Reading Disability With Motor Problems May Be an Inherited Subtype

(1)

Reading

Disability

With

Motor

Problems

May

Be an Inherited

Subtype

Sonya

M. Regehr,

MSc,

and

Bonnie

J. Kaplan,

PhD

From the Departments of Pediatrics and Psychology, University of Calgary, and Alberta Children ‘a Hospital Research Centre, Calgary, Alberta, Canada

ABSTRACT. There is evidence from a number of studies that various forms of reading disability are inherited. The familial patterns of one specific type of reading disability that occurs together with impaired coordination and

bal-ance were evaluated. Ten reading-disabled children with

these motor problems, ten children with only reading

problems, and ten control children without reading dis-abilities were examined along with their siblings and

parents. The control children were matched to the

read-ing-disabled children for age, sex, and family

socioeco-nomic status. A high prevalence of reading and motor

problems was found in the relatives of the children with

reading disability and motor problems, whereas a high

prevalence of only reading problems was found in the

relatives of the children with only reading disability.

There were no differences among the three groups of

children or their siblings on number of pregnancy and

birth complications or prevalence of attention deficit

disorder. It was concluded that reading disability with

motor problems may be an inherited type of reading

disability, distinct from reading disabifity without motor problems, which is also inherited. Pediatrics 1988; 82:204-210; reading disability, motor deficit, genetias.

There is evidence that reading disability in

gen-eral is inherited. The concordance rate for the

disorder is 70% for monozygotic twin pairs and only

31% for dizygotic twin pairs.’ Furthermore, in

nu-merous studies evidence of strong familial patterns

of the disorder has been found. For instance,

Hallgren2 tested 90 families in which one of the

parents was reading disabled and found that 45.7%

of their offspring were also. None of the studies on

reading disability in general, however, has been able

to establish the mode of genetic transmission. For

instance, Lewitter et al3 performed segregation

Received for publication Sept 1, 1987; accepted Nov 23, 1987. Reprint requests to (B.J.K.) Alberta Children’s Hospital Re-search Centre, 1820 Richmond Ed, SW, Calgary, Alberta, Can-ada T2T 5C7.

analysis for 133 family pedigrees and found no

evidence of a single dominant or recessive gene

responsible for the disorder. They concluded that

reading disability is heterogeneous and that any

further attempts at genetic analysis would require

prior classification of the reading-disabled families

into subtypes.

Investigators have conducted research on the

in-heritance of subtypes of reading disability.4 Two

major subtypes have been studied: reading

disabil-ity associated with verbal/language problems and

reading disability associated with visual/spatial

problems. Several types of research indicate that

the verbal/language subtype of reading disability is

inherited. It tends to occur frequently within only

certain families, making a relative of a

reading-disabled individual of this subtype not only more

likely to be reading disabled but also more likely to

be of this particular subtype.’ One group of

re-searchers found evidence that a gene on chromo-some 15 may be responsible for the transmission of

this subtype of reading disability.5 Investigators

have also found evidence of familial patterns for

the visual/spatial type of reading disability,6’7 but

no one has yet isolated a gene that may be

respon-sible for this subtype.

These two subtypes of reading disability are by

no means the only ones that have been postulated8;

they are, however, the only ones that have been

examined for genetic cause. Another subtype of

reading disability which has been hypothesized,9

but not yet analyzed for genetic involvement, is

reading disability characterized by the cooccurrence of impaired balance and coordination. Researchers

have noted that many individuals with reading

disability have motor as well as reading problems:

they have been found to perform abnormally on

tests of balance, coordination, posture, body

equi-librium, spatial visualization, oculomotor control,

(2)

sug-gested that these motor problems are indicative of

cerebellovestibular dysfunction’5 and that the

rea-son these children have trouble reading is because

of impaired ocular fixation and scanning abilities,

as well as limited cerebellar capacity to regulate the

order and speed of sensory input to the cortex.’2”5

The aim of the current study was to examine the

inheritance ofthis subtype of reading disability that

is associated with signs of cerebellovestibular

dys-function. Ten reading disabled children with motor

problems (group RDMP), ten children with only

reading disability (group RDO), and ten control

children with no problems in either area (group

NC), plus the siblings and parents of each type,

were compared on a variety of motor and reading

tests to determine whether (1) the relatives of one

type of child were more likely to be of that type as

well and (2) children in group RDMP appeared to

be a genetically separate disorder from those in

group RDO.

METHODS

Subjects

Ten children from each of the three groups were

selected from referrals to the local children’s

hos-pita! and from the public and private school sys-tems. A total of 74 suitable families were contacted

before 30 willing participants could be found. The

most typical reasons for families declining to

par-ticipate were that the parents were embarrassed

about their own reading disability, that families

were too busy, or that the parents thought that

their child had been overtested already.

The control children were matched as closely as

possible to the children in group RDMP for gender,

age, and family socioeconomic status, using the

socioeconomic index for occupations in Canada.’6

Most families consisted of the child initially re-ferred to the study (the proband), a biologic sibling

of that child, and the children’s biologic parents. In

three cases, however, the father was unwilling to be

tested, and so only the proband, a sibling, and the

mother participated. When more than one sibling

was available for the study, the one closest in age

to the proband was chosen. All children were

be-tween the ages of 7 and 14 years, and each child

from the reading disability groups had a full-scale

IQ score of at least 85 and a performance or verbal scale score of at least 90 as assessed by the Wechsler

Intelligence Scale for Children-Revised

(WISC-R).’7 There were no significant differences between

the three groups or the siblings for age, sex, or

socioeconomic status, according to analysis of

var-iance.

Procedure

Each family member was tested individually for

30 to 60 minutes by one of the authors (S.R.).

Subjects were first given the balance, bilateral

co-ordination, and upper limb coordination subtests

of the Bruininks-Oseretsky Motor Proficiency

Test,’8 the motor accuracy and design copy subtests

of the Southern California Sensory Integration

Test,’9 and a test of fast successive and alternating

finger, hand, and foot movements.20’2’ Next, the

reading recognition, reading comprehension, and

spelling subtests of the Peabody Individual

Achievement Test (PIAT)22 were administered.

The parents were also given a self-report

question-naire concerning their reading history and their

own perception of their current reading ability. In

addition, the mothers were given a questionnaire

concerning pregnancy and birth complications24

and the Conners’ Abbreviated Symptom

Question-naire for attention deficit disorder for both of

their children in the study.

The examiner who tested all of the children was

aware of the experimental hypotheses and the

re-ferral status of the children. Final assignment of

children to one of the three groups, however, was

not determined until after all assessments were

finished. In addition, the objective nature of the

standardized tests was thought to minimize

sub-stantially any potential experimenter bias. All of

the reading and motor tests used in this study had

objective, quantifiable formats (primarily multiple

choice), in contrast to interpretive, qualitative

eval-uations such as those found in ,informal reading

inventories or the typical neurologic examination.

Group Assignment

When probands were initially referred to the

study, they were temporarily assigned to one of the

three groups based on available records. Final group

assignment was determined by performance on the

measures. For classification as only reading dis-abled or reading disabled with motor problems, a child had to have a reading quotient of less than

0.90, calculated as observed age divided by expected

age, where observed age was defined as the mean

score of three PIAT subtests (reading recognition,

reading comprehension, and spelling) and expected

age was defmed as the mean of the child’s

chrono-logic age, mental age, and age for grade. Other

(3)

reading quotient scores of our Canadian sample are

inflated by the PIAT, which overestimates reading

ability. For instance, our control children were

re-ported by their parents to be average readers in school, yet their average percentile scores were high

relative to the American norms: 78.20 on reading

recognition, 81.20 on reading comprehension, and

74.40 on spelling. Furthermore, plotting the reading

quotient scores revealed a bimodal and

nonoverlap-ping distribution with all children referred to the

study as reading disabled scoring below 0.90 and all

children referred to the study as normal readers

scoring above 1.00.

To be classified as having motor problems, a child

had to score at least 1 SD below the test mean in

two of the following areas: balance, bilateral

coor-dination, upper limb coordination, and fme motor

coordination. As with the reading tests, some of the

balance and coordination tests appeared to

over-estimate the children’s ability relative to their

peers, and therefore these somewhat liberal

diag-nostic criteria were adopted. It is important to note

that even with these somewhat liberal diagnostic

criteria, none of the control children referred to the

study were found to have motor problems.

RESULTS

To confirm that the proband groups did differ

from each other on the reading and motor variables

as intended, a multivariate analysis of variance was

done with proband diagnosis as the grouping

van-able and balance, bilateral coordination, and upper

limb coordination (as assessed by the

Bruininks-Oseretsky test), fine motor coordination (as

as-sessed by the design copy, motor accuracy, and foot

and hand movement tests), and reading

recogni-tion, reading comprehension, and speffing (as

as-sessed by the PIAT) as the dependent variables.

Overall, the three groups differed significantly from

each other (F (24, 32) = 12.14, P < .0001), and so

the univariate tests were examined. Significant

dif-ferences were found on each of the seven variables

(P < .0001). Newman-Keuls multiple comparisons

(a 0.05) confirmed that the probands in group

RDMP were impaired on the motor tests relative

to the other two groups, which did not differ from

each other, and the children in the RDO and RDMP

groups were impaired on the reading tests relative

to the control group but did not differ from each

other. Furthermore, there were no significant

dif-ferences among the three groups of probands or

their siblings on reports of pregnancy and birth

complications, attention deficit disorder, age, sex,

or socioeconomic status. The IQ data for probands

in groups RDO and RDMP were also examined,

and it was found that although the probands in the

RDO group had much higher performance (mean

110.0) than verbal (mean 101.9) scale scores (t (9)

= 2.35, P < .05), the probands in the RDMP group

had relatively similar verbal (mean 101.2) and

per-formance (mean 102.1) scores.

A multivariate analysis of variance test was then

performed on the sibling data, with proband

diag-nosis as the between-group variable and the seven

reading and motor tests as the dependent variables.

Overall, the three groups of siblings differed

signif-icantly from each other (F (16, 40) = 3.63, P <

.0005) and specifically on bilateral coordination (F

(2, 27) = 15.75, P < .0001), upper limb coordination (F (2, 27) = 6.95, P < .005), fine motor coordination (F (2, 27) = 8.62, P < .001), reading comprehension (F (2, 27) = 7.07, P < .005), and reading recognition (F (2, 27) = 5.84, P < .008). Newman-Keuls

com-parisons revealed that siblings of children in the

RDMP group were impaired on bilateral and upper

limb coordination relative to siblings of those in

the other two groups, who did not differ from each

other, and siblings of children in the RDMP and

RDO groups were impaired on fine motor

coordi-nation relative to siblings of control children but

did not differ from each other. Finally, siblings of

children in the RDMP and RDO groups were

im-paired relative to siblings of control children but

did not differ from each other, on reading

recogni-tion and reading comprehension (Fig 1).

Similar analyses of parent data resulted in

com-parable separation of groups. Overall, there were

significant differences between the parent groups

(F (18, 32) = 4.38, P < .001), and specifically there

were differences on balance (F (2, 24) = 8.41, P <

.005), bilateral coordination (F (2, 24) = 7.50, P <

.005), upper limb coordination (F (2, 24) = 17.42,

P < .0001), fine motor coordination (F (2, 24) =

17.24, P < .0001), and reading recognition (F (2,

24) = 5.98, P < .01). Newman-Keuls comparisons

revealed that parents of children in group RDMP

were impaired relative to the parents of the other

two groups of children on balance, bilateral

coor-dination, upper limb coordination, and fine motor

coordination and that parents of children in groups

RDO and RDMP were impaired relative to parents

of control children on reading recognition (Fig 2).

For the purposes of drawing up pedigrees of the

30 families (Fig 3), criteria were established for

diagnosing reading disability and motor problems.

For siblings, an average reading and spelling score

below the 50th percentile was required for a

diag-nosis of reading disability. Scores more than 0.5 SD

below test means in two of the four motor areas

were required for a diagnosis of motor problems.

(4)

U)

z

5z

LLfl-cU)

Lu

Za::

U)

18

U)

Lu

c 16

0 C-)

U)

14

4

c 12

10

RDMP

‘:i-’

.RDO T

T

_Fc

‘-.-..---‘ IF.4,

.

,. *

F ::.,

-

.- L’ .;:

;,:

BALANCE

BILATERAL

UPPER LIMB

COORDINATION

READING

RECOGNITION

cOMPREHENSION

FINE MOTOR

COORDINATION

Ld

0 0

Li

-I pa:

zO

L&jL

0

a:

Lu

Fig 1. Means and standard errors of three groups of

siblings on motor and reading tests. Only variables for

which univariate analyses revealed significant group

dif-ferences are displayed. An asterisk indicates that two

groups marked by brackets differ from each other at P <

.05 by Newman-Keuls post hoc test. Abbreviations:

RDMP, reading disability and motor problems; RDO,

reading disability only; NC, control.

because (1) evidence discussed earlier indicated that

the children’s scores were being inflated by both

the reading and motor tests, and (2) it has been

argued that, in a genetic study, to diagnose mild

reading disability in family members as the absence

of a reading disability would be misleading.27

Clas-sifying parents required using the control adults in

this study as reference, because many of the tests

used did not provide norms for adults. Reading

disability was diagnosed in parents if they had an

average reading and spelling score at least two grade

levels below the average score for the control par-ents. For a diagnosis of motor problems, a parent

had to score more than 1.5 SD below the control

means in two of the four motor areas.

None of the 20 control parents or ten siblings of

control children met the diagnostic criteria for read-ing disability, but one control father met the criteria

78

76

fl

w

74

72

70-68____

READING

RECOGNITION

Fig 2. Means and standard errors of three groups of

parents on motor and reading tests. Only variables for

which univariate analyses revealed significant group

dif-ferences are displayed. An asterisk indicates that two

groups marked by brackets differ from each other at P <

.05 by Newman-Keuls post hoc test. Abbreviations:

RDMP, reading disability and motor problems; RDO,

reading disability only; NC, control.

for motor problems. Two families with only reading

disability had more than one child aged 7 to 14

years, and these additional children were tested for

inclusion in the pedigrees, although their data were

not included in the statistical analyses. The

evi-dence of the pedigrees supports the findings of the

statistical analyses. Two distinct familial patterns

emerged, with reading disabilities often occurring

alone among the relatives of the probands with

reading disability only but rarely occurring without

motor problems among the relatives of the

pro-bands from the RDMP group. It is likely that these

familial patterns were genetic, because the two

reading disability groups did not differ from the

control groups on socioeconomic status or on

re-ports of pregnancy and birth complications. Also,

parents of the probands from the reading disability

(5)

pro-____

j:

#{149}RDMP

0

MP

Q

RDO

0

Normal j Not Tested

Fig

3.

A, Pedigrees of families with a proband with

reading disability and motor problem (RDMP). B, Pedi-grees of families with a proband with only reading

disa-bility (RDO). Asterisk indicates an RDO family which is

probably more appropriately characterized as RDMP.

Child referred to study happened to be RDO, however,

and so their data were analyzed with IWO group. Arrows

point to probands in each family.

bands on their responses to the question “How

often do (did) you read to your children?”,

suggest-ing that there was no difference in environmental

variables which might facilitate reading.

To determine whether the reading and motor

problems were associated in families with reading

disabilities and motor problems, a

x2

analysis was

performed on the sibling data. There was a

signifi-cant association between the two

(x2

= 6.24, P =

0.035), suggesting that reading and motor problems

in this group were not of separate origins but were

related, possibly sharing one common genetic cause.

DISCUSSION

It had been predicted that the relatives of the

children with reading disability without motor

problems would be more likely to have the same

difficulty as well and that the relatives of the

chil-dren with reading disability and motor problems

would be more likely to be reading disabled and

have motor problems. This prediction was

con-firmed by the results of the study: the siblings and

parents of the children in group RDMP performed

worse than the relatives of the children in the other

two groups on the tests of motor ability, and the

siblings and parents of children in both

reading-disabled groups were impaired relative to the

rela-tives of the control children on reading. The

differ-ences between the groups on reading and spelling

ability, however, were not as striking as the

differ-ences on motor ability. Perhaps this was because

our society places more emphasis on remediating

reading difficulties than on correcting coordination

and balance. Thus, a strong environmental effect

could have diminished the expected familial

pat-terns on reading and have little effect on the

pat-terns of motor ability. In general, the siblings and

parents of the probands who were reading disabled

only were more likely to be reading disabled only

as well, and the siblings and parents ofthe probands

with motor problems and reading disability were

more likely to be reading disabled and have motor

problems.

The data also supported the idea that reading

disability with motor problems is a separate

disor-der from reading disability only: the siblings of the

probands with reading disability and motor

prob-lems had deficits on bilateral coordination and

up-per limb coordination relative to the control

sib-lings, but the relatives of the probands with only

reading disability did not. Similarly, the parents of

the probands with reading disability and motor

problems had deficits on balance, bilateral

coordi-nation, upper limb coordination, and motor

accu-racy relative to the control group, whereas the

parents ofthe probands with reading disability only

did not. Thus, among the siblings and parents,

motor problems were primarily confined to the

relatives of the probands with motor problems.

Examination of the family pedigrees (Fig 3)

con-firms this idea: there were certain families with a

history of both reading and motor problems and

others with a history of only reading problems.

One finding, however, did contradict the

hypoth-esis that motor problems were unique to the

rela-tives of probands with both reading and motor

problems. The siblings ofthe probands with reading

disability only, although scoring significantly better

than the siblings of the RDMP probands with both

reading and motor problems, scored significantly

worse than the control siblings on fine motor

co-ordination. This could mean that, although balance

and gross motor coordination problems are specific

(6)

motor problems, deficits in fine motor coordination

can occur in the children of families with a history

of reading disability only. It should be noted,

how-ever, that the tests of fine motor coordination often

resembled academic tests, requiring children to

draw or trace using a pen or pencil. The tests of

gross motor coordination and balance involved

tasks such as walking on a balance beam or

throw-ing a ball, which are less like academic tests than

tasks involving drawing and tracing. Perhaps, then,

the children with only reading disability, who had

already experienced a considerable amount of

fail-ure on traditional academic tasks, became more

nervous and unsure on the tests of fine motor

coordination than on the other motor tests, and as

a result performed more poorly.

Efforts were made in this study to control for a

number of the nongenetic factors that can affect

familial patterns of reading ability. For instance, it

was found that the groups of reading-disabled

chil-dren did not differ from the group of control

chil-dren on socioeconomic status or reports of

preg-nancy and birth complications. Also, the parents of

the reading-disabled probands did not differ from

the parents of the control probands on their

re-sponses to the question “How often do (did) you

read to your children?” Indeed, the parents of the

reading-disabled children placed great emphasis on

the importance of being able to read well, and many

were paying considerable tuition to have their

chil-dren educated in special schools for the learning

disabled. Thus, it appears relatively unlikely that

the observed familial patterns of reading disability

were due to nongenetic factors such as family

so-cioeconomic status or negative attitudes toward

reading shared in certain families.

The sample may have been biased in that biologic

families willing to volunteer a considerable amount of time for research were required. Many families

who were called about the study refused to

partici-pate, partly because of the time commitment

in-volved and partly because many parents were

em-barrassed about their inability to read and did not

want to be tested. Embarrassment about a reading

disability was particularly common among the

fa-thers: 44 of 74 families who were called regarding

the study declined to participate, and of these 44

families who declined, 15 did so because the father

was unwilling to be tested. Eventually, some

fami-lies were tested without the father to keep the

sample from becoming too biased. Even so, the final

sample of families with reading-disabled members

was middle to upper class, which is probably not

typical of all families with reading-disabled

chil-dren. In this case, the bias of the sample probably

served to weaken the results. One would expect that

inclusion of the more severely impaired parents

would result in even stronger evidence of familial

patterns.

The size of the sample in this study (n = 10 for

each proband group) was small, but power analyses

performed before the study began indicated that

reducing the heterogeneity among the

reading-dis-abled children by considering subgroups yielded

sufficient power to detect differences among the

groups. Indeed, because conservative a levels were

used to ensure that we did not make any type I errors (reporting the existence of an effect which was due to chance alone), the significant differences

among the groups that emerged are especially

corn-pelling.

The physiologic basis of the coordination and

balance problems in children with reading disability

and motor problems is not known, although the

type of dysfunction seen suggests cerebellovestibu-lar involvement. It is also not known whether the

children’s motor problems are related to their

read-ing problems, but it seems feasible that their

cere-bellovestibular dysfunction could result in an

ina-bility to precisely control eye-hand movements in

the gathering of sensory information and then to

accurately relay this information from the retina to

the cortex. This could, in turn, result in reading

difficulties.

Several studies have reported abnormal and

er-ratic eye movements in reading-disabled

chil-dren.#{176} Furthermore, in one of these studies,#{176}the

eye movements of children with gross and fine

motor deficits similar to those of the children with

motor problems in this study, were examined,

in-cluding three children who were actually part of the

sample of children with both motor and reading

problems in the present study. These researchers

found that their reading-disabled subjects with

mo-tor problems had significantly worse fixational

con-trol when viewing a simple stationary stimulus than

did a group of normal control children.

Further-more, the degree of fixational instability exhibited

by these children has been shown to substantially

degrade visual resolution,3’ and so it might be

pre-dicted that their oculomotor problems would

di-rectly affect their visual processing ability when reading. It has also been reported that patients with

a neurologic disorder and acquired fixational

con-trol problems have great difficulty reading because

the words and letters appear to “jump about.”32

Thus, oculomotor control could be a mechanism

whereby the motor problems of the

reading-dis-abled children in the present study could affect

reading ability. That these children may have had

such visual/motor problems was indicated by the

(7)

disa-bility only, like many reading-disabled children,

had much higher verbal than performance IQ

scores, the children with motor problems had

sim-ilar verbal and performance scores. One could argue

that the children with reading disability only had a

primary verbal/language problem, which affected

their verbal but not performance IQ scores.

Con-versely, the children with both reading and motor

problems may have had a primary visual/motor

problem affecting their performance score and a

verbal/language problem secondary to their visual

deficits. Specifically, the visual/motor problem

might have resulted in an inability to read and

therefore to develop the vocabulary that affects

verbal IQ scores. There is some evidence to suggest

that the children with reading disability and motor

problems in this study had motor deficits which

may have included oculomotor problems resulting

in a visually based reading deficit. More research is

needed, however, regarding this controversial issue

of oculomotor dysfunction in children with reading

disabilities.

SUMMARY

In conclusion, this study has shown that two

types of reading disability do show strong familial

patterns, with some families having a history of one

type or the other. Furthermore, in the families with

a history of reading disability and motor problems,

it appears that the reading and motor problems are

related, perhaps as part of one genetic disorder

characterized by underlying cerebellovestibular

dysfunction. More research is now needed to

repli-cate these findings and also to use segregation

analysis to investigate the possible mode of

inher-itance.

ACKNOWLEDGMENT

This research was funded by the Alberta Heritage

Foundation for Medical Research.

REFERENCES

1. Pennington BF, Smith SD: Genetic influences on learning disabilities and speech and language disorders. Child Dev 1983;54:369-387

2. Hallgren M: Specific dyslexia (“congenital word blindness”):

A clinical and genetic study. Acta Psychiatr Neural Scand (suppl) 1950, p 65

3. Lewitter F!, DeFries JC, Elston RC: Genetic models of reading disability. Behau Genet 1980;10:9-30

4. Regehr 5: The genetic aspects of developmental dyslexia. Can J Behav Sci 1987;19:240-253

5. Smith SD, Kimberling WJ, Pennington BF, et a!: Specific reading disability: Identification of an inherited form through linkage analysis. Science 1983;219:1345-1347

6. Childs B, Finucci JM: The genetics of learning disabilities. Ciba Found Symp 1979;66:359-376

7. Omenn GS, Weber BA: Dyslexia: Search for phenotypic and genetic heterogeneity. Am J Med Genet 1978;1:333-342

8. Rourke BP: Overview of learning disability subtypes, in Rourke BP (ed): Neuropsychology of Learning Disabilities: Essentials of Subtype Analysis. New York, Guilford Press,

1985, pp 3-14

9. Doehring, DG: Reading disability subtypes: Interaction of reading and nonreading deficits, in Rourke BP (ed): Neu-rOpSyChO1y of Learning Disabilities: Essentials of Subtype Analysis New York, Guilford Press, 1985, pp 133-146

10. Ayres AJ: Deficits in sensory integration in educationally handicapped children. J Learning Disabil 1969;2:160-168

11. De Quiros JB: Diagnosis ofvestibular disorders in the

learn-ing disabled. J Learning Disabil 1976;9:50-58

12. Ottenbacher K: Excessive postrotary nystagmus duration in learning disabled children. Am J Occup Ther 1980;34:40-44

13. Steinberg MA, Rendle-Short J: Neuro-sensori-motor dys-function in young children with educational and behavioural problems and controls. Except Child 1980;27:107-114

14. Younes RP, Rosner B, Webb G: Neuroimmaturity of learn-ing-disabled children: A controlled study. Dev Med Child

Neurol 1983;25:574-579

15. Levinson HN: A Solution to the Riddle Dyslexia. New York, Springer-Verlag, 1980

16. Blishen BR, McRoberts HA: A revised socioeconomic index for occupations in Canada. Can Rev Social Anthropol 1976;13:71-79

17. Wechsler D: Manual for the Wechsler Intelligence Scale for Children-Revised. New York, Psychological Corporation,

1974

18. Bruininks RH: Bruininks-Oseretsky Test of Motor

Profi-ciency. Circle Pines, MN, American Guidance Services, 1978

19. Ayres AJ: Southern California Sensory Integration Test

Bat-teiy. Los Angeles, Western Psychological Services, 1972 20. Denckla MB: Development of speed in repetitive and

suc-cessive finger-movements in normal children. Dev Med

Child Neurol 1973;15:635-645

21. Denckla MB: Development of motor co-ordination in

nor-mel children. Dev Med Child Neurol 1974;16:729-741

22. Dunn L, Markwardt F: Peabody Individual Achievement Test. Circle Pines, MN, American Guidance Services, 1970

23. Finucci JM, Whitehouse CC, Isaacs SD, et al: Derivation and validation ofa quantitative definition ofspecific reading disability for adults. Dev Med Child Neurol 1984;26:143-153

24. Levine MD: The Anaer System Parent Questionnaire, Form

2P. Cambridge, MA, Educators Publishing Service mc, 1980

25. Goyette CH, Conners CK, Ulrich RF: Normative data on revised Conners parent and teacher rating scales. JAbnorm

Child Psychol 1978;6:221-236

26. Finucci JM, Isaacs SD, Whitehouse CC, et al: Empirical validation of reading and spelling quotients. Dev Med Child Neurol 1982;24:733-744

27. Pennrngton BF: Issues in the diagnosis and phenotypic analysis of dyslexia: Implications for family studies, in Smith SD (ed): The GenetiCs of Learning Disabilities: Gene-typic Studies. San Diego, College Hill Press, 1986, pp 69-96

28. Zangwill OL, Blakemore C: Dyslexia: Reversal of

eye-move-ments during reading. Neuropsychologia 1972;10:371-373

29. Pavlidis GTh: Do eye movements hold the key to dyslexia? Neuropsychologia 1981;19:57-64

30. Raymond J, Ogden N, Fagan J, et al: Fixational instability and saccadic eye movements ofdyslexic children with subtle cerebellar dysfunction. Am J Optom Physiol Opt 1988;65: 174-181

31. Murphy BJ: Pattern threshold for moving and stationary gratings during smooth eye movement. Vision Res 1978; 18:521-530

32. Natheson M, Bergman PS, Bender MB: Reading disturb-ances in the presence of oculomotor disorders. Eur Neurol

(8)

1988;82;204

Pediatrics

Sonya M. Regehr and Bonnie J. Kaplan