Abstract
Current theories of reading development (e.g. Frith, 1985) propose an initial logographic phase in which printed word recognition is based on salient visual features of the print. An alternative proposition (Stuart and Coltheart, 1988) is that phonologically skilled beginning readers might be able to set up word recognition units from their phonological analyses of spoken words, and their knowledge of correspondences between sounds and letters. Neither view is supported by the present study. It is shown in a series of experiments that phonological knowledge is indeed helpful even to pre-reading children asked to discriminate printed targets from distractors in an auditory-visual matching task. However, this effect occurs for non-words as well as words and pseudohomophones, suggesting the operation of a non-lexical rule-based system rather than the establishment of word recognition units.

This publication has 1 reference indexed in Scilit: