Sequential adjustments of task-pair control in dual-task contexts: Examining the role of repetition priming effects at the level of task-pair sets and abstract control states.

In dual-task situations, two stimuli are presented in rapid succession, requiring participants to perform two tasks simultaneously. Prior studies suggested that when two tasks are performed simultaneously, information about the identity of the two tasks is represented in a joint cognitive representation, referred to as the task-pair set. This evidence comes from studies showing that switching between different task pairs results in performance costs, called task-pair switch costs (i.e., performance in task-pair switches vs. repetitions). In the present study, we focused on the adjustive characteristics of task-pair switching by investigating whether task-pair switch costs are sequentially modulated by the previous experience with a task-pair switch (vs. repetition). First, we reanalyzed the data of four published experiments and observed a reduction of task-pair switch costs after a task-pair switch trial relative to after a task-pair repetition trial. Second, we confirmed this novel finding in a new experiment. The new experiment also showed that performance in a current task-pair repetition was better after a task-pair repetition than after a task-pair switch, whereas the performance in a current task-pair switch was not modulated by the task-pair sequence in the previous trial. These findings suggest that automatic bottom-up repetition priming at the level of task-pair sets, rather than repetition priming at the level of abstract control states, contributes to the sequential adjustment of task-pair switch costs.