Four dual-task experiments are reported in which a short-term memory task is performed concurrently with a random interval repetition task, which was designed to interfere with functions normally attributed to the central executive in the working memory model of Baddeley and Hitch (1974). The task was found to interfere with supra-span serial recall and with backward memory span, but did not disrupt performance on a forward-memory-span task. The effects were observed in dissociation with effects of articulatory suppression and matrix tapping, so that the locus of the effects of the new task is not due to the slave systems. In addition, single-task random-interval repetition performance was sampled and compared to performance in the dual-task conditions of all four experiments. Although quality of tapping performance differed between the single-task and the dual-task conditions, it was not related to recall performance. All the results are discussed with reference to the working memory model.

The present article reports two experiments testing the use of working memory components during reasoning with temporal and spatial relations in four-term series problems. In the first experiment four groups of subjects performed reasoning tasks with temporal and with spatial contents either without (control) or with a secondary task (articulatory suppression, visuo-spatial suppression or central executive suppression). The second experiment tested the secondary task effects in a within-subjects design either on problems with a spatial content or on problems with a temporal content, and within each content domain either under conditions of self-paced or of fixed presentation of the premises. Both experiments found effects of all three secondary tasks on reasoning accuracy. This supports the hypothesis that the subjects construct spatial representations of the premise information with the support of visuo-spatial resources of working memory. The second experiment also showed that during premise intake, only visuo-spatial and central executive secondary tasks had an effect. The implications of the data for the working memory requirements of reasoning and for theories of linear reasoning are discussed.

According to the mental model theory, reasoners build an initial model representing the information given in the premises. In the context of relational reasoning, the question arises as to which kind of representation is used to cope with indeterminate or multimodel problems. The present article presents an array of possible answers arising from the initial construction of complete explicit models, partial explicit models, partial implicit models, a single "isomeric" model, or a single annotated model. Predictions generated from these views are tested in two experiments that vary the problem structure and the number of models consistent with the premises. Analyses of the premise processing times, answering times and accuracy show that the annotated model yields the best fit of the data. Implications of these findings for the mental model theory as developed for relational reasoning are discussed.In linear syllogisms such as George is heavier than Tony and Tony is heavier than Jacques, people easily infer that George is heavier than Jacques. The findings collected in numerous studies substantiate the hypothesis that people solve such problems by constructing a linear array with the largest values at the top and the smallest ones at the bottom or alternatively by constructing a left-to-right array (see, e.g., De Soto, London, & Handel, 1965;Huttenlocher, 1968;Potts, 1974). The review by Evans, Newstead, and Byrne (1993, chap. 6) shows that this view yields the best summary of the findings reported in the literature. The usage of such an integrated spatial representation seems to be independent of the relational content and works both for determinate problems (as in the example above) and for indeterminate problems as, for example, in George is heavier than Jacques and George is heavier than Tony, where the relation between Jacques and Tony remains unresolved. Support for all this has been accumulated in

No abstract

No abstract