REPORT RESUMESED 012 120 EC 000 106

VISION SCREENING OF THE PRESCHOOL CHILD, REPORT OF A STUDY.

BY SAVITZ, ROBERTA A. AND OTHERSCHILDRENS BUREAU, WELFARE ADMIN., WASHINGTON, D.C.REPORT NUMBER CB PUG - 414 -1964 PUB DATE 64

EDRS PRICE MF$0.18 HC$3.20 80P.

DESCRIPTORS *VISION, *VISION TESTS, *PRESCHOOL CHILDREN,SCREENING TESTS, TESTING, PRESCHOOL TESTS, *HANDICAPDETECTION, VISUALLY HANDICAPPED, VISUAL DISCRIMINATION,OSTERCERG CHART, ALLEN PICTURE CARDS, AMERICAN OPTICALKINDERGARTEN CHART, STYCAR VISION TEST, DISTRICT OF COLUMBIA

USING A SAMPLE OF 94 CHILDREN (31 TO 54 MONTHS OLD),THIS STUDY COMPARED EIGHT VISION SCREENING PROCEDURES FOR

YOUNG CHILDREN IN THE ABILITY TO TEST FOR SEVERAL VISUALFUNCTIONS; AND PREFERENCE AMONG THEM BY CHILDREN, THESUBJECTS WERE ORIGINALLY TESTED USING THE EIGHT SCREENINGTESTS, AND 6 MONTHS LATER 40 OF THE CHILDREN WERE RETESTED

FOR CHANGES IN VISUAL ACUITY AND EYE DOMINANCE DURING THEINTERIM PERIOD. RESULTS INDICATED THAT THE RELATIVE SCREENING

ABILITY OF THE FROCEDURES WAS UNDETERMINED FOR THE VISUAL

FUNCTIONS OF VISUAL ACUITY, MUSCLE BALANCE, AND COLORPREFERENCE, DUE TO NONTESTABILITY OF SIGNIFICANT NUMBERS OF

THE SUBJECTS. THE RESULTS INDICATED THAT EYE DOMINANCE COULD

BE ESTABLISHED. THE CONCLUSION SUGGESTS THAT PRESCHOOL

CHILDREN 30 MONTHS OF AGE AND OVER CAN BE SCREENED, ALTHOUGH

50 PERCEt.T MAY BE NONSTABLE. THIS DOCUMENT IS AVAILABLE FROM

THE SUPERINTENDENT OF DOCUMENTS, U.S. GOVERNMENT PRINTING

OFFICE, WASHINGTON, D.C. 20402, FOR $0.45. (KH)

a

U.S. DEPARTMENT OF HEALTH, EDUCATION & WELFARE

OFFICE OF EDUCATION

THIS DOCUMENT OAS BEEN REPRODUCED EXACTLY AS RECEIVED FROM THE

PERSON OR ORGANliATION ORIGINATING IT. POINTS OF VIEW OR OPINIONS

STATED DO NOT NECESSARILY REPRESENT OFFICIAL OFFICE OF EDUCATION

POSITION OR POLICY.

Page 6.

Page 7.

Page 15.

ERRATAPicture caption should read

"Allen Picture Cards."Picture caption should read

"Osterberg Chart."Line 10, right column, should read

"project at 962 Parker Street."

vision

screening

of the

preschool

ChildREPORT OF A STUDY

ROBERTA A. SAVITZ, M.D.

ROBERT B. REED, Ph.D.

ISABELLE VALADIAN, M.D., M.P.H.

Departments of Biostatistics and Maternal and Child Health, Harvard School of Public Health

U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE

WELFARE ADMINISTRATION Children's Bureau 1964

For sale by the Superintendent of Documents, U.S. Government Printing Office,Washington, D.C. 20402. Price 45 cents.

contents

I. INTRODUCTION ..

II. MATERIALS AND METHODS:

1 Findings on QuestionnaireComposite Test Findings: Test-

ability and Nontestability.

34

36Background and Description. 4 Visual Acuity 36

Tests 4 Muscle Balance 39Visual Acuity"Stycar".. . . 4 Performance 39Visual AcuityNaming pic- Dominance 40

tures and toys 6 Color 40Visual AcuityIndicating di- Composite Test Findings: Pass-

rections 7 ing and Failing 42Visual AcuityEliciting Opti-

cokinetic NystagmusVisual Acuity"Stycar"..Visual AcuityNaming pic-

42

(OKN) 8 tutes and toys 45Muscle Balance 8 Visual A cuityIndicating di-Performance 9 rections 45Dominance 11 Visual AcuityElicitingColor 12 OKN 45

Questionnaire 12 Muscle Balance 46Performance 46III. THE SUBJECTS 14 Dominance 46

IV. OPERATING PROCEDURES.. 17 ColorQuestionnaire

4647

V. FINDINGSTestability and Nontestability

on Individual Tests

19

19

VI. FOLLOWUP OF SUBOPTIMALFINDINGS 48

Visual Acuity"Stycar".. . 20 Children Recommended for Fol-

Visual AcuityNaming pic- lowup 48tures and toys 22 Retests 51

Visual AcuityIndicating di-rections 24 VII. DISCUSSION 55

Visual AcuityEliciting Tests 55

OKN 26 Visual Acuity"Stycgr".. 55Muscle Balance 27 Visual AcuityNaming pic-Performance 29 tures and toys 57Dominance 32 Visual AcuityIndicating di-Color 34 rections 57

Visual AcuityElicitingOKN

Muscle BalancePerformanceDominanceColor

QuestionnaireReferrals

VIII. CONCLUSIONS AND RECOM-MENDATIONS

Possible Program for Immedi-iate Use

Age of children to be screenedTestsPlace

Major Problems of PreschoolVision Screening Programs

TestsFollowup

Areas for Further Investigation .

Objective vision testsNormal visual development.Current preschool screening

programs

tables

1. Summary of Definitions of Testabil-ity

2. Testable Children, Totals for EachTest

3. Visual Acuity"Stycar." Testableand Nontestable Children.Breakdown by Sex, Race, andAge

4. Occluder Rejection by Sex and Age_5. Visual AcuityNaming Pictures

and Toys. Testable and Non-testable Children. Break-down by Sex, Race, and Age_

6. Visual AcuityIndicating Direc-tions. Testable and Nontest-able Children. Breakdownby Sex, Race, and Age

7. Visual AcuityEliciting OKN.Testable and Nontestable Chil-dren. Breakdown by Sex,Race, Age, and Type of Re-sponse

8. Muscle Balance. Testable and Non-58 testable Children. Break-58 down by Sex, Race, and Age__ 2859 9. Stereoscope : Patterns of Response.60 Breakdown by Sex, Race, and61 Age 2961 10. Performance. Testable and Non-62 testable Children. Break-

down by Sex, Race, and Age 3011. Dominance. Testable and Nontest-

64 able Children. Breakdown bySex, Race, and Age 31

64 12. Patterns of Dominance. Break-64 down by Sex, Race, and Age 3364 13. Color. Testable and Nontestable64 Chi I d r e n. Breakdown by

Sex, Race, Age, and Type ofResponse 33

14. Testability of 19 Children with Pos-sible Eye Problem by History.Breakdown by Sex, Race, andAge 34

15. Testability of 57 Children with Pos-66 itive Family History of Eye

Problem s. Breakdown bySex, Race, and Age ' 35

16. Testability of 10 Premature Chil-dren. Breakdown by Sex,Race, and Age 36

17. Comparison of Ability to Match Let-ters and Directions. Break-down by Sex, Race, and Age 37

18. OKN and Miniature ToysCompos-ite Testability. Breakdown bySex, Race, and Age 39

23 19. Nearp..)int of Convergence DeviatingEye versus Eye Dominance bySighting. Breakdown by Sex,Race, and Age 40

23 20. Performance Tests : Composite Rat-ings. Breakdown by Sex,Race, and Age 41

21. Eye Preference as an Indication of26 Testability. Breakdown by

Sex, Race, and Age 4122. Passing Criteria for Tests 4223. Total Testable Children Passing and.

Failing each Test and Ques-27 tionnaire 43

656565656565

19

20

21

1 I

24. Composite Failures on Tests andQuestionnaire, by Age

25. Failures on Miniature Toy Test.Breakdown by Sex, Race, Age,and Other Test Failure

26. Basis of Recommendation for Retestfor 49 Children. According toSex, Race, and Age

27. Retests of 12 Children with Poten-tial Amblyopia. According toSex, Race, and Age

28. Retests of 25 Immature and Incom-pletely Test Able Children.Findings According to Sex,Race, and Age

29. Eye Preference as an Indication ofTestability. Supplement toTable 21, based on Nontestable

44

45

50

51

Children, Table 28. FindingsAccording to Sex, Race, andAge

30. Rate of Amblyopia in Vision Screen-ing Surveys

Figures

53

62

1. Age Distribution of Subjects by Sexand Race 16

2. Correct Responses to Direction Test52 (Hand) by Position, by Age

and Sex 253. Comparative Testability by Age for

Various Visual Acuity Tests_ 38

I. INTRODUCTION

This study was undertaken after an ex-tensive review of the available literature onpreschool vision. This is a body of literaturewhich is often vague and inconsistent andleaves many questions unanswered. The pres-ent study is descriptive and limited to a smallnumber of subjects. In order to define the pres-ent work within the literature and within thetotal subject of preschool vision screening, somebasic questions are posed and discussed below.

.What are the vision problems ofpreschool children?

The vision problem of the preschoolchild which has been the major impetus to thedevelopment of screening procedures is am-

.

blyopia ex anopsia (unilateral amblyopia ofdisuse, "lazy eye blindness"). This is con-sidered a problem which cannot await laterdetection and treatment. Refractive errors andheterophorias in this age group have not beenwell-defined and generally screening programshave only aimed indirectly at detection of theseproblems by detecting their consequences likelow vision.1-9

Even less well-defined are general devel-opmental problems involving visionproblemswhich might bear some direct relationship tolater difficulty, physical and educational. Thereis growing interest in normal visual develop-ment and the consequences of deviations fromthis normal. The Gesell Institute of Child De-velopment has given considerable attention tothis subject, and work in this area is being

pursued by some optometrists, educators, andpsychologists. 10-12 Medical and public healthpractitioners have not yet directBd particularattention to this area in an organized way.

What are the vision tests availablefor preschool children which might beadapted for screening purposes?

Most of the vision tests recommendedfor children aged 5 years and younger are testsof distant visual acuity. These tests detect lowvision, and the child is then given a professionalexamination to determine diagnosis and treat-ment.

Vision tests available for infants arebased upon the elicitation of reflexes which, bytheir presence, indicate that visual pathwaysare intact.13-18 The newborn is not blind if hedemonstrates either pupillary reaction to light,the constriction of the pupil when light isshown into either eye, or vestibular reaction, theslow following of the eyes in the direction theinfant is turned. The visual acuity of the new-born has been estimated by production ofopticokinetic nystagmus, the response to seeinga moving series of lines. After 1 month, thepresence of vision may be demonstrated byoptically elicited movement, the movement ofthe eyes toward a peripherally appearing object.Experimentally, devices based on this reflexand using graded stimuli have been used tomeasure the vision present. These objectivevision tests, however, seem to underestimatevision when the results are compared to sub-jective tests.

1

Visual acuity of the child from 10 monthsto 2 years has been estimated with various sub-jective tests such as Worth's marble balls 29' 20 andBock's Candy Bead 74 eSt.21-24 Ambulatory in-fants and toddlers are expected to retrievesmall, standard-sized objects from varying dis-tance using each eye in turn.

Communicative preschool children areusually requested to give a subjective response,to indicate in some way that the test object isperceived. In picture tests, the child names orshows a similar picture or object to the exam-iners 29-36. In direction tests, the child names orindicates with his hand the direction of the testobject 21-40. Several types of visual acuity testswere included in this study, and these will bedescribed in de.; il in Section II.

Tests of eye muscle balance are those usedfor adults; 41-43 techniques of administrationand superficial apparatus changes make thesetests more suitable for children.

rsion test batteries for preschool chil-dren have been established to detect abnormali-ties of vision as they relate to general develop-ment.12 Several of these functionally orientedtests, e.g. tests of performance and eye domi-nance, were included in this study and againwill be described dater.

Whatever vision tests are suitable forscreening must be fairly nontechnical. Sub-jective vision tests, however, are also part ofthe professional examination; they cannot bereplaced by the cycloplegic examination whichdoes not serve the same purpose of indicatingwhat use the child is making of his eyes.

Have the available vision tests beenused for testing large numbers of childrenand adapted for screening purposes?

Several of the visual acuity tests, e.g."Stycar", Allen pictures,25-27 and Osterbergchart,28 have been used over a period of yearsfor testii-g many children. The Snellen Etest 39-49 is the only test so far which has beenstandardized for use by nonprofessional per-sonnel as a screening too1.44 Most acuity testshave never been used in controlled settings byimpartial observers.

Other vision tests are administered tolarge groups of children, but not necessarily in

2

the context of an easily reproduced screeningprogram. For instance, British orthoptistsgive routine cover tests to children enteringschool; 45 there are not enough profesLionalorthoptists in this country to establish a similarprogram.

The functional vision tests 10.12 have beengiven to many children as individuals on an ex-perimental basis. These have not been adaptedfor screening.

How reliable are these tests whenadministered at different times, by differ-ent personnel, in different places?

Carefully controlled studies of the reli-ability of the available preschool vision testshave not been done. The reliability of schoolvision tests has been investigated more thor-oughly," but this information cannot begeneralized.

In most screening programs a certainlack of reproducibility is expected and childrenare retested under new circunrtances beforereferrals are made.

How valid are the results of thescreening tests when compared to profes-sional findings on the same child?

Any screening test should bear some defi-nite relationship to a more refined procedurefrom which it is derived. The referrals fromvision screening tests should, for the most part,be those children whom professional eye doctorswould select for specialized treatment if theyhad examined all the children screened. Thisis a particular problem in vision testing wherethe number of concerned professions and thevariation within each profession make the defi-nition of a correct referral very difficult. Thisvariation in standards of referral was clearlydemonstrated in the study by Lancaster et al.47Variation in the findings of even a single pro-fessional examiner is demonstrated by Sloane.48In the extreme situation, one person might beconsidered a needless referr41 by one profes-sional consultant, and a correct referral byanother. When standards are set arbitrarilyfor the purposes of a screening program, thevalidity of the screening procedures can be de-

termined. This determination has not beenmade completely for a large population of pre-school children. In general, clinical follow-upof referrals has been inadequate, and the num-ber of under-referrals has not been determined.

What is the expenditure of time andmoney involved in screening preschoolchildren?

There is no definite answer to this ques-tion to be found in the literature. Only whenthere is agreement about screening procedures,can the expense incurred by such programs bedetermined. Then, screening expenditure willhave to be justified by proven vision conserva-tion. This will be the subject of further inves-tigation.

At what age are preschool visiontests applicable?

The vast majority of authors state thatage 3 years is the optimal time to screen foramblyopia. One reason offered is that 3 yearsis when retinal reflexes are established and thatdisuse at this age frequently precedes unilateralamblyopia.4 Another pragmatic reason to testat age 3 is because subjective t,ests first becomepossible at this time. H- wever, the most fre-quently employed, and the best standardized"preschool test", the illiterate E, is frequentlynot successful before age 4 years.5°

Variations on the E continue to appear

735-670 0-65-2

as "new tests", but the basic principle, a sub-jective response to direction, is unchanged andyoung 3-year-olds are still nontestable. TheSnellen E is an excellent acuity test for anolder child or illiterate adult, but may be tooabstract for the young child. Pictures of toyanimals, on the other hand, are not too interest-ing to adults and are very appealing to pre-school children.

Age-appropriateness of vision tests war-rants some attention. If age is considered animportant factor in prognosis, then perhapschildren 3 years old and even younger shouldbe screened; this would necessitate the de-velopment and standardization of some testingmethod for this age group. If children 4 yearsold can be as successfully treated as youngerchildren, then there is no need to develop newmethods for testing visual acuity. Since thereseems to be general agreement that prognosis isrelated to age and children should be tested asyoung as possible, the subjects chosen for thisstudy were around 3 years old.

The aims of the present study were:

(I) To test several visual functions withavailable procedures.

(2) To clarify the abilities and prefer-ences of young preschool childrenwith respect to these procedures.

(3) To detect vision problems in the pop-ulation studied.

3

II. MATERIALS AND METHODS:

background and description

Tests

Visual acuity"Stycar"Screening Tests for Young Children and

Retardates or "Stycar" 29.32 is a set of acuitytests designed by Sheridan (a medical officer)and Pugmire (an ophthalmologist) , which arenot well known in this country, but have beenused successfully in Britain for almost 25 years.These tests were designed, unlike previoustests, with facts of normal child developmentin mind. In addition to high validity forvisual acuity, they were to provide informationabout the child's meaningful visual abilities,those involved in comprehending the environ-ment. The tests were to be applied to youngchildren or handicapped children including thepartially sighted, deaf, cerebral palsied andmentally handicapped.

Matching letter test. For the older pre-school child (3 years on), a letter test is -used;the child matches the letter shown to him withthe same letter on a key card he holds in hislap or reproduces the letter seen by drawing it.In this study, children were sometimes askedto cover the letters with pentiles instead ofpointing; this technique was used in a Lottoacuity test in France,33 which is basically likethe Stycar letter test. Singl, letters are pre-sented to 3-5 year olds; a letter chart is used forchildren more than 5 years old. The youngest

4

group, about 36-48 months, match 5 letters(0, X, V, T, H), presumably easily recog-nized because of shapes that can be drawn andare frequently used in mass media. When fiveletters are there are five possible responses,or four if V and X are confused; thus the samenumber of choices is available as with the Ewhen direction sense is well developed. Olderchildren 5-9 years match seven letters (A and Uadded) or nine letters (C and L added ).

Near letter test. A near vision test (12-14 inches) is included for those children whomay have defects at near but test normally ondistant tests, and for those whose maturity isnot great enoagh to connuct a test at any dis-tance. Acuity is graded from N36 to N6, Aboutequivalent to type sizes 36 point to 6 point.

Distant letter test. This test is adminis-tered at feet; with the ye- --ger child the dis-tance is established without loss of contact bytesting in a mirror placed 10 feet away, permit-ting examiner to sit with the child. Acuity isgraded from 6/36 to 6/4.

Miniature toy test. The Miniature ToyTest is used for children between 20 and 36

months. The child receives a set of 10 toys con-sisting of familiar objects and is asked to namethem if he can; these include a car (2-inch,colored plastic), plane (2-inch, metal), dell (2-inch, plastic with colored cotton dress), chair

---" ,Of44", 71""r;

M,

4.y tt-

MINIATURE TOYS AND MASK OCCLUDER

(2 x 1 inch, colored wood), knife (31/2 -inch,

meta:), fork (31/2-inch, prongs %o inch,metal), spoon (31/2-inch, bowl 1 x 3/4 inch,metal), small knife (21/1-inch, metal), smaller'fork (21/4-inch, 4 prongs yid inch apart,metal), smaller spoon 21/4 inch, 3/4 x 1/2 inchbowl, metal) .*

He then must match or name his toys toduplicates shown him by the examiner on a dullblack background 10 feet away. This is donefirst with both eyes, then with each eye. Thereis no numerical Snellen equivalent determined

These measurements are based on the set of toys used in this

study, and differ slightly from those stated in the test manual.

".

No.

for this test. It is, rather, a comparison of thetwo eyes. The best acuity detectable is theability to distinguish the small fork from thesmall spoon, and this is said to approximate6/6. A comparable near test for these youngchildren is their ability to detect small threadsand toys close at hand (about 2u inches).

In this study, black cloth Hallowe'enmasks were used as occluders. An eye on each

was covered with adhesive tape and black paperto serve as an occluder for one eye while theother peeked through a 1 x 11/2-inch. hole.

There was no consistent occlusion of one eyefirst, but the order of occlusion was noted.

The Stycar Vision Test is distributed by

the National Foundation for Educational Re-

5

%. 11:1.4As

OSTERBERG CHART

search in England and Wales. The tests werenot originally designed for large scale routinescreening, but rather they were to be given by amedical examiner or psychologist.* They canand hL .-e been easily given by nonprofessionalpersonnel, but in such situations the authorsadvise over-referring (any doubtful cases) andfrequent checks on the testers by the concernedmedical officer.

Visual acuity --- Naming pictures andtoys

Picture tests. Several commonly em-ployed visual acuity tests are based on theability of the preschool child to recognize some-what schematically drawn black pictures on awhite background. These tests depend fortheir success upon the child's verbal ability; rec-ognition is indicated by an appropriate andconsistently used name. Many pictures arethings one might not expect a child to recognizebecause of lack of experience with similarthings; others seem to be poor reproductions ofthe actual object.

In this study, the children were askedto identify pictures in three different pictureseries, usually only at reading distance and withboth eyes open : The Osterberg chart 28 (4th and5th editions made by Nyrop and Maag, Copen-hagen, Denmark) consists of pictures inspiredby children's drawings, drawn in black on awhite background to subtend a visual angle ofone minute at appropriate distance, and found

*The test kit was obtained through Dr. Lenin Baler of the Har-vard School of Public Health, who provided supervision in the useof "Stycar" materials.

6

recognizable with increasing distance, are usedfor his chart. Figures included were recognizedby at least 80 percent of Danish children testedbetween ages 2 and 5 years. Figures includeswan (6/60), steamboat and house (6/36), man,car, and horse, or horse, train, and skeleton key(C/24) ; skeleton key, horse and wagon, scissorsand boat or swan, steamboat, horse and wagon(6/18) ; house, horse and wagon, man, plane,Christmas tree, man on bicycle--or man on bicy-cle, Christmas tree, horse and -wagon, house,man (6/12) ; scissors, swan, man on bicycle,skeleton key, house, car, horse and wagon, orman, cup, scissors, swan, skeleton key, man onbicycle (6/9) ; horse, plane, Christmas tree,house, steamboat, scissors and boat (6/6) ; man,skeleton key, house, horse and wagon, plane,swan, scissors, car and steamboat, or house, scis-sors, cup, Christmas tree, horse and wagon,skeleton key and man (6/4). The StandardWall Chart was shown and also a photographicreduction (1 x 11/2-inches) of this chart. Thissmall chart, obtained from Dr. Trygve Gun-dersen, is a near vision test; the figures are com-parable in size to the standard Jaegar types.

The Allen picture cards 25 are originaldrawings borrowed from Massachusetts Eye &Ear Infirmary Ocular Motility Clinic. Thesearti bound in a booklet, unlike the single cardsof commercially produced test, but could bepresented singly. Pictures included were :teddy bear, plant, flower, clown, birthday cake,Xmas tree, steamboat, telephone, and horse withrider. The test now is available from Ophthal-mix, LaGrange, Ill.; it includes a car, and theplant, flower, and clown have been omitted.Cards measure 4 x 4 inches and pictures arewithin an area of 1 square inch, approximately

4 gw

A r * a., e, 41,

ALLEN PICTURE CARDS

the size which subtends a 1-minute arc whenviewed at 30 feet.

The American Optical Kindergarten Chart(1946) consists of pictures which include a sail-boat (20/200) ; circle and cross (20/100) ; flag,star, and heart (20/70) ; heart, cross, and sail-boat (20/50) ; star, moon, circle, and flag(20/40) ; circle, hand, star, cup, cross (20/30) ;star, cross, circle, heart, flag (20/20) ; and moon,heart, cup, star, and cross (20/10) .

Toys. Children were also asked to namesome three dimensional toys, from a dime store,which were representations of the same objectsas those in the picture tests, e.g. house, tree,horse, telephone, teddy bear, scissors, man. Thepurpose of this was to see if recognition couldbe enhanced by a more realistic representationthan that in the picture. Recognition could beindicated verbally or by pantomime play withthe toys. This test was included to to see if ver-bal ability might be significantly improved bythis change from picture to toy targets, and if

those children who did not respond verballyroight indicate recognition by play.

Visual acuityIndicating directionsThe most widely used preschool vision

tests depend upon the child's perception andimitation of direction as a response; if thechild responds correctly, it is assumed that henot only saw the stimulus, but he interpreted itas well. With such tests, if the response is in-correct, it is not known if the child failed to seethe stimulus or failed to interpret or communi-cate wAlat he did actually see. There has beenmuch work in child development and psychol-ogy 31-94 based upon conceptualization of thedirectionality of figures. The findings in thiswork point to a tenuous grasp of the directionconcept in the preschool child. It is commoneven for first graders in school to demonstratesome right-left confusion as evidenced by fre-quent reversals, e.g. letters or words like "b"and "d", "was" and "saw". Young childrenwho have difficulty recognizing abstract formsmight appear more visually acute when famil-iar concrete forms are used.

In contradistinction to these findings ofpsychologists and educators, the direction-basedtests have remained the tests of choice for pre-school children. The Snellen E test has beenstandardized and is used frequently by theNational Society for the Prevention of Blind-ness.44-50 Perhaps because mainly older pre-school children, 4 and 5 years old, have been in-cluded in these studies, and probably because ofa concretization of the E figure into a table,these programs have met with some success.

Drawing. Children were asked to copylines, geometric forms and letters after demon-strations were drawn in front of them. Thiswas done on a pad of paper with crayons. Thehand used and manner of holding the crayonwere noted. Lines were vertical, horizontal,two obliques, a cross. Forms were a circle, tri-angle, diamond, and square. Letters were anE (in one or more directions), C, U, and D.

Hand and "E". An attempt was made toteach the Sjogren hand 39 and the Snellen E,4°in order to elucidate the learning procedure in-

7

volved and the comprehension attained. A

20/40 size hand card, and then a 20/20 E card

were shown at conversation distance, and thechild was asked to match his hand to it as itappeared in different directions. Usually noattempt was made to test actual acuity by doing

this at 20 feet and occluding each eye; children

who learned successfully were asked to repeat

the test after stepping back a few paces. The

E card was also matched by a cutout E figure

(3 x 3 inches) made of white cardboard, and

held by the child. Notations were made of the

way the child responded and the duration andamount of interest was estimated.

Visual acuityEliciting opticokineticnystagmus (OKN)

Opticokinetic nystagmus (OKN) is anystagmus which can be used to objectivelymeasure visual acuity; it is produced when a

simple repetitive moving pattern is seen by the

eyes.66 The stimulus is usually vertical lines(stripes) moving horizontally ; the eyes follow

a line (slow phase), then jerk back to fixate asucceeding line (rapid phase). Also known as"train nystagmus", this is the response of the

eyes to multiple passing telephone poles. Verti-cal opticokinetic nystagmus, the response to ahorizontal pattern moving vertically, is moredifficult to elicit. This nystagmus differs fromvestibular nystagmus in that it continues aslong as the eliciting stimulus is present. Theeye must see the stimulus in order for the nys-tagmus to be produced, and if one eye perceivesthe stimulus, both eyes will show the nystagmus.The stimulus must be presented to one eye at atime or to a visual field not shared by the twoeyes in order to be sure that each eye sees. Asthe basis of a test of visual acuity," more acuityvision is measured when lines and spaces arenarrower and the distance from the subject is

greater. A technique to produce nystagmushas been used to test newborn infants for blind-ness and to determine how well normal new-borns see." A more refined test has been usedfor adult acuity testing, but its use has beenlimited to detection of malingerers and hysteri-cal blindness; subjective tests are more accuratethan this when subjects are cooperative." Pre-cise measurements of nystagmus with recording

8

electrodes for the extraocular muscles has been

used as a test of ocular dominance."There is at present no workable objective

measure of visual acuity for preschool children.Such a test might answer some of the problemsof administration and cooperation with theseyoung subjects. Possibilities for an opticoki-netic nystagmus test with preschool children:lave never been fully investigated, and this willbe the subject of a further study.

In this study the nystagmus is elicited bypulling a striped tape, e.g. a tape measure, infront of the subject's eyes. This simple tech-nique was used because the purpose of includ-ing this test was only to see if children this agewould be interested and respond with nystag-mus; there was no effort made at that time toactually measure acuity. No attempt was madeto occlude one eye during the test. A 5-foottape measure, white plastic ribbon with mark-ings in black separated by 1A6 inch, was used.The child was instructed to look at the exam-iner; the rolled tape was then held at eye leveloff to the left side of the child's face and the enddrawn across in front of the eyes from left toright.. This was done quickly and frequentlywith much verbalization, "Look at it, look atit, etc." The reverse side of the tape with alight green background was presented to somechildren as well. Notation was made of theparallelism of response between the eyes andany difference in response between the twodirections in which the stimulus was presented.

Muscle balanceTests of extraocular muscle coordination

between the eyes themselves and the eyes andhands were included in an attempt to detectexternal pathology, latent strabismus, nystag-mus, immature eye-hand coordination, andsubclinical cerebral palsy.

Nearpoint of convergence and penlightfollowing. Following a penlight as it movesout to and in from the peripheral visual fieldsgave an indication of child's attention as well asan opportunity to find field defects. Nearpointof convergence (N.P.C.) was determined withthe light, and the distance and deviating eye atthe near point were recorded. Touching the

light at several distances and threading it withthe rings of the toy scissors gave an indicationof hand coordination, and behavior on thesetests could be compared to expected behaviorbased on many such examinations by Ape 11 andLowry. 12

A Hirshberg test or "conical reflection"test, the observation of the relative positions ofthe corneal reflections was done incidentallyduring observation of the extraocular move-ments. Relative eccentricity of one reflectionindicates a possible deviation of that eye. Thistest was not systematically used as an absoluteindication of muscle imbalance because of theview that disparity between the position of cor-neal reflections may frequently be a normalfinding. Any positive finding, however, wasconsidered when recommending followup for aparticular child.

Cover test. A cover test, the most fre-quently used and criticized test for eye musclebalance, was attempted. The examiner's handwas used as an occluder and the lighted pen-light, 12 inches away, was the focus. A roughdistance cover test was done without a stand-ardized target; the child was directed to lookat some small object outside the window ap-proximately 20 feet away.

Red glass test. The red glass test fordiplopia was attempted. A 2 x 2 inch red plas-tic square was placed before each eye in turn;the penlight held about 10 inches away was thetarget. The child was asked how many lightshe saw and what color they were. "One redlight" indicates fusion of images is occurring;"two lights", one white and one red, indicatessome degree of muscular imbalance is present.

Stereoscope. A hand stereoscopeequipped with three cards was used to determinereactions to such an instrument and detectfailure of fusion. The cards used were :

Bird and cage (colored, Kroll Al).

Pumpkin with one eye on each side(Hale 02).

Jumping dog and lion holding a hoop(Kroll A3).

The cards were presented at a position 5 cm.from the lenses and moved out along the 15 cm.length when cooperation permitted.

PerformanceTasks which are appealing to preschool

children as familiar games give an indicationof how well they function visually in practicalterms. All of the performance tests used weresimilar to tests in the child development andoptometric literature,"-12 but none was exactlythe same. The tasks were comparable and thetools, if anything, easier than those used byGesell. It is behavior in approaching thesetasks which assumes comparable patterns ratherthan performance on the tasks themselves.This behavior comparison permits an estimationof maturity of the children tested. It wasanticipated that these children would be per-formance oriented and might respond favorablyto this type of task.

Blocks. Six colored plastic cubes (threered, one blue, one green, one yellow, 21/2 incheson a side) were used. The child was asked tobuild a tower; this was demonstrated if thereseemed to be poor comprehension. The numberof blocks piled, the hand used to pile, and thestance of the child were noted. A bridge ofthree blocks was demonstrated and, with thedemonstration intact, the child was asked toreproduce this with the other three blocks. Atrain (five horizontal blocks, and an engine)was demonstrated, and after the blocks wereknocked askew, the child was asked to buildsuch a train.

Peg bench. A commercial toy designedfor 2 years and older, was purchased to serve asa test requiring coordinated aim, judgment ofdepth and awareness of a third dimension. Thetoy is composed of a red and neutral coloredbench measuring 10 x 4 x 5 inches, 12 coloredpegs (2 each of red, blue, purple, yellow, orange,and green) cylinders measuring 1% inches longand 5/8 inches diameter, and a hammer with 5-inch handle 1/2 inch diameter, and a head 23/4 x11/2 inches deep. Six of the pegs sit in 1/2 inchdeep holes on top of the bench, and a seventh

9

1111..-

fkultis I iIr4

PERFORMANCE TESTS

hole can hold up to three pegs; when a fourthis hammered in this hole, a peg is ejected fromthe side of the bench. This method of hammerpegs "through the tunnel" was demonstrated toeach child who was then permitted free playwith the toy. The approach to the game, focusof interest, coordination and aim were noted.Particular note was taken of the developmentof an awareness of different depth holes andactivity like probing holes during the game.Pegs also served as a color test (see below) .

Some children were asked to play thisgame while wearing occluder masks; these werethe same Hallowe'en masks as used in the toytest above. The purpose of this was to seeif aim was significantly different with one eye.alone, and whether there was any consistent

10

deviation to either side. This idea of markedand consistent deviation from a central linewas used in the next two performance tasks aswell, line walking and ball rolling. It wasthought that significant eccentric fixation, i.e.extramacular fixation might be suspected bysuch simple performance tasks. Actual deter-mination of fixation. necessitates a specializedophthalnioscope-like instrumeat, the Visu-skop.69-71

Walking lines. Walking lines was at-tempted in some children a3 a measure of thegeneral ey e-foot coordination, and a possiblemeans of detecting eccentric fixation. Withboth eyes open, and then with each eye occludedin turn, the child was asked to walk a straight

line made by the tiles on the living room floorwhen these were available. This was demon-strated as a heel-to-toe walk which might alsodetect ataxia.

Ball playing. Ball playing had severalpurpose :. A bright orange rubber ball with a7-inch diameter was used. General coordina-tion and aim were noted; the manner of spon-taneous throwing and catching were comparedto patterns expected at the age of the child.The child was asked to roll the ball straightlooking with both eyes, and then with each eyeoccluded. While each eye was occluded, the ballwas rolled toward the child's side, and he wasasked to fetch it. Observing the child's be-havior in searching for this very large objectgave an impression to compare with his behaviorwith two eyes open. Thus it was possible tojudge whether he was normally binocular, i.e.,the child who functioned well enough andequally with each eye alone but not as well as hedid with two. The functionally monocularchild could be detected by good function equi va-lent to that shown with both eyes in one eye anddistinctly poorer function in the second eye.

Puzzle. A commercial puzzle graded"18 months on" was used in lieu of a geometricform board. It is wood and consists of a char-treuse background board and four separatepieces shaped to resemble types of fruit : purplegrapes, red apple, orange pear, and yellowbanana. This was chosen over a form boardfor its concretely represented form variationand its appeal as well as for its availability.The puzzle was taken apart and put togetheronce in front of the child, and he was thenasked to do the same; his motions were notedand timed. Reinstruction with verbal descrip-tion was given when necessary, usually after3 minutes of trial and error. Ability to matchforms, distinguish landmarks of forms, and theright from the wrong side of a puzzle piece werenoted.

Dominance

Interest in eye dominance and its relationto handedness and hem spheric dominance, andthe correlation of the, =3 factors with reading-

735-670 0 65 3

disability, speech disorders, visual acuity, havebeen the subject of much discussion and muchcontroversy periodically. Numerous more orless complicated tests have been devised, andthe literature is full of testimonies to the signif-icance of one or more tests, and also to theworthlessness of almost all of them 72 -93

The tests used in this study are subjectto much of this criticism. They were performedto determine at what age dominance started tobe established, and what the fact of having adominant side meant. The eye dominance testswere three, all simple sighting tests : (1) an81/2 X 11 inch gray cardboard with a 1 inch cen-ter hole was handed to the child who was askedto peek through the hole while holding the card-board with both hands; (2) a cardboard paper-towelling tube, with a 11/2-inch hole, was usedto sight while held in both hands, the apparentlydominant and the apparently nondominanthand; (3) a lorgnette-type occluder,25 with a1 inch hole, made of black paper was used tosight while held in each hand. The influenceof eye dominance on effective occlusion espe-cially with the lorgnette has not been deter-mined.

Handedness, first, of all, was not deter-mined in any standardized manner; rather itwas inferred from the child's use of one hand towrite, indicate directions, hammer, and playball, to the exclusion of the other hand, and, onthe mother's statement made with certainty,that the child strongly preferred to use onehand for eating. When there was any use ofthe second hand for these activities, and whenthe mother expressed any doubt, handednesswas judged "not established." Children whowere ambidextrous with good de_terity, how-ever, were judged to be left -handed, particu-larly when they seemed clearly left-eyed andthere was a family history of left-handednesssince left-handed people often tend to be ambi-dextrous. Secondly, none of the factors enter-ing into the judgment of handedness wasinfluenced by eyedness, i.e. the child was nevertotally blindfolded (except perhaps thoseseverely amblyopic children who, with the goodeye occluded, were functionally blind). Theinfluence of handedness on eyedness was ac-counted for by notation of which band was

11

used while sighting.Foot dominance was determined in some

children by consistent use of one foot kickinga ball.

ColorColor awareness was observed in an

attempt to judge when this becomes a factorin the child's dealings with his environmentboth for normal learning and growing and fordesign of vision tests. Children were asked tofind another peg the same color as one theexaminer picked upusually red, yellow, orblue. If this was done successfully, the childwas asked to name colors. This was done dur-ing the peg game and/or the block building.Ishihara series 94 were not used because theyrequire that the subject identify a figure usuallya number, within a mass of colored dots, and

12

this task was considered beyond the capabilityof these children.

Questionnaire

Questions asked the mother concernedthe child's development, his preference for cur-rent activities and his current abilities, pasthistory of general medical illness, symptomsand signs suggestive of visual disorders. Thisafforded an opportunity to learn about thechild himself past and present and to comparehim to his peers. But it also gave some indica-tion of the mother's awareness of the child, herconcerns about his health, and her willingnessto seek preventive medical care. It served toplace any visual findings in the total frameworkof the child's medical and social background.

PARENT QUESTIONNAIRE

1. What name do you usually call child?

2. Perinatal historya. pregnancyb. deliveryc. birth weight

3. Developmental historya. When did baby seem to see?b. When did baby first smile?c. When did baby start to pick up toys?d. When did baby sit? pull to stand? walk? run?

e. When did he start to talk?f. How does he compare with your other children at the same ages?

same, faster, slowerHow?

g. Did he ever seem cross-eyed or wall-eyed?.....when?....how long?..h. Does he dress himself?

4. What are his favorite activities? Indoors? Outdoors?How long does he play at these?Games, puzzles, blocks, drawingBall, bicycleTelevision since favorite programsBeing read to since pictures identified ..since

pictures favoredEating Does he feed himself? How long? How well?

5. What illnesses has he had? Age? Complications?Did he take any drugs? Does he now?

6. Current eye symptomsa. rub or close eyes? Do they water? Are they bloodshot?b. complain of burning, headache, etc., after watching TV?

carsickness? ... .

c. seem uncomfortable in bright light? Does he shut an eye?d. have: trouble finding his way in the dark?e. hold things close to face? .......sit close to TV?f. seem clumsy in games?

If child does squint, ask:(1) Has he had any illness with high fever?....how high?

How song did it last?(2) Has he ever been unconscious?....following a head injury?...(3) Has he had convulsions?....fever fits? other ?.

(4) Measles? mumps?. . .whooping cough?.. polio?(5) Has one eyc ever been bandaged for any reason?

7. Has he ever been to an eye doctor? where? when why?

8. Does anyone in the immediate family have any eye trouble?Did anyone ever fail a driver's test or military exam because of eyes?

Does anyone wear glasses Who? Why? Since

Does anyone have a squint now? As a child? .....

13

III. THE SUBJECTS

This study was carried out througha Well Child Conference (hereafter calledW.C.C.) of the Boston Health Department sit-uated in the Bromley-Heath Housing Projectsat 10 Lamartine Street Extension, JamaicaPlain, Mass. This W.C.C. is conducted fordemonstration and teaching by Harvard Schoolof Public Health and Harvard Medical School.There are about 1,150 families living in the hous-ing project, and about 2,300 of the 4,000 resi-dents are children. There are equal numbersof Negro and white families. The social, eco-nomic, occupational and educational levels arelow.

All children listed in the records of theW.C.C. public health nurses who lived withinthe project and had birthdays between Decem-ber 1958 and December 1959 were included forthis study which was done August 1 to Octo-ber 16,1962. This group was chosen because ofthe "borderline" age for testability in tradi-tional preschool vision testingabout 3 years,and it was hoped that some visual trend in thefew months before age 3 and in those monthsafter the third birthday could be established.There were 105 children in this age group in theW.C.C. records, and an additional 9 were pickedup as the study progressed, making a total of114 children. The "pickups" were siblings whowere also of preschool age, or 3-year-old chil-dren who appeared in the W.C.C. for the firsttime during the course of the study.

Of the 114 children, 10 had moved sometime in the 6 months prior to testing. Of the104 children contacted, 10 refused to participatefor various reasons. Of particular interestamong the refusals were two children who had

14

eye problems and were under care elsewhere:one, a child with esotropia, had been referredfrom the W.C.C., was under treatment by anophthalmologist, and the mother did not wantto confuse the child with another eye test; theother was a child with allergies manifested byconjunctivitis, and again the parents thoughtshe was having "enough eye care" (local, privatedoctor). Two mothers were ill and unable tocooperate because the children were placed infoster homes for the duration of the mother'sindisposition. Two mothers, with poor recordsof clinic attendance, were "too busy." Three,fearing the mad strangler, would not open thedoor. One child with phenylketonuria was soretarded that the mother thought it pointless t'detect any other problems he might have. Thus,94 of the 104 available children (90 percent) ac-cepted and were tested.

Of the 94 children studied, 25 were fromfamilies called "uncooperative" for poor clinicattendance. Five of these children had neverbeen to clinic, and the other 20 had not beensince early infancy. The families were knownbecause of initial home visits made by the nurseor clinic visits with other children. Eight ofthese families showed evidence of extremelypoor general care and low motivation whenvisited. The other 17, however, seemed fairlywell organized and at least 14 of these statedthey were receiving ongoing medical care fromother sourceshospital out-patient depart-ments (Boston City, Children's Medical Center,Chelsea Naval, Boston Floating, and BethIsrael), and local private doctors.

The 94 children tested (see fig. 1) ranged

in age from 31 to 54 months; the "extra" chil-dren at the extremes are accounted for by errorsin recording ages and the testing of several oldersiblings. The mediim age was 39 months; thiswas, in other words, an unusually young popu-lation of children to' vi ion screening. Therewere 47 males and 47 females. In eight situa-tions, two different children were tested; sevenof these were families with two children in theage range (three sets of twins), and one was afoster home. for two children. Thus, there were87 families involved in the study. The childrenstudied had an average of 2.9 siblings each, 2.3older and 0.6 younger.

There were 53 white and 41 nonwhite(part or all Negro) children. Three Puerto

Rican children were included. The families of66 children were "intact", and 28 children weremotherless, fatherless, or both,

Seven children (three males, four fe-males) had attended nursery school, and an-other 21 (12 males, 9 females) were startingthat September. For all of these .:;8 children,nursery school association was with the Associ-ated Day Care Center located in the housingproject at 262 Parker Street.

All except five children had checks or nocomment about their eyes on their clinic records.The five notations were for transient erythema,torticollis, early inability to fallow lights,plugged lacrimal duct, and an unexplained re-minder to "check eyes next visit."

15

number of children

rn 1X1

4S* * 41 4$4$4W

::.X.X.:04001101,1001,11044040"

,::::::::::::::::::::::::::::::::::44.:0$0 444444444

:::::::.::::<::4444444440.***

411.4.41..11.4..*.V.V.O.%,11.0y4A00.0,40.,

5 E. CD E.

CD

91

IV. OPERATING PROCEDURES

Preliminary tests were given to two chil-dren in one family after arrangement wasmade with the district nurse. All other meet-ings were arranged directly, following an in-troductory letter. Groups of 10 mothers at atime, arranged geographically, were contactedby this letter which explained the nature of thestudy and stated when the examiner wouldvisit. District nurses were informally awareof which families were being contacted. Thisfirst visit was intended to make the testing datefor some convenient time in the family sched-ule, and to obtain answers to the questionnaire.It took anywhere from 5 to 30 minutes perfamily depending on the number of questionsthe mother asked and whether the questionnairewas actually completed. It became apparentsoon after this visiting started that time andschedules are not important factors in the livesof many of these families. Although almostall of them acknowledged having received theletter, most said they had forgotten the day ofthe visit. There was at most an interval of oneweek between mailing the letter and visiting.It was often necessary to revisit numerous timeson subsequent days in order make a testingappointment. It became much lore efficient todelay the questionnaire for the testing sessionwhen there tended to be more time for this,during those test activities which seemed to con-sume time beyond their immediately informa-tive value. Thus, the first visit was shortenedto 5 or 10 minutes. This overall average timewas almost 30 minutes though because of thelarge number of visits necessary in some cases.This first visit co-aid be unnecessary if a famil-

iar person like the public health nurse were todo the testing, but multiple visits for contactingsome families seem inevitable.

It was decided that testing in the homeswould be preferable to testing at the W.C.C.It was thought that the disadvantages of poorlystandardized environment, interference fromparents and siblings, and imposition on thefamily would be outweighed by the advantagesof a flexible schedule, opportunity to talk withthe mother at leisure, and observe the child inhis natural environment. The available build-ing was very often occupied with day campand clinic activities, and the aura of a clinicvisit to most children might have detractedfrom their rtadiness to cooperate. Fifty ofthe children might have come to the W.C.C. fortesting; these were generally cooperative fami-lies and mothers who were concerned about thepossibility of eye problems. Thirty-seven saidthey would not come to the W.C.C. for the testbut were willing to have the tests brought tothem. Seven were not even willing, before ex-tended clarification and discussion, to have thetests done at home; reasons ranged from fear ofeyedrops to concern about what the procedurewould cost.

The entire procedure was construed asa play session, and it was not unusual for thechildren to request a return visit despite oc-casional, initial anxiety at the word "doctor."Mothers tended spontaneously to call the exam-iner "teacher" and many children seemed toenjoy feeling this was like school.

Most mothers were home at the ap-pointed time, but many volunteered that they

17

had forgotten the appointment, and the childwas often outdoors, asleep, or away. The en-tire procedure took an average of 11/2 hours.The most time consuming tests were consist-ently the performance tasks; because of theirappeal as real games it was difficult to cut themshort, and they often tended to function moreas transitional energy expenditure than as testsin themselves. The time for each individualtest varied greatly with the child's capacity andinterest, and it is impossible to state meaningfulaverages.

The tests were carried in a shopping bag,and an effort was made to use all the tests ex-cept where there was obvious lack of coopera-tion or interference. The order varied fromchild to child, and there was no set number oftimes for reintroduction of tests which failedinitially. An appealing toy was used as an in-troduction--usually the ball or a puzzle. Some-what abstract tests were interspersed with phys-

18

ical activity, and it was thus possible to retainthe child's attention for long periods of time,at least up to 2 hours, and to relieve both fatigueand tension. Notes were taken during testingand later transcribed onto mimeographed forms.

When 40 children were retested 6 monthslater, it was interesting to note that although 5mothers stated they had misinterpreted the let-ter and thought they were supposed to go to theW.C.C., only 1 actually did go. The revisitingnecessary was again in excess of what might beexpected. Optimally, i.e. when consecutivefamilies were home, 5 retests an hour could bedone. However, a total of 11 hours were neces-sari. in order to retest 36 childrenthe actualtests taking little over 7 hours. This is againa problem that might be obviated if the nursewere doing the testing.

Summary cards were kept for each childlisting all basic identifying information, ap-pointments and a summary of the test findings.

Y.

V. FINDINGS

Testability and nontestability onindividual tests

Most of the eight types of tests were at-tempted with most of the subjects. Childrenwere testable with each procedure accordingto the definition established for the purposes ofthe study and summarized in table 1 The totalnumbers and percentages of children tested andfound testable with each procedure are shownin table 2.

Table 1. SUMMARY OF DEFINITIONS OFTESTABILITY

Test

Visual acuity"Stycar:"

Matching letters

Near letter test

Distant letter test....

Testability is defined as

Three or more letters success-fully matched.

Consistent matching of 3 ormore letters with or withoutoccluder at 12-16 inches.

Consistent matching of 3 ormore letters with or withoutoccluder at a distance greaterthan 5 feet.

Miniature toy test:2 eyes Naming or matching 2 or more

toys consistently and cor-rectly at a distance of atleast 5 feet.

Each eye Naming or matching 2 or moretoys consistently and cor-rectly at a distance of atleast 5 feet while wearingoccluder.

735-610 0-65-----4

Teat Testability is defined as

Visual acuityNaming pictures and toys:

Picture tests(Allen, Osterberg,A.0.)

Toys

Naming 4 or more pictures onany of the 3-picture series atclose range without occluder.

Naming 4 or more toys atclose range without occluder.

Visual acuityIndicating directions:

Drawing Copying the demonstrationswell enough to accuratelyreproduce at least the ver-tical and horizontal line.

Indicating 2 or more directionsconsistently and correctly atclose range without occluder.

Indicating 2 or more directionsconsistently and correctly atclose range without occluder.

Sjogren hand

Sne llen E

Visual acuityEliciting:

Opticokineticnystagmus. Showing nystagmus when tape

moves right to left and leftto right, without occluder.

Muscle balance:Penlight following

and nearpoint ofconvergence.

Cover test

Red glass test

Stereoscope

Following the light in all direc-tions and converging on it.

Cooperating sufficiently toallow reproducible observa-tion.

Giving plausible answers to thetest question.

Looking correctly and report-ing an observation for atleast one of test cards.

19

Test Testability is defined as

Performance:Blocks Reproducing at least 1 of 3

constructions demonstrated(tower, bridge, or train).

Peg bench Participating in any way whichmakes possible an estima-tion of depth perception andcomparison to peers.

Walking a straight line withand without an occluder.

Playing with the ball withanother person in any way,with or without occluder.

Participating in any way whichgives indication of ability toperceive and recall shapesand permits comparison topeers.

Dominance Sighting (even partially)through at least one of thedevices used.

Color Matching and/or naming colorsconsistently and correctly.

Walking line

Ball playing

Puzzle

All of the tests could not be attemptedwith all of the children. Those tests which re-quired some abstract thought, particularly theStycar letter test, and the direction tests, some-times did not attract enough attention to makeinstruction and testing possible. Some equip-ment, e.g. the stereoscope and nystagmus tapemeasure, periodically disappeared and thesetests were necessarily omitted until test mate-rials were recovered. Since the tests were nevergiven in a prescribed order, there was an oc-casional omission caused by oversight, usuallyin situations where more than one child at atime was being tested. Some tests were pur-posely discontinued after a number of childrenhad been examined, because they seemed inap-propriate to this age or especially uniformative.

Visual acuity "Stycar"Matching letters. In general, the chil-

dren did not find this letter test attractive;there was little enthusiasm displayed even bythose who could competently match letters.The test could not be attempted with 25 chil-dren who either ignored these particular testmaterials when they were shown repeatedly orwere so generally uncooperative and hyper-active that the futility of proceeding beyond

20

the showing was obvious. Those 59 childrenwith whom the test was attempted (tables 2and 3) were those whose attention was goodenough to permit explanation of the procedure.

The 26 children who were able to per-form the required matching procedure for aminimum of three letters were considered test-able (table 1) and the other 33 children werenontestable. Many of these nontestable chil-

Table 2.TESTABLE CHILDREN

Totals for Each Test

TestAt-

temptedNo.

Test-ableNo.

Per-cent

Visual acuity "Stycar":Matching letters 59 26 44Near letter test 23 22 96Distant letter test 21 10 48Miniature toy test -

2 eyes 94 80 85each eye 94 60 64

Visual acuityNaming picturesand toys:

Pictures 93 67 72Toys 43 29 67

Visual acuityIndicating direc-tions:

Drawing 89 61 69Hand 72 42 58E 70 21 30

Visual acuityEliciting optico-kinetic nystagmus. 77 50 65

Muscle balance:Penlight following and

N.P.0 80 67 84Cover_ 59 32 54Red glass 12 2 17Stereoscope 68 26 38

Performance:Blocks 85 80 94Peg bench 91 91 100Walking lines 19 16 84Ball playing 89 82 92Puzzle 94 94 100

Dominance 94 57 61

Color_ 84 36 43

Table 3.VISUAL ACUITY"STYCAR"

Testable and Nontestable Children (Breakdown by Sex, Race, and Age)

TestSex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Matching letters:Attempted_ 28 31 24 35 ___ 1 ___ 4 1 5 5 5 4 2 4 3 7 4 5 3 3 1 1 1

Testable 14 12 11 15 ___ ___ ___ ___ 1 1 1 1 2 ___ 1 3 5 1 2 3 3 1 ___ 1

Nontestable 14 19 14 20 ___ 1 ___ 4 ___ 4 4 4 2 2 3 ___ 2 3 3 ___ ___ _. 1 __

Near letter test:Attempted. 12 11 9 14 ___ ___ ___ ___ 1 2 1 ___ 2 ___ ___ 3 5 1 1 2 3 1 ___ 1

Testable 11 11 9 13 ___ __ ___ ___ 1 1 1 __ 2 ___ __ 3 5 1 1 2 3 1 ___ 1

Nontestable.Distant letter test:

Attempted 11 10 10 11 ___ ___ ___ ___ ___ 1 ___ ___ 2 ___ ___ 3 6 1 2 2 3 ___ ___ 1

TestableNontestable

Miniature toy test:Attempted 47 47 41 53 1 2 5 7 4 11 7 5 8 4 7 5 8 5 5 4 3 1 1 1

Testable-2 eyes_ _ _ 39 41 34 46 ___ 2 4 5 4 9 5 3 7 4 7 5 7 5 4 3 3 1 1 1

Nontestable-2eyes 8 6 7 7 1 ___ 1 2 ___ 2 2 2 1 ___ ___ ___ 1

Testableeach eye_ 30 30 28 34 ___ 2 1 5 1 7 2 3 5 3 4 4 7 4 4 3 3 1 ___ 1

Nontestableeacheye 17 17 15 19 1 ___ 4 2 3 3 5 2 4 1 3 1 1 1 1 1 __- ___ 1 _-

dren were able to match only the 0" ; a fewmore could distinguish the "X" or "V" butconfused these two letters. There seemed to beenhanced interest and accuracy in this test whenthe child used pennies to cover the appropriateletters on the key card instead of pointing tothe matching letter.

This ability to match letters made possi-ble actual visual acuity measurement in thechildren who continued to match under theconditions of (b) the near letter and (c) thedistant letter tests.

Near letter test. Of the 26 children whowere testable (table 1), there were 23 whosustained enough interest in the letter matchingoperation to take the near visual acuity test;the 3 children with whom the near test wasnot attempted had refused to play at matchingor the acuity test after they had successfully

matched the three letters required for minimumtestability.

With one exception, the children seemedto enjoy this near test more than the practicematching. A possible explanation was offeredby one child's mother who suggested that hewas overwhelmed by the mass of letters on thenear test card, and that his ability to match de-pended on the presentation of single largesymbols.

The near test was done with both eyes inall 23 children and also with each eye sepa-rately in 20 children. Those who were nottested with mask occluders were children whodid not accept occlusion for any test. Two ofthese three children were called testable (tables1 and 2) since they could perform the requiredoperation. If, however, the primary objectiveof including this test had been amblyopia detec-tion, i.e. if it were the only visual acuity test

21

performed, these two children would not havebeen considered testable.

Distant letter test. None of the childrenwho successfully matched letters was able tosustain interest long enough to learn the pro-cedure and then take near and distant testsconsecutively. The distant test was thereforeusually given some time after the near test.It was possible to attempt this in only 21 of the23 children who took the near test; the othertwo children became too tired and distractedto attempt the distant test. There was gener-ally great fascination with the mirror placed10 feet away, but this arrangement made match-ing more difficult.

Although, no doubt, distance itself was alarge factor, the mirror seemed to be distract-ing, and it was often difficult to illuminate theletters well. Of the 10 children who were ableto successfully match letters at least with botheyes, at some distance, 8 matched 20 feet away(the mirror was 10 feet away), 1 matched 10feet %way (the mirror was 5 feet away), and1 matched letters 6 feet away without themirror.

Miniature toy test. This was the only testthat could be systematically administered as adistant visual acuity test to most of the children.In general, the children found these toys andthis test very appealing. Children who couldnet name pictures, could either name these or in-dicate their function in play, and match them.Sometimes, the toys seemed to be almost tooattractive; there was much dawdling, initialplay v ith the toys, and pauses during the testto put the doll on the chair or see if the carwheels worked. The airplanes were so attrac-tive that they both disappeared, by the fourthtest, and other toys sporadically followed. Itwas possible to function with only a single setwith those children who named rather thanmeched ; the less verbal children who needed tomatch seemed content to match not quite identi-cal toys, i.e. dime store replicas, to the test mate-rials. Thus, it was possible to attempt the testwith all 94 children.

There were two children who matched ac-cording to color rather than form; this sup-posedly more primitive type of match was more

22

noteworthy for its absence in the other children,many of whom had toys arranged so that thewhite chair might have been matched with thewhite car. After these two children, specialcare was taken to keep all objects of one colorin one set.

Two eyes: The 80 testable children (table1) were more or less able to follow the test in-structions. The 14 children who could not betested at all with this test did not display anycomprehension or interest in any of the othertests. On this test they did not accept any pre-paratory instructions although they did playwith the toys.

Eacb eye: The test was only successfulas an actual measure of visual acuity in those 60children who matched or named the toys at adistance while wearing occluders.

The 20 children who were not differenti-ally testable would not accept the mask occluderfor this test; 5 of them did accept occlusion forperformance tasks, especially the ball. Table 4shows rejection of the occluder by sex and age;this rejection was sporadic out frequent before37 months, and acceptance was the rule after41 months. Of the 60 who were testable, 59 usedthe mask occluder, and one who refused themask was tested with his mother holding thelorgnette. Of the 59 who wore the masks forthe test, 47 accepted them readily and acceptedthe right as well as the left mask; the other 12were slow to accept the masks and obviously un-happy with them, taking them off several timesduring the test, needing reassurance that the oc-clusion was temporary, and having to be testedwith great speed to take advantage of the short-lived mask acceptance.

Visual acuityNaming pictures andtoys

Pictures. At least one of the three pic-ture series was shown to all subjects (with oneoversight), and an attempt was made to testrecognition of these figures. Actual measure-ment of visual acuity was not attempted for-mally, i.e. there was no distance assumed be-tween examiner and subject and no occluderswere used.

All the children, except two, preferred

Table 4.OCCLUDER REJECTION (by Sex aad Age)

Sex Age (months)

M F 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Number rejecting mask 9 11 3 3 1 3 3 1 3 1 1 1

Number children tested 39 41 2 4 5 4 8 5 3 4 7 5 7 5 4 3 3 1 1 1

Percent rejecting mask 23 27 75 75 13 60 37 25 43 20 20 100

the Osterberg chart and four children wouldname these pictures and none of the other pic-ture tests; the two who preferred the Allenseemed intrigued with the little booklet form ofthe test rather than with the pictures them-selves. In general, there was most delight withthe Osterberg, and it was surprising that manyof the figures were correctly named despite anexpectation that they would be unfamiliar, e.g.the skeleton key and scissors. The Allen cardswere called more unusual names than the Oster-berg figures, e.g. the Christmas tree was calleda "lamp", and a "lady's hat," and the birthdaycake was often "fire" or "fire on plates." TheA.O. was generally unappealing; few childrenknew four of the figures, hand, "coffee" ( for

cup) and "valentine" (for heart) were the mostfrequent responses. Few children recognizedthe sailboat, and the other figures were namedonly when a concrete figure was made of them,e.g. "wheel" for circle, "window" for cross,"sky" for moon or star or flag (with star on it).

Children were considered testable whenfour pictures were correctly named (table 1)because the Allen test series 25,27 which has thefewest pictures utilizes a passing criterion offour out of the seven cards. The Osterbergchart with 12 pictures and the A.O. with 8 couldbe used as acuity tests if 4 pictures are con-sistently identified.

A Spanish speaking child was consideredtestable (tables 2 and 5).

Table 5.VISUAL ACUITYNAMING PICTURES AND TOYS

Testable and Nontestable Children (Breakdown by Sex, Race, and Age)

TestSex Race Age (months)

MF NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

PicturesAttempted 46 47 41 52 1 2 5 7 4 10 7 5 8 4 7 5 8 5 5 4 3 1 1 1

Testable 31 36 29 38 ___ 1 3 2 3 8 4 3 5 3 5 4 8 5 4 3 3 1 1 1

Nontestable 15 11 12 14 1 1 2 5 1 2 3 2 3 1 2 1 ___ ___ 1 1 ___ ___ ___ __

ToysAttempted 23 20 18 25 1 2 3 4 2 5 1 2 2 1 5 2 3 3 2___ 2 1 1 1

Testable 13 16 14 15 ___ 1 1 2 1 3 1 ___ 2 ___ 4 1 3 3 2 ___ 2 1 1 1

Nontestable 10 4 4 10 1 1 2 2 1 2 ___ 2 ___ 1 1 1 ___ ___ ___ ___ ___ ___ ___ __

23

One child communicated by cleverlymatching pictures on the charts with objectsin her home; she was, for thiLl tabulation (tables2 and 5) , however, considered nontestable be-cause she could not have been tested under ordi-nary conditions.

There were 17 children in the group of 26nontestables (tables 2 and 5) who hardly spokeat all during the exam. Nine of these could notexpress themselves verbally to anyone besidesparents. Of these nine children there were fivewhose mothens expressed particular concern andthe other four had a family history which madethis seem normal. Of the testable children 38did not speak well for their age; only 29 (13boys, 16 girls) had very good verbal ability.

Toys. The toys were not shown, alterthe first few tests, to those who readily identi-fied pictures because they were more distractingthan informative. It is assumed that the 51children who were not specifically asked to nametoys (other than those in the toy acuity test)might have done so, based on their verbal prow-ess on other tests. A specific attempt to deter-mine recognition of toys was made with only43 children (table 2 and 5).

Three dimensional toys, including thosein the miniature toy test, were not named by14 children (tables 2 and 5) ; however, 3 of the"nonspeakers" could name these toys and 12children who could not name enough pictures oncharts to be considered testable (table 1) couldname the toys. These children seemed to lackinterest in the diagrammatic pictures; thereseemed not to be so much a failure of perceptionas a lack of appeal. The toys were named onlyby withholding them from use until they werenamed, and this motivation to find a word wasnot so great for the picture charts as for thereal toys.

Visual acuityIndicating directionsDrawing. Drawing was attempted with

the 89 children who cooperated to the extent ofholding a crayon and writing on a paper. Itcould not be attempted at all with five childrenwho would not accept these materials. The 29children who were nontestable did not followdemonstration drawings, walked away andscribbled on their own, or in spite of apparent

24

attention were unable to control their writingwell enough to imitate the examiner's demon-strations.

It was striking how few children coulddraw more than the vertical and horizontal linesand the circle; 3-year-olds are normally ex-pected to be able to copy these. Those who werediligent and copied the other forms indicatedoblique directions incorrectly as often as cor-rectly, and paid no attention to where the open-ing in the circle was when making a "C" or a"U." Children who could copy the E had diffi-culty with the number of lines as well as withtheir direction ; they frequently made five paral-lel horizontal lines as a reproduction of thisletter. In general, the older the child, the morecareful the copying and the fewer were thespontaneous scribbles.

Although children were considered test-able (table 1) if only a vertical and horizontalline were produced with fidelity, this degree ofability might not be adequate for an acuitymeasurement. Children seemed to enjoy writ-ing with direction more than indicating direc-tion alone, and direction sense seemed betterdeveloped when the child wrote. The crayonsand paper seemed appealing in themselves, andhaving to do exactly what the examiner haddone seemed to increase motivation.

Hand. The hand card on repeated show-ings elicited some response from only 72 of thechildren ; with these the test was attempted,i.e. instruction in directions was pursued.Most of the '72 children enjoyed looking at thehand card and calling it a "hand" or "glove,""mitten" or "elephant."

However, the ability to match directionswas limited, even at conversation distance whiletouching the test card, and the assumption ofany greater distance between examiner andchild diminished this matching ability (seefig. 2) particularly for horizontal (right andleft) positions. Only 42 of the 72 childrencould be considered testable (table 1) by thecriterion used, because only 2 correctly indi-cated positions are necessary to make a childtestable. This was the criterion set for thisstudy, although the usual passing criterion isthree of four positions. Figure 2 shows the

0

lee:4:4-.......-.......v...-...........-

A1130.1.103 2uipuodsaJ uaJputp Jo Jaqumu

25

number of children correctly responding to eachof the four positions of the hand.

Despite some difficulty in coordination,the children enjoyed using their own hands tomatch with the Sjogren hand card. The mostdifficulty was experienced with horizontal posi-tions; children seemed to find it easier to usethe right hand to point left and the left hand topoint right or sometimes vice versa. All thechildren who could indicate left but not right,for instance, were strongly right handed butcould not seem to turn their right hands towardthe right (fig. 2) .

Testability seemed to be related to chron-ological age somewhat more obviously on thistest than on others (see tables 2 and 6). Thetest was attempted with 9 children under 36months, and 8 of them or 89 percent were non -testable; after 36 months, 41 of 63 children or65 percent were nontestable. This decrease innontestability is most striking when 42 monthsis used as the dividing line (fig. 2). The testwas attempted with 42 children under 42months, and 24 of them or 57 percent were non -testable; after 42 months, only 6 of 30 childrenor 20 percent were nontestable.

"E." This test was less appealing thanthe hand, except to one child, and it could be

attempted with only 70 of the 72 children withwhom the hand test was done. The E wascalled a "letter," "fork," "table," "hand,""comb," "tooth," or "nothing;" it generally wasnot the object of much interest. The hand wastried before the E after it became obvious inthe first few tests that the hand was easier andmore appealing. Some of the lack of successwith the E might be attributable to fatiguesince it followed the hand in most sessions;however, the E also seemed to be intrinsicallymore difficult than the hand. Those childrenwIlo could match the E preferred to use the cut-out E rather than their own hands; it seemed tobe a help to match something exactly like theE figure, whatever that was, and there wasadded confusion when the child's hand had toserve as a replica of the abstract and unfamiliarletter.

Visual acuity Eliciting OKN

In general, the children found this testintriguing for a short period of time. In the77 children who were presented with the tapemeasure, an attempt was made to elicit nystag-mus (table 2). The failure to attempt thiswith the other 17 children is attributable to the

Table 6.VISUAL ACUITYINDICATING DIRECTIONS

Testable and Nontestable Children (Breakdown by Sex, Race, and Age)

TestSex Race Age (months)

MF NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Drawing:Attempted 42 47 39 50 __ 2 4 6 4 10 7 4 9 4 7 5 8 5 5 4 3 1 1 _Testable 27 34 23 38 ______ 1 2 2 5 4 3 8 3 4 5 7 5 4 3 3 1 1 _Nontestable

liand:15 13 16 12 ___ 2 3 4 2 5 3 1 1 1 3 ___ 1 ___ L 1 ___ ___ ___ __

Attempted 33 39 32 40 ___ 1 ___ 6 2 9 5 2 8 3 6 4 S 4 5 3 3 1 1 1

Testable 21 21 20 22 _-__________ 1 4 2 2 3 2 4 3 8 3 3 2 3 ___ 1 1

Nontestable 12 18 12 18 _-_ 1 ___ 6 1 5 3 ___ 5 1 2 1 ___ i 2 1 ___ 1 ___ .._

E:Attempted 32 38 31 39 __ 1 ___ 5 2 8 5 2 8 3 6 4 8 4 5 3 3 1 1 1

Testable 9 12 11 10 _________ ___ 1 ______ 2 1 1 3 2 2 1 3 1 2 ___ 1 1

Nontestable 23 26 20 29 _ _ _ 1 ___ 5 1 8 5 ___ 7 2 3 2 6 3 2 2 1 1 _____

26

periodic disappearance of the tape. Thosechildren considered testable (table 1) were ableto follow the instructions and sustain interestlong enough to complete the task. The non-testable children showed partial response andwere easily distracted or did not respond at all.Those who did not sustain interest seemed tofollow the end of the tape rather than fixatethe lines; the response was elicited in the firstdirection or in the first moments but not con-tinued. Those with no nystagmus either neverapproached closely enough or looked past themoving tape. These patterns of response aresummarized in table 7 below by sex, race, andage.

There were four children who had goadhorizontal nystagmus in whom vertical nys-tagmus was stimulated, but at best there were afew jerks of small amplitude, and this directionseemed to have nothing to recommend it.

An incidental observation was that thegreen side of the tape produced noticeablyslower nystagmus than the white side. Thischange of background color or decrease in con-trast may be useful for slowing down theresponse enough to observe it more carefully,

but probably tends to lower the level of acuitymeasurement.

Muscle balanceMuscle balance tests were uniformly dif-

ficult to administer because of inadequate co-operation from the children, and because theyusually called for totally subjective judgmentson the part of the examiner.

Nearpoint of convergence and penlightfollowing. Determination of N.P.C. aid abilityto follow the light was attempted with 80 ofthe 94 children, those who were tested when thepenlight was present and intact. All the chil-dren who were willing to follow the light alsoconverged upon it and since this was one opera-tion, the findings are tabulated together. Only13 children, those called nontestable (tables 2and 8), did not approach the examiner anddid not follow the light.

It was very helpful in gaining coopera-tion on this test to have the child try to "blowout the light ;" this drew the children closeenough for observation, and seemed to be such

Table 7.VISUAL ACUITYELICITING OKNTestable and Nontestable Children (Breakdown by Sex, Race, Age, and Type of Response)

Response Pattern

Attempted

Testable

Equal sustainedresponsebothdirections

Response one direc-tion more thanother consistently_

Nontestable

No responseResponse in one

direction only_ _ _ _

Sex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

38 39 33 44 1 2 2 5 5 11 6 3 7 3 6 4 6 3 5 4 2 1 ___ 1

23 27 24 26 ___ 1 1 3 2 8 3 3 6 2 4 3 4 1 2 4 2 1. ___ __

22 25 22 25 ___ 1 1 3 2 8 3 3 5 2 4 3 4 i 1 3 2 1 __- __

15 12 9 18 1 1 1 2 3 3 3 ___ 1 1 2 1 2 2 3 ___ ___ ___ ___ 1

6

9

9

3

3

6

12

6

___

1

___

1

1

. ._

1

1

3

___

1

2

2

1

___

___

1

___

___

1

1

1

___

1

1

1

2

__

2

1

--_

___

___

___

___

___

___

___

--

1

735-4370 0 -65 -5 27

Table 8.MUSCLE BALANCE

Testable and Nontestable Children (Breakdown by Sex, Race, and Age)

TestSex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

N.P.C. and PenlightFollowing:

Attempted 36 44 34 46 1 2 1 6 4 10 4 5 8 4 5 5 8 5 5 3 2 1 ___ 1Test able 30 37 30 37 1 2___ 4 2 8 3 4 8 4 4 5 8 5 4 3 2 ___ ____Nontestable 6 7 4 9 ___ __ 1 2 2 2 1 1 ___ ___ 1 __ __ ___ 1 ___ ___ 1 ___ 1Cover test:Attempted 25 34 30 29 ___ 2 1 4 3 E 3 2 4 3 4 5 6 5 3 1 3___ 1 1Testable 11 21 19 13 ___ 1 1 _ _ _ 2 3 1 1 1 2 2 2 5 4 2 1 2 ___ 1 1Nontestable 14 13 11 16 ___ 1 ___ 4 1 5 2 L 3 1 2 3 1 1 1 ___ 1 ___ ___ _Red glass test:Attempted 7 5 6 6 ___ ___ ___ ___ __ 3 2 1 1 __ ___ 1TestableNontestable 6 4 5 5 _____ ___ ___ ___ 2 2 1 1 __ ___ 1 __ 2 1 ______ __Stereoscope:Attempted 31 37 28 40 _ _ 1 1 5 3 7 6 3 8 4___ 6 5 7 2 5 3 1 ___ 1Testable 12 )4 12 14 _____ ___ 1 1 1 3 1 3 2 ___ 2 3 3 1 2 2 1 ___ __Nontestable 19 23 16 26 ___ 1 1 4 2 6 3 2 5 2___ 4 2 4 1 3 1 ___ ___ 1

a fascinating phenomenon to them that it madetheir cooperation worthwhile.

Behavior on touching the light at varyingdistances and threading it with the scissorsvaried with general maturity. Those childrenwho were better coordinated did this easilyusually with the index finger of one hand orholding the scissors in one hand. The youngerand less coordinated used the right hand forthe right visual fields, the left for the left, andgrasped the scissors in both hands. In sevenchildren there seemed to be definite tremorsand distinctly poorer coordination to one side.Only one of these children had a significantintention tremor which seemed to be combinedwith a field defect; she functioned normallyaccording to her family, had been a prematureand had had "spinal meningitis." Tne othersix were difficult to interpret, considering othertest results in the same cases.

Cover test: The limitations of the ex-aminer's eye and experience make the accuracy

28

of this test particularly questionable. Impres-sions of imbalance on this test were usually notcorroborated on other tests despite optimal co-operation. The near cover test was attemptedonly with 59 of the 67 testable children on theNPC ; these 59 seemed able to fixate the light.It was possible to make reproducible observa-tion repeatedly in those children called testable(tables 1, 2, and 8), but even in these cases theremight have been minor deviations which werenot noticed. In most cases, it was easier andless ambiguous to observe the position of cor-neal reflections relative to one other, and to ques-tion muscle imbalance on this basis.

Red glass test. Soon after the studybegan, it became clear that the concept of num-ber and color was inadequately developed atthis age to make this test at all useful. The testhad been attempted with 12 children when thedecision to discontinue it was made. Only 2children (testable on tables 2 and 8) gave an-swers which were at all plausible possibilities

and these may well have been fortuitous; theother 10 reported seeing e.g. "three lightsblack and blue."

Stereoscope. The stereoscope was avail-

able when 68 children were tested and an at-tempt was made to get these children to lookinto the instrument and report their observa-tions. Their reactions fell into five categoriesonly one of which could be considered testable(tables 2, 8, and 9). These categories and thechildren showng each type of response are sum-marized in table 9 below by sex, race, and age.Those six who looked and reported (a) unfeasi-ble answers made it impossible to determine ifthere was fusion. The 17 who looked into theinstrument correctly (b), but did not reportany observation were not nonspeaking children,but rather a group made speechless by the task.The eight who never looked through correctly(c), peeked over the top, to the sides, and ap-proached with the wrong eye to the lens.Eleven (d) refused to approach the stereoscopeclosely enough to make an observation.

PerformanceBlocks. This test was attempted with

85 children (tables 2 and 10). When this was

not included, the usual reason was interferencefrom siblings (especially when two childrenwere tested together) or extreme uncooperative-ness. Children who were shown demonstrationsbut either ignored or destroyed them were non-testable, while those who copied or imitated atleast one demonstration were considered test-able (table 1) ; a minimal amount of perform-ance was necessary to make the child testableor comparison with peers possible.

This was usually very appealing to thechildren despite the fact that, in describingfavorite play activities, most mothers said blockshad been outgrown, or confiscated because theyhad been used to hit siblings. The patterns ofbuilding with these large blocks followed pat-terns described by Gesell for smaller blocks.The youngest children could not at all estimatedistance between the two base blocks of thebridge and typically the older 3.-year-old childplaced the base blocks together, then movedthem slightly apart after finishing the bridge.Only the better coordinated children could pilethe six plastic blocks used in this study ; thisrequired some staggering of juxtaposed blocksbecause of the unevenness of their surfaces.Most children built towers of four to five blocksspontaneously, and did attempt to add the lastblocks(s) when the tower fell. Despite well

Table 9.STEREOSCOPE: PATTERNS OF RESPONSE (Breakdown by Sex, Race, and Age)

TestSex Race Age (months)

MF NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Testable:Looks, reports, fu-

sion determinable.. 12 14 12 14 ___ ____ _ 1 1 1 3 1 3 2 __ _ 2 3 3 1 2 2 1 ___ _ _

Nontestable:Looks, reports, fu-

sion undetermin-9 ble

Looks, no reporting_ 5 12 6 11 ___ ___ ___ 3 ___ 1 1 1 4 2 __No correct looking

(at ins trument)__ _ 6 2 3 5 ___ ___ _ __ ___ ___ 2 2 1 ___ __ __ 1 1 1 ___ __ __ _ ___ ___ _ _

No correct looking(several feet frominstrument) 4 7 5 6___ 1 1 1 1 2

29

Table 10. -PERFORMANCETESTABLE AND NONTESTABLE CHILDREN (Breakdown by Sex,Race, and Age)

Sex RaceTest

M F N

Blocks:Attempted 40 45 36Testable 37 43 35Nontestable 3 2 1

Peg bench:Attempted 46 45 41Testable 46 45 41Nontestable

Walking lines:Attempted 9 10 11Testable 7 9 8Nontestable 2 1 3

Ball playing:Attempted 44 45 39Testable 40 42 37Nontestable 4 3 2

Puzzle:Attempted 47 47 41Testable 47 47 41Nontestable

Age (months)

w

4945

4

5050

88

50455

5353

31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

2 3 6 4 10 6 5 8 4 7 5 8 5 5 3 3 1

2 2 6 4 10 5 3 8 4 7 5 8 5 4 3 3 1

1 1 2 1

2 5 7 4 10 7 5 8 3 7 5 8 5 5 4 3 1 1 1

2 5 7 4 10 7 5 8 1 3 7 5 8 5 5 4 3 1 1 1

None

1 2 2 1 3 2 4 2 1 1

2 1 1 3 2 3 2 1 1

1 1 1

1 2 4 6 4 10 .7 5 8 3 7 5 8 5 5 4 3 1 1

1 2 4 5 3 8 6 5 8 3 7 5 7 5 4 4 3 1 1

1 1 2 1 1 1

1 2 5 7 4 11 7 6 8 4 7 5 8 5 5 4 3 1 1 1

1 2 5 7 4 11 7 5 8 4 7 5 8 5 5 4 3 1 1 1

None

developed handedness, the most usual patternwas alternation of hands or use of both handstogether in performing these tasks. The trainwas easily reproduced but most children fai ledto add an engine and placed all six blocks on thesame level.

Peg bench. The peg bench activity wasattempted with 91 children (table 2 and 10) andall of them were considered testable (table 1).The three children who did not attempt thistest were not tested alone, and the other childbeing tested monopolized the game.

There tended to be three outstandingpatterns of response which overlapped onlyslightly :

(1) Most disorganized, uncoordinated, im-mature. Children held the hammer with thehand pronated on the handle near its head, fre-quently had the hammer head on its side, and

30

often changed hands. They banged on thebench and on the solidly placed pegs, settingthese out of their holes, enjoying the noise andthe flying pegs. They turned the bench to itsside, hanged it there, never understood the gameas it was demonstrated.

(2) More organized, coordinated, and ma-ture. Children were less destructive in otheractivities than the first group. They held thehammer in a preferred hand in a more usualmanner with the head upright. They tendedto hit the solidly placed pegs endlessly, holdingthese like nails, with the opposite hand, andnever letting them fly off the bench. Their aimwas good, for the most part. Despite reinstruc-tion, however, these children never showed com-prehension of the fact that a peg could be hitthrough the bench, and showed no evidence ofawareness of this difference in the hole depth..This activity was particularly frustrating to

onlooking parents and older siblings, but thechildren did not seem to think they were miss-ing anything.

(3) Most organized, coordinated, and ma-ture. Children were generally chronologicallyolder. They held the hammer in one hand andhit with a solid stroke and good aim. Theyrecognized the game, took out all pegs and usedthem in turn, banging pegs into the hole andthrough the tunnel in the bench. They inter-rupted their hammering to probe the depth ofthe holes and pull pegs out of the exit hole.

Observation of this activity performedmonocularly gave no new information, but defi-nitely helped to corroborate suspicions based onother tests.

Walking lines. This test was attemptedwith 19 of the more cooperative children (tables2 and 10). This was discontinued, however,because lines were frequently not available, andthe pattern cf response was such that the desiredinformation was not forthcoming. Also, thistest added no information to that derived frombail playing.

Those 16 called testable (table 1) wereable with both eyes and each eye to follow theinstruction to walk a straight line; but the typi-cal response was a running walk with one footto each side of the line, and the rapidity ofmovement made it impossible to detect devia-

tion. These young children were not coordi-nated enough to walk precisely heel to toe; thefew who could be persuaded to do this neededbalance steps to each side. The three childrencalled nontestable did not even begin to walkwhen this test was done.

The mask did not seem to make a differ-ence; once a roughly straight line had beenwalked with the two eyes, the testable childrenreproduced the performance with each eye.

Ball playing. This test was attemptedwith 89 children (tables 2 and 10) to whom theball was given to initiate play. The five chil-dren who did not get a chance at the ball werenot tested alone. Of the 89, 82 were consideredtestable (table 1) and the 7 children who neverheld and released the ball in any way werenontestable.

In general, this was a well accepted andinformative test. Usually the game was playedwith the examiner, but it sometimes helped tohave another child participate, especially inthose situations where more than one child wasPresent. With this particular ball and thesAort distances involved most children were ableto do quite well. This "test" was a treat formany children who are usually forbidden toplay ball in their homes.

Younger children tended to catch witheyes closed, mouth open, and two hands. Thisadvanced to catching with the whole body; and

Table 11. EYE DOMINANCETESTABLE AND NONTESTABLE CHILDREN (Breakdown by

Sex, Race, and Age)

Sex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Atempted 47 47 41 53 1 2 5 7 4 11 7 5 8 4 7 5 8 5 5 4 3 1 1 1

Testable 26 31 23 34 0 2 1 5 0 7 1 3 4 3 4 3 8 4 4 3 3 1 0 1

Near eye dominance 10 14 8 16 ___ _ __ 1 4 ___ 5 ___ 3 2 _ __ 2 1 4 1 ___ ___ 1 ___ ___ _ _

Clear eye domi-nance 16 17 15 18 ___ 2 ___ 1 ___ 2 1 ___ 2 3 2 2 4 3 4 3 2 1 ___ 1

Tontestable (no eyepreference) 21 16 18 19 1 0 4 2 4 4 6 2 4 1 3 2 0 1 1 1 0 0 1 0

31

finally, in the best coordinated, to spotting theball and moving only the arms to reach it. Allthe children could roll the ball although manyhad never done this and preferred the noiserand more dynamic bouncing. Throwing un-derhand was the method used by all except themost coordinated who were able to throw over-hand as well.

It was easy to detect differences in per-formance created by the mask, i.e. binocularversus uniocular performance. This made asignificant difference in most childrena factwhich was taken to indicate binocularity as thepreferred mode of seeing. The tasks were sosimple, though, that adaptation to uniocularitytook place rapidly, usually in a minute. Thecolor and size of the ball made it an easy targetto find even when it was purposely thrown offcourse.

Ball kicking was attempted with 26 chil-dren and was successful in 25; only 1 who wasnontestable on all procedures involving the ballrefused to kick at all. Repeated kicks yieldedreproducible results in almost all cases; it wasassumed that the few who changed feet to kickmight not yet have established foot dominance.

Puzzle. The puzzle was attempted with93 children (tables 2 and 10) to whom it wasdemonstrated, and all of them could be con-sidered testable (table 1) . Even childrentested together made certain they each had aturn at this extremely appealing toy, and other-wise nontestable children enjoyed walking awaywith this and doing some work on it.

As with the peg bench there were definitepatterns of response :

(1) Least mature and coordinated. Chil-dren placed pieces haphazardly over the spacesin the board, turned pieces upside down, andpaid no attention to which piece might fit whichposition. They hammered on the piece with afist or pushed and shoved to get the piece intothe board, but never turned or lifted andreplaced.

(2) More mature and coordinated. Mostof the children who had not done puzzles previ-ously used trial and error first but got eachpiece into its right position after much maneu-

32

vering, usually taking between 3 and 10 min-utes. Turning the board 180 degrees oftencompletely destroyed their learning of the task,i.e. the perception of the board and shapes wasvery concrete and dependent upon the singledirection in which it had been tried.

(3) Most mature and coordinated. Chil-dren who were especially capable and all thosewho had been to nursery school (regardless ofgeneral capability and testability) were able tocomplete the puzzle with minimal errors on thefirst trial in less than 1 minute. Also most ofthese children made no additional mistakeswhen the board was turned 180 degrees.

This test was not informative in propor-tion to the time it consumed. It was very use-ful as a means of gaining cooperation initially,but it seemed pointless to prolong this activityby using occluders since the fumblers continuedto fumble and the better performers were ableenough with one eye to do well.

DominanceEye dominance tests were attempted with

all 94 children; everyone was asked to sightthrough at least one of three devices the card-board, tube, or lorgnette. Observations ofhandedness were made in all cases. Childrenwho inspected the cardboard devices but neversighted, and those who used fingers, mouth, ornose rather than eyes to "sight" were called non-testable (tables 2 and 11) ; there were 37 suchchildren, with a mean age of 38 months, whoeither had not established or did not demon-strate any eye preference with this test. Theother 57 children called testable (tables 1, 2, and11) showed some eye preference; 24 childrenwith a mean age of 39 months had not estab-lished a dominant eye but seemed definitely toprefer one over the other; and 33 children witha mean age of 42 months had clearly establisheda dominant eye.

In general, the lorgnette was preferredto the cardboard a-id the cardboard to the tube.Those children who had not clearly establisheddominance would use the favored eye whenholding the cardboard with both hands or thehand of that side, and would use the nose oreven the opposite eye when using the opposite

Table 12.PATTERNS OF DOMINANCE (Breakdown by Sex, Race, and Age)

PatternSex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Nearly establisheddominance:

R eye, R hand (15).. 4 11 5 10 ___ ___ ___ 4 ___ 4 ___ 2 1 ___ ___ ___ 2 1 ___ ___ 1

L eye, L hand (6)___ 3 3 2 4 ___ ___ 1 __ ___ 1 ___ 1 1 ___ 1 ___ 1

R eye, L hand (1)_ _ 1 0 0 1 ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ 1 ___ ___ ___ ___ ___ ___ --L eye, R hand (2).._ 2 0 1 1 ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ 1 1 ___ ___ ___

Clearly establisheddominance:

R eye, R hand (21)_ 12 9 9 12 ___ 1 ___ 1 ___ 1 ___ ___ 2 1 1 ___ 3 2 3 2 2 1 ___ 1

L eye, L hand (2)___ 1 1 1 1 ___ ___ ___ ___ ___ ___ ___ ___ __ 1 ___ 1 ___ ___ ___ ___ ___ ___ ___ --R eye, L hand (1)_ 0 1 1 0 ___ 1 ___ ___ ___ ___ ___ ___

L eye, R hand (9).._ 3 6 4 5 ___ ___ ___ ___ ___ 1 1 ___ ___ 1 1 1 1 Y. 1 1 ___ ___ __ __

hand. Perhaps because of naivete on the partof subjects and examiner, these findings seemedvery reproducible.

Based on the observations of handednessand determinations of eyedness, table 12 showshow many children preferred the same hand andeye, i.e. were "straight" dominant, and howmany showed preference for an eye contralat-eral to the hand, i.e. were "mixed" dominant.Of the 24 with nearly established eye domi-

nance, 21 (15 right and 6 left) preferred theeye on the same side as the favored hand (andfoot where this was tested), while the other 3(1 right eyed, left handed, and 2 left eyed, righthanded) showed eye preference opposite to handpreference. Of the 33 with clearly establishedeye preference, 23 (21 right and 2 left) showed"straight" dominance and 10 (1 right eyed, lefthanded, 9 left eyed, right handed) demonstrated"mixed" dominance.

Table 13.COLORTESTABLE AND NONTESTABLE CHILDREN (Breakdown by Sex, Race, Age,

and Type of Response)

ResponseSex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Attempted 42 42 34 50 1 2 4 7 3 10 43 5 8 4 7 4 8 3 5 3 1 ___ ___ __

Testable 16 20 18 18 0 2 0 2 0 3 1 4 3 1 5 3 6 2 2 1 0 ___ ___ __

Matches onlyMatches, knows

names, but can-not apply them 5 6 5 6 ___ ___ ___ 1 ___ 1 1 2 1 1 2 ___ 2 ___ ___ ___ ___ ___ ___ __

Names and matches.. 4 7 6 5 ___ ___ ___ ___ ___ 1 ___ 2 ___ ___ ___ 3 2 1 2 ___ ______ ___ _ _

Attended nurseryschool

Nontestable 26 22 16 32 1 0 4 5 3 7 5 1 5 3 2 1 2 1 3 2 1 ___ ___ __

33

Foot dominance was noted in 25 of 26children who were asked to kick the ball. Use ofone foot only was less common than use of onehand. This finding was considered when deter-mining sidedness. Since this was not a system-atic test, however, the findings were much lessimportant than observations of handedness.

Color

Color awareness was looked for, i.e. thistest category was attempted, in 84 children. Nospecific observation of color sense was made in10 children (tables 2 and 13) who did not playwith colored toys long enough or quietly enoughto have made this feasible. The 36 childrencalled testable (table 1) showed various degreesof color awareness which are summarized intable .3. The other 48, nontestable children,showed varying degrees of color awarenesswhich are summarized in table 13.

Colors, like numbers and the alphabet,were frequently learned by rote either from anolder sibling or a parent anxiously attemptingto climb the educational ladder. The ability toname colors varied with such previous drillrather than with age, and it is noteworthy thatat least half of the children who had learnedcolor names still could not correctly use them atthis age. These were often the same familieswhere mothers insisted children count and theresult was a duly recited "one, six, seven, thir-teen, nineteen years old !"

Findings on questionnaire

The questionnaire was attempted withall and answered satisfactorily in 90 cases, thoseconsidered testable in table 2. The four un-satisfactory responses were the only situationswhere a parent was not the informant; two werea foster mother, one a grandmother, and one anaunt. The parents in many cases were probablynot more reliable than these other informants,but they at least attempted to answer thequestions.

There were 19 children (table 14) whosemothers reported symptoms or signs possiblyindicative of eye problems. Fifteen of these

34

O

0/

O

1.1

Co

C71 1.1

ti

CO

4 eq

1.4

Co 1.1

eq

1.4

0

C71Co

C4

00Co

1.1

tiCo

1.1

Co1.1 eq

Coeq

Co

CoCo

C4Co

1.1

a)0,

Co

teD 1.4 4,

Co C4 CO

a)

3oHz

CO 0 CeJ

La el

00 el 00

nineteen had at some time had a turned eye.Intermittent turning in the first 3 months of lifewas not considered in this tabulation. Occa-sional turning of one eye, most noticeable withfatigue and fever, was the most frequent report.Two twin girls had eyes which, according to themother, "converge for 15 minutes every morningon awakening, a family trait." Several mothersreported they had not noticed any deviation buta neighbor had seen the eye turn. Most of the15 said this turning had become more evidentor had had its onset in the last year, i.e. in thethird or fourth year of the child's life. Onlyone child had constant strabismus and this wasfirst noticed several months before examination.The four children without turned eyes had torti-collis, clumsiness with and without excessiveblinking, tearing and "floating out" of an eye.In 15 of the 19 children with suspicious his-tories, there was some family history of eyeproblems and heightened awareness and interestin the child's eyes; tables 14 and 15 thereforeoverlap in many cases. Testability of thesechildren used in compiling table 14 (and 15 and16) was overall testability, but primarily testa-bility on differential visual acuity testing, i.e.on tests which detect low vision and possibleamblyopia.

A total of 57 children (table 15) had apositive family history. A variety of visualproblems was reported : 17 children had a totalof 20 siblings with strabismus (all but 4 not de-tected until school vision tests). Six parentsand six aunts, uncles, a ^,ousins had strabismusas well. Longstanding refractive errors existedin 31 parents; and 29 siblings (of 25 children)wore glosses presumably 18 of these aiding thecorrection of stabismus. Other severe visualproblemsamblyopia, cataracts, traumawerereported for five parents and four siblings.

Developmental histories could not be col-lected. Except in cases of flagrant slow devel-opment, e.g. lack of language, all mothersclaimed this tested child was faster than hissiblings ! Developmental milestones wereaverages of all the children in the family ratherthan individual steps, and the time of toilettraining tended to be the primary individual-ized fact of development retained by mostmothers.

3

N

4

00M

cqM

M

)0

ell CNI

MM

Cq

1

M

M

z

O

Hz

35

36

cr4

+.1 +.4

9+1 r+1

+.4

"DI

cr) +.1

co +.1

+.4 +.1

ev

+.4

Ten of the 94 children (11 percent com-pared to the national average 7 percent) hadbeen premature by birthweight (table 16) ; five

had required oxygen and prolonged hospitaliza-tion. Four of five nontestable children hadbeen severely ill during infancytwo with men-ingitis, one with some hemorrhagic episode, and

one with neonatal convulsions.There were two children, not prematures,

who had also had prolonged illnesses requiringhospitalization : a girl with Vitamin D resistantrickets, and one with jaundice and fever of un-known origin ; one was testable (table 1) , andthe other, a twin, was nontestable primarily be-cause of environmental confusion. Many of thechildren in the study had spent short times inthe hospital for minor illnesses and operations,but the exact frequency of such occurrences isdifficult to state; the mothers were surprisinglyvague about such things, tended to confuse their

children.

Composite test findings: Testability

and nontestability

Visual acuityVisual acuity tests are compared in figure

3; while this comparison is observed, the dif-ference in the tests themselves and the, criteriaof testability should be recalled (table 1) . Eachbar represents the number of children of thatparticular age with whom the test was at-tempted; the area shaded represents the numbertestable with each visual acuity measurement.Testable for all five sets of bar graphs does notmean actual measurements were made butrather that acuity might be measured with thetest in question. These values might be over-estimates; children called testable for the pur-poses of this compilation would not necessarilybe testable if a formal test of the same type wasadministered. The values plotted in figure 3called "testable" really represent the ability ofthe child to communicate his understandingof the particular visual acuity testwithoutan occluder, and without any distance from the

examiner.The five bar graphs on figure 3 were

plctted from the findings on the Stycar minia-ture toy test for two eyes, the Stycar matchingletters, the hand direction test, the picture nam-ing, and the opticokinetic nystagmus.

The top graph representing the minia-ture toy test is noteworthy for the high ratio oftestable children even for the younger childrentested, and for its overall high testability com-pared to the other four tests. The Stycar lettertest and the OKN seem to show less of a definiteincrease in the ratio of testable children withage than do the picture naming and direction,indicating tests. Picture naming shows highlevels of testability but not as high as the toytest. Ability to perform on picture and direc-tion tests seems definitely age dependent, andability to perform on Stycar letters seems lessobviously age dependent. However, it shouldbe noted that 17 of the 26 children who weretestable (table 1) on the Stycar letters wereolder than 42 months (tables 2 and 3). OKNtestability seems less age dependent or at leastnot dependent on development which normallyoccurs during the range of ages covered in thisstudy; the percentage of testable children atmany of the younger ages is the same as at thehigher extreme.

The Stycar letter test and the directiontests are somewhat more comparable to eachother than to the others since they seemed to

require more abstract thought from the child.The results of the direction tests were betterin general, but the ability to match letters andthe ability to indicate directions did not neces-sarily overlap. It can be seen from table 17that 17 children could do both, but 25 childrencould indicate directions with tie hand and notmatch letters while only 9 could match lettersand not indicate directic ns sufficiently well tobe considered testable (tables 1 and 2).

There was only one child who was notcompletely testable with miniature toys (test-able with two eyes but not with each eye) andwas testable on any of the other visual acuitytests; this girl had good direction indicatingability. The nonspeaking children, who werenontestable on picture naming, were usually alsonontestable on Stycar letters and direction tests;several were testable with miniati tre toys andone could indicate directions.

Those children found testable on theOKN test (tables 1, 2, and 7) were not the samechildren as those considered testable on otheracuity tests; the children in whom OKN couldnot be elicited were often testable with othermethods (table 18).

If an objective measure of acuity likeOKN could be refined, and performed withoutan obvious occluder, some of the presently non-testable children might thus be tested effec-tively. A combination of this and the minia-ture toy test, for instance, yielded a 17 percent

Table 17.COMPARISON OF ABILITY OF MATCH LETTERS AND DIRECTIONS (Breakdownby Sex, Race, and Age)

Sex Race Age (months)

MF NW 31 32 33 34 35 36 37 38 3i 40 41 42 43 44 45 46 47 49 53 54

Matches letters andindicates direc-tions (17) 10 7 9 8 1 1 1 1 1 3 2 1 2 1 1

Matches letters anddoes not indicatedirections (9) 4 5 2 7 1 1 2 2 2 1

Indicates directionsand does not matchletters (25) 11 14 11 14 2 2 1 2 2 3 2 5 2 1 1 1

37

FIGURE 3: COMPARATIVE TESTABILITY BY AGE FOR VARIOUS VISUAL ACUITY TESTS

cr) +.1

14

+.1 CID

1+1

C:bCO

00CO

tiCO

COCO

CO

N ...4 1-4 N

= N 1-4 N

1.4 _4 1-4 N

CO

CO

eg

a)

z

si)rr

+.1

+.1

+.1

+.1

co1+.1

0+.1

r` 1+1-4

CO+1

C1 r` Cq

911 CO r.. LO

0aN1

CO ti 00

3oE-4

Z

, .1 z z

ffi

0 0 wz'amz- ocoo.:1w 1c=wz.ci. m a)o 7:11 715

CS -45 :4 Isa40.4-14,.a 0

, 100Z

nontestable rate compared to the 35 percent non-testable with either test alone.

Muscle balanceTestability on muscle balance U-As did

not correlate well with testability on othertests. The children who cooperated with thesetests so that they could be attempted had allbeen testable with at least the miniature toyacuity tIst. The response to the stereoscopemade 26 children testable (tables 1, 2, and 9),and these were all children who were testableon a number of acuity tests and did especiallywell on performance tests. Child who lookedinto the stereoscope and did not speak wereslightly less testable on acuity and performancetests in general, but were not the nonspeakingchildren found on picture naming tests. Devia-tion of one eye at the near point of convergencehas been used as an indication of eye domi-nance, but a comparison of the deviating eyeand dominant eye based on sighting tests (table19) did not show such a relationship.

PerformancePerformance tasks were rated all to-

gether for summary purposes (see table 20).The ratingsexcellent, very good, good, fair,and poorcould be applied because the abilityto perform one task well was directly related tothe ability to perform other tasks involving theuse of forms and space. Unusual facility at onetask usually meant that the child had had pre-vious experience with the particular puzzleused, and in these cases, poorer performanceswere weighted more heavily in the final rating.In general, older children received higherratings. The 12 children rated excellent on thebasis of an expected 36 month norm (based onGesell) had a mean age of 43 months; the 7rated very good had a clean age of 40 months;the 26 rated good had a mean age of 39 months;the 24 rated fair had a mean age of 39 months;and the 25 rated poor had a mean age of 38months. Females rated higher than males.

This summary rating on performancetasks expresses motor coordination and thechild's ability to act upon his perceptions of

39

Table 19. NEARPOINT OF CONVERGENCE DEVIATING EYE vs. EYE DOMINANCE BYSIGHTING (Breakdown by Sex, Race, and Age)

TotalSex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 4 46 47 53 54

Deviation, domi-nance:

Deviating eye,plus dominanteye 24 12 12 8 16 ___ 1 ___ 2 ___ 1 1 1 4 1 1 2 2 ___ 3 3 2 __ _

Deviating eyeequals domi-nant eye 15 5 10 10 5 ___ ___ ___ ___ ___ 2 2 ___ 1 1 1 1 3 2 1 ___ ___ -__

Deviating eye,no dominanteye established_ 19 8 11 5 14 1 ___ ___ 2 2 3 __ 3 1 2 1 1 2 1 ___ __ ___ ___ __

Deviating eye "L" 13 6 7 4 9 1 _____ 1 1 2 ___ 2 1 1 1 1 1 1 1 ___ _____ ___ __Deviating eye "R" 2 ___ 2 ___ 2 ___ ___ ___ 1 ___ -- __ ___ -_- ___ ___ --- 1 ___ ___ ___ ___ ___ __Deviating eye, alter-

natesNeither deviates 9 5 4 6 3 _ _ _ 1 ___ _ _ _ 2 __ . ___ 1

space. The ratings were roughly correlatedwith overall testability, but there were childrenwho were more testable on acuity tests andothers who were uniquely testable and capableon these more active tests.

There was an overlap between superiorperformance ratings and ability to indicate di-rections this correspondence was not so strikingfor other tests with limited testability. Twenty-seven of the children who were testable on thehand direction test (tables 1, 2, and 6) receivedratings of good, very good, or excellent in over-all performance. However, 15 others were notconsidered so well coordinated; and 18 withsimilar superior performance (including 1 ex-cellent) ratings failed to indicate directionsadequately.

Dominance

There is a striking relationship betweenthe lack of dominance and overall nontesta-bility particularly nontestability on differen-tial acuity tests (table 21) : of the 37 childrenwith no eye dominance demonstrated, 33 wererecommended for retests, usually because im-

40

maturity and occluder rejection had preventedcomplete appraisal of the vision of each eye.Of the 57 children who showed some eye pref-erence, 18 were recommended for follow-up, butonly 2 of these 18 were considered nontestablebecause of immaturity ; the others had been test-able and received this recommendation for otherreasons, e.g. a history of strabismus, question ofamblyopia, or poor environment for tffiting.

Color

Ability to match or name colors did notseem causally related to testability. There wasa general relationship in that children whoknew colors were older and more testable forthat reason (tables 1, 2, and 13). Inability toshow recognition of colors did not seem to spe-cifically decrease performance on other testswhich contained colored materials '-nit did notdepend upon their color. The childre who hadbeen to nursery school did better at tois (table13), perhaps because of a more meaningful ex-perience involving colors. However, nurserychildren also tended to be somewhat older

a

Tab

le 2

0.PE

RFO

RM

AN

CE

TE

STS:

CO

MPO

SIT

E R

AT

ING

S (B

reak

dow

n by

Sex

, Rac

e, a

nd A

ge)

Rat

ing

Tot

alSe

xR

ace

Mea

nA

ge

Age

(m

onth

s)

MF

NW

3132

3334

3536

3738

3940

4142

4344

4546

4749

5354

Exc

elle

nt12

48

57

43__

___

___

___

___

_1

___

___

1__

___

_3

21

1__

_3

___

___

___

Ver

y G

ood

73

43

440

___

___

___

___

12

___

_

Goo

d26

1412

1115

39__

___

_2

2__

_3

22

22

3__

_3

21

2__

___

_

Fair

2410

1412

1239

___

11

3__

_3

11

41

11

11

21

___

___

___

__ _

Poor

2516

910

1538

11

22

32

42

11

21

Tab

le 2

1.E

YE

PR

EFE

RE

NC

E A

S A

N I

ND

ICA

TIO

N O

F T

EST

AB

ILIT

Y (

Bre

akdo

wn

bySe

x, R

ace,

and

Age

)

No

eye

pref

eren

ceno

ntes

tabl

eN

o ey

e pr

efer

ence

test

able

Eye

pre

fere

nce

nont

esta

ble

Eye

pre

fere

nce

test

able

; fol

low

-up

rec

omm

ende

d _

Eye

pre

fere

nce

test

able

; no

follo

wup

Tot

al 33 4 2 16 39

Sex

Rac

eA

ge (

mon

ths)

M 17

4 1 11 14

F 16 0 1 5 25

N 15

3 0 9 18

W 18 21

31 ____

3233 4

34 2 3

35 3 1 1

36 3 1 3

37 5 2

38 2 ,., 2

39 4 1 5

40 1 2 2

41

3 2 2

42 1 2

43 1 5

44

1 2 3

45

1 2

46 3

4749

5354

1 1

(mean 45 months) than the whole group tested Table 22. PASSING CRITERIA FOR TESTS(mean 39 months).

Composite test findings: Passing andfailing

Although children could be consideredtestable or nontestable on each test, there wereonly some tests on which it was appropriate torate them passing or failing. This is becausemany of the tests were not formalized to the ex-tent of examining a visual ability; the goal ofthese, as stated previously, was to tell how chil-dren would respond to a test of this type if itwere an actual examination. Some tests wereused to examine visual functions, and childrenwho were testable either passed or failed thesetests. Table 22 summarizes the tests and definesthe passing criterion for each test where this isapplicable.

Visual acuityStycarMatching letters. This test looked for an

ability which was then used in the near and dis-tant letter tests to measure visual acuity. Chil-dren who had the ability were testable (table 1),but did not pass or fail this particular test (table22).

Near letter test. Of the 22 testable chil-dren (tables 1, 2, and 3) only 20 accepted theoccluder for this test and could thus be consid-ered as passing or failing (tables 22 and 23).The three children who failed the near lettertest did not have equal near visual acuity intheir two eyes. Of the three failures (table24) two failed distant visual acuity tests andone of these two also failed the stereoscope andperformance tests. One child failed no othervisual acuity * st; his attention had been ade-quate but less than optimal for this test, but theonly other area he failed was the color tests.

Distant letter test. Of the 10 testablechildren (tables 1, 2, and 3), 4 had failed thedistant letter test (table 23). One of thesealso failed the near letter and miniature toy

42

Test

Visual acuity-5year:Matching lettersNear letter test

Distant letter test..._

Miniature toy test_ _

Passing is defined as

N6 in each eye and equal acuityin 2 eyes.

6/9 or better in each eye andequal acuity in 2 eyes.

Matching all toys (except smallfork) with each eye and thesame toys (equal acuity) inboth eyes.

Visual acuityNaming pictures and toys

20/40 or better in each eye andequal acuity in 2 eyes (onlyapplicable when actual meas-urement made).

PicturesToys

Visual acuityIndicating directionsDrawingHand.E

Visual Acuity ElicitingOKN Equal nystagmus elicited when

each eye stimulated.

Muscle Balance:NPC and penlight

following

Cover test (andHirshberg).

Red Glass TestStereoscope

NPC less than 5 cm. from nose;parallel extra ocular move-ments.

No horizontal or vertical devia-tion under cover or inequali-ties in corneal reflection.

Reporting single red light.Reporting fused picture (com-

bination of right and lefteyes).

Performance..

Blocks

Peg benchWalking lines

Ball

Puzzle

Dominance.

Color

Excellent, very good, or goodcombined ratings (table 19).

When done with occluder, noconsistent deviation to oneside, or inequality of mon-ocular performance.

Consistent correct naming ormatching colors.

Table 23.TOTAL TESTABLE CHILDREN PASSING AND FAILING EACH TEST AND

QUESTIONNAIRE

Test No.testable

Passing Failin

Number Percent Number Percent

V isual acuity-Stycar:Matching letters 26

Near letter test (22*) 20 17 85 3 15

Distant letter test 10 6 60 4 40

Miniature toy test 60 45 75 15 25

Visual acuityNaming pictures and toys:Pictures (67*) 2 1 50 1 50

Toys 29

Visual acuityIndicating directions:Drawing 61

Hand 42

E 21

Visual acuity eliciting OKN 50 47 94 3 6

Muscle balance:NPC and penlight following 67 62 92 5 7

Cover test (and Hirshberg) 32 26 81 6 19

Red glass test 2 1 50 1 50

Stereoscope 26 2', 88 3 12

Performance 94 (45*) 6 (48*) 6 (49*) 88 (52*) 94

Blocks 80

Peg bench 91

Walking lines 16

Ball 82

Puzzle 94

Dominance 57

Color.. 36 33 92 3

*See text for explanation of numbers in parentheses.

tests (table 14) and his failure on this test wascaused by a wide disparity in visual acuity be-tween the two eyes. The other three failureson this test passed other tests. One child had6/9 visual acuity in one eye tested, refused theoccluder so the second eye was not tested; onthe toy test she accepted the occluder and thesecond eye proved as good as the first. Theother 2 "failures" had a difference of one linebetween the eyes although both eyes had goodvision (6/9 and 6/6, and 6/6 and 6/4) ; thesechildren were considered failures for the tabula-tion but not considered worthy of followup.One of these children saw one fewer toy withthe "worse" eye despite its normal acuity and

the other saw one more toy with the eye thattested worse with the letters. Both thesechildren had a strong family history of eyedisorder; each had a parent with unilateralblindness.

Miniature toy test. Of the 60 childrenwho could be tested with each eye alone (tables1, 2, and 3), 15 failed (table 23) this test whenall toys (except small fork) or at least the smallspoon and knife had to be identified in order topass (table 22). Of the 15 failures, 10 failedother tests as well including 1 who failed thecover but had not been testable on any othertests which could be failed. One child failed

43

TABLE 24: FAILURES ON TESTS AND QUESTIONNAIRE BY AGE

Indicates failure.

Blank space means nontestable or pass; only when of particular interest, a sass is noted.

cn

OCI) 0

42

tia)i..a).4,..-,a)4iata)Z

cni..

..-,.4,a)4

...,

2CO

.....A

cn>,oEia)io4..Z

2

cna)140

..-,Q..a 0

..-,

.0k n.,trVcuai

TBi..a).n.0Q

cl) cn> .!.4,

c.) ....,cox

cncn0rIbn

"C"(1)

1:4

a)04oQcn014

MM

a)Q0cts

E14o

.....r...a)ta4

i..o=40

C.)

Questionnaire

0o......-,

.11004 a)g ,1

Eo 0rI Qo cu44 c4

32 I :NI. 4I

.41104.II, * F . Hx. brother w/squint retest

334 II

VOA41,

04 Nontestable retest

34010,001

I41 V.V.;* Premature twin retest

34 1

110.44 4141111.VA. 4%.4, Premature & meningitis retest

364141Il 40.441Il 41$ I PI I+440,

lb: 7

Vel410::%14

F. Hx. sister wjmyopia retest36 Pass

41. 4804 IJI.-:640 ,

-*4 1 F. Hx. ; cleft palate retest

37 Pass kV2 . Premature retest

374ii4.. 41 tSeM;

t it**P. Hx. squint, 2 siblings retest

38 Pas.111.. V

. 4***. F. Hx. squint, 2 siblingsF. retest

40 Pas..... ....4%.%S.:

F. Hx. squint retest

41 Pas .4 nursery

424.411 ....... F. Hx. squint & amblyopia ***none

42 ke.

retest

42 ' 4

1 44 4

HoX44:4 P. Hx. s q u i n t refer

42ye",44.......64.e.

4*11.-.. FF. Hx ***none

43 eeee.noloolgo,4 r4 IN qr.

..', r*'"'44.1141

retest

43,... 11. V4

....-.41*'44........ P. Hx. squint retest

44vegreee......4.4...4...., *----

r43$ P. Hx. squint retest44 Pass

I..**Ivow Premature ***none

45A

WA%4101. At&

V.VO41404OW*1

P. P Hx. clumsiness refer

45 4.11 411i. 11 ill airP. Hx. squint x etest

46,0.s*,

V:. ***none

46 ass 4444401,14 nursery

47Voi4ee, , . a s s ***none

49,,....,4I, V'P P. Hx. squint retest

54%V. :V.V.

Ve%00.4 4 OP .. .. .sII *II. P. Hx. clumsiness refer

Totals 1 3 5 6 1 3

*F. Hx. = Family History **P. Hx. = Personal History ***See text for explanation

44

the questionnaire but no test; four failed onlythe toy test (tables 24 and 25).

The patterns of failure were inabilityto match or name the same toys with each eyeand inability to see the smaller toys with eithereye. Children who recognized all toys witheach ey(3 but could not distinguish the smallfork from the small spoon with either eye werepassed; this distinction was considered beyondexpected visual acuity for the age, approxi-mately 6/9 or 20/30. Children who recognizedaii toys with one eye, however, were failed ifthey did not do so with their other eyeeven ifit was just a difference of one small utensil.Table 25 shows the distribution of children whofailed the miniature toy test.

Visual acuity Naming pictures andtoys

Pictures. Only 2 of the 67 children wereasked to name pictures at a distance with eacheye occluded, both using the Osterberg chart.One child passed (tables 22 and 23) and the

other failed. The child who failed also failedthe miniature toy test because he did not differ-entiate small utensils with either eye; he hadbeen shown the miniature Osterberg chart Andwas ablc to name many more pictures on thisthan the large chart placed 10 feet away. Onthe questionnaire, this child's mother had re-ported that he tended to hold things close to hisface and sit close to the television set.

Toys. Passing and failing do not applyto this test as given.

Visual acuity Indicating directions

Drawing, Hand, "E." None of these testswas given as a formal test for distant visualacuity ; therefore, passing and failing criteriaare not applicable.

Visual acuity Eliciting OKN

Of the 50 testable children, 3 consist-ently showed inequality of response which

Table 25.FAILURES ON MINIATURE TOY TEST (Breakdown by Sex, Race, Age and OtherTest Failure)

Sex Race Age (months)

M F NW 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 49 53 54

Failures_

Failed only toys_Also failed

Near letter3Distant lettersPicture namingOKNNPC, penlightCoverRed glassStereoscopePerformance_History_

10 5 5 10 ___ 1 1 1 ___ 2 _ __ 4 2 1 2 ___ ___ ___ ___ 1

28

23

23

2 ___ ___ 1

1 2 ___ ___ ___ ___ 1

21

4

22

___1

___1

1

1

1

l

2

--

--

45

might be considered failure (tables 22 and 23).Children who responded with nystagmus inone direction but not at all in the other wereconsidered nontestable because this inequalityseemed attributable to poor attention. Thesethree failures, however, repeatedly showedqualitatively different nystagmus in each direc-tion usually a slower less smooth nystagmus6 greater amplitude in one direction than it theother. Two of the three failures showed ine-quality of vision on several other tests (tables24 and 25). The other one was testable andhad equal vision on the minature toy test.

Muscle balancePenlight following and NPC. The five

children who failed this test (table 23) did sobecause of apparent unparallelism during move-ment of the eyes. This impression was verysubjective and was always considered in con-junction with other tests before requesting fol-lowup ; four of the five children could be r-tn-sidered questionnaire failures but faile,. noother test.

NPC of 5 cm. was found in three chil-dren and another eight had an NPC of 4 cm.Since none had NPC greater than 5 cm., therewere no failures on this basis (table 22).

Cover test (Hirshberg). The obser-vation of corneal reflections or Hirshberg testwas used together with the cover in decidingfailures for this test (table 22). There weresix such failures (table 23) ; one failed only thistest (besides color test which was unrelated) andhad a positive family history and five othersalso Liled at least the miniature toy test (table25)'.

Red glass test. Of the two children whowere testable (table 1), one failed (tables 22 and23) ; she gave an answer which indicated sup-pression of the image of one eye and was con-sistent with the findings on other tests, the min-iature toys and stereoscope (tables 24 and 25).

Stereoscope. Of the 26 testable children(tables 12 and 9), 3 indicated by their answersthat they suppressed one image. All three alsofailed the minfolire toy test (tables 24 and 25) ;

4

and each one failed a single additional test:the cover, near letter, and red glass.

PerformanceWhen overall ratings of performance

(table 19) are used, 49 children who receivedfair and poor ratings might be considered fail-ures. These numbers appear in parentheses ontable 23. The failure is in relation to the 36month expected performance. This populationtended to run about 6 months slower thanGesell's "normals;" some tasks were exactly thesame as anticipated, e.g. tricycle riding andfeeding of self were things all children in thisstudy could do, but other tasks, particunrlytested ones, ran behind expectations.

The six children considered failures forthis tabulation wore occluders for at least oneof the performance tasks (ball, line, or pegbench) and demonstrated different abilitieswith each eye or poor ability which did notimprove when the occluder was removed. Threeof these failures also failed the miniature toytest (tables 24 anti 25), two of them showingconsistent deviation to one side and the thirdfailing the cover test as well. A fourth failurealso had unparallel extraocular movements.Two children passed all other tests but were sogenerally uncoordinated with and withoutoccluder that they were considered worthy ofsollow-up on this basis alone.

DominancePassing and failing cannot be applied to

these tests; children who had not establisheddominance were considered nontestable (tables1, 2, and 11) rather than failures (tables 22 and23). Those with mixed dominance might beconsidered "failures" but the evidence did notseem to warrant this.

ColorOf the 36 children considered testable

(tables 1 and 2), there were 3 who obviouslyknew that colors existed and could name sev-eral colors correctly but consistently confusedred and green. Thase children, all boys, wereconsidered failures (tables 22 and 23) and prob-

4

able examples of color blindness or achroma-topsia. All three were testable and failed atleast one other test (table 24), none of whichdepended upon color awareness.

Questionnaire

Of the 90 children for whom there weresatisfactory questionnaires (table 2), a total of66 could be considered failures on the basis ofpersonal history, family history, prematurity

or combinations of these factors (table 23).Of these 66 failures, 25 were generally nontestable, and particularly could not be tested withoccluders on the visual acuity tests. Of theremaining 41 failures who were testable, therewere 18 who failed other tests in addition and23 failed only the questionnaire (table 24).Children who had suspicious histories andfailed only one test, particularly those four whofailed only the penlight following were rec-ommended for followup on the basis that theymight later develop clinically apparent in-equality between their eyes.

IF

47

VI. FOLLOWUP OF SUBOPTIMAL FINDINGS

Children recommended for followupIf all suboptimal or questionable findings

on every test and questionnaire had been giventhe same consideration in formulating follow-up recommendations, all of the 94 childrentested would have deserved followup. How-ever, tests were weighted differently; thosewhich could be failed, especially the visualacuity tests and particularly the miniature toytest were most important in deciding if follow-up was necessary. The other tests like OKNand color provided findings which were of con-cern in many cases, but the significance of thesefindings is still somewhat speculative. Mostchildren who were severely negligent in areasother than visual acuity also failed the toy testand were therefore included among those rec-ommended for followup.

Since one of the original objactiv es ofthis study had been to survey the W.C.C. chil-dren for amblyopia and since the recommenda-tions were made on W.C.C. records, it seemedwise to limit this record entry to factual find-ings and to base the recommendation on whetherthere was low vision in one or both eyes. Chil-dren who were nmtestable or failed the minia-ture toy test constitute most of the 54 children;children who passed the toy test were recom-mended for followup only if there seemed tobe from history or another test, some imbalancebetween the eyes which might later producea difference in visual acuity. The 40 childrenwho received no recommendation seemed tohave little likelihood of developing amblyopia.

The summaries for the records includeda statement of pertinent history, acuity in botheyes, or left and right eye, a statement about

48

testability with picture charts and direction-based tests a statement about eye preference,and a rating of performance. A sample sum-mary of a case follows :

Date of test. Vision Study.

Visual acuity leftright eye with min-iature toy test. Names pictures withunusual terms. Direction sense inade-quate for test. Looks right esotropic onfollowing light but shows no evidence ofsuppression on stereoscope. Dominancenearly established left eye and left hand.Fair spatial perception.

Recommendation. Retest in 6 monthsto see if muscle imbalance may producevision loss.

There were three types of recommenda-tion given to a total of 54 children : (1) recom-mendation for immediate referral given to 3children; (2) recommendation for followup innursery school given to 2 children; and (3)recommendation for retest given to 49 children.

1. Recommended for Immediate Re-ferral

Three children (two boys, one girl) werethought to have striking unilateral amblyopiaand they were referred back to the W.C.C. forimmediate referral to an eye doctor. Case his-tories follow ; see table 24 for a summary of testand questionnaire findings and comparison toother children.

G.F., 42-month-old white male had shown

slow development, particularly of language,and had looked "lazy-eyed" for several months.

His mother was considering a visit to Massa-

chusetts Eye and Ear Infirmary ( M.E.E.I.).He did not talk at all during the test session

but did seem interested in games and followedinstructions well with the miniature toy test.Visual acuity in the right eye was much betterthan the left; there was readier acceptance ofmask for occluding left eye. He could not per-form tasksball, peg bench, and puzzlewithleft eye alone and fair performance with theright alone was the same as that with both eyes.Cover test revealed left esotropia. Dominancewas nearly established for left eye and righthand. Two months later he was seen atM.E.E.I. for a fourth visit and Mrs. F. was toldto return in 5 months because he was nontesta-ble ; she had been given "E" cards to train himand found his comprehension of this limited.His speech retardation was studied at Chil-dren's Hospital Medical Center where he is be-ing seen by the psychiatry department and at-tending the Sarah Fuller Foundation ( for

speech and hearing) Nursery School. This at-tendance has been interrupted by family prob-

.lems; parents were separated and child wasunder Division of Child Guardianship forseveral months. Seven months after initial test-ing, at the the mother was told hisvision was "good enough for a 4-year-old," andhe was to return in 6 months.

C.G., 45-month-old Negro female hadalways been A'clumsy" and "sloppy", had hadperiods of excessive blinking and frequently re-quested sunglasses, all of which the mother con-sidered perversity. A 6-year-old brother failedthe school vision test and has strabismus, an 8-year -old sister also failed the school vision testand wore glasses for 2 years because of "slowfocus" and nearsightedness. This child waseasily tested and showed good visual acuity withthe right eye and very poor vision in the lefteye with miniature toys. She had excellentability at indicating directions. The left imagewas not reported on either red glass or stereo-scope tests. Fair spatial perception was wors-ened when right eye was occluded. Dominancewas established for right eye and right hand.

An appointment was made for the child withthe family's optometrist but, when the date ar-rived, "it seemed a shame to waste an afternoonout by taking that brat along," and mother wentalone. Intervening pregnancy has further de-

layed this child's care. Seven months after theinitial test, she was seen by the optometrist whoadvised the mother that vision in the child's lefteye was weak and they should go to M.E.E.I.for treatment. Mother had hesitated to go be-cause of the expense involved and thought shewould go to City Hospital Eye Clinic instead,"in several weeks."

C.W., 54-month-old Negro male who ap-pears to have partial albinism had always beer,slow in development, antisocial, and clumsy,held things close to his face, and frequently hadbloodshot eyes. Testing environment was com-plicated by a younger sibling being tested andconstant disparaging comments about this childfrom observing mother and grandmother. Onthe Stycar distant letter test, he had an acuity of6/4 in the left eye and 6/18 in the right eye,and on the near letter test held the test card 6inches from his right eye to see the small print(6 point) that he could see 12 inches away withthe left eye. On cover test and penlight follow-ing, he showed right hyperphoria. Recom-mendation was for referral through W.C.C.One month later he had not been to W.C.C. andfamily was urged to go directly to an eye clinic.Three months later he was taken to City Hospi-tal by his grandmother who decided while wait-ing in the pediatric clinic that what he reallyneeded was a circumcision and this should bethe order of business. "First things first," shesaid, an "I thought about having him circum-cised long before I thought of getting his eyesexamined." Nine months after the initial test,he had neither received eye care or been circum-cized and the grandmother was waiting untilafter he had had his teeth extracted.

2. Recommended for Pollourup inNursery School

Two children who did poorly on test 6,performance tasks, and were not recommendedfor retest based on visual acuity tests, seemed tohave incoordination particularly inconsistent

49

with their age and other abilities. There was noquestion of amblyopia, but it did not seem thatwithout good coordination, vision could not beoptimal. Nursery school was mentioned byboth mothers and it was decided to follow thesechildren there.

T.E., 41-month-old Negro male had anunremarkable history. Vision was equal withminiature toys. Spatial perception was poor.There was incompletely established left eye andright hand dominance. At nursery school hewas described as awkward, slow, unenthusiastic.

D.O., 46-month-old white male had anunremarkable history. Vision was equal withminiature toys and with near letter test. Spa-tial perception was fair. Dominance was estab-lished for left, eye and right hand. He wasnever enrolled at nui c,!Pry school.

3. Recommended for Retest

Forty-nine children were recommendedfor retests either because of initial nontestabil-ity or failure not severe enough to warrant im-mediate referral. Nontestability was related totwo factors : immaturity and adverse environ-mental conditions. Children recommended forretests and the primary reasons the initial testwas unsuccessful are summarized in table 26according to sex, race, and age.

The 12 children who were testable andrecommended for retest with a question of de-veloping amblyopia showed slight inequalitiesin the acuity of the two eyes, or had a history ofoccasional strabismus and a suggestive stereo-scope cover test. These children were consid-ered potentially amblyopic and particularly inneed of followup. Two children who failed thetoy and distant letter tests (table 24) were notrecommended for retest because the inequalitybetween their eyes was in the range of better-than-expected vision and did not represent sig-nificant loss of vision in one eye. Other find-ings on these children showed that they wereusing their eyes together and had excellent abil-ity to coordinate visual and motor activities.

The 25 children who were considerednontestable because of immaturity includedchildren who were completely nontestable,others who rejected occluders so that each eyecould not be tested separately, and children

50

a)

11

11

11

11 11 11

00

N

c#D

11

z

Nkit N1-1 1-1

who were testable for the purposes of individualtest tabulations but whose general demeanorindicated that their apparent failure was notreal. Such failure seemed attributable pri-marily to immaturity, easy distractability, rest-lessnew, short attention span. There wereseven childen who showed better ability withwhatever eye was tested first and lower toler-ance for occlusion of the second eye.

There were 12 children who could not betested adequately because of the environmentalfactors, both physical factors, e.g. lighting andsocial factors, e.g. excessive interference fromsiblings or parents. Six of these might havebeen too immature to test even if the environ-ment had been different. The other six weremainly hindered by environment alone. Ad-verse environmental factors did not deter allchildren, however. There were some coopera-tive and testable children who managed in themidst of as many as eight other interested chil-u :en to attend to the tests.

Retests

Retests, performed about 6 months afterinitial testing, included two of the originaltests : test of eye dominance with the cardboardin two hands, and the miniature toy acuitytest. These were given as described above.Mothers were asked one question, "Has the childbeen well this winter ?" These tests werechosen because the main 'Jason, for retestingwas to detect additional children with ambly-opia and a simple visual acuity test seemed themost expedient way. The relation between test-ability and dominance (table 21) was furtherinvestigated by the inclusion of an eye domi-nance test. The question concerning the child'shealth was asked in order to discern if there wassome reason for continued nontestability.

Of the 49 children recommended forretest 37 were actually approached for retests.Eleven of the twelve children whose environ-ment had seemed the major deterrent to testingwere not approached at all for retests whichwere performed in the homes; one who had astrongly suggestive history was approached at

vI

vI

oo vI

vI

vI

CY3

CV

CY3

00IYD

t--CY3

z

i Cpl CY3 Cpl

9-1 TM 0

vI vI vI

t-

E1

n1

51

cc)

C1 I1

I1

I1

cr) I1

oo I1

ti

cc)

I1

C1

o)Co

ooCo

tiCo

0:2

N CO

CNI 1 CO

CD

CO ek.1 C)

52

nursery school, after phone contact with themother, but the child completely refused to betested. The 37 children approached for retestsincluded the 12 who were potentially amblyopic(table 27) and the 25 who were incompletelytestable because of immaturity (table 28) .

Three children who received no recom-mendation on the basis of the first test werealso retested to see if they remained testable andpassing and to possibly enable an attempt atprognostigation about previous patterns oi re-sponse. A 46-month-old boy who had relativeamblyopia on the near letter test and hadshowed equal vision on the toy test (table 24)

was retested; it was thought that he might intime develop amblyopia in the eye which wasworse on the letter test. He continued to showequal vision on the toy test 6 months later. Asecond child had identified the same toys withcu,ch eye, but could not differentiate the small:f!ork and spoon on the first test, and it wasthought that she might show some myopia orknow all the utensils when retested a few monthslater. However, she was not testable at all atthe time of retest : A third child had equal andgood vision on the toy test, but somewhat incon-sistent ability on various performance tests; shereceived an overall performance rating of good(table 20) but her ability with both eyes openwas not much better than with each eye aloneand the possibility of alternating monocular fix-ation was considered. On retest, she continuedto be testable and showed equal vision.

Thus a total of 40 children were ap-proached for retests at home; 35 of these wereretested at home and 1 was retested at nurseryschool. The four who were not retested in-cluded two who moved and two refusals. Thechildren retested ranged in age from 37 to 56

months, with a In( dian age of 43 months. Therewere 22 males, 14 females, 12 nonwhite and 24white children retested.

Three children who had previously beentestable were nontesthble on retests for no ap-parent reason; they not only refused the maskoccluders, but refused to take part in any way.Parents of the two who were considered po-tentially amblyopic on the first test were advisedto have the children's eyes checked soon, but 4months later had failed to do so. It is note-worthy that these were the only two children

who had shown consistent deviaiion to one sideon performance tests and for whom the diag-nosis of eccentric fixation might be considered.

The 12 immature children who were non-testable on retest (table 28) had also been non-testable originally ; 11 of these children actedexactly as they had 6 months previously and onecooperated completely excvt that he would notwear the occuluder or allow his mother to coverone eye. Several of these children might havebeen more testable if they had not recognizedthe tester and equipment and immediately re-called previous failure.

In this small sample, testability of boysimproved more than the testability of girls; the6 months seemed to make a greater differenceat this age for the 12 males than for the 10 fe-males retested because of previous immaturity(table 28). There was no correlation betweenrecent illness and change in testability; somechildren had been sick all winter and becametestable during that time while others who wereIN ell and waling strides according to theirmothers remained nontestable.

As in the original tests (table 21) therewas a relationship between eye preference andtestability (table 29). All of the testable chil-dren showed some eye preference, although twoof them would not look through the cardboarduntil after the toy test. Of the 12 immaturenontestable children S failed to show eye prefer-ence (table 29). The two potentially amblyopicnontestable children had had eye preferenceon initial testing and continued to show this, asdid the one previously testable child not recom-mended for follow-up who was retested.

There were three additional referralsoased on the retest failures (asterisks on tables27 and 28). Referral slips were given to thefamilies by the district nurse in an attempt todirect the children tc appropriate care. Theiroriginal test failures can be identified in table

14 I"'

24.Two of these who had seemed possibly

amblyopic on first testing showed slight but def-inite decrease in vision in one eye.

R.M., 56-month-old white male (49months on first test) had a history of intermit-tent right cross-eye for 1 year prior to originaltest but not appearing in the last few months.On initial testing, he seemed to be using both

cO

C1

0

Ct 7.1

00.44

Tt4

CO

TtsI

aiCf.!

0rri

ti

00 r-1

0 C+,

C 0

- 4.44

00 !tl

53

eyes together and the only inequality was on thetoy test where he was able to identify one moretoy with the left than with the right eye. Onretest, he identified all the toys with the right,eye but none of the small utensils with the left,on repeated trials with good cooperation. Heis the second oldest of five children, and 4months after retest, the mother is still planningto take him to 1:n eye clinic but has not ht'.d thetime.

R.B., 42-month-old white male (36months on first test) seemed to have a rightsquint on first testing, but right eye was slightlybetter than left on toy acuity test and he seemedto be using both eyes together. Parentheti-cally, OKN was faster moving right to left thanleft to right. On retest, he consistently missedall small utensils with right eye and recognizedall toys with the left. He was referred to theMassachusetts Eye and Ear Infirmary wherethe mother was planning to take a 9-year-oldsister who had failed the Massachusetts Vision

54

Test. At the eye clinic they were asked to re-turn for refraction, but the family was out whenthree revisits were made and there has been nofollow-up.

The third failure had probably been mis-interpreted on first testing as a case .1 immatu-rity rather than recognized as an ainblyope.

P.G., 48- month -old white male (42months on first test) had never received anywell-child care, and his mother was particularlypreoccupied and uninformative. He refusedboth occluders and was tested with both eyesand with his left, sighting or dominant eye, us-ing the cardboard lorgnette. He would not usethe lorgnette to occlude the left eye hile sight-ing with his right. On retest, he accepted themasks, and was able to identify only two largetoys at 2 feet with the right eye and all the toysat 10 feet with his left eye. Four months afterretest, his mother was waiting for an older childto have some free time so he could take this childto Boston City Hospital Eye Clinic.

Tests

VII. DISCUSSION

Visual acuityStycar

Letter test. The poor response to this testin this study is contrary to previous experiencein Britain.29-31 An estimated 80 percent ofchildren 36-48 months have been able to matchfour to five letters; here, only 44 percent couldmatch a minimum of three letters. The litera-ture on this test gives no breakdown of subjectsaccording to age (no finer increment than one-half year) and cultural background. The testis based on those letters most often used in pop-ular British culture, and it may be that childrenwho have been exposed to these letters but notto exactly the same culture show poorer recog-nition. However, the test is also based uponchild development studies which were done inthis country ; these Gesell norms 95 were estab-lished with a group of upper middle class cul-turally privileged children but are expectedwithin limits to be universal. Children in thisstudy who matched all five letters could notmatch any more than five consistently. Testa-bility depended upon age and it was the olderand generally more capable children who suc-ceeded at this test. The findings are thereforenot inconsistent with those of other authors,bitt neither are they identical.

Sheridan and Pugmire 96 claim that thenormal visual acuity by 5 years of age shouldbe 6/6 or 20/20 and that children who cannotread the 6/6 line are few in number and prob-ably have some visual defect. Some optome-

trists contend that preschool children are nor-mally hyperopic and the child who sees 20/20before age 6 years is abnormal and has a goodchance of becoming myopic at age 10-11 years;plus lenses are, sometimes prescribed for nearwork in the early school grades.12

In this study, children who matched let-ter- -t a distance could see 6/6 and even 6/4.

The usual screening test in this countryuses a referral level of 20/30 for 5-year-olds,20/40 for 3- and 4-year-olds, and sometimeseven 20/50 for 3-year-olds. The normal visualacuity of the preschool child, particularly theyounger preschool child and the infant, is notdefinitely known. The levels used are chosenfor operational facility. Although the 3-year-old may actually see as well as the 5- or 6-year-old, he has less familiarity with visual stimuli,less capacity to interpret what is presented tohim from the visual clues he receives, and lessability to communicate his perception to thetester. By making the size of the target hemust see larger, his chance of passing the testincreases and the number of incorrect refer-rals decreases. However, if indeed the normalvision is 20/20, then 20/50 represents signifi-cant vision loss and some children who shouldbe referred will not be. It would seem, fromthe limited evidence presented here, that 3-year-olds have better visual acuity than they areusually asked to exhibit on vision screeningtests.

Vision might seem better if the targetsand the means of communicating one's percep-tion of the target were made more interesting.In subjective vision tests, partcularly tests

55

for preschool children, motivation plays an im-portant role in the results. The problem ofmatching letters is not so much more difficultthan the problem of matching toys that theresults should he so different (tables 2 and 3) ;but the interest, of the children in these twoproblems and their motivation to perform testsbased on these problems differed widely. Thus,the letters were not intrinsically interesting andthere was little desire to see and match small let-ter targets. Distance and the mirror made thisless appealing and the findings for the distancetest were worse than for instruction in lettermatching and the near test (tables 2 and 3).The mirror was too interesting in itself and dis-tracting. The matching improved when amore appealing method of giving answers wasintroduced in the form of the Lotto game; . 33

motivation to see the small letters increasedwhen seeing meant that a penny could be usedto cover a similar figure on the key card.

The near test was performed with easeby all children who passed the distant lettertest. There is questionable significance to theone child who failed the near letter test andpassed the toy test. This information does notcontribute much to the controversy over thevalue of near vision tests, e.g. for detection ofamblyopia in myopes,26 or for detection of acause of reading failure." Whether or notthese tests should be included in vision screen-ing batteries is an unsolved question. The mainadvantage of the near test would seem to bethe enhanced cooperation at close range. Inthe Stycar near letter test, the smallest print,6 point, was easily read by all the children whopassed the test, and they probably had sufficientaccommodative ability to read much smallerprint. However, here again the idea of makingthe target appealing enters. The miniatureOsterberg chart was intriguing as a whole, butonce the children recognized that it was thesame as the large chart or named a few of thelarger pictures, there was little interest in dili-gently naming pictures on the smaller lines.Thus th3 small picture chart was too small andthe small letter chart could have been smaller.While the near vision tests fog preschool chil-dren might be useful, e.g. for detection of am-blyopia in a myopic child, they cannot, in theirpresent form, replace distant vision tests.

56

Twenty feet, even created by a mirror at 10 feet,is too great a distance for a 3-year old. Tenfeet seems to be a comfortable distance, at leastwhen the test is being given on familiar groundand the targets are interesting; presumablythis is far enough to represent optical infinity.

Miniature toy test. The Stycar miniaturetoy test would seem to be an improvement overother currently used acuity tests for the 3-yearold (table 2, fig. 3). This test is appealing,easy, and Ims the theoretical advantage, notshared by most acuity tests, of testing realvision, i.e. the sort that is useful and bears aresemblance to visual tasks a person performsin life. People do not have to discern the direc-tion of E's at 20 feet, but they do have to spottoys across the room. It also has the theoreticaldisadvantage of being impossible to standardizeaccurately. The targets are such differentshapes and textures that the test itself couldnever be adminiztered exactly the same wayfrom time to time. It would be impossible tocalculate the visual angle subtena 3d by eachtoy, a three-dimensional object which can appearfrom any number of aspects.

Thus some of the features which makethis test attractive to young children are thevery features to which ophthalmologists mightobject. If the test itself is not very accurate,how correct will the referrals be? Unfortu-nately, referrals based on this test have not yetreceived sufficient followup to begin to answerthis question. If, however, children who can-not be tested otherwise, are testable with thistest, it is a worthwhile procedure. If the num-ber of children tested increases, the number ofcorrect referrals as well as needless referralsincreases.

This test was recommended for children20-36 months old, but in this study it, waspopular with all the children (up to 54 months)and also with many older siblings and parents.Although its use may be relegated to theyounger preschool child, it could probably alsoenhance the testability of a number of olderchildren who are given more accurate tests.

The miniature toy test can be admin-istered in under 10 minutes, in most cases isclose to 4 minutes. Fewer toys could be used,but understanding the principle rather than

making the toys match is the time-consumingaspect.

Occluder acceptance (table 4) is a de-finite problem which does not seem soluble bychanging the shape of the occluder itself. Theidea of covering the face and eyes is what seemsto bother the children who reject occlusion,even by mother's liqnd. This characteristicbehavior 98 is not usually overcome by using thelorgnette instead of the mask although thedesire to peek through the lorgnette 25 occasion-ally allowed testing of one eye when the maskhad been rejected. The children who did notflatly reject occlusion were reassured whenmother or the examiner also wore a mask andwhen they were told they could remove itshortly.

Visual acuityNaming pictures andtoys

The most noteworthy finding whichcame to the foreground with these bsts was thelarge number, almost 10 percent, of the childrenwho did not speak at all and the 57 percent oftestable children who did not speak w ell fortheir age. Slow and inadequate language de-velopment seems unusually prevalent in thispopulation. It is impossible to state with anycertainty whether these young ciiildren willlater show more far-reaching language prob-lems, e.g. reading disability or be among schooldropouts for related reasons, but it is temptingto speculate along these lines. Such culturallyunderprivileged children and their familieshave not received the attention of the vari-ous professional groups studying specificdyslexia.93-99

The finding of such children would seemto be an important byproduct of vision screen-ing with subjective tests. These children maybe testable and pass, or nontestable if standardprocedures are used; in such situations, they arenot immediately referred. Yet, their problemmay later involve vision, at least as writtenlanguage. Perhaps they should be referred toa neurologist, or a speech and hearing clinic;they probably should not be ignored.

The nine children in this study who didnot speak were followed. Five mothers hadexpressed concern about this; four seemed to

expect speech retardation, usually because olderchildren had had a similar pattern of duelop-ment. Of the five who had concerned mothers,three boys received evalui,Con for this problem;two at Children's Hospital Medical Center (one,an amblyopia referral) and one at Boston Dis-pensary and Putnam Clinic Nursery, and twostarted to talk well soon after the vision test.

The picture tests were well liked, andthere was no problem of inability to identifypictures when the corresponding toy was recog-nized. Having like toys helped children whocould not talk communicate the fact that theycould see the picture presented, and these testsmight have better success generally if matching,as well as naming, was used as a means ofcommunication.

Ability to name pictures defititely seemsto increase with age, but there is overlap withother abilities like indicating directions (fig. 3).A child who can indicate directions may- not yelspeak well ; most children who speak well canalso indicate direction. A combination of thesetwo tests probably does enhance testability inthe younger 3-year-olds, but not as much as thecombination of the miniature toy and directiontests, or perhaps miniature toy and OKN(table 18).

Visual acuityIndicating directionsDespite relaxed criteria of testability

(table 1) and the use of instructibiFty in thehand test as a base line, the ability of 3-yearolds to indicate directions seems limited. Itseems clear from the direction tests is this studythat most 3-year-olds do not have direction sensedeveloped to the point of testabiliV, especiallyif direction based tests with more than twochoices (up and down) are used. Testabilitymay be increased by requiring two of fourrather than three of four directions to be cor-rectly indicated (fig. 2). There is some indi-cation that 3-year-olds do not memorize patternsin a two-choice guessing game, and so the ac-curacy of the test might not be greatlyimpaired.1"

The hand seems to be easier than the "E,"but this is probably significant mainly for chil-dren around age 3, a transitional period whenthis ability to recognize directions is develop-

57

ing. Then the more concrete symbol of thehand is easier than the strange figure "E."For the (developmentally) older child. the dif-ference is probably not significant, and for theyounger child both the hand and the "E" are toodifficult.

That the fourth year of life is the timewhen direction sense becomes testable has beendemonstrated inadvertently by several pre-school vision surveys.45, 50, 101 Whenever abreakdown of results by age (in increments ofmonths) is included, it is apparent that test-ability increases markedly after 42 months. Inone study 50 testability increased from 25 to 75percent.

In California now a new direction-basedtest,101 the "Do-As-I-Do Clown" is being used.The child matches directions to the clown'shand as it changes its direction. The advan-tage over the Snellen E is a shortened testingtime in children over 48 months. This tends toimprove methods for the older child, butchanges the trimmings alone and does notchange the fact of developmental stages whichpreclude the younger child's testability.

A greater increase in testability is gainedby the use of a conceptually different test likethe minature toys. For instance, 37 of 66 (56percent) children under 42 months and 21 of 25(84 percent) children 43-17 months were test-able with the toys compared to 1 of 8 (12.5percent) children under 42 months and 5 of 10(50 percent) children 43-47 months testablein the Berkeley preschool survey. 101

While the direction tests, especially the"E" and the Landolt ring are theoretically thebest vision tests available, they are not the bestin practice, when young preschool children arethe subjects.

Visual acuity Eliciting OKNThe 3-year-olds studied often showed

nystagmus when presented with a movingstriped stimulus (tables 2 arid 7). The combi-nation of this and another acuity test like thetoy test may be a way of improving testabilityof preschool children which seems self-limitedon subjective tests alone (table 18). The factthat the design for a more refined OKN appa-ratus will have to include an obvious occluder

58

and/or some binocular device (like the stereo-scope) will tend to diminish the success of thistest with 3-year-olds.

In the testing of even younger children,an objective test like OKN is the only standard-ized method now potentially available. Hori-zontal aystagmus which is easier to elicit andobserve should be the basis of such a visiontest. The contrast between the lines and spacesbetween the lines should be great, e.g. blackand white since less contrast may producenystagmus based on seeing every other line andthe level of visual acuity may not be determi-nable.

Muscle balanceNPC and penlight following. NPC and

penlight following did not add much informa-tion; positive findings here were difficult toreconcile with other findings.

Cover test. The cover test was the mostuseful of the muscle balance tests given, underthe circumstances of this study ; poor coopera-tion, inexperience and the fact that only a neartest was given make the findings debatable. Itwas assumed that both corneal reflections shouldbe centered at the same time, but this is a fal-lacious assumption for those children who dohave some lack of parallelism which is withinlimits. The findings of this study indicate that19 percent of testable children on the cover testshow some deviation of one eye. This unusuallyhigh rate of strabismus may be attributable tomisinterpretation of the test, but this may alsomean that these children differ from expecta-tions in this way. Gesell's dynamic retinoscopicstudies 1° have shown that many 3-year-oldsnormally alternate fixation, and function mo-nocularly at times. This apparent strabismuson cover test may then represent a develop-mental stage. Thirteen percent of a group of3-year-olds screened by an orthoptist in Man-chester, England 45 showed strabismus com-pared to 2 percent of school-age children. Isthis possibly a finding which is normal at a cer-tain age, and in most cases tends to resolve itselfwithout symptoms and without amblyopia?The actual prevalence of strabismus at variouspreschool ages is not known and the prognosis

for the development of amblyopia has to bedetermined.

In the group of 12 children retested be-cause of potential amblyopia (table 27), 1 ofthe 2 testable failures had shown muscle im-balance on the cover test (table 24). The poorereye on the retest in each case was the one whichhad seemed to turn in, although the oppositeeye had tested slightly worse the first time.

Of the six referrals, four showed rn ascleimbalance on the cover test and one showed sup-pression of an image with the red glass andstereoscope (table 24). These findings werebetter than those based on the questionnaire :only 2 of the referrals had a history of a_ abis-mus and only 1 was a strong history ; there were13 other children with a positive history but nostrabismus on examination.

Red glass test. The red glass test cannotbe successfully used in a young preschool popu-lation of this type. Its dependence upon colorand numbe;. makes most children nontestable.

Stereoscope. Stereoscopic devices seemnot to have their optimal success at this age;38 percent of the children were testable in thisstudy with a hand stereoscope. The commer-cially produced vision tests, based on a Brewsterstereoscope have many advantages over tradi-tional vision tests done at 20 feet; these ad-vantages include rapidity of administration,uniform lighting, the need for only one tester,and easy storage.1°2, 103 However, older chil-dren seem to be intrigued by sach machinesmore than these younger children were; theyhave been successful as school tests, but have notbeen adequately tried with preschool children.The disadvantage of a poor response outweighsany advantage the machine might hav,.. Dis-tance of 20 feet is simulated in these machines,but children in this study seem to respond asadversely to simulated distance (of the mirroron the Stycar letter test) as they do to actualdistance. Even when a hand stereoscope is usedto detemine fusion suppression of an image,interpretation is difficult. The testable child of3 years sometimes does not report what he seesin sufficient detail or omits details because theydo not interest him, and the decision of whetherthis is real or apparent suppression is not easily

made. Even adults who are known to haveequal and normal binocular vision will reportthat they see something consistent with com-plete suppression of the image from one eyewhen the picture shown to that eye is rejectedby the subject. 104, 105 The stereoscope is mostuseful when the subject reports a single fusedimage and can hold this image ; then, the ex-aminer knows the subject has stereor,,opicvision. When the subject does not report, afu.,:ed image, the examiner cannot be certain thesubject does not usually function with stereo-scopic vision.

This situation applies to other tests ofstereopsis as well. For instance, the fly ob-served through Polaroid glasses appears threedimensional if there is stereopsis and two dimen-sional if there is not. If the subject respondsdramatically to the fly, or pinches his wingsaway from the page he is assumed to be func-tioning binocularly; if he does not show astartled reaction or pinches the wings close tothe page, he may or may not have stereopsis.This test (Ind others like it are difficult to inter-pret with young preschool children who maynot understand the instructions, may not havelearned to be startled by flies, and may actuallybe monocular at times. People of all agesalthough they have binocular vision may ex-press themselves, e.g. in their art work as if theyperceived a very flat two-dimensional world ;this cultural factor hes to be considered whenexamining physiologicd, phenomena.

Performance

These performance "tests" were morepopular than any others with the children.They have limited use as means of detection ofvision loss. However, in the otherwise non-testable child an estimation of visual abilitymay be made from watching activities like theseand this may be the only available subjectivetest. Performance of the same task with andwithout occluders gives some idea of relativevisual acuity and can be used to detect severedefects. This becomes a less refined version oftests for toddlers like the Candy Bead test 21-24and Worth's Marble Balls." 20 Very accurateperformance tests, i.e. active games which arevision tests can probably never be devised.

59

Performance tests are most useful as ameans of observing the chi/ s coordination andthe way he uses his eyes with the rest of hisbody in dealing with his environment. Chil-dren in this study were poorly coordinatedcompared to *.heir peers who were used to estab-lish Gesell's norms.93 Thus the children whocould do as well as Gesell's 36-month- olds hada mean age of 43 months (table 20), i.e. were 7months later in their development; all of theother children were more than . months behindthe expected normal. Thus all of the childrentechnically failed this section. This representsa problem for these children. But is this aproblem with which vision screening should be

concerned ? Is it the role of preschool visionscreening to detect these children with anyvision problems other than amblyopia ? Anybasis of referral other than poor acuity tendsto increase the number of referrals and to detectproblems which are not detected and not treatedby most eye doctors.

The dilemma is reminiscent of the situ-ation with the Betts Sensation and PerceptionTests performed in schools in the 1930's. Thesetests were functional and the referral rate wasabout 85 percent, most considered unnecessaryby the consultant opthalmologist. 102. 106

If the goal of screening is limited to thedetection of low vision, especially unilateralamblyopia, then many problems which are notsuch purely visual problems will be missed.Such a screening program detects children whodo not have two good eyes; those who cannotbe helped will probably suffer little as long astheir one good eye remains intact. It is notcorrect to publicize such a program as a detectorof visual handicaps which may later hinder thechild's school work. In addition to or in con-junction with ophthalmological care, there arechildren who need training in perception, co-ordination, etc. "Visual training" is being doneby some optometrists and receiving much pub-licity,107 but the substantiation of this work isyet to come.93

DominanceThe findings on the sighting tests indi-

cate that children start to demonstrate an eyepreference during the fourth year. This pref-

60

erence seemed to be cons:stent for each childtested and retested. However, it is possible tha:actual establishment of eye dominance occurslater."8 There is some change in laterality be-tween age 3 and 4 years which one may observein nursery schools. Most children in theyounger class (3 years old) eat with both handsand use both almost interchangeably in playwhile in the older class (4 years old) most chil-dren eat with one hand keeping the other be-neath the table and use one hand much morethan the other in play. True handedness, again,may not be fully established until much later,but the preference is shown at this age. Somechildren show strong side preference evenearlier, of course, e.g. the newborn with a strongtonic neck reflex.

The most interesting finding on thesighting tests is the high association of demon-stration of eye preference with testability, par-ticularly the acuity of each eye. Most childrenwho had not developed eye preference rejectedthe occluder and most who were developing eyepreference accepted the occluder (tables 21 and29). Thus a rapid test of testability would bea simple sighting test, e.g. giving the child acardboard tube to hold with both hands aridasking him to look through it. Children whodid not use an eye to sight after several momentswould probably not be testable and there mightbe no need to proceed with instruction on thevision test and struggle with occluders. Ac-cording to the findings in this study, 50 percentof children judged nontestable on the firstattempt by the use of this sighting screeningmethod will become testable in 6 months. Afew testable children will be missed by this time-saving method; there were children who demon-strated no eye preference and were testable(table 21). The use of sighting as a prescreen-ing test should therefore be limited to situationsin which time is especially important.

There is no reason to state from the find-ings in this study that children with "mixed"dominanoe are any less achieving than thosewith "straight" dominance (table 12). The 18recommendations for follow-up from the domi-nance testable, group included 14 children with"straight" dominance and 4 with "mixed"dominance ; the reason for follow-up was somequestionable finding rather than nontestability.

There wa3 no striking difference in the occur-rence of straight and mixed dominance in thetotal testable group between male and female,nonwhite and white children. However, 12 ofthe 18 who received follow-up recommendationswere males, including all 4 with "mixed" domi-nance. Of the three children who were recom-mended for immediate referral, two showed"straight" dominance and the third was"mixed."

Color

Matching by colors is supposed to occurbefore matching of -`orris in development.Howay. in this study only 2 children matchedby color on the miniature toy test an(' 80matchej by form. Ability to match Ind namecolors probably develops later in children whoare culturally underprivileged. Children whoare taught colors at home usually learn to namecolors in the preschool years. However, thechild's bl.ckground must be considered beforeany action is taken for inability to name colors.Harper 109 lists failure to learn colors by age 4years as an indication for ophthalmologicalreferral, but in a group similar to the onestudied, this action would seem premature andproduce many unnecessary referrals.

From the limited vantage point of thisstudy, color did not seem to be an important con-tributing factor to the way children deal withtheir environment in the early preschool yearsunless specific color conditioning had takenplace. This is not to say that color was notappealing to these young children ; they do not,however, seem to be aware of colors as qualitiesof some objects shared by other objects regard-less of the nature of the objects themselves.

Questionnaire

The questionnaire gave much informa-tion which was not directly useful either aspreparation for testing or as data considered forfollowup recommendations. Some of this infor-mation sounded unreliable as it was related andwas not therefore weighted too heavily. If the

questionnaire had been filled in by parentsrather than at an interview, this judgment couldnot have been made. The background informa-tion v bout the child did not seem to fit the childtested in many cases; children described al,"pills" with a very short attention span werecompletely testable while others described as"smart" with long attention spans would notcooperate at all. The most important contribu-tion of the questionnaire was its provision of anopportunity to talk with the mother about thechild and get a feeling about her interest inand expectations for him. This very subjectivefinding might easily differ from another ob-server's and cannot be used as a fact to be col-lated with other findings.

The important facts on the questionnairewere the personal history of signs and symp-toms possibly referable to eye disorders (table14), family history of eye disorders (table 15),and predisposing factors to eye problems inthe child's history (table 16). This high riskgroup of children are those most likely to haveor develop visual difficulties and should be par-ticularly checked by the pediatrician and per-haps periodically by an ophthalmologist aswell. If every child with a suggestive historyin one of these three categories was referred,the referral rate from the questionnaire alonewould be quite high. When there is a stronglysuggestive history, e.g. several siblings undertreatment for strabismus, a questionnaire niightbe a sufficient screening device. However, manycases of visual problems, like amblyopia, areasymptomatic and some mothers are not par-ticularly observar4) factors which lead to under-referral; and only some children with a posi-tive history have visual problems, a factorwhich leads to overreferral.

In this study, the questionnaire was notused by itself in considering who deserved fol-lowup. Children who were testable and passedwere not followed even if they had positivehistories. The questionnaire findings on thesix referrals are illustrative, of the problems aquestionnaire presents. There were 4 childrenwho had symptoms or signs; 2 had turned eyesand 2 were clumsy, sensitive to the sun, andblinked excessively ; there were 15 other chil-dren with symptoms and signs who were notconsidered in need of referral. All 6 referrals

61

had a positive family history but so did 51 otherchildren who were not referred. None of thereferrals had been premature. Obviously allquestionnaire findings cannot be given equiva-lent consideration, and it is helpful to use thisin conjunction with or as a supplement to visiontests.

Referrals

Referrals from this study of vision testswere all based on the finding of unequal visionin the two eyes, i.e. unilateral amblyopia. Therehas been no followthrough on referrals and it istherefore impossible to judge the validity of thetests used from results of professional examina-tion. An ophthalmologist examining these chil-dren would detect abnormalties not demon-strated by these procedures, and some of thesetests give cause for concern in children whowould not be referred to an ophthalmologist.

Recommendations for referral were basedon visual acuity because low vision should in-clude refractive errors and amblyopia on thebasis of refract ive differences, muscular or otherimbalance. Cocroboration of low vision by theophthalmologist and elucidation of the reasonbehind this can be expected from the clinicalexamination. There would riot necessarily beany additional information about the child'sability to deal with the visual stimuli of hisenvironment.

Perhaps it is the nontestable children,those with language problems, or those withpoor coordination who deserve followup.Should the pediatrician choose these childrenas vision problems and send them to the ophth-almologist ? What could the ophthalmologistdo with them. Perhaps there is no disciplineyet developed which can deal diagnosticallyand therapeutically with all developmentalproblems relating to vision; perhaps the prob-lems themselves have yet to be clearly definedand recognized.

The finding of 6 probable cases of am-

Table 30.RATE OF AMBLYOPIA IN VISION SCREENING SURVEYS

Bibliographic

Percentamblyopes

withoutreference Number of Sex Age Percent apparent

No. subjects amblyopia, etiology(strabismus,

trauma.etc.)

94 M & F 2%-4 years 6. 4 33

110, 45 301 M & F 3 years 1, 7 none

1 1 1 245 M & F 1 4 years 1. 2, *1. 6

112 1,572 M & F 3-5 years_ _ * *.6

113 12, 000 M & F School 2

24 25, 000 M & F School . 6

24 1, 200 M & F School 1. 3

114 10, 000 M & F 4-85 years * * *5. 3

115 60, 000 M Army inductees 4. 9 29. 2

116 21, 446 M 17-44 years ****2. 4 20

117 190, 012 M 18-36 years 5. 5 23. 8

*Two separate methods used.**Based on actual cases of amblyopia among followed referrals.

***Population of successive patients at eye clinic un:.:-e other surveys.* * * *Amblyopia defined as 20/70 or less unlike other surveys (less than 20/40).

blyopia in 94 children tested seems to be a largenumber. Table 30 compares the rates of am-blyopia found in different types of vision sur-veys. Although an effort has been made tomake the basis of these rates comparable, thishas not been entirely possible because of differ-ences in the surveys themselves and in reportsabout them. The definition of amblyopia inalmost all studies was corrected vision of lessthan 20/40. In previous preschool and schoolsurveys,24, 45, 110-113 which list amblyopia, the ratevaried from 0.6-2 percent; in the small samplereported here the rate was 6.4 percent. In asurvey of consecutive ophthalmic patients,114the rate is higher as might be anticipated; chil-dren under 10 years had a higher rate of am-blyopia (7 percent) than older patients did.The often quoted rates of amblyopia,115-117 basedupon army induction physical examinations,are higher than those for younger groups ofmales and females. One reasob s that trau-matic amblyopia accounts for almost one-thirdof these cases. It is possible too that the rate ofamblyopia increases with age.

At least two of the six possible am-blyopes were idiopathic; this agrees with otherlarger studies in which 20-30 percent, of a m-blyopes Lid no apparent etiology on examina-tion or history (table 30). In this group ofsix, only one had definite strabismus by historyand anothe had a suggestive history, two othersshowed slight muscle imbalance on testing buthad no such history.

These children are cukurally deprived,

and it is interesting to speculate about the pos-sible role of sensory deprivation in the etiologyof amblyopia. Can it be that when there is noreason to develop optimal binocular vision, itdoes not develop ? Certainly development ofvision after birth depends upon stimulation ofthe visual apparatus; just how stimulated doesit need to be ? Numerous infants in these largefamilies are seen lying .in dark rooms in cribswhich face blank walls and ceilings with bottlespropped in their mouths. The vision of theneglected child is said to develop prematurely;these children are seen under 3 months of agewith little general physical activity and mucheyeing of the environment." This is perhapsa crucial time for the development of goodvision, and the lack of response by the environ-ment to the child's looking may possibly be re-lated to loss of vision. This hypothesis canonly be tested by parallel observations ofinfants and actual visual tests. The fate ofsuch children can only be known by followingthem for many years.

It is interesting that in this particulargroup of referrals there were five boys and onegirl, two nonwhite and four white children.Although the sample is small these findingsseem consistent with other impressions. Girlsare valued more than boys in this population,and may receive more of the necessary attentionas infants. There is some evidence to indicatethat Negroes have better vision than membersof the white rece.:19

63

VIII. CONCLUSIONS AND RECOMMENDATIONS

Possible program for immediate use

Age of children to be screened

Children from 30 months on can bescreened even though about 50 percent of theyounger subjects might be nontestable. It isbetter to start early not only because early de-tection permits more effective treatment,, butalso because it seems to take so long for referralsto seek professional attention.

The recommended tests can be used effec-tively with children less than 3 years old and sohave the decided advantage over more conven-tional methods of increasing the yield of posi-tive early findings.

Tests

In order to save time and test only thosechildren who will probably accept the occluder,a sighting test can be done, then only proceedwith those children who show some eyepreference.

The miniature toy test, which is appeal-ing to the children and makes fewer demandson their abilities to speak and coordinate,should be used. Although less accurate thanother available tests, this test does enable earlydetection of low vision in one or both eyes, andgives an indication of the child's ability to usehis eyes in a way which is important to hisnormal development.

64

PlaceHome. Testing in the home seems to

have decided advantages over testing in theWCC. Many more children can be testedthrough home visits than in the WCC. Al-though some children who at least partially werenontestable for reasons pertaining to the home,there are equally adverse environmental factorsat a WCC, e.g., other children, noise, many newpeople, associations with previous painful visits.Mothers are more often helpful than meddle-some, and being near one's own mother, refrig-erator, and toilet are very important aids tocooperation.

WCC. The major advantage to testingat the WCC is that it takes less time than hometesting where the need for excessive visiting istime consuming. Children can be tested atroutine visits, but many preschool children arenot brought in for checkups. Physical exami-nations for nursery school entrance should in-clude a vision test.

Nursery school. It is feasible to take avision test, particularly such a simple one, toa nursery school. This is a captive preschoolpopulation and vision testing is part of aschool's general health supervision.

Tester. It would be possible for the dis-trict nurse to do vision screening. The factthat the nurse is familiar to the child would

probably tend to outweigh the adverse associa-tion she might have in some children's minds.The main quality necessary would be the abilityto approach the test as a game with a sense offun and Way. Any youthful nurse who enjoysworking in a children's clinic would be capableat this.

It is generally considered unwise to haveprofessional people perform vision screeningbecause of the false sense of security parentsmay get knowing that a doctor tested theirchild's eyes but not realizing that the test wasnot a professional examination. Teachers andnurses frequently do vision screening, and doc-tors should test children's vision in their offices.Volunteers or specially trained lay techniciansare preferred for large scale screening pro-grams.

Major problems of preschool visionscreening programs

Tests

The tests available for screening pre-school vision need to be improved and perhapschanged fundamentally. The best techniquesand screening tests available do not always de-tect those children in need of special attention.There is no good data on the number of chil-dren missed by screening programs, and therehas been more attention paid to eliminatingoverreferrals than to preventing underref-als. Part of the problem is that not all eyedoctors agree upon what constitutes a correctreferral.

Even if some available screening test didcorrelate well with the professional examina-tion, these screening procedures are limited bythe children themselves. Under optimal cir-cumstances, there will be a certain percentageof nontestable children. These nontestable chil-dren may include some with vision problems.

Follow-up

There is a major practical problem ofgetting children who fail screening tests to have

professional consultation with necessary diag-nosis and treatment. A screening program isworthless if there is no followup. Part of thesolution to this problem will involve widespreadpublicity about preschool vision problems andthe need for early eye care. Parents who arethus informed may be more willing to make aneffort to seek appropriate attention. The morecomplex aspect to solving this followup prob-lem is part of the la-zger problemwhy doparents seek and reject medical care from vari-ous sourcesbut this is not in the domain of thepresent investigation.

Areas for further investigation

Objective vision tests

The development of an effective objectivevision test which could be used effectively withpreschool children and infants would answermany questions. It would permit even earlierdetection of low vision, perhaps in the first yearof life. It would aid in defining what is normalvisionsomething which is not really knownfor young children. It would decrease the num-ber of nontestable children because it would re-quire minimal cooperation from the child. Itwould enable children who are deaf or retardedto be tested and thus better separate the prob-lem of communicating what is seen from thefact of having seen it.

Normal visual development

In order to interpret the findings of ascreening test or professional examination at asingle stage of development, it helps to knowwhat normally precedes and what might be ex-pected to follow this stage. If some large groupof children could be followed from early infancywith periodic vision tests which were widelyapproved, then it would be possible to learnwhat is normal, what is pimsual and transitory,what is abnormal and oermanent. Visual de-velopment should not be the concern of one pro-fessional group alone (which has been pro-posed) ; 120 this is an area which would profit

65

from a multidisciplinary approach.Several current longitudinal studies

might be used as structures within which tostudy normal visual development and the pre-valence and prognosis of various visual abnor-malities.

a. Harvard Longitudinal Studies ofChild Health and Development, HarvardSchool of Public Health:

Children, who were followed from birthto maturity, received a visual acuity during thepreschool years, and received the MassachusettsVision Test periodically from age 6 to 18 years.In addition, physical examination notes includereferences to signs of eye disorder. This infor-mation has not yet been tabulated, but somepatterns of visual development are potentiallydiscernible in it.

b. Child Research Council, Denver,Colorado:

The Child Research Council, anotherlongitudinal study of child growth and develop-ment, included vision screening t'sts but dis-continued them because they were found to beinadequately accurate. Again this was a well-studied group of children and visual findings

66

might have been related to numerous otherfactors.

c. Maternal Infant Study, Collabora-tive Research, Perinatal Research Branch,National Institute of Neurological Diseaseand Blindness:

This is a nationwide program which wasstarted more recently than the two previouslymentioned. Here is an opportunity then tofollow children from a very young age. Un-fortunately, at present, there is no formal visiontest included in the protocols.

Current preschool screening pro grams

The Brookline Health Department is anexample of an ongoing vision survey of severalthosuand preschool children. It includes oph-thalmological examinations of children referredfrom screening, and could serve as a basis for astudy of diagnosis, treatment, and prognosis ofpreschool vision problems.

Only after the expected development isknown can the effect of various treatments beevaluated. There should be long term followupof cases treated now, but determination of prog-nosis for future cases should include considera-tion of normal growth and development.

biblicgraphy

1. Gundersen, T. Amblyopia ex Anopsia : A Pre-ventable Form of Blindness. Pediatric Clinicsof North America 2 (2) : 625, 1955.

2. Allen, H. F. Make sure your child has two goodeyes. Today's Health, 1956.

3. King, R. W. Visual screening of the preschoolchild. J. Oklaho:na Med. Assoc. 49 (8) : 341,1950.

t. Blackhurst, R. T. A plea for prechool eye care.J. Mich. State Med. Soc. 56 (3) : 336, 1957.

5. Falls, H. F. Care of the nreschool child's eyes.J. Mich. State Med. Soc. 58 (3) : 412, 1959.

6. Cunningham, F. Preschool vision screening. TheNational Society for Prevention of Blindnesspublication P-253 (187).

7. Lepard, C. W. Early diagnosis of visual impair-ment in children. Postgraduate Medicine 24:101, 1958.

8. Dockery, R. W. The value of vision screening inthe preschool age child. J. Nat. Med. Assoc. 52:122, 1960.

9. Parks, N. M. The usual pediatric visual prob-lems. Clin. Proc. Children's Hospital, Wash-ington 16: 204, 1960.

10. Gesell, A., Ilg, F., and Bullas, B. "Vision : ItsDevelopment in Infant and Child." Huebner,New York, 1949.

11. Gesell, A. Developmental Aspect of Child Vision.J. Pediatrics 35: 310, 1949.

12. Apell, R. J. and Lowry, R. W., Jr. "PreschoolVision: Tests, Diagnosis, Guidance." TheAmerican Optometric Association, Inc., 1959.

13. Hill, K. and Stocks, J. F. Assessment of Vision ofthe Pre-literate Child. J. Maine Med. Assoc.,53: 30, 1962.

14. Gorman, J. J., Cogan, D. D. and Gellis, S. S. AnApparatus for Grading the Visual Acuity of In-fants. Pediatrics 19: 1088, 1957.

15. Brown, C. A. Development of Visual Capacity inInfant and Young Child. Cerebral Pals Bul-letin 3: 364, 1961.

16. ,Schwarting, B. H. Testing Infants' Vision-Ap-paratus for Estimating Visual Acuity of Infants

and Young Children. Amer. 3. Ophth. 38: 714,1954.

17. MacKeith, R. The Baby Who May Be Blind.Trans. Ophth. Soc. U.K. 80: 13, 1960.

18. Evans, J. N. Visual Tests for Infants. Amer. J.Ophth. 29: 73, 1946.

19. Worth, Trans. Ophth. Soc. 21: 245, 1901 cited byDuke, Elder, S. Textbook of Ophthalmology IV,p. 3900, St. Louis, C. V. Mosby.

20. Rychener, R. 0. Vision Tests in Infants andYoung Children. Pediatrics Clinics of N.Amer., February 1958, p. 231.

21. Bock, R. H. Amblyopia Detection in she Practiceof Pediatrics. Arch. Fed. 77: 335, 1960.

22. Bock, R. H. Vision Screening for Infants. Oph-thalmologic Supplement 47: 405, 1957.

23. Bock, R. H. Occlusor Mask. Ame. J. Ophth.48: 847, 1959.

24. Vaughn, D., Cook, R., and Bock, R. H. Eye Testsfor Preschool and School Age Children withSpecial Emphasis on the Detection of Am-blyopia ex Anopsia in Pre-school Children. TheCalifornia Medical Eye Council, Stockton, 1960.

25. Allen, H. F. A New Picture Series for PreschoolVision Testing. Amer. J. Ophth. 44: 3g, 1957.and Lorgnette Occluder. Amer. J. Ophth. 44:262, 1957.

26. Allen, H. F. A Near Vision Test for PreschoolChildren. Amer. J. Ophth. 52: 712, 1961.

27. Allen. H. F. Testing of Visual Acuity in Pre-school Children. Pediatrics 19: 1093, 1957.

28. Osterberg, G. A sight-test chart for children.Acta Ophth. 14: 397, 1936.

29. Sheridan, M. D. Vision Screening of Very Youngor Handicapped Children. Brit. Med. J. 2: 453,1960.

30. Sheridan, M. D. Vision Tests for Young Children,Normal and Handicapped. Spastics Quarterly11: 15, 1962.

31. Sheridan, M. D. and Pugmire, G. E. RevisedVision Screening Chart for Very Young or Re-tarded Children. Med. Officer 98: 53, 1957.

32. Sheridan, M. D. "Manual for the Stycar Vision

67

Test." Distributed by National Foundation forEductional Research in England and Wales,1960.

33. Carlevaro, G. and Ouillon, H. "La Methode Opto-psychopedagogique." Minerva Medica, 1958.

34. Berens, C. Kindergarten Visual Acuity Test.Amer. J. Ophth. 21: 667, 1938.

35. Berens, C. Visual Acuity and Color RecognitionTests for Children. Amer. J. Ophth. 46: 219,1958.

36. Evans, J. N. Optotype for Young Children. Amer.J. Ophth. 2: 425, 1919.

37. Blackhurst, R. T. and Austin, C. A ConstantSearch for Blindness. J. Mich. State Med. Soc.59: 1840, 1960.

38. Sloane, A. E. and Gallagher, J. R. A Vision Testfor Pediatricians' Use. J. Ped. 28: 140, 1946.

39. Sjogren, H. New Series of Test Cards for Deter-mining Visual Acuity in Children. Acta Ophth.17: 67, 1939.

40. Albini, Tavoli Ottimetriche, Napoli, 1885, andSnellen, H. Scala tipografica per measure itvisus. Utrecht, 1862, cited by Duke-Elder, S.Textbook of Ophthalmology, IV, p. 3900, St.Louis, C. V. Mosby.

41. Lancaster, W. B., Refraction, and Motility, Spring-field, Charles 0 Thomas, 1952, p. 216.

42. Kramer, M. E. Clinical Orthoptics, St. Louis,C. V. Mosby, 1953, p. 224.

43. Vaughn, D., Cook, R., and Asbury, T. GeneralOphthalmology, 2nd edition, Los Altos, Lange,1960, p. 196.

44. A Guide for Eye Inspection and Testing VisualAcuity. A Screening Process. The NationalSociety for Prevention of Blindness. Publica-tion P-200 (180).

45. Brown, C. M. Annual Report on the School HealthService, 1960. City of Manchester EducationCommittee. Education Offices, Deansgate."Unilateral Aniblyopia," p. 49.

46. Blum, H. L., Peters, H. B., and Bettman, J. W.Vision Screening for Elementary Schools, TheOrinda Study. University of California Press,1959.

47. Lancaster, W. B., Dunphy, E. B., Leahy, B. D.,Regan, J. J., Sloane, A. E., Yasuma, E. R.,Stuart, H. C., Gallagher, J. R., and Schmidt, W.Standards of referral of school children for aneye examination. Amer. J. Ophth. 37: 696, 1954.

48. Sloane, A. E., Dumphy, E. B., Emmons, W. V., andGallagher, J. R. A comparison of refractionresults on the same individuals. Amer. J.Ophth. 37: 696,1954.

49. Keeney, A. H. "Chronology of Ophthalmic De-velopment." American Lecture Series, CharlesThomas, 1951.

50. Colasuonno, T. M. Preschool vision screening inDouglas County, Oregon. The National Societyfor Prevention of Blindness. Publication P-255(280).

51. Gibson, Eleanor J., et al. "A Developmental Study

68

of the Discrimination of Letter-Like Forms."Unpublished report. Cornell University, July 11,1962.

52. Vernon, D. "A Further Study of Visual Per-cep lion." Cambridge Univ. Press, 1952.

53. Ghent, L. Form and Its Orientation. Amer. J.Psychology, 74: 177,1961.

54. Graham, F. K. et al. Development in PreschoolChildren of Ability to Copy Forms. Child De-ve'iopment 31: 339, 1960.

55. Lark-Horovitz, B. and Norton, J. Children's ArtAbilities Child-Development, 31: 1960.

56. Wohlwill, J. F. Developmental Studies of Per-ception. Psychol. Bull. 57: 294, 1960.

57. Gesell, A. and Ames, L. B. Development of Direc-tionality in Drawings. J. Genetic Psychol. 68:45, 1946,

58. Meyer, E. Comprehension of Spatial Relations inPreschool Children. J. Genetic Psychol. 57:119, 1940.

59. Ames, L. B. et al. Development of Perception inthe Young Child. J. Genetic Psychol. 82: 183,1953.

60. Rush, G. D. Visual Grouping in Relation to Age.Arch. Psychol. 31: (217), 1937-38.

61. Munn, N. L. "The Evolution and Growth ofHuman Behavior." Houghton Mifflin Company,Boston, 1955. Chapter 8, Development of Sen-sory Processes.

62. Peckham, R. H. Visual Discrimination in Pre-school Children. Child Development 4: 292,1933.

63. Rice, C. The Orientation of Plane Figures as aFactor in Their Perception by Children. ChildDevelopment 1: 111, 1930.

64. Newhall, S. M. Identification by Young Childrenof Differently Oriented Visual Forms. ChildDevelopment 8: 105, 1937.

65. Preschool Vision Screening in Louisville (Re-print). The National Society for Prevention ofBlindness. Publication P-254 (189).

66. Walsh, F. B. Clinical Neuro-Ophthalmology, 2ndedition, Williams and Wilkins, Baltimore, 1957.

67. Reinecke, R. D., and Cogan, D. G. Standardiza-tion of Objective Visual Acuity Measurements :

Opticokinetic Nystagmus vs. Snellen Acuity.AMA Arch. Ophth. 60: 418,1958.

68. Enoksson, P. Method for Investigation of OcularDominance Based on Opticokinetic Nystagmus.Acta Ophth. 39: 115, 1961.

69. Von Noorden, G. K. Treatment of squint ambly-opia with the after-image method. Amer. J.Ophtb. 47: 809, 1959.

70. Priestley, B. S. and Hermann, J. S. The respon-sibility of the pediatrician in amblyopia. J.Ped. 63: 190,1963.

71. Schllossman, A. Prognosis and therapeutic re-sults of pleoptics. International Ophthalmol-ogy Clinics 1: 829, 1961.

72. Miles, W. Ocular Dominant*: Methods and Re-sults. Pschol. Bull. 25: 155,1928.

73. Miles, W. Ocular Dominance Demonstrated byUncon:Tious Sighting. J. Exper. Psychol. 12:113, 1929.

74. Miles, W. Ocular Dominance in Human Adults.J. General Psycho'. 3: 412, 1930.

75. Sheard, C. The Dominant or Sighting Eye.Amer. J. Physiol. Optics 7: 167, 1926.

76. Mills, L. Eyedness and Handedness. Amer. J.Ophth. 8: 933, 1925.

77. Berner, G. E. and Berner, D. E. Relations ofOcular Dominance, Handedness, and the Con-trolling Eye in Binocular Vision. Arch. Ophth.50: 603,1953.

78. Pascal, J. I. The Chromatic Test for the Dom-inant Eye. Amer. J. Ophth. 9: 357, 1926.

79. Lavery, F. S. Ocular Dominance. Trans. Ophth.Soc. U.K. 63:409,1943.

80. Wile, I. S. Eye Dominance, Its Nature and Treat-ment. Arch. Ophth. 28: 780, 1942.

81. Buxton, C. E. and Crosland, H. L. The ConceptEye Preference. Amer. J. Psychol. 49: 458, 1937.

82. Gahagan, L. Visual Dominance Acuity Relation-ships, J. General Psycho', 9: 455, 1933.

83. Walls, G. L. A Theory of Ocular Dominance.Arch. Ophth. 45: 387, 1951.

84. Washburn, M. F. et al. A Comparison Betweenthe Miles ABC Method and Retinal Rivalry asTests of Ocular Dominance. Amer. J. Psycho'.46: 633, 1934.

85. Cohen, J. Eye Dominance. Amer. J. Psycho!.65: 634 1952.

86. Hughes, H. An Investigation into Ocular Donal-nance. grit. J. Physiol. Optics 10: 119, 1953.

87. Dolman, Tests for Determining the SightingEye. Amer. J. Ophth. 2: 867, 1919,

88. Dolman, P. Relation of Sighting Eye to Measure-. ment of Heterophofia. Aiiier. J. Ophth. 3: 258,

1920.89. Crider, B. A New Test for

Amer. J. Psychol. 49: 669, 19390. Jasper, H. H. and Raney, E. T.

Lateral Dominance. Amer. J.1937.

91. Selzer, C. A. "Lateral Dominance and Visual Fu-sion." Harvard Monographs in Education #12,Harvard University Press, Cambridge, Mass.,1933.

92. Johnston, P. W. "The Relation of Certain Anoma-lies of Vision and Lateral Dominance to Read-ing Disability." Monographs of the Soc. forResearch in Child Development VII, Serial #32,No. 2. National Research Council, Washington,D.C., 1942.

93. John Money, Editor. "Reading Disability, Prog-ress and Research Needs in Dyslexia." JohnsHopkins Press, 1962.

94. Synolds, D. L. and Pronko, N. H. An explora-tory study of color discrimination in children.J. Genetic Psychology 74: 17, 1949.

95. Gesell, A. and Amatruda, C. S. "DevelopmentalDiagnosis." New York, Hoebner, Inc., 1947.

Eye Dominance.7.

The Phi Test ofPsycho'. 49: 450,

96. Pugmire, G. E. and Sheridan, M. D. The VisualAcuity of Five-year-old Children. The MedicalOfficer 103: 177, 1960.

97. Vetterli, C. H. How good is 20/20 vision? Edu-cation 80 (1) : 41,1959.

98. Wolff, W. The Personality of the PreschoolChild-The Child's Search for Ilia Self. NewYork, Grune and Stratton, 1946.

99. Gallagher, J. R. Specific language disability (dys-lexia). Proc. Children's Hosp., Washington,D.C. 16:3, 1960.

100. Kessen, W. and Dessen, M. L. Behavior of youngchildren in a 2 choice guessing problem. ChildDevelopment 32: 779, 1961.

101. Dr. John Nelson, California State Department ofPublic Health, Prevention of Blindness Pro-gram, Marin County Schools Study. Draft ofan unpublished report, February 1963.

102. Sloane, A. E. and Rosenthal, P. School VisionTesting. Arch. Ophth. 64: 763, 1960.

103. Gentile, J. F. and Johnston, P. W. A comparisonof two school vision screening tests, Amer. J.Ophth. 52: 511, 1961,

104. Pettigrew, T., Allport, G. W. and Barnett, E. O.Binocular resolution and perception of race.Brit. J, Psycho'. 49: 265, 1958.

105. Davis, J. M. Personality, perceptual defense andstereoscopic perception. J. Abnormal Psychol.58: 398. 1959.

106. Oak, L. and Sloane, A. E. The Betts Visual Sen-sation and Perception Tests. Arch. Ophth. 22:832, 1939.

107. Goldman, R. P., They crawl to improve eyesight.Boston Sunday Globe Parade, Mar. 10, 1963,p. 16.

108. Belmont, L. and Birch, H. G. Lateral Dominanceand Right Left Awareness in Normal Children.Child Development 34: 257, 1963.

109. Harper, P. A. "Preventive Pediatrics." Apple-ton-Century-Crofts, New York, 1962. Ch. 6.Communication.

110. DaCunha, D. and Blaxter, P. L. Amblyopia inThree Year Olds. Med. Off. 106: 146, 1961.

111. Ricci, A. Sui Valore practico di alcumi testsproposti per la messa in evidenza della "Amby-opia ex Anopsia" Nella Prima Infanzia. Ann,Ottalm. (Parma) 83(3): 141, 1957.

112. (a ) Hufhines, D. M. Preschool Vision Screeningin Orange County California, I. CommunityOrganization. The Sight-Saving Review31: 210. 1961.

(b) Russell, E. L., Kada, J. M., Hufhines,D. M. Orange County Vision ScreeningProject, II. Ophthalmological Evaluation.The Sight-Saving Review 31: 215, 1961.

113. Bangerter, A. Notre devoir envers les enfantsamblyopes. Bull Soc. Ophtal. Franc. 62(5) :332, 1962.

114. Cole, R. B. Unilateral amblyopia in 10,000consecutive National Health Service patients.Brit. Med. J. #5116 : 202, 1959.

69

115. Downing, A. H. Ocular Defects in 60,000 se-lectees. Arch. Ophth. 88: 137, 1945.

16. Glover, L. P. and Brewer, W. R. Ophthalmologicreview of more than 20,000 men at Altoona in-

duction center. Amer. J. Ophth. 27: 346, 1944.117. Theodore, F. H., Johnson, R. M., Miles, N. E., and

Bonser, W. H. Causes of impaired vision in

recently inducted soldiers. Arch. Ophth. 31:399, 1944.

70

118. Robertson, J. The Psychoanalytic Study of theChild, vol. XVIII, 1962.

119. Karpinos, B. D. Racial difference in visualacuity. Public Health Reports 75: 1045, 1960.

120. Optometric Weekly, Oct. 18, 1956, cited in Medi-ine, and The Public Welfare, The

N :clonal Medical Foundation of Eye Care, 1957.

U.S. GOVERNMENT PRINTING OFFICE : 1965 0-735-670

acknowledgments

This research was supported in part by fundsfrom the Commonwealth of Massachusetts, De-partment of Public Health, Division of Mater-nal and Child Health Services, and conductedwith the cooperation of the Bostoi, HealthDepartment.

The authors are indebted to Drs. WilliamM. Schmidt and Philip W. Johnston who gavegenerous cour,sel and encouragement through-out the study. We also wish especially tothank Dr. Albert E. Sloane and Dr. Pauline G.Stitt, who have been most helpful with every-

/ thing from planning to reviewing the finalmanuscript.

Miss Ann E. Stromberg and her staff atthe Ocular Motility Clinic of the MassachusettsEye and Ear Infirmary provided the opportu-nity to observe and discuss vision testing ofchildren by able and experienced orthoptists.

Drs. Trygve Gundersen, Richard G,Domey, David G. Cogan, Robert Reinecke,John W. Ferree, Lenin A. Baler, Lawrence W.MacDonald, Harold H. Haynes, Peter H. Wolff,Charles Drake, Richard Brodie, F. Good-enough-Pitcher, Miriam F Fiedler, Steward H.Clifford, T. Berry Braze lton, John S. Robey,Mary Louise Scholl, David S. Johnson, andSamuel McLaughlin, all deserve recognition fortheir willingness to demonstrate, discuss, andmake suggestions about various aspects of thestudy.

We also wish to acknowledge the cooper-ation and support of Miss Helen Cohn, Mrs.Ruth Cowin, Mrs. Bay la Steinberg, and theBoston Health Department staff nurses at theBromley-Heath Well Child Conference, MissShirley Glass and Mrs. Susan Mueller.

n.

children's bureau publication number 414 -1964

U.S. DEPARTMENTOF HEALTH,EDUCATION,AND WELFARE

WELFARE ADMINISTRATION

Children's Bureau