Phonological Processing in Printed Word Recognition: Effects of Age and Writing System

Abstract

Do readers of orthographies in which there is a simple relation between spelling and sound (transparent orthographies) depend more on decoding for word recognition than do readers of opaque orthographies? Participants named printed target words that were preceded (at 300 msec) by a spoken pseudoword that either rhymed or did not rhyme with the target. To the extent that readers depend on decoding (assembled phonology) to recognize the target word, preceding that assembly process with a spoken rhyme ought to facilitate it. A transparent orthography (Turkish), in which each letter has only one pronunciation, was compared to a more opaque one (English). Three ages of participants were studied: second- and fifth-grade children and adults. Rhyme had a stronger effect in Turkish than in English and a stronger effect on younger than on older readers. A second experiment indicated that the difference between languages was not likely to have been an artifact of the proportion of rhymes used. The results support previous work suggesting that orthographic transparency is a determiner of the degree to which readers use phonology during word recognition. This study suggests that readers become less dependent on phonological mediation with experience and that this reduction is more rapid for readers of opaque orthographies.