Dynamic vergence eye movements instrabismus and amblyopia:

symmetric vergenceRobert V. Kenyan, Kenneth J. Ciujfreda,* and Lawrence Stark

Dynamic vergence eye movements in response to step target displacements along the midlinewere measured by an infrared reflection technique in 11 patients having either intermittentstrabismus, constant-strabismus amblyopia, or amblyopia without strabismus. We found theabsence of normal disparity (fusional) vergence in all patients having strabismus and in somepatients having amblyopia without strabismus. A characteristic response consisting of a bin-ocular accommodative vergence component and an early binocular saccadic component wasused to foveate the target of interest with the dominant eye. Vergence responses in our controlsubjects and patients with the nondominant eye occluded were similar to those recorded in ourpatients during binocular vieioing. These results suggest that disparity information is notutilized by patients, probably as a result of long-term, ongoing suppression in the deviated oramblyopic eye. Accommodative vergence with the aid of an early foveating saccade was theprimary mechanism for tracking targets in three-dimensional space.

Key words: strabismus, amblyopia, vergence, eye movements,Hering's law, accommodation

V,ergence eye movements in patients withstrabismus and amblyopia have been inferredfrom patients' subjective responses to dispa-rate stimuli presented simultaneously to eacheye, rather than objectively recorded andquantified (for a review of early work, seeBurian and von Noorden1). Burian,2 follow-ing Bielschowsky3 and Schlodtmann,4 be-

From the School of Optometry, University of California,Berkeley.

This research was supported by N.I.H. training grantEY00076 (R. V. K. and K. J. C.) and an Auxiliary tothe American Optometric Association Research grant(K. J. C ) .

Submitted for publication July 26, 1978.Reprint requests: Robert V. Kenyon, Man-Vehicle Lab-

oratory, Department of Aeronautics and Astronautics,Room 37-215, Massachusetts Institute of Technology,Cambridge, Mass. 02139.

*Present address: Department of Basic Optometric Sci-ences, State College of Optometry, State University ofNew York, 100 E. 24th St., New York, N. Y. 10010.

lieved that some strabismic patients couldproduce "fusional movements" when dispa-rate stimuli, placed in the peripheral retina ofeach eye, were reported fused. Patients re-ported changes in the localization of two hap-lopic fiduciary marks located in the center ofthe visual field, and this was interpreted asevidence for actual eye rotations. Hallden5

confirmed Burian's finding in patients havinganomalous retinal correspondence (ARC). Hesimilarly defined fusional movements as a re-turn to the no-disparity condition in the cen-tral field following the introduction of a dis-parity. In Hallden's study, changes in sub-jective localization of targets were estimatedby haploscopic techniques, and from this hecalculated changes in vergence angle. Thesefusional movements took several minutes tocomplete, and static measurements of targetlocations were made only about once a min-ute. However, in none of the above studieswere eye movements objectively recorded.

60 0146-0404/80/010060+15$01.50/0 © 1980 Assoc. for Res. in Vis. and Ophthal., Inc.

Volume 19Number 1 Symmetric vergence in strabismus, amblyopia 61

Mariani and Pasino6 criticized Burian'sstudy and pointed out that other factors couldaccount for his results. They referred toKretschemer,7 who showed that changes inthe angle of anomaly rather than true fusionalmovements could be the mechanism em-ployed to fuse disparate targets. (The angle ofanomaly is measured from the fovea in thenondominant eye to a point in the nondomi-nant eye that corresponds to the same visualdirection as the fovea of the dominant eye.)From their own experiments on fusion instrabismic patients, still without objectiveeye movement recordings, Mariani and Pa-sino6 inferred that most patients changed theangle of anomaly instead of performing truefusional movements, thus confirming Kret-schemer's findings.

In an attempt to reconcile these differ-ences and further extend these observations,we objectively recorded dynamic vergenceeye movements in 11 patients having ei-ther constant-strabismus amblyopia, ambly-opia without strabismus, or intermittentstrabismus, while changing binocular fixationbetween near and far midline targets. Thesetargets produced disparity by changing dis-tance from the patients as is found in normaleveryday binocular viewing. Quantificationof the dynamic aspects of the vergence re-sponse in these patients shows the absence ofnormal disparity vergence movements andthe use of normal accommodative vergenceeye movements and saccades to track targetsin three-dimensional space.

Methods

Binocular horizontal eye position was recordedby an infrared reflection method.8- 9 The recordingsystem had an overall bandwidth of 150 Hz, a lin-ear range of at least ±7°, and a noise level of 6 minarc. The infrared method does not distinguish be-tween eye rotation and eye translation; hence,translational movements, if large enough, couldintroduce artifacts in the eye movement record-ings. However, Krishnan and Stark,10 using amethod that distinguishes between rotation andtranslation, have shown that eye translation duringsaccadic or vergence eye movements is minimal.Furthermore, using data from Fry and Hill," theyestimated that the translational movement of the

eye would only be 0.01 mm or 0.05° for a 10°rotation, below the noise level of our recordingsystem.

To ensure that the responses were independentof the target system, two systems were used tostimulate vergence. In each system, targets wereplaced along the subject's midline at 25 and 50 cmaway from the estimated center of rotation of eacheye. These targets were carefully adjusted in bothhorizontal and vertical planes to minimize occur-rence of eye movements resulting from targetmisalignment. In one system, the two targetsconsisted of small Lucite plates with fine crossesetched in the front surfaces subtending visual an-gles of 1.5° and 3.0° for the far and near targets,respectively. The fine lines of the crosses sub-tended angles of 2 and 4 min arc for the near andfar targets, respectively. A miniature bulb (GE222) was installed to provide target illumination ofthe etched cross alone. In the other system,targets were made from strips of exposed film withfine crosses etched into the emulsion, and thetargets were mounted in front of a 2 cm squareholder and back-illuminated by a grain-of-wheatfilament bulb through a light diffuser. The holdersfor the far and near targets subtended angles of2.3° and 4.5°, respectively. The fine lines compos-ing the crosses subtended angles of 2 and 4 min arcfor far and near targets, respectively.

The luminance of the targets was measuredwith an S.E.I, photometer under the same lowphotopic illumination conditions used during theexperiments. The luminance for the Lucite platesmeasured 0.5 log ft-lamberts. The contrast forthese targets with the formula (lmax — tarn/Lax +Imin) was 0.53. For the second system using theexposed film, the target luminance measured 1.7log ft-lamberts and the contrast was 0.93.

Experimental procedures were adapted to theclinical situation and thus kept as simple and asuniform as possible. Targets were alternately il-luminated in a pseudorandom sequence by theexperimenter. Each subject was instructed tokeep the illuminated target clear and single at alltimes. These procedures were performed (1) withboth eyes viewing and (2) with the nondominanteye covered but still allowing eye movements tobe recorded from both eyes. Calibrations wereperformed before, during, and after each brief ex-perimental run.

Eleven patients having strabismus and/or func-tional amblyopia and two normal control subjectsparticipated in the study. Patients were recruitedfrom the general clinic at the School of Op-tometry, University of California, Berkeley. All

62 Kenyon, Ciuffreda, and StarkInvest. Ophthalmol. Vis. Sci.

January 1980

Table I. Clinical data of subjects

Subject Age PrescriptionVisualacuity

Vergenceabnormality

Eccentricfixation

Correspondenceand stereoacuity*

Constant-strabismus amblyopia:B. S.

D. F.

J. K.

B. B.

25

23

15

33

LE +2.00RE +2.25LE +3.75RE +0.50LE -1.50RE -1.75LE +0.75RE +0.25

Amblyopia without strabismus:S. H.

L. T.

B. W.

24

25

19

LE -0.75RE pi =LE -2.50RE -5.00LE +5.00RE +3.00

= -0.25 x 130

= -0.50 x 165

= -0.50 x 40= -0.50 x 180

= -2.00 x 90-0.50 x 19= -1.25 x 172= -0.75 x 5

Intermittent strabismus without amblyopia:J. L.

B. N.

FormerJ. W.

J. B.

ControlB. L.

A. D.

13

31

strabismus24

25

subjects:29

23

LE +0.75RE +0.50LE -5.00RE -4.50

patients:LE -5.00RE -2.50LE -0.75RE -2.75

LE PianoRE PianoLE +8.0 =RE +8.0

- -0.75 x 20

= 0.75 x 165= -0.50 x 8= -0.75 x 90

= 0.25 x 175

20/2520/1520/3020/1520/12220/2020/63020/10

20/3820/2020/2520/4020/11020/15

20/2020/2020/2020/20

20/1520/2520/2520/30

20/2020/2020/2020/20

1-2* ET LE

18* ET LE

10* ET LE;1* HT LE5-6*ETLE;2* HT LE

None

None

None

20* ET RE;6* HT RE15* XT LE

6*EPH

10* EXPH2*HPH

None

None

V2* nasal LE

1* nasal LE

2.5* nasal;2* superior LE2.5^.5* nasal;3-4* superior LE

2* nasal; 2*inferior LE2* Nasal; 2*inferior RE2* temporal LE

Jerk nyst.

Central, steadyLE, RE

None

None

None

None

ARC200"

400"ARC300"

——

NRC100"NRC60"NRC400" -> 60"

UHARCt—

ARC40"

NRC140"NRC

—

NRC40"NRC40"

ET = esotropia; XT = exotropia; ARC = anomalous retinal correspondence; NRC = normal retinal correspondence.*At least as indicated.tUnharmonious ARC.

patients had a thorough vision examination andwere free from ocular or neurological disease.They were classified into three major groups:constant strabismus with varying degrees of am-blyopia, amblyopia without strabismus, and in-termittent strabismus without amblyopia. In somepatients classified as amblyopic, the visual acuitieswere less than 20/40 in the nondominant eye. Inthese cases, the criterion of a two-line differencebetween the visual acuities of the two eyes wasused to define amblyopia. This method is standardclinical practice.12' l3 The subjects' ages rangedfrom 10 to 33 years, with an average age of 25. Asummary of each subject's clinical findings is pre-sented in Table I.

It should be noted that what is called "dispar-ity" vergence in this paper has been known as"fusional" vergence in the past. We prefer to usethe term disparity rather than fusional becausedisparity drives the vergence system and fusion isusually a higher-level perceptual result followingthese movements.

Results

Absence of disparity vergence in patients.Typical disparity vergence eye movementsproduced by changing fixation between nearand far midline targets were characterized bya 160 to 200 msec latency followed by

Volume 19Number 1 Symmetric vergence in strabismus, amblyopia 63

None

None

Previoussurgery or

therapy

Age 16SurgeryAge 6SurgeryNone

Age 8Surgery

None

None

Age 17Orthoptics

Age 2SurgeryAge 16Surgery

Age 3SurgeryAge 8Surgery

Vergence abilities

40 cmBase tnA

Base out*(break/

refusion)

8/418/106/2

24/1424/15

——

24/1614/1038/1027/22

——

—40/doubled24/doubled

28/1832/18

——

6 mBase inBase out*(break/

refusion)

14/8——

15/106/3——

36/264/2

32/144/2——

—40/doubled8/doubled

16/1032/30

——

smooth, equal, disjunctive movement of theeyes.14"18 The response in Fig. 1, a, of a con-trol subject exemplifies a normal response.At times, the vergence contained a small sac-cade as seen during the divergence responseof Fig. 1, a.

In contrast, most patients with strabismusand/or amblyopia did not perform normaldisparity vergence responses. Although thelatencies were normal, these patients madeunequal vergence movements in each eye asshown in Fig. 1, b. The vergence amplitudeswere always smaller in the dominant eye thanin the fellow eye; the average vergence

amplitude of the dominant eye was approxi-mately 18% of that of the fellow eye. In theseresponses, the unequal vergences were ac-companied by a binocular saccade that servedto place the fovea of the dominant eye closeto the new target. The combination of thissaccade and the small amplitude vergencefunctioned to place and maintain the domi-nant eye on the new target.

This pattern of unequal vergence am-plitudes accompanied by a saccade charac-terized all the responses made by patientshaving intermittent strabismus without am-blyopia. Eye movement responses in Fig.1, b, display the typical response found inPatient B. N. (see Table I) and the other pa-tient in this diagnostic group. In each re-sponse, the vergence in the dominant eyewas much less than that found in the felloweye. For example, the convergence ampli-tudes in Fig. 1, b equaled 0.8° and 9.0° forthe dominant and nondominant eyes, re-spectively. For both convergence and di-vergence responses, the percent vergenceamplitude in the dominant eye equaled 11%of that found in the nondominant eye.

Patients with constant-strabismus ambly-opia responded similarly. A representativeexample is seen in the eye movement re-sponses of Subject J. K. (Fig. 2, a). Re-sponses showed an absence of normal dispar-ity vergence similar to those found in theprevious patient (Fig. 1, b), having intermit-tent strabismus. Here, too, the unequal ver-gence amplitudes that occurred during bothconvergence and divergence, had lower am-plitudes in the dominant left eye than in thefellow eye. For example, the convergenceamplitudes equaled 0.8° and 4.5° for thedominant and nondominant eyes, respec-tively. For these records, dominant eye ver-gence amplitudes measured 18% of those inthe fellow eye for both the convergence anddivergence.

In contrast to patients with strabismus, notall amblyopic patients exhibited these ab-normal vergence repsponses. Of three pa-tients having amblyopia without strabismus,only L. T. performed normal vergence re-sponses at all times; another amblyope,

64 Kenyon, Ciuffreda, and Stark Invest. Ophthalmol. Vis. Sci.January 1980

°RE * °T

LF

NEAR

500 ms

-f+14+*++1

Lt

'RE l * - _ . -

'< i _

•"V

• r I - 4 ;F1J500 m«

Fig. 1. Symmetric vergence condition. Symbols for all figures unless otherwise noted:0Le = left eye position; BRe = right eye position; 0 ^ = left eye velocity; 0R e + 0 T = right eyevelocity summed with target position. Downward deflections represent rightward movements;calibration bars for position represent 2°, and time markers represent 500 msec, a, ControlSubject B. L. shows normal response to symmetric vergence stimuli, b, Patient B. N. havingintermittent strabismus without amblyopia exhibits abnormal response, contrary to normalcontrol subject responses in a. Following target changes, vergence occurs primarily in left eyeonly. Saccade occurring early in response is used to foveate new target with right eye; noteunequal saccades occurring during vergence response, but equal saccades during extendedfixation periods.

B. W., always produced abnormal vergenceresponses similar to those of the strabismicpatients; the last amblyope, S. H., had briefintermittent episodes of normal vergence butonly during convergence. Eye movementsfrom Patient S. H. are shown in Fig. 2, h; thevergence amplitudes in the dominant eyeequaled 6% of those in the amblyopic eye.

Monocular viewing of midline targets.When one eye was occluded, the responsesof the control subjects to the symmetric ver-

gence stimuli resembled those responsesproduced by all strabismic patients and somenonstrabismic amblyopic patients under bin-ocular viewing conditions. The response ofControl Subject B. L. with the right eye cov-ered is shown in Fig. 3. The vergence re-sponses seen in this figure contained only anaccommodative vergence component due tothe occlusion of one eye. This is to be ex-pected since disparity vergence requires abinocular stimulus; occluding one eye dis-

Volume 19Number 1 Symmetric vergence in strabismus, amblyopia 65

©LE

ORE

©RE • eT

NEAR

FAR

500 mi

Fig. 2. Symmetric vergence condition, a, Patient J. K. with constant strabismus amblyopia,showing similar abnormal responses consisting of predominant vergence in amblyopic left eye,early foveating saccade in dominant right eye, and non-Hering's law saccades during re-sponse, especially for convergence, b, Patient S. H. having amblyopia without strabismus.Similar abnormal responses also found in this patient, but they only occurred intermittently.

ables the disparity vergence control system.Characteristically, the amplitude is smaller inthe viewing eye than in the covered eye dur-ing accommodative vergence.9j 19 For in-stance, the amplitude of the accommodativevergence movements in Fig. 3 equaled 4.2°in the covered eye and 1.2° in the view-ing eye for both convergence and diver-gence. The binocular saccade, due to the ret-inal error in the viewing eye, moves the foveaof the dominant eye close to the new target.Both qualitatively and quantitatively, the re-sponses from the control subjects' monoculartracking were similar to those recorded in ourpatients under binocular viewing conditions

{Figs. 1, b, and 2). Each contained unequalvergence; the smaller amplitude occurred inthe viewing eye for the control subject and inthe dominant eye for the patients; and finally,each used a saccade to bring the fovea of thedominant eye near the new target.

When patients with strabismus and/oramblyopia were monocularly occluded, sothat only the dominant eye viewed thesymmetrically positioned targets, responsesagreed both qualitatively and quantitativelywith the symmetric vergence responses fromthese patients under binocular viewing con-ditions. As with the control subjects, block-ing of one eye caused accommodation to

66 Kenyon, Ciuffreda, and StarkInvest. Ophtfialmol. Vis. Sci.

January 1980

500 m»

Fig. 3. Monocular tracking of midline targets with dominant eye. Normal Control SubjectB. L. with left eye occluded. Note vergence occurring predominantly in occluded eye; viewingeye saccade is used to foveate new target.

drive vergence, producing unequal vergenceamplitudes. The monocular retinal errorproduced by the target change caused a bin-ocular saccade placing the dominant eye onthe new target.

Monocular line of sight targets. To studythe ability of these patients to make normalaccommodative vergence responses, the tar-gets were carefully aligned along the visualaxis of the dominant eye with the fellow eyeoccluded, (the Miiller experiment).19 Underthis condition, a large accommodative ver-gence response occurs in the covered non-dominant eye.19 We have recently shownthat besides the large vergence in the cov-ered eye, the viewing eye makes small ver-gence responses approximately 12% of theamplitude of that measured in the coveredeye.9 A clear example of this accommodativevergence movement is seen in the responsesfrom Control Subject B. L. (Fig. 4, a). In thedivergence response, the small vergence inthe viewing eye equaled 25% of that found inthe covered eye. The small corrective sac-cades were used to keep the viewing eye onthe target and thus counteracted the smallvergence driving the viewing eye away fromthe target. The responses from our patients

under these conditions appeared similarto the accommodative vergence responsesfound in normals. Typical accommodativevergence responses from our patients areshown in Fig. 4, b. This figure shows the eyemovement response of Patient B. N. exhibit-ing normal characteristics of accommodativevergence: smaller amplitude vergence in theviewing eye (25% of the amplitude in thecovered eye) and corrective eye movementpatterns.9 In general, accommodative ver-gence with the dominant eye viewing wasnormal in all patients.

Special cases. Neither degree of strabis-mus nor grade of amblyopia in patients withconstant strabismus amblyopia had any effecton the responses to the symmetric vergencestimuli. Patients with small-angle strabismusand deep amblyopia (B. B.) produced sym-metric vergence responses similar to those ofpatients who had mild amblyopia and small-angle strabismus (B. S.). Furthermore, thepresence of ARC, as found for Patient B. S.but not B. B., did not alter the basic re-sponse.

Within our group of patients, three weresuccessfully treated for amblyopia or strabis-mus. Patient B. W. received orthoptics ther-

Volume 19Number 1 Symmetric vergence in strabismus, amblyopia 67

'LE

3RE

FAR

°RE * °T NEAR f j

FLE

LE I

500 ms

tf500 ms

Fig. 4. Accommodative vergence with dominant eye viewing, a, Normal Control SubjectB. L., with left eye occluded. Note vergence occurring primarily in occluded eye and smallvergence in viewing eye, similar to that recently reported in other normal subjects.9 b, PatientB. N. having intermittent strabismus without amblyopia. Accommodative vergence showsnormal dynamical response characteristics.

apy for amblyopia and eccentric fixation inthe left eye. After training, his visual acuityincreased from 20/110 to 20/20, and his ec-centric fixation reduced from 2 prism diop-ters (PD) to zero. His vergence movements,recorded shortly after termination of orthop-tics treatment, contained the same character-istic abnormal vergence response recorded inmost of our patients. Thus, even after nor-malization of visual acuity and centralizationof fixation, normal disparity vergence re-sponses were never recorded in this formeramblyope. Interestingly, some other oculo-motor findings such as saccadic latency re-mained abnormal in this patient duringand subsequent to successful amblyopiatherapy.20"22

Patients with a history of strabismus as achild and surgical correction to reduce thestrabismus were also included in our sampleand showed the typical abnormal vergence

responses. Patient J. B, who now has anexophoria was operated on at age 8 to correcta squint of his left eye. His symmetric ver-gence responses were similar to those of theother patients in our study having strabismusbut who were not surgically corrected. How-ever, another patient, J. W., who had an eso-tropia surgically corrected at age 3 and pres-ently has an esophoria, had intermittentepisodes of normal convergence interspersedwith more frequent abnormal responses.

Non—Hering's law saccades during ver-gence. During the analysis of vergence re-sponses from both patients and control sub-jects, we found that saccades which occurredduring vergence were often unequal in eacheye. Fig. 5 shows symmetric vergence re-sponses that contained saccades during dis-parity vergence from the control subjects. Al-though the vergences in each eye wereapproximately equal, the saccades were

68 Kenyon, Ciuffreda, and StarkInvest. Ophthalmol. Vis. Sci.

January 1980

*«

300 IM

-

-m * •

-_

_

•

\

zm

l-

&

"-

"_-1?

s

H ft m M

-

m

• •• ™ r —

9 0 0 * * fc

Fig. 5. Non-Hering's law saccades in normal con-trol subjects, a, Subject A. D. In left record, note0.6° saccade in left eye and 2.0° saccade in righteye occurring early in convergence response.Similar markedly unequal saccades found early indivergence responses (right record) of same sub-ject; yet observe similar size of saccades followingcompletion of divergence, clearly demonstratingthat calibrations are accurate, b, Subject B. L. Asfound in Subject A. D., markedly unequal sac-cades occur during vergence response but not dur-ing fixation periods following, or prior to, the largevergence responses.

markedly unequal. For example, the saccadicpair 60 msec after the start of convergence inFig. 5 a (A. D.), had amplitudes of 3.1° and0.8° for the right and left eyes, respectively.The smaller saccade measured 26% of theamplitude of the larger saccade. The nextsaccadic pair was not as dramatically unequal.Also, the smaller saccade was now in the

right eye, whereas it appeared in the left eyefor the previous pair. This same pattern wasrepeated in the divergence response. Thesaccadic properties found during vergence inall subjects were best illustrated by the re-sponses of control subject B. L. (Fig. 5, b).The divergence response showed dramaticdifferences in the amplitudes of the first sac-cadic pair during vergence, with relative sac-cadic amplitude equaling 31%. For the firstpair, the smaller saccade was in the left eye.The following saccadic pair was not as un-equal as the first, and relative amplitudeequaled 52%. Also, the smaller saccade nowappeared in the right eye. For all subjects thesmaller saccade always occurred opposite tothe direction of the vergence movement ofthat eye, and the amount of inequality wasgenerally larger near the start of the vergencethan at the end.

Looking back through the responses of ourpatients, one finds that several showed thesetwo characteristics. As each saccadic pair isdiscussed, notice that the smaller saccadeopposes the vergence movement. During thefirst 150 msec of the vergence, the differencein the amplitudes of the two saccades wasusually greatest. In Fig. 2, a (J. K.), the sac-cade at the beginning of the divergence had44% of the amplitude of the fellow eye sac-cade. As the saccades occurred later in thevergence, the amplitude differences were notas great. In Fig. 1, b (B. N.), the saccade 600msec after the start of the divergence hadrelative amplitude of 67%. In Fig. 2, b(S. H.), the smaller saccade 300 msec afterthe start of the divergence equaled 60% ofthe amplitude of the saccade of the other eye.Finally, saccades following completion of thevergence response were equal. For example,saccades after the convergence in Fig. 5, a,prior to the vergence in Fig. 2, a, and be-tween the vergence responses in Fig. 1, b,were all equal. These equal saccades occur-ring after the vergences provide additionalevidence that saccades are indeed unequaland are not due to a nonlinearity in the eye-movement recording system.

To summarize the non-Hering's law sac-cade result, the relative saccadic amplitudes

Volume 19Number 1 Symmetric vergence in strabismus, amblyopia 69

140

120

* 100

8 0 -

< 6 0 -

< 40

20

a >ft • o

O DF6 AD0 BL

J I-500 0

STARTOF

VERGENCE

100 200 300 400 500 600 700 800 900 1000

TIME (ms)

Fig. 6. Relative amplitude of saccades in each eye as a function of time during vergenceresponse for seven patients and two control subjects. Plot clearly shows reduction in non-Hering's law saccades as they occur later in the vergence movement.

Table II. Percent relative vergence

Subject

Normal controlsubjects:B. L.A. D.

Constant strabismusamblyopia:D. F.

Intermittent strabismuswithout amblyopia:B. N.

Amblyopia withoutstrabismus:S. H.

Former strabismuspatient:J. B.

Symmetric vergence targets(binocular)

92.1% ± 389.0% ± 8

16.0% ± 4

7.6% ± 2

7.0% ± 2

11.0% ± 5

Symmetric vergence targets(nondominant eye covered)

26.0% ± 317.0% ± 5

15.0% ± 3

8.5% ± 4

9.0% ± 1

5.3% ± 2

Accommodative vergencealong dominant eye

(nondominant eye covered)

20.0% ± 514.5% ± 9

13.0% ± 5

8.0% ± 1

7.5% ± 3

6.0% ± 2

from all subjects during vergence are plottedin Fig. 6. The ordinate represents percentrelative amplitude, and the abscissa is timebefore and after the start of the vergencemovement. The relative amplitude is definedas the ratio, in percent, of the smaller saccadeto the larger saccade. The solid line at 100%represents the ratio expected if equal am-plitude saccades occurred in each eye. Thegraph shows that most saccadic pairs duringvergence had relative amplitudes of 60% or

lower. Furthermore, the lowest relativeamplitudes occurred early in the vergenceand then slowly increased until, at about 700msec, both saccades were approximatelyequal. Also notice that saccades beforethe vergence movement were approximatelyequal; it is only during the vergence move-ment that unequal saccades were found. Theplot also shows that not all saccades that oc-curred at a particular time after the start ofthe vergence were reduced by the same

70 Kenyon, Ciuffreda, and StarkInvest. OphthalmoL Vis. Sci.

January 1980

1000

ai 100

MAGNITUDE (DEG

Fig. 7. Main sequence for eye movements used toidentify normal saccades. The solid line representslocus of points for normal saccades. Symbols rep-resent saccades from normal subjects and PatientsJ. K. and B. N., during the vergence. The circles,squares, and triangles represent Control SubjectB. L. and Patients J. K. and B. N., respectively.Saccades cluster around the main sequence line,indicating that despite the inequality of the sac-cades they remain normal.

amount. Some of these differences were dueto intersubject variability, as can be seen bythe placement of the different symbols on thegraph. Still, some intrasubject variability wasevident when data from an individual subjectwere examined. Despite this scatter, the sac-cades were clearly unequal during the ver-gence in both subjects and patients. Al-though the gradual increases in saccadicequality have been plotted as a function ofthe timing between the saccade and ver-gence, it may also be related to the velocityof the ongoing vergence and the saccade.Further experiments will be needed to clarifythis relationship.

An analysis of these unequal saccades re-veals that they all have normal main se-

quence23 peak velocity-amplitude character-istics. The peak velocity-amplitude plot forsaccades for normals is represented by thesolid line in Fig. 7. The cluster of symbolsaround the normal saccadic line indicates thatour patients' saccades during vergence move-ments had normal dynamics.

Discussion

Absence of disparity vergence. Normaldisparity vergence eye movements, inducedby fixating near and far midline targets, arecharacterized by smooth, equal, disjunctivemovements of both eyes. The eye move-ments of our control subject (Fig. 1, a) illus-trate this characteristic response, well-knownin the literature on vergence eye move-ments.14"18 In contrast, under binocularviewing conditions where disparity vergenceshould be operating, most patients displayeda conspicuous lack of normal disparity ver-gence movements. A qualitative comparisonof the responses of the patients and the con-trol subjects supports this hypothesis.

Possible mechanisms blocking disparityvergence. The absence of disparity vergenceunder binocular viewing conditions indicatesthat a disparity blocking mechanism is oper-ating in all groups of patients. Suppression isthe most likely condition. Strabismus pa-tients with deviations greater than 10 PD, asfound in many of our patients, generally usesuppression to eliminate diplopia and confu-sion caused by the strabismus.24"28 The sizeof the suppression scotoma has been found tobe a function of the angle of deviation, en-compassing the region between and includ-ing the fovea and the diplopia point.24' 25

Suppression is also commonly found in pa-tients with amblyopia only and may explainthe lack of disparity vergence found in someof our amblyopic patients. Of particular in-terest is the intermittent nature of the ver-gence recorded in patients having amblyopiawithout strabismus. For example, PatientS. H. at times made normal disparity ver-gence, and L. T. always did so. These results,together with the absence of normal disparityvergence eye movements in strabismus pa-tients with and without amblyopia, suggest

Volume 19Number 1 Symmetric vergence in strabismus, amblyopia 71

that strabismus has a stronger effect onthe disparity vergence system than does am-blyopia.

Some of our patients with strabismus alsohad ARC. Interestingly, these patients alsoshowed an absence of disparity vergence.Jampolsky25 showed that suppression charac-teristics are dependent on stimuli conditions,whereas the results of Bagolini28 suggest thatwhen patients with ARC are allowed to viewfreely, the suppression found is usually min-imal. Bagolini29 has established, by measur-ing haplopia horopters, a markedly increasedPanum's area in such patients. Indeed, plas-ticity of Panum's area in normal subjects isdependent upon the nature of the stimuli andthe state of the feedback control system, ashas been demonstrated by Fender and Ju-lesz30 and Diner.31 Perhaps an expandedPanum's area could eliminate the need forvergence and thereby provide a possiblemechanism underlying the absence of dis-parity vergence in our patients with ARC.

For our patients J. B. and J. W. who have ahistory of strabismus, and B. W. who has ahistory of amblyopia, the lack of disparityvergence movements suggests some connec-tion between their former abnormal clinicalcondition and their vergence responses.Here, the important studies of single corticalunits in animals with experimental strabis-mus and amblyopia may provide importantclues toward the understanding of humanstrabismus and amblyopia. These reportsshow that if strabismus and amblyopia areproduced early in life, the number of corticalcells that respond to disparity or binocularstimuli is much less than that usually found innormal animals.32"36 If this abnormal visualexperience is continued throughout the ani-mal's "sensitive period," the response pat-terns of the cortical cells are difficult to mod-ify later.37 Perhaps, in the case of our PatientsJ. B., J. W., and B. W., their abnormal earlyvisual experince may have affected cells thatrespond to disparity to such an extent thatthese cells no longer respond normally todisparity information. This also may be anadditional mechanism operating in patientswho still have strabismus and/or amblyopia.

If, in the future, animal experiments involv-ing induced strabismus and/or amblyopiavergence eye movements were measured,this would provide an additional criterion fortheir being adequate models of human stra-bismus and amblyopia and would also corre-late these hypotheses with neurophysiologi-cal experiments.

Accommodative vergence. Several piecesof evidence strongly suggest that these un-equal vergence responses found in our pa-tients are not disparity vergence but ratheraccommodative vergence. Accommodativevergence studies on normal subjects show anasymmetry in the amplitudes of the vergencein each eye. Miiller19 first described the largevergence in the covered eye, and we9 haverecently shown that the viewing eye vergesapproximately 12% of the amplitude mea-sured in the covered eye. The responses ofControl Subject A. D. (Fig. 4, a) demonstratethis characteristic accommodative vergencemovement. Similar characteristics are alsofound in the binocular viewing response ofour patients indicating that these unequalvergence movements are accommodative ver-gence. A test of this hypothesis, performedby covering the patient's nondominant eye,showed that the response under monocu-lar conditions contained the same generalcharacteristics as found in these same pa-tients under binocular viewing conditions.Furthermore, when one eye of a normal sub-ject was occluded, responses were similar tothose of patients under binocular viewingconditions. Finally, the accommodative ver-gence responses of our patients showed nor-mal characteristics similar to the characteris-tics of our control subjects. Table II summa-rizes the relation between the vergenceamplitudes in each eye for control subjectsand one patient from each category underbinocular and monocular conditions. There isa clear difference in the relative amplitudesduring binocular conditions between the con-trol subjects and the patients. Under mon-ocular conditions, the controls show a largedifference between the monocular and thebinocular condition, but patients show littledifference in relative vergence under each

72 Kenyon, Ciuffreda, and Stark Invest. Ophthalmol. Vis. Sri.January 1980

condition. Also the relative vergences underboth monocular and binocular conditions forpatients are similar to those of the controlsunder monocular conditions. Therefore weconclude that patients in our study madenormal accommodative vergence and notdisparity vergence under binocular viewingconditions. Blake et al.38 speculated that con-vergence changes observed in his kittensreared with alternating monocular occlusionmight be due to binocular disparity, changesin image size, or changes in accommodation;in our human subjects we find that accommo-dative vergence is the principal mechanismresponsible for the vergence movement.

Non-Hering's law saccades. One of themost interesting new findings of our studywas the presence of non-Hering's law sac-cades during vergence. Non-Hering's lawsaccades were present in at least 85% of thevergence movements found in all diagnosticgroups of patients. We were alerted to thispossibility because of our previous finding ofnon-Hering's law saccades during accomo-dative vergence in normal subjects.9 Aspointed out in Results, the smaller saccadeopposed direction of the ongoing vergence;the direction of the saccades is determinedby the retinal error in the dominant eye. Theproperties of a smaller saccade opposing thevergence and a smaller vergence in the dom-inant eye combine, so that the inequalitiescontribute to the amplitudes of the ongoingconvergence or divergence. For example, inresponse to convergence stimuli, a patientwith strabismus of the left eye will use a sac-cade to place the dominant right eye onto thenear target. A leftward saccade in both eyeswill move the right eye on the target and theleft eye away from the target. The smallersaccade opposing the vergence causes the lefteye to move less than the right; the result isthat both eyes are converged more after thesaccade than before. The addition of a largervergence in the left eye than the right keepsthe dominant right eye on the target whilemoving the left eye towards the target. Thusthe saccade carries about half the burden ofthe vergence movement, and the accom-modative vergence does the other half.

The inequalities of these saccades are notrestricted to patients,9' 39 since non-Hering'slaw saccades appear during vergence move-ments in normals (Fig. 5). In control sub-jects, the saccades occur sparsely under bin-ocular conditions, usually during divergence.In patients, the saccades are present in al-most all vergences. Only when the controlsubjects occlude one eye does the saccadeoccur with the regularity of that found in pa-tients. In the control subjects, the saccadesare due to the monocular retinal error in theviewing eye. The occurrence of a similarfunction in most patients may be construed asevidence for suppression of the nondominanteye in these patients.

Possible mechanisms for non—Hering's lawsaccades. Knowledge of the fine structure ofsaccades may provide a clue to the underly-ing physiological mechanism responsible forproduction of unequal saccades. We haveruled out simple linear summation of the sac-cade with the vergence movement. If simplelinear summation of saccadic and vergencepositions were responsible for a change in thesaccadic magnitudes, then the amplitude dif-ferences should be equal to the additionalvergence amplitude the eye moved duringthe saccade. Compensating the saccadic am-plitudes for the vergence during the saccadeshould yield equal saccadic pairs. However,saccades compensated for the vergencemovement still have amplitudes that are 70%of their uncompensated value. This small re-duction is due to the short duration of thesaccade, circa 25 msec, and the slow velocityof the vergence, circa 20°/sec. The combina-tion results in little additional eye move-ment. In addition, we chose to analyze sac-cades that are large enough so that theamount of additional amplitude due to thevergence is insignificant.

Nonlinear summation in the muscle orin the motoneuronal pool is possible butunlikely, since the smaller saccades haveshorter duration than the larger saccades oc-curring concomitantly in the other eye andthus may be already programmed to besmaller at a higher interneuronal level. Al-ternatively, it may be a complex interaction

Volume 19Number 1 Symmetric vergence in strabismus, amhlyopia 73

of muscle forces, viscosities, elasticities, ac-celerations, velocities, and positions sepa-rately controlled or influenced by vergenceand saccadic neural control signals. In addi-tion, the main sequence analysis shows thatthese saccades are not abnormally fast or slowwith respect to saccades made free from ver-gence influences. The peak velocity and am-plitudes correspond well with the normalmain sequence measurements. Therefore anymechanism which will reduce or enhancethese saccades must equally affect the peakvelocity as well as the amplitude. Furtherexperimental, simulation, and neurologicalstudies will be necessary to clarify the gen-eration of the non-Hering's law saccades.

Summary. Vergence eye movements inmost patients with strabismus and/or am-blyopia are restricted to accommodative ver-gence movements. The characteristic re-sponse to disparity produced by targets at dif-ferent distances involves binolcular saccadeswhich fixate the dominant eye on the targettogether with an accommodative vergence.Saccades of both patients and control subjectsthat occur during vergence are unequal, withthe saccade moving against vergence beingthe smaller. Main sequence analysis showsthat both equal and unequal saccades havenormal peak velocity-magnitude relation-ship. Strabismus appears to have a strongerdisruptive effect on normal disparity ver-gence than does amblyopia; surgery and or-thoptics to relieve strabismus and amblyopiahave mixed effects on vergence responses.

We are grateful to Dr. Kenneth Poise, Clinic Direc-tor, and Dr. J. David Grisham for their assistance inobtaining patients for this study. We also thank Dr. AllanN. Freid for his helpful suggestions.

R E F E R E N C E S

1. Burian, H.M., and von Noorden, G.K.: BinocularVision and Ocular Motility, St. Louis, 1974, TheC. V. Mosby Co.

2. Burian, H.M.: Fusional movements in permanentstrabismus; a study of the role of central and periph-eral retinal regions in the act of binocular vision insquint, Arch. Ophthalmol. 26:626, 1941.

3. Bielschowsky, A.: Untersuchungen iiber das Sehender Schielender, Graefes Arch. Ophthalmol. 50:406, 1900; cited in Burian and von Noorden.'

4. Schlodtmann, W.: Studien iiber anomale Sehrich-

tungsgemunschaft bei Schielenden, Graefes Arch.Ophthalmol. 51:256, 1900; cited in Burian and vonNoorden.'

5. Hallden, U.: Fusional phenomena in anomalous cor-respondence, Acta Ophthalmol. 37(suppl. 1): 1952.

6. Mariani, G., and Pasino, L.: Variations in the angleof anomaly and fusional movements in cases of smallangle convergent strabismus with harmonious anom-alous correspondence, Br. J. Ophthalmol. 48:439,1974.

7. Kretschemer, S.: La fausse correspondence re-tinienne, Doc. Ophthalmol. 9:46, 1955; cited inBurian and von Noorden.'

8. Stark, L., Vossious, G., and Young, L.: Predictivecontrol of eye tracking movements, IRE Trans.Human Fac. Electr., HFE-3:52, Sept., 1962.

9. Kenyon, R.V., Ciuffreda, K.J., and Stark, L.: Bin-ocular eye movements during accommodative ver-gence, Vision Res. 18:545, 1978.

10. Krishnan, V.V., and Stark, L.: A heuristic model forthe human vergence eye movement system, IEEETrans. Biomed. Eng., BME-24 1:44, 1977.

11. Fry, G.A., and Hill, W.W.: The center of rotation ofthe eye, Am. J. Optom. Arch. Am. Acad. Optom.39:581, 1962.

12. Ciuffreda, K.J., Kenyon, R.V., and Stark, L.: Ab-normal saccadic substitution during small-amplitudepursuit tracking in amblyopic eyes, INVEST. OPH-THALMOL. VISUAL SCI. 18:506, 1979.

13. Griffin, J.E.: Binocular anomalies: Procedures forVision Therapy, Chicago, 1976, Professional PressInc., pp. 74-75, 95-96.

14. Westheimer, G., and Mitchell, A.M.: Eye move-ment responses to convergent stimuli, Arch Oph-thalmol. 55:848, 1956.

15. Riggs, L.A., and Niehl, E.W.: Eye movements re-corded during convergence and divergence, J. Opt.Soc. Am. 50:913, 1960.

16. Rashbass, C., and Westheimer, G.: Disjunctive eyemovements, J. Physiol. 159:339, 1961.

17. Zuber, B.L., and Stark, L.: Dynamical characteris-tics of fusional vergence eye-movement system,IEEE Trans. System Sci. Cyb., SSC 4 1:72, 1968.

18. Westheimer, G., and Mitchell, D.: Sensory stimu-lus for disjunctive eye movements, Vision Res.9:749, 1969.

19. Miiller, J.: Elements of Physiology, trans, byW. Baly, London, 1843, Taylor & Walton, vol. 2,pp. 1147-1148.

20. Ciuffreda, K.J., Kenyon, R.V., and Stark, L.: Pro-cessing delays in amblyopic eyes: evidence fromsaccadic latencies, Am. J. Optom. Physiol. Opt.55:187, 1978.

21. Ciuffreda, K.J., Kenyon, R.V., and Stark, L.: In-creased saccadic latencies in amblyopic eyes, IN-VEST. OPHTHALMOL. VISUAL SCI. 17:697, 1978.

22. Ciuffreda, K.J., Kenyon, R.V., and Stark, L.: Dif-ferent rates of functional recovery of eye movementsduring orthoptics treatment in an adult amblyope,INVEST. OPHTHALMOL. VISUAL SCI. 18:213, 1979.

74 Kenyon, Ciuffreda, and StarkInvest. Ophthalmol. Vis. Sri.

January 1980

23. Bahill, A.T., Clark, M.C., and Stark, L.: The mainsequence, a tool for studying human eye move-ments, Math. Biosci. 24:191, 1975.

24. Travers, T.: Suppression of vision in squint and itsassociation with retinal correspondence and ambly-opia, Br. J. Ophthalmol. 22:577, 1938.

25. Jampolsky, A.: Characteristics of suppression instrabismus, Arch. Ophthalmol. 54:683, 1955.

26. Pratt-John son, J., Wee, H.S., and Ellis, S.: Sup-pression associated with esotropia, Can. J. Oph-thalmol. 4:284, 1967.

27. Pratt-Johnson, J., and Wee, H.S.: Suppression as-sociated with exotropia, Can. J. Ophthalmol. 2:136,1969.

28. Bagolini, B.: Sensory anomalies in strabismus. Doc.Ophthalmol. 41:1, 1976.

29. Bagolini, B.: Anomalous correspondence: definitionand diagnostic methods, Doc. Ophthalmol. 23:346,1967.

30. Fender, D., and Julesz, B.: Extension of Panum'sfusional area in binocularly stabilized vision, J. Opt.Soc. Am. 57:819, 1967.

31. Diner, D.: Hysteresis in human binocular fusion: asecond look, Dissertation, California Institute ofTechnology, 1978.

32. Hubel, D., and Wiesel, T. N.: Binocular interactionsin striate cortex of kittens reared with artificialsquint, J. Neurophysiol. 28:1041, 1965.

33. Ikeda, H., and Wright, M.J.: Is amblyopia due toinappropriate stimulation of the sustained pathwaysduring development? Br. J. Ophthalmol. 58:165,1974.

34. Ikeda, H., and Tremain, K.E.: Different causes foramblyopia and loss of binocularity in squinting kit-tens, J. Physiol. 299:26P, 1977.

35. Baker, F.H., Grigg, P., and von Noorden, G.K.:Effects of visual deprivation and strabismus on theresponse of neurons in the visual cortex of monkeys,including studies on the striate and prestriate cortexin the normal animal, Brain Res. 66:185, 1974.

36. Eggers, H., and Blakemore, C.: Physiological basisof anisometropic amblyopia, Science 201:264, 1978.

37. Blakemore, C , and Van Sluyters, R.C.: Reversal ofthe physiological effects of monocular deprivation inkittens: further evidence for a sensitive, period, J.Physiol. 237:195, 1978.

38. Blake, R., Crawford, M., and Hirsch, H.: Conse-quences of alternating monocular deprivation of eyealignment and convergence in cats, INVEST. OPH-THALMOL. 13:121, 1974.

39. Ono, H., Nakamizo, S., and Steinbach, M.: Non-additivity of vergence and saccadic eye movements,Vision Res. 18:735, 1978.