Orthographic input coding: A review of behavioural evidence and current models

Successful identification of a printed word requires the reader to construct an orthographic input code-a representation that encodes both the identity and the order of the letters in the input stimulus. There is now a very considerable body of evidence relating to orthographic input coding. In this chapter, I review and synthesize this evidence in order to establish a set of behavioural criteria that a successful model of orthographic input coding must satisfy. I then assess the ability of existing models to satisfy these criteria. None of these models succeed in satisfying all of the criteria, and I will argue that this is due to fundamental problems with the way in which letter order is coded in these models. In the final part of the chapter, I very briefly review an alternative approach to orthographic input coding, which may offer a more promising approach (Davis, submitted, 1999).