No abstract

Existing computational models of the Stroop task differ in predictions concerning the set-size effect, which is the relation between a number of stimuli/responses and the magnitude of the Stroop interference. However, relevant empirical data is not unequivocal, as some studies reported no set-size effects, while others found substantial set-size effects. We administered two experiments in order to resolve this discrepancy in the case of the manual Stroop task. Experiment 1 compared conditions including four, six, and eight stimulus/response mappings in the picture-word task. No reliable set-size effects were found, apart from a weak effect observed when a working memory load imposed by the task was deliberately decreased. Experiment 2 tested conditions consisting of four versus eight mappings in the color-word task, and it replicated results of Experiment 1. As both experiments had sufficient power to detect set-size effects if they existed, our data are inconsistent with models predicting such effects. Key words:Stroop task, cognitive control, interference, set-size effectsThe Stroop task is a widely used test to determine how humans control (organize, coordinate) their own cognitive processing. In its standard color-word variant, the task involves the naming of the ink color of a word that denotes the same (congruent trials) or another color (incongruent trials). The picture-word variant requires naming a picture, while ignoring an included word. The crucial observation is the interference effect: an increased response latency (and sometimes error rate) in the incongruent trials, compared to the congruent trials, or to trials called neutral, in which meaningless colored strings (e.g., XXXXX) are shown. This observation means that overcoming a well-learned action (i.e., reading), in order to perform a relatively novel process (naming a color or a picture), requires an additional effort from the cognitive system, and results in significant processing costs.Research on cognitive control using the Stroop task involves showing that the interference effect can change as a result of cer-

A meta-analysis was carried out to demonstrate the existence of publication bias in research on the relationship between measures of fluid intelligence and working memory. Reanalysis of data collected in Ackerman, Beier, & Boyle, 2005 was conducted. A heterogeneous distribution of correlation coefficients in the absence of asymmetry in the distribution of coefficients was observed. According to the author of the analysis, there are no arguments for the presence of publication bias in this particular set of results drawn from research on intelligence and working memory.

No abstract