Cognitive control during a spatial Stroop task: Comparing conflict monitoring and prediction of response-outcome theories

Abstract

Cognitive control allows information processing and behaviour to vary adaptively from moment to moment depending on current goals. Two of the most prominent theories that have been proposed to account for the processing of cognitive control are the Conflict Monitoring Theory (CMT) and the Prediction of Response-Outcome Theory (PRO). According to both theories, the implementation of cognitive control during a trial in a conflict task reflects processing events that occurred in the preceding trial. Both CMT and PRO advocate that the detection of conflict situations leads to the recruitment of cognitive control, but they differ regarding the processing underpinnings of cognitive control during conflict resolution. CMT proposes that conflict between alternative responses is resolved by enhancing the task's relevant dimension, reducing interference from the task's irrelevant dimension(s). This control setup promotes conflict adaptation in the subsequent trial. PRO proposes that conflict is resolved by means of a cost-effectiveness analysis that identifies and suppresses action plans linked to the less appropriate responses, facilitating conflict resolution in the subsequent trial. To adjudicate between these alternatives, we manipulated contingencies pertaining to two-trial sequences (n-1; n), namely, the congruency between task relevant/irrelevant dimensions in trial n-1 and response repetition in trial n. A spatial Stroop task was used, in which task-relevant and irrelevant information were integrated within the same stimulus. In this task, participants were required to attend to the direction of an arrow while ignoring its position. The arrow's direction and position could be congruent (C) or incongruent (IC). In one experiment, trials in which the participant was required to respond according to the position of a circle (PO; position only trials), occupying the sequential position n, were the focus of the analyses. Three experiments were conducted manipulating the trials' sequence structure. In Experiment 1, we studied a low control/low conflict condition (cC trials), and two high control/low conflict conditions (icC with and without response repetition). In Experiment 2, we studied two low control/no conflict conditions (cPO with and without response repetition) and two high control/no conflict conditions (icPO with and without response repetition). In Experiment 3, we studied a high control/high conflict condition (icIC) and two low control/high conflict conditions (cIC with and without response repetition). Overall, our findings are in agreement with previous studies in which both bottom-up processing, linked to response and stimulus position repetition, and top-down processing, linked to cognitive control, were shown to contribute to sequence effects in conflict tasks. Specifically, our observations mainly support PRO's account of conflict resolution, in which the intervention of top-down processing is substantially more complex than in CMT's account.