The Science of Reading. Группа авторов

they argue that sublexical learning takes precedence early in development (at a time when the child must learn the relationships between print and sound), and lexical processes gain precedence later in development with increasing exposure to print. Key to the acquisition of expertise is the process of statistical learning – while explicit instruction in phonics encourages the development of sublexical processes, rich and varied reading experience supports orthographic learning more implicitly. They discuss interactions between sublexical and lexical knowledge, and describe the task of learning to read as one in which the reading system comes to embody the structure of the writing system, paving the way to the markers of skilled word recognition discussed in the first part of this handbook.

      There are synergies here with what Ehri (2005) referred to as the development of “sight‐word reading” in the first edition of this book. She proposed that following a stage of early reading when children are beginning to use the alphabetic principle, development is a process of assimilating mappings between orthographic and phonological word forms and meaning in a single memory system. Castles and Nation’s model does not distinguish stages of development, but rather argues for item‐level developmental interactions between sublexical and lexical systems during a protracted phase of learning with a “lexical” tuning process ensuring the adaptation of the reading system for the orthography it is learning, as children’s reading experience increases.

Learning to spell is arguably a more daunting task than learning to read for most children. Complementing Castles and Nation’s discussion, Kemp and Treiman describe how children progress from drawing on rudimentary knowledge of printed forms, to being able to use a precise lexical representation of all the symbols, in the right sequence, to spell proficiently. In the initial phases, spelling is pre‐phonological although young children quickly begin to demonstrate sensitivity to some of the graphotactic conventions of their writing system, as they begin to build lexical knowledge and phonological spelling skills. However, learning to spell well, like acquiring reading fluency, takes time and even when a child has mastered the ability to spell with phonological accuracy, there is much still to learn about orthographic patterns and morphological structure (and in the digital era, about contractions and emojis that comprise the “textese” used to convey messages). Drawing together the multiple sources of knowledge that underpin fluent spelling in the Integration of Multiple Patterns framework, they argue for a strong role of statistical learning alongside explicit teaching in learning to spell, and draw out implications for education.

      Like Bowey (2005) in the first edition, Compton, Steacy, Petscher, Rigobon, Edwards, and Gutiérrez are concerned with the predictors of individual differences in word reading. In a review of these across the age range, they describe contemporary analytical tools (including dominance analysis and quantile regression) that allow a more nuanced analysis of sources of variation than before. They proceed to discuss the varied use of longitudinal data to reveal how trajectories in one domain of literacy may affect another domain, how gains in word reading predict gains in reading comprehension, for example. They also consider data from intervention studies in the context of longitudinal designs and note that these elucidate the mechanisms that underpin effective intervention. Such methodological approaches are important as they can inform the design of assessment measures of reading and its subskills and throw light on the recurrent question of “What works and for whom?”. Compton et al.’s review dovetails nicely with the other chapters in this section that emphasize the interaction between sources of variance that are extrinsic and intrinsic to the child and drive the statistical learning that fuels literacy development. Their review also highlights the utility of child‐level, word‐level, and child‐by‐word predictors for answering questions about how children learn to read words.

      Savage takes up many of these issues when considering how to teach children to read as Torgesen (2005) and Snow and Juel (2005) did in the first edition. Like Compton et al., Savage argues that randomized controlled trials (RCTs) represent the gold standard of evidence that an intervention is effective, providing data that can be used to confirm a causal hypothesis (such as the claim that phoneme awareness is causally related to word reading skill). However, he also highlights the limitations of some meta‐analytic approaches in which data from disparate studies are combined and moderating factors such as the size of the intervention group (small group or whole class), the fidelity of delivery, or the professional training of the teacher are not taken into account. Nevertheless, this critical review shows that there are educationally significant effects of whole class phonics instruction on reading attainment, and that these effects are substantially larger when combined with wider language support and tutoring approaches. Consistent with the child by word‐level interactions discussed in earlier chapters, not all children require tutoring in phonics.

      The evidence is less robust in relation to how best to support children with reading difficulties or dyslexia; broadly, phonic approaches are favored but many studies confound reader ability with delivery mode (small group for struggling reader vs. whole class for more typical reader) and there is danger in generalizing from typical readers to poor readers. Similarly, there is some limited evidence for the efficacy of morphological interventions in the teaching of reading, but no consensus yet on what to teach and when. What is clear is that there is not a “one‐size‐fits‐all” approach and there is considerable room for the use of well‐designed “Assessment to Intervention” approaches that can specify the optimal blend of “code‐” and “meaning‐” focused instruction to secure the foundations for reading with understanding.

      References

      1 Bowey, J. A. (2005). Predicting individual differences in learning to read. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 155–172). Blackwell Publishing. https://doi.org/10.1002/9780470757642.ch8.

      2 Byrne, B. (2005). Theories of learning to read. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 104–119). Blackwell Publishing. https://doi.org/10.1002/9780470757642.ch6.

      3 Ehri, L. C. (2005). Development of sight word reading: Phases and findings. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 135–154). Blackwell Publishing. https://doi.org/10.1002/9780470757642.ch8.

      4 Mattingly, I. G. (1972). Reading, the linguistic process, and linguistic awareness. In J. F. Kavanagh & I. G. Mattingly (Eds.), Language by ear and by eye: The relationship between speech and reading. Massachusetts Inst. of Technology P.

      5 Phillips, B. M., & Lonigan, C. J., (2005). Social correlates of emergent literacy. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 173–187). Blackwell Publishing. https://doi.org/10.1002/9780470757642.ch7.

      6 Snow, C. E., & Juel, C. (2005). Teaching children to read: What do we know about how to do it? In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 501–520). Blackwell Publishing. https://doi.org/10.1002/9780470757642.ch26.

      7 Torgesen, J. K. (2005). Recent discoveries on remedial interventions for children with dyslexia. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 521–537). Blackwell Publishing. https://doi.org/10.1002/9780470757642.ch27.

      8 Treiman, R., & Kessler, B. (2005). Writing systems and spelling development. In M. J. Snowling & C. Hulme (Eds.), The science of reading: A handbook (pp. 120–134). Blackwell Publishing. https://doi.org/10.1002/9780470757642.ch7.