Fractionating the neural mechanisms of cognitive control.

Modifications of the Wisconsin Card Sorting Test were established. In these new task variants, participants were asked to exert sequential control over attentional sets or over intentional sets (task domain factor). Attentional set shifting requires changing the priorities by which sensory stimuli are selected, whereas intentional set shifting requires changing the priorities by which motor responses are selected. Auditory stimuli that signaled to maintain or shift set were presented immediately before (precuing) or after (postcuing) the selection of cards (cue timing factor). Twenty-four healthy young individuals participated. Performance data (response times, error percentages) indicated that intentional tasks were easier to perform than attentional tasks. The electroencephalogram was recorded during task performance, and the N1, medial frontal negativity (MFN), P3a, and sustained potential (SP) components of the cue event-related brain potentials (ERPs) were analyzed. Irrespective of the task domain, shift precues led to increased N1 amplitudes compared to shift postcues. When intentional sets had to be shifted, the MFNs in the postcuing condition were more pronounced than in the precuing condition. On the other hand, shifts of attentional sets resulted in a more prominent P3a in response to postcues compared to precues. Irrespective of the task domain, the shift effect that was evident in SPs was more pronounced in precue ERPs compared to postcue ERPs. We conclude that ERPs provide valid measures to empirically constrain theories about the neural mechanisms of cognitive control. The domain hypothesis of the fractionation of the neural mechanisms of cognitive control is introduced.