van 't Wout, F. , Monsell, S. & Lavric, A.
School of Psychology. University of Exeter. Exeter, UK.
A growing body of evidence suggests that linguistic, and specifically phonological, representations might be involved in task-set preparation and maintenance (e.g. Miyake et al., 2004). But the precise nature of this contribution remains unclear. Accounts of task-switching generally assume that the current task’s stimulus-response (S-R) rules must be elevated to and maintained in a privileged state of activation. The two experiments reported here test the hypothesis that the phonological buffer is used to represent them. To this end, two variables that should reveal phonological buffer involvement - the word length and phonological similarity of the stimulus terms - were manipulated within a task-cueing paradigm. Specifically, participants were required to switch between classifying sets of images depicting nouns of longer or shorter spoken duration (Experiment 1), or between classifying sets of phonologically similar or dissimilar consonants (Experiment 2). The results demonstrate that neither word length nor phonological similarity affected task switching performance, or indeed performance in general. Only at the very start of Experiment 2 were reaction times reliably longer for phonologically similar than dissimilar consonants, suggesting a very transient role of the phonological buffer in representing S-R rules.