Skip over navigation

Journal Issue: Special Education for Students with Disabilities Volume 6 Number 1 Spring 1996

Learning Disabilities
G. Reid Lyon

New Knowledge of Reading Disabilities

Since the inauguration of the NICHD Learning Disability Research Network in 1987, researchers have learned the most about learning disabilities that affect linguistic, reading, and spelling abilities and the least about learning disabilities in mathematics.2 A number of new findings have also been obtained in the area of attention deficit disorder (ADD) and its relationship to different types of LD, particularly disorders in reading.12 For brevity, the major discoveries made during the past several years are presented in Table 1. Selected findings are reviewed here. The reader should note that many findings have been replicated by multiple research groups, as cited in Table 1, and that the findings are primarily based on large longitudinal samples. Finally, readers should note that studies being conducted in Canada by Stanovich and Siegel at the Ontario Institute for Studies in Education are included in Table 1 because of their impact on the field and because Stanovich and Siegel serve as consultants to the Yale Learning Disability Research Center (LDRC).

As Table 1 shows, a majority of discoveries made during the past decade have been in the area of reading disabilities. This is appropriate. As Lerner pointed out from her analysis of public school referral data in 1989,33 approximately 80% of children identified as having learning disabilities have their primary difficulties in learning to read. This high rate of occurrence of reading difficulties among youngsters with LD has also been reported by Kavale in his meta-analytic studies.34 More recent longitudinal and cross-sectional studies have supported the high rate of reading difficulty among children with learning disabilities, but have also found that reading deficits frequently co-occur with other academic and attentional difficulties. For example, Fletcher and his associates at the Yale Center for the Study of Learning and Attention have, as part of a larger classification effort, studied 216 children, 7.5 to 9.5 years of age, who were identified as normal readers, reading disabled, math disabled, both reading and math disabled, normal reading with ADD, and reading disabled with ADD.21,35 From this sample of children with a variety of learning disabilities, only 25 youngsters were reading at age-appropriate levels.

Research indicates that reading disorders reflected in deficient decoding and word-recognition skills are primarily caused by deficiencies in the ability to segment syllables and words into constituent sound units called phonemes.16,22,36-38 For example, in a large study of 199 seven- to nine-year-old children who had significant difficulties in decoding and word recognition, more than 85% of the youngsters manifested deficits on measures of phonological awareness. In this investigation, children with and without IQ-reading-achievement discrepancies appeared equally impaired on both the phonological and reading measures.21 This extremely high frequency of phonological awareness deficits in children with reading disabilities has led Share and Stanovich to conclude: "We know unequivocally that less-skilled readers have difficulty turning spellings into sounds. . . . This relationship is so strong that it deserves to be identified as one, if not the defining, feature of reading disability."39

Biological Bases

Several NICHD investigations have indicated that these phonologically based reading disabilities are linked to neurobiological and genetic factors.2,8,13,40 Functional and structural neuroimaging studies indicate that the poor phonological skills, which limit the development of basic reading abilities, are highly related to aberrant neurophysiological processing.22,40 Moreover, there is increasing evidence from behavioral and molecular genetic studies that the phonological deficits observed in reading disability are heritable.41,42 Taken together, longitudinal studies of the linguistic, neurobiological, and genetic factors in reading disabilities provide strong and converging evidence that reading disability is primarily caused by deficits in phonological processing and, more specifically, phonological awareness.8,13,30,37,38,40

Likewise, the data derived from genetic and neurobiological studies suggest that some reading disabilities are associated with subtle chromosomal42 and neurological differences,22,40 indicating that such disabilities are biologically "real" rather than sociopolitically created.

Discrepancy Standard

In addition to the previously discussed problems of the discrepancy standard, Table 1 indicates that the use of a discrepancy formula, which calculates differences between IQ and reading scores, is not a valid indicator of reading disability; that is, children with reading disabilities both with and without such discrepancies have similar deficits in phonological awareness and similar genetic and neurophysiological characteristics.36 At this time, it is not clear whether children with higher IQs respond more favorably to intervention.7

Persistent Deficit

Unfortunately, as Table 1 indicates, reading disabilities appear to reflect a persistent deficit rather than a developmental lag. That is, children with delays in understanding phonological concepts in first grade are unlikely to catch up later without explicit and informed teaching. Longitudinal studies show that, of the youngsters who are identified in the third grade, approximately 74% remain reading disabled through the ninth grade.19,43 This appears to be true even when special education has been provided. It should be made clear, however, that interventions applied after a child has failed in reading for two or three years may not be effective for several reasons, including the student's declining motivation and impaired self-concept. Instructional difficulties in later intervention abound. For example, the teacher carrying out the interventions may not be properly trained, the interventions may not include explicit and informed instruction in the development of phonological awareness and sound-symbol relationships, the interventions may not be consistently applied and/or may be limited in intensity and duration, and there may be insufficient follow-up or explicit instruction to enable the student to generalize the specific concepts learned to material presented in regular classroom settings.

Distribution of Severity

A significant finding from the Yale LDRC is that reading disability represents the extreme of a normal distribution of reading ability so that there is an unbroken continuum from reading ability to reading disability.43 The finding that reading disability is part of a continuum now places the disorder in the context of other biologically based disorders such as hypertension and obesity.43 The discovery that reading disability is best conceptualized as occurring along a normal distribution of reading skills underscores the fact that children will vary in their level of severity of the disorder running along a mild-to-severe spectrum, with the majority of children with reading disabilities falling at the mild end. This finding has significant implications. For example, what are the criteria for identifying a child as having a severe reading disability, and does this degree of disability warrant entitlement to a greater intensity and duration of specialized interventions?

To answer such questions, the NICHD is embarking on a series of studies to identify the most valid points along the distribution of reading scores that distinguish levels of severity. In part, the validity of different cutoff points for mild, moderate, and severe reading disability is being determined by how children in each severity group respond to different types and intensities of intervention. At this writing, some initial results derived from the Florida State Intervention Project show that children with scores at the extreme lower end of the distributions for both phonological awareness skills and basic reading skills are much more difficult to remediate than children who fall along the distribution in the mild and moderate ranges.44-46 It is as yet unclear whether children in the more severe range can achieve age- and grade-approximate reading skills, even with intense, informed intervention provided over a protracted period of time.

While children with severe reading disabilities will most likely require a greater amount of time in high-impact intervention programs than children with less severe deficits, as discussed earlier, it is clear that the longer children at any level of severity go without proper identification and intervention, the more difficult the task of remediation and the harder it becomes for the children to respond. It is also clear that even children with relatively subtle linguistic and reading deficits require the expertise of a teacher who is well trained and informed about the relationships between language development and reading development.30 Unfortunately, such teachers are in short supply, primarily because of a lack of programs providing this training.31