An fMRI and TMS investigation of response, semantic, and task conflict in the Stroop task.

Abstract

Overcoming Stroop interference (i.e., ignoring the meaning of a word whilst naming the colour it is printed in) requires selective attention which calls upon several different processes. A Stroop stimulus generates cognitive conflict at different stages of the processing stream (i.e., at the task, semantic, and response level). An enduring goal of research investigating attentional control is to establish the point in the processing stream at which information can be selected or ignored, and in turn establish whether dissociable neural networks underpin these different cognitive processes. The anterior cingulate cortex (ACC) and dorsolateral prefrontal cortex (DLPFC) are two regions thought to be central to the circuitry of attentional control, however the precise role that each plays is still widely debated. The present thesis attempts to elucidate the function of these regions in two experiments employing a Stroop paradigm. Experiment 1 (n=20) used high frequency rTMS to stimulate the left DLPFC and observed the behavioural effect on each of the various forms of cognitive conflict. Our results showed a significant main effect of stimulation in reducing response times (RTs) compared to sham-stimulation, however stimulation had no effect on any of the underlying components of interference. In line with previous findings our results were consistent with the role of the left DLPFC in implementing top–down attentional control. Experiment 2 (n=13) employed the same Stroop paradigm while the BOLD response of participants was recorded using fMRI. We demonstrated evidence for hemispheric differences in attentional control, as the right DLPFC appeared more engaged by response conflict, while left DLPFC activation was unique to pre-response conflict. Surprisingly ACC activation was only associated with semantic conflict. We concluded that our results favour multi-stage selection accounts of attentional control but highlight the need for additional neuroimaging studies with more consistent methods before our findings can be validated. While the Stroop effect is one of the most studied phenomena in psychology our findings make clear that there remains a significant amount for future research to address. The results are discussed in context of the limitations of our experiments and directions for future research are proposed.