The effect of word length in short-term memory: Is rehearsal necessary? Guillermo Campoy
University of Murcia, Spain
Three experiments investigated the effect of word length on a serial recognition task when rehearsal was prevented by a high presentation rate with no delay between study and test lists. Results showed that lists of short four-phoneme words were better recognized than lists of long six-phoneme words. Moreover, this effect was equivalent to that observed in conditions in which there was a delay between lists, thereby making rehearsal possible in the interval. These ﬁndings imply that rehearsal does not play a central role in the origin of the word length effect. An alternative explanation based on differences in the degree of retroactive interference generated by long and short words is proposed.
The capacity to retain phonological information in the short term is an important and intriguing characteristic of human memory. A satisfactory understanding of the mechanisms underlying this ability depends on our capacity to explain phenomena such as the word-length effect: the ﬁnding that immediate recall is better for lists of short words than for lists of long words (Baddeley, Thomson, & Buchanan, 1975). The most comprehensive explanation of the word-length effect is that offered within the phonological loop model (Baddeley, 1986; Burgess & Hitch, 1999). According to this model, verbal information is kept in a phonological store, but decays in a few seconds unless it is reactivated by a subvocal
rehearsal process. The word-length effect occurs because rehearsal is faster with short than with long words and therefore more efﬁcient in counteracting the consequences of decay between presentation and recall. A crucial factor in this account is the ﬁnding that the effect disappears when rehearsal is disrupted by articulatory suppression, a procedure in which participants were instructed to repeat an irrelevant word or phrase over and over during the memory task (Baddeley, Lewis, & Vallar, 1984). According to the phonological loop model, since the word-length effect is due to differences in rehearsal rate, the disappearance of the effect under articulatory suppression is the logical consequence of rehearsal prevention.1
1 This explanation is conﬁned to situations in which items are presented auditorily, so they enter straight into the phonological store. The reason for this is that articulatory suppression prevents subvocalization. According to the model, subvocalization supports both rehearsal of items kept in the phonological store and recoding of visually presented stimuli into phonological traces. When items are presented visually under suppression, the word-length effect would disappear merely because items are not phonologically recorded, and, therefore, there are not items in the phonological store to be rehearsed.
´ ´ ´ Correspondence should be addressed to Guillermo Campoy, Departamento de Psicologıa Basica y Metodologıa, Universidad de Murcia, Campus de Espinardo, Apartado de correos 4021, 30080 Murcia, Spain. E-mail: email@example.com I thank Alan Baddeley, Eamon Fulcher, Luis Fuentes, John Xu, and Stephen Hasler for helpful comments on earlier drafts of this ´ ´ ´ manuscript and Lucıa Colodro, Beatriz Cuevas, Alejandro Hernandez, Cristina Hernandez, Yolanda Jaldo, and Violeta Provencio for their help in data collection. I also thank Gerry Tehan and two anonymous reviewers for constructive comments and suggestions.
# 2007 The Experimental Psychology Society DOI:10.1080/17470210701402364
THE EFFECT OF WORD LENGTH
However, there are a number of reasons why this interpretation should be considered with caution. On the one hand, apart from preventing rehearsal, articulatory suppression also involves the generation of abundant phonological noise as occurs in irrelevant speech...