Surface dyslexia and surface dysgraphia: Treatment studies and their theoretical implications

We report studies of a patient with acquired surface dyslexia and dysgraphia. His reading impairment was treated using the methods previously reported to be successful for surface dyslexia by Byng and Coltheart (1986); these methods were also successful with our patient. In addition, we observed, as Byng and Coltheart did, that when these methods are used there is incomplete but significant generalisation to the reading of untreated words. Connectionist simulations of the effects of damage to the language-processing system have taken this generalisation effect to be evidence that words are represented in a distributed fashion in that system; we challenge this inference, on the basis of data from our and Byng and Coltheart's patient. Our patient's spelling impairment was treated using the methods previously reported to be successful for surface dysgraphia by Behrmann (1987); these methods were also successful for our patient, and we found in addition, as did Behrmann, that the treatment effects upon spelling did not generalise to the spelling of untreated words. We consider the implications of our results for the theoretical issue of whether lexical reading and lexical spelling depend upon a common orthographic lexicon or upon separate input and output orthographic lexicons. Although our results were not entirely unequivocal in relaton to this issue, we interpret the evidence from our patient, and from analyses we report of data from the surface dyslexic and surface dysgraphic patient described by Behrmann and Bub (1992), as favouring the two-lexicon view.