ABSTRACT

The experiments reported in this chapter tested whether phonologic mediation affects proofreading and, by implication, whether phonologic mediation affects reading generally. Printed words activate phonologic codes 1 , and these phonologic codes are primary constraints in word identification. This phonologic mediation hypothesis is hotly debated, and variants of this hypothesis span the history of the psychology of reading. Recently, my colleagues and I proposed a new phonologic mediation hypothesis which we called the phonologic coherence hypothesis (Van Orden, Pennington, & Stone, 1989). This hypothesis was motivated by computational assumptions common to subsymbolic connectionism (e.g., see Smolensky, 1986a ; 1986b; 1988).