Current Research Informing the Conceptualization,Identification,and Treatment of Dyslexia Across Orthographies: An Introduction to the Special Series

Genetic Basis for Dyslexia

Dyslexia is indeed highly heritable (Defries & Alarcón, 1996; Pennington et al., 1991; Snowling & Melby-Lervåg, 2016). When dyslexia is defined as word reading accuracy more than 1.5 SD below the mean, estimates of its prevalence in the population are roughly 7% (Peterson & Pennington, 2012); among children with a parent with dyslexia, however, prevalence estimates according to this same definition are closer to 35% to 45% (Pennington et al., 1991; Snowling & Melby-Lervåg, 2016). Although research has not reliably identified specific candidate genes, there are robust findings identifying nine regions of the genome that contain genes or genetic material associated with individual differences in word reading (Grigorenko et al., 2019).

Brain Basis for Dyslexia

Brain structure and patterns of brain activation also differ for children with dyslexia relative to their non-dyslexic peers. Children with dyslexia have reduced volume of gray matter and/or integrity of white matter pathways in regions of the brain that are activated during reading (Gabrieli, 2009; Seidenberg, 2017) even when they are in preschool (Im et al., 2016; Raschle et al., 2017) and perhaps as early as a few days after birth (Molfese, 2000). When children undergo brain scans while engaged in reading or reading-related activities, there are differences in degree and pattern of brain activation that correspond to word reading ability (Gabrieli, 2009; Kearns et al., 2019; Verhoeven et al., 2019), with atypical activation patterns sometimes normalizing after effective intervention (Barquero et al., 2014).

Early Behavioral Indicators of Dyslexia

Finally, a large proportion of children at family risk for dyslexia manifest observable behavioral symptoms long before reading instruction begins (e.g., word pronunciation accuracy and length/complexity of sentence production in early childhood; for example, Lyytinen et al., 2004; Richardson et al., 2009), in addition to demonstrating difficulties with phonological awareness and phonemic decoding after the onset of early reading instruction. In light of this, extensive research that supports the genetic, neural, and cognitive bases of dyslexia while also suggesting the presence of environmental influences on development, it becomes clear that our energies are best spent debating the best ways to identify and treat individuals with dyslexia rather than debating the validity of the construct (Elliott & Grigorenko, 2014).

Issues Related to Identification and Treatment of Dyslexia

There are numerous controversies related to dyslexia identification. Research has long demonstrated lack of evidence for the validity of an IQ-achievement discrepancy approach to identification (Stuebing et al., 2002, 2015). There is also lack of evidence for the validity of approaches that involve administering a comprehensive battery of cognitive processing assessments and identifying uneven patterns of strengths and weaknesses in achievement across domains (Miciak et al., 2014; Taylor et al., 2017). In addition, response to intervention (RtI) approaches can be problematic for identification because of poor agreement between different methods and measures used to identify inadequate instructional response (Fuchs et al., 2008; Schatschneider et al., 2016) and because of challenges schools face in providing evidence-based reading instruction and intervention (Balu et al., 2015). There is emerging consensus that it may be most appropriate to use multiple indicators for the identification of dyslexia, including academic underachievement and inadequate instructional response and while considering exclusionary factors (Bradley et al., 2002; Fletcher et al., 2018; Wagner, 2018).

Treatment of dyslexia is also not without controversy. Dyslexia has had a rather checkered history of fads and other quick fixes that have diverted attention from legitimate instructional practices that are necessary to support word reading acquisition for individuals with learning difficulties in this domain. Even today, there exist products that claim to improve word reading ability through optometric training, colored overlays or lenses, or cognitive exercises that target cognitive processing in the absence of letter or word reading, despite lack of evidence that they are effective (Melby-Lervåg et al., 2016; Pennington, 2009). In addition, there continues to be a tendency to misunderstand learning to read as a “natural” process much like learning language (Seidenberg, 2017), rather than a radical rewiring of the brain’s circuitry to support a relatively recent human activity to which the brain is not well adapted (i.e., reading/writing developed about 5,400 years ago, which is recent in the context of the history of human speech; Perreault & Mathew, 2012). Misunderstanding reading to be a natural process to which the brain is well adapted can lead to resistance in adopting explicit modes of instruction.

This Special Series

The first three articles in this special series focus on studies with implications for the identification and treatment of dyslexia. In the first article, Jack Fletcher et al. describe the process by which they developed and validated a series of short, teacher-administered dyslexia screeners in the domains of phonological awareness (Kindergarten, Grade 1), letter-sound knowledge (Kindergarten, Grade 1), and word reading (Grades 1 and 2). Authors prioritized the minimization of “false negatives” (i.e., students with dyslexia that the screener fails to identify), even if this necessitated increased identification of “false positives” (i.e., children who may eventually have performed within the normal range even without receiving intervention). They also differentiate between dyslexia screening and diagnosis in a way that is important given current confusion that exists related to this distinction. The detailed way in which authors describe the development, scaling, and validation process makes this article instructive for others who are currently endeavoring to develop dyslexia screeners in response to state legislation mandating the use of such screeners and to ensure efficient identification and effective treatment of children with dyslexia in public schools. Such a screener will be a critical component of any multi-indicator approach to identifying dyslexia and will inform decisions that leverage the intensity of instruction to meet children’s needs.

In the second article in this special series, Laura Steacy and her colleagues explore the effect of training corpora on word reading accuracy in children with dyslexia. Their findings suggest that children with dyslexia, unlike their typically developing peers, fail to benefit from corpus feedback (i.e., they fail to correctly infer word pronunciations for new words based on exposures to words with similar orthographic-phonological sublexical components). Authors describe the ways in which their findings align with the hypothesis that children with dyslexia are more likely to form whole-word phonological representations rather than developing word representations that prioritize sublexical orthographic-phonological correspondences (Harm et al., 2003). Their results suggest that dyslexic children need extended support to develop knowledge of sublexical orthographic-phonological correspondences that facilitate generalization, thus enabling students to benefit from exposures to training corpora that tune vowel pronunciations.

The third article, by Carolyn Denton et al., report results of an initial evaluation of a reading intervention (Idea Detectives) for elementary-grade students with word reading difficulties, including students identified with dyslexia. The intervention included word study (i.e., phonemic awareness, phonemic decoding, high frequency word recognition, and spelling), text reading, comprehension (both text-based questioning and metacognitive reading comprehension strategy instruction), and self-regulation training (e.g., training in the use of positive self-talk informed by a growth mind-set, goal-setting, and self-monitoring of strategy use) components. Although participating teachers perceived a great need to incorporate self-regulation training into instruction, it proved difficult to develop instructional materials targeting self-regulation that were feasible to deliver in typical school settings. Authors discuss potential reasons for their finding that there were no significant differences at posttest between the group that received Idea Detectives and the group that received business-as-usual instruction.

The next three articles address the degree to which different component skills of reading contribute to dyslexia, and the extent to which there are distinct profiles of students with reading difficulties, both within samples of students learning to read in English and within samples of students learning to read in more transparent orthographies. The first article, by Capin et al., uses latent profile analysis to identify three subtypes of English-speaking Grade 4 students with significant reading comprehension problems. The vast majority of students fit into a profile that experienced moderate difficulties with both word reading and listening comprehension. A smaller proportion (5%) had severe word reading difficulties paired with moderate difficulties with listening comprehension, and an even smaller number (4%) had severe difficulties with listening comprehension paired with moderate difficulties with word reading. Authors’ findings suggest that, even beyond Grade 3, students with reading comprehension difficulties very often experience word reading difficulties. Authors cite research suggesting that a large number of interventions designed for students in Grades 4 and above provide inadequate word reading instruction, which is problematic in light of their results. Capin and colleagues also report on analyses that investigated the degree to which cognitive processing skills predicted group membership within latent classes.

In the second article, Aehwa Kim and colleagues ask similar questions to the ones posed by Capin et al., but for a sample of elementary-school students learning to read in Korean. Authors use latent profile analysis to classify students into subgroups based on (a) both reading achievement measures and (b) measures of cognitive–linguistic processing skills that contribute to reading; they also investigate the relationship between reading achievement profiles and cognitive–linguistic profiles. Authors identified four subtypes of native Korean readers on the basis of reading achievement and four subtypes on the basis of cognitive–linguistic skills. Reading achievement subgroups were associated with different patterns of cognitive–linguistic strengths and weaknesses. Most students who were found to have word recognition difficulties also demonstrated significant fluency and comprehension difficulties. Little is known about learning to read in alphasyllabary languages such as Korean (Nag et al., 2011) and there are few studies that endeavor to identify subtypes of poor readers in transparent or semitransparent orthographies more broadly. For this reason, the findings reported by Kim et al. are noteworthy contributions to the literature.

The third article, by Tatiana Logvinenko and colleagues, provides an overview of Russian efforts to address the problem of reading difficulties in school-aged students across the last century. It also investigates the component skills (e.g., phonemic awareness, decoding, pseudoword repetition, rapid naming, and lexical decision-making) that contribute to reading difficulties in Russian. Again, due to the lack of research documenting the nature of reading difficulties in transparent or semitransparent orthographies in general and in Russian more specifically, this study is an important contributionto the field. Lessons gleaned both from the historical account and from the empirical findings are transferable to different nations/orthographies. We interpret these studies as contributing to a growing body of research improving our knowledge about screening, identification, and treatment of students with dyslexia in English-speaking countries and around the world.

The final article, a commentary by Donald L. Compton, examines dyslexia through a multifactorial lens. That is, multifactorial models of dyslexia serve as the framework for examining issues of etiology, identification, and instruction in dyslexia. These models are based on the idea that cognitive/linguistic risk factors are not deterministic but instead probabilistic. Difficulties may stem from an interaction between risk and protective factors, which include exogenous and endogenous influences.