Phoneme-Specific Interference in Reading: The Tongue-Twister Effect Revisited

Abstract

IN TWO STUDIES, the authors examine the tongue-twister effect (McCutchen & Perfetti, 1982) in order to help determine the role of phonological information during silent reading. In the first experiment, U.S. college students took longer to judge the semantic acceptability of sentences containing several words with the same initial consonants (tongue-twisters) than of matched control sentences. In addition, when their working memory was pre-loaded with digits whose names repeated the same consonants as occurred in the sentences, a specific phonetic interference was observed. The authors conclude that the tongue-twister effect results from phonetic rather than visual confusion, and that the locus of the effect is within working memory. In a second experiment employing a similar methodology but using a lexical decision task, no tongue-twister effect was found. Thus, the tongue-twister effect appears to occur during the memory and comprehension processes involved in sentence processing, not during processes involved in isolated word reading.