Basic processes in reading: Two phonological codes.

Publisher Website
Google Scholar
Abstract
The phonological code that can be used to gain lexical access from print is assumed to be identical to that which can be used for the maintenance of information in short term memory. Two immediate memory span experiments were carried out which demonstrated that these phonological codes are dissociable. In experiment I subjects viewed letter strings which varied in phonological similarity and lexical status. Under a no interference condition, phonologically distinct lists were better recalled than phonologically confusable ones, and lists with entries in phonological lexicon (e.g., BRANE) were better recalled than lists without lexical status (e.g., SLINT). When articulation of the test items was prevented by requiring subjects to articulate irrelevent sounds, the phonological similarity effect was completely eliminated, but a lexicality effect persisted. In experiment II subjects viewed letter strings which varied in syllable length and lexical status. Pseudohomophones were better recalled than control nonwords under both quiet and articulation conditions, but a syllable length effect was obtained only in the no articulation condition. Apparently, at least 2 phonological codes underly performance in a memory span task. The 1st of these codes permits lexical access from print, suppression does not prevent this code from accessing lexical memory. The 2nd code underlies both word length and phonological similarity effects in span; suppression prevents either the formation or utilization of this code. Implications for understanding normal reading and developmental reading problems are noted.

This publication has 6 references indexed in Scilit: