The compass of control: Congruency sequence effects, boundaries, and task sets

Abstract

The congruency sequence effect literature is inconsistent when it comes to the issue of the boundaries of this phenomenon. Some have argued that control is implemented at the level of a stimulus dimension while other have claimed that control operates across dimensions and tasks. Flexible control boundaries defined by task sets might explain the inconsistent findings. Response set manipulations have been shown to influence control boundaries. Unitary response sets can produce cross-dimension congruency sequence effects but applying separate response sets to a task can lead to the absence of the congruency sequence effect on dimension switch trials. This thesis is concerned with the extension of these findings. The study applied response set manipulations to a paradigm (Stroop Trajectory Task) that has exhibited robust crossdimension congruency sequence effects. In addition, the influence of switching routine on congruency sequence effects was tested. It was expected that separate response sets for different stimulus dimensions would eliminate the congruency sequence effect on switch trials. Switching routine was anticipated to act as a weaker boundary marker but nevertheless it was expected that systematic switching routines should have at least attenuated the congruency sequence effect on switch trials. Contrary to expectations it was found that the congruency sequence effect was present in switch trials across all conditions.