Printer Friendly


Abstract. Dyslexia is a common developmental disorder of unknown etiology. Behavioral and biological studies of dyslexia are complicated by its phenotypic heterogeneity and the lack of uniformly applied diagnostic criteria. In the past 20 years, increasingly powerful genetic technologies and statistical methodologies have been applied to identify genomic locations for genes involved in this complex heterogeneous disorder. This article reviews studies addressing the genetic contributions to dyslexia.


Dyslexia, or specific reading disability, is a common and complex disorder manifested by unexpected difficulty in learning to read not attributable to general cognitive delay, psychiatric or neurologic disorder, sensory impairment or inadequate instruction. It is characterized by difficulties in learning to recognize letters, subsequent difficulties in pronouncing words out of sentence context, and often persistent difficulties in fluent oral reading and spelling (Berninger, 2000). Diagnosis is usually made during the elementary school years, but is complicated by variations in educational experience, changes in phenotypic expression across development (Berninger, Abbott, Thomson, & Raskind, 2001), and the lack of a standard protocol for which assessment measures to use.

Three related language processes useful in making the diagnoses and predictive of both normal and disabled reading include (a) phonological processes -- the ability to code spoken words into short-term memory, manipulate component sounds, and reproduce words without the aid of semantic cues; (b) rapid automatized naming (Denckla & Rudel, 1976; Wolf, Bally, & Morris, 1986) or switching (Wolf, 1986) of familiar visual symbols; and (c) orthographic processes -- the ability to code written words into short-term memory and to represent them in long-term memory (reviewed in Berninger et al., 2001).

Dyslexia is believed to be a language-based disorder, typically associated with a phonological core deficit primarily manifested by difficulty in phoneme awareness and phonological decoding (e.g., Liberman, Shankweiler, Fischer, & Carter, 1974; Pennington, Van Orden, Smith, Green, & Haith, 1990; Vellutino, 1979; Wagner & Torgesen, 1987), but it may also involve deficits in physiological mechanisms of the visual system (e.g., Eden et al., 1996). There is general agreement that dyslexia is not a simple matter of letter reversals. Although with appropriate educational intervention, most affected individuals eventually achieve some proficiency in reading or writing skills, perceived or documentable deficits often persist into adulthood (Bruck, 1990, 1992, 1993; Felton Naylor, & Wood, 1990; Maughan & Hagell, 1996; Pennington et al., 1990; Shaywitz et al., 1999) and these disabilities may have long-term educational, economic, and social repercussions.

When there is a genetic component to the etiology of a disorder, it is possible to find contributing genes and delineate the biochemical and physiologic pathways involved. Such information may be of practical benefit for therapy as well as for diagnosis. Reading and spelling disabilities are not single distinct entities but comprise a spectrum of disorders. Studies of the genes involved and their interactions with each other hold promise for determining whether there are distinct genetic subtypes on a molecular level or whether the clinically heterogeneous phenotypes are different manifestations of the same gene defects. Not all children with reading disabilities respond to any given intervention. Knowledge of the biologic distinctions among different learning disabilities subtypes might provide a means to triage individuals to receive specific types of instruction.

The wide variation in age during which children acquire basic reading and writing skills contributes to the delay that often occurs in making a diagnosis of dyslexia. Educators may be reluctant to refer a student for evaluation for fear of labeling a child dyslexic when he/she is in the lower portion of the normal range for these developmental milestones. Therefore, a dyslexic child may struggle for several years before receiving optimal intervention. It would be advantageous to be able to identify children who are at higher risk for dyslexia before they have experienced repeated academic failure. Alterations in genes for monogenic Mendelian disorders can serve as "markers" for the disorders even before they are clinically apparent and, if preventive therapies exist, developing "screening tests" may be appropriate.

For complex disorders in which multiple genes of small effect contribute to the phenotypes, identifying the genes is the relatively "easy" phase. Understanding how these genes work in conjunction with other genes and environmental factors will be much more difficult. It is likely that no one gene is sufficient to cause the disorder with high frequency. The genotypes of some set of genes may merely alter the probability that a person will be affected. Whether and to what degree at-risk individuals develop reading or spelling problems may involve stochastic processes (chance events) as well as environmental factors such as quality of instruction.

For these reasons, genetic information will more likely be a tool for assigning relative risk for reading or spelling disability than a true diagnostic test. Despite the complexity of potentially multiple interacting genes and environmental influences, inherited risk factors have recently been identified for other Complex disorders, including late-onset Alzheimer's disease (apo E4; Corder et al., 1993) and type II diabetes (calpain-10; Horikawa et al., 2000) and may also be found for dyslexia.


A variety of observations have led to the consensus that reading and spelling disabilities have genetic bases. A complete description of the molecular biology and statistical methods is beyond the scope of this review; a more detailed explanation of these concepts is contained in an excellent tutorial by Olson, Witte, and Elston (1999).

Familial Aggregation

When a phenotype has a genetic basis, there is greater concordance or correlation among relatives than would be predicted by the overall population frequency. Familial aggregation of reading and spelling disabilities has been observed repeatedly since the first description of "word blindness" in the late 1890s (e.g., Gilger, Borecki, DeFries, & Pennington, 1994; Hinshelwood, 1907; Raskind, Hsu, Berninger, Thomson, & Wijsman, 2001; Schulte-Korne, Deimel, Muller, Gutenbrunner, & Remschmidt, 1996; Tallal, Ross, & Curtiss, 1989; Vogler, DeFries, & Decker, 1985; Wolff & Melngailis, 1994). The risk of reading impairment in first-degree relatives of a person with reading disability exceeds that in the general population (e.g., Gilger, Pennington, & DeFries, 1991; Hallgren, 1950; Pennington et al., 1991), and the risk and severity of the disorder increases as the number of parents affected increases (Gilger, Hanebuth, Smith, & Pennington, 1996; Wolff & Melngailis, 1994). However, the observation of familial aggregation is not sufficient to prove a genetic basis, as many environmentally mediated phenotypes also cluster in families. Examples include diseases such as lead poisoning or asbestosis and behaviors such as speaking French or religious observance.

Twin Studies

The magnitudes of the genetic and environmental contributions to the variance in a phenotype can be estimated by comparing the concordance rates in monozygotic (MZ) twin pairs and same-sex dizygotic (DZ) twin pairs. MZ twins should be more alike than DZ twins for traits that have a genetic component. In one extreme, for a trait that is entirely genetically determined, MZ twins should be entirely concordant because they share all their genes, whereas DZ twins would be less alike because they share, on average, only 50% of their genes. For the same reason, the DZ co-twins of probands selected for a phenotype in one tail of a distribution should regress halfway to the population mean. In contrast, for a phenotype that is caused entirely by environmental factors, MZ and DZ twin pairs should be equally concordant because, in theory, co-twins of both types of pairs share environmental exposures. Nonshared environmental factors must also be considered, and the relative contributions of all three influences (genetic, shared environmental, and nonshared environmental) on a phenotype can be estimated from the concordance patterns observed. In fact, the observed concordance rates for reading and spelling disabilities in twins suggest that both genetic and environmental factors are important. Although there is a higher concordance for deficits in reading and spelling in MZ twin pairs than in DZ twin pairs, even in MZ twins the concordance is not 100% (Olson, Forsberg, & Wise, 1994; Pennington et al., 1991; Reynolds et al., 1996).

MZ twins are also more similar than DZ twins on continuous component measures of reading and spelling (DeFries, Fulker, & La Buda, 1987; Reynolds et al., 1996; Stevenson, Graham, Fredman, & McLoughlin, 1987) as well as combined factor scores based on a combination of continuous measures (DeFries, Olson, Pennington, & Smith, 1991a). In addition, co-twins of DZ probands in the low end of the distribution for reading and spelling measures show greater regression to the mean than do co-twins of MZ probands (DeFries et al., 1987; DeFries, Stevenson, Gillis, & Wadsworth, 1991b).

Twin studies have also supported the hypothesis that the processes involved in reading and spelling are substantially heritable. Although heritability of phonological coding was significant and heritability of orthographic coding, measured by the ability to distinguish between written real and pseudoword homonyms, was reported to be nonsignificant in one study (Olson, Wise, Connors, Rack, & Fulker, 1989), the same authors subsequently found that heritability of orthographic coding was similar in magnitude to that of phonological coding (Olson et al., 1994). The heritability of components of reading and spelling disability -- word recognition, phonological coding, phonological awareness and orthographic coding -- is approximately 47-60% whereas shared environmental factors account for 29-48% of the phenotypic variance. Nonshared environmental factors appear to make a smaller contribution.

Patterns of Familial Transmission

The frequent finding of reading and spelling disabilities in multiple relatives of a child with reading disabilities has been observed empirically by many researchers, but the pattern in the pedigrees often does not fit a simple Mendelian mode of inheritance (e.g., Finucci, Guthrie, Childs, Abbey, & Childs, 1976; Omenn & Weber, 1978). The mode of inheritance is of great practical importance. If dyslexia were simply the lower tail of the normal distribution of reading skill (Shaywitz et al., 1992a) in which multiple genes contribute equally small effects (i.e., a polygenic or multifactorial trait), even with the current rapidly advancing technology it may be difficult, if not impossible, to identify the genes, because no one gene accounts for enough of the variation to be distinguished from the rest. On the other hand, if there are genes that contribute even modest effects above the polygenic background (called a "major" gene effect), they may be identifiable.

A quantitative trait locus (QTL) is one that contributes to a phenotype that may be measured on some continuous scale. The phenotype may be influenced by more than one gene and the phenotype associated with a particular combination of alleles (genotype) may be variable, in part because of environmental contributions to the variation. A mixed model is one that has both major gene and polygenic or multifactorial components. Single-gene inheritance (and the individual genes that contribute to an oligogenic or multifactorial phenotype) can be autosomal recessive, autosomal dominant, additive (codominant), sex-chromosome linked, or mitochondrial.

One way to clarify the genetic contribution to a phenotype is to develop a mathematical model of the mode of inheritance that explains the observed patterns in families. For example, how often does a child have dyslexia when one parent has dyslexia and the other parent does not? How do age, gender, test scores on other language measures, or other measurable quantities affect these predictions? What is the likelihood that a person will manifest dyslexia if he or she carries the particular contributing gene and how much influence does a single allele of the gene have on the type and severity of the problem? This process, called segregation analysis, fits models of transmission to empiric data and provides parameters that can be used in linkage analyses (reviewed in Pennington et al., 1991). These parameters include estimates of the disease gene frequency, the relative effects of genetic and nongenetic influences, and the existence of etiologic heterogeneity. In complex segregation analysis, a single Mendelian gene is modeled, other genetic factors are combined into one polygenic factor, and a single model that best fits the data can be chosen. There are now analysis approaches that employ a Bayesian Monte-Carlo Markov-Chain stochastic method and allow an estimate to be made of the number of genes that are involved (briefly described in Wijsman et al., 2000).

There are potential pitfalls to these modeling studies, however. Even when a disorder has no major gene component, rare families with what appears to be single-gene transmission may be found. If studies are done on a biased collection of such families, the incorrect conclusion that the disorder is a single-gene defect can be reached. Genetic heterogeneity -- when different genetic mechanisms can result in the same phenotype -- must also be considered. In addition, unless accounted for, age, gender or other influences on the phenotype may confound the analyses, especially when a categorical trait is modeled.

Three different approaches have provided evidence for family transmission patterns of dyslexia-associated phenotypes consistent with Mendelian modes of inheritance. Segregation analyses of reading disability defined categorically produced evidence for a sex-influenced, dominant or additive major gene effect in three independently sampled populations of dyslexic families (Pennington et al., 1991). The disorder-allele frequency at the major locus was estimated at between 3% and 5%. However, a multifactorial-polygenic mechanism seemed to be operating in a fourth population. Analysis of a quantitative discriminant score based on multiple measures found evidence for involvement of one or more dominant genes with a high-frequency "disease" allele of low penetrance in a population of nuclear families ascertained through nonreading disabled probands (Gilger et al., 1994). Evidence for a major-gene mode of inheritance for quantitative scores on measures of two reading-related processes -- phonological nonword memory and digit span -- was obtained by a combination of oligogenic-trait and complex segregation analyses (Wijsman et al., 2000). Taken together, these lines of research suggest that reading disability, as well as the phonological and orthographic skills contributing to reading, are heritable, but only 40-70% of the individual differences result from genetic factors.


Positional Cloning

Studies that exploit the naturally occurring DNA variations between individuals can lead to identification of genes that are involved in a disorder even when the function of the gene is not known. This process, called "reverse genetics" or "positional cloning," can involve a variety of approaches such as linkage analyses, association studies, and linkage disequilibrium studies. The gene is found primarily on the basis of its location in the genome. The objective of a genetic linkage analysis is to identify regions of the genome that are nonrandomly associated with a phenotype. With the exception of genes on the sex chromosomes, all people carry two copies of each gene -- one copy inherited from each of the two parents. For some of these pairs of genes, the two copies (the alleles) differ within or between individuals. A straightforward example is the ABO blood type system that has three common alleles. The A and the B alleles produce a protein that can be detected on the cell surface, whereas the O allele does not. Therefore, people who have an AA or AO genotype have type A blood, people with a BB or BO genotype have type B blood and people with an A and a B allele have type AB blood. Each of the children from two parents with blood types AB and O, respectively, will have the A or B blood type, because each child will inherit the O type from one parent and either the A or the B from the other parent. About half of the children will fall in each category, if we look at a large enough number of children from matings of this type.

Genes comprise only a minority of the DNA in a cell and are not very polymorphic. That is, most people have the same DNA sequence for the portions of the gene that code for the amino acids. However, the noncoding spacer DNA can have many different compositions in the population. These differences usually have no effect on a person but they make it possible to track the transmission of short pieces of DNA, called markers, from a person to his/her offspring. The alleles of the most widely used type of marker -- short tandem repeat polymorphisms (STRP) -- vary by number of repeats of a sequence of nucleotides (i.e., length). During the process of meiosis, which results in formation of the ova and sperm, the two chromosomes in each pair (homologues), one maternally derived and the other paternally derived, physically mix their genetic material (recombine) in one or more places along their length. The closer two segments of DNA are to each other, the more likely they are to remain together during this process. For the linkage analysis, in a method analogous to typing blood, many polymorphic markers scattered throughout the genome are genotyped and then examined in each family to determine whether a marker and the specific phenotype are co-inherited more often than would be predicted by chance under the proposed model of transmission of the phenotype. This occurrence suggests that the DNA marker is near ("linked to") a gene that is involved in the phenotype. The marker is not the cause of the phenotype, and different alleles may be associated with the phenotype in different families. Unless a single family is extremely large, it is necessary to study many families to detect a significantly skewed pattern of co-inheritance. Because the genomic locations of the markers are known, finding linkage to a marker reveals the chromosomal region containing the gene. The remarkable progress of the Human Genome Project is facilitating the identification of specific genes within a chromosomal region because the location and some sequence information is known for most of the approximately 30,000 to 50,000 human genes.

Model-Based Versus Model-Free Analyses

For some phenotypes, the mode of transmission in families is obvious and straightforward. For example, cystic fibrosis is autosomal recessive (both alleles must be mutated for a person to be affected; both parents are carriers of one mutated and one normal allele; and each child of two carrier parents has a 25% chance of inheriting the disease); Huntington's chorea is autosomal dominant (a single mutated allele is sufficient to cause the disease and each child of an affected person has a 50% chance to inherit the disease); and hemophilia A is X-linked recessive (males, who have only one X chromosome, are affected if they have the mutation; a female who has one mutated allele and one normal allele is not affected, but her sons each have a 50% chance of being affected).

For dyslexia, however, the pattern of inheritance is much harder to discern because of the likely complexity of the genetic portion (that is, there are probably multiple genes that interact to produce the observable phenotype) and the contribution of environmental factors. Furthermore, unlike Mendelian disorders, which are usually caused by highly penetrant but rare alleles in a single gene, complex disorders are likely to be associated with common, less penetrant functional polymorphisms in multiple genes. As mentioned, each of these susceptibility alleles may increase the risk of disease, but does not necessarily cause disease.

Analyses can be performed that do not place many constraints on the model of transmission in families. These methods are often referred to as "model-free" and include sib-pair analyses (e.g., Thomson, 1986). Sib-pair methods calculate how often sibs who are concordant for a phenotype have inherited the same marker allele from a parent. As each parent has only two alleles at any marker, if the gene and the marker are not linked, the two affected sibs will inherit the same allele by chance approximately 50% of the time. If the gene and the marker are linked, sibs who are concordant for the phenotype will tend also to be concordant for the allele at the marker. The alleles are tallied in multiple sib pairs and the deviation from the random expectation is a measure of linkage between the marker and the gene involved in the phenotype being studied.

Model-based methods, in which details of the inheritance and expression patterns of a disorder are incorporated into the analyses, are inherently more powerful than sib-pair methods, but the power is dependent on the correctness of the model used. In family-based linkage analysis, the cotransmission of marker alleles and the phenotype can be tracked through multiple generations and can utilize all family members, apparently unaffected as well as affected (reviewed in Borecki & Suarez, 2001). In this method, the empiric data from each family are compared to theoretical results expected if the two loci are linked. The predicted proportion of recombinant gametes is the recombination fraction ([Theta]), with fewer recombination events expected for more tightly linked loci. The genetic distance between two loci (measured in Morgans) is correlated in a more variable way with physical distance (measured in DNA base pairs). The linkage analyses are repeated for various values of [Theta] and results are reported as the logarithm of the odds of likelihood of linkage, or lod score, for each family and summed for the total families. Given the number of genes, the a priori likelihood of linkage of any given marker to the gene involved in the phenotype is low. Therefore, regardless of the method of analysis, the convention has been to require a high lod score, at least 3 (or 1,000 to 1 odds in favor of linkage), as significant evidence in favor of linkage for a single gene defect (Morton, 1955) and higher thresholds for complex disorders (Lander & Kruglyak, 1995).

If there is a known gene whose properties suggest it is a good candidate to be involved in the phenotype, or if there is linkage of the phenotype to a short genomic region, one can look for an association between the phenotype and a specific allele or haplotype. A mutation arises in the context of a haplotype formed by the specific alleles at flanking loci on the chromosome. Over time, with succeeding generations, the process of recombination separates the mutated allele from the flanking alleles and eventually equilibrium, based on allele frequency, is reached between the alleles at all these loci. The number of generations needed to arrive at equilibrium is a function of the physical distance between the loci, but is not uniform across the genome. An association between specific alleles at two loci is called "linkage disequilibrium," and it is generally restricted to very short segments of DNA. In essence, in linkage disequilibrium mapping, the transmitted and nontransmitted parental alleles are scored in their offspring. The finding of biased co-inheritance of an allele and a phenotype supports the hypothesis that the candidate gene contributes to the phenotype or, if the gene is unknown, the extent of the co-inherited region can be used to further localize the gene. On one hand, the a priori likelihood of success for association studies is low for several reasons: (a) the large number of genes in the genome, (b) the lack of complete information about the functions of most genes necessary to ensure that the causative gene is among those selected for study, and (c) the variability across the genome in the distances over which linkage disequilibrium persists for sufficient generations to allow detection. On the other hand, one great advantage of association studies is their ability to study family groups as small as trios of single affected offspring and their parents.

Linkage Analyses for Reading- and Spelling-Related Phenotypes

As would be predicted for a disorder that is complex and phenotypically diverse, linkage analyses suggest that reading and spelling disabilities are also genetically heterogeneous. Possible localizations for genes for reading and spelling disabilities have been reported on chromosomes 1, 2, 6, and 15 (see Table 1). The studies are not directly comparable as there are substantial differences in the phenotypes evaluated, the ascertainment schemes, the diagnostic criteria for affected status, and the analysis methods.
Table 1
Comparison of Current Linkage Analyses in Reading and Spelling

Reference Samples Phenotype

Chromosome 1p22 or 2q22
Froster et al., 1993 1 two- severely
 generation delayed speech,
 family school history,
 writing disability

Chromosome 1p34-36
Rabin et al., 1993 9 three- composite --
 generation details of
 families assessment not given

Grigorenko et al., 8 multi- single-word
2001 generational reading

 rapid naming

 phonological decoding

 phonological awareness

Chromosome 2p15-16 (DYX3)
Fagerheim 1 family: 36 composite, weighted
et al., 1999 members in 3 for history and
 generations tests of phonological
 decoding and
 awareness, word
 or spelling

Chromosome 6p21.3 (DYX2)
Cardon et al., 114 sib pairs composite,
1994, 1995 from 19 including word
 families recognition,
 46 DZ comprehension,
 twin pairs spelling recognition
 same composite

Grigorenko et al., 6 multi- phonological
1997 generational awareness

Grigorenko 8 single real-
et al., 2000 multigenerational word reading,
 families vocabulary
 and spelling

Gayan et al., 101-126 DZ twins orthographic
1999 and their sibs choice of real
 same population word from
 same population nonword homonyms
 phonological decoding
 phonological awareness

Chromosome 6p21.3 (DYX2)
Fisher et al., 1999 77-119 phonological decoding
 sib pairs

 orthographic coding of
 irregular words

Chromosome 6q13-16.2
Petryshen et al., 1999 96 families phonological decoding

 228 sib pairs phonological decoding
 357 sib pairs spelling from

Chromosome 15q (DYX1)
Smith et al., 1983 8 families with oral prose reading
 history of

Grigorenko 6 families with single real-word
et al., 1997 extended reading
 history of

Schulte-Korne 7 families spelling from
et al., 1998 dictation

Morris et al., 2000 178 parent/ oral prose reading
 offspring trios

Fagerheim et al., 2000 1 family: same phenotypes as for
 36 members DYX3
 in 3

 Categorical (C) vs.
Reference Quantitative (Q) Method(a)/Score(b)

Chromosome 1p22 or 2q22
Froster et al., 1993 C LOD/1.5

Chromosome 1p34-36
Rabin et al., 1993 C LOD/1.95

Grigorenko et al., C APM

 NPL p < 0.05


 NPL p < 0.05
 C NPL p < 0.05

Chromosome 2p15-16 (DYX3)
Fagerheim LOD/3.525
et al., 1999 MP-NPL/0.O16-0.001

Chromosome 6p21.3 (DYX2)
Cardon et al., Q [MPNP.sub.p])/0.0027
1994, 1995 to 0.0417

 Q NPL/0.0094

Grigorenko et al., C APM/<0.005

Grigorenko C APM/[approximately
et al., 2000 equal] 0.01

Gayan et al., Q [MPNP.sub.1]/3.10
 Q [MPNP.sub.1]/2.42
 Q [MPNP.sub.1]/1.46

Chromosome 6p21.3 (DYX2)
Fisher et al., 1999 Q NPL/<0.0005;


 Q NPL/<0.0005;

Chromosome 6q13-16.2
Petryshen et al., 1999 C LOD/2.6

 C NPL/.016
 Q NPL/.00076
Chromosome 15q (DYX1)
Smith et al., 1983 C LOD/3.241

Grigorenko C LOD/3.15
et al., 1997

Schulte-Korne C LOD/1.26-1.38
et al., 1998

Morris et al., 2000 C eTDT/p = 0.006

Fagerheim et al., 2000 C LOD/2.25

 Map location (KcM(c))
Reference Flanking markers

Chromosome 1p22 or 2q22
Froster et al., 1993 --

Chromosome 1p34-36
Rabin et al., 1993 27.6-41.2

Grigorenko et al., 45.33-48.53
2001 D15199 - D15478 and
 [approximately equal] 60 -
 [approximately equal] 71
 D15253 - D1S507

 14.59-[approximately equal] 71
 D15253 - PPT

 D15199 - D15478

 D15199 (45.33)
 14.59-[approximately equal] 71
 D15253 - PPT

Chromosome 2p15-16 (DYX3)
et al., 1999 76.34-80
 D252352 - D251337

Chromosome 6p21.3 (DYX2)
Cardon et al., 44.4-45.2
1994, 1995 D65105 - TNFB


Grigorenko et al., 34.23-44.4
1997 D65109 - D65276
 D65109 - D65306

Grigorenko 40.9-41.1
et al., 2000 D65464 - D65306

Gayan et al., 44.40
1999 D65276 - D65105

Chromosome 6p21.3 (DYX2)
Fisher et al., 1999 40.13-44.4
 D651660 - D65276
 D65276 - D65105
 D6S1660 - D65291
 D65276 - D65105

Chromosome 6q13-16.2
Petryshen et al., 1999 82.59-90.43
 D65254 - D65251
Chromosome 15q (DYX1)
Smith et al., 1983 peri-

Grigorenko 40.16-47.8
et al., 1997 D155214 - D155209

Schulte-Korne 40.16-45.56
et al., 1998 D155214 - D155126

Morris et al., 2000 39.72-40.25
 D155146 - D15S994

Fagerheim et al., 2000 6.92

(a) LOD = parametric family-based loci score method; a lod score
of slightly greater than 3 is accepted as evidence in favor of linkage
at p [is less than or equal to] .05.

[MPNP.sub.p] = multipoint nonparametric regression method for
interval mapping of an extreme quantitative trait phenotype, p value.

[MPNP.sub.i] = multipoint nonparametric regression method, lod score.

NPL = nonparametric scoring method for a categorical phenotype.

MP-NPL = multipoint nonparametric lod score for a categorical

APM = affected pedigree member; a nonparametric method that includes
all affected relatives, not only sibs.

[APM.sub.M] = multipoint analysis.

eTDT = transmission disequilibrium test.

(b) p value or maximum lod score. The empirical p values for
the eTDT are based on the number of times the observed heplotype
frequency exceeded simulated expected frequencies, expressed as
an empirical p value for a [chi square].

(c) Marshfield sex-averaged map location as measured in Kosambi
centimorgans from the top of the short arm of the chromosome
(Marshfield website). A Kosambi centimorgan is a unit of
recombination frequency, adjusted for region-specific
interference. The likelihood of a recombination event
in a region is not equally distributed along the
chromosomes. Interference is the phenomenon whereby
recombination is suppressed below what would be predicted
for the physical length of the segment.

(d) Reported in abstract and meeting poster; details about
flanking markers not provided.

The first report of linkage of dyslexia to a chromosomal region, now assigned the locus name DYX1, was published in 1983 (Smith, Kimberling, Pennington, & Lubs, 1983). The STRP markers had not yet been described and linkage studies were limited to the few polymorphic gene and cytogenetic markers available. The investigators noticed cosegregation of dyslexia, defined by oral reading ability at least two years below expected grade level, and a cytogenetic polymorphism at the top of the acrocentric chromosome 15 in eight families selected on the basis of apparent autosomal dominant transmission of the disorder. A dichotomous scheme was employed (disabled or not disabled) and history was preferentially used for categorization of adults. One family contributed the majority of the significance. When additional families and family members were added and DNA markers replaced the cytogenetic markers for the linkage analysis, the lod score was reduced below the level accepted as evidence for linkage (Smith et al., 1986; Smith, Pennington, Kimberling, & Ing, 1990). However, there was evidence that reading disability might be genetically heterogeneous. That is, only a subset of the families might have the disorder on the basis of a gene on chromosome 15.

Although linkage associations between dyslexia and a locus on chromosome 15 have not been confirmed in all studies (Bisgaard, Eiberg, Moller, Niebuhr, & Mohr, 1987; Cardon et al., 1994; Fagerheim et al., 1999; Lubs et al., 1991; Rabin et al., 1993; Sawyer et al., 1998), suggestive evidence favoring the existence of a chromosome 15 locus involved in reading and spelling disability has been published by at least three other groups using categorical definitions for the phenotypes. Linkage to markers on the proximal long arm ([Theta]) of chromosome 15 was found for deficits in single real word reading (Grigorenko et al., 1997) and spelling (Schulte-Korne et al., 1998) and association of a 3-marker haplotype was found for deficits in oral prose reading (Morris et al., 2000). Dyslexia was observed to cosegregate with a balanced translocation involving chromosome band 15q21 in three individuals in one family (Nopola-Hemmi et al., 2000). However, a fourth translocation carrier in that family was not affected by dyslexia and in a second family, only one of four carriers of a translocation involving the same region of chromosome 15q was affected by dyslexia. It has also been suggested that a gene at the DYX1 locus might modify the expression of the dyslexia phenotype in a family whose dyslexia appears to be linked to the short arm (p) of chromosome 2 (Fagerheim et al., 2000).

Suggestive evidence in favor of linkage to chromosome 1p34-36 was obtained in a study of nine three-generation families (Rabin et al., 1993), including the one that had previously provided much of the evidence for linkage to chromosome 15 (Smith et al., 1983). Supportive evidence has recently been reported for a contribution of a locus on the short arm of chromosome 1 to the phenotypes of single-word reading, phonological decoding and rapid automatized naming in eight extended dyslexic families previously studied for linkage to chromosomes 6 and 15 (Grigorenko et al., 2001). The region of interest was very broad, spanning approximately 57 Kosambi centimorgans. No evidence for linkage to this region of chromosome 1 was found in another study (Cardon et al., 1994). Cosegregation of a balanced chromosomal translocation, t(1;2)(p22;q22) and a syndrome of speech delay and dyslexia was reported in one family (Froster, Schulte-Korne, Hebebrand & Remsschmidt, 1993), but the significance of this study is limited by the small size of the family. Bands 1p22 and 1p34-36 are genetically and physically distant from each other.

Evidence for a locus on chromosome 6p was originally suggested by Smith, Kimberling, and Pennington (1991) in a study of the multigenerational families in whom linkage of dyslexia to chromosome 15 was now in doubt. The region of chromosome 6p containing the HLA locus had been targeted for genetic studies because of previous research suggesting an increased incidence of autoimmunity in dyslexic individuals and their relatives (Geschwind & Behan, 1982; Pennington, Smith, Kimberling, Green, & Haith, 1987). However, this association has not been confirmed in other studies (Gilger et al., 1998). Cardon et al. (1994, 1995) later reported linkage to chromosome 6p for a compound continuous phenotype defined by a discriminant score in a study population of DZ twins. The discriminant score was derived from performance on measures of reading recognition, reading comprehension, and spelling. Additional evidence for the existence of a locus on chromosome 6p that may contribute to reading difficulties has been reported by several groups.

The phenotypes that showed linkage to chromosome 6p were: (a) a categorical phenotype assessed by different composite measures of phonological awareness based on the ability to segment words into phonemes and to recognize and sequence phonemes appropriately (Grigorenko et al., 1997); (b) a categorical phenotype based on single real-word reading, spelling or vocabulary (Grigorenko et al., 2001); (c) a continuous component measure of orthographic processing based on the ability to identify a real word from nonword homonyms and to choose the correct word from a pair of real word homonyms when given the word's meaning (Gayan et al., 1999); (d) a continuous component measure of orthographic processing based on the ability to read irregular real words (Fisher et al., 1999); (e) a continuous component measure of phonological awareness based on the ability to delete or transpose the order of phonemes (Gayan et al., 1999); and (f) a continuous measure of phonological decoding of nonwords (Fisher et al., 1999; Gayan et al., 1999).

On the other hand, failure to find evidence for linkage to chromosome 6p has been reported for a categorical phenotype of spelling deficiency in a set of eight German families (Schulte-Korne et al., 1998), for a categorical measure of phonological decoding (Field & Kaplan, 1998) in a set of 96 Canadian families, and in the same Canadian sample for continuous measures of phonological awareness, phonological decoding, spelling, and RAN (Petryshen, Kaplan, Liu, & Field, 2000). The latter group, has recently reported suggestive evidence of a locus on chromosome 6q for categorical and quantitative phenotypes based on phonological decoding and for a quantitative measure of spelling from dictation (Petryshen, Kaplan, & Field, 1999).

Significant evidence for linkage of dyslexia to chromosome 2q was found in a large Norwegian family (Fagerheim et al., 1999) in the first reported genome-wide scan for a dyslexia locus. This locus has been assigned the name DYX3. A qualitative affectation status was determined by weighting the history and test results for isolated real-word and nonword recognition, with and without time constraints, spelling, and phoneme manipulation. Although mildly positive linkage results were obtained to the pericentromeric region of chromosome 15, where the DYX1 locus was originally specified, linkage to DYX2 was excluded in this family (Fagerheim et al., 1999, 2000). It may be of interest that linkage of dyslexia, as categorically determined from tests of orthographic and phonological decoding and spelling, to DYX2 was also not found in another large Norwegian family (Heiervang et al., 1995). Failure to confirm a given localization is difficult to evaluate because it may be due to inadequate sample size, differences in diagnostic criteria, differences in ascertainment methods, differences in statistical methods and genetic differences between different populations.


Rationale for Identifying Subphenotypes

None of the studies cited above is independently strong enough to provide conclusive evidence for linkage of a particular phenotype to a specific locus, given the complexity of the trait. However, consistent findings by several groups make it likely that loci at chromosomes 6 and 15 are involved. The information available currently does not allow a precise estimate of the relative contributions of each locus to each of the processes comprising the "dyslexia" phenotype, nor is it known how many genes contribute to each of the processes. Perhaps there is a threshold of dysfunction for dyslexia and, in an additive model, each of several loci can contribute relative protection or burden towards this threshold. Some loci may play a unique role and some may contribute to more than one phenotype. Others may have a more restricted range of effect. More than one gene may independently lead to a similar phenotype and other genes may modify the expression of the phenotype. None of the psychometric measures used in the evaluation of dyslexia are pure measures of discrete and unique cognitive processes. The phenotypes identified by these psychometric measures are related to each other and performance on any two measures is often significantly correlated (Berninger et al., 2001; Gayan et al., 1999; Grigorenko et al., 1997; Olson, Forsberg, & Wise, 1994).

Information is lost and power is reduced when a complex disorder is treated as a unified entity (Wijsman & Amos, 1997). Therefore, we have chosen a systematic approach: (a) to identify the phenotypes most likely to be determined by genetic factors; (b) to evaluate the genetic relationship between these measures; and (c) to develop models of transmission that will enable use of more powerful model-based linkage analyses rather than model-free methods. These studies are being performed using the quantitative measurement data rather than dichotomous diagnostic categories.

Rationale for Including Measurements of Verbal IQ in Dyslexia Research

There is an ongoing controversy regarding the use of IQ in the diagnosis of dyslexia (Shaywitz et al., 1992b; Siegel, 1999; Stanovich, 1994). However, there are compelling reasons to account for Verbal IQ in genetic research on dyslexia. First, the low performance group is likely to be more heterogeneous than the remainder of the distribution because it will contain individuals who have neurological, psychiatric, and other medical disorders that interfere with learning or testing (e.g., fetal alcohol syndrome, brain trauma), as well as those for whom environmental exposures are the main causes for the learning disability. Some support for this conjecture is provided by Pennington, Gilger, Olson, and DeFries (1992). Second, there is some evidence that reading disability defined by the discrepancy criterion may be more heritable than reading disability defined by the low performance criterion (Olson, Datta, Gayan, & DeFries, 1999) and that reading disability in high-IQ groups is significantly more heritable than is reading disability in low-IQ groups, with estimates of .72 and .43, respectively (Wadsworth, Olson, Pennington, & DeFries, 2000). Third, the Verbal IQ exerted a significant positive influence on the correlation between siblings and between parents and offspring for all measurements of reading and spelling performance (Raskind et al., 2001). There is evidence for a genetic contribution to intelligence (e.g., Alarcon, Plomin, Fulker, Corley, & DeFries, 1998; Bouchard, 1998; Light, DeFries, & Olson, 1998; Petrill et al., 1997; Rietveld, van Baal, Dolan, & Boomsma, 2000), and assortative mating for cognitive abilities as well as for achievement has been repeatedly observed (e.g., Alarcon, DeFries, & Gillis, 1994; Gilger et al., 1991; Wolff & Melngailis, 1994).

Failure to adjust for IQ effects might result in an incorrect model of transmission of reading and spelling related phenotypes. It is important to note that although inclusion of IQ measurements in research on the etiologies of reading disability is justifiable, it may not be appropriate to use IQ as an exclusionary criterion in schools, as all children with a variety of reading disorders may benefit from educational intervention, regardless of cause.


It has been suggested that the genetics of complex disorders might be more tractable if the phenotypes studied could be reduced to simpler or more phenotypically precise subphenotypes. In the University of Washington Learning Disabilities Center, aggregation and segregation analyses are being carried out to identify dyslexia subphenotypes whose properties make them good candidates for further genetic study. Families were identified in which there was a child with dyslexia and/or dysgraphia in grades 1 to 6. These families were ascertained without regard for family size, existence of siblings, or additional history of reading or writing problems. One hundred and two of these children (probands) qualified for the study. Details of the procedures used for selecting probands and for evaluation of probands and their nuclear family members have been published (Berninger et al., 2001; Raskind et al., 2001). A combination of low achievement and Verbal IQ-Achievement discrepancy criteria were employed, rather than low achievement only or Verbal IQ-Achievement discrepancy only, to determine whether a prospective proband qualified for the study and for the analyses based on categorical diagnoses reported in this article.

Inclusion as a proband required a prorated Verbal IQ (proVIQ) of at least 90 ([is greater than or equal to] 25th %tile), a discrepancy of at least one standard deviation between the proVIQ and the score on one or more reading or writing measure in a screening battery, and performance on the measure below the population mean. The proVIQ is the same as the Verbal Comprehension Factor, calculated by prorating all verbal scale subtests of the Wechsler Intelligence Scale for Children -- Third Edition (WISC-III, Wechsler, 1992a) with the exception of the Digit Span and Arithmetic subtests. A total of 23 measures were administered to 409 members of 102 nuclear families who were at least 6 years old.

The familial aggregation pattern of the quantitative data from each measure in the battery was analyzed (Raskind et al., 2001). If there is a genetic basis for a phenotype and the sample set is of sufficient size, blood-related family members should show positive correlations, with magnitudes proportional to the degree of genetic relatedness. The phenotypes of non-consanguineous (genetically unrelated) parents should not be correlated. Significant correlations among relatives in a pattern consistent with a genetic basis were observed for five measures: the Phonological Decoding Efficiency (PDE) subtest of the prepublication version of the TOWRE (Torgesen, Wagner, & Rashotte, 1999), the Nonword Memory (NWM) subtest of the Comprehensive Test of Phonological Processing (CTOPP) (Wagner & Torgesen, 1999), the Digit Span subtest of the WISC-III for individuals younger than 17 or the Wechsler Adult Intelligence Scale -- Revised (WAIS-R; Wechsler, 1981) for those 17 and older, the Word Attack (WA) subtest of the Woodcock Reading Mastery Test-Revised (WRMT-R; Woodcock, 1987), and the spelling subtest of the Wide Range Achievement Test -- Third Edition (WRAT-3; Wilkinson, 1993). Weaker evidence of such an aggregation pattern was obtained for five additional measures: the spelling subtest of the Wechsler Individual Achievement Test (WIAT; Wechsler, 1992b), the accuracy, rate and comprehension subtests of the Gray Oral Reading Test -- Third Edition (GORT-3; Wiederholt & Bryant 1992), and Rapid Automatized Naming for Switching Letters and Numerals (RAS; Wolf, 1986).

To obtain information about the interdependence of the measures on shared genetic factors, pairs of related measures were examined, using each measure in the pair once as the response variable and once as the covariate. In addition to the reading- and writing-related measures NWM, Digit Span, WA, PDE, WRAT-3 spelling, RAS and GORT-3 rate, a quantitative measure of inattention (difficulty in self-regulation of mental processes) was also evaluated. The effect of the respective covariate on the aggregation pattern of the response variable suggested that there may be a genetic contribution to phonological NWM in addition to the genetic contribution it shares with Digit Span, a genetic contribution to efficiency (rate) of phonological decoding in addition to the genetic contribution it shares with accuracy of phonological decoding, a genetic contribution to phonological nonword memory in addition to the genetic contribution it shares with written spelling, a genetic contribution to written spelling in addition to the genetic contribution they share with accuracy of phonological decoding, and a genetic contribution to inattention ratings in addition to the genetic contribution it shares with either RAS or oral reading rate (Hsu, Berninger, Thomson, Wijsman, & Raskind, 2001).

In the first segregation study two converging measures of phonological short-term memory were considered -- Digit Span and phonological NWM (Wagner & Torgesen, 1999; Wechsler, 1981, 1992a). When analyzed individually, evidence for a major gene inheritance was obtained for each phenotype, with ~2.4 and ~1.9 QTLs contributing to NWM and Digit Span, respectively (Wijsman et al., 2000). When analyses were performed with one measure incorporated as a covariate for the other, the most parsimonious model suggested that genes that influence Digit Span also influence NWM, but that additional QTLs are involved in NWM. Identification of genes for these more basic phenotypes will be an important step towards understanding the biology of the more complex behavior of reading.


Effect of Gender on Expression of Dyslexia Almost all studies of reading and spelling disabilities have found a higher prevalence in males, ranging from 1.3-4:1 (e.g., Allred, 1990; Decker & DeFries, 1981; DeFries, 1989; James, 1992; Pennington et al., 1991; Schulte-Korne et al., 1996; Wadsworth, DeFries, Stevenson, Gilger, & Pennington, 1992; Wadsworth et al., 2000). It has been argued that this observation may result from biased ascertainment rather than from an actual gender influence on the phenotype (Shaywitz, Shaywitz, Fletcher, & Escobar, 1990; Vogel, 1990). Referred samples might reflect societal prejudices and/or gender differences in behavioral coping mechanisms. However, it is also possible that there is a true difference in the prevalence, severity or etiology of dyslexia between males and females. In a population-based study of oral reading performance in twins, with no correction for IQ, there was greater variability in scores for males than females (Reynolds et al., 1996). As a larger percent of males fell in the tails of the distribution, more males than females would meet the criteria for disability in oral reading using an achievement criterion. A study of 75 first-degree relatives of 20 children with specific reading disability found that a significantly greater number of affected male relatives displayed difficulty on tests of spelling, oral reading and reading nonsense words contained in text than did the female relatives (Finucci et al., 1976). The gender ratio among spelling disabled relatives of 32 spelling disabled children was 2.4:1 (Schulte-Korne et al., 1996). Brothers of 125 children with reading disabilities showed a greater deficit on average than did sisters on a continuous composite measure of reading recognition, reading comprehension and spelling (Decker & DeFries, 1980).

It has been suggested that females may have a higher threshold for expression of reading and spelling disabilities than males. One prediction of a gender-influenced threshold model is that relatives of dyslexic females would be at greater risk than relatives of dyslexic males, because dyslexic females would have higher genetic liability, that is, more dysfunctional genes inherited from their parents. This hypothesis has been supported by empiric data in several studies (DeFries, 1989; DeFries & Decker, 1982; Gilger et al., 1991), but not in others (Schulte-Korne et al., 1996; Wadsworth, Knopik, & DeFries, 2000). In one of these negative studies, it was observed that 42.9% of the first-degree relatives of the 7 female probands were spelling disabled, compared with 33.3% of the relatives of 12 male probands, but the difference was not significant (Schulte-Korne et al., 1996). However, there is no evidence to suggest that the differential prevalence of dyslexia in males and females reflects a gender difference in the etiology of the disability (Alarcon, DeFries, & Fulker, 1995; Wadsworth et al., 2000).

Although a debate remains regarding the differential risk for reading and spelling disabilities in males and females, it is generally agreed that there is an age effect on the expression of these disabilities. Many adults with a history of dyslexia in childhood continue to show residual deficits on careful testing, but there is a wide variation in the severity of the disability. Even when they report subjective difficulties with reading facility, some individuals no longer evidence signs of dyslexia by achievement or discrepancy criteria (e.g., Gilger et al., 1996). Such individuals are often described as "compensated" adults. In a study of twins, the estimated heritability of real-word reading was significantly higher in the subset of twins who were younger than 11.5 years old than it was in the subset who were older than 11.5 years. The converse pattern was found for spelling recognition and spelling generation, with greater heritability in the older twins, although only the results for spelling recognition reached statistical significance (DeFries, Alarcon, & Olson, 1997). Therefore, the phenomenon of compensation must be considered in family-based genetic analyses, which include subjects of many ages, especially studies that rely on quantitative test data rather than on historical report.

We addressed the issues of age- and gender-influenced penetrance and expressivity in our sample of 102 nuclear families. To evaluate the frequency with which siblings and parents of the probands evidenced difficulties with component reading skills, several categorical analyses were performed for nine measures that displayed strong or weak aggregation patterns consistent with a genetic etiology (see last section). Digit Span was not included in these analyses because it is confounded by inclusion in the proVIQ for adults. The criteria described above for qualifying a proband for the study were used to determine affected status. Assuming a population prevalence of 5-10%, we observed a markedly increased incidence of dyslexia and dysgraphia in families of children with the disorder, as has been reported by other groups (Decker & DeFries, 1980; Wolff & Melngailis, 1994). For a single measure, the frequencies of affected siblings and parents ranged from 12% to 68% and 1% to 55%, respectively. Probands were affected for a mean of 6.3 of these nine measures. Despite the distorted gender ratio among the probands (2.2 males to 1 female), sisters were equally likely as brothers to exhibit difficulties on each of nine measures analyzed categorically (see Table 2). The number of measures for which a sibling was affected did not differ by sex. The size of the discrepancy between proVIQ and achievement on reading and spelling measures was similar for affected brothers and affected sisters, with the exception of WIAT spelling on which brothers demonstrated larger discrepancies (p [is less than] .05, see Table 3). The prevalence of affected status was more skewed by gender for adults. With the exception of GORT-3 Comprehension, fathers tended to be affected more often than mothers (see Table 2). Gender was statistically associated with prevalence of deficits for GORT-3 Accuracy ([chi square] (1, N = 203) = 8.05, p [is less than] .005) and WIAT Spelling ([chi square] (1, N = 203) = 16.09, p [is less than] .0001). In addition, affected fathers tended to evidence larger discrepancies than affected mothers on WA, GORT-3 Rate and Accuracy, and WIAT and WRAT-3 Spelling, but only the latter measured skill reached statistical significance at p [is less than] .05 (see Table 3).
Table 2
Frequency of Affected Status for Component Reading Skills in Probands
and Their First-Degree Relatives(a)

 Probands Siblings
 (70 males, 32 females) (46 males, 57 females)

 Male Female Male Female
Measure % Aff(b) % Aff(b) % Aff(b) % Aff(b)

WA 84.3 (70) 71.9 (32) 32.6 (46) 47.4 (57)
PDE 67.6 (68) 67.7 (31) 15.9 (44) 16.4 (55)
GORT-3 Rate 92.5 (67) 84.4 (32) 36.4 (44) 27.8 (54)
GORT-3 Accuracy 91.0 (67) 71.9 (32) 34.1 (44) 33.3 (54)
GORT-3 Comprehension 58.2 (67) 40.6 (32) 31.8 (44) 22.2 (54)
WRAT-3 Spelling 84.3 (70) 78.1 (32) 35.6 (45) 35.1 (57)
WIAT Spelling 85.7 (70) 67.7 (31) 47.7 (44) 36.4 (55)
RAS 80.9 (68) 78.1 (32) 63.0 (46) 71.9 (57)
NWM 24.6 (69) 21.9 (32) 10.9 (46) 12.3 (57)

 (102 pairs)

 Male Female
Measure % Aff(b) % Aff(b)

WA 20.8 (101) 11.8 (102)
PDE 2.0 (102) 0 (102)
GORT-3 Rate 18.8 (101) 8.8 (102)
GORT-3 Accuracy 21.8 (101) 6.9 (102)
GORT-3 Comprehension 5.0 (101) 8.8 (102)
WRAT-3 Spelling 31.4 (102) 11.8 (102)
WIAT Spelling 20.8 (101) 2.0 (102)
RAS 61.8 (102) 48.0 (102)
NWM 3.9 (102) 2.0 (102)

(a) Some people were not tested for all the measures. In most cases
this occurred when the subject was struggling with the testing and
declined to take the measure. The Statistical Package for the
Social Sciences (SPSS) was used to analyze the data. Percent affected
did not vary systematically with gender of probands or siblings based
on [chi square] test with a continuity correction and 1 degree of
freedom. However, percent of parents affected was significantly
related to gender for GORT-3 Accuracy (at p < .005) and WIAT
spelling (p <. 0001), respectively. In both instances, fathers were
more often affected than mothers.

(b) Number tested is given in parenthesis.
Table 3
Severity of Disability for Component Measures of Reading
and Spelling Disabilities in Affected First-Degree Relatives(a,b)

 Brothers (46) Sisters (57)

 Mean Mean
 Number Discrepancy Number Discrepancy
Measure Affected (s.d.) Affected (s.d.)

WA 15 27.0 27 25.7
 (8.1) (8.9)

PDE 8 2.5 12 2.0
 (0.72) (0.6)

GORT-3 16 6.0 15 5.8
Rate (2.2) (2.2)

GORT-3 15 7.0 18 5.8
Accuracy (2.7) (1.9)

GORT-3 14 5.4 12 5.3
Comprehension (1.8) (1.8)

WRAT-3 16 27.7 20 24.8
Spelling (10.2) (5.8)

WIAT 21 28.6 20 23.0
Spelling (10.7) (6.0)

 Siblings Fathers (102)

 Number Discrepancy
Measure t test Affected (s.d.)

WA t = 0.47 21 22.7
 p = 0.64 (6.7)

PDE t = 1.69 2 1.6
 p = 0.11 (0.3)

GORT-3 t = 0.27 19 5.6
Rate p = 0.78 (1.7)

GORT-3 t = 1.48 22 5.0
Accuracy p = 0.15 (1.5)

GORT-3 t = .16 5 3.8
Comprehension p = 0.88 (.8)

WRAT-3 t = 1.09 32 24.6
Spelling p = 0.28 (7.6)

WIAT t = 2.04 21 26.6
Spelling p < .05 (7.2)

 Mothers (102) Parents

 Number Discrepancy
Measure Affected (s.d.) t test

WA 12 22.3 t = 0.144
 (9.2) p = 0.89

PDE 0 NA(c) NA(c)

GORT-3 9 4.6 t = 1.49
Rate (1.8) p = 0.15

GORT-3 7 4.6 t = 0.56
Accuracy (1.5) p = 0.58

GORT-3 9 4.6 t = -1.31
Comprehension (1.2) p = 0.22

WRAT-3 12 20.1 t = 2.01
Spelling (2.9) p = 0.05

WIAT 2 19.5 t = 1.35
Spelling (2.1) p = 0.192

(a) Shown as average discrepancy from proVIQ. Not all individuals
were tested for all measures.

(b) The severity of disability for male and female relatives was
analyzed by two-tailed t tests corrected for discontinuity
using SPSS.

(c) NA -- not applicable.

Effect of Age on Expression of Dyslexia

As observed by many others, we did not detect residual disabilities in all adults who had a strong history of reading impairment as children. Our observations on the prevalence of reading and spelling deficits in parents (see Table 2), based entirely on test results, support a suggestion, based on comparison of historical and test information by Wolff and Melngailis (1994), that females may compensate more fully for reading and spelling disability than males. Although the sizes of the IQ-achievement discrepancy of mothers and fathers categorized as affected using the strict criteria of the research protocol were not statistically different, gender effects were found when only the subset whose achievement was below the population mean was analyzed.

In these subsets of parents, the mean WRAT spelling scores for fathers was 82.7 and for mothers it was 91.4 (t(78) = -3.85, p [is less than] 0.0001), the mean WIAT spelling score was 83.5 for fathers and 92.7 for mothers (t(60) = -2.88, p = 0.006), and the mean WA score was 89.5 for fathers and 93.1 for mothers (t(71) = -2.393, p [is less than] 0.019). The mean proVIQ-WRAT spelling discrepancy was 18.73 for fathers and 10.10 for mothers (t(78) = 3.65, p [is less than] 0.001) and the mean proVIQ-WIAT spelling discrepancy was 17.67 for fathers and 7.16 for mothers (t(60) = 3.84, p [is less than] 0.001). The mean discrepancies between proVIQ and WA score for fathers and mothers were not significantly different -- 13.19 for fathers and 10.9 for mothers (t(71) = 0.797, p. = 0.428). The fathers' and mothers' mean proVIQ scores were not statistically different for any of the subsets analyzed.

Of interest, a study of spelling disability in German families found an excess of affected male relatives compared to female relatives, but the data were not provided to determine if the excess was entirely contributed by the parents (Schulte-Korne et al., 1996). It is possible that severity of disability correlates with probability and degree of compensation. This possibility has implications for family studies because offspring of parents with a childhood history of reading disability who still show evidence of reading disability by age- or IQ-discrepant criteria appear to be at increased risk for reading disorders than offspring of compensated parents (Gilger et al., 1996). To account for the phenomenon of age- and gender-related expressivity, age and gender were included as covariates in all our quantitative aggregation analyses.


In the past decade, genetic studies of reading and spelling disabilities have advanced our understanding of this complex heterogeneous behavioral disorder. Possible sites have been identified in the genome for genes involved in these phenotypes. Although the next steps -- those of cloning the genes for dyslexia and identifying their roles in normal development -- are daunting, there is reason to be optimistic, given the remarkable progress of the human genome project, technological advances in high throughput genotyping, and development of statistical methods to handle the vast quantities of data that will be collected. Careful assessment of the dyslexia phenotype is a crucial component of the research endeavor for finding the genes and applying that knowledge to identifying children at risk for this heterogeneous disorder for purposes of instructional intervention.


Alarcon, M., DeFries, J. C., & Fulker, D. W. (1995). Etiology of individual differences in reading performance: A test of sex limitation. Behavior Genetics, 25, 17-23.

Alarcon, M., DeFries, J. C., & Gillis, J. J. (1994). Familial resemblance for measures of reading performance in families of reading-disabled and control twins. Reading and Writing: An Interdisciplinary Journal, 6, 93-101.

Alarcon, M., Plomin, R., Fulker, D. W., Corley, R., & DeFries, J. C. (1998). Multivariate path analysis of specific cognitive abilities data at 12 years of age in the Colorado Adoption Project. Behavior Genetics, 28, 255-64.

Allred, R. A. (1990). Gender differences in spelling achievement in grades 1 through 6. Journal of Educational Research, 83, 187-193.

Berninger, V. (2000). Dyslexia, the invisible, treatable disorder: The story of Einstein's Ninja Turtles. Learning Disability Quarterly, 23, 175-195.

Berninger, V. W., Abbott, R. D., Thomson, J. B., & Raskind, W. H. (2001). Phenotype for reading and writing disability: A life span approach. Scientific Studies of Reading, 5, 59-105.

Bisgaard, M. L., Eiberg, H., Moiler, N., Niebuhr, E., & Mohr, J. (1987). Dyslexia and chromosome 15 heteromorphism: negative LOD score in a Danish material. Clinical Genetics, 32, 118-119.

Borecki, I. B., & Suarez, B. K. (2001). Linkage and association: basic concepts. Advances in Genetics, 42, 45-66.

Bouchard, T. J., Jr. (1998). Genetic and environmental influences on adult intelligence and special mental abilities. Human Biology, 70, 257-279.

Bruck, M. (1990). Word recognition skills of adults with childhood diagnoses of dyslexia. Developmental Psychology, 26, 439-454.

Bruck, M. (1992). Persistence of dyslexics' phonological awareness deficits. Developmental Psychology, 26, 874-888

Bruck, M. (1993). Word recognition and component phonological processing skills of adults with childhood diagnosis of dyslexia. Developmental Review, 13, 258-268.

Cardon, L. R., Smith, S. D., Fulker, D. W., Kimberling, W. J., Pennington, B. F., & DeFries, J. C. (1994). Quantitative trait locus for reading disability on chromosome 6. Science, 266, 276-279.

Cardon, L. R., Smith, S. D., Fulker, D. W., Kimberling, W. J., Pennington, B. F., & DeFries, J. C. (1995). Quantitative trait locus for reading disability: correction [letter]. Science, 268, 1553.

Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel, D. E., Gaskell, P. C., Small, G. W., Roses, A. D., Haines, J. L., & Pericak-Vance, M. A. (1993). Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families. Science, 261, 921-923.

Decker, S. N., & DeFries J. C. (1980). Cognitive abilities in families with reading disabled children. Journal of Learning Disabilities, 13, 517-522.

Decker, S. N., & DeFries J. C. (1981). Cognitive ability profiles in families of reading-disabled children. Developmental Medicine and Child Neurology, 23, 217-227.

DeFries J. C. (1989). Gender ratios in reading disabled children and their affected relatives: a commentary. Journal of Learning Disabilities, 22, 544-545.

DeFries, J. C., Alarcon, M., & Olson, R. K. (1997). Genetic aetiologies of reading and spelling deficits: Developmental differences. In C. Hulme & M. Snowling (Eds.), Dyslexia: Biology, cognition and intervention (pp. 20-37). London: Whurr Publishers, Ltd.

DeFries J. C., & Decker, S. N. (1982). Genetic aspects of reading disability: a family study. In R. H. Malatesha & P. G. Aaron (Eds.), Reading disorders: Varieties and treatments (pp. 225-279). New York: Academic Press.

DeFries J. C., Fulker, D. W., & La Buda, M. C. (1987). Evidence for a genetic aetiology in reading disability of twins. Nature, 329, 537-539.

DeFries, J. C., Olson, R. K., Pennington, B. F., & Smith, S. D. (1991a). Colorado reading project: Past, present, and future. Learning Disabilities, 2, 37-46.

DeFries, J. C., Stevenson, J., Gillis, J. J., & Wadsworth, S. J. (1991b). Genetic etiology of spelling deficits in the Colorado and London twin studies of reading disability. Reading and Writing: An Interdisciplinary Journal, 3, 271-283.

Denckla, M., & Rudel, R. (1976). Rapid "automatized" naming (R.A.N.): Dyslexia differentiated from other learning disabilities. Neuropsychologia, 14, 471-479.

Eden, G., Van Meter, J., Rumsey, J., Maisog, J., Woods, R., & Zeffiro, T. (1996) Abnormal brain processing of visual motion in dyslexia revealed by functional brain imaging. Nature, 383, 66-69.

Fagerheim, T., Raeymaekers, P., Tonnessen, F. E., Pedersen, M., Tranebjaerg, L., & Lubs, H. A. (1999). A new gene (DYX3) for dyslexia is located on chromosome 2. Journal of Medical Genetics, 36, 664-669.

Fagerheim, T., Raeymaekers, P., Tonnessen, F. E., Sandkuijl, L. A., H. A. Lubs, H. A., & Tranebjaerg, L. (2000). A genome wide search for dyslexia loci in a large Norwegian family. American Journal of Human Genetics, 67 (supplement), A1712.

Felton, R. H., Naylor, C. E., & Wood, F. B. (1990). Neuropsychological profile of adult dyslexics. Brain and Language, 39, 485-497.

Field, L. L., & Kaplan, B. J. (1998). Absence of linkage of phonological coding dyslexia to chromosome 6p23-p21.3 in a large family data set. American Journal of Human Genetics, 63, 1448-1456.

Finucci, J. M., Guthrie, J. T., Childs, A. L., Abbey, H., & Childs, B. (1976). The genetics of specific reading disability. Annals of Human Genetics, 40, 1-20.

Fisher, S. E., Marlow, A. J., Lamb, J., Maestrini, E., Williams, D. F., Richardson, A. J., Weeks, D. E., Stein, J. F., & Monaco, A. P. (1999). A quantitative-trait locus on chromosome 6p influences different aspects of developmental dyslexia. American Journal of Human Genetics, 64, 146-156.

Froster, U., Schulte-Korne, G., Hebebrand, J., & Remsschmidt, H. (1993). Cosegregation of a balanced translocation (1;2) with retarded speech development and dyslexia. Lancet, 342, 178.

Gayan, J., Smith, S. D., Cherny, S. S., Cardon, L. R., Fulker, D. W., Brower, A. M., Olson, R. K., Pennington, B. F., & DeFries, J. C. (1999). Quantitative-trait locus for specific language and reading deficits on chromosome 6p. American Journal of Human Genetics, 64, 157-164.

Geschwind, N., & Behan, P. O. (1982). Left-handedness: association with immune disease, migraine, and developmental learning disorder. Proceedings of the National Academy of Sciences, 79, 5097-5100.

Gilger, J. w., Borecki, I. B., DeFries, J. C., & Pennington, B. F. (1994). Commingling and segregation analysis of reading performance in families of normal reading probands. Behavior Genetics, 24, 345-355.

Gilger, J. W., Hanebuth, E., Smith, S. D., & Pennington, B. F. (1996). Differential risk for developmental reading disorders in the offspring of compensated versus noncompensated parents. Reading and Writing: An Interdisciplinary Journal, 8, 407-417.

Gilger, J. W., Pennington, B. F., & DeFries, J. C. (1991). Risk for reading disability as a function of parental history in three family studies. Reading and Writing, 3, 205-217.

Gilger, J. W., Pennington, B. F., Harbeck, J., DeFries, J. C., Kotzin, B., Green, P., & Smith, S. (1998). A twin and family study of the association between immune system dysfunction and dyslexia using blood serum immunoassay and survey data. Brain and Cognition, 36, 310-333.

Grigorenko, E. L., Wood, F. B., Meyer, M. S., Hart, L. A., Speed, W. C., Shuster, A., & Pauls, D. L. (1997). Susceptibility loci for distinct components of developmental dyslexia on chromosomes 6 and 15. American Journal of Human Genetics, 60, 27-39.

Grigorenko, E. L., Wood, F. B., Meyer, M. S., & Pauls, D. L. (2000). Chromosome 6p influences on different dyslexia-related cognitive processes: further confirmation. American Journal of Human Genetics, 66, 715-723.

Grigorenko, E. L., Wood, F. B., Meyer, M. S., Pauls, J.E.D., Hart, L. A., & Pauls, D. L. (2001). Linkage studies suggest a possible locus for developmental dyslexia on chromosome 1p. American Journal of Medical Genetics, 105, 120-129.

Hallgren, B. (1950). Specific dyslexia (congenital word-blindness): A clinical and genetic study. Acta Psychiatrica et Neurologica Supplement, 65, 1-287.

Heiervang, E. R., Apold, J., Lund, A., Smievell, A. I., Lovie, R., Mollerhaus, L., & Hugdahl, K. (1995). Failure to replicate chromosome 6 as probable locus for reading disability. Psychiatric Genetics, 5, S87 (Poster 179).

Hinshelwood, J. (1907). Four cases of congenital word-blindness occurring in the same family. British Medical Journal, 2, 1229-1232.

Horikawa, Y., Oda, N., Cox, N. J., Li, X., Orho-Melander, M., Hara, M., Hinokio, Y., Lindner, T. H., Mashima, H., Schwarz, P. E., del Bosque-Plata, L., Horikawa, Y., Oda, Y., Yoshiuchi, I., Colilla, S., Polonsky, K. S., Wei, S., Concannon, P., Iwasaki, N., Schulze, J., Baier, L. J., Bogardus, C., Groop, L., Boerwinkle, E., Hanis, C. L., & Bell, G. I. (2000). Genetic variation in the gene encoding calpain-10 is associated with type 2 diabetes mellitus. Nature Genetics, 26, 163-175.

Hsu, L., Berninger, V. W., Thomson, J. B., Wijsman, E. M., & Raskind, W. H. (2001). Familial aggregation of dyslexia phenotypes: paired correlated measures. Submitted for publication.

James, W. H. (1992). The sex ratios of dyslexic children and their sibs. Developmental Medicine and Child Neurology, 34, 530-533.

Lander, E., & Kruglyak, L. (1995). Genetic dissection of complex traits. Science, 265, 2037-2048.

Liberman, I., Shankweiler, D., Fischer, F., & Carter, B. (1974). Explicit syllable and phoneme segmentation in the young child. Journal of Experimental Child Psychology, 18, 201-212.

Light, J. G., DeFries, J. C., & Olson, R. K. (1998). Multivariate behavioral genetic analysis of achievement and cognitive measures in reading-disabled and control twin pairs. Human Biology, 70, 215-237.

Lubs, H. A., Duara, R., Levin, B., Jallad, B., Lubs, M. L., Rabin, M., Kushch, A., & Gross-Glenn, K. (1991) Dyslexia subtypes -- genetics, behavior, and brain imaging. In D. D. Duane & D. B. Gray (Eds.), The reading brain: The biological basis of dyslexia (pp. 89-117). Parkton, MD: York Press.

Marshfield Clinic Website:

Maughan, B., & Hagell, A. (1996). Poor readers in adulthood: psychosocial functioning. Developmental Psychopathology, 8, 457-476.

Morris, D. W., Robinson, L., Turic, D., Duke, M., Webb, V., Milham, C., Hopkin, E., Pound, K., Fernando, S., Easton, M., Hamshere, M., Williams, N., McGuffin, P., Stevenson, J., Krawczak, M., Owen, M. J., O'Donovan, M. C., & Williams, J. (2000). Family-based association mapping provides evidence for a gene for reading disability on chromosome 15q. Human Molecular Genetics, 9, 843-848.

Morton, N. E. (1955). Sequential tests for the detection of linkage. American Journal of Human Genetics, 7, 277-318.

Nopola-Hemmi, J., Taipale, M., Haltia, T., Lehesjoki, A. E., Voutilainen, A., & Kere, J. (2000). Two translocations of chromosome 15q associated with dyslexia. Journal of Medical Genetics, 37, 771-775.

Olson, R. K., Datta, H., Gayan, J., & DeFries, J. (1999). A behavioral-genetic analysis of reading disabilities and component processes. In R. M. Klein & P. A. McMullen (Eds.), Converging methods for understanding reading and dyslexia (pp. 133-151). Cambridge, MA: MIT Press.

Olson, R., Forsberg, H., & Wise, B. (1994). Genes, environment, and the development of orthographic skills. In V. W. Berninger (Ed.), The varieties of orthographic knowledge I: Theoretical and developmental issues (pp. 27-71). Dordrecht, The Netherlands: Kluver Academic Publishers.

Olson, R., Wise, B., Connors, F., Rack, J., & Fulker, D. (1989) Specific deficits in component reading and language skills: genetic and environmental influences. Journal of Learning Disabilities, 22, 339-348.

Olson, J. M., Witte, J. S., & Elston, R. C. (1999). Genetic mapping of complex traits. Statistics in Medicine, 18, 2961-2981.

Omenn, G. S., & Weber, B. A. (1978). Dyslexia: search for phenotypic and genetic heterogeneity. American Journal of Medical Genetics, 1, 333-342.

Pennington, B. F., Gilger, J. W., Olson, R. K., & DeFries, J. C. (1992). The external validity of age- versus IQ-discrepancy definitions of reading disability: Lessons from a twin study. Journal of Learning Disabilities, 25, 562-573.

Pennington, B., Gilger, J. w., Pauls, D., Smith, S. A., Smith, S. D., & DeFries, J. C. (1991). Evidence for a major gene transmission of developmental dyslexia. JAMA, 266, 1527-1534.

Pennington, B. F., Smith, S. D., Kimberling, W. J., Green, P. A., & Haith, M. M. (1987). Left-handedness and immune disorders in familial dyslexics. Archives of Neurology, 44, 634-639.

Pennington, B., Van Orden, G., Smith, S., Green, P., & Haith, M. (1990). Phonological processing skills and deficits in adult dyslexics. Child Development, 61, 1753-1778.

Petrill, S. A., Saudino, K., Cherny, S. S., Erode, R. N., Hewitt, J. K., Fulker, D. W., & Plomin, R. (1997). Exploring the genetic etiology of low general cognitive ability from 14 to 36 months. Developmental Psychology, 33, 544-548.

Petryshen, T. L., Kaplan, B. J., & Field, L. L. (1999). Evidence for a susceptibility locus for phonological coding dyslexia on chromosome 6q13-q16.2. American Journal of Human Genetics, 65, A163.

Petryshen, T. L., Kaplan, B. J.; Liu, M. F., & Field, L. L. (2000). Absence of significant linkage between phonological coding dyslexia and chromosome 6p23-21.3, as determined by use of quantitative-trait methods: confirmation of qualitative analyses. American Journal of Human Genetics, 66, 708-714.

Rabin, M., Wen, X. L., Hepburn, M., Lubs, H. A., Feldman, E., & Duara, R. (1993). Suggestive linkage of developmental dyslexia to chromosome lp34-p36. Lancet, 342, 178.

Raskind, W. H., Hsu, L., Berninger, V. W., Thomson, J. B., & Wijsman, E. W. (2001). Familial aggregation of dyslexia phenotypes. Behavior Genetics, 30, 385-395.

Reynolds, C. A., Hewitt, J. K., Erickson, M. T., Silberg, J. L., Rutter, M., Simonoff, E., Meyer, J., & Eaves, L. J. (1996). The genetics of children's oral reading performance. Journal of Child Psychology and Psychiatry, 37, 425-434.

Rietveld, M. J., van Baal, G. C., Dolan, C. V., & Boomsma, D. I. (2000). Genetic factor analyses of specific cognitive abilities in 5-year-old Dutch children. Behavior Genetics, 30, 29-40.

Sawyer, D. L., Krishnamani, M.R.S., Hannig, V. L., Garcia, M., Kim, J. K., Haines, J. L., & Phillips, J. A. (1998.). Genetic analysis of phonological core deficit dyslexia (PCDD). American Journal of Human Genetics, 63, A1776.

Schulte-Korne, G., Deimel, W., Muller, K., Gutenbrunner, C., & Remschmidt, H. (1996). Familial aggregation of spelling disability. Journal of Child Psychology and Psychiatry, 37, 817-822.

Schulte-Korne, G., Grimm, T., Nothen, M. M., Muller-Myshok, B., Cichon, S., Vogt, I. R., Propping, P., & Remschmidt, H. (1998). Evidence for linkage of spelling disability to chromosome 15. American Journal of Human Genetics, 63, 279-282.

Shaywitz, S. E., Shaywitz, B. A., Fletcher, J. M., & Escobar, M. D. (1990). Prevalence of reading disability in boys and girls. JAMA, 264, 998-1002.

Shaywitz, S. E., Escobar, M. D., Shaywitz, B. A., Fletcher, J. M, & Makuch, R. (1992a). Evidence that dyslexia may represent the lower tail of a normal distribution of reading ability. New England Journal of Medicine, 326, 145-150.

Shaywitz, S. E., Fletcher, J. M., Holahan, J. M., Schneider, A. E., Marchione, K. E., Stuebing, K. K., Francis, D. J., Pugh, K. R., & Shaywitz, B. A. (1999). Persistence of dyslexia: the Connecticut longitudinal study at adolescence. Pediatrics, 104, 1351-1359.

Shaywitz, S. E., Fletcher, J. M., Holahan, J. M., & Shaywitz, B. A. (1992b). Discrepancy compared to low achievement definitions of reading disability: results from the Connecticut Longitudinal Study. Journal of Learning Disabilities, 25, 639-648.

Siegel, L. S. (1999). Issues in the definition and diagnosis of learning disabilities: A perspective on Gluckenberger v. Boston University. Journal of Learning Disabilities, 32, 350-361.

Smith, S. D., Kimberling, W. J., & Pennington, B. F. (1991). Screening for multiple genes influencing dyslexia. Reading and Writing, 3, 285-298.

Smith, S. D., Kimberling, W. J., Pennington, B. F., & Lubs H. A. (1983). Specific reading disability: identification of an inherited form through linkage analysis. Science, 219, 1345-1347.

Smith, S. D., Pennington, B. F., Kimberling, W. J., Fain, P. R., Ing, P. S., & Lubs H. A. (1986). Genetic heterogeneity in specific reading disability. American Journal of Human Genetics, 39 (Suppl), A169.

Smith, S. D., Pennington, B. F., Kimberling, W. J., & Ing, P. S. (1990). Familial dyslexia: use of genetic linkage data to define subtypes. Journal of the American Academy of Child and Adolescent Psychiatry, 29, 204-213.

Stanovich, K. E. (1994). Annotation: does dyslexia exist? Journal of Child Psychology and Psychiatry, 35, 579-595.

Stevenson, J., Graham, P., Fredman, G., & McLoughlin, V. (1987). A twin study of genetic influences on reading and spelling ability and disability. Journal of Child Psychology and Psychiatry, 28, 229-247.

Tallal, P., Ross, R., & Curtiss, S. (1989). Familial aggregation in specific language impairment. Journal of Speech and Hearing Disorders, 54, 167-173.

Thomson, G. (1986). Determining the mode of inheritance of RFLP-associated diseases using the affected sib-pair method. American Journal of Human Genetics, 93, 207-221.

Torgesen, J. K., Wagner, R. K., & Rashotte, C. A. (1999). Test of Word Reading Efficiency (TOWRE). Austin, TX: Pro-Ed.

Vellutino, F. (1979). Dyslexia, theory, and research. Cambridge, MA: MIT Press.

Vogel, S. (1990). Gender differences in intelligence, language, visual-motor abilities, and academic achievement in wtudents with learning disabilities: A review of the literature. Journal of Learning Disabilities, 23, 44-52.

Vogler, G. P., DeFries, J. C., & Decker S. N. (1985). Family history as an indicator of risk for reading disability. Journal of Learning Disabilities, 18, 419-421.

Wadsworth, S. J., DeFries, J. C., Stevenson, J., Gilger, J. W., & Pennington, B. F. (1992). Gender ratios among reading-disabled children and their siblings as a function of parental impairment. Journal of Child Psychology and Psychiatry, 33, 1229-1239.

Wadsworth, S. J., Knopik, V. S., & DeFries, J. C. (2000). Reading disability in boys and girls: No evidence for a differential genetic etiology. Reading and Writing: An Interdisciplinary Journal, 13, 133-145.

Wadsworth, S. J., Olson, R. K., Pennington, B. F., & DeFries, J. C. (2000). Differential genetic etiology of reading disability as a function of IQ. Journal of Learning Disabilities 33, 192-199.

Wagner, R., & Torgesen, J. (1987) The nature of phonological processing and its causal role in the acquisition of reading skills. Psychological Bulletin, 101, 192-212.

Wagner, R., & Torgesen, J. (1999). Comprehensive Test of Phonological Processing (CTOPP). Austin, TX: Pro-Ed.

Wechsler, D. (1981). Wechsler Adult Intelligence Scale -- Revised (WAIS-R). San Antonio, TX: The Psychological Corporation.

Wechsler, D. (1992a). Wechsler Intelligence Scale for Children -- Third Edition (WISC-III). San Antonio, TX: The Psychological Corporation.

Wechsler, D. (1992b). Wechsler Individual Achievement Test (WIAT). San Antonio, TX: The Psychological Corporation.

Wiederholt, J., & Bryant, B. (1992). Gray Oral Reading Test -- Third Edition (GORT-3). Odessa, FL: Psychological Assessment Resources.

Wijsman, E. M., & Amos, C. I. (1997). Genetic analysis of simulated oligogenic traits in nuclear and extended pedigrees: summary of GAW10 contributions. Genetic Epidemiology, 14, 719-735.

Wijsman, E. M., Peterson, D., Leutenegger, A.-L., Thomson, J. B., Goddard, K.A.B., Hsu, L., Berninger, V. W., & Raskind, W. H. (2000). Segregation analyses of individual measures of dyslexia I: Nonword Memory and Digit Span. American Journal of Human Genetics, 67, 631-646.

Wilkinson, G. (1993). Wide Range Achievement Test -- Third Edition (WRAT-3). Wilmington, DE: Wide Range, Inc.

Wolf, M. (1986). Rapid alternating stimulus naming in the developmental dyslexias. Brain and Language, 27, 360-379.

Wolf, M., Bally, H., & Morris, R. (1986). Automaticity, retrieval processes, and reading: a longitudinal study in average and impaired reading. Child Development, 57, 988-1000.

Wolff, P. J., & Melngailis, I. (1994). Family patterns of developmental dyslexia: clinical findings. American Journal of Medical Genetics (Neuropsychiatric Genetics), 54, 122-131.

Woodcock, R. (1987). Woodcock Reading Mastery Tests -- Revised (WRMT-R). Circle Pines, MN: American Guidance Service.


I am indebted to my colleagues Drs. Virginia Berninger and Ellen Wijsman for their invaluable collaboration in the research and for many helpful editorial suggestions, and to my colleague Dr. Robert Abbott, whose statistical expertise has contributed to many aspects of the research, including the categorical analyses reported here.

This work was supported by P50 33812 from the National Institute of Child Health and Human Development.

Requests for reprints should be addressed to: Wendy Raskind, Department of Medicine, Box 35-7720, University of Washington, Seattle, WA 98195.

WENDY H. RA. SKIND, M.D., Ph.D., is associate professor of medicine, Division of Medical Genetics, University of Washington.
COPYRIGHT 2001 Council for Learning Disabilities
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Raskind, Wendy H.
Publication:Learning Disability Quarterly
Date:Jun 22, 2001

Related Articles
The relationships of phonemic awareness and rapid naming to reading development.
Comparison of faster and slower responders to early intervention in reading: differentiating features of their language profiles.
Reading by design: Evolutionary psychology and the neuropsychology of reading.
Bad readout from DNA: genes that act on brain may promote dyslexia.
Classification of students with reading comprehension difficulties: the roles of motivation, affect, and psychopathology.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters