Abstract
A recent article by Teasdale, Proctor, Lloyd, and Baddeley (1993) has shown the usefulness of the working memory paradigm for the research on stimulus independent thoughts (SIT). SIT are thoughts that come into our minds without elicitation by immediate external sensory input. The results of the four experiments reported, converged to suggest that the central executive (CE) plays an important role in the production of SIT. However, no hard evidence could be obtained, as the task used to suppress CE resources, also interferes with the activity of the phonological subsystem. The present experiment used a random interval generation task (Vandierendonck, De Vooght, & Van der Goten, submitted), which is a CE suppression task that is “pure”, in the sense that the task interferes with the CE but not with the phonological or the visuo-spatial slave system. The influence of a slow and a fast version of the “pure” CE task was compared with the effect of a control condition and with a random letter generation task. The results showed that the three interference tasks reduced the production of SIT, with the largest effect created by the random letter generation task. The results corroborated the finding of Teasdale et al. (1993) that the disruptive effect of tasks that involve the CE, is restricted to thought segments that arc part of a stream of connected thoughts. The results allowed for the conclusion that the production of connected sequences of SIT requires the involvement of the CE.