Dissociating Levels of Cognitive Control: The Case of Stroop Interference

Attention is often imperfect; cognitive control is needed to counteract the tendency to attend to distractors that are incompatible with current goals. Cognitive psychologists have long explored cognitive control by examining Stroop interference-the slowed naming of colors on incongruent trials (e.g., "RED" displayed in blue ink), as compared to congruent trials (e.g., "RED" displayed in red ink), in the color-word Stroop task. The magnitude of interference reflects the effectiveness of cognitive control, but it does not reveal the precise processes used to minimize attention to the distracting word. The need for experimental approaches that accomplish this objective is underscored by the existence of qualitatively different cognitive control processes. Prior accounts stressed the use of top-down filtering processes at a task- or list-wide level to avoid word reading, but recent findings have shown that control of word reading is sometimes stimulus-driven-that is, triggered by the processing of stimuli or stimulus features. In this article, I highlight the critical findings that dissociate top-down and stimulus-driven control in the Stroop task, dissociations that are central to the view that cognitive control operates at multiple levels.