Abstract
Operating in a dynamic environment often requires inhibition of responses. Countermanding tasks have been used extensively to probe the mechanisms behind inhibition. We used a countermanding task to investigate the processes at play during failure of inhibition. Specifically, we investigated the reaction times for correcting mistakes. Participants were instructed to respond to a 30° visual target presented randomly left or right of central fixation by orienting their eyes, head, and arm to the target. A visual stop signal was presented in 30% of the trials at central fixation, 25, 75, 125, 175 or 225 ms after target appearance and prompted participants to cancel their response. In case of inhibition failure, participants were instructed to reorient all effectors back to center. Reaction times (RT) and times to correct a failed stop (CRT) were measured for all effectors using video eye tracking (Chronos Vision), 3D infrared marker tracking of the head and arm (Optotrak), and electromyography of shoulder and neck muscles (DelSys EMG). Estimated stop signal reaction times (SSRT) measured the efficiency of response inhibition. The RT distributions and SSRTs of all effectors were consistent with predictions from a dual LATER model describing a race between a ‘go’ and a ‘stop’ process. We extended this model to capture CRT distributions. We found a delay between the end of the go/stop race and the corrective response onset. This delay could be characterized by a second ‘go’ process starting after the previous stop signal reached threshold. Correlations of latencies (RTs and CRTs) between effectors suggest a supervisory control mechanism for both initial and corrective responses, pointing towards effector-specific decision processes receiving input from the common controller. Our results demonstrate that correcting incorrect responses relies on processes similar to the ones governing response initiation and inhibition and that this is true across all effectors.
NSERC, CFI, ORF (Canada).