G. Brian Thompson

Learning correspondences between letters and phonemes without explicit instruction

Three studies examined the sources of learning by which children, very early in learning to read, formed correspondences between letters and phonemes when these were not explicitly taught in the whole language instruction they received. There were three classes of predicted knowledge sources: (a) induced sublexical relations (i.e., induction of orthographic–phonological relations from the experience of print words), (b) acrophones from letter names, and (c) transfer from spelling experience. The results of Study 1 indicated that children used both sources (a) and (b). Study 2 results showed that source (a) dominated when the letters were initial components of pseudowords rather than isolated items. The transfer from phoneme-to-grapheme correspondences of the childrens spelling was examined in Study 3. The results were not consistent with the use of source (c). The findings of these studies have implications for the question of how early in learning to read children are able to use knowledge from their experience of print words as a source for phonological recoding.

Is dependence on phonological information in children's reading a product of instructional approach?

Eight-year-old British children were found to be less accurate at rejecting pseudohomophones than ordinary nonwords in a lexical decision task, but 8-year-old New Zealand children did not show this effect. A subsequent homophone decision task showed that this difference was not due to the New Zealand children being unable to distinguish pseudohomophones from other nonwords. The New Zealand children, however, were less accurate than the British children in pronouncing the pseudohomophones they had identified. It was argued that the British children tended to sound out the items before making a choice in the lexical decision task, which gave salience to phonological rather than visual information, resulting in increased errors to the pseudohomophones. It was concluded that where the British children showed this dependency on use of phonological information it was a product of the teaching approach they had experienced.

Learning to read with underdeveloped phonemic awareness but lexicalized phonological recoding: a case study of a 3-year-old.

Case studies on very precocious readers are useful for examining what sources of knowledge and processes are necessary in the acquisition of reading. This is a case study of a 40-month-old child with a word reading age of 8 years 6 months. Tests indicated that she had no phoneme awareness beyond initial phonemes, and that her productive spelling was undeveloped. In reading she was highly proficient at rapid phonological recoding, both by correspondences that were contextually sensitive and those that were not. The former determined her high level of irregular pronunciations for irregular consistent non-words. Experiments indicated that she had well-specified orthographic lexical representations. It was concluded that her phonological recoding was an implicit process based on sublexical relations induced from her lexical representations rather than explicitly taught letter-sound correspondences. The implications of the results for major developmental models of reading acquisition are examined.

A mechanism of implicit lexicalized phonological recoding used concurrently with underdeveloped explicit letter-sound skills in both precocious and normal reading development

These are findings of theoretical interest from: (i) follow-up of a case study of a precocious reader; and (ii) normally developing readers who served as comparison groups. The precocious reader was first reported when 2-3 years of age (Cognition 74 (2000) 177). From 3 to 7 years of age her precocious reading development continued, her word reading accuracy increasing from the 8- to the 16-year-level, although her phonemic awareness skills remained underdeveloped relative to word reading. Nonword reading continued to develop rapidly. Her word reading, however, was more than phonological recoding. At 5 years of age, in comparison with reading-level matched normal 11-year-olds she exhibited strong effects of semantic characteristics of words and evidence of well-specified lexical orthographic representations. In common with normal comparison 11-year-olds, who had not received instruction in explicit phonics, her explicit letter-sound skills were underdeveloped but she possessed high speed and accuracy in nonword reading, a result most theories of the acquisition of reading fail to explain. Her responses to irregularly spelt nonwords indicated higher proficiency than the 11-year-olds in acquiring lexical orthographic representations not predictable from prior phonological recoding knowledge. It is considered that this proficiency contributes to an explanation of her precocious reading development. A mechanism of implicit lexicalized phonological recoding is involved which explains the dissociation of skills in both the precocious reader and normally developing readers.

Are Nonword and Other Phonological Deficits Indicative of a Failed Reading Process

Phonological processing problems have been consideredcritical in explaining developmental readingdisability. Reading disabled children were comparedwith two matched reading-level normal control groupson indicators of phonological processing. The readingdisabled children had lower nonword readingperformance than the phonics taught controls. However, performance was equivalent to that of thecontrols without phonics teaching. Therefore anonword reading deficit was not in itself diagnosticof developmental reading disability. The readingdisabled children and the non-phonics control groupwho exhibited lower nonword reading did not differfrom the phonics taught control group in phonemeawareness, nor in magnitude of the word regularityeffect. Nevertheless, within all groups thosechildren with higher phonemic awareness skills showedlarger word regularity effects and better nonwordreading. Processes involving two sources of knowledgefor phonological recoding are discussed asexplanations of these and many previous results onphonological deficits and of the phonological effectsof phonics instruction.

The nature of skilled adult reading varies with type of instruction in childhood

Does the type of reading instruction experienced during the initial years at school have any continuing effect on the ways in which adults read words? The question has arisen in current discussions about computational models of mature word-reading processes. We tested predicted continuing effects by comparing matched samples of skilled adult readers of English who had received explicit phonics instruction in childhood and those who had not. In responding to nonwords that can receive alternative legitimate pronunciations, those adults having childhood phonics instruction used more regular grapheme-phoneme correspondences that were context free and used fewer vocabulary-based contextually dependent correspondences than did adults who had no phonics instruction. These differences in regularization of naming responses also extended to some low-frequency words. This apparent cognitive footprint of childhood phonics instruction is a phenomenon requiring consideration when researchers attempt to model adult word reading and when they select participants to test the models.

The functions of phonology in the acquisition of reading: lexical and sentence processing.

It has been claimed (V. Coltheart, Laxon, Rickard, & Elton, 1988) that learners as well as skilled readers use phonology for multiple functions in reading-for-meaning tasks. This claim was examined using lexical decision and sentence evaluation tasks. It was found in the first experiment that the type of instruction learners had received determined whether there was prelexical use of phonology in responding to items out of sentence context. Type of instruction had no effect when the items were in context. In the second experiment, performances on a homophone sentence evaluation task and a homophone semantic decision task, which excluded sentence processing, were examined. The results suggest that phonology served the function of access to lexical meanings in addition to any function in postlexical sentence processing. The obtained relationships between relative frequencies of the presented and unpresented homophone mates and item accuracy on these tasks were inconsistent with exclusive use of “direct access” but consistent with access of lexical meanings via phonology and application of a “spelling-check” procedure when multiple homophonic meanings are activated.

Discrimination of mirror-image shapes by young children

Abstract Experiments were conducted to examine the difficulty six-year-old children have in discriminating visually between a shape and its lateral reversal, relative to the difficulty in discriminating between a shape and its inversion. Contrary to results with simultaneous presentation, under successive presentation there was no difference between the difficulty of discriminating lateral reversals and difficulty of discriminating inversions. This was also the case for successive presentation with trial-to-trial variation in position of shapes. Spatial adjacency of the shapes to a reference feature of the visual background accounted for the variations in difficulty between discrimination of lateral reversals and discrimination of inversions which obtained under simultaneous presentation.

The long learning route to abstract letter units

This is a review of theory and evidence on how abstract letter units (ALUs) are initially learnt by the developing individual. Despite the predominance of the lower-case form of letters in the print environment, naming identification of upper case has precedence over lower case among preschool children. Such children showed a significant lag in extending their categories of upper-case variants to include the corresponding lower-case forms that are visually dissimilar. As late as 11 years of age children gave longer naming latencies for the lower-case than the upper-case forms. Initial learning of ALUs proceeded slowly over many months, consistent with the “common contexts” hypothesis but not consistent with the early acquisition predicted by the “common letter name” hypothesis. Evidence from cross-case transfer in a training experiment indicated that prior to acquiring full use of ALUs the children had formed representations of words that were letter based but specific to lower-case forms.

Discovering and accounting for limitations in applications of theories of word reading acquisition

More attention to the discovery of the limitations of current theories of word reading acquisition would enable progress in development of theories with a wider and more varied range of valid and useful applications. This general opinion is illustrated here with work that makes such an attempt. There have been recent occasional attempts to apply computational connectionist models of adult word reading in simulations of childrens normal progress in word and pseudoword reading (Hutzler et al., 2004; Powell et al., 2006), but they failed unless modifications were made that included adding “context-free” letter-sound correspondences to the initial training of the model. Taught phonics sounds for letters are such, as they are not bound to features within a word, such as position and/or the context of other (adjacent or otherwise) letter-sound correspondences of the word. Adding the phonic sounds was justified as a representation of the way children learnt because it was how they were taught reading. This introduces a major potential limitation in application of the model, in so far as it was improved for only one type of teaching, that with phonics. A new multiple-route theory (Grainger et al., 2012) of learning to read words has been proposed which may appear to avoid that problem but much of the learning of the beginner reader is modeled as in the theory of Share (1995). This, however, also requires full knowledge of “context-free” letter sounds for the initial development of word reading (Share, p. 164), as does the widely recognized theory of Ehri (1999, 2005, 2012). The illustration for my opinion focuses on the discovery of the limitations of this feature common to these theories, and on development of theory that accounts for evidence beyond the limitations.