ERC Advanced Grant
Cracking the Orthographic Code
The proposed research is designed to test a new theory of orthographic processing - that is a theory about how information concerning letter identity and letter position is encoded during reading. The theory is couched within a general framework for word recognition that makes a critical distinction between a coarse-grained orthographic code that provides a fast-track to semantics, and a fine-grained orthographic code that is used to generate a prelexical phonological code, hence providing the connection with auditory word processing. The project is divided into three sections, each examining a specific component of this theoretical framework. Section 1 examines low-level visual constraints on the earliest phase of orthographic processing - the retinotopic mapping of visual features onto letter identities. Section 2 examines how this preliminary orthographic information can be most efficiently used to constrain lexical identity via a coarse-grained, word-centered orthographic code. Section 3 examines a further critical constraint on orthographic processing - the fact that the orthographic system is grafted onto a pre-existing phonological system during the course of reading acquisition. Each section of the research program will combine the methods of experimental cognitive psychology with visual psychophysics, brain imaging (ERPs and MEG), and computational modeling. This multi-methodological approach is one of the keys to success of the present project, along with the strong theory-driven nature of the proposed research. It is this unique combination that is expected to generate the breakthroughs that will provide the foundations for a general account of skilled reading and its breakdown in reading disabled persons.