The Task-Switch Cost is Still Absent After Selectively Stopping a Response in Cued Task Switching

The task-switch cost is one of the most robust phenomena in human task performance, but it can disappear after nogo trials where the actors decide not to respond to the target. According to the response-selection account, it is the occurrence of response selection that generates a task-switch cost on the subsequent trial, and the absence of a switch cost after nogo trials has been attributed to a nonoccurrence of response selection on nogo trials. However, an alternative account is that a task-switch cost is generated but is abolished on nogo trials because of the interference from the nogo signal with the activated task set, suggesting that the absence of a task-switch cost does not necessarily imply the nonoccurrence of response selection. The present study tested these competing accounts by using selective go/nogo procedures for which withholding a response would require selecting a response and inhibiting the selected response. Bayes factors in five experiments provided evidence for the absence of a task-switch cost after selective nogo trials, indicating that the occurrence of response selection does not necessarily result in a task-switch cost on the subsequent trial. The present results are consistent with the task-set interference account that a task-switch cost could be generated on nogo trials but is abolished because a nogo signal interferes with the activated task-set.