ABSTRACT
Neuropsychologists began to get truly interested in spelling in the 1980s. Until then, the dominant position was that the production of written language1 was essentially dependent on the mechanisms responsible for the production of oral language. Consequently, the disorders of written production-the agraphias2-were mainly viewed as being secondary to the deficits affecting oral language-the aphasias-resulting themselves from lesions of the left hemisphere. Nonfluent agraphias were usually associated with frontal lesions and fluent agraphias with more posterior lesion sites (Benson & Cummings, 1985; Hécaen, Angelergues, & Douzenis, 1963). However, even if most authors admitted that the intensity and the types of disorders affecting the oral and the written domains could vary (Marcie & Hécaen, 1979), the existence of selective deficits of written production, called pure agraphias, was difficult to conciliate with this perspective. That is why some authors, following Exner (1881), hypothesized the existence
1The production of written language requires a set of pragmatic, syntactical, semantic, and lexical processes as well as perceptuomotor mechanisms allowing the movements necessary for the production of letters in the graphic space of the page. We limit ourselves to considering the processes responsible for the generation of strings of graphemes forming words or pseudowords.
