Double-deficit Hypothesis Research Articles

Evaluation of the Double-Deficit Hypothesis Subtype Classification of Readers in Spanish

The double-deficit hypothesis acknowledges both phonological processing deficits and serial naming speed deficits as two dimensions associated with reading disabilities. The purpose of this study was to examine these two dimensions of reading as they were related to the reading skills of 29 Spanish average readers and poor readers (mean age 9 years 7 months) who met the criteria for either single phonological deficit (PD), double deficit (DD), or no deficit. DD children were the slowest readers and had the weakest orthography processing skills. No significant differences were found between PD and DD groups on word and pseudoword reading. Word reading and reading comprehension skills were average or above average in the three studied groups. As in previous studies in transparent orthographies, word reading was not a salient problem for Spanish poor readers, whereas for the DD group, reading speed and orthographic recognition skills were significantly affected.

Subtyping of children with developmental dyslexia via bootstrap aggregated clustering and the gap statistic: comparison with the double‐deficit hypothesis

The marked degree of heterogeneity in persons with developmental dyslexia has motivated the investigation of possible subtypes. Attempts have proceeded both from theoretical models of reading and the application of unsupervised learning (clustering) methods. Previous cluster analyses of data obtained from persons with reading disabilities have suffered from the inherent limitations of unsupervised learning methods. Specifically, the reliability and stability of cluster solutions have proven difficult to determine. Recent developments in the clustering literature have addressed these concerns by permitting checks on the internal validity of the solution. Resampling methods produce consistent groupings of the data independent of initialization effects, while the gap statistic provides a confidence measure for the determination of the optimal number of clusters present in the data. Combining these methods produces a robust data-driven classification that can be compared with theoretically based subtypes to produce better-informed models of developmental dyslexia. The present study is a novel application of resampling (bootstrap aggregating or bagging) methods and the gap statistic to the subtyping of children with developmental dyslexia. The specific aims of this study are: (1) to illustrate the use of bagging methods and the gap statistic in multivariate data obtained from children with developmental dyslexia; and (2) to compare the bagged clustering thresholded by the gap statistic against the predictions of the double-deficit hypothesis. The double-deficit hypothesis is a prominent theoretical model of developmental dyslexia, which predicts three subtypes: phonological, rate, and phonological-rate impaired readers. Three simulated data sets with known cluster structure were created to check the validity and illustrate the utility of the bagged clustering with the gap statistic in data with known structure. Subsequently, a clinical database of standardized test data (eight tests) from 93 children with developmental dyslexia was clustered using these methods. This procedure was repeated on a database of 93 children without reading disability matched for gender and age as a control. Finally, the clustering was repeated on the entire database of 186 participants. Cluster solutions were obtained for an increasing number of clusters (1-10) and were tested against the null hypothesis that no subtypes were present, i.e. the data represented a single cluster. Four clusters were identified in the children with developmental dyslexia. There was no evidence of significant cluster structure in the children without dyslexia. Two clusters were identified when children with and without reading impairments were considered together. Among the participants with developmental dyslexia, there was evidence of a phonological-deficit cluster, a rapid-naming cluster, and a cluster showing both depressed phonological processing and rapid naming. These accounted for 73 of the 93 participants (78%). All three are predicted by the double-deficit hypothesis. The fourth cluster consisted of children with normal phonological and rapid naming ability incommensurate with their high verbal ability. An analysis of variance with post-hoc multiple comparisons demonstrated that the phonological, rapid-naming, and double deficit clusters did not differ significantly in age, but the fourth cluster was comprised of significantly older children. The mixed data set revealed two clusters. One cluster consisted almost entirely of the double-deficit and phonological subtypes. The other consisted of the participants without dyslexia and the children with dyslexia demonstrating either a single rapid naming deficit or standardized test scores in the normal range. A silhouette analysis indicated that the four-cluster solution for the children with developmental dyslexia was superior to the two-cluster solution obtained for the entire data set. The study provides support for the presence of distinct subtypes in children with developmental dyslexia and for the double-deficit hypothesis. Specifically, this study finds three subtypes predicted by the double-deficit hypothesis without the assumption of an a priori theoretical model of reading. Taken together, these subtypes account for 79% of the participants with dyslexia. Further, the percentages of children in each subtype are in good agreement with previous studies. The participants in the subtype not predicted by the double-deficit hypothesis were significantly older than the other three groups. Recent advances in unsupervised learning can be expected to aid the improvement and refinement of the definition of developmental dyslexia. If reliable and consistent subtypes can be identified among persons with developmental dyslexia, it is reasonable to assume that diagnostic and intervention efforts will be greatly improved.

Testing the double-deficit hypothesis in an adult sample

The double-deficit hypothesis of dyslexia posits that reading deficits are more severe in individuals with weaknesses in phonological awareness and rapid naming than in individuals with deficits in only one of these reading composite skills. In this study, the hypothesis was tested in an adult sample as a model of reading achievement. Participants were parents of children referred for evaluation of reading difficulties. Approximately half of all participants reported difficulty learning to read in childhood and a small subset demonstrated ongoing weaknesses in reading. Structural equation modeling results suggest that the double-deficit hypothesis is an accurate model for understanding adult reading achievement. Better reading achievement was associated with better phonological awareness and faster rapid automatized naming in adults. Posthoc analyses indicated that individuals with double deficits had significantly lower reading achievement than individuals with single deficits or no deficits.

The Double-Deficit Hypothesis

The double-deficit hypothesis of developmental dyslexia proposes that deficits in phonological processing and naming speed represent independent sources of dysfunction in dyslexia. The present article is a review of the evidence for the double-deficit hypothesis, including a discussion of recent findings related to the hypothesis. Studies in this area have been characterized by variability in methodology--how dyslexia is defined and identified, and how dyslexia subtypes are classified. Such variability sets limitations on the extent to which conclusions may be drawn with respect to the double-deficit hypothesis. Furthermore, the literature is complicated by the persistent finding that measures of phonological processing and naming speed are significantly correlated, resulting in a statistical artifact that makes it difficult to disentangle the influence of naming speed from that of phonological processing. Longitudinal and intervention studies of the double-deficit hypothesis are needed to accumulate evidence that investigates a naming speed deficit that is independent of a phonological deficit for readers with dyslexia. The existing evidence does not support a persistent core deficit in naming speed for readers with dyslexia.

Processing Abilities Associated with Math Skills in Adult Learning Disability

This study evaluated college adults (N = 138) referred for learning problems using a Cattell-Horn-Carroll based intelligence measure (Woodcock Johnson-Revised: WJ-R) and spatial and executive function neuropsychological measures to determine processing abilities underlying math skills. Auditory and visual perceptual (WJ-R Ga and Gv), long- and short-memory (WJ-R Glr and Gsm), crystallized and fluid intellectual (WJ-R Gc and Gf), and spatial and executive function (Judgment of Line Orientation [JLO] and Category Test) measures differentiated those with and without math deficits. Multiple regression revealed selective processing abilities (Gf, JLO, and Category) predicting about 16% of the variance in math skills after variance associated with general intelligence (also about 16%) was removed. Cluster analysis found evidence for a selective spatial deficit group, a selective executive function deficit group and a double deficit (spatial and executive function) group. Results were discussed in relation to a double deficit hypothesis associated with developmental dyscalculia.

Instability of Double-Deficit Subtypes Among At-Risk First Grade Readers

The double-deficit hypothesis provides a framework for identifying students at-risk for persistent reading difficulties. I examined the temporal stability of four double-deficit subtypes (no-deficit, naming-speed deficit, phonological deficit, and double-deficit) in 197 low-performing, first-grade readers. Concurrent analyses in fall and spring revealed that students with poor phonemic awareness and naming speed were the lowest readers in the sample; however, having a double-deficit in fall was not a good predictor of subsequent progress, most likely because subtypes were not stable over the first half of grade 1. Only about half the sample retained the same subtype from fall to spring.

Evaluation of the Double-Deficit Hypothesis in College Students Referred for Learning Difficulties

The present study explored the double-deficit hypothesis (DDH) in a sample of 146 college students with and without reading disabilities (RD). The results indicated that although both phonological awareness (PA) and visual naming speed (VNS) contributed to performance on measures of decoding and comprehension, their relative contribution was influenced both by the nature of the stimulus (word vs. nonword vs. text) and by the conditions of the task (timed vs. untimed). Similar results were obtained using an individual differences approach, or when between-group comparisons were made of individuals with deficits in PA or VNS. The relative representation of DDH subgroups in groups of adults with RD varied based on the classification criteria used to define RD. These results support the DDH, extend its applicability to adults, and have implications for diagnostic decision making.

Neurodevelopmental characteristics of children with learning impairments classified according to the double-deficit hypothesis.

The double-deficit model has been examined primarily in relation to reading. We investigated whether children classified according to the double-deficit model would exhibit differences in other neuropsychological domains. Children referred for learning problems (N = 188), ages 7 to 11, were classified by double-deficit subtype. Only three of the four groups predicted by the model could be identified. There were no group differences in IQ or attention problems. The three groups showed different neuropsychological profiles, involving functional domains other than reading and language. Differences also emerged in nonverbal low-level information processing. The double-deficit group was generally most severely affected. The double-deficit groupings identify children with different neuropsychological profiles and variation in the efficiency of basic online information processing, extending beyond the oral and written language domain.

Causal models of reading disability: a twin study.

The genetic and environmental relationships among measures of phoneme awareness, naming speed, Intelligence Quotient (IQ), and reading performance were investigated in 623 identical and fraternal twin pairs tested in the Colorado Learning Disabilities Research Center. A Cholesky decomposition analysis of these measures provided evidence supporting the double deficit hypothesis that difficulties in phonological processing and naming speed both contribute to reading disability. Additionally, the model revealed marginally significant genetic and significant non-shared environmental relationships between IQ and reading independent of naming speed and phoneme awareness. Thus a more complete causal model of reading disability should include IQ as well as measures of phonological processing and naming speed.

Rapid Naming and Phonemic Awareness: Contributions to Reading, Spelling, and Orthographic Knowledge

The double-deficit hypothesis suggests that deficient skills on two dimensions, rapid naming and phonemic awareness, are associated with poor reading. We studied the reading, spelling, and orthographic skills of Grade 3 children who met our criteria for double asset (DA), single phonological deficit (PD), single naming speed deficit (NSD), and double deficit (DD) groups. Analyses of variance revealed main effects of each factor, oftentimes modified by significant interactions, on the varied achievement measures. All deficit groups performed below the DA children. Compared to children with NSD, children with PD (a) were less accurate decoders, (b) were faster readers, (c) had weaker spelling dictation, and (d) had comparably poor spelling recognition scores. Children with DD showed a mixed pattern of stronger, equal, or weaker skills compared to younger reading level controls.

Understanding the Relations Between RAN Letter Subtest Components and Word Reading in First-Grade Students

First-grade students (N = 221) were individually tested on a battery of cognitive and achievement measures of verbal fluency, visual attention, phonological awareness, orthographic recognition, rapid automatized naming (RAN) of letters and objects, and reading. All tests were subjected to postacquisition scoring, and all RAN measures were segregated into measures of articulation time, pause time, and consistency of the pause time. Structural equation modeling demonstrated that word reading was directly and significantly predicted by RAN letter naming and general RAN cognitive processing time of objects. Moreover, RAN letter reading constructs were significantly and directly predicted by the latent variables of phonological awareness, orthographic recognition, and general RAN object articulation and cognitive processing times. RAN letter naming constructs were also significantly and indirectly predicted by visual attention. The reading model was found to be consistent with a total mediation of the relation of phonological awareness and reading through RAN letter naming and supported the validity of the RAN letter naming subtest as a basic letter reading test. These findings supported the double-deficit hypothesis for letter reading. We suggest that phonological memory is a basic factor underlying general RAN cognitive processing time of objects and domain-specific information associated with phonemes and their graphic representations.

Using Semantics, Grammar, Phonology, and Rapid Naming Tasks to Predict Word Identification

This study investigated the relative importance of semantic, grammatical, phonological, and rapid naming abilities in predicting word identification in a large, representative sample of children enrolled in first through sixth grade, using correlation, factor analysis, multiple regression, and predictive outcome analysis techniques. Composite measures of these abilities were found to correlate significantly with word identification, even after controlling for the effects of nonverbal intelligence. Factor analysis indicated that the spoken language composites and the word identification composite loaded on one factor, whereas the perceptual speed composites loaded on a second factor. Multiple regression analyses showed that among younger children in the early stages of learning to read and children whose word identifying skills were below average, the phonology and rapid naming composites accounted for the most variance in predicting word identification skills. Among older children and children who were proficient in word identification, the semantics composite accounted for the most variance. The most important analyses in this study (i.e., the calculations of the sensitivity indexes, the specificity indexes, and the positive predictive values) evaluated the practical value of using the composites to predict poor word identification skills in children. To be considered practically useful, all predictive outcome values had to be .75 or greater. None of the composites studied, including an application of the double-deficit hypothesis, met this criterion. The results from this study question the accuracy and utility of using any of the abilities studied to predict which students are at risk for or have poor word identification skills.

"I like to take my own sweet time"

Three types of dyslexic readers exist, according to the Double-Deficit Hypothesis: those with (a) a single phonological deficit, (b) a single naming-speed deficit, or (c) a combination of deficits. Although there is considerable information about phonologically based reading disabilities, there is insufficient information about poor readers who have intact phonological skills but severe naming-speed problems. This case is the first in-depth study of a reading disabled student with a profound single naming-speed deficit. Through an analysis of the student's performance on a variety of educational and neuropsychological tasks, we explore how the single naming-speed pattern of disruption affects reading performance, and whether this performance is consistent with theoretical assumptions regarding the role of naming speed in reading. We also describe a theoretically based reading program (RAVE-O) designed to address deficits in processing speed and fluency. Through pre/posttest analysis, we discuss whether the reading performance of this reader changed after participation in the program.

Are RAN- and phonological awareness-deficits additive in children with reading disabilities?

The double-deficit hypothesis (Wolf, M. and Bowers, P.G. (1999) The double-deficit hypothesis for the developmental dyslexias. Journal of Educational Psychology, 91, 415-438) proposes that deficits in phonological processing and rapid automatized naming (RAN) are separable sources of reading dysfunction. Further, the double-deficit hypothesis predicts that the presence of deficits in both phonological processing and RAN have an additive negative influence on reading performance above and beyond that of a single deficit. The purpose of this study was to examine the additive nature of phonological awareness (PA)- and RAN-deficits on written language skill in children with reading disabilities (RD). Concurrent relationships between PA, RAN, and written language skills were examined in 476 children with RD, ranging in age from 8 to 18 years of age. Hierarchical regression analysis revealed that PA and RAN skill have an additive effect on a majority of the reading and spelling measures. When participants were classified into three deficit subtypes based on the double-deficit model (i.e. phonological-, rate-, and double-deficit), comparisons across the subtypes confirmed that individuals with double-deficits performed below the single-deficit groups on both subtyping variables (RAN and PA) and all measures of written language. When the double- and single-deficit groups were matched on the subtyping variable (i.e. double- and rate-deficit groups matched on RAN and double- and phonological-deficit groups matched on PA) differences between the double- and rate-deficit groups remained in non-word reading, whereas differences between the double- and phonological-deficit groups remained in timed word recognition and reading comprehension. These results support an additive model in which RAN-deficits primarily affect tasks that require speeded/fluent response, and PA-deficits primarily affect tasks that emphasize phonological processing skill. Results are also presented that illustrate several statistical problems associated with the formation of deficit groups by dichotomizing the RAN and PA variables.

Comparing the phonological and double deficit hypotheses for developmental dyslexia

This study tested the predictions of thephonological and double deficit hypotheses byexperimentally examining speech perception,phoneme awareness, lexical retrieval (serialand discrete), articulatory speed, and verbalSTM in school age child (N = 35) and adolescent(N = 36) dyslexics, and both chronological age(CA) and reading age (RA) controls. Theresults confirmed the findings of previousstudies of a deficit in phoneme awareness indevelopmental dyslexia. At both age levels,dyslexics performed significantly more poorlythan both their CA and RA controls. Althoughdeficits in the other processes investigated,particularly in rapid serial naming, were alsoapparent, they were not as clear-cut as thedeficit in phoneme awareness. In general,definite evidence of a deficit in rapid serialnaming was limited to the more severelyimpaired dyslexics. Furthermore, although rapidserial naming contributed independent variationto various literacy skills, its contributionwas modest relative to the contribution ofphoneme awareness, regardless of whether theliteracy skill relied more or less heavily onphonological or orthographic coding skills. Further analyses suggested that variation inrapid serial skill is particularly importantfor fluent reading of text, whereas phonemeawareness is particularly important for thedevelopment of the ability to read byphonologically recoding letters or groups ofletters in words into their phonological codes.This explains the relatively strongcontribution of phoneme awareness to readingand spelling ability in general. In sum, thephonological hypothesis offers a moreparsimonious account of the present resultsthan the double deficit hypothesis.

Naming-Speed Processes, Timing, and Reading

This article integrates the findings in the special issue with a comprehensive review of the evidence for seven central questions about the role of naming-speed deficits in developmental reading disabilities. Cross-sectional, longitudinal, and cross-linguistic research on naming-speed processes, timing processes, and reading is presented. An evolving model of visual naming illustrates areas of difference and areas of overlap between naming speed and phonology in their underlying requirements. Work in the cognitive neurosciences is used to explore two nonexclusive hypotheses about the putative links between naming speed and reading processes and about the sources of disruption that may cause subtypes of reading disabilities predicted by the double-deficit hypothesis. Finally, the implications of the work in this special issue for diagnosis and intervention are elaborated.

Remediating the Core Deficits of Developmental Reading Disability

The double-deficit hypothesis (Wolf, 1997; Wolf & Bowers, 1999, this issue) contends that deficits in phonological awareness and deficits in visual naming speed represent two independent causal impediments to reading acquisition for children with developmental reading disabilities (RD). One hundred and sixty-six children with severe RD from 7 to 13 years of age were classified into three deficit subgroups according to a double-deficit framework. A total of 140 children with RD, 84% of the sample, were classified; 54% demonstrated a double deficit (DD), 22% a phonological deficit only (PHON), and 24% a visual-naming speed deficit only (VNS). Diagnostic test profiles highlighted the joint contributions of the two core deficits in depressing written language acquisition. The children in the DD group were more globally impaired than those in the other subgroups, and the VNS group children were the highest achieving and most selectively impaired readers. Following 35 hours of word identification training, sizable gains and significant generalization of training effects were achieved by all subgroups. A metacognitive phonics program resulted in greater generalized effects across the domain of real English words, and a phonological training program produced superior outcomes within the phonological processing domain. The greatest non-word reading gains were achieved by children with only phonological deficits.

Retrieval, Automaticity, Vocabulary Elaboration, Orthography (RAVE-O)

The most important implication of the double-deficit hypothesis (Wolf & Bowers, in this issue) concerns a new emphasis on fluency and automaticity in intervention for children with developmental reading disabilities. The RAVE-O (Retrieval, Automaticity, Vocabulary Elaboration, Orthography) program is an experimental, fluency-based approach to reading intervention that is designed to accompany a phonological analysis program. In an effort to address multiple possible sources of dysfluency in readers with disabilities, the program involves comprehensive emphases both on fluency in word attack, word identification, and comprehension and on automaticity in underlying componential processes (e.g., phonological, orthographic, semantic, and lexical retrieval skills). The goals, theoretical principles, and applied activities of the RAVE-O curriculum are described with particular stress on facilitating the development of rapid orthographic pattern recognition and on changing children's attitudes toward language.

Naming Speed, Phonological Awareness, and Orthographic Knowledge in Second Graders

Concurrent relationships among measures of naming speed, phonological awareness, orthographic skill, and other reading subskills were explored in a representative sample of second graders. Hierarchical regression analyses revealed that naming speed, as measured by the the rapid automatized naming (RAN) task, accounted for a sizable amount of unique variance in reading with vocabulary and phonemic awareness partialled out. The unique contribution of naming speed to reading was relatively stronger for orthographic skills, whereas the contribution of phonemic skills was stronger for nonword decoding. In further analyses, marked difficulties on a range of reading tasks, including orthographic processing, were seen in a subgroup with a double deficit (slow naming speed and low phonemic awareness) but not in groups with only a single deficit. These findings are broadly consistent with Bowers and Wolf's (1993a, 1993b; Wolf & Bowers, 1999) double-deficit hypothesis of reading disability.

The double-deficit hypothesis and difficulties in learning to read a regular orthography.

In 2 large longitudinal studies, we selected 3 subgroups of German-speaking children (phonological awareness deficit, naming-speed deficit, double deficit) at the beginning of school and assessed reading and spelling performance about 3 years later. Quite different from findings with English-speaking children, phonological awareness deficits did not affect phonological coding in word recognition but did affect orthographic spelling and foreign-word reading. Naming-speed deficits did affect reading fluency, orthographic spelling, and foreign-word reading. Apparently, in the context of a regular orthography and a synthetic phonics teaching approach, early phases of literacy acquisition (particularly the acquisition of phonological coding) are less affected by early phonological awareness deficits than are later phases that depend on the build up of orthographic memory. The double-deficit hypothesis was developed by Maryanne Wolf and Patricia Bowers as an extension of the dominant phonological-deficit explanation of developmental dyslexia (e.g., Bowers & Wolf, 1993; Wolf & Bowers, 1999). The phonologicaldeficit hypothesis postulates an early difficulty in acquiring phonological awareness, which interferes with the acquisition of grapheme-phoneme coding as a word recognition mechanism, which in turn results in reduced self-teaching of orthographic word representations (Share, 1995). The early problem with phonological awareness is seen as resulting from less sharp phoneme boundaries in speech perception (Fowler, 1991) or from less distinct phonological word representations (Elbro & Peterson, 1998). The double-deficit hypothesis acknowledges the phonological awareness deficit of dyslexic children but stresses a deficit in naming speed as a second and equally important cause of reading difficulties. The first demonstration that dyslexic children show impaired naming speed was provided by Denckla and Rudel (1976), whose Rapid Automatized Naming (RAN) Test became the standard assessment of naming speed. In the RAN test, a large array of repeatedly presented, well-known visual patterns (pictured objects, color patches, digits) has to be named as quickly as possible. Wolf, Bally, and Morris (1986) were the In:st to show that early differences in rapid naming are predictive of later reading difficulties. The double-deficit hypothesis is based on findings (reviewed by Wolf & Bowers, 1999) showing that, typically, there are only modest correlations between phonological awareness measures and RAN performance in groups of dyslexic children and that dyslexic children as a group exhibit both phonological awareness deficits and naming-speed deficits. From these findings,