Despite an extensive literature on the neural substrates of response inhibition, when and where this process occurs in the brain remain unclear. The present study aimed to shed light on this issue by exploiting the high temporal resolution of the event-related potentials (ERPs) and recent advances in source localization. Temporo-spatial principal component analysis was employed to define more precisely the two ERP components most often associated with response inhibition (i.e., frontocentral N2

more »

... and frontocentral P3), as well as to improve the accuracy of source localization. In addition, participants (N = 40) performed a modified Go/ Nogo task composed of three types of stimuli (frequent-Go, infrequent-Go, and infrequent-Nogo), which allowed us to dissociate neural activity associated with response inhibition from that related to novelty processing by directly contrasting nogo and go trials matched with respect to frequency of occurrence. Scalp ERP data indicated that the frontocentral P3, but not the frontocentral N2, showed larger amplitudes for infrequent-Nogo than for infrequent-Go trials. Source localization data parallel the results obtained at the scalp level: only P3-related activity showed differences between infrequent-Nogo and infrequent-Go trials. This increased activation was observed predominantly in the presupplementary motor area (preSMA). Present results suggest that the frontocentral P3 and the preSMA play a core role in response inhibition. The findings of this study substantiate and complement previous results obtained by hemodynamic procedures.