We examined the role of the phonological loop in task switching performance in a prevalently used research paradigm, that is, the alternating runs paradigm. This paradigm basically requires participants to alternate tasks on every second trial (i.e., AABBAABB). In this study, participants were required to alternate the letter and numerical decision tasks in cue-present and cue-absent conditions under control, articulatory suppression and foot tapping conditions. The results showed that error rates were larger and reaction times were longer in the articulatory suppression condition than in the control and tapping conditions for both switch and nonswitch trials in two cue conditions. These findings indicate that the phonological loop contributes to task performance not only in the switching trials but also in situations where the maintenance of task sequence information is required.

Key words: working memory, phonological loop, task switching, alternating runs paradigm, articulatory suppression