Inhibitory control is a crucial cognitive-control ability for behavioral flexibility, which has been extensively investigated through action-stopping tasks. Multiple neurophysiological features have been proposed as 'signatures' of inhibitory control during action-stopping, though the processes indexed by these signatures are still controversially discussed. The present study aimed to disentangle these processes by comparing simple stopping situations with those in which additional action revisions were needed. Three experiments in female and male humans were performed to characterize the neurophysiological dynamics involved in action-stopping and -changing, with hypotheses derived from recently developed two-stage 'pause-then-cancel' models of inhibitory control. Both stopping and revising an action triggered an early, broad 'pause'-process, marked by frontal EEG β-frequency bursting and non-selective suppression of corticospinal excitability. However, EMG showed that motor activity was only partially inhibited by this 'pause', and that this activity could be modulated during action-revision. In line with two-stage models of inhibitory control, subsequent frontocentral EEG activity after this initial 'pause' selectively scaled depending on the required action revisions, with more activity observed for more complex revisions. This demonstrates the presence of a selective, effector-specific 'retune' phase as the second process involved in action-stopping and -revision. Together, these findings show that inhibitory control is implemented over an extended period of time and in at least two phases. We are further able to align the most commonly proposed neurophysiological signatures to these phases and show that they are differentially modulated by the complexity of action-revision.Significance Statement Inhibitory control is one of the most important control processes by which humans can regulate their behavior. Multiple neurophysiological signatures have been proposed to index inhibitory control. However, these play out on different time scales and appear to reflect different aspects of cognitive control, which are controversially debated.Recent two-stage models of inhibitory control have proposed that two phases implement the revisions of actions: 'pause' and 'retune'. Here, we provide the first empirical evidence for this proposition: Action revisions engendered a common, low-latency 'pause' process, during which motor activity is broadly suppressed. Later activity, however, distinguishes between simple stopping of actions and more complex action revisions. These findings provide novel insights into the sequential dynamics of human action control.
Read full abstract