What is the process of reading development?

3.2 Reading Development Theoretical Underpinnings

Other reading development models and vocabulary and comprehension theories that are not unique to ELs but that also can be operationalized when students are learning L1 or L2 are Anderson and Freebody's (1981) knowledge hypothesis, Kintsch's (1988) discourse comprehension theory, and Ehri's phases of reading development. The knowledge hypothesis (Anderson & Freebody, 1981) suggests that knowledge of words reflect background knowledge and having a schema of the topic or theme. Thus, learning the meaning of words and phrases used in texts and the meaning of words in the context of specific types of texts is what helps students comprehend text (Stahl, 1999).

Kintsch's (1988) discourse comprehension theory suggests that it is important for adults to ask young children questions and to design activities that stimulate thinking and construction of deep knowledge of vocabulary. First, there is the notion of levels of understanding, varying from superficial to deep understanding. Our knowledge of the conditions that foster different levels of understanding can guide instruction. Deep learning requires integration of prior knowledge with new information and results in the ability to use this information constructively in new contexts (the formation of a situation model, Kintsch, 1998). Second, the type of mental representation students form (superficial understanding versus adequate situation model) is determined by the comprehension strategies they use (Paris, Wasick, & Turner, 1991; Pressley & McCormick, 1995). Thus, ELs benefit from rich multimodal representations of words and associated concepts in addition to the support from adults to reason about and express their understandings of new information.

Ehri's phases of reading development model (Ehri, 2005) helps to explain some potential language bottlenecks ELs may face as they learn to read in English. Ehri suggests three phases of reading acquisition, the prereading phase, the learning to read phase, and the reading to learn phase. In the prereading phase, young children begin to develop the fundamental language skills that are necessary for learning to read and, as they hear books being read to them, they learn the foundation of code-focused skills: print awareness and phonological awareness. As children increase their receptive language, they understand better spoken language and increase their listening comprehension. The prereading phase may be substantially different for young ELs who at kindergarten entry, may not have yet mastered skills their EO peers typically learn during that phase.

The learning to read phase begins around the time children enter kindergarten, when they begin to understand the correspondence between letters and sounds and benefit from systematic and explicit instruction to understand how to crack the alphabetic code. They learn to decode words of increasing difficulty with automaticity and they learn to recognize many irregular words by sight. Thus, by the time they move from this phase to “reading to learn,” they will reliably use the correspondences between letters and sounds in words to identify words they have never seen before in print. Yet, they can only master this phase by also using their vocabulary and language skills to support the integration of letter sound correspondences to background knowledge and the meaning of the sentence or story, to find a word that matches the sounds they have decoded and that also “makes sense.” For ELs, who in general might be able to crack the alphabetic code as fast or faster than their EO peers (Baker & Baker, 2008; Lesaux & Siegel, 2003), their limited vocabulary and language proficiency can create a powerful roadblock to entering the next phase of reading, reading to learn. In this later phase, students are expected to independently gain information and build knowledge and critical thinking skills as a result of what they have read. Moreover, they are expected to decode increasingly more complex words that require them to understand the meaning of the morphemes in the word to support their text comprehension.

Therefore, an important way to prepare ELs for the learning to read and reading to learn phase is by providing them with strong language support in their L1 and in English. Strong language skills in the L1 can reduce the language demands of children, particularly in the area of vocabulary where children can learn new labels for words they already know in their L1 (e.g., the word in English for “silla” is “chair”). In addition, given that language is the foundation for learning to read in any language, developing children's language proficiency in their native skills could also support their reading development in L2 if this support is provided intentionally and with high quality. In summary, it will generally be easier to learn an L2, such as English, once students have mastered their L1. This is particularly relevant to Hispanic students where extensive research indicates a positive transfer of certain reading skills from Spanish to English. Less research exists, however, for other languages (with markedly different structures) and for other orthographies that are not alphabetic.

B Reading strategies in Down syndrome

Theories of reading development describe the stages through which typically developing readers progress from the early stages of literacy to a skilled level (e.g., Frith, 1985). At first, readers rely on visual recognition of the whole written word and link this to its pronunciation; this is the logographic stage. This strategy depends on learning every printed word along with its spoken form. Later, children reach the alphabetic stage of reading and are able to “sound out” new words based on a knowledge of letter sounds. This phonological route to reading offers a more flexible strategy enabling children to read new words and nonsense words. Skilled readers no longer need to read most words letter-by-letter but recognize them directly. This visual route to reading is also important for reading irregular words, such as yacht, where a phonological approach would not be helpful.

The phonological route to reading depends on the development of PA. PA is established as a predictor of typically developing children's reading success (e.g., Bradley & Bryant, 1983). It has been assessed using a wide range of tasks including letter sound knowledge, and those that require segmentation or blending of words into or from constituent phonemes or larger segments such as rhyme endings. Early reports described children with DS as logographic readers, developing substantial sight word vocabularies (Buckley, 1985), and suggested that they learned to read without PA (Cossu, Rossini, & Marshall, 1993). However, numerous subsequent studies have shown that some individuals with DS do develop PA (for review, see Lemons & Fuchs, 2010). Group performance in these studies is generally significantly poorer than that of comparison groups of typically developing children matched on any criterion. There may also be differences in the order in which particular PA skills are acquired with children more successful on tasks that require detection of onset phonemes than those that depend on sensitivity to rhyme (e.g., Snowling, Hulme, & Mercer, 2002). Despite lower levels of PA, there are similar relationships between PA scores and word reading to those found in MA- or reading-matched groups (e.g., Fowler et al., 1995; Roch & Jarrold, 2008). As in typical development, longitudinal studies show that early PA is predictive of later reading ability (Cupples & Iacono, 2000; Kay-Raining Bird, Cleave, & McConnell, 2000; Laws & Gunn, 2002). In our own recent study, Time 1 PA scores were related to later reading scores but there was no difference between the size of the correlations between regular and irregular words and PA, and no significant correlation between PA and nonword reading. These results suggest that children were not relying on phonological strategies to advance their reading. Interestingly, early PA was associated with later scores on tests of receptive vocabulary, VSTM, and speech articulation; partial correlations between PA and these functions remained significant after accounting for differences in CA and nonverbal MA, and even early reading scores (Laws et al., 2010).

Despite the importance of PA and letter sound knowledge for facilitating the more flexible phonological reading route, and advice to teach phonics to children with DS once they have established a sight word vocabulary (Buckley et al., 1996), relatively few intervention studies have been published (Cupples & Iacono, 2002; Goetz et al., 2008; Kennedy & Flynn, 2003; van Bysterveldt, Gillon, & Moran, 2006). These studies suggest that improvements to letter sound knowledge, and other PA skills such as onset and rhyme segmentation and blending, are possible following explicit instruction (e.g., Kennedy & Flynn, 2003; van Bysterveldt et al.). However, there is less evidence for improvements in word or nonword reading following letter sound and PA instruction (Cupples & Iacono, 2002) suggesting that children may not necessarily use the PA skills they acquire.

For the most part, intervention studies have provided relatively short periods of teaching with sessions totaling 4–8 h spread over 4–6 weeks. Goetz et al. (2008) offered a more substantial intervention involving daily 40-min sessions of one-to-one teaching by specially trained learning support assistants over 16 weeks for one group and 8 weeks for a second group that also provided a waiting list control. The first group made gains in letter sound knowledge and early word reading after 8 weeks, relative to the controls. The second group also made small gains after 8 weeks. These gains were statistically nonsignificant but it was thought that this was probably due to a combination of small sample size and wide variation in participants' abilities. The gains made by both groups were maintained after several weeks suggesting that intervention can be successful with higher intensity of teaching over longer periods.

Summary

Learning to read is a specific phase in reading development that all children must pass through. Research has provided a wealth of data to inform practice. Attention to both the content of instruction and the delivery of instruction is necessary as is the use of assessment to guide instruction. In planning instruction that will meet students’ needs in a cohesive and systematic manner, ensure that there are common goals across the curriculum and contexts. Response to instruction provides a means for providing such instruction. When implemented with fidelity it includes: (1) high quality, research-based instruction; (2) data-based decision-making; (3) clearly articulated procedures; and (4) clearly articulated criteria for movement between more intensive tiers of instruction.

Neural Networks for Reading

Another area of concentrated research interest is the study of reading development, both in typically developing and dyslexic children. Acquiring literacy skills impacts the functional organization of the brain, differentially recruiting networks for language, visual, and sound representation in both hemispheres, as well as increasing the amount of white-matter tissue connecting brain areas. Work on individual differences in the cognitive paths to reading has enriched the interpretation of the neurological research (e.g., Knight and Fischer, 1992), and helped to bridge the gap between the neuroscience findings and classroom practice (Katzir and Pare-Blagoev, 2006; Wolf and O'Brien, 2006). In dyslexic readers, progress is being made toward better understanding of the contributions of rapid phonological processing (Gaab et al., 2007), orthographic processing (Bitan et al., 2007), and visual processing to reading behaviors, as well as to thinking in other domains (Boets et al., 2008). For example, the visual field of dyslexics may show more sensitivity in the periphery and less in the fovea compared to nondyslexics, leading to special talents in some dyslexics for diffuse-pattern recognition (Schneps et al., 2007). Most recently, research looking at developmental differences in neurological networks for reading across cultures has begun to appear (e.g., Cao et al., 2009), which ultimately may contribute to knowledge about how different kinds of reading experiences shape the brain.

The neural networks for learning reading and math have important implications for education, as the most effective lessons implicitly scaffold the development of brain systems responsible for the various component skills. For example, successful math curricula help students to connect skills for calculation with those for the representation of quantity, through scaffolding the development of mental structures like the number line (Carey and Sarnecka, 2006; Griffin, 2004; Le Corre et al., 2006). While different students will show different propensities for the component skills, all students will ultimately need to functionally connect the brain systems for quantity and calculation to be successful in math.

Conclusions

In this chapter, we have argued that in order to fully understand reading development and disabilities, both within and across different languages, we need a better meta-theoretical framework to guide both the research itself and the interpretation of research findings. We presented briefly one such framework, the Multiple Systems Model of Reading, that considers normal reading development and disabilities as different states of the same complex developing system, with no specific interactant being allocated the causal control of the developmental process. This, naturally, does not mean that researchers should give up the search for specific genetic, neural, or cognitive mechanisms that contribute to reading development and disabilities, or the development of theories that explain the contribution of these mechanisms. On the contrary, such work is necessary and has greatly increased our understanding of the genetic, neural, and cognitive interactants that any explanatory model needs to include. What the Multiple Systems Model of Reading leads us to argue is that all of these theories are enriched if they conceptualize development at all levels as construction and focus on identifying the interactants of this process (see e.g., Galaburda, LoTurco, Ramus, Fitch, & Rosen, 2006, for an example in the genetic-neural level), and that none of these theories per se can constitute a sufficient explanation of development of the target behavior, be it reading or academic achievement. In particular, educational practice is well served by respecting the complex, and sometimes idiosyncratic, nature of the developing organism, and by not allocating causal responsibility—and all of the assessment and remedial efforts—for possible reading problems to any single factor.

Cognitive Characteristics of Developmental Dyslexia

For the majority of children, early phonological coding skills predict the course of their later reading development. These skills encompass sound manipulation at the phoneme level (phonemic awareness), rapid naming of letters and numbers (phonological retrieval), and short-term verbal memory (phonological recoding). Children who fail to show awareness of the phonemic structure of spoken words at the kindergarten level are more likely to struggle in acquiring grapheme–phoneme correspondence rules, and this affects their long-term ability to decode words. These observations have led to ways by which to identify children with dyslexia early, as those children with poor phonological representation are considered to be at higher risk for reading disability, even prior to the advent of formal schooling. This longitudinal research, together with the fact that treatments aimed at promoting the phonological code are fairly successful in helping children with reading disabilities, has given rise to the notion that a phonological core-deficit hypothesis represents the best explanation for dyslexia.

However, while phonological processes are impaired in many poor readers, these children may also exhibit deficits in other language-related processes, including vocabulary, morphology and syntax, and text comprehension. Likewise, measures of vocabulary and grammar attained at kindergarten are strongly associated with reading outcome in second grade. Finally, skills in the domain of orthography, such as the ability to name words based on visual information or appreciating letter patterns that are legitimate, affect reading fluency and reading comprehension in the later grades.

In sum, the most prominent cognitive deficit associated with dyslexia is that pertaining to phonological coding. However, it should not go unnoticed that other language skills make strong predictions for later reading development and that the process of reading itself promotes phonological representation. Hence it would be fair to say that dyslexia is associated with a number of language-based deficits, with poor decoding serving as the identifying feature. However, because the definition of dyslexia is largely exclusionary, it is not surprising that the result is the identification of heterogeneous populations of dyslexics. This in turn results in variability in the description of the dyslexic phenotype. Although the phonological and other language deficits are prominent and most are directly related to the process of reading acquisition, other behavioral manifestations have been observed and will be described next.

Levels of Reading

When children are learning to read, the children's levels of reading and the difficulty of the text must be taken into account. Children's decoding and comprehension of text and their ability to read fluently are affected not only by their reading development, but also by the text they are reading. The difficulty or ease in reading a text is influenced by textual features, such as the size and placement of the text; the presence or lack of pictures; difficulty of the words; the complexity of the sentence structure; and the children's familiarity with the content of the text. Recognizing that texts will pose different levels of challenge for children as they learn to read, it is important to match texts with children in ways that foster reading development. To facilitate this process, texts are generally classified as being at children's independent, instructional, borderline instructional, or frustration level (see Table 1).

Table 1. Children's levels of reading in relationship to text difficulty

Texts at children's independent level are ones children can read fluently without assistance. Reading texts on their independent level provides children with an opportunity to practice and apply the reading strategies and skills they already possess and improves their decoding, comprehension, and fluency. Children's reading of instructional-level texts promotes reading development when support or scaffolding is provided by a more adept reader. The use of instructional-level texts during reading instruction enables teachers to instruct within the children's zone of proximal development and optimizes the potential effects of the instruction provided. This is also true when reading materials at the borderline instructional level are read during instruction. However, texts at this level require a high level of teacher support. This is often the level used when children are working with a reading specialist or tutor who can provide more intensive instruction and scaffolding. Texts at children's frustration level require such a high level of support that children typically do not benefit from instruction. With these levels in mind, teachers can enhance reading development by providing children with opportunities to read a variety of texts on their independent and instructional levels with appropriate levels of support and scaffolding.

Introduction

J. P. Das has contributed numerous texts in the field of reading. In the context of his work, PASS (Planning, Attention, Simultaneous, and Successive processing) theory has informed and shaped our understanding about the cognitive processes that are involved in reading development and reading disabilities. Traditionally, the components of PASS theory have been explicitly linked to reading decoding and fluency, as well as to visual or orthographic processing, but less often to reading comprehension. The aim of this chapter is to address this issue by exploring the relation between reading comprehension and the four PASS theory processes and to provide initial empirical evidence to support this relation.

It has been demonstrated that the ability to read is largely determined by the individual’s cognitive ability. Indeed, researchers have examined the relations between intelligence and reading achievement across the lifespan, putting to the test various theories of reading and intelligence. This work has shown that it has been extremely challenging to reach consensus as to what is an acceptable relation between reading and intelligence. In this chapter, our purpose is to examine the relation between reading and intelligence in the context of current theories of reading comprehension and intelligence and provide initial empirical support of our position. It is our contention that both reading and intelligence are multidimensional constructs, and by definition, any attempt to reduce their operationalization to a single score is imperfect.

First, we present theoretical evidence for, and discuss the implication of the multidimensional nature of reading comprehension. In developing our case, we consider current theories and models of reading comprehension with an emphasis on either lower-level or higher-order processes. Second, we present PASS theory, a multidimensional view of intelligence, and attempt to identify the relations of its process components (i.e., planning, attention, simultaneous, and successive processing) and reading comprehension. In developing our case, we provide empirical evidence from an ongoing study exploring the relations between the four PASS processes and reading comprehension. To bring this theoretical contribution to a close, we then discuss several challenges that we believe hold promise for those who are committed to future research in these areas.

1 Literacy

The WS language profile (described above) has important implications for the development of literacy in children with WS. With a reported relative strength in phonological short-term memory and a common advantage of some aspects of verbal ability over nonverbal abilities, one might expect relatively good reading skills. Indeed, reading abilities appear to be correlated with verbal intellect, as is the case in the typically developing population (Howlin et al., 1998). However, it is critical to consider that: (i) reading involves a visuospatial element; (ii) reading involves auditory–visual cross-modal (integrative) learning and (iii) reading, especially reading comprehension, relies heavily on semantics and verbal working memory; all areas of relative impairment in WS. Also, there is some evidence to suggest that nonverbal (spatial) skills correlate more highly with reading abilities in WS than verbal intellect (Howlin et al., 1998), suggesting that deficits of spatial ability play an important role in the development or demise of reading skills.

Consistent with relative strengths in phonological short-term memory and verbal intellect, phonemic awareness skills and single word reading abilities, at least on average, appear to be at or above mental age (e.g., see Mervis, 2009 for a review). However, as Mervis (2009) points out, mental age equivalence scores are commonly reported within the literature, but other indices such as standard scores may be more reliable (whereby the mean is 100 and standard deviation is 15). Without the use of standard normative scores, it is difficult to gauge whether reading ability is commensurate with the level expected for IQ. Also, even when standardized scores are reported, they are usually directly compared to other standardized scores (e.g., an IQ standard score of 63 may be directly compared to a reading score of 63). We know, however, that there is no direct relationship between standardized scores on IQ tests and standardized scores on academic tests (i.e., a standard score of 70 on an IQ test does not equate to an expectation that reading ability should be 70 to be on par with IQ). Instead, psychometric tables should be used to predict reading ability based on IQ. To the best of our knowledge, no study has assessed whether reading ability is commensurate with FSIQ or VIQ in WS using psychometric tables. Thus, it remains unclear whether single word reading abilities are consistent with FSIQ, VIQ, or indeed PIQ.

So, are reading abilities delayed or deviant in WS? Laing (2002) suggested that single word reading was a relative strength, but critically, despite reasonably good abilities, individuals with WS may arrive at their level of single word reading by an atypical route. Thus, it is possible that some aspects of reading development are delayed and others deviant. For example, it has been suggested that (similar to other phonological processes), phonological awareness skills are delayed, but are unlikely to develop atypically. In contrast, the relationship between reading and semantic processing may be atypical in WS. Semantics play a role in reading development; at least within the typically developing population (e.g., see Mervis, 2009). Therefore, it is possible that the atypical development of semantics has implications for the development of reading skills. Indeed, individuals with WS seem to rely less on semantics than typically developing children when learning to read (Laing, Hulme, Grant, & Karmiloff-Smith, 2001) and for some individuals with the disorder there may be evidence of atypical semantic errors being made4 (Temple, 2003).

In summary, the jury is still out regarding whether (i) reading and reading comprehension abilities are consistent with intellectual capabilities in WS or (ii) whether reading develops typically or atypically. Again, there is striking variability of reading skills amongst individuals with the disorder and to accommodate this variability it has been noted that “reading skills range from an inability to read at all to age-appropriate decoding and comprehension” (Mervis, 2009; p. 150).

Reading

Converging lines of research (Fletcher et al., 1994; Shaywitz, Fletcher, & Shaywitz, 1996) have emphasized the primacy of core phonological-processing deficits in disabled readers. Unlike language competencies which unfold naturally in a fairly predictable fashion, reading represents an acquired skill. As such, not only constitutional but environmental determinants may contribute to failures in reading. Without an explicit model of normal reading development, patterns of impairment cannot be described.

Reading development has been traditionally dichotomized across two component skills which must be mastered. These have been described within a dual-route model of reading as involving (a) a phonological, and (b) a direct lexical route in which whole-word or orthographic-recognition skills facilitate active word-recognition (Morton, 1969). In the earliest stages of the development of pre-reading competencies, the logographic stage, mastery of the visual-orthographic properties of letters, memorization of the visual gestalts of a limited repertory of words, and utilization of visual associative skills to foster word recognition from pictures that accompany text, are accomplished. The early reader is simultaneously developing sound-symbol associative skills. Such abilities are predicated upon a child’s ability to first decompose speech into component structures (phonological awareness). Via these processes, the early reader is learning to associate visual symbols (graphemes) with their corresponding sound equivalents (phonemes). With increasing experience, direct orthographic-associative competency is established and familiar words are decoded on “sight". Confrontation with low frequency, novel words is presumed to require some combination of orthographic- and phonological-coding abilities (Coltheart, 1978). Skilled readers are presumed to have developed automatized orthographic skills in reading by the fourth grade. As such, phonological skills are relegated primarily to the processing of less familiar words. A competent reader is presumed to have developed equivalent proficiency in both aspects of word analysis. In addition, with advancing age, the capacity to incorporate morphological cues as well as lexical-semantic, and other contextual cues, further contributes to the act of reading. As such, words rich in meaning tend to be decoded with greater faculty than more ambiguous function words.

The complexity of reading-skill acquisition expressed over time requires recognition of the reciprocal contributions of higher-level processing systems, beyond the dual-route model. Cognitive models of reading development (see Chase & Tallal, 1991, for review) take into consideration increased capacity to bring on-line higher cognitive faculties (the simultaneous development of not only “bottom-up” but “top-down” processing) to the development of reading fluency and comprehension (McClelland & Rumelhart, 1981). Progression in reading proficiency requires the act of word recognition becoming automatized (Liberman, Liberman, Mattingly, & Shankweller, 1980). Thus, on-line cognitive activities are directed less at the act of decoding and permit the use of metalinguistic awareness, selective attention and working memory to support semantic encoding and comprehension monitoring. Although phonological-coding skills (word- analysis abilities) represent the most widely studied aspect of reading development, more recent research has focused upon the contribution of other types of linguistic-rule knowledge (semantic, morphological, and syntactic conventions) to the development of higher-level reading skills, ie., comprehension skills.

Associated with research on phonetic processing in deficient readers has been the identification of specific linguistic deficiencies associated with reading disabilities. In a review by Mann (1994), deficiencies associated with naming and verbal productivity, expansion of semantic knowledge, auditory sequential memory, sentence recall, and grammatical and syntactical analysis skills (particularly comprehension demands requiring the processing of more complex grammatic structures) have been identified. Catts (1989) futher identified higher rates of oral-speech deficits, including early history of articulation inefficiencies among the reading disabled. A commonality linked to these deficits involves the processing of sound patterns of language. These deficits have been broadly conceived as evidence that specific maturational lags in the systems supporting language development represent a secondary source of the cognitive deficit expressed by disabled readers. The specificity of language dysfunction related to reading inefficiency is conveyed by Tallal’s finding (1987) that 85 percent of children exhibiting language disorders in the preschool years develop language-related learning disabilities (i.e., reading problems).

In addition to phonological and language-processing abilities, visual-feature analysis is also required in the act of grapheme-phoneme correspondence. Although low-level visual deficits have been identified among reading-disabled populations (Lovegrove, Martin, & Slaghuis, 1986), their impact as factors significantly impinging upon reading-skill development appears minimal (Hulme, 1988; Vellutino & Scanlon, 1987). In addition, select memory inefficiencies have been identified in some impaired readers. A sparse network of associations in working memory and retrieval deficits have been hypothesized in these instances. Employing hierarchical-regression analysis, Vellutino, Scanlon and Tanzman (1994) utilized measures of phonological coding and analysis abilities, verbal-memory measures, semantic- and syntactical-analysis tests, and visuo-perceptual task demands as dependent variables in predicting word-analysis proficiency. Phonological-processing skills accounted for the majority of the variance associated with word-identification proficiency. Semantic and syntactic measures were identified as intermediate predictors of reading proficiency and visual abilities.

Findings such as these have reshaped conventional wisdom applied to the assessment of reading disabilities. Dissociations between normal versus impaired readers have been traditionally specified by decision rules, i.e., aptitude-achievement test discrepancies. Children with a discrepancy between IQ and an objective reading measure are classified as disabled and deemed eligible for special-education supports. Low-achieving readers are conceptualized as reading below normative standards for age. But because of associated lower level IQ scores and imputed, more generalized cognitive inefficiencies, these low-achieving readers are not presumed eligible or appropriate for special-education services. These long-standing practices have led to a bimodal conceptualization of reading deficiencies, with reading disabilities representing a hump on the lower tail of this distribution (Shaywitz, Escobar, Shaywitz, Fletcher, & Makuch, 1992). Data derived from the Connecticut Longitudinal study (Shaywitz et al., 1992) support the contention of Stanovich (1991) that there actually may be no qualitative differences between disabled and low-achieving readers, and that phonological-processing deficits represent a core deficit indentifiable in both groups. These investigators have argued that reading abilities exist on a continuum which includes superior, average, and impaired readers. This model argues against the use of any arbitrary cut point indicating normalcy versus disability (i.e., discrepancy models) and instead suggests that intervention for any individual with reading inefficiencies be driven by identification of his or her unique processing deficiencies (phonological and associated cognitive limitations). As such, virtually all poor readers will exhibit a primary phonological-processing deficit. Variance in the expression of deficiencies in impaired readers, that is, heterogeneity in the expression of reading deficiencies, is a function of the severity of the core phonological processing deficits and the nature and extent of any underlying or associated cognitive dysfunction.

These findings have aided in validating Stanovich, Nathan, and Zolman’s (1988) initial hypothesis regarding the variability expressed among impaired readers. This model presumes all disabled readers manifest a phonological-processing deficit. The most severe forms of reading disability are characterized by a fundamental or “core” deficit in the ability to establish grapheme-phoneme correspondence. As the nature of manifest impairment extends beyond core phonological-processing deficiencies, the term “variable” is attributed to the idiosyncratic manifestation of other language, attention, memory, or perceptuo-integrative skill deficits that may be additionally expressed. The model takes into account the remarkable heterogeneity expressed in reading deficiencies and why categorical models of reading, e.g, subtyping schemas, may not satisfactorily characterize the unique attributes expressed in individual cases.

Conceiving of reading problems in this fashion emphasizes the importance of defining the individual array of strengths and weaknesses expressed by any reader. This represents an alternative to models which posit more discrete subtypes of disabled readers and permits a means to conceive of reading on a continuum from normal variability in reading proficiency to the heterogeneous expression of impaired reading development.