Understanding Oral Spelling: A Review and Synthesis

Abstract

This article reviews a number of studies that report discrepancies between written and oral spelling that cannot easily be explained in terms of deficits to modality-specific output routines. For instance, some patients appear to be ‘surface dysgraphic’ for oral spelling but ‘phonological dysgraphic’ for written spelling (e.g. Lesser, 1990). A model is proposed to account for these findings which contains the basic premise that the mappings between phonemes and letter names (in oral spelling) is generally non-arbitrary, whereas the mappings between phonemes and letter shapes (in written spelling) is entirely arbitrary. For example, the phoneme /t/ sounds similar to the oral letter name “Tee” but is arbitrarily related to the written letter ‘T’ or ‘t’. An important consequence of this is that oral spelling, in normal spellers, may be more biased by phonological correspondences than written spelling. Other discrepancies between written and oral spelling are reviewed and accommodated within this model, and the model is extended to include recognition of oral spellings and transcoding between spoken and written letter forms.