Non-decision time: the elephant in the executive control room

Response inhibition is seen as a key facet of executive control, investigated in thousands of studies across psychology, psychiatry and neuroscience. It is most frequently studied using the stop-task, and its main outcome measure, the stop-signal reaction time (SSRT), is overwhelmingly interpreted as indicating top-down inhibitory ability or speed. Here we provide converging evidence that undermines this central assumption, from reanalyses of archival datasets and a novel preregistered study. We find that low-level visuo-motor delays account for 2/3rds of manual SSRT duration and 40% of its individual differences, and predictably affect SSRT when manipulated. Sensory and motor contributions must be ruled out before linking SSRT results to inhibition or cognition. We evaluate a method for doing so, then introduce and evaluate a new measure, the selective stopping delay, that factors out visuomotor delays and captures the original aim of the SSRT to measure how fast response inhibition can be deployed.