Can the Stroop effect serve as the gold standard of conflict monitoring and control? A conceptual critique.

Abstract

The Stroop effect has been a key to the assay of selective attention since the time of the epoch-making study by J.R. Stroop almost a century ago. However, recent work based on computational modeling and recording of brain activations ignored the primary meaning of the Stroop effect as a measure of selectivity-with the Stroop test losing its raison d'être. Espousing the new framework, numerous studies in the past 20 years conceived performance in the Stroop task in terms of conflict-induced adjustments governed by central control on a trial-to-trial basis. In the face of this tsunami, we try to convince the reader that the Stroop effect cannot serve as a testing ground for conflict-monitoring and control, because these constructs are fundamentally unsuited to serve as a candidate theory of Stroop processes. A range of problems are discussed that singly and collectively pose grave doubts regarding the validity of a control and conflict monitoring account in the Stroop domain. We show how the key notion of conflict is misconstrued in conflict-monitoring models. Due to space limitations and for sake of wider accessibility, our treatment here cannot be technical.