Speech and spelling interaction: the interdependence of visual and auditory word recognition

Abstract

AbstractA critical feature of interactive networks is the bidirectional flow of activation. In an interactive model, cross-code consistency is crucial. The phonological coherence constraint puts a major emphasis on the correlations between orthographic and phonological subunits of the system. This framework makes a number of intriguing predictions with regard to spelling-to-phonology interactions. For example, it predicts that phonology should routinely be involved in visual word recognition, and that inconsistency in the mapping between orthography and phonology should hurt word recognition. Moreover, the framework predicts that inconsistency should matter not only between spelling and phonology but also between phonology and spelling (“feedback consistency”). This article summarizes the empirical evidence in favor of these predictions, reviews the behavioral data regarding speech and spelling interactions, and discusses the extent to which current computational models of word recognition are able to account for these effects. It also considers the influence of orthography on spoken language, and how print and speech interact online during word recognition.