Dyslexia Screening Test Review of the Dyslexia Screening Test by KAREN T. CAREY, Professor of Psychology, California State University-Fresno, Fresno, CA: DESCRIPTION. The Dyslexia Screening Test (DST) is designed as a screening instrument for use with children from 6 years 6 months to 16 years 5 months of age. Based on the World Federation of Neurology definition of 1968, dyslexia is "a disorder in children who despite conventional classroom experience, fail to attain the language skills of reading, writing, and spelling commensurate with their intellectual abilities" (manual, p. xi). The measure consists of 11 subtests including three Tests of Attainment (One Minute Reading, Two Minute Spelling, and One Minute Writing); eight Diagnostic Tests (Rapid Naming, Bead Threading, Postural Stability, Phonemic Segmentation, Backwards Digit Span, Nonsense Passage Reading, Semantic Fluency, and Verbal Fluency); and an At-Risk Quotient. Tests of Attainment include reading, writing, and spelling based on the definition from the World Federation of Neurology. The Diagnostic subtests are tests that the authors believe, based on their literature review, give "positive indicators of dyslexia" (manual, p. 3). The kit comes with an audiocassette tape for administering the Backwards Digit Span subtest and scoring keys. The test can be used by teachers and other school personnel and was normed in England on 1,000 children. It takes approximately 30 minutes to administer. In addition, the authors provide a list and brief description of programs available to assist children with dyslexia. This includes specific remedial academic program packages, books, information on associations, and journal references on dyslexia. DEVELOPMENT. The authors' goal in developing the test was to "construct a quick, simple enjoyable test that could be used by school professionals and provided a simple 'at risk' index for dyslexia" (manual, p. 3). The United Kingdom Education Act of 1993 led to substantial changes in the country for meeting the needs of special education students and the tool was developed to be used as an objective measure to identify special needs children and assist in the development of intervention plans for such children. According to the authors, the tool was developed after several years of research and testing, although the specifics are not discussed in the manual. The test format was selected and then norms were established for each of the subtests so that a child's performance on the test could be compared to what would be expected of a child of the same age. Each of the 11 subtests was developed in different ways based on a review of the literature. For example, the Phonemic Segmentation test was developed from the work of Rosner and Simon (1971) and Nonsense Passage Reading was selected and modified from work by Finucci, Guthrie, Childs, Abbey, and Childs (1976). One subtest, Postural Stability, was developed from work by Nicolson, Fawcett, and Dean (1995). The authors of the DST state that "dyslexic children show difficulties consistent with slight abnormalities in the cerebellum.... This test is based on clinical procedures for establishing cerebellar abnormalities" (manual, p. 46). A "balance tester" is included in the materials, which is a plastic device with a collar that slides back and forth. The examiner stands behind the child and holds the balance tester on the two vertebrae above the small of the child's back. This procedure seems quite dangerous as the examiner could potentially hurt the child. Individual subtests were developed in this manner and pilot studies were conducted to ensure that children "enjoyed doing it" (manual, p. 9) and teachers were able to use the tool reliably. Several preliminary subtests were eliminated and others were modified due to poor reliabilities. The tests are the same for all age groups with four exceptions. The Nonsense Passage Reading, One Minute Writing Tests, the One Minute Reading, and the Two Minute Spelling Tests were modified for age differences, making the tests appropriate for children of different ages. TECHNICAL. Children from selected schools in Sheffield, London, and Wales participated in the norming sample with at least 100 children at ages 6 years 6 months to 7 years 5 months, 7 years 6 months to 8 years 5 months, and then every 2 years (e.g., 8 years 6 months to 10 years 5 months). Norms were derived for each subtest for each age resulting in every score assigned a percentile point on average performance. The scores for each age group were ranked resulting in percentile ranks for each test. The actual percentile ranks used for scoring were developed through interpolation and smoothing for each age group. The percentile scores are collapsed into five categories: very highly at risk, which the authors refer to as triple minus (---) (percentile ranks 1-4); highly at risk, double minus (--) (percentile ranks 5-11); at risk, single minus (-) (percentile ranks 12-22); normal performance, no minus (0) (percentile ranks from 23-77); and well above average performance, a + (percentile ranks 78100). To obtain the At-Risk Quotient (ARQ), the mean of the at-risk scores (triple minus, double minus, single minus, and 0) was weighted; those individuals scoring not at-risk were not included. An ARQ of 1.0 or greater is considered "at-risk." Construct validity was investigated through the administration of the DST to 17 children who had been diagnosed as dyslexic on other scales. Fifteen of the students ranging in age from 10 to 15 years had an ARQ of 1 or higher. Unfortunately, the sample for these studies is very small and must be interpreted with caution. Test-retest reliability was estimated from 34 children, ages 6 years 6 months to 12 years, who were retested about 1 week apart. No information related to the ages of these children or whether they were at-risk is included in the manual. The correlations range from .72 to .99 on the subtests. A test-retest correlation for two forms of the One Minute Reading Test was .959. Interrater agreement for the postural stability subtest, which requires some judgment on the part of the examiner, was .98. COMMENTARY. The DST needs to be further researched before it is widely used. The test has not been adequately evaluated for technical requirements including reliability and validity. In addition, the sample is limited to children in the United Kingdom, and as it has not been used in the United States or other countries, the norms must remain in question for use in countries other than the U.K. As the norms are limited and with the use of interpolation and smoothing of the norms to the degree done in this instrument, the DST does not appear appropriate for making individual decisions regarding students' with academic difficulties. Reliability and validity were not adequately assessed, resulting in questions regarding the consistency and accuracy of the tool. In general, the academic subtests including Rapid Naming, One Minute Reading, Phonemic Segmentation, Two Minute Spelling, Nonsense Passage Reading, One Minute Writing, and Verbal and Semantic Fluency selected for inclusion in the test seem appropriate for this type of assessment. However, although the authors attempt to make cases for including Bead Memory, Postural Stability, and Backwards Digit Span, there is not enough evidence for including such subtests on a scale assessing dyslexia. In addition, the Postural Stability has the potential for harming children and would not be recommended for use in many school districts in the United States. The scaling is difficult to understand with the minuses and plus and the development of these procedures. The resulting categories identified from the minuses and plus of highly at risk to well above average are not defined and the specifics of how the categories discriminate from one another are not explained. In addition, the authors do not address the potential problem for errors of measurement in their categorizations. In the final section of the manual, prior to the descriptions of the subtests, the authors do provide some information for users related to developing interventions for children identified as dyslexic. However, the material provided is simply a list of possible packages and readings for the examiner. Specific interventions for each subtest would be far more beneficial. SUMMARY. The DST needs further evidence of technical adequacy before it can be used on a large scale basis. No information other than the ages of the norming sample are provided so it is unknown whether or not the test would be appropriate for boys and girls, children of different ethnic backgrounds, and children in different social economic classes. The reliability and validity evidence for the test is insufficient for individual decision making and although the authors state the test can be used to help develop interventions, it is unclear how this can be done. Overall, it appears more useful information could be obtained from conducting a thorough observation in the student's classroom. REVIEWER'S REFERENCES 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, 1-23. Nicolson, R. I., Fawcett, A. J., & Dean, P. (1995). Time estimation deficits in developmental dyslexia: Evidence for cerebellar involvement. Proceedings of the Royal Society: Biological Sciences, 259, 43-47. Rosner, J., & Simon, D. P. (1971). The auditory analysis test: An initial report. Journal of Learning Disabilities, 4, 384-392. Review of the Dyslexia Screening Test by JAMES E. YSSELDYKE, Associate Dean for Research, College of Education and Human Development, University of Minnesota-Twin Cities, Minneapolis, MN: DESCRIPTION. The Dyslexia Screening Test (DST) is an individually administered, normreferenced test designed to provide an "at risk" dyslexia index for children ranging in age from 6-6 to 16-5. The DST was created to be a dyslexia screening instrument that could be administered by school professionals other than educational or clinical psychologists. The authors explain that the test can also be used (a) to derive a profile of a child's strengths and weaknesses in order to guide in-school supports, (b) as a basis from which to request formal assessment, and (c) to provide educational psychologists with an instrument that provides a valid index of dyslexia, thereby establishing a foundation for further assessment. The DST test battery takes about 30 minutes to administer and includes 11 subtests: 3 Attainment tests and 8 Diagnostic tests. The purpose of the Attainment tests is to assess those areas in which dyslexic children typically demonstrate difficulties: Reading, Writing, and Spelling. The Diagnostic tests are designed to assess skills that the authors believe are affected by dyslexia, thereby providing a profile of a child's difficulties that is intended to inform (a) the causes of attainment difficulties, and (b) the skills that need to be targeted for intervention. Raw scores for each individual subtest are converted into age-appropriate "at-risk index" scores, which are derived from the norm sample. Five "at-risk index" scores are possible for each subtest: triple minus (--- ; bottom 4%), double minus (-- ; bottom 5-11%), minus (- ; bottom 1222%), zero (0 ; midrange, 23-77%), and plus (+ ; top 22%). The "at-risk quotient" (ARQ) is found by adding the individual subtest indices (triple minus = 3; double minus = 2; minus = 1; zero and plus = 0) and dividing that sum by 10. Although there are 11 subtests, Semantic Fluency is not included in this division because this is a supposed strength of dyslexic children. If the ARQ is 1 or greater, a child is considered "at-risk" for dyslexia. DEVELOPMENT. Although the authors claim that test development "entailed a lengthy series of studies" (manual, p. 9), the manual provides little detail of the item development process. After reviewing the dyslexia literature, the authors originally developed 14 possible tests to be included in the DST. Through a series of studies, the authors established norms for performance on each test while also examining reliability, consistency of implementation, and student enjoyment of tasks. Following these studies, three tests were removed due to problems with reliability and/or test equipment, and other tests were modified to improve reliability and ease of use. Attainment tests. The DST was developed under the assumption that dyslexic children typically demonstrate difficulties in reading, writing, and spelling, so the three attainment tests included in the DST specifically assess these areas. A description and the rationale or origin of each attainment test is included below. One Minute Reading-This test measures how many words a child can read correctly from a onepage list in 60 seconds. The authors felt that a measure that assessed both accuracy and speed would more accurately estimate a child's reading skills than a measure that assessed only accuracy. It is based on the Dutch EMT test (Brus & Voeten, 1980, as cited in DST manual, p. 31). Two Minute Spelling-During this test of spelling fluency, a child has 2 minutes to spell each word that the tester reads. (The tester reads the next word as soon as the child has completed the previous word.) One Minute Writing-During this test, a child must copy text as quickly and accurately as possible in the 1-minute time limit. The authors describe this task as an indication of "pure" writing speed. One reason this test is included is to provide documented evidence for "slowness of writing" (manual, p. 32) so that children can apply to examination boards for extra time in taking tests. Diagnostic tests. Because reading, writing, and spelling are all learned skills that can be at least partially alleviated through extensive instruction, the authors additionally included diagnostic tests to assess other difficulties typically associated with dyslexia (i.e., phonological skill, fluency, working memory, and balance). A description and the rationale or origin of each diagnostic test is included below. Rapid Naming-This test measures the amount of time it takes a child to name a series of familiar outline drawings (e.g., hat, ball). This test is based on the "Rapid Automatised Naming" test introduced by Martha Denckla in the 1970s. In explaining the inclusion of this activity, the authors claim there is evidence that dyslexic children are typically slower than normal when asked to name a series of pictures. Bead Threading-This test measures how many beads a child can thread in 30 seconds. The authors explain that this activity is included because evidence suggests dyslexic children have difficulties in fine motor skill activities that involve hand-eye coordination. Postural Stability-This test measures how well a child can balance following a slight push in the back. It is a clinical test of cerebellar abnormality and is included presumably because the authors have found evidence for cerebellar abnormality in dyslexia. Moreover, the authors claim this task successfully distinguishes between dyslexic children and nondyslexic poor readers (i.e., Unlike nondyslexic poor readers, dyslexic children face difficulties in balance). Phonemic Segmentation-This test measures a child's ability to break down a word into its component sounds and then manipulate those sounds (e.g., say "grandma" without the "ma"). It is included because phonological difficulties are one of the most clearly established difficulties in dyslexia. The test is based on earlier measures developed by Jerome Rosner. Backwards Digit Span-This test measures a child's ability to repeat a sequence of numbers in the opposite order in which they were verbally presented. It is a common measure of working memory found on many IQ tests. Dyslexic children typically perform poorly on this task. Nonsense Passage Reading-This test assesses a child's ability to read a passage that is interspersed with nonsense words. This test is included based on evidence that dyslexic children continue to have difficulty with unfamiliar words even after they improve their scores on standard tests of reading. Verbal Fluency and Semantic Fluency-In the Verbal Fluency test, children have 1 minute to name as many words as they can that begin with a specific letter (e.g., G-good, grape, golf, etc.). In the Semantic Fluency test, children have 1 minute to name as many animals as they can. These tasks are included based on work by Frith and colleagues that suggests dyslexic children do well on tasks of semantic fluency but do poorly on tasks of verbal fluency. In order to justify the inclusion of these diagnostic tests, the authors reference a study they conducted in which they distinguished dyslexic and nondyslexic children on the basis of an index that combined "indications of difficulties in phonological skills, balance and speed of processing" (manual, p. 4). Few details of this study are presented, making it difficult to examine the validity of the claim. TECHNICAL. Information about the norm sample is vague. The authors explain that through a series of studies, "norms were derived for each test for each age" (manual, p. 9). The details included in the manual indicate that the DST was standardized on whole classes of children who attended school in Sheffield, London, and Wales. For the first 2 years (i.e., 6-6 to 7-5, and 7-6 to 8-5), at least 100 children represent each age group. At later ages, this number is spread over 2year increments. No information is presented about procedures for ensuring the sample was representative of the population of children in the U.K., and no data are presented about how well the norm sample matches this population. The authors present evidence for test-retest, interform, and interrater reliability, and argue that measures of internal consistency (another commonly reported reliability) are inappropriate for this test. For test-retest reliability, the authors report on a study in which 34 children aged 6-6 to 12 years were administered the test on two separate occasions (there was about 1 week between administrations). The majority of the test-retest reliabilities for each subtest exceed .80, with three subtests having reliabilities between .70 and .80 (i.e., Bead Threading, Postural Stability, and Semantic Fluency). The interform reliability for the One Minute Reading Test was found to be .96 in a study with 22 children. Postural stability was the only subtest for which interrater reliability was measured. It was measured by showing videotapes of 14 children participating in this test, and having three experimenters independently rate the task. Reliabilities ranged from .94-.98. Evidence for validity is limited. The authors highlight the face validity of the DST by reiterating that reading, writing, and spelling are three areas in which dyslexic children typically struggle, and a poor performance in at least one of those areas is a prerequisite for a diagnosis of dyslexia. Moreover, they claim that evidence has shown that dyslexic children face difficulties with the remaining tests (except Semantic Fluency) as well. However, the reader is expected to take them at their word as no detailed evidence of these links is provided in the manual. The authors do reference a limited number of articles that might provide evidence regarding the inclusion of certain subtests. The only evidence for construct validity that is presented comes from a study in which DST was administered to 17 children ranging in age from 10-6 to 15-7 who had previously been diagnosed with dyslexia, and 20 children who showed no sign of dyslexia. All but 2 of the children diagnosed with dyslexia were identified as at-risk by their scores on the DST, whereas none of the children who showed no signs of dyslexia were identified as at-risk on the DST. Yet, no details are provided about the measures used initially to identify children as dyslexic. Moreover, no children below the age of 10-6 were included in the sample even though the test is supposedly valid for children as young as 6-6. COMMENTARY. This is a strange mix of subtests that are to give the educator information that is predictive and diagnostic. The authors do not provide evidence of the validity of the test for the purposes they identify (see introductory paragraph). Those who use this test in a normative manner are comparing students to an unspecified population in the U.K. SUMMARY. The DST is intended as a screener for dyslexia, a nebulous condition. It consists of a mixture of subtests that assess reading, writing, and spelling attainment, and measures of areas thought to be affected by dyslexia. The norm sample is vague, scores are reliable, yet evidence for validity is very limited. Dyslexia Dyslexia What you need to know Dr Handler is a pediatric ophthalmologist in private practice in Encino, California. She is one of the authors of the American Academy of Pediatrics, American Academy of Ophthalmology, American Association for Pediatric Ophthalmology and Strabismus, and American Association of Certified Orthoptists 2011 Joint Technical Report, Learning Disabilities, Dyslexia, and Vision. She has nothing to disclose in regard to affiliations w ith or financial interests in any organizations that may have an interest in any part of this article. 18 By being vigilant to signs of dyslexia, dispelling the myths, and coordinating care, pediatricians can help children w ith dyslexia enjoy success in school and in daily life. SHERYL M HANDLER, MD Learning to read is an extremely complex pro­ cess, which has been described to be as chal­ lenging as learning rocket science.1Therefore, it is not too surprising that, for many reasons, over 60% of children in America fail to meet standards for reading proficiency.2 M u ltip le issues m ay u n d e rlie th is reading difficulty, including poor early language development, inadequate instruc­ tion, insufficient reading practice, lack of background knowledge, and intellectual disability. In some children, however, the problem is the specific learning disability called dyslexia. Dyslexia is by far the m ost com m on learning disability and is present in some degree in up to 20% of c h ild ren .3 Just as early detection and intervention are crucial in m edical diseases, the same is tru e in learning disabilities. The conse­ quences of untreated dyslexia are broad and can be significant, including effects on academic success and psychosocial well­ being. Children with dyslexia experience intense frustration; may act aggressively CONTEMPORARYPEDIATRICS.COM | A UGUS T 2016 or withdraw; frequently become targets of bullying and ridicule; have low self-esteem; and may even develop mental health prob­ lems, including anxiety and depression. Pediatricians have the opportunity and responsibility to enable detection and proper treatment of dyslexia in children. This article aims to provide information and strategies that will allow clinicians to best assist and advise patients and their parents. Dyslexia defined Dyslexia is a language-based learning dis­ ability characterized by difficulties with d e c o d in g (so u n d in g out) w o rd s, flu ­ ent w ord reco g n itio n , a n d /o r readingcomprehension skills. Children with dys­ lexia often develop secondary problem s with comprehension, spelling, writing, and knowledge acquisition. The difficulties found in dyslexia are usually caused by a phonological deficit (an auditory processing problem involving hearing the sounds in speech). The pho­ nological deficit leads to difficulty con­ necting speech sounds to letters, which is a skill needed to decode the written word. peer-reviewed 1 tends to be identified earlier and more often in boys, perhaps because boys tend to “act out” when they are unable to do a difficult task versus girls who are inclined to make them­ selves “invisible” in the classroom. RISK FACTORS AND SIGNS OF DYSLEXIA RISK FACTORS o Family history POSSIBLE SIGNS o Prematurity or birth problems ° Language or speech problems, including articulation o Fetal exposure to drugs or alcohol o Difficulty with rhymes o Hearing, language, or speech problems, including articulation ° Difficulty learning the names of letters o Developmental delay o Trouble connecting letters to their sounds o ADHD or other neurological problems O Difficulty sounding out words o Other chronic health problems that may have caused school absences o Difficulty with sight word recognition o Slow reading o Poor spelling Abbreviation: ADHD, attention-deficit/hyperactivity disorder. Alternatively, dyslexia in some chil­ dren results from problems with oral language skills, sight word recogni­ tion, processing speed, comprehen­ sion, attention, or verbal working memory. Anatomical and imaging stud­ ies investigating brain development and function show a corresponding physical basis for dyslexia in language-related areas of the brain. The brains of persons with dys­ lexia function differ­ ently than the brains of “typical readers” before they even start to read, as dyslexics use an alternative pathway for reading. Specifically, these investigations reveal that persons with dyslexia have dysfunction in the left-hemisphere posterior reading areas with corresponding compensa­ tory use of the bilateral inferior fron­ tal gyri of both hemispheres and the right occipitotemporal area.3 In discussing the definition of dyslexia and its causes, it is also use­ ful for pediatricians to be aware of the many myths and mispercep­ tions that exist. Dyslexia is not a condition where readers see letters or words upside down or back­ ward. It is not related to visual or eye-tracking problems.4 In addition, dyslexia is not a developmental issue that children may be expected to outgrow; rather, it is a persistent lifelong condition. Dyslexia also is not related to intelligence or laziness in a child. Dyslexia occurs in persons with low, nor­ mal, and high intel­ ligence quotients (IQs) alike. The fact that dys­ lexia is not related to IQ, however, creates the potential for a significant learning disability to be overlooked in an otherwise bright child. Dyslexics are often perceived to be “lazy” or “not working up to their potential” when, in fact, they often work harder and longer than their peers. In addition, there is no male pre­ dominance for dyslexia. It is found almost equally in boys and girls, but AUGUST 2016 I D e te c tio n a n d d ia g n o s is A diagnosis of dyslexia is established clinically based on history, observa­ tion, and a battery of age-appropriate educational tests interpreted by a knowledgeable, qualified profes­ sional. Although it is not up to pedi­ atricians to make the diagnosis, an understanding of dyslexia can help to identify children by being attentive to risk factors and signs that are elic­ ited in the medical and social history during well-child exams (Table 1). Dyslexia is heritable and familial, and so the history should ascertain whether there is a family history of speech and language problems or dyslexia. Other risk factors for dys­ lexia include prematurity, neurolog­ ical problems, and developmental or language delays. Early warning signs for dyslexia in preschool-aged children include trouble learning nursery rhymes or playing rhyming games; confus­ ing words that sound alike; mis­ pronouncing words; and trouble recognizing letters of the alphabet.5 Early elementary school children with dyslexia often have difficulty learning the names and sounds of the letters; separating and blending sounds; sounding out words; rec­ ognizing sight words; and spelling. They often read slowly and dislike reading. A parent’s complaint that a child is not doing well in school should prompt questions to explore the presence of reading difficulties. CONTEMPORARYPEDIATRICS.COM 19 peer-reviewed Although many dyslexic students are identified in the prim ary grades, dyslexia’s possible presence should not be overlooked in older children and teenagers. Signs in adolescents include a history of phonologically based reading difficulties, slow read­ ing, choppiness when reading aloud, poor com prehension, and requir­ ing more time to finish assignments or tests.5 These adolescent students also may have multiple school tru an ­ cies and/or behavioral issues, such as anger, aggression, depression, or even suicidal tendencies and possibly alcohol or drug use. If a family history of dyslexia or other risk factors exists, the child’s early language developm ent and school progress should be carefully monitored. A formal psychoeducational school evaluation is needed to identify dyslexia and will p ro ­ vide an understanding of the child’s s tre n g th s a n d w eak n esses, th e severity of the problem, and whether the child has a “specific learning disability” that is eligible for special education and support program s. Severe dyslexia ty p ic a lly q u a li­ fies a child for an Individualized Educational Plan, special education, and related services. The psychoeducational or broader neuropsycho­ logical or developmental-behavioral pediatric evaluation will identify comorbid conditions, and the find­ ings will provide the foundation for a treatm ent plan. Fortunately, individuals who have severe dyslexia are in the minority. The other side of the coin, however, is that those students whose disorder is more moderate or mild may not be readily recognized and/or may not qualify for treatment, although they would benefit from those services. 20 T STRATEGIES FOR TREAT NG DYSLEX Explicitly teach reading skills o Phonemic awareness o Intensive systematic phonics o Morphology— analysis of words and word parts o Memorization of sight words o Fluency training— guided oral reading o Vocabulary building o Comprehension techniques Spelling— including the rules Writing Treating dyslexia The prognosis for a child with dys­ lexia depends not only on the spe­ cific features of the disorder and its severity, but also on the timeliness and appropriateness of intervention (Table 2). Studies show that begin­ ning remediation in first grade versus waiting 2 more years greatly increases the chance that a child will later be able to read at grade level.6 Neverthe­ less, it is never too late to help. Effective intervention for dyslexia targets its etiology as a languagebased disorder and should be pro­ vided by a professional w ith the appropriate training. Children with dyslexia are best served w ith les­ sons provided in small group set­ tings, which bring together students who are at the same reading level and include no m ore th an 5 stu ­ dents. Remediation program s that follow the International Dyslexia A ssociation guidelines call for a “S tru ctu red Literacy A pproach.” CONTEMPORARYPEDIATRICS.COM I AUGUS T 2016 These program s explain language in an explicit, systematic, sequen­ tial, an d m u ltise n so ry m anner. They include training in the 5 read­ ing skills: 1) phonem ic awareness (h earin g th e so u n d s in words); 2) phonics (correlating sounds with letters); 3) fluency (ability to read w ith speed, accuracy and expres­ sion); 4) vocabulary; and 5) compre­ hension. Emphasis is also placed on learning word structure, letter pat­ terns, spelling, and writing. W hereas early m anagem ent of dyslexia focuses on remediation of the reading problem, for older stu­ dents there is a shift toward p ro ­ viding tools and accommodations. Accommodations allow the student to access his/her higher-level think­ ing and reaso n in g skills. These measures include access to assistive technology (eg, recorded books, text reading software, note takers, spell checkers) as well as provisions to enable test taking (eg, extended test­ ing time, a special quiet room, or preferential seating). Pediatrician's role Parents of children w ith dyslexia often seek counsel from their pedia­ trician about dyslexia treatments, so clinicians should be aware of both the proven and unproven interven­ tions being recommended and heav­ ily prom oted. By being inform ed about the v ariety o f approaches, as well as being able to discuss the lack of evidence supporting the effi­ cacy for th e unp ro v en in te rv e n ­ tions, pediatricians will be equipped to educate patients and their fam ­ ilies, encourage positive proven approaches, discourage the use of unproven techniques, and thus pre­ vent families from wasting valuable peer-reviewed time and resources pursuing expen­ sive, nonproductive therapies. Non-evidence-based approaches promoted for remediation of read­ ing difficulties in children with dyslexia include medication for ves­ tibular dysfunction, chiropractic manipulation, physical exercises, and dietary supplem entation or restrictions.5 In particular, however, the erroneous concept that dys­ lexia is a vision-based disorder has spawned a variety of interventions with nearsightedness may have diffi­ culty seeing the board. Symptomatic convergence insufficiency, which occurs more rarely in elementary school-aged children, makes reading at near blurry and uncomfortable. Consequently, affected children may read only for short periods of time, and on that basis may be thought to have dyslexia or attention-deficit/ hyperactivity disorder (ADHD), especially the inattentive type. Pediatric ophthalmologists play Pediatric ophthalmologists play a role in the assessment of children with suspected learning disabilities, including dyslexia. involving the use of training glasses, eye exercises, behavioral/perceptual vision therapy, and colored lenses and overlays. Although visual prob­ lems, including convergence insuf­ ficiency, can hamper reading, they are not the cause of dyslexia. Vision problems are not more common in children with reading im pair­ ment.4 Not only are vision-related approaches misdirected in theory, but authors of systematic literature reviews have concluded there is no scientific evidence of their efficacy for learning disabilities.3,7'9 Eye and vision problems occur in approximately 5% to 10% of early elementary students and 25% of high school students, and pediatricians should recognize that some treat­ able vision problems can masquer­ ade as a learning problem or may be coexisting. For example, children with large amounts of farsighted­ ness or astigmatism may be unin­ terested in books whereas children 22 a valuable role in the assessment of children with suspected or estab­ lished learning disabilities, including dyslexia, to determine whether there are any eye or vision problems that could be interfering with learning or reading. As part of the manage­ ment team for children with dys­ lexia, pediatric ophthalmologists can help families by reinforcing informa­ tion on the condition, dispelling the myths, and providing guidance to resources available online, in print, and in the community. Referral for an ophthalmologic evaluation is ind icated when a vision problem is suspected or the child fails vision screening. The pediat­ ric ophthalmologist will perform a complete dilated eye and vision examination, thoroughly evaluate near vision, and perform a cycloplegic refraction. Most ocular condi­ tions found in children, including strabism us, am blyopia, conver­ gence or focusing deficiencies, and CONTEM PORARYPEDIATRICS.COM | AUGUST 2016 refractive errors, are treatable with glasses and will improve quickly. Children diagnosed with symptom­ atic convergence insufficiency are usually prescribed convergence eye exercises to be performed at home. These exercises usually result in improved reading comfort within a few weeks. In-office exercises with at-home reinforcement are consid­ ered if a child continues to show signs and symptoms of convergence insufficiency. It is crucial to understand that although these exercises may relieve eye strain, they are not a treatment for coexisting dyslexia. Furthermore, pediatricians should know that con­ vergence insufficiency is frequently overdiagnosed. Any child who has been diagnosed with convergence insufficiency should be referred to a pediatric ophthalmologist for a second opinion as should any child who has been recommended vision therapy or tinted lenses or filters as treatment for a reading disability. Caring for other conditions The pediatric evaluation of a child presenting with reading difficul­ ties or dyslexia should also include assessments for other potentially contributing, m asquerading, or coexisting conditions. In addition to vision screening, these investiga­ tions should include hearing screen­ ing and evaluations for behavioral, medical, and mental health disor­ ders. Notably, ADHD and reading disability are highly comorbid. The inattentive type of ADHD has been found in 18% to 42% of children with dyslexia, and rates of dyslexia among children with ADHD have been reported to be within a similar peer-reviewed range.10 Dyslexia also may be asso­ ciated with oppositional defiant disorder, obsessive-compulsive dis­ order, anxiety, and depression. Counselling and advocacy t ONLINE RESOURCES FOR INFORMATION ON DYSLEXIA International Dyslexia Association www.eida.org Learning Disabilities (LD) OnLine www.ldonline.org Some p ed iatrician s can help to National Center for Learning Disabilities advocate for the child as needed, for example, by writing a letter to Reading Rockets request an evaluation or accom­ Yale Center for Dyslexia and Creativity modations. To ensure that children University of Michigan— Dyslexia Help get appropriate care, pediatricians should be prepared to provide fam­ ilies with names of professionals in available educational materials and/ their area who are qualified to evalu­ or lists of resources where families ate and treat children with dyslexia. can get information about dyslexia, Children who have not been making its evaluation and treatment, and progress after a prior psychoeducatheir child’s rights for evaluation, tional evaluation and interventions special services, and accommoda­ could benefit from referral for fur­ tions as defined by the Individuals ther evaluation by a developmentalwith Disabilities Education Act behavioral pediatrician or neuro­ (IDEA), and Section 504 of the psychologist. Although private eval­ Americans W ith Disabilities Act uation of a child with suspected (ADA). Table 3 lists several useful dyslexia may be expensive, a proper websites for more information. diagnosis may save m any years Finally, the role of the pediatri­ of frustration and failed educa­ cian in helping children to become tion, as well as thousands proficient readers begins of dollars on unneces­ early in the providersary and unsupported fam ily relationship, FAST FACT treatments. Attention-deficit/ long before consid­ hyperactivity Pediatricians ering whether chil­ disorder and reading should be sym pa­ dren are at risk for disability are highly thetic to the stress or showing signs of comorbid. that a child with dys­ dyslexia. Recognizing lexia and his or her fam­ that oral language is the ily are experiencing and foundation for reading and should provide encouragement and that children with speech and lan­ em otional support. In addition, guage delay or difficulties have a while counselling parents to support 50% likelihood of having difficul­ their child’s efforts at reading, pedi­ ties learning to read, pediatricians atricians should also advise them to can have an im portant impact by pursue their child’s strengths and promoting early language develop­ allow time for activities in which the ment. The Books Build Connections child finds enjoyment and excels. Toolkit, developed by the American Pediatricians also should have Academy of Pediatrics and Reach AUGUST 2016 I www.ncld.org www.readingrockets.org www.dyslexia.yale.edu http://dyslexiahelp.umich.edu Out and Read, contains material for pediatricians and parents and offers a practical resource to assist in these efforts.11 Conclusion Dyslexia is a common and lifelong language-based learning disabil­ ity that may be detected as early as kindergarten or first grade. Affected children are most likely to obtain the greatest benefit when the prob­ lem is identified early followed by prompt initiation of appropri­ ate language and reading instruc­ tion. Care should be taken to avoid using ineffective methods of treat­ ment that waste time and family resources and often critically delay proper remediation. By being vigilant to signs of dyslexia, dispelling the myths sur­ rounding this condition, helping to coordinate care, and providing support and encouragement, pedia­ tricians with the aid of pediatric ophthalmologists can help children with dyslexia enjoy success in school and in their daily life. ■ For references, go to ContemporaryPediatrics.com/ dyslexia O CONTEMPORARYPEDIATRICS.COM 23 Copyright of Contemporary Pediatrics is the property of Advanstar Communications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. 59 OCTOBER 15, 2017 :: Ophthalmology Times clinical diagnosis Pediatric ophthalmology role vital in detection, treatment of dyslexia Involvement key in determining if vision problems may be interfering with learning, reading By Sheryl M. Handler, MD LEARNING TO READ is an extremely complex process that has been described to be as challenging as learning rocket science.1 Therefore, it is not too surprising that, for many reasons, more than 60% of children in America fail to meet standards for reading proficiency.2 Multiple issues may underlie this reading difficulty, including poor early language development, inadequate instruction, insufficient reading practice, lack of background knowledge, and intellectual disability. In some children, however, the problem is the specific learning disability called dyslexia. Dyslexia is by far the most common learning disability and is present in some degree in up to 20% of children.3 Just as early detection and intervention are crucial in medical diseases, the same is true in learning disabilities. The consequences of untreated dyslexia are broad and can be significant, including effects on academic success and psychosocial well being. Children with dyslexia experience intense frustration; may act aggressively or withdraw; frequently become targets of bullying and ridicule; have low self-esteem; and may even develop mental health problems, including anxiety and depression. Physicians—including pediatricians with the aid of pediatric ophthalmologists—have the opportunity and responsibility to enable detection and proper treatment of dyslexia in children. This article aims to provide information and strategies that will allow clinicians to best assist and advise patients and their parents. DY SL E X I A DE F I N E D Dyslexia is a language-based learning disability characterized by difficulties with decoding (sounding out) words, fluent word recognition, and/or reading-comprehension skills. Children with dyslexia often develop secondary problems with comprehension, spelling, writing, and knowledge acquisition. The difficulties found in dyslexia are usually caused by a phonological deficit (an auditory processing problem involving hearing the sounds in speech). The phonological deficit leads to difficulty connecting speech sounds to Table 1: Risk factors and signs of dyslexia RISK FACTORS POSSIBLE SIGNS > Family history > Language or speech problems, including articulation > Prematurity or birth problems > Difficulty with rhymes > Fetal exposure to drugs or alcohol > Difficulty learning the names of letters > Hearing, language, or speech problems, including articulation > Trouble connecting letters to their sounds > Developmental delay > Difficulty sounding out words > ADHD or other neurological problems > Difficulty with sight word recognition > Other chronic health problems that may have caused school absences > Slow reading > Poor spelling Abbreviation: ADHD, attention-deficit/hyperactivity disorder. TAKE-HOME letters, which is a skill needed to and the right occipitotempodecode the written word. Alterral area.3 In discussing the definition natively, dyslexia in some chilBy being vigilant of dyslexia and its causes, it is dren results from problems with to signs of dyslexia, also useful for pediatricians to oral language skills, sight word dispelling the myths, be aware of the many myths and recognition, processing speed, and coordinating misperceptions that exist. Dyscomprehension, attention, or care, physicians — lexia is not a condition where verbal working memory. including pediatric readers see letters or words upAnatomical and imaging ophthalmologists side down or backward. It is not studies investigating brain de—can help children related to visual or eye-tracking velopment and function show with dyslexia enjoy problems.4 a corresponding physical basis success in school and In addition, dyslexia is not a for dyslexia in language-related in daily life. developmental issue that chilareas of the brain. The brains dren may be expected to outof persons with dyslexia funcgrow; rather, it is a persistent tion differently than the brains of “typical readers” before they even start to lifelong condition. Dyslexia also is not related to intelligence read, as dyslexics use an alternative pathway or laziness in a child. Dyslexia occurs in perfor reading. Specifically, these investigations reveal that sons with low, normal, and high intelligence persons with dyslexia have dysfunction in the quotients (IQs) alike. The fact that dyslexia is not related to IQ, left-hemisphere posterior reading areas with corresponding compensatory use of the bilat- however, creates the potential for a signifiContinues on page 60 : Dyslexia eral inferior frontal gyri of both hemispheres 60 OCTOBER 15, 2017 :: Ophthalmology Times clinical diagnosis DYSLEXIA ( Continued from page 59 ) cant learning disability to be overlooked in an otherwise bright child. Dyslexics are often perceived to be “lazy” or “not working up to their potential” when, in fact, they often work harder and longer than their peers. In addition, there is no male predominance for dyslexia. It is found almost equally in boys and girls, but tends to be identified earlier and more often in boys, perhaps because boys tend to “act out” when they are unable to do a difficult task versus girls who are inclined to make themselves “invisible” in the classroom. DETECTION, DI AGNOSIS A diagnosis of dyslexia is established clinically based on history, observation, and a battery of age-appropriate educational tests interpreted by a knowledgeable, qualified professional. Although it is not up to pediatricians to make the diagnosis, an understanding of dyslexia can help to identify children by being attentive to risk factors and signs that are elicited in the medical and social history during wellchild exams. (Table 1 on Page 59) Dyslexia is heritable and familial, and so the history should ascertain whether there is a family history of speech and language problems or dyslexia. Other risk factors for dyslexia include prematurity, neurological problems, and developmental or language delays. Early warning signs for dyslexia in preschoolaged children include trouble learning nursery rhymes or playing rhyming games; confusing words that sound alike; mispronouncing words; and trouble recognizing letters of the alphabet.5 Early elementary school children with dyslexia often have difficulty learning the names and sounds of the letters; separating and blending sounds; sounding out words; recognizing sight words; and spelling. They often read slowly and dislike reading. A parent’s complaint that a child is not doing well in school should prompt questions to explore the presence of reading difficulties. Although many dyslexic students are identified in the primary grades, dyslexia’s possible presence should not be overlooked in older children and teenagers. Signs in adolescents include a history of phonologically based reading difficulties, slow reading, choppiness when reading aloud, poor comprehension, and requiring more time to finish assignments or tests.5 These adolescent students also may have multiple school truancies and/or behavioral Table 2: Strategies for treating dyslexia Explicitly teach reading skills > Phonemic awareness > Intensive systematic phonics > Morphology—analysis of words and word parts > Memorization of sight words > Fluency training—guided oral reading > Vocabulary building > Comprehension techniques Spelling—including the rules Writing issues, such as anger, aggression, depression, or even suicidal tendencies and possibly alcohol or drug use. If a family history of dyslexia or other risk factors exists, the child’s early language development and school progress should be carefully monitored. A formal psychoeducational school evaluation is needed to identify dyslexia and will provide an understanding of the child’s strengths and weaknesses, the severity of the problem, and whether the child has a “specific learning disability” that is eligible for special education and support programs. Severe dyslexia typically qualifies a child for an Individualized Educational Plan, special education, and related services. The psychoeducational or broader neuropsychological or developmental-behavioral pediatric evaluation will identify comorbid conditions, and the findings will provide the foundation for a treatment plan. Fortunately, individuals who have severe dyslexia are in the minority. The other side of the coin, however, is that those students whose disorder is more moderate or mild may not be readily recognized and/or may not qualify for treatment, although they would benefit from those services. T R E AT I NG DY SL E X I A The prognosis for a child with dyslexia depends not only on the specific features of the disorder and its severity, but also on the timeliness and appropriateness of intervention (Table 2). Studies show that beginning remediation in first grade versus waiting 2 more years greatly increases the chance that a child will later be able to read at grade level.6 Nevertheless, it is never too late to help. Effective intervention for dyslexia targets its etiology as a language-based disorder and should be provided by a professional with the appropriate training. Children with dyslexia are best served with lessons provided in small group settings, which bring together students who are at the same reading level and include no more than 5 students. Remediation programs that follow the International Dyslexia Association guidelines call for a “Structured Literacy Approach.” These programs explain language in an explicit, systematic, sequential, and multisensory manner. They include training in the five reading skills: 1) phonemic awareness (hearing the sounds in words); 2) phonics (correlating sounds with letters); 3) fluency (ability to read with speed, accuracy and expression); 4) vocabulary; and 5) comprehension. Emphasis is also placed on learning word structure, letter patterns, spelling, and writing. Whereas early management of dyslexia focuses on remediation of the reading problem, for older students there is a shift toward providing tools and accommodations. Accommodations allow the student to access his/her higher-level thinking and reasoning skills. These measures include access to assistive technology (e.g., recorded books, text reading software, note takers, spell checkers) as well as provisions to enable test taking (e.g., extended testing time, a special quiet room, or preferential seating). PEDIATRICIAN ROLE Parents of children with dyslexia often seek counsel from their pediatrician about dyslexia treatments, so clinicians should be aware of both the proven and unproven interventions being recommended and heavily promoted. By being informed about the variety of approaches, as well as being able to discuss the lack of evidence supporting the efficacy for the unproven interventions, pediatricians will be equipped to educate patients and their families, encourage positive proven approaches, discourage the use of unproven techniques, and thus prevent families from wasting valuable time and resources pursuing expensive, nonproductive therapies. Non–evidence-based approaches promoted for remediation of reading difficulties in children with dyslexia include medication for vestibular dysfunction, chiropractic manipulation, physical exercises, and dietary supplementation or restrictions.5 In particular, however, the erroneous concept that dyslexia is a vision-based disorder has spawned a variety of interventions involving the use of training glasses, eye exercises, behavioral/perceptual vision therapy, and colored lenses and overlays. Although visual problems, including convergence insufficiency, can 61 OCTOBER 15, 2017 :: Ophthalmology Times clinical diagnosis hamper reading, they are not the cause of dyslexia. Vision problems are not more common in children with reading impairment.4 Not only are vision-related approaches misdirected in theory, but authors of systematic literature reviews have concluded there is no scientific evidence of their efficacy for learning disabilities.3,7–9 Eye and vision problems occur in about 5% to 10% of early elementary students and 25% of high school students, and pediatricians should recognize that some treatable vision problems can masquerade as a learning problem or may be co-existing. For example, children with large amounts of farsightedness or astigmatism may be uninterested in books whereas children with nearsightedness may have difficulty seeing the board. Symptomatic convergence insufficiency, which occurs more rarely in elementary school-aged children, makes reading at near blurry and uncomfortable. Consequently, affected children may read only for short periods, and on that basis may be thought to have dyslexia or attention-deficit/hyperactivity disorder (ADHD), especially the inattentive type. Pediatric ophthalmologists play a valuable role in the assessment of children with suspected or established learning disabilities, including dyslexia, to determine whether there are any eye or vision problems that could be interfering with learning or reading. As part of the management team for children with dyslexia, pediatric ophthalmologists can help families by reinforcing information on the condition, dispelling the myths, and providing guidance to resources available online, in print, and in the community. Referral for an ophthalmologic evaluation is indicated when a vision problem is suspected or the child fails vision screening. The pediatric ophthalmologist will perform a complete dilated eye and vision examination, thoroughly evaluate near vision, and perform a cycloplegic refraction. Most ocular conditions found in children, including strabismus, amblyopia, convergence or focusing deficiencies, and refractive errors, are treatable with glasses and will improve quickly. Children diagnosed with symptomatic convergence insufficiency are usually prescribed convergence eye exercises to be per- formed at home. These exercises usually result in improved reading comfort within a few weeks. In-office exercises with at-home reinforcement are considered if a child continues to show signs and symptoms of convergence insufficiency. It is crucial to understand that although these exercises may relieve eye strain, they are not a treatment for co-existing dyslexia. Furthermore, pediatricians should know that convergence insufficiency is frequently overdiagnosed. Any child who has been diagnosed with convergence insufficiency should be referred to a pediatric ophthalmologist for a second opinion, as should any child who has been recommended vision therapy or tinted lenses or filters as treatment for a reading disability. CAR ING FOR OTHER CONDITIONS The pediatric evaluation of a child presenting with reading difficulties or dyslexia should also include assessments for other potentially contributing, masquerading, or co-existing conditions. In addition to vision screening, these investigations should include hearing screening Continues on page 62 : Vision screening Making a WORLD of Difference Lions Eye Institute for Transplant & Research in Tampa proudly introduces One World Sight Alliance. Through an unparalleled global network, One World Sight Alliance endeavors to eradicate preventable blindness by way of international tissue distribution, physician training and eye bank development. LEARN MORE AT AAO BOOTH #502 62 OCTOBER 15, 2017 :: Ophthalmology Times clinical diagnosis VISION SCREENING Table 3: Online resources for information on dyslexia ( Continued from page 61 ) and evaluations for behavioral, medical, and mental health disorders. Notably, ADHD and reading disability are highly comorbid. The inattentive type of ADHD has been found in 18% to 42% of children with dyslexia, and rates of dyslexia among children with ADHD have been reported to be within a similar range.10 Dyslexia also may be associated with oppositional defiant disorder, obsessive-compulsive disorder, anxiety, and depression. COU NSELING, A DVOC ACY Some pediatricians can help to advocate for the child as needed, for example, by writing a letter to request an evaluation or accommodations. To ensure that children get appropriate care, pediatricians should be prepared to provide families with names of professionals in their area who are qualified to evaluate and treat children with dyslexia. Children who have not been making progress after a prior psychoeducational evaluation and interventions could benefit from referral for further evaluation by a developmental-behavioral pediatrician or neuropsychologist. Although private evaluation of a child with suspected dyslexia may be expensive, a proper diagnosis may save many years of frustration and failed education, as well as thousands of dollars on unnecessary and unsupported treatments. Pediatricians should be sympathetic to the stress that a child with dyslexia and his or her family are experiencing and should provide encouragement and emotional support. In addition, while counseling parents to support their child’s efforts at reading, pediatricians should also advise them to pursue their child’s strengths and allow time for activities in which the child finds enjoyment and excels. Pediatricians also should have available educational materials and/or lists of resources where families can get information about dyslexia, its evaluation and treatment, and their child’s rights for evaluation, special services, and accommodations as defined by the Individuals with Disabilities Education Act (IDEA), and Section 504 of the Americans With Disabilities Act (ADA). Table 3 lists several useful websites for more information. Finally, the role of the pediatrician in helping children to become proficient readers begins early in the provider-family relationship, long before considering whether children are International Dyslexia Association www.eida.org Learning Disabilities (LD) OnLine www.ldonline.org National Center for Learning Disabilities www.ncld.org Reading Rockets www.readingrockets.org Yale Center for Dyslexia and Creativity www.dyslexia.yale.edu University of Michigan—Dyslexia Help http://dyslexiahelp.umich.edu at risk for or showing signs of dyslexia. Recognizing that oral language is the foundation for reading and that children with speech and language delay or difficulties have a 50% likelihood of having difficulties learning to read, pediatricians can have an important impact by promoting early language development. The Books Build Connections Toolkit, developed by the American Academy of Pediatrics and Reach Out and Read, contains material for pediatricians and parents and offers a practical resource to assist in these efforts.11 CONCLUSION Dyslexia is a common and lifelong languagebased learning disability that may be detected as early as kindergarten or first grade. Affected children are most likely to obtain the greatest benefit when the problem is identified early followed by prompt initiation of appropriate language and reading instruction. Care should be taken to avoid using ineffective methods of treatment that waste time and family resources and often critically delay proper remediation. By being vigilant to signs of dyslexia, dispelling the myths surrounding this condition, helping to coordinate care, and providing support and encouragement, pediatricians with the aid of pediatric ophthalmologists can help children with dyslexia enjoy success in school and in their daily life. ■ References 1. Moats LC. Teaching reading is rocket science: What expert teachers of reading should know and be able to do. Washington, DC: American Federation of Teachers; 1999. Available at: http://www.aft.org/ sites/default/files/reading_rocketscience_2004.pdf. Accessed July 19, 2016. 2. Nation’s Report Card. Nine subjects, three grades, one report card. Available at: http://www. nationsreportcard.gov/. Accessed July 19, 2016. 3. Handler SM, Fierson WM, Section on Ophthalmology; Council on Children with Disabilities; American 4. 5. 6. 7. 8. 9. 10. 11. Academy of Ophthalmology; American Association for Pediatric Ophthalmology and Strabismus; American Association of Certified Orthoptists. Learning disabilities, dyslexia, and vision. Joint Technical Report. Pediatrics. 2011;127(3):e818-e856. Creavin AL, Lingam R, Steer C, Williams C. Ophthalmic abnormalities and reading impairment. Pediatrics. 2015;135:1057-1065. Shaywitz SE, Gruen JR, Shaywitz BA. Management of dyslexia, its rationale, and underlying neurobiology. Pediatr Clin North Am. 2007;54:609-623. Torgesen JK. Avoiding the devastating downward spiral: the evidence that early intervention prevents reading failure. Available at: http://www.aft.org/ periodical/american-educator/fall-2004/avoidingdevastating-downward-spiral. Accessed July 19, 2016. Fletcher JM, Francis DJ, Morris RD, Lyon GR. Evidence-based assessment of learning disabilities in children and adolescents. J Clin Child Adolesc Psychol. 2005;34:506-522. Barrett BT. A critical evaluation of the evidence supporting the practice of behavioural vision therapy. Ophthalmic Physiol Opt. 2009;29:4-25. Ritchie SJ, Della Sala S, McIntosh RD. Irlen colored overlays do not alleviate reading difficulties. Pediatrics. 2011;128:e932-e938. Germanò E, Gagliano A, Curatolo P. Comorbidity of ADHD and dyslexia. Dev Neuropsychol. 2010;35:475493. American Academy of Pediatrics. Books Build Connections Toolkit. Available at: https://littoolkit.aap. org/Pages/home.aspx. Accessed July 19, 2016. SHERYL M. HANDLER, MD P: 818/789-2226 Dr. Handler is a pediatric ophthalmologist in private practice in Encino, CA. She is one of the authors of the American Academy of Pediatrics, American Academy of Ophthalmology, American Association for Pediatric Ophthalmology and Strabismus, and American Association of Certified Orthoptists 2011 Joint Technical Report, Learning Disabilities, Dyslexia, and Vision. She has nothing to disclose in regard to affiliations with or financial interests in any organizations that may have an interest in this article. This article originally appeared in sister publication ContemporaryPediatrics. com/Dyslexia Copyright of Ophthalmology Times is the property of Advanstar Communications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. Learning and Individual Differences 49 (2016) 209–215 Contents lists available at ScienceDirect Learning and Individual Differences journal homepage: www.elsevier.com/locate/lindif Predicting responsiveness to intervention in dyslexia using dynamic assessment Sebastián Aravena a,b,c,⁎, Jurgen Tijms a,b,c, Patrick Snellings a,c, Maurits W. van der Molen a,c a b c Department of Developmental Psychology, University of Amsterdam, Nieuwe Achtergracht 129B, 1018 WT Amsterdam, The Netherlands IWAL Institute, Amsterdam, Tweede Hugo de Grootstraat 45a, 1052 LB, The Netherlands Rudolf Berlin Center, Amsterdam, Valckenierstraat 65-67, 1018 XE, The Netherlands a r t i c l e i n f o Article history: Received 13 June 2015 Received in revised form 30 March 2016 Accepted 24 June 2016 Keywords: Dyslexia Dynamic assessment Letter–speech sound learning Treatment success Response to intervention (RTI) a b s t r a c t In the current study we examined the value of a dynamic test for predicting responsiveness to reading intervention for children diagnosed with dyslexia. The test consisted of a 20-minute training aimed at learning eight basic letter–speech sound correspondences within an artificial orthography, followed by a short assessment of both mastery of these correspondences and word reading ability in this unfamiliar script. Fifty-five (7- to 11-yearold) children diagnosed with dyslexia engaged in specialized intervention during approximately 10 months and their reading and spelling abilities were assessed before and after. Our results indicated that the dynamic test predicted variance in reading skills at posttest, over and above traditional static measures, such as phonological awareness and rapid naming. These findings indicate that responsiveness to learning new letter–speech sound correspondences has a prognostic value for the success of specialized reading intervention. © 2016 Elsevier Inc. All rights reserved. 1. Introduction Developmental dyslexia, henceforth referred to as dyslexia, is characterized by a specific and significant impairment in the automatic recognition of written words (Fletcher & Lyon, 2008; Peterson & Pennington, 2012; Snowling, 2012). There is ample evidence that specialized intervention is effective in ameliorating reading and spelling proficiency of children with dyslexia (see Galuschka, Ise, Krick, & Schulte-Körne, 2014 for an overview). Unfortunately, not all dyslexic readers benefit to the same extent and there is a substantial amount of non-responders as well (Galuschka et al., 2014; Singleton, 2009; Torgesen, 2005). Gaining more insight into factors that can predict responsiveness to intervention in dyslexia would be very welcome as it could help us to identify nonresponders at an early stage and, by doing so, to prevent wasting time, effort, and resources on interventions that are not effective. A framework that is particularly important in this context is response to intervention (RTI), which is a common practice in educational settings across the United States and several European countries nowadays. RTI is an approach in which a tutor provides a pupil with progressively intense and individualized tiers of instruction with the aim of finding the best possible way to educate children and of identifying ⁎ Corresponding author at: IWAL Institute, Amsterdam, Tweede Hugo de Grootstraat 45a, 1052 LB, The Netherlands. E-mail address: s.aravena@uva.nl (S. Aravena). http://dx.doi.org/10.1016/j.lindif.2016.06.024 1041-6080/© 2016 Elsevier Inc. All rights reserved. children with learning disabilities (Fuchs & Fuchs, 2006; Grigorenko, 2009; Gustafson, Svensson, & Fälth, 2014). Pupils who do not respond to Tier 1 receive more intensive and individualized instruction within Tier 2, and those who are unresponsive to Tier 2 proceed with even more rigorous instruction within Tier 3. Depending on the educational system, the framework is sometimes complemented by a fourth tier, which consists of placement in special education or referral to assessment and therapy within the health care system. Although many pupils benefit from RTI as they receive high-quality instruction as soon as learning difficulties arise, the notion that intervention should initially be of modest intensity has been questioned (Denton et al., 2011; Vaughn, Denton, & Fletcher, 2010). Especially the value of Tier 2 intervention for the most learning disabled continues to be a subject to debate (Compton et al., 2012; Fuchs, Fuchs, & Compton, 2010). Indeed, there is evidence that engaging in less intensive tiers of intervention may not be effective for addressing the reading difficulties of children with dyslexia (Vaughn et al., 2010). Early identification of nonresponders could thus potentially improve their chance to benefit from intervention by intensifying initial intervention. A convenient starting point for identifying factors predicting intervention success would be to focus on the standard assessment of dyslexia, which typically consists of a combination of reading and writing tasks along with a set of phonology-related tasks, such as phonological awareness, rapid naming, and verbal short-term memory, as well as some general cognitive measures. Indeed, several studies indicate that some of these factors, among which poor phonological awareness in 210 S. Aravena et al. / Learning and Individual Differences 49 (2016) 209–215 particular, can predict unresponsiveness to early literacy intervention within children at risk for dyslexia (see Al Otaiba & Fuchs, 2002 and Nelson, Benner, & Gonzalez, 2003 for an overview), but it is far less clear whether these findings hold for children diagnosed with dyslexia (Frijters et al., 2011; Hatcher & Hulme, 1999; Morris et al., 2012; Tijms, 2011). For this group there is a paucity in our knowledge of factors moderating responsiveness to intervention (Démonet, Taylor, & Chaix, 2004; Frijters et al., 2011; Hoeft et al., 2011; Shaywitz, Morris, & Shaywitz, 2008; Tijms, 2011). A recent meta-analysis including twenty-two randomized controlled trial studies of reading disabled children failed to identify subject-related moderators of responsiveness to intervention (Galuschka et al., 2014). Dynamic assessment (DA) might be a viable approach for examining potential moderators of responsiveness to intervention. The focus of DA is on learning potential rather than learning outcome (Grigorenko, 2009; Gustafson et al., 2014). A typical DA procedure requires the pupil to engage in a training in which feedback is provided. The effect of training is then used to estimate the pupils' learning potential. There is ample evidence that this kind of process-oriented testing better predicts future learning than conventional testing within various academic domains, including reading skill (Caffrey, Fuchs, & Fuchs, 2008; Fuchs, Compton, Fuchs, Bouton, & Caffrey, 2011; Grigorenko & Sternberg, 1998; Gustafson et al., 2014; Jeltova et al., 2007; Spector, 1992). However, other studies have shown little advantage of dynamic testing over static testing (Caffrey et al., 2008). In a recent study Petersen, Allen, and Spencer (2014) compared the utility to predict reading difficulty at first grade of two DA reading measures and two commonly used one-point-in-time pre-reading measures administered to 600 kindergarten children and found both DA measures to be superior to the common static measures. DA has also been used to examine moderators of responsiveness to intervention recently. Cho, Compton, Fuchs, Fuchs, and Bouton (2014) showed that DA predicted the responsiveness to a validated reading intervention program. In this study, firstgrade students received Tier 2 reading intervention within small groups during 14 weeks. DA of decoding was found to be a significant predictor of the growth in word identification fluency and the final level attained. In the current study, we applied DA to children diagnosed with dyslexia in order to predict the success of subsequent specialized Tier 4 intervention. The DA we developed consists of a 20-minute training aimed at learning eight new basic letter–speech sound correspondences, followed by a short assessment of both mastery of the correspondences and word reading ability in this unfamiliar script. Letter– speech sound learning is the central focus of the training, because recent research suggests that a fundamental letter–speech sound learning deficit is a key factor in dyslexia (Blomert, 2011; Kronschnabel, Brem, Maurer, & Brandeis, 2014; McNorgan, Randazzo-Wagner, & Booth, 2013; Mittag, Thesleff, Laasonen, & Kujala, 2013; Peterson & Pennington, 2015; van Atteveldt & Ansari, 2014; Žarić et al., 2014). The advantage of adopting an artificial script is that differences in previous exposure to experimental stimuli can be ruled out, allaying concerns about noncontrolled factors influencing performance. In a previous study we demonstrated that our DA procedure differentiates between dyslexic readers and normal readers and predicts individual differences in reading and spelling ability (Aravena, Tijms, Snellings, & van der Molen, 2015). In the current study we examined whether, in addition to its diagnostic value, the DA procedure has prognostic value as well. The participating children engaged in specialized Tier 4 intervention during approximately 10 months. We tested their reading and spelling abilities before and after intervention and related these to the scores on our DA, as well as to the scores on two conventional static measures frequently used for the assessment of dyslexia, namely a phonological awareness task and an alphanumeric rapid naming task. Unlike most approaches to DA (Grigorenko, 2009; Grigorenko & Sternberg, 1998), our assessment did not involve instruction but just associative learning from exposure and implicit feedback. The 20-minute training consisted of a computer game in which children had to match speech sounds to unfamiliar letters. As correct responses led to success in the game and incorrect responses were penalized, children learned the letter–speech sound correspondences just by playing the game, without being aware of learning. Instructions were only related to the specifics of the game and did not reveal the underlying learning objective. This approach was chosen to approximate letter–speech sound learning as it naturally occurs and to measure the capacity to master new letter–speech sound correspondences, without interference from more general factors related to instruction, such as intelligence, verbal comprehension, or attention. In brief, in the current study we examined whether a new DA procedure predicted the success of a subsequent specialized intervention within a group of children diagnosed with dyslexia. We expected this procedure to be an adequate candidate for this purpose for two reasons. First, because it focuses on the formation of letter–speech sound correspondences, a process that appears to be disrupted in children with dyslexia. Second, because it focuses on learning rate rather than on learning outcome. 2. Method 2.1. Participants Participants were 55 primary education pupils (30 boys and 25 girls) diagnosed with dyslexia recruited from a nation-wide center for dyslexia in the Netherlands. The children had a mean age of 9 years and 3 months (SD = 12.39 months, age range = 7.33–11.08 years). An estimate of general intelligence was obtained by averaging the standardized C-scores (M = 5, SD = 2) of the subtest Analogies from the SONR (Laros & Tellegen, 1991), a non-verbal reasoning-by-analogy task in which the child had to extract a principle and to apply it to a new situation (r = 0.79, test–retest), and the subtest Vocabulary from the WISCIII (Kort et al., 2005), a measure of expressive vocabulary requiring the child to describe the meaning of words of increasing complexity (r = 0.90, test–retest). The IQ estimates ranged from 3 to 8.5 (M = 5.57, SD = 1.37). Informed consent was obtained from the parents of each child. Consistent with standard norms for severe dyslexia in the Dutch health care system (Blomert, 2006), children were diagnosed with dyslexia when they met all of the following three inclusion criteria: (1) either word reading speed was 1.5 standard deviation (SD) or more below average or, word reading speed was at least 1 SD below average together with a spelling skill of 1.5 SD or more below average; (2) performance on at least two out of six administered phonology-related tasks was at least 1.5 SD below average; and (3) the child had shown a poor response to intervention provided at school. Exclusionary criteria were uncorrected sensory disabilities, broad neurological deficits, low IQ (b80), poor school attendance, and ADHD. Because we incorporated Hebrew graphemes in our assessment, previous experience with Hebrew script was also an exclusionary criterion. All participants were native speakers of Dutch. The study was approved by the University's Ethics Committee. 2.2. Dynamic assessment The dynamic assessment (DA), which had a total duration of approximately 30 min, consisted of a 20-minute training dedicated to learning non-existent letter–speech sound correspondences followed by a short assessment of both mastery of the newly learned correspondences and word reading ability in the artificial script. A summary of the different components of the DA is provided below. 2.2.1. The letter–speech sound training The training consisted of a computer game in which the child had to match speech sounds to their corresponding orthographic representations (Aravena et al., 2015). Correct associations were rewarded while S. Aravena et al. / Learning and Individual Differences 49 (2016) 209–215 incorrect associations were penalized. Fast playing was reinforced by progressive time restrictions and by providing bonuses for fast playing. More specific, children operated a cannon at the bottom of the screen, moving it horizontally. The upper part of the screen was composed of columns of balloons containing single graphemes. Children were required to act on speech sounds that were presented repeatedly in the game. The response consisted of releasing bullets from the cannon and associating them to their corresponding grapheme. When children managed to clear a field of balloons, a new field was presented. As the amount of distractor graphemes increased during the game, fields became gradually more complex. Fig. 1 depicts some screenshots from the game. The goal of the training was to learn a set of letter–speech sound correspondences from an artificial orthography. At the start of the game the child was presented with a standardized instruction that was integrated in the software. This instruction provided information regarding the specifics of the game but did not reveal the underlying learning objective. After the instruction, children received a short practice trial to become familiar with the set-up and the controls of the game. During the training session children were wearing headphones. 2.2.2. The artificial orthography The artificial orthography consisted of eight Hebrew graphemes, which were randomly matched to highly frequent Dutch phonemes, thereby providing eight basic non-existing letter–speech sound pairs. We adopted Hebrew script to capture the characteristics of graphemes as they naturally occur. Evidence exists that letter shapes are not an arbitrary cultural choice but rather a product of our neural architecture (Dehaene, 2009, p. 173). The phonemes, three vowels and five consonants, were selected based on their high frequency and their ability to, by combining, create a large corpus of words. Combinations of phonemes producing strong coarticulation effects were avoided. Table 1 presents the letter–speech sound correspondences that were used. The directionality of the script was left-to-right. 2.2.3. Letter–speech sound identification task within the artificial orthography In this task a phoneme was presented over headphones, while simultaneously two graphemes from the artificial orthography were displayed at the screen. One of these graphemes corresponded with the presented phoneme, while the other was as a distractor. By striking the corresponding button the child had to decide, as fast as possible, which of the graphemes belonged to the presented phoneme. The task consisted of 56 items. Response speed and accuracy were recorded automatically by the software. The score for response speed was the median speed of correct responses and the score for accuracy was the number of correct responses (respectively r = 0.96 and r = 0.90, split-half). 211 Table 1 Letter–speech sound correspondences within the artificial orthography. Letter ‫ט‬ ‫כ‬ ‫ם‬ ‫ף‬ ‫פ‬ ‫צ‬ ‫ך‬ ‫ש‬ Speech sound (IPA) [u] [ε] [α] [a] [k] [r] [l] [t] [n] Note. IPA, International Phonetic Alphabet. 2.2.4. Reading task within the artificial orthography We administered a time-limited test (3MAST) consisting of a list of 22 high-frequent Dutch words written within the artificial orthography. The words were presented in lowercase Arial typeface, font size 24, and arranged in two columns of equal length. The child had to read (column-wise) as many words as possible within 3 min. The score consisted of the number of words read correctly per second. 2.3. Traditional measures used for the assessment of dyslexia 2.3.1. Phonological awareness We assessed phonological awareness with a phoneme deletion task from the 3DM, a standardized and computerized battery for assessing dyslexia (Blomert & Vaessen, 2009). In this task the child had to delete consonants from aurally presented pseudowords (CVC or CCVCC structure) as fast as possible (for example /FOT/ minus /F/ makes /OT/). The score consisted of the percentage of correct responses (r = 0.85, internal consistency). 2.3.2. Rapid naming We assessed both rapid naming of letters and digits with a rapid naming task from the 3DM (Blomert & Vaessen, 2009). The child had to name aloud items presented on the computer screen as fast and accurate as possible. Within both domains sheets containing 15 items each were presented two times. The score per subtask was the mean response time of the two sheets (r = 0.80 for letters and r = 0.83 for digits, split-half reliability). In the current study we used a composite measure of alphanumeric rapid naming consisting of the scores of both the rapid naming of letters and digits. 2.4. Reading and spelling measures 2.4.1. Word reading We assessed word reading with a time-limited task from the 3DM (Blomert & Vaessen, 2009). This word-reading task included three different levels comprising high-frequency words, low-frequency words and pseudo-words. Each level contained 75 words, displayed on 5 sheets with 15 items each. The difficulty of each level increased systematically from monosyllabic words without consonant clusters to 3 or 4 syllabic words with consonant clusters in the fifth sheet. The child was instructed to read as many words as possible while maintaining accuracy within a time-limit of 30 s per level. Both accuracy (percentage of Fig. 1. Screenshots from the training. 212 S. Aravena et al. / Learning and Individual Differences 49 (2016) 209–215 correctly read words) and speed (number of words read correctly) were measured (respectively r = 0.73 and r = 0.95, test–retest). 2.4.2. Spelling We assessed spelling with a task from the 3DM (Blomert & Vaessen, 2009). In this task a word was presented over headphones while it was also visible on the screen. In the visually presented word a letter or letter combination was missing. By striking a key the child had to decide as fast as possible which of four different letters or letter combinations represented the missing part. Word frequencies varied systematically and words were either phonetically transparent (18 items) or needed the application of a Dutch spelling rule (36 items). Scores consisted of the percentage of accurate responses (r = 0.80, internal consistency). 2.5. Specialized intervention The Dutch educational system utilizes a three-tier approach to reading instruction. Children who do not respond to intensive Tier 3 intervention are assessed for their reading deficiency and those diagnosed with dyslexia receive specialized intervention within the health care system, which in the Netherlands represents the fourth tier. The intervention used in this study was a Dutch computer-based Tier 4 intervention program for treating dyslexia (LEXY). Intervention was provided by speech therapists and psychologists, on a one-to-one basis in weekly 45-min sessions. Sessions took place in a dyslexia center in the child's neighborhood. Besides these sessions at the center, participants were required to practice at home three times a week for 15 min. LEXY provides insight into the way written language transcribes the characteristics of the spoken language system by clarifying the phonological and morphological structure, and by explicitly training the rule-systems that are essential for the graphic representation of spoken language, using step-by-step algorithmic plans. All elements within the learning environment (like phonemic and orthographic units and mapping operations) are graphically represented on the computer screen (see Tijms & Hoeks, 2005 for a more detailed review of LEXY). The LEXY program aims at achieving a mastery level for each element of the program, which implies that participants do not pass through it at a fixed pace. On average the duration of the intervention program is 48 to 60 sessions, but in the current study the posttest was administered before the end of intervention, at 39 weeks. The program is in line with guidelines regarding effective intervention for children with dyslexia (Galuschka et al., 2014; Singleton, 2009) and its efficacy has been demonstrated repeatedly (Tijms, 2011; Tijms & Hoeks, 2005). 2.6. Design and procedure The DA was administered as part of a standard diagnostic assessment consisting of two 3.5 hour-sessions within one week interval. A trained psychologist administered the DA on a one-to-one basis during the second session. The assessment took place in a silent room in the dyslexia center. Children started with the intervention program approximately four months after the assessment. The pretest consisted of the word reading and the spelling task and took place during the first session of the intervention. The posttest, which included the same tasks, took place during the 39th session, which was after approximately 10 months (M = 43.0 weeks, SD = 1.9 weeks), depending on the amount of cancelled sessions due to illness or holidays. In total, 33 sessions were used for intervention and 6 sessions were used for assessment purposes. (M = 50, SD = 10) obtained at pretest from those obtained at posttest for each individual. A one-sample t-test was then conducted to determine whether the mean of these gain values was significantly different from zero. To determine whether the three dynamic assessment (DA) variables predict the improvement in reading and spelling skills during intervention we conducted a series of two-step fixed-entry multiple regression analyses with the posttest scores of reading and spelling measures as the dependent variables. In each of the analyses we entered the pretest score in the first step to filter out variance due to differences at the start of intervention. The three DA measures as well as phonological awareness (PA) and alphanumeric rapid naming (RAN) were added alternately in the second step to determine their individual contribution. In an additional series of multiple regression analyses we compared the predictive potential of the combined DA measures to the combined traditional measures by entering both phonological awareness and rapid naming in the second step and the three DA measures in the third step and vice versa. 3. Results 3.1. Overall effect of the intervention Table 2 presents the standardized T-scores for the reading and spelling tasks at pretest and posttest. A one-sample t-test showed that the treatment had a significant beneficial effect on reading accuracy (t(52) = 3.032, p = 0.004, d = 0.84) and on reading speed (t(52) = 7.071, p b 0.001, d = 1.96) as well as on spelling (t(52) = 5.937, p b 0.001, d = 1.65). Note that the gains are expressed in standardized scores, and thus reflect a shift in position within the normal distribution. In other words, the reading disabled children that received intervention made significantly more progress than their peers (from the national norm) during the same period. Despite the improvements that were made, the average accuracy and speed scores of reading were still below the normal range after 39 sessions of treatment. This was not surprising, however, given that the posttest was administered mid-term. The improvements found at the end of this treatment are typically more substantial (Tijms, 2011; Tijms & Hoeks, 2005). 3.2. Predicting reading and spelling gains during intervention The results from the multiple regression analyses, presented in Table 3, indicate that neither PA nor RAN made a significant contribution to predicting the improvement in any of the reading and spelling skills during intervention. The same was true for the artificial orthography-related accuracy measure of the letter–speech sound identification task (LSIa). The speed measure of the letter–speech sound identification task (LSIs), however, accounted for 17% of the variance in reading accuracy and 6% of the variance in reading speed at posttest. The contribution to the variance in spelling at posttest was negligible. The amount of words read per second within the artificial orthography (WPS) accounted for 12% of variance in reading accuracy at posttest, but did not contribute to the variance of any of the other measures. The results from the additional analyses, which are shown in Table 4, indicate that the three DA measures combined accounted for 23% of Table 2 The development of standardized t-scores for reading and spelling during intervention. t-Score at pretest 2.7. Analyses In order to understand the predictive potential of the pertinent variables, we first had a look at the overall effect of the intervention. Gain values were calculated by subtracting the standardized T-scores Reading accuracy Reading speed Spelling t-Score at posttest Gain values (t-score) M (SD) M (SD) M (SD) 32.44 (11.37) 29.60 (5.09) 37.07 (7.28) 37.25 (11.64) 34.94 (7.39) 44.49 (8.69) 5.11 (12.28)⁎⁎ 5.45 (5.61)⁎⁎ 7.75 (9.51)⁎⁎ ⁎⁎ Significant at the 0.01 level. S. Aravena et al. / Learning and Individual Differences 49 (2016) 209–215 Table 3 Regression models predicting reading and spelling measures at posttest. Reading accuracy Steps 1 2 2 2 2 2 Pretest PA RAN LSIa LSIs WPS R2 ΔR2 0.19 0.21 0.21 0.19 0.36 0.31 0.19⁎⁎ 0.02 0.02 0.00 0.17⁎⁎ 0.12⁎⁎ Reading speed Spelling R2 ΔR2 R2 ΔR2 0.42 0.43 0.42 0.42 0.48 0.43 0.42⁎⁎ 0.09 0.09 0.09 0.10 0.10 0.10 0.09⁎ 0.00 0.00 0.01 0.01 0.01 0.01 0.00 0.00 0.06⁎ 0.01 Note. PA, phonological awareness; RAN, rapid naming of alphanumeric items; LSIa, letter– speech sound identification accuracy; LSIs, letter–speech sound identification speed; WPS, number of words read per second. ⁎⁎ Significant at the 0.01 level. ⁎ Significant at the 0.05 level. variance in reading accuracy at posttest when entered in the second step and for 19% of additional variance when entered in the third step. The three DA measures thus predicted variance in reading accuracy at posttest, over and above traditional static measures, such as PA and RAN. 4. Discussion In the current study we used dynamic assessment (DA) for children diagnosed with dyslexia to examine whether it would predict the success of a subsequent specialized intervention. In a previous study we demonstrated that our DA predicts individual differences in reading and spelling ability and differentiates between dyslexic readers and normal readers (Aravena et al., 2015). The results from the current study indicate that in addition to its diagnostic value our DA has prognostic value as well. More specifically we found that the speed measure from the letter–speech sound identification task (LSIs) made a significant contribution to explaining variance in response to intervention on reading accuracy and speed and that the amount of words read per second within the artificial orthography (WPS) accounted for another significant portion of variance in reading accuracy at posttest. Our findings are consistent with previous findings demonstrating the added value of DA in forecasting reading development (Caffrey et al., 2008; Fuchs et al., 2011; Grigorenko & Sternberg, 1998; Gustafson et al., 2014; Jeltova et al., 2007; Petersen et al., 2014; Spector, 1992) and, more importantly, in predicting responsiveness to reading intervention (Cho et al., 2014). The current study strengthens and extends available data by showing that DA has the potential to predict Table 4 Regression models predicting reading and spelling measures at posttest. Reading accuracy Steps 1 2 3 2 3 Pretest PA RAN LSIa LSIs WPS LSIa LSIs WPS PA RAN Reading speed Spelling R2 ΔR2 R2 ΔR2 R2 ΔR2 0.19 0.19⁎⁎ 0.42 0.42⁎⁎ 0.09 0.09⁎ 0.23 0.04 0.43 0.01 0.09 0.00 0.42 0.19⁎⁎ 0.50 0.07 0.11 0.02 0.41 0.23⁎⁎ 0.48 0.06 0.11 0.02 0.42 0.00 0.50 0.02 0.11 0.00 Note. PA, phonological awareness; RAN, rapid naming of alphanumeric items; LSIa, letter– speech sound identification accuracy; LSIs, letter–speech sound identification speed; WPS, number of words read per second. ⁎⁎ Significant at the 0.01 level. ⁎ Significant at the 0.05 level. 213 responsiveness to intervention beyond Tier 3 intervention within a sample of children diagnosed with dyslexia. This is of particular interest because so far the quest for predictors of responsiveness to intervention for this group has not been very fruitful (Frijters et al., 2011; Hoeft et al., 2011; Tijms, 2011). Our results indicate that a dynamic approach to assessment provides new opportunities to predict responsiveness to intervention even for the most reading disabled. From a clinical point of view early identification of potential non-responders is valuable because it may assist practitioners adapting their educational strategies at an initial stage or even start off a prompt deployment of alternative ways of accessing written information, such as computer-based readers. A possible explanation for the current success of our DA approach is that it not only identifies an essential underlying factor, namely a letter– speech sound binding deficit, but that it also provides an index of the extent to which this underlying problem interferes with learning to read. This explanation is in line with findings from longitudinal studies indicating that deficits in the initial learning of letter–speech sound associations are an important risk factor for developing reading difficulties (Caravolas et al., 2012; Lyytinen, Ronimus, Alanko, Poikkeus, & Taanila, 2007). According to Lyytinen et al. (2007) it is a serious reason for concern when a child struggles storing grapheme–phoneme connections in memory in stable form. We think it is this ‘struggle’ that manifests itself within the 20 min of playing the DA game and, thus, it might provide a proxy for the responsiveness to reading intervention. It is noteworthy that we did not find any moderating effect of phonological awareness or rapid naming on the responsiveness to intervention. Research has consistently demonstrated that these factors are important predictors of variance in reading skills (see Melby-Lervåg, Lyster, & Hulme, 2012 and Norton & Wolf, 2012 for reviews). Moreover, some studies did obtain evidence to suggest that these factors can predict responsiveness in children at risk for dyslexia (Al Otaiba & Fuchs, 2002; Nelson et al., 2003 for reviews). The current lack of an association between phonological awareness and rapid naming and response to intervention is in line with the notion that phonological factors may be less important than is often assumed (Byrne, 2011). Observations that, although phonological deficits are common in individuals with dyslexia, a single phonological deficit is not necessary or sufficient to cause the disorder, have led to the idea that poor phonological awareness and rapid naming are two of multiple factors that interact in causing dyslexia (Pennington, 2006; Peterson & Pennington, 2012; Snowling, 2008; Moll, Loff, & Snowling, 2013). Although the current study did not focus on the effectivity of the intervention per se, but rather aimed at gaining insight into factors that can predict responsiveness to intervention, the intervention gains were derived from national normative data rather than from a direct comparison with a control group within a randomized control trial (RCT) design. Our study thus indicates that the DA procedure is able to predict changes between pretest and posttest, but cannot establish whether these changes result from the intervention. However, it is important to add that findings from a recent RCT-study on the effectivity of the LEXY program demonstrated that children with dyslexia showed substantial reading and spelling gains after the intervention and improved at a faster rate than both typical readers and waiting-list controls (Fraga González et al., 2015). Although our findings indicate that our DA is able to significantly predict progress in reading and spelling skills during specialized intervention, with up to 19% of uniquely explained variance, its predictive power is modest from a clinical perspective. It should be noted, however, that, as we focused on Tier 4 intervention, all children within our sample were characterized by severe and persistent reading and spelling disabilities, limiting variability. Based on the results from other studies (for example Cho et al., 2014 and Petersen et al., 2014) it seems plausible that the predictive potential of the DA will increase when applied to Tier 3 or even Tier 2 intervention. The findings of the current study raise a number of interesting questions. First, why is LSIs the best predictor among the three DA variables? 214 S. Aravena et al. / Learning and Individual Differences 49 (2016) 209–215 One possible explanation is that, while LSIa is related to the understanding of the newly learned letter–speech sound correspondences, LSIs, in addition, provides an index of the ability to instrumentally use these correspondences. This interpretation is in line with the literature on dysfunctional letter–speech sound learning, suggesting that the amount of automation of the concerning units at a neuronal level and in identification latencies at a behavioral level, reflects the extent to which the quality of the learned association enables fluent reading (Aravena et al., 2015; Blomert, 2011; van Atteveldt & Ansari, 2014; Widmann, Schröger, Tervaniemi, Pakarinen, & Kujala, 2012). Within this context, LSIs seems to be the purest measure of one's ability to automate the learned associations and it is this ability that may to a large extent determine one's responsiveness to intervention. A second interesting question is why progress in reading speed seems much more difficult to predict than prog...
Purchase answer to see full attachment