Single or dual orthographic representations for reading and spelling? A study of Italian dyslexic–dysgraphic and normal children

Abstract

Italian children with surface dyslexia and dysgraphia show defective orthographic lexical processing in both reading and spelling. It is unclear whether this parallelism is due to impairment of separate orthographic input and output lexicons or to a unique defective lexicon. The main aim of the present study was to compare the single- versus dual-lexicon accounts in dyslexic/dysgraphic children (and in normal but younger children). In the first experiment, 9 Italian children with surface dyslexia and dysgraphia judged the orthographic correctness (input lexicon) of their phonologically plausible misspellings (output lexicon) and of phonologically plausible spellings experimentally introduced for words they consistently spelt correctly. The children were generally impaired in recognizing phonologically plausible misspellings. Parallel deficits in reading and spelling also emerged: Children were more impaired in judging items they consistently misspelt and more accurate in judging items they always spelt correctly. In a second experiment, younger normal children with reading/spelling ability similar to that of the dyslexic/dysgraphic children in the first experiment (but younger) were examined. The results confirmed a close parallelism between the orthographic lexical representations used for reading and spelling. Overall, findings support the hypothesis that a single orthographic lexicon is responsible for reading and spelling performance in both dyslexic/dysgraphic and normal (but younger) children.

Keywords:

• Development
• Dyslexia
• Spelling
• Orthography
• Lexicon

We would like to thank Professor Brenda Rapp for permission to reproduce Figure 1 and for helpful suggestions on a preliminary version of this paper. We would also like to thank Claudio Luzzatti for his precious advice for implementing the experimental task.

Notes

¹ The asterisk indicates nonlexical spelling errors.

² Ambiguities in (c) and (d) are related to the pronunciation of specific linguistic areas. Furthermore, in these cases, the incorrect orthographic versions are phonologically close to the targets but are not true homophones.

³ Only the syllable [kw] can be transcribed using four different orthographic solutions (i.e., CU, QU, CQU, and QQU). In the other cases, there are only two phonologically plausible spellings, and only one is correct.

⁴ Gough, Juel, and Griffith (1992) found that inconsistent responses on successive reading and spelling tests were frequent in novice readers and argued that the discrepant responses in reading and spelling the same stimuli in beginning reader–spellers could reflect inconsistency in performance rather than a lack of association between reading and spelling performance. In this framework, analysis of the children's performance in judging inconsistent misspellings could be informative.