence sys-
tem has a neural integrator with a short halflife of 10–15 s [7, 14]. By contrast,
phoria adaptation consists of a more stable neural integrator that leaks with a much
longer time constant (varying from minutes to several days) [5, 9]. The output of the
slow neural integrator allows for a reciprocal reduction of the output of fast fusional
vergence by means of the negative feedback loop in which the system is trying to
adapt until the phoria is controlling the entire response [9]. As the slow fusional
vergence system charges up, the fast fusional vergence system decays to maintain a
constant total output [6, 7]. This feedback loop presumably reduces the fatigue and
eyestrain that would result from the prolonged use of the fast vergence system [7–9].
Measurement of Phoria Adaptation
Phoria adaptation can be estimated but not precisely quantified, due in part to the
fact that it is slower to dissipate after an extended period of fusional demands [6, 9,
13, 17]. Measurement of fusional vergence amplitudes using a prism bar placed
before one eye under binocular conditions and prolonged monocular occlusion are
two examination techniques that help to uncover the degree to which phoria adapta-
tion is operative [9]. Depending on the preexisting level of phoria adaptation, prism
bar testing can serve to induce or dissipate phoria adaptation (a critical distinction
not determinable by the examiner), whereas prolonged occlusion removes the input
of disparity vergence and presumably dissipates any preexisting phoria adaptation.
Although monocular occlusion may dissipate phoria adaptation, it remains unclear
whether prism bars and monocular occlusion operate by equivalent mechanisms
and produce similar endpoints, since monocular occlusion seems to exert other
Fast
fusional
Ac vergence Tc
Target + Motor output
+
binocular + FD
+ + Vergence
paraliax - + + response
Pc
Phoria
adaptation
Fig. 18.1 Diagram depicting feedback interaction between slow and fast fusional vergence. The
fast fusional vergence system uses binocular retinal image disparity as its input signal. Phoria
adaptation uses the motor output of the fast fusional vergence system as its error signal to reset
tonic vergence and reduce fixation disparity. The output of this slow neural integrator then allows
for reciprocal reduction of fast fusional vergence by means of a negative feedback loop to keep the
final innervational output constant
effects that go beyond disrupting fusional vergence and dissipating phoria adapta-
tion (similar to those of Marlow occlusion) [6, 24]. Monocular occlusion may also
disrupt other adaptive processes that have different time constants.
Resilience of Phoria Adaptation
In patients with intermittent strabismus, phoria adaptation seems to operate most

effectively at fixation distances that place binocular alignment within its range [25].
It can also correct different degrees of deviation in different fields of gaze [20, 21,
25, 26]. This phenomenon explains patients ready adjustment to the prismatic
effects of anisometropic spectacles in different fields of gaze [6, 27–30]. The finding
that noncomitant vertical phoria adaptation dissipates much more quickly than com-
itant phoria adaptation (31 min versus 83 min) [31], demonstrates the coexistence
of both global and local mechanisms that subserve phoria adaptation, each having
different capabilities and time courses of action [5, 26].
Time Course of Phoria Adaptation
Phoria adaptation begins to initiate quickly but builds and dissipates more slowly [1,
6, 9]. Even brief presentations of small disparities may be enough to alter phoria
adaptation [14]. Estimates for initiation have ranged from as little as 100 ms to
30 min of binocular visual experience [9, 32, 33]. These wide ranges probably
reflect some degree of immediate initiation followed by a slower buildup period [7].
According to Henson and North, phoria adaptation is substantially complete after
2–3 min of binocular visual experience [33]. An earlier study by Ogle et al. had
found that vertical prism adaptation was complete within 3–7 min and that
Protean Clinical Manifestations 211
adaptation followed a logarithmic curve [18]. A larger prism takes longer to adapt
to but does not increase adaptation time in a linear manner [7].
However, the rate at which phoria adaptation dissipates varies considerably,
ranging from 30 s in some individuals to several days in others [7]. This measured
variability may in part reflect the inherently different time constants for phoria
adaptation between individuals, as well as the inverse correlation between the
degree of phoria adaptation and the duration of the previous vergence effort [7, 13,
34]. Rosenfield et al. found that 20 min of monocular occlusion was sufficient to
estimate the latent deviation [35]. However, a longstanding phoria adaptation may
require hours to fully dissipate [5, 31, 36–38].
Protean Clinical Manifestations
Phoria adaptation infuses our binocular alignment with a kind of inertia [13], which
can corrupt or override our clinical measurements. Conversely, when phoria adapta-
tion is fully operational, the phoria that we measure with prism and alternate cover
testing can be spuriously low. Prolonged patching often increases the measured
deviation by breaking down phoria adaptation to reset binocular misalignment
toward its uncompensated value.
Phoria adaptation comes in many guises, with clinical effects that are much more
protean than generally appreciated. What follows are some examples of conditions
in which alternate cover testing of the phoria measurement fails us when the super-
imposed effects of phoria adaptation are not recognized.
1. Stillness: Over a century ago, Sherrington remarked that “The role of muscle as
an executant of movements is so striking that its office in preventing movement
and displacement is somewhat overlooked” [39]. An overlooked but essential
quality of phoria adaptation is the stillness that it confers upon our binocular
alignment. While friction accounts for the stillness of many inanimate objects,
stillness is a formidable task for biological systems. As articulated by Carpenter:
“Why do we take stillness so much for granted? … That while a muscle is main-
taining a constant length and thus doing no work, the actin-myosin bonds
between the sliding filaments are ceaselessly breaking and reforming, consum-
ing ATP and so wastefully turning precious energy into heat rather than work.
To maintain a stationary gaze demands that the brain sends continual com-
mands to the contractile fibers in the ocular muscles and poses as many prob-
lems as moving the eyes fast. Thus, stillness is not an easy task” [40]. In fact,
biological systems are never truly still, as evidenced by the tiny fixational intru-
sions which promote stability of visual perception via the Troxler phenomenon
[41]. Phoria adaptation also explains the recent observation that the brain does
not require visual input to maintain binocular alignment in the dark on a short-
term basis [4].
2. Orthophorization: Despite the baseline divergent anatomical position of the
eyes within the orbits, normal humans have little or no measurable horizontal
phoria on alternate cover testing [10]. Phoria adaptation provides the interstitial
force to maintain approximate orthophoria, which explains why “distance pho-

rias are not normally distributed but instead showed abnormally high clustering
near orthophoria prior to an extended period of monocular occlusion” [6]. An
extended period of reading prior to a phoria measurement may generate a more
convergent phoria [14].
3. “Latent” phoria: When measuring a binocular deviation with progressive
prisms, neutralization frequently occurs long before the deviation can be
reversed with a stronger prism. This finding of a “latent phoria” probably
reflects the hidden presence of phoria adaptation [1, 6, 14]. It is therefore cus-
tomary to assign the value of the full deviation to the maximum measured angle
that produces no reversal of the initial deviation. When we say “This patient has
a small intermittent exotropia that slowly builds to 35 prism diopters,” we are
tacitly acknowledging the effects of phoria adaptation.
4. Accommodative esotropia: Phoria adaptation may mask the true esodevia-
tion in children with accommodative esotropia, who are known to “eat up
prisms,” meaning that when temporary prisms are placed on top of the full
cycloplegic refraction to neutralize the esodeviation, a portion of the esode-
viation will slowly recur through the prismatic correction. Accordingly, sur-
gery based on the initial measurements may lead to undercorrection [42]. In
this setting, it is unknown whether phoria adaptation unmasks the full esode-
viation or whether it actively generates a larger esodeviation [43]. There is
further controversy as to whether the robust prism adaptation response in
accommodative esotropia reflects or requires the presence of anomalous reti-
nal correspondence [44].
Some children with accommodative esotropia have a greater esodeviation at
near, associated with a high AC/A ratio or convergence excess. Nevertheless,
strabismus surgery targeted to the larger near esodeviation does not produce
surgical overcorrection of the distance esodeviation. This finding suggests that
the measured distance deviation is artificially low, a finding which implicates
phoria adaptation [6, 12]. The recent finding that both prism adaptation and
prolonged monocular occlusion can increase the distance deviation further sup-
ports this notion [45, 46]. So phoria adaptation may explain why you can target
surgery for the near deviation and not overcorrect for distance.
Although phoria adaptation is a disparity-driven response, vergence aftereffects
can also adapt to accommodative vergence (and vice versa), demonstrating a
versatile complementary crosslinking and plasticity between the tonic vergence
and accommodative systems to match changing demands [47–49] (Fig. 18.2).
Thus, in some people (like those with accommodative esotropia), accommoda-
tion drives vergence to produce an AC/A ratio while in others (like those with
intermittent exotropia), convergence drives accommodation to produce a CA/C
ratio [12].
Phoria adaptation can sometimes result from sustaining a vergence effort for as
little as 30 s [5]. “Cellphone vision,” which follows prolonged reading of small
font at close range, causes distant objects to appear distorted (not quite blurred
but not quite double) for several minutes. This symptom may be driven by the
Protean Clinical Manifestations 213
Accommodative Phoria
esotropia adaptation
–
+ ++
Accommodation Distance
Diopters Blur ++ realignment
CA/C
Target
distance
AC/A
Angle Disparity +
+ Near
+ Convergence
++ realignment
–
Intermittent Phoria
exotropia adaptation
Fig. 18.2 Diagram depicting the general mechanism of phoria adaptation as it influences
vergence-accommodation interaction. Blur-driven accommodation implements a similar neural
integrator as does disparity-driven vergence to program tonicity. Note the cross-linking that gener-
ates accommodation-linked convergence (AC/A) and convergence-linked accommodation (CA/C).
For both convergence and for the accommodative system, the fast system provides the immediate
phasic response to a change in disparity or blur. The tonic adaptation uses the motor output of the
fast system to provide a slower adjustment in the tonic level of accommodation or vergence. In
both systems, phoria adaptation is driven by the motor signal and not by sensory input to the motor
signal. And in some people (like those with accommodative esotropia), accommodation drives
vergence to produce an AC/A ratio while in others like those with intermittent exotropia, conver-
gence drives accommodation to produce a CA/C ratio. Phoria adaptation explains why strabismus
surgery targeted for the distance deviation does not overcorrect the near deviation in intermittent
exotropia, and why strabismus surgery targeted to the near deviation does not overcorrect the dis-
tance deviation in accommodative esotropia. (Reprinted from Ref. [12], with permission)
transitory effects of phoria adaptation, which throws the vergence system into
disarray by causing an esophoria, perhaps contributed to by the secondary
effects of tonic accommodation [49].
5. Intermittent exotropia: Patients with intermittent exotropia display a greater
distance than near exodeviation on initial alternate cover testing. It was Scobee
who noted that the near exodeviation usually increases to approximate the dis-
tance exodeviation after 45 min of monocular occlusion [2]. This “tenacious
proximal fusion” [50] reflects the role of phoria adaptation, which impedes
expression of the full near exodeviation [12]. Although it is generally believed
the presence of large convergence amplitudes are necessary to control intermit-
tent exotropia, it may be that phoria adaptation serves this function [12], which
would explain why correlative fusional convergence amplitudes (i.e. equal to
the amplitude of the exodeviation plus superimposed convergence amplitudes)
are no longer detectable immediately following surgical correction of the exo-
deviation (Fig. 18.2). So phoria adaptation also explains why we can target
surgery for the distance deviation and not overcorrect the near deviation (except
in the rare patients who have a true high AC/A ratio).
Prism adaptation or prolonged monocular occlusion can also increase the mea-
sured angle in patients with intermittent exotropia and more accurately direct
surgical dosing [51–53]. However, its use as a stand-alone treatment to reduce
hemiretinal suppression and restore binocular control in intermittent exotropia
has never met with success, perhaps owing to the correlative dissipation of pho-
ria adaptation that ensues. While some patients with intermittent exotropia
immediately break down to their full deviation (indicating control by fast ver-
gence), others break down slowly with prism and alternate cover testing, sug-
gesting that phoria adaptation is being used to control the deviation. The
prognosis for relative postoperative control in these two overlapping groups is
unknown.
6. Congenital superior oblique palsy: Patients with congenital superior oblique
palsy are said to have “increased fusional vergence amplitudes,” meaning that
these patients will “eat up prisms” and show a large induced vertical deviation
when a base down prism bar is increased progressively before the paretic eye.
This stepwise dissipation of fusional control to yield a measured deviation far
beyond the normal range of vertical divergence probably reflects the slow dis-
sipation of phoria adaptation [6, 54]. As with intermittent exotropia, these
“increased vertical fusional amplitudes” disappear immediately following suc-
cessful strabismus surgery, signifying that phoria adaptation rather than periph-
eral fusion was masking the true hyperdeviation.
7. Postoperative prism adaptation: When performing strabismus surgery in
adults who have worn prisms to treat diplopia, we admonish them to discon-
tinue their prism glasses postoperatively [6, 8]. We do this not only to prevent
postoperative diplopia, but to prevent a prism-induced recurrence of the devia-
tion. Instructing patients not to wear their prisms after strabismus surgery
betrays an unconscious awareness of phoria adaptation.
8. Physiologic skew deviation: Despite the predicted effects of otolithic input to
the extraocular muscles, normal binocular individuals have no measurable
hyperphoria during head tilt to either side [55]. Phoria adaptation may provide
the “antivestibular” force that maintains orthophoria during head tilt. Despite its
slow decay, it retains a fluidity which allows it to kick in to different degrees in
different fields of gaze. This is not surprising since the vestibular system and the
binocular system each rely upon a central neural integrator with a slow decay
(velocity storage for the vestibular system and phoria adaptation for the eyes) to
produce a perseverative signal that stabilizes balance and perception. This inter-
connection may explain why phoria adaptation can modify vestibulo-ocular
adaptation [56] and vice versa [5, 57]. As pointed out to me by Cameron Parsa,
MD, and Stephen Archer, MD, many patients have tiny amounts of skew devia-
tion (on the order of 1–2 PD) yet they are unable to fuse it, and they do not
prism adapt when a vertical prism is used to eliminate the diplopia. So the cen-
tral vestibular pathways that generate skew deviation may also be integral to the
pathogenesis of phoria adaptation.
9. Fusional vergence amplitudes: As divined long ago by Ogle et al., phoria
adaptation is not a measure of fusional reserves, since the same fusional reserves
Neural Substrate 215
are measurable before and after adaptation [18]. Strong evidence for this mech-
anism for phoria adaptation is evident in patients with intermittent exotropia as
well as those with congenital superior oblique palsy. Phoria adaptation can cre-
ate the false appearance of large fusional vergence amplitudes when it masks a
latent phoria [14]. Phoria adaptation explains why the measured divergence
break and recovery points are lower when tested after convergence [7, 23,
58–60]. It also explains Scobee’s observation that fusional vergence ampli-
tudes, as measured at near, seem to correlate with the horizontal phoria mea-
surement as measured at distance [61]. Scobee found that patients with greater
amounts of exophoria at distance tend to have greater amounts of measured
fusional divergence at near, while patients with esophoria at distance tend to
have smaller amounts of prism divergence at near. Thus, the slow dissipation of
phoria adaptation is indistinguishable from the measurement of fusional diver-
gence amplitudes.
10. Spread of comitance: Although we generally attribute spread of comitance to
extraocular muscle contracture, phoria adaptation can initiate and perpetuate
spread of comitance long before extraocular muscle contracture develops [22,
25, 26, 31]. For example, Kolling et al. found that prolonged occlusion restored
incomitance in patients with superior oblique palsy and spread of comitance
[62]. Similarly, the esodeviation in sixth nerve palsy may become more comi-
tant over time, a phenomenon sometimes attributed to ipsilateral medial rectus
contracture. However, medial rectus contracture should produce a spread of
incomitance. So this paradoxical spread of comitance is more likely attributable
to phoria adaptation, which can vary in different fields of gaze. As emphasized
by Guyton [20], phoria adaptation can eventuate in muscle length adaptation
and secondary anatomical extraocular muscle contracture.
Neural Substrate
Phoria adaptation has been considered by some [10, 63], but not others [64], to be a
cerebellar response, but midbrain vergence-related neurons may also play a role
[65]. Visual adaptations to reversing prisms in the cat are preempted by lesions in
the vestibulocerebellum [66]. In monkeys, removal of the cerebellar vermis (but not
the flocculus) has been shown to disrupt phoria adaptation [67, 68].
When prisms are placed before one eye of an orthotropic patient, the visual
image discordance is interpreted as an error signal. Adaptation and learning to cor-
rect motor error signals is modulated at the level of the vestibulocerebellum [10,
69]. It, therefore, seems likely that phoria adaptation is similarly mediated by both
climbing and mossy fibers within the cerebellum, which implement modifiable,
adaptive, and “plastic” responses [10] (Fig. 18.3) [70]. The cerebellum receives
continuous information via the mossy fiber system. The climbing fiber system origi-
nates from the inferior olivary nucleus and provides a powerful timing and error
signal to Purkinje cells. The inferior olivary nucleus acts as a comparator of motor
commands from the cerebral cortex, brainstem nuclei and receives feedback from
Inhibitory
interneuron
Parallel fiber
Purkinje
cell
Granule
cell
Deep Climbing
cerebellar fiber
nucleus
Mossy
fiber
Inferior
olive
Phoria
adaptation
Tonic Fast
vergence Binocular vergence motor
innervation alignment error signal
Fig. 18.3 Diagram depicting cerebellar candidate pathways for the modulation of phoria adapta-
tion. Complex spikes from mossy fibers transmit ongoing vergence information from the two eyes
while slow simple spikes from climbing fibers input through inferior olive provide the error signal
to modulate Purkinje cell output to the deep cerebellar nuclei and generate phoria adaptation.
(Modified from Ref. [70], with permission)
receptors via the spinal cord, visual system, or vestibular organs. The inferior olive
senses the error and recalibrates the tonic firing of the Purkinje cells. The increased
frequency of inferior olivary nucleus discharge and complex spikes in the Purkinje
cells triggers long-term depression of the synapse between the parallel fibers and
the Purkinje cells, thereby resetting the single spike discharge rate to produce the
necessary motor learning and adaptation. The Purkinje cell provides profound inhi-
bition via GABA to the cerebellar nuclei, which provide the output of the cerebel-
lum. It is therefore likely that the neural circuitry subserving phoria adaptation is
not localized to a specific area but modulated by cortical, midbrain, and cerebellar
circuitry [10, 65, 66].
Conclusions 217
Future Questions
Is there a critical time period for the development of phoria adaptation? Could
deficient phoria adaptation (due to a leaky neural integrator) engender some com-
mon forms of strabismus? [6, 8, 71, 72] Can hemiretinal suppression impede pho-
ria adaptation to allow intermittent deviations to perpetuate with time? [25] Since
horizontal phoria adaptation is stronger in the convergent direction [6, 9, 73],
could phoria adaptation function as an on-off system rather than a push-pull sys-
tem? Accordingly, could accommodative esotropia reflect a state of augmented
phoria adaptation (driven by the accommodative cross-linkage instead of by ver-
gence input), and intermittent exotropia reflect a state of deficient phoria adapta-
tion? Do measured horizontal fusional divergence amplitudes simply represent
the dissipation of horizontal phoria adaptation rather than the effects of an active
divergence mechanism? Is deficient phoria adaptation associated with specific
symptomatology? [8] Conversely, could some patients grow sensorially accus-
tomed to a high level of phoria adaptation and feel subjectively uncomfortable
when strabismus surgery removes the need for it? Do prism adaptation and pro-
longed monocular occlusion ultimately uncover the effects of phoria adaptation to
the same extent? Should strabismus surgery be calibrated according to the mea-
sured phoria or to the deviation produced by the induced phoria adaptation? Does
phoria adaptation supersede dissociated esotonus following surgical realignment
of the eyes? Are the horizontal, vertical, and torsional components of phoria adap-
tation [25] anatomically segregated but mediated by the same final common path-
ways? Can a focal neurological lesion selectively disable a single component of
phoria adaptation?
Conclusions
Phoria adaptation provides the central tonus mechanism that recalibrates binocular
alignment to minimize the demands on fusional vergence, and thereby promote sen-
sorimotor fusion. It is generated by a central neural integrator, probably within the
cerebellum, that provides inertia, stability, and plasticity to human binocular vision.
Although this shadow control system is often considered synonymous with “tonic
vergence,” phoria adaptation generates a deeper level of perseverative tonicity to
maintain baseline extraocular muscle tonus so that sensorimotor fusion can be
quickly and accurately superimposed. Clinically, its activation cannot readily be
distinguished from its dissipation. Phoria adaptation is deeply entrenched before
most of our prism measurements take place, making its effects virtually invisible to
the clinician. As it so often adulterates our clinical strabismus measurements, it is
truly the ghost in the machine.
As the effects of phoria adaptation may linger even after the largest possible
phoria is measured [74], this analysis begs the question of whether a “true phoria”
can even exist [17], or whether all measured phorias inevitably incorporate some
degree of phoria adaptation. The ubiquity and robustness of this cerebellar learning
mechanism in frontal binocular animals, and its augmentation in acquired forms of

intermittent strabismus, would explain why our measured deviations so often under-
estimate the larger deviations that are obtained after prolonged occlusion or prism
adaptation. When the effects of phoria adaptation fly under our radar, the resulting
clinical measurements are often artificially low, and surgical undercorrections can
be predicted. Thus, phoria adaptation may help to explain the trend toward progres-
sively higher surgical dosing over the last century in intermittent exotropia and
accommodative esotropia. Conversely, the malleability and plasticity of this shadow
control system due to its inherent ability to “learn” may also explain why strabismus
surgery is so often successful [75].
References
1. Hofmann FB, Bielschowsky A. Uber due der Willkue entzogenen Fusions-
Bewegungen der Augen. Arch Gesamte Physiol Menschen Tiere.
1900;80:1–40. https://doi.org/10.1007/BF01661926.
2. Scobee RG. The oculorotary muscles. St Louis: C.V. Mosby; 1952. p. 1–359.
3. Lancaster WB. Terminology with extended comments on the position of rest
and on fixation. In: Allen JH, editor. Strabismus ophthalmic symposium II. St.
Louis: Mosby; 1958. p. 503–22.
4. Jung J, Klaehn L, Brodsky MC. Stability of human binocular alignment in the
dark and under conditions of nonfixation. J AAPOS. 2016;20:357–63. https://
doi.org/10.1016/j.jaapos.2016.05.011.
5. Leigh J, Zee DS. The neurology of eye movements. 4th ed. New York: Oxford
6. Cooper J. Clinical implications of vergence adaptation. Optom Vis Sci.
1992;69:300–7. https://doi.org/10.1097/00006324-199204000-00008.
7. McDaniel C. Phoria adaptation in clinical vergence testing. Optometry.
2010;81:469–75. https://doi.org/10.1016/j.optm.2010.01.012.
8. Carter DB. Fixation disparity and heterophoria following prolonged wearing of
prisms. Am J Optom. 1965;42:141–51. https://doi.org/10.1097/00006324-
196503000-00001.
9. Schor CM. The influence of rapid prism adaptation upon fixation disparity. Vis
Res. 1979;19:757–65. https://doi.org/10.1016/0042-6989(79)90151-2.
10. Milder DG, Reinecke RD. Phoria adaptation to prisms. A cerebellar dependent
response. Arch Neurol. 1983;40:339–42. https://doi.org/10.1001/
archneur.1983.04050060039005.
11. Spencer S, Firth AY. Stereoacuity is affected by induced phoria but returns
toward baseline during vergence adaptation. J AAPOS. 2007;11:465–8. https://
doi.org/10.1016/j.jaapos.2007.03.014.
12. Brodsky MC, Jung J. Intermittent exotropia and accommodative esotropia: dis-
tinct disorders or two ends of a spectrum. Ophthalmology.
2015;122:1543–6. https://doi.org/10.1016/j.ophtha.2015.03.004.
13. Ellerbrock VJ. Tonicity induced by fusional movements. Am J Optom
Arch Am Acad Optom. 1950;27:8–20. https://doi.org/10.1097/00006324-
195001000-00003.
Conclusions 219
14. Toole AJ, Fogt N. The forced vergence cover test and phoria adaptation.
Ophthalmic Physiol Opt. 2007;27:461–72. https://doi.org/10.1111/j.1475-1313
.2007.00498.x.
15. Hering E. Die Lehre von binocularen Sehen. Leipzig: Englemann; 1868.
16. Maddox EE. The clinical use of prisms: and the decenring of lenses (Enlarged
5th Edition). London: John Wright & Co; 1907.
17. Bielschowsky A. Lectures on motor anomalies. Hanover: Dartmouth College
Publications; 1940. p. 15–7.
18. Ogle KN, Martens TG, Dyer JA. Oculomotor imbalance in binocular vision and
fixation disparity. London: Henry Kimpton; 1967.
19. Ogle KN, Prangen AD. Observations on vertical divergences and hyperphorias.
Arch Ophthalmol. 1953;49:313–34. https://doi.org/10.1001/archo
pht.1953.00920020322009.
20. Guyton DL. The 10th Bielschowsky lecture. Changes in strabismus over time.
The roles of vergence tonus and muscle length adaptation. Binocul Vis
Strabismus Q. 2002;21:81–92.
21. Taylor MJ, Roberts DC, Zee DS. Effect of sustained cyclovergence on eye
alignment: rapid torsional phoria adaptation. Invest Ophthalmol Vis Sci.
2000;41:1076–83.
22. Dysli M, Abegg M. Gaze-dependent phoria and vergence adaptation. J Vis.
2016;16(2):1–12. https://doi.org/10.1167/16.3.2.
23. Fray KJ. Fusional amplitudes: exploring where fusion falters. Am Orthopt
J. 2013;63:41–54. https://doi.org/10.3368/aoj.63.1.41.
24. Graf EW, Maxwell JS, Schor CM. Changes in cyclotorsion and vertical align-
ment during prolonged monocular occlusion. Vis Res. 2002;42:1185–94. https://
doi.org/10.1016/S0042-6989(02)00047-0.
25. Schor CM, Maxwell JS, McCandless J, Graf E. Adaptive control of vergence in
humans. Ann N Y Acad Sci. 2002;956:297–305. https://doi.org/10.1111/
nyas.2002.956.issue-1.
26. Maxwell JS, Schor CM. Mechanisms of vertical phoria adaptation revealed by
time-course and two dimensional spatiotopic maps. Vis Res.
1994;34:241–51. https://doi.org/10.1016/0042-6989(94)90336-0.
27. Cooper H, Dharamsh Sethi B, Henson DB. Vergence adaptive change with a
prism induced noncomitant disparity. Am J Optom Physiol Optic. 1985;62:247–63.
28. Henson DB, Dharamshi BG. Oculomotor adaptation to induced heterophoria
and anisometropia. Invest Ophthalmol Vis Sci. 1982;22:234–40.
29. Cooper Ellerbrock VJ, Fry GA. Effects induced by anisometropic corrections.
Am J Optom Arch Am Acad Optom. 1942;19:444–59. https://doi.
org/10.1097/00006324-194211000-00002.
30. Ellerbrock VJ, Fry GA. Effects induced by anisometropic corrections. Am J
Optom Arch Am Acad Optom. 1942;19:444–59. https://doi.
org/10.1097/00006324-194211000-00002.
31. Graf WE, Maxwell JS, Schor CM. Comparison of the time courses of concomi-
tant and nonconcomitant vertical phoria adaptation. Vis Res.
2003;43:567–76. https://doi.org/10.1016/S0042-6989(02)00597-7.
32. Henson DB, North R. Adaptation to prism-induced heterophoria. Am J Optom

PhysiolOptic.1980;57:129. https://doi.org/10.1097/00006324-198003000-00001.
33. Larson W. An investigation of prism adaptation latency. Optom Vis Sci.

1994;71:38–42. https://doi.org/10.1097/00006324-199401000-00008.
34. Fogt N, Toole A. The effect of saccades and brief fusional stimuli on phoria
adaptation. Optom Vis Sci. 2001;78:815–24. https://doi.
org/10.1097/00006324-200111000-00011.
35. Rosenfeld M, Chun TW, Fischer SE. Effect of prolonged dissociation on the
subjective measurement of near heterophoria. Ophthalmic Physiol Opt.
1997;17:478–82. https://doi.org/10.1111/j.1475-1313.1997.tb00086.x.
36. Howard IP, Allison RS, Zacher JE. The dynamics of vertical vergence. Exp
Brain Res. 1997;116:153–9. https://doi.org/10.1007/PL00005735.
37. Hwang JM, Guyton DL. The Lancaster red-green test before and after occlu-
sion in the evaluation of incomitant strabismus. J AAPOS. 1999;3:151–6. https://
doi.org/10.1016/S1091-8531(99)70060-1.
38. Niekter B. Horizontal and vertical deviations after prism neutralization and
diagnostic occlusiogn in intermittent exotropia. Strabismus. 1994;2:13–22.
https://doi.org/10.3109/09273979409105049.
39.
Sherrington CS. Postural activity of muscle and nerve. Brain.
1915;38:191–234. https://doi.org/10.1093/brain/38.3.191.
40. Carpenter RHS. What Sherrington missed: the ubiquity of the neural integrator.
Ann N Y Acad Sci. 2011;1233:208–13. https://doi.org/10.1111/j.1749-6632
.2011.06110.x.
41. Moses RA, editor. Adler’s physiology of the eye. 7th ed. St. Louis: CV Mosby;
1981. p. 678–9.
42. Repka MX, Connett JE, Scott WE. One year surgical outcome after prism adap-
tation for the management of acquired esotropia. Ophthalmology.
1996;103:922–8. https://doi.org/10.1016/S0161-6420(96)30586-1.
43. Kushner BJ. Strabismus: practical pearls you won’t find in textbooks, vol. 68.
New York: Springer; 2017.
ment of strabismus. St. Louis: CV Mosby; 1996. p. 507.
45. Garretty T. The effect of prism adaptation on the angle of deviation in conver-
gence excess esotropia and possible consequences for surgical planning.
Strabismus. 2018;26:111–7. https://doi.org/10.1080/09273972.2018.1481435.
46. Garretty TL. Convergence excess esotropia: a proposed new classification and
the effects of monocular occlusion on the AC/A ratio. J Pediatr Ophthalmol
Strabismus. 2010;47:308–12. https://doi.org/10.3928/01913913-20100118-03.
47. Schor CM, Kotular JC. Dynamic interactions between accommodation and
convergence are velocity sensitive. Vis Res. 1986;26:927–42. https://doi.
org/10.1016/0042-6989(86)90151-3.
48. Schor C. Influence of accommodative and vergence adaptation on binocular
motor disorders. Am J Optom Physiol Opt. 1988;65:464–7. https://doi.
org/10.1097/00006324-198806000-00006.
Conclusions 221
49. Gowrisankaran S, Satgunam P, Fogt N. Phoria adaptation can be induced by

accommodative convergence. Invest Ophthalmol Vis Sci. 2007;48:907.
50. Kushner BJ. Tenacious proximal fusion: the Scobee phenomenon. Am Orthopt
J. 2015;65:73–80. https://doi.org/10.3368/aoj.65.1.73.
51. Othsuki H, Hasebe S, Kono R, et al. Prism adaptation response is useful for pre-
dicting surgical outcome in selected types of intermittent exotropia. Am J
Ophthalmol. 2001;131:117–22. https://doi.org/10.1016/S0002-9394(00)00704-2.
52. Dadeya S, Kamlesh NS. Usefulness of the prism adaptation test in patients with
intermittent exotropia. J Pediatr Ophthalmol Strabismus. 2003;40:85–9. https://
doi.org/10.3928/0191-3913-20030301-07.
53. Zahavi A, Friling R, Ron Y, et al. Evaluation of ocular motility deviation
changes in exotropic patients after cycloplegic eye drops versus prism adapta-
tion test. Eur J Ophthalmol. 2019;29(5):482–5. https://doi.
org/10.1177/1120672118803518.
54. Ohtsuka H, Hasabe S, Furuse T, et al. Contribution of vergence adaptation to
difference in vertical deviation between distance and near viewing in patients
with superior oblique palsy. Am J Ophthalmol. 2002;134:252–60. https://doi.
org/10.1016/S0002-9394(02)01519-2.
55. Brodsky MC, Haslwanter T, Kori AA, Straumann D. The role of voluntary
effort in the Bielschowsky head tilt test: a clinical and oculographic assessment.
Binocul Vis Strabismus Q. 2000;15:325–30.
56. Lewis RF, Clendaniel RA, Zee DS. Vergence-dependent adaptation of the
vestibulo- ocular reflex. Exp Brain Res. 2003;152:335–40. https://doi.
org/10.1007/s00221-003-1563-9.
57. Sato F, Akao T, Kurkin S, et al. Adaptive changes in vergence eye movements
induced by vergence-vestibular interaction training in monkeys. Exp Brain Res.
2004;156:164–73. https://doi.org/10.1007/s00221-003-1777-x.
58. Alpern M. The zone of clear vision at the upper levels of accommodation and
convergence. Am J Optom Arch Am Acad Optom. 1950;27:491–513. https://
doi.org/10.1097/00006324-195010000-00003.
59. Rowe FJ. Fusional vergence measures and their significance in clinical assess-
ment. Strabismus. 2010;18:9–17. https://doi.org/10.3109/09273971003758412.
60. Lanca CC, Rowe FJ. Variability of fusion vergence measurements in heteropho-
ria. Strabismus. 2016;24:63–9. https://doi.org/10.3109/09273972.2016.1159234.
61. Scobee RG, Green EL. Relationships between lateral heterophoria, prism ver-
gence, and the near point of convergence. Am J Ophthalmol.
1948;31:427–55. https://doi.org/10.1016/0002-9394(48)92164-3.
62. Kolling GH, Steffen H, Baader A, et al. Diagnostic occlusion test in cases of
unilateral strabismus sursoadductorus. Strabismus. 2004;12:41–50. https://doi.
org/10.1076/stra.12.1.41.29016.
63. Kono R, Hasebe S, Ohtsuki H, et al. Impaired vertical phoria adaptation in
patients with cerebellar dysfunction. Invest Ophthalmol Vis Sci. 2002;43:673–8.
64. Hain TC, Luebke AE. Phoria adaptation in patients with cerebellar dysfunction.
65. Morley JW, Judge SJ, Lindsey JW. Role of monkey midbrain near-response
neurons in phoria adaptation. J Neurophysiol. 1992;67:1475–92. https://doi.
org/10.1152/jn.1992.67.6.1475.
66. Robinson DA. Adaptive gain control of the vestibulo-ocular reflex by the cerebel-
lum. J Neurophysiol. 1976;39:954–69. https://doi.org/10.1152/jn.1976.39.5.954.
67. Takagi M, Tamargo R, Zee DS. Effects of lesions of the cerebellar oculomotor
vermis on eye movements in primate: binocular control. Prog Brain Res.
2003;142:19–33.
68. Judge SJ. Optically-induced changes in tonic vergence and AC/A ratio in nor-
mal monkeys and monkeys with lesions of the flocculus and the ventral para-
flocculus. Exp Brain Res. 1987;66:1–9. https://doi.org/10.1007/BF00236195.
69. Ito M, Shiida T, Yagi N, et al. There cerebellar modification of the rabbit’s hori-
zontal vestibulo-ocular reflex induced by sustained head rotation combined
with visual stimulation. Proc Jpn Acad. 1974;50:85–9. https://doi.org/10.2183/
pjab1945.50.85.
70. Kandel R, Schwartz JH, Jessell TM, Siegelbaum SA, Hudspeth AJ. Principles
of neural science. 5th ed. New York: McGraw Hill; 2013. p. 967–8.
71. Prezekoracka-Krawczyk A, Michalak KP, Pyzalska P. Deficient vergence prism
adaptation in subjects with decompensated heterophoria. PLoS
ONE. 2019;14(1):e0211039. https://doi.org/10.1371/journal.pone.0211039.
72. Przekoracka-Krawczyk A, Michalak KP, Pyzalska P, Lappe M. Deficient ver-
gence prism adaptation in subjects with decompensated heterophoria. PLoS
One. 2019;14(1):e0211039. https://doi.org/10.1371/journal.pone.0211039.
73. Alpern M. The zone of clear single vision at the upper levels of accommodation
and convergence. Am J Optom Arch Am Acad Optom. 1950;27:8–20. https://
doi.org/10.1097/00006324-195010000-00003.
74. North RV, Begumpara S, Henson DB. Effects of prolonged forced vergence
upon the adaptation system. Ophthalmic Physiol Opt. 1986;6:391–6. https://
doi.org/10.1111/j.1475-1313.1986.tb01158.x.
75. Archer S. Why strabismus surgery works: the legend of the dose-response
curve. J AAPOS. 2018;22(1):1.e1–1.e6. https://doi.org/10.1016/j.
jaapos.2017.12.001.
Postscript
In this Richard G. Scobee lecture, I argue that phoria adaptation is the major source
of binocular tonus. I extrapolate this notion to explain the tenacity of normal bin-
ocular vision. I grapple with the notion that much of what we designate as fusional
vergence amplitudes represents phoria adaptation. Finally, I propose that phoria
adaptation is a normal cerebellar learning process that is driven by a neural integra-
tor that provides plasticity to real world binocular function. Phoria adaptation pro-
vides a unifying explanation for myriad clinical effects that exert themselves in
normal binocular vision and in different disease states.
Postscript 223
As the binocular tonus mechanism for phoria adaptation differs from that which
superintends dissociated eye movements in infantile strabismus, one can conclude
that human binocular vision can be superintended by two distinct tonus pools, each
having a strong contribution from the neural integrators within the cerebellum, with
the operative coordinate system dictated by genetic, neurologic, epigenetic, and
environmental factors as yet undetermined. It is clear, however, that newer this bin-
ocular tonus system can still kick in to maintain binocular alignment after surgical
realignment of the eyes in children with essential infantile esotropia. Perhaps, a
secondary maldevelopment of phoria adaptation could explain the delayed onset of
consecutive exotropia in some surgically-treated patients.
Monocular Nasotemporal Optokinetic
Asymmetry—Unraveling the Mystery 19
Pediatric ophthalmologists encounter a unique visuomotor phenomenon known as

monocular nasotemporal optokinetic asymmetry (MNTA) [1, 2]. This condition is
characterized by brisk nasalward and poor temporalward optokinetic responses
under monocular conditions of viewing (Fig. 19.1A–B). Babies with normal vision
display this subcortical optokinetic bias within the first few months of life, until
cortical binocular vision is firmly established [1–3]. When infantile strabismus pre-
cludes normal binocular development, however, this optokinetic asymmetry persists
throughout life [4–7]. In this way, MNTA provides a diagnostic “footprint in the
snow” of infantile strabismus. It also gives rise to latent nystagmus and underlies its
unique association with infantile strabismus [8–11].
What is the origin of MNTA, and how can we explain its unique association with
infantile strabismus? MNTA is an ancestral visual reflex that is present in most
lateral-eyed afoveate animals [12, 13]. The afoveate retina in lateral-eyed animals,
such as fish and rabbits, corresponds to the human nasal retina, because it views the
ipsilateral (temporal) visual field. Its retinotectal pathways project contralaterally to
the nucleus of the optic tract and the dorsal terminal nucleus of the accessory optic
tract (NOT-DTN) of the midbrain [14–17], which are sensitive to full-field motion
and facilitate optokinetic stabilization of retinal slip during body movements. The
subcortical circuitry for optokinetic asymmetry is shared by all vertebrates [16].
During navigation, survival depends on distinguishing real-world motion from
body motion [15–17]. To this end, the optokinetic system encodes full-field visual
motion separately from labyrinthine signaling of head motion. MNTA enables the
animal to see where it is going during both translational and rotational body move-
ments [15–17]. Consider a fish with two eyes that each see the ipsilateral side of
visual space. When it swims forward, it would be visually disruptive to have both
eyes driven backward by optokinetic responses to radial optic flow as it passes sta-
tionary contours that move posteriorly relative to the body (Fig. 19.1C). Because
fish are buoyant, they may also need to use posterior optic flow to detect self-motion
226 19 Monocular Nasotemporal Optokinetic Asymmetry—Unraveling the Mystery
a b
c d
Fig. 19.1 Monocular nasotemporal optokinetic asymmetry. (A–B) In the human with infantile
strabismus, monocular viewing results in strong nasalward optokinetic responses (solid lines) and
poor temporalward optokinetic responses (dashed lines) relative the viewing eye. (C) In a fish that
is swimming forward, temporalward optokinetic responses (dashed lines) in response optic flow
would rotate the eyes posteriorly and impede visualization of the oncoming visual scene. (D) In a
fish that is turning to the right, nasalward rotation of nearby contours in the right hemifield of
visual space necessitates a strong and accurate optokinetic response in the right eye (solid curved
line). As the fish turns away from the left hemifield of visual space, more distant visual contours
seen by the left eye (dashed curved line) are moving at slower velocities; therefore, the nasalward
visual input to the right eye dictates the optokinetic responses of both eyes
[18]. It is therefore evolutionarily advantages to minimize temporalward optoki-

netic responses to a receding visual world [15–18]. As aquatic visibility can be low,
fish swim closer to features that provide optokinetic input, enabling them to use
temporal optic flow to judge self-motion [18].
Monocular Nasotemporal Optokinetic Asymmetry—Unraveling the Mystery 227
But this is only half the story. When the animal turns to one side, it needs to
track the visual world that is moving nasally, because that is the area it is turning
toward (Fig. 19.1D). Because the nose and mouth are median (ie, nasal) to the
eyes, accurate tracking of nasalward circumferential optic flow may help to stabi-
lize the rapid visual motion of the approaching hemifield, as the animal rotates
toward a salient nearby nutrient [17] to inspect it with the nose and ingest it with
the mouth. At the same time, the opposite eye is seeing an area from which the
animal is turning away, with objects likely at a further distance and therefore pro-
ducing less retinal slip, so that the eye receiving nasalward visual input must dic-
tate the optokinetic response for both eyes. So MNTA provides an ingenious
evolutionary solution to enable the lateral-eyed animal to use optokinetic rotation
to visually navigate both translation and rotation. In the zebrafish pretectum, mon-
ocularly and binocularly driven clusters of directionally selective horizontal motion
detection cells are hierarchically organized to distinguish between translational
and rotational optic flow [19].
Ontegeny recapitulates phylogeny for primitive reflexes within the developing
human visual system. In humans with infantile strabismus, the cortical optokinetic
pathways reconfigure to the monocular subcortical template, so that MNTA can be
modulated through the cortical pursuit centers as they establish ipsilateral connec-
tions to each NOT-DTN [20–23]. The resulting cortical pursuit asymmetry can now
generate MNTA to foveated optokinetic targets [24–26]. In this setting, infantile
esotropia confers the binocular advantage of restoring horizontal optokinetic bidi-
rectionality in the presence of MNTA [24]. Horizontal visual motion can be fol-
lowed in either direction simply by switching fixation to the eye receiving nasalward
input [27]. In infantile esotropia, the human nasal retinas realign in the frontal plane
but continue to function as they originally did in lateral-eyed afoveate animals.
What an evolutionary wonder!
References
1. Atkinson J. Development of optokinetic nystagmus in the human infant and
monkey infant: an analogue to development in kittens. In: Freeman R, editor.
Developmental neurobiology of vision. New York: Plenum Press Nato
Advanced Study Institute Series; 1979. p. 277–87.
2. Naegele J, Held R. The post-natal development of monocular optokinetic nys-
tagmus in infants. Vis Res. 1982;22:341–6.
3. Flynn JT. Vestibulo-optokinetic interactions in strabismus. Am Orthopt
J. 1982;32:36–47.
4. Mein J. The asymmetric optokinetic response. Br Orthopt. 1983;40:1–3.
5. Wright KW. Clinical optokinetic nystagmus asymmetry in treated esotropes. J
Pediatr Ophthalmol Strabismus. 1996;33:99–109.
2003;44:5142–50.
7. Schor C, Fusaro R, Wilson N, McKee SP. Prediction of early-onset esotropia
from components of the infantile squint syndrome. Invest Ophthalmol Vis Sci.
1997;38:719–40.
228 19 Monocular Nasotemporal Optokinetic Asymmetry—Unraveling the Mystery

ment. London, UK: Macmillan; 1988. p. 99–109.
9. Kommerell G, Mehdorn E. Is an optokinetic defect the cause of congenital and
latent nystagmus? In: Lennerstrand G, Zee DS, Keller EL, editors. Functional
basis of ocular motility disorders. New York: Pergamon Press; 1982. p. 159–67.
11. Brodsky MC. Latent nystagmus: Vestibular nystagmus with a twist. Arch

Ophthalmol. 2004;122:202–9.
12. Tauber ES, Atkin A. Optomotor responses to monocular stimulation. Relation
to visual system organization. Science. 1968;160:1365–7.
13. Masseck OA, Hoffmann KP. Comparative neurobiology of the optokinetic
reflex. Ann N Y Acad Sci. 2009;1164:430–9.
14. Collewijn H, Noorduin H. Conjugate and disjunctive optokinetic movements in
the rabbit, evoked by rotatory and translator motion. Pflugers Arch.
1972;335:173–85.
15. Ohmi M, Howard IP, Everleigh B. Directional predominance in human optoki-
optokinetic system. In: Miles FA, Wallman J, editors. Visual motion and its role
in the stabilization of gaze. Amsterdam: Elsevier Science; 1993. p. 393–403.
18. Scholtyssek C, Dacke M, Kröger R, Baird E. Control of self-motion in dynamic
fluids: fish do it differently from bees. Biol Lett. 2014;10:20140279.
19. Kubo F, Hablitzel B, Dal Maschio M, et al. Functional architecture of an optic
flow responsive area that drives horizontal eye movements and zebrafish.
Neuron. 2014;81:1344–59.
20. Hoffman KP. Control of the optokinetic reflex by the nucleus of the optic tract
in primates. Prog Brain Res. 1989;80:173–82.
21. Hoffmann K-P. Neural basis for optokinetic defects in experimental animals
with strabismus. In: Kaufmann H, editor. Transactions of the 16th Meeting of
the European Strabismological Association. Giessen, September 1987. Giessen:
Gahmig Druck Giessen; 1987. p. 35–6.
22. Distler C, Hoffmann K-P. Development of the optokinetic response in macaques:
A comparison with cat and man. Ann N Y Acad Sci. 2003;1004:10–8.
23. Braddick O. Where is the naso-temporal asymmetry? Curr Biol. 1996;6:250–3.
24. Brodsky MC. Is infantile esotropia subcortical in origin? Prog Brain Res.
2019;248:183–93.
reflex is normal in infantile strabismus. Arch Ophthalmol. 1985;103:536–9.
Postscript 229
26. Joshi AC, Aqaoglu MN, Das V. Comparison of naso-temporal asymmetry dur-
ing monocular smooth pursuit, optokinetic nystagmus, and ocular following
response in strabismic monkeys. Strabismus. 2017;25:47–55.
27. Brodsky MC, Klaehn LD. An optokinetic clue to the pathogenesis of crossed
fixation in infantile esotropia. Ophthalmology. 2017;124:272–3.
Postscript
This article examines the reason why monocular nasotemporal asymmetry (MNTA)
is integral in lateral-eyed afoveate animals, and shows how subcortical visual
reflexes cause the visual cortex to secondarily reconfigure when the substrate for
binocular vision fails to develop within a critical time window. Many pediatric oph-
thalmologists do not routinely check for MNTA, which is surprising, as I have found
it to be a critical determinant of timing of strabismus that often dictates surgical
dosing. MNTA readily distinguishes essential infantile esotropia from accommoda-
tive esotropia, which requires higher surgical dosing to restore binocular alignment.
As clinical overlap can exist (e.g. essential infantile esotropia with an accommoda-
tive component), and a parent’s estimate of onset can be misleading, the absence of
MNTA suggests that you may need to augment the standard recession. However, it
is important to use an optokinetic drum with colored animals rather than stripes to
engage a young child’s attention.
Although the argument that ontology recapitulates phylogeny, first popularized
by Haeckel in 1866, has been refuted as a general neurodevelopmental principle, it
is indisputable that this concept applies to the development of human visual motor
reflexes. The persistence of MNTA in humans with infantile strabismus demon-
strates this principle. The fact that this optokinetic reflex is phylogenetically ancient
(present long before the development of the visual cortex), further shows that its
cortical representation is secondary to a reconfiguration of the visual motion cortex
to the subcortical visual template in binocular humans with early onset strabismus
(actually, fish are also “binocular,” as both eyes see at the same time). Once embed-
ded in patients with strabismus, MNTA persists indefinitely throughout life, making
it a valuable neurodiagnostic sign. MNTA is easy to detect clinically, and it anchors
the examiner to the evolutionary nature of infantile strabismus. So give it a go.
Infantile Nystagmus—Following
the Trail of Evidence 20
As a fastidious ophthalmology resident 35 years ago, I remember thinking that there

would someday be an easy way to understand and explain the constant horizontal
oscillations of the eyes that characterize infantile nystagmus. Although its terminol-
ogy has since evolved from congenital nystagmus [1] to infantile nystagmus (as the
oscillation is usually not present at birth) [2], this terminology is itself somewhat
shaky, because the genetic or anatomical conditions that lead to the oscillation are
clearly present at birth.
At that time, it was becoming increasingly apparent that a surprising number of
cases had underlying anatomical abnormalities affecting retinal or optic nerve func-
tion to reduce central visual acuity from birth [3]. Estimates of the prevalence of
patients with underlying sensory visual anomalies approximated 80%-90%, as close
clinical examination [3] and hemispheric visual evoked potentials [4, 5] disclosed
cases with subtle albinism, and electroretinography demonstrated other cases aris-
ing from congenital retinal dystrophies such as achromatopsia and congenital sta-
tionary night blindness [6, 7]. The recent advent of optical coherence tomography
(OCT) [8] and targeted next-generation genetic sequencing [9, 10] as core constitu-
ents of the infantile nystagmus evaluation may uncover sensory visual abnormali-
ties in a higher proportion of cases. Nevertheless, many hereditary forms of infantile
nystagmus have no detectable sensory visual defect [6]. The x-linked forms have
been isolated to the FRMD7 gene [11], whereas the autosomal dominant forms have
been isolated to a handful of other genes [12].
Fast forward 30 years, when I became interested in the role of subcortical optokine-
sis in human vision. Lateraleyed animals use a subcortical optokinetic system mediated
by the nucleus of the optic tract and the dorsal terminal nucleus of the accessory optic
system (AOS) in the midbrain to track full-field visual motion by generating horizontal
optokinetic responses to minimize retinal slip [13]. Ontogeny recapitulates phylogeny
in the developing visual system, as this phylogenetically older subcortical visual sys-
tem remains active within the first 2–3 months of human life [14, 15]. Foveal pursuit
pathways from the binocular visual cortex then mature and establish connections with
232 20 Infantile Nystagmus—Following the Trail of Evidence
the nucleus of the optic tract and the dorsal terminal nucleus of the AOS as the subcorti-
cal visual system permanently ceases to function [15].
In 2014, Lou Dell’Osso and I proposed that this critical period for subcortical
neuroplasticity may be necessary to prevent conflicting optokinetic signals from
arising, since following a foveated object to the right (cortical pursuit) causes the
entire visual world to sweep across the retina to the left, creating a full-field optoki-
netic stimulus that would drive the eyes away from the foveated target [16]. Delayed
cortical visual development, as has been documented in infantile nystagmus [17],
could allow the subcortical optokinetic system to remain operational. If the subcor-
tical AOS failed to become extinguished, both optokinetic systems would trigger
each other in a positive feedback loop. This reverberating circuit would generate an
“optokinetic tug-of-war,” causing the eyes to oscillate horizontally, producing what
we have come to recognize as infantile nystagmus. This mechanism seemed to
explain a number of clinical findings, most notably: (1) infantile nystagmus is iso-
lated to the visual system; (2) infantile nystagmus begins at 2–3 months of age,
when the AOS normally turns off; (3) humans with infantile nystagmus usually have
“reversed” optokinetic responses, consistent with the notion that the still-functional
subcortical optokinetic system overrides foveal pursuit to generate the optokinetic
response; and (4) humans with cortical visual loss don’t have infantile nystagmus,
again showing that the cortical and subcortical visual motion systems both have to
remain operational to produce this palindromic effect on eye position [16].
More recently, Yonehara and colleagues [18] studied the mouse model idiopathic
infantile nystagmus caused by the FRMD7 gene and found that the gene affected a
starburst amacrine cell within the retina, providing asymmetric inhibition to
direction-selective retinal ganglion cells that project to the nucleus of the optic tract
and the dorsal terminal nucleus of the AOS and subserve horizontal optokinesis.
This experimental study was confounded by the finding that FRMD7 mice do not
display infantile nystagmus despite the fact that they have complete loss of horizon-
tal optokinetic responses. Nevertheless, it dovetailed nicely with our proposed
mechanism that a primary retinal abnormality involving the optokinetic system
could lead to infantile nystagmus. More importantly, it suggested, that, if you drill
down deep enough to hone in on the starburst amacrine cell, what we call “idio-
pathic” infantile nystagmus could actually arise from defective electrochemical
transmission at the cellular level, qualifying it nosologically as a sensory disorder
associated with infantile nystagmus.
In a landmark study published last year, Winkelman et al [19] studied 3 patients
with infantile nystagmus associated with an X-linked form of congenital stationary
night blindness due to NYX or CACNA1F mutations, along with a mouse model
that had the same NYX gene mutation and also lacked the functional nyctalopin
protein on bipolar cells at their synapse with the photoreceptors. Both the patients
and the mice showed horizontally oscillating eye movements at a frequency of
4–7 Hz. Impressive was the fact that, in the mouse model, the ON direction-selec-
tive ganglion cells, which detect global motion and project to the AOS, showed
electrical oscillations at the same frequency as the ocular oscillations. The indi-
vidual retinal ganglion cells oscillated asynchronously in the dark, but the oscilla-
tions became synchronous when the retina was stimulated with light and contrast,
Infantile Nystagmus—Following the Trail of Evidence 233
together sending a strong oscillating global motion signal to the AOS. Intravitreal

injection of pharmacologic agents to modify the frequency of ganglion cell oscil-
lations produced correlative changes in the nystagmus frequency. This study
showed that, at least in congenital stationary night blindness, the ocular oscilla-
tions can be driven directly by electrical oscillations at the level of the retinal
ganglion cells. Given that ON and OFF retinal ganglion cells were firing out of
phase by 180 degrees, they considered the likely presynaptic source to be the AII
amacrine cells, which drive both cells with an opposite sign to produce an oscilla-
tory firing in antiphase. These amacrine cells contain an intrinsic, membrane-
potential oscillator, consisting of a fast sodium channel and both a fast and slow
potassium channel. Altered input from the ON-bipolar cells places the amacrine
membrane potential outside of its normal working range, which causes the AII
amacrine cells to oscillate and drives the ON and OFF retinal ganglion cells in
antiphase to produce pendular ocular oscillations.
These exciting new findings raise the possibility that electrochemical retinal dys-
function at the cellular level can directly modulate motion input to the subcortical
optokinetic system to eventuate in infantile nystagmus. It therefore seems likely that
similar electrochemical cellular aberrations will be found in other genetic forms of
isolated infantile nystagmus to explain the similar timing of onset and nystagmus
waveforms in all patient groups. The Austrian psychotherapist Alfred Adler (origi-
nally an ophthalmologist) famously stated that “the only normal people are the ones
you don’t know very well.” [20] As we zoom down on the cellular and electro-
chemical aberrations within the retina, we may come to find that the same holds true
for individuals with infantile nystagmus. If so, then any meaningful distinction
between isolated and sensory infantile nystagmus may soon disappear.
References
1. Cogan DG. Congenital nystagmus. Can J Ophthalmol. 1967;2:4–10.
2. Hertle RW, Maldanado VK, Maybodi M, Yang D. Clinical and ocular motor
analysis of infantile nystagmus syndrome in the first 6 months of life. Br J
Ophthalmol. 2002;86:670–5.
3. Simon JW. Albinotic characteristics in congenital nystagmus. Am J Ophthalmol.
1984;97:320–37.
4. Creel D, Witkop CJ, King RA. Asymmetrical visual evoked potentials in human
albinos: Evidence for visual system anomalies. Invest Ophthalmol Vis Sci.
1974;13:430.
5. Apkarian P, Shallo-Hoffmann J. VEP projections in congenital nystagmus;
VEP asymmetry in albinism: a comparison study. Invest Ophthalmol Vis Sci.
1991;32:2653–61.
6. Gelbart SS, Hoyt CS. Congenital nystagmus: a clinical perspective in infancy.
Graefes Arch Clin Exp Ophthalmol. 1988;226:178–80.
7. Weiss AH, Biersdorf WR. Visual sensory disorders in congenital nystagmus.
8. Lee H, Sheth V, Bibi M, et al. Potential of handheld optical coherence tomogra-
phy to determine cause of infantile nystagmus by using foveal morphology.
234 20 Infantile Nystagmus—Following the Trail of Evidence
9. Thomas MG, Maconachie G, Sheth V, McLean RJ, Gottlob I. Development and

clinical utility of a novel diagnostic nystagmus gene panel using targeted next-
generation sequencing. Eur J Hum Genet. 2017;25:725–34.
10. O’Gorman L, Normal CS, Michels L, et al. A small gene sequencing panel
realises a high diagnostic rate in patients with congenital nystagmus following
basic phenotyping. Sci Rep. 2019;9:13229.
11. Thomas S, Proudlock FA, Sarvananthan N, et al. Phenotypical characteristics of
idiopathic infantile nystagmus with and without mutaitons in FRMD7. Brain.
2008;131:1259–67.
12. Gottlob I, Proudlock F. Aetiology of infantile nystagmus. Curr Opin Neurol.
2014;27:83–91.
2011;51:1588–609.
15. Hoffmann KP. Development of the optokinetic response in macaques: a com-
parison with cat and man. Ann N Y Acad Sci. 2003;1004:10–8.
16. Brodsky MC, Dell’Osso LF. Neurologic mechanisms of infantile nystagmus.
JAMA Ophthalmol. 2014;132:761–8.
17. Weiss AH, Kelly JP. Acuity development in infantile nystagmus. Invest

18. Yonehara K, Fiscella M, Drinnenberg A, et al. Congenital nystagmus

gene FRMD7 is necessary for establishing a neuronal circuit asymmetry for
direction selectivity. Neuron. 2016;89:177–93.
19. Winkelman BJH, Howlett MHC, Hölzel MB, et al. Nystagmus in patients with
congenital stationary night blindness (CSNB) originates from synchronously
firing retinal ganglion cells. PLoS Biol. 2019;e3000174:17.
20. Patel A. Person of the month: Alfred Adler (1870-1937). Int J Indian Psych.
2016;3:1.
Postscript
Many subclinical retinal disorders still fly beneath the radar of our clinical detection
and are therefore diagnosed as “idiopathic” infantile nystagmus. It is not my inten-
tion to state categorically that underlying retinal dysfunction is strictly necessary for
infantile nystagmus to develop. It is certainly possible that some cases of infantile
nystagmus arise spontaneously from an inherent instability of the ocular motor sys-
tem. Most cases of so-called “idiopathic” infantile nystagmus have underlying
genetic mutations. As the downstream effects of these mutations are further eluci-
dated, it will be interesting to see whether the cellular aberrations they produce can
be confined to the ocular motor system, or whether they inevitably perturb the sen-
sory visual system to some degree.
Glossary of Terms
Accessory optic system (AOS) the three terminal midbrain nuclei that receive and
process crossed retinal input related to full-field motion of the visual world
Anterior canal predominance excessive central innervation from the portions of
the otoliths corresponding to the anterior semicircular canals, leading to tonic
upgaze with bilateral inferior oblique muscle overaction.
Atavism a modification of a biological structure whereby an ancestral trait reappears
after having been lost through evolutionary change in successive generations.
Central vestibular imbalance A disruption in input to the vestibular system due
to stimulation or injury to the brainstem or cerebellar neurologic pathways that
transmit information from the labyrinths and eyes.
Cerebellar flocculus a portion of the ancient vestibulocerebellum concerned with
integrating visual and vestibular information to provide ocular stabilization, and
with coordination of the eyes and head during gaze pursuit.
Cortical suppression Inhibition of visual input from one eye within the visual
cortex, usually seen in individuals with strabismus.
Dissociated deviations the tonic deviations and variable binocular movements that
arise from unequal binocular visual input in the setting of infantile strabismus
Dissociated esotonus a differing degree of tonic medial rectus innervation (unre-
lated to convergence) generated by monocular fixation with each of the two eyes
in infantile strabismus.
Dissociated vertical divergence A dorsal drift of either eye evoked by monocular
occlusion or spontaneous cortical suppression of the hyperdeviating eye in indi-
viduals with infantile strabismus.
Dorsal referring to either the back of the organism or the top of the skull
Dorsal light reflex (DLR) a body tilt (or a vertical divergence of the eyes) induced
by unequal luminance input to the two eyes in lateral-eyed animals.
Dorsal terminal nucleus of the accessory optic system (DTN-AOS) the com-
ponent of the midbrain accessory optic system that, together with the nucleus
Children, https://doi.org/10.1007/978-3-030-62720-1
236 Glossary of Terms
of the optic tract (NOT), generates optokinetic responses to full-field nasalward

visual motion.
Luminance the intensity of light per unit area emitted from a surface.
Monocular nasotemporal optokinetic asymmetry (MNTA) a nasalward optoki-
netic predominance under conditions of monocular fixation in lateral-eyed ani-
mals and in primates with infantile strabismus.
Nucleus of the optic tract (NOT) a pretectal nucleus that generates horizontal
optokinesis in response to full-field motion of the visual world.
Ocular tilt reaction (OTR) the triad of vertical divergence of the eyes, binocular
torsion, and head tilt evoked by unequal static graviceptive input from the two
otoliths.
Otolithic pertaining to static vestibular input from the two otoliths.
Optokinesis the ocular motor response to visual motion consisting of repetitive
series of alternating slow and fast ocular rotations.
Phoria adaptation (PA) a slow central adaptation of tonic vergence to reset the
baseline phoria that is induced by prisms or by the attempt to fuse in individuals
with strabismus.
Pitch a fore-or-aft rotation around a side-to-side axis.
Pitch plane corresponding to the sagittal plane (for head and eye rotations).
Posterior canal predominance excessive central innervation from the portions of
the otoliths corresponding to the posterior semicircular canals, leading to tonic
downgaze with bilateral superior oblique muscle overaction.
Roll rotation about the front-to-back axis (or nasooccipital axis of the head) result-
ing in tilt.
Roll plane corresponding to the coronal plane (for head and eye rotations).
Skew deviation One component of the ocular tilt reaction consisting of a verti-
cal misalignment of the eyes induced by unequal input from the two otoliths,
usually due to a neurologic lesion involving their central connections within the
vestibular system.
Subjective vertical the internal sense of vertical, or the perceived orientation of
“true vertical” based on labyrinthine and visual binocular input.
Subjective visual tilt a tilt of the perceived visual world (real or imagined) relative
to the subjective vertical.
Subjective visual vertical (SVV) the perceived vertical position of the visual
world relative to the subjective vertical.
Teleotactic a physical response that occurs without the necessity of maintaining
bilateral (or binocular) balance.
Tonus the excitatory effects of baseline innervation on skeletal or extraocular mus-
culature in the awake alert state.
Tropotactic a physical response that reestablishes bilateral (or binocular)
equilibrium
Ventral referring to the front or the lower side of an organism.
Vestibular nucleus (VN) an embryologically displaced cerebellar nucleus within
the midbrain that modulates input from the labyrinths and the two eyes to stabi-
lize balance
Glossary of Terms 237
Visuo-vestibular Referring to the subcortical visual pathways that transmit bin-

ocular visual input to the vestibular nuclei to generate infantile strabismus and
its accompaniments.
Yaw rotation about the vertical or longitudinal axis resulting in a turn.
Yaw plane referring to the horizontal plane (for head and eye rotations).
Bibliography
Chapter 1, Brodsky MC. Dissociated vertical divergence: a righting reflex gone wrong. Arch
Ophthalmol. 1999;117:1216–22.
Chapter 2, Brodsky MC. DVD remains a moving target. J AAPOS. 1999;3:325–7.
Chapter 3, Brodsky MC, Donahue SP. Primary oblique muscle overaction: the brain throws a wild
pitch. Arch Ophthalmol. 2001;119:1307–14.
Chapter 4, Brodsky MC. Do you really need your oblique muscles: adaptations and exaptations.
Chapter 5, Brodsky MC, Tusa RJ. Latent nystagmus: vestibular nystagmus with a twist. Arch
Ophthalmol. 2004;122:202–9.
Chapter 6, Brodsky MC. Dissociated vertical divergence: perceptual correlates of the human dor-
sal light reflex. Arch Ophthalmol. 2002;120:1174–8.
Chapter 7, Brodsky MC, Gräf MH, Kommerell G. The reversed fixation test: a diagnostic test for
dissociated horizontal deviation. Arch Ophthalmol. 2005;123:1085–7.
Chapter 8, Brodsky MC, Fray KJ. Does infantile esotropia arise from a dissociated deviation? Arch
Ophthalmol. 2007;125:1703–5.
Chapter 9, Brodsky MC. The accessory optic system: the fugitive visual control system in infantile
strabismus. Arch Ophthalmol. 2012;130:1055–9.
Chapter 10, Brodsky MC. Visuo-vestibular eye movements. Arch Ophthalmol. 2005;123:837–42.
Chapter 11, Brodsky MC. An expanded view of infantile esotropia: bottoms up! Arch Ophthalmol.
2012;130:1199–202.
Chapter 12, Brodsky MC. The lizard’s tail: an ocular allegory. Ophthalmology. 2013;120:225–6.
Chapter 13, Brodsky MC, Klaehn L. The optokinetic uncover test: a new insight into infantile
esotropia. Arch Ophthalmol. 2014;131:761–5.
Chapter 14, Brodsky MC, Dell’Osso LF. A unifying neurological mechanism for infantile nystag-
mus. JAMA Ophthalmol. 2014;132:761–8.
Chapter 15, Brodsky MC, Jung J. Infantile exotropia and accommodative esotropia: distinct disor-
ders or two ends of a spectrum? Ophthalmology. 2015;122:1544–6.
Chapter 16, Brodsky MC, Klaehn L. An optokinetic clue to the pathogenesis of infantile esotropia.
Chapter 17, Brodsky MC. Is infantile esotropia subcortical in origin? Prog Brain Res.
2019;248:183–93.
Chapter 18, Brodsky MC. Phoria adaptation: the ghost in the machine. J Binoc Vis Ocular Motility.
2020;70:1–10.
Chapter 19, Brodsky MC. Monocular nasotemporal optokinetic asymmetry: unravelling the mys-
tery. J AAPOS. 2019;23:219–21.
Chapter 20, Brodsky MC. Infantile nystagmus: following the trail of evidence. J
AAPOS. 2020;24:70–1.
Index
A B
Accessory optic system (AOS), 115, 117, 123, Bagolini bar, 15
135, 137, 163, 166, 167, 231 Balance organs, 26, 66, 68, 76, 89, 126,
canal-based coordinate system, 121 132, 147
decortication, 118 bilaterally symmetrical organs func-
DTN and LTN neurons, 120 tion as, 126
neuroanatomical connections, 116 Balancing movement, 20
photic stimulation, 120 Base-in prism, 97, 99–102
primary visual system, 117 Bidirectional horizontal motion
signals self-motion, 117 detection, 193
spherical enclosure, monocular optokinetic Bielschowsky Head Tilt Test, 90
stimulation, sensitivity, 119 Binocular control mechanisms,
visual system, 117 1, 11, 107, 183
Accessory optic tract (AOT), 115, 166 Binocular deviations, 14, 115, 212
Accommodation-linked convergence (AC/A), Binocular disparity, 6, 15, 83, 185, 186, 209
184, 185, 189, 213 Binocular eye movements, 95
Accommodative esotropia, 183–187, Binocular misalignment, 81, 115, 137, 145,
212–213 192, 201, 204, 211
Adaptation, 36, 45–46, 50, 56–58, 89, 108, Binocular tonus mechanism, 223
183–187, 189 Binocular torsional control, 50
reversion from exaptation to, 56–58 Binocular vertical cyclodisparity, 54
Adler, Alfred, 233 Binocular vision, 7, 10, 11, 26, 34, 35,
Alexander’s law, 71, 73 37, 38, 47, 58, 62, 65, 66, 83, 89,
Amacrine, 232, 233 115, 121, 124, 127, 129, 130, 132,
Angular momentum, 147, 148 138, 144, 145, 148, 151, 155, 166,
Anterior canal or posterior canal 173, 198, 199
system, 30, 31 Binocular visual disturbance, 43
A-pattern strabismus, 25, 31–34, 38, 86 Binocular visual imbalance,
Asymmetrical esodeviation, 96 88, 129, 135, 170
Asymmetry, 5, 6, 19, 66, 86 Binocular visual input
monocular nasotemporal optokinetic, 81, abrupt fluctuations, 85
127, 151, 152, 154, 155, 157–159, physiologic imbalance in, 129
192, 193, 196, 197, 199, 200, 225, Binocular visual system, 62
226, 229 Bipolar cells, 232
nasotemporal, 66, 67, 86 Blur-driven accommodation
Nasotemporal optokinetic, 67 convergence, 175, 176, 213
Atavism, 148, 196, 197 Body tilt, 4, 5, 9, 10, 17, 37, 46, 83, 84, 86,
Atavistic torsional movements, 141 90, 93, 128
242 Index
C Dissociated vertical divergence (DVD), 1, 3,

Canal-based coordinate system, 68, 135, 141 14, 17, 128
Canal-based planes, 43 additivity with oblique muscle overac-
Cartesian coordinate system, 130, 135 tion, 18, 19
Causation, theories of, 175–176 association with torticollis, 19
Cellphone vision, 213 balancing movement, 20
Central tonus mechanisms, primary oblique clinical algorithm, differential diag-
muscle overaction, 26 nosis, 22
Central vestibular disease, 58, 65, 131 clinical phenomenology of, 22
Central vestibular innervation, 36, 37 dorsal light reflex, 7–9
Central vestibular imbalance, visuo-vestibular fundamental treatment goals for, 20
eye movements from, 130, 131 latent nystagmus, 20, 23
Cerebellar pathways, 141, 173–175 neurophysiological explanation, 3
Chavessian hypothesis, 145 primary oblique muscle overaction and, 37
Chiari malformation, 34, 58, 141 pseudoinferior oblique muscle over-
Coexistent exodeviation, 102 action, 18
Compensatory balancing movements, 150 skew deviation, 9–11
Congenital esotropia, 65, 91, 108 superimposition of, 18
inferior oblique muscle overaction and variable torsional component, 18
V-pattern strabismus in, 34, 35 visuo-vestibular eye movements, 129
Congenital strabismus, dissociated compo- Dorsal cap (DC) of the inferior olive, 116
nent of, 100 Dorsal light reflex, 3–6, 37, 46, 57, 83, 84, 88,
Congenital superior oblique palsy, 214 89, 91, 120, 128, 131, 196
Congenital visual loss, 172 dissociated vertical divergence, 7–9
Convergence accommodation to convergence Dorsal rotation, 83–84, 89
(CA/C), 185, 212, 213 Dorsal terminal nucleus (DTN), 116, 117, 138,
Cortical motion asymmetries, 157, 199, 201 154, 159, 166, 196, 201, 225,
Cortical pursuit movements, 152 231, 232
Cortical visual motion, 14, 53, 139, 204, 205 Double elevator palsy, 2
Cortico-mesencephalic-cerebellar pathways, Dynamic imbalance, 43, 131
141, 174, 176 Dynamic torsional movements, 195
Craniosynostosis, 42
Crossed fixation, 151, 152, 191–192, 196
Cyclodisparities, 53 E
binocular vertical, 54 Electrical microstimulation, 118
vertical, 48 Esodeviation, 95, 100, 102, 107, 108,
Cyclovergence, 46–48, 50, 52–54, 57 184–187, 201, 212, 215
Cycloversion, 46, 54–56, 58 Esotonus, 139
Cyclovertical divergence Essential infantile esotropia (EIE), 144, 145,
movement, 7, 8, 89 163, 193, 196, 204
crossed fixation, 195
evolutionary basis of, 196–197
D optokinetic etiology for, 200
Darwin, Charles, 147 prenuclear, 195
Decortication, 118, 120, 140, 154, 155 progressive nasalward deviation, 195
Diopter base-in prism, 97–102 subcortical origin, 195
Directionality, 157, 169, 201, 227 subcortical pathophysiology in, 197–199
Direction-sensitive ON-type retinal ganglion Exaptation, 47, 54, 56–58, 89, 147
cells, 117 Exodeviation, 95–97, 99–103, 107, 110, 113,
Dissociated deviation, infantile esotro- 183, 189, 213
pia, 107–111 Extraocular muscle overaction, 42
Dissociated esotonus, 104, 107–111, 113, 217 Extraocular muscles, 6, 9, 11, 26, 29, 30, 43,
Dissociated horizontal deviation (DHD), 95, 45, 62, 65, 68, 71, 108, 123, 126,
96, 99, 102, 104, 109, 110, 113 129, 132, 134, 145, 156, 189, 197
Index 243
F conjugate oscillation, 165

Fine-motor modulation, 53 directionality, 169
Fixational innervation, 8, 17, 19, 31, 33, 35 fixation, 172
Foveation periods, 165, 169, 171 foveation periods, 169
FRMD7 gene, 232 infantile nystagmus, 173
Frontal binocular vision, 47, 58, 65, 127, 132 null position, 170
Fundamental dichotomy, 45 pendularity, 167
Fusional vergence amplitudes, 209, pursuit, 172
214–215, 223 reversed optokinetic nystagmus, 171
Fusion maldevelopment nystagmus, 81 seesaw nystagmus, 171
Fusion maldevelopment syndrome, 205 time of onset, 170
Infantile strabismus, 3, 14, 62, 95, 107, 120,
124, 125, 129, 132, 137, 139, 141,
G 151, 152, 154, 155, 166, 225, 227
Gait, 34 Inferior oblique muscle overaction, 18, 25, 26,
Goldfish, dorsal light response in, 4, 5 28, 29, 32, 34, 35, 65, 86, 128, 129
Intermittent exotropia, 98, 105, 109, 110,
183–187, 189
H Interoceptive stimulus, 14
Handheld rotating drum, 152 Intrinsic ocular motor instability, 182
Hering law, 8, 33 Inverse skew deviation, 10
Hoffmann hypothesis, 198, 204 Invoked vergence adaptation, 108
Horizontal nystagmus, 71
Horizontal optokinetic responses, video-
oculography tracing of, 153–155 L
Human cyclovergence, 52, 53 Labyrinthectomy, 6, 68
Human dorsal light reflex Labyrinthine imbalance, 85, 131
amplitude of DVD, 86 Labyrinthine input, 4, 73, 129, 138
binocular visual input, abrupt fluctua- Labyrinths, 5, 43, 57, 65, 70, 76, 83, 85, 87,
tions, 84 89, 90, 117, 126, 130, 131, 135,
central vestibular system, 90 148, 167, 197
dorsal rotation, 84 Latent nystagmus, 65, 66, 80, 127, 225
monocular occlusion, 86 dissociated vertical divergence, 20
patients and methods, 85 evolutionary underpinnings of, 73, 74
subjective tilt, 91 experimental evidence, 74, 75
subjective vertical, 87, 89 nasotemporal asymmetry and, 66, 67
subjective vertical retinal meridian, 90 neuroanatomy of, 68, 69
true vertical, 87 superimposition of, 172–173
twofold movement, 89 vestibular nystagmus, 67, 68
vision and balance, sensory organs for, 83 vestibular origin, clinical signs of, 70–73
Human ocular torsion, 46 visuo-vestibular eye movements, 129
Hydrocephalus, 32–34, 38, 58 Lateral eyes, 126
Hyperdeviation, 1–3, 6, 8, 11, 17–19, Lateral-eyed animals, 3, 11, 29, 38,
94, 95, 214 46, 50, 57, 67, 74, 76, 83,
88, 115, 127–129, 132, 135,
140, 166, 173, 176, 196, 200,
I 201, 225, 227
Idiopathic infantile nystagmus, 182, 232, 234 Lateral valvuli cerebelli, 6
Infantile esotropia, 22, 42, 109, 110, 113, Listing’s law, 50
137–141, 145, 191, 192, 227 Lichtruckenreflex, see Dorsal light reflex
Infantile Nystagmus, 165–182, 231–233 Lower lateral-eyed animals, 11, 38, 196
absence of oscillopsia, 169 Luminance, 14, 15, 37, 38, 46, 74, 84, 108,
cerebellar pathways, 173, 174 120, 123, 126–128
congenital visual loss, 172 Luminance pathways, 120, 123
244 Index
M primary adaptations in, 45, 46

Maddox Rods, 34, 49, 50 reversion, exaptation to adaptation, 56–58
Medial superior temporal (MST), 68, 137, Oblique muscle overaction, dissociated
138, 154, 157–159, 169, 174, 201 vertical divergence, 18
Medioterminal nucleus (MTN), 117 Oblique muscles, 46, 148, 149
Melanopsin pathways, 14 inferior overaction, 18, 25, 26, 28, 29, 32,
Methadone moms, 182 34, 35, 65, 86, 128, 129
Middle temporal (MT) visual area, 68 ocular, 150
Monocular fixation, 8, 19, 72, 87, 95, 107, primary overaction, 25, 26
113, 200, 207 pseudoinferior overaction, dissociated
Monocular nasotemporal optokinetic vertical divergence, 18
asymmetry (MNTA), 81, 127, 151, superior overaction, 25, 31, 33, 34
152, 154, 155, 157–159, 192, 193, Oscillopsia, absence of, 169
196, 197, 199, 200, 225, 226, 229 Ocular motor incursions, visuo-vestibular eye
neuroanatomy of, 197 movements, 127–129
Monocular occlusion, 11, 85–88, 209, Ocular motor system, vestibular interactions
211–214, 217 with, 26, 28–32, 173
Monocular optokinetic stimulation, sensitiv- Ocular oblique muscles, 150
ity to, 119 Ocular stabilization, 31, 186
Monocular visual loss, 108, 173 Ocular tilt reactions, 4, 10, 132
Monofixation syndrome, 145 Ohm’s visual balancing metaphor, for
Motion pathways, 80, 120 dissociated vertical divergence, 128
Muscle length adaptation, 36, 108, 215 Optical coherence tomography (OCT), 231
Optic flow, 67, 75, 138, 152, 154, 225–227
Optokinesis, 81, 123, 148, 163, 200, 231, 232
N Optokinetic, 138, 141, 151–159, 191,
Nasalward movement, 66 192, 231–233
Nasotemporal asymmetry, 86, 105, 127, 128, Optokinetic eye movements, 69, 138, 170
192, 196, 204 Optokinetic input, 116, 131, 154, 155, 163,
Nasotemporal optokinetic asymmetry, 67, 120, 172, 176, 191–193, 201, 204, 226
138, 151, 173, 192, 193 Optokinetic pathways, 70, 138, 151, 152, 154,
Neuroanatomy, latent nystagmus, 68, 69 155, 158, 166, 170, 172–174, 176,
Neurologic disorders, superior oblique muscle 198, 227
overaction and A-pattern strabismus Optokinetic responses, 68, 74
in, 31, 34 horizontal, video-oculography trac-
Neurologic lesion, 10, 35, 38, 129, 141 ing of, 153
Nonstereoscopic perception, 54–66 subcortical, 69
Nucleus of the optic tract (NOT), 67, 68, 75, Orchid evolution, 63, 147
117, 127, 138, 154, 157, 159, 166, Otolithic pathways, 31, 36
168, 174, 192, 198, 225, 231, 232 Otoliths, 4, 6, 19, 29, 31, 57, 90, 131, 214
Nucleus of the optic tract–dorsal terminal
nucleus (NOT-DTN), 157, 159,
166, 225 P
Null position, 170, 171 Panum’s space, 54
Paradoxical movement, 150
Pendularity, 167–168
O Perceived visual tilt, depiction of, 88
Oblique muscle function Perceptual correlates, human dorsal
cyclovergence, stereoscopic perception, light reflex
and Pitch Plane, 47, 48, 50, 52–54 amplitude of DVD, 86
cycloversion, nonstereoscopic perception binocular visual input, abrupt fluctua-
and the Roll Plane, 54–56 tions, 84
frontal binocular vision, from visual central vestibular system, 90
panorama, 47 dorsal rotation, 84
Index 245
monocular occlusion, 86 ocular motor system, vestibular interac-

patients and methods, 85 tions with, 26, 28–32
subjective tilt, 91 phenomenon of, 42
subjective vertical, 87, 89 Primary oblique overaction, visuo-vestibular
subjective vertical retinal meridian, 90 eye movements, 129
true vertical, 87 Primitive reflexes, 1, 73, 126, 227
twofold movement, 89 Primitive visual reflexes, 127, 129, 132
vision and balance, sensory organs for, 83 Prolonged occlusion, 35, 71, 186, 187,
Peripheral vestibular disease, 65 215, 218
Periventricular leukomalacia, 34, 86, 137, 139 Prolonged subcortical neuroplasticity,
Phoria adaptation, 183–187, 189, 208, 223 potential role of, 199–200
accommodative esotropia, 212–213 Pseudoinferior oblique muscle overaction, 18
binocular tonus mechanism, 223
congenital superior oblique palsy, 214
fusional vergence amplitudes, 215 R
intermittent exotropia, 213, 214 Retinal disparity, 62, 185
latent phoria, 212 Retinotectal input, 124
measurement, 209 Reversed fixation test (RFT), 96, 97, 100, 104
neural substrate, 215–216 advantage of, 103
orthophorization, 212 horizontal deviation, 103
physiologic skew deviation, 214 Reversed optokinetic nystagmus, 171
postoperative prism adaptation, 214 Roll plane, 54–56
protean clinical manifestations, 211–215
resilience, 210
spread of comitance, 215 S
stillness, 211 Seesaw nystagmus, 171
stimulus, 209 Semicircular canals, 29–31, 67, 68, 70, 71, 85,
time course, 210 117, 132, 157, 159, 167, 174
Photic stimulation, 120 Sensory balance organs, 66, 76, 126, 132
Physiologic skew deviation, 215 Sensory disconjugate torsion, 42
Physiologic V pattern, 63 Single binocular vision, 34, 35, 37, 38,
Pitch plane, 47, 49, 50, 52–54 47, 83, 204
Pitch-down body movement, 27, 28 Skew deviation, dissociated vertical diver-
Postoperative prism adaptation, 215 gence, 9–11
Postrotational nystagmus, 71 Stereoisomers, vestibular reflexes, 132
Postural control, 34 Stereopsis, 7, 47, 50, 52, 54, 62, 63, 85, 127,
Predominant mechanism, 205 144, 145, 204, 205, 209
Primary adaptations, oblique muscle Stereoscopic effect, 50, 53
function, 45, 46 Stereoscopic perception, 47, 49, 50, 52–54
Primary inferior oblique muscle overaction, Structural neurologic disease, 58
25, 28, 35, 128, 129 Subcortical, 138, 140, 151, 152, 154, 155, 159,
Primary oblique muscle overaction, 25, 26 196–201, 225, 227
and dissociated vertical divergence, 37 Subcortical accessory optic system, 145, 163
and Hering’s law, 35, 36 Subcortical central vestibular pathways, 130
central tonus mechanisms for, 26 Subcortical mechanism, 14, 43, 155
congenital esotropia, inferior oblique Subcortical optokinesis, 163, 231
muscle overaction and V-pattern Subcortical optokinetic responses, 69, 154,
strabismus in, 34, 35 167, 172, 192
gravistatic and visual input, physiologic Subcortical visual pathways, 76, 120, 124,
effects of, 27 138, 145, 155, 158, 159, 200
neurologic disorders, superior oblique Subjective visual tilt, 87, 88, 90, 91
muscle overaction and A-pattern Subjective visual vertical, 20, 57, 88, 89, 93
strabismus in, 31, 34 Superior oblique muscle overaction, 18, 25,
nonneurologic causes of, 37, 38 31–34, 58, 86
246 Index
Superior oblique palsy, 54 experimental evidence, 74–75

congenital superior oblique palsy, 214–215 latent nystagmus, 68–70
vestibular origin, 70–73
Vestibular origin, clinical signs of, 70–73
T Vestibular reflexes, 115, 120, 132
Telotactic, 6 Vestibulocerebellum, 33, 117, 120, 173–175,
Tenacious proximal fusion, 183, 213 196, 215
Titmus stereoacuity test, 47 Vestibulo-ocular movements, 29, 132
Tonic imbalances, 43, 131, 173 Vestibuloocular pathways, 32
Tonic vergence, 207, 208, 210, 212, 217 Vestigial function, 148
Tonus, 19, 26, 27, 29, 32, 34–38, 43, 47, 66, Video-oculography, 23, 152, 153
72, 80, 107, 130, 189, 208, 217 Visual acuity, 85, 97, 109, 173, 231
Torsional eye movements, 15, 45, 46, 56, 58, Visual field, 7, 28, 67
90, 93, 115, 118, 120, 121, 123, Visual righting reflex, 4
135, 137, 138, 145, 148, 171 Visual tilt
Torticollis, 19, 22 perceived, depiction of, 88
Tropotactic, 5, 6 subjective, 87
True hypertropia, 94 Visual stabilization mechanisms, 68
True vertical, 23, 87, 88, 93 Visuo-vestibular, 8, 138, 140, 154, 158
Visuo-vestibular eye movements, 135, 196
balance organs, bilaterally symmetrical
U organs function as, 126
Umstimmung, 6 central vestibular disease, 131
Unequal accommodative convergence, from central vestibular
96, 97, 100 imbalance, 130, 131
Utricular ocular tilt reaction, 9 latent nystagmus, primary oblique
overaction, and dissociated vertical
divergence, 129
V ocular motor incursions, 127–129
Variable torsional component, 18 primitive reflexes, 126, 127
Ventral tegmental tract, 31, 32 primitive visual reflexes, 129
Vergence adaptation, 108 sensory balance organs, lateral
Vertical cyclodisparity, 48, 54 eyes, 126
Vertical head-tilt test, 36 subcortical central vestibular pathways,
Vestibular eye movements, 45, 67, 71, generated by, 130
120, 130 visual reflexes, 132
Vestibular interactions, with ocular motor Visuo-vestibular imbalance, 35, 65, 74, 131
system, 26, 29–31 Visuo-vestibular tonus predominates, 189
Vestibular nystagmus, 65, 67–76 Volitional body movements, 15, 195
evolutionary underpinnings, latent von Holst, Erich, 6
nystagmus, 73–74 V-pattern strabismus, 32, 34–35, 86

The Evolutionary Basis of Strabismus and Nystagmus in Children

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Evolutionary Basis of Strabismus and Nystagmus in Children

Uploaded by

Copyright:

Available Formats

The

The Evolutionary Basis

ISBN 978-3-030-62719-5 ISBN 978-3-030-62720-1 (eBook)

Pediatric ophthalmologists manage a unique collection of ocular motility distur-

The intrusion of these subcortical eye movements reflects a primitive binocular

1 Dissociated Vertical Divergence���������������������������������������������������������������� 1

Cycloversion, Nonstereoscopic Perception, and the Roll Plane������������ 54

Primitive Visual Reflexes are Evoked by a Physiologic

16 An Optokinetic Clue to the Pathogenesis of Crossed

In the course of evolution, primitive responses to external stimuli are suppressed by

Dissociated vertical divergence is an enigmatic disorder characterized by a slow

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 1

A cogent neurophysiological explanation of DVD must account for the follow-

1. Both eyes never drift up simultaneously (Fig. 1.1) [7].

What Is a Dorsal Light Reflex?

Dissociated vertical divergence recapitulates a primitive visuo-vestibular righting

Fig. 1.3 In fish, visual and

function of the degree of visual input asymmetry, as evidenced by the Bielschowsky

that visuo-­vestibular information contributes to postural adjustment through the

Is DVD a Dorsal Light Reflex?

b Visuo-vestibular + Fixational Innervation

Is DVD a Skew Deviation?

Skew deviation is a descriptive term used to denote an acquired, supranuclear, verti-

Table 1.2 Complementary Ocular Tilt Reactions in Humans

function together as complementary righting responses in lower lateral-eyed ani-

7. Spielmann A. Les divergences verticales dissociées: excès desursum version lié

26. Yanagihara D, Watanabe S, Mitarai G. Neuroanatomical substrate for the dorsal

Dissociated vertical divergence (DVD) is an ocularmotor phenomenon that is

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 17

The superimposition of DVD on the static and dynamic innervational derange-

Variable Torsional Component

Pseudoinferior Oblique Muscle Overaction

Additivity with Oblique Muscle Overaction

muscles), oblique muscle overaction is generally caused by sensory torsion of the

Association with Torticollis

Relationship to Latent Nystagmus

3. Graf W, Meyer DL. Central mechanisms counteract visually induced tonus

21. Pritchard C. Incidence of dissociated vertical divergence in intermittent exotro-

In this extension of my original treatise, I address the clinical phenomenology of

Horizontal Realignment of the Eyes

Symmetrical DVD Asymmetrical DVD Unilateral DVD

No head tilt Compensatory Compensatory Noncompensatory

Binocular Occlusion OS Occlusion OD

Primary oblique muscle overaction is a common ocular motility disorder character-

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2021 25

primary oblique muscle overaction in humans. Clinical observations suggest that an

 entral Tonus Mechanisms for Primary Oblique

Vestibular Interactions with the Ocular Motor System

To understand why primary oblique muscle overaction so often accompanies early-­

Fig. 3.3 The close Midline

the semicircular canals and Superior

modified with permission

cerebellum, peripheral vestibular input is summated to produce appropriate innerva-

 uperior Oblique Muscle Overaction and A-Pattern Strabismus

FLO NOD FLO

Right Labyrlnth Vestibular Innervation Left Labyrlnth

Vestibular and Fixational Innervation

A-pattern strabismus, bilateral superior oblique muscle overaction, and bilateral

of or injury to those vestibulocerebellar pathways activating the anterior canals

I nferior Oblique Muscle Overaction and V-Pattern Strabismus

Clinical observations and eye movement recordings have documented abnormal

Primary Oblique Muscle Overaction and Hering’s Law

Right Labyrinth Vestibular Innervation Left Labyrinth

1 Dissociated Vertical Divergence�� 1

Cycloversion, Nonstereoscopic Perception, and the Roll Plane�� 54

16 An Optokinetic Clue to the Pathogenesis of Crossed

What Is a Dorsal Light Reflex?

that visuo-vestibular information contributes to postural adjustment through the

Is DVD a Dorsal Light Reflex?

Is DVD a Skew Deviation?

entral Tonus Mechanisms for Primary Oblique

Vestibular Interactions with the Ocular Motor System

To understand why primary oblique muscle overaction so often accompanies early-

uperior Oblique Muscle Overaction and A-Pattern Strabismus

I nferior Oblique Muscle Overaction and V-Pattern Strabismus

Primary Oblique Muscle Overaction and Hering’s Law

blique Muscle Overaction and Dissociated

Nonneurologic Causes of Oblique Muscle Overaction

Primary Adaptations in Oblique Muscle Function

Oblique Muscle Exaptations

From Visual Panorama to Frontal Binocular Vision

Cyclovergence, Stereoscopic Perception, and the Pitch Plane

Cycloversion, Nonstereoscopic Perception, and the Roll Plane

Reversion From Exaptation to Adaptation

tinkering-it is not design. The visual system never had an opportunity to be

What Is Latent Nystagmus?

Nasotemporal Asymmetry and Latent Nystagmus

Is Latent Nystagmus a Vestibular Nystagmus?

Neuroanatomy of Latent Nystagmus

Clinical Signs of Vestibular Origin

Evolutionary Underpinnings of Latent Nystagmus

xperimental Evidence That Latent Nystagmus Is Vestibular

Reversed Fixation Test

primarily in the relative activation of binocular fusion (which behaves as an all-or-