Serial short-term memory is markedly impaired by the presence of irrelevant speech so long as the successive tokens within the irrelevant speech are phonologically (or acoustically) dissimilar (Jones & Macken, 1995b). In two experiments in which consonant-vowel-consonant syllables were used as irrelevant speech tokens, we sought to evaluate the relative disruptive potency of changes in the final consonant only (Experiment 1), in the initial consonant, or in the vowel portion (Experiment 2) of each token. The results suggest that the vowel changes are the dominant source of disruption. This dominance may be explained, at least in part, by the role played by vowel sounds in the perceptual organization of speech and, in turn, the particular propensity for vowel changes to yield information about serial order. The results are consistent also with the view that the factors that promote order encoding in sound are also the ones that promote disruption.
In a seminal study, Colle and Welsh (1976) found that background speech disrupted serial recall of visually presented items, despite the fact that the participants had been instructed to ignore the speech. This so-called irrelevant sound effect has been widely and frequently replicated and has contributed to the understanding of the interrelations between perception, attention, and memory (for recent discussions, see, e.g., Baddeley, 2000; Jones & Tremblay, 2000; Neath, 2000). In the present study, we investigated a phenomenon first reported by Jones and Macken ( 1995b)-namely, that a sequence of irrelevant speech tokens that are phonologically dissimilar from each other causes more disruption than does a sequence of phonologically similar tokens. Using consonant-vowel-consonant (CVC) syllables, we examined whether any kind of phonological change between successive speech tokens gives rise to a phonological dissimilarity effect or whether some attributes (e.g., vowels, as compared with consonants) assume more importance than others.
Generally, many aspects of the irrelevant sound effect command a good degree of consensus: The intensity of the sound is not an influential variable, at least within the range of 40-76 dB(A) (Colle, 1980; see also Tremblay & Jones, 1999); the effect occurs within memory, not at encoding (Miles, Jones, & Madden, 1991 ); the meaning of sound (when speech is used) plays little if any role in the effect (Buchner, Irmen, & Erdfelder, 1996); and tasks that rely heavily on, or encourage, a serial rehearsal strategy (e.g., serial recall) are particularly susceptible to disruption (e.g., Jones & Macken, 1993).
An early account of the irrelevant sound effect proposed that the effect is confined to speech and, more specifically, that the effect is a function of the degree of phonological similarity between the irrelevant material and the to-be-remembered items (Salamé & Baddeley, 1982). However, several lines of evidence now converge to refute this account, including the finding that irrelevant nonspeech sounds (e.g., tones) can disrupt serial recall (e.g., Jones & Macken, 1993). Moreover, a study by Jones and Macken (1995b) provided more direct evidence against the phonological interference account by demonstrating that the degree of phonological similarity between the relevant and the irrelevant material was not, in fact, a strong predictor of the level of disruption (see also Larsen, Baddeley, & Andrade, 2000; LeCompte & Shaibe, 1997).
The study of Jones and Macken ( 1995b; Experiments 3 and 4) did reveal, however, that the degree of phonological dissimilarity within the irrelevant sequence does strongly dictate the degree of disruption. Specifically, an irrelevant sequence made up of the nonrhyming words hat, cow, nest . . . or deaf, pay, bell . . . was more disruptive than a sequence made up of the rhyming words sea, flea, key . . . or door, war, more ... In a more recent study, Larsen et al. …