(Extract from the doctoral thesis of Vincent Goetry, translated from the French)

As we have often said, phonemic awareness is essential to the acquisition of reading. This capacity lays down a system that generates the translation from writing to pronunciation thanks to the associations which progressively develop an automatic phonological lexicon (Alegria et al, 1994). This system plays a major, dynamic role in learning to read.

Share and colleagues have suggested a mechanism which could explain this process (Jorm & Share, 1983; Jorm, Share, McLean & Matthews, 1984; Share, 1995, 1999). Phonological recoding works as a self-teaching mechanism allowing the child (and adult) to acquire more and more detailed orthographic representations of the words they meet. The concept of self-teaching implies that every accurate decoding of an unfamiliar word not only reinforces the decoding system, making it increasingly automatic and sophisticated, but also plays a part in progressively enhancing the orthographic code for the same word. This allows the same word met later to be accessed directly.

Phonological recoding as a self-teaching function has three essential facets. First, auto-instruction of orthographic representations takes place word by word and not in stages. Put another way, the method of treating each word depends at the same time on the frequency of exposure to this word as well as the successful outcome of the previous phonological decoding of the same word. Fast access to the orthographic information embodied in each specific word (Manis, 1985, Reitsma, 1983, Share, 1999) allows direct access, with minimal phonological processing, from the earliest stages of learning to read.

Second, this concept implies that the process of phonological recoding progressively ‘self-lexicalizes’ as the child learns to read. When starting, the child can only carry out a partial recoding on the basis of a limited set of available grapheme-phoneme correspondences (see for example Ehri & Wilce, 1985; Ehri & Sweet, 1991). Rudimentary associations will be progressively enriched and modified in the light of the lexical constraints imposed by the development of orthographic awareness, and this will promote the establishment of the rules for association integrating contextual, positional and morphological limitations. This automatic processing means in fact that the fluent reader is no longer consciously aware of the sub-lexical relationships between graphemes and phonemes when reading words by direct access.

Share (1995) also insists that self-teaching by phonological recoding applies equally to irregular words. In fact most irregular words present a sufficient amount of regular correspondences, especially for consonants, to enable the best choice amongst the set of possible pronunciations, above all when the word is presented in context, which is generally the case in schools. Therefore context plays an essential role.

Third, phonological decoding is a condition sine qua non for all children learning to read. The development of orthographic representations depends first and foremost on the efficient performance of the decoding system. Each different word has an important and individual bearing on the capacity to store and recover its orthographic representations, and this will influence the quality of its representations and the speed of recovering them. Even so, in view of the mechanism for self-teaching, visuo-orthographic treatment is thought of ‘not as a second source of variance, but as a secondary source of individual differences in the acquisition of literacy’ (Share, 1995, p. 156). 

The theory of self-teaching allows the important prediction that the failure to establish efficient mechanisms for decoding will compromise the entire system, including orthographic representation. Numerous studies support this prediction. For example Jorm et al. (1984) divided first year primary pupils into two groups: ‘good decoders’ and ‘poor decoders’ according to the results of reading pseudo-words (which cannot be treated by lexical access). The two groups were equal as to age, sex, verbal intelligence, type of school and reading age. A longitudinal study of reading ability after two years showed that the ‘good decoders’ were nine months ahead of the ‘poor decoders’. In agreement with the hypothesis of self-teaching, the first group had developed better reading skills than the second group.

(See also Share, 1999, for a more direct test of the hypothesis.)

References

Alegria, J., Leybaert, J., & Mousty, P. (1994). Acquisition de la lecture et troubles associés : évaluation, remédiation et théorie. In J. Grégoire & B. Piérart (Eds.), Evaluer les troubles de la lecture : les nouveaux modèles théoriques et leurs implications diagnostiques (pp. 105-126). Bruxelles: De Boeck-Wesmael s. a.
Baron, J. (1977). Mechanisms for pronounciation printed words: Used and acquisition. In D. L. S. J. Samuels (Ed.), Basic processes in reading: perception and comprehension. Hillsdale, NJ: Erlbaum.
Byrne, B., Freebody, P., & Gates, A. (1992). Longitudinal data on the relations of word-reading strategies to comprehension, reading time, and phonemic awareness. Reading Research Quarterly, 27, 141-151.
Ehri, L., & Sweet, J. (1991). Fingerpoint-reading of memorised text: What enables beginners to process the print? Reading Research Quarterly, 26, 442-462.
Ehri, L., & Wilce, L. (1985). Movement into reading: Is the first stage of printed word learning visual or phonetic? Reading Research Quarterly, 20, 163-179.
Jorm, A., & Share, D. (1983). Phonological recoding and working memory. Applied Psycholinguistics, 4, 103-147.
Jorm, A., Share, D. L., MacLean, R., & Matthews, R. (1984). Phonological recoding skills and learning to read: A longitudinal study. Applied Psycholinguistics, 5, 201-207.
Manis, F. R. (1985). Acquisition of word identification skills in normal and disabled readers. Journal of Educational Psychology, 77, 78-90.
Reitsma, P. (1983). Printed word learning in beginning readers. Journal of Experimental Child Psychology, 36, 321-339.
Share, D. L. (1995). Phonological recoding and self-teaching: sine qua non of reading acquisition. Cognition, 24, 139-168.
Share, D. L. (1999). Phonological recoding and orthographic learning: A direct test of the self-teaching hypothesis. Journal of Experimental Child Psychology, 72, 95-129.