Word Recognition in Reading (Tutorial Paper)

Abstract

This chapter offers a critical survey of evidence for functionally separable mechanisms in word recognition, and a theoretical framework for interpreting the contributions of linguistic, especially lexical, knowledge in both pre-attentive processing and conscious perception of written words. Section 1 reviews evidence for independent parallel systems responsible for graphemic access to semantic and conceptual knowledge, and for conversion of alphabetic script into a phonological representation. The role of phonological coding in silent reading is discussed, including its role in interpreting certain syntactic components of text. Section 2 discusses the effects of orthographic regularity and lexicality in the identification of words and letter strings. Evidence is summarized favouring an intermediate, tacit stage of letter identification. Contrasts between the recognition of words and of pictures are explored. Section 3 introduces a procedural model for the mental lexicon, illustrated by means of the MARGIE system, in relation to both referential and syntagmatic word-meanings. It emphasizes the role in reading of pre-attentive expectations (contextual facilitation effects) generated by parallel, production-like mechanisms. Evidence of preattentive lexical access is surveyed. In Section 4 theoretical issues concerning perceptual integration and its central place in the mechanism of word perception are raised. Some models of the integration process are also reviewed. Some of the best camouflaged conceptual traps for psychologists lie in the way ordinary language is used to perform a quasi-technical role in the description of behaviour. “Word recognition” is full of such traps. “Recognition,” like “perception,” appears to refer both to the “discriminative uptake” of information from the senses and (in the adult human) to the ability to report or comment upon it, in answer to the question “What did you perceive?” Yet there are many illustrations of the dissociation between these two abilities. One striking example is the phenomenon that Weiskrantz and colleagues have termed “blindsight” (Weiskrantz, Warrington, Sanders, & Marshall, 1974). Although subjectively blind in one quadrant of the visual field after surgical removal of a major portion of the calcarine visual cortex, one patient could reach with great accuracy to small visual objects and make crude shape-discriminations in the “blind” region. Throughout testing the patient remained convinced that he was guessing, and denied any awareness of “seeing” the stimuli.