Task Preparation and the Switch Cost: Characterizing Task Preparation through Stimulus Set Overlap, Transition Frequency and Task Strength

摘要

Behavioral switch costs are commonly thought to reflect additional control processes necessary to change to a new way of responding to the environment. However, it is not clear whether task switching recruits additional control processes during the performance of task switch, but not task repeat trials. The current set of studies focused on preparatory processes involved in actively setting up a new task. Three experiments examined whether a. task switch preparation recruits unique processes, which are not necessary for task repeat preparation, or whether b. task switch preparation reflects greater recruitment of general preparation processes. The aim of the current studies is to identify whether there are unique task switch processes that may be influenced selectively by experimental factors. Several factors influencing task preparation were manipulated to determine whether each factor affects task switch preparation alone or whether each factor affects both switch and repeat trial preparation. First, the overlap in task set was examined to determine whether switch preparation is selectively affected. It was hypothesized that preparation demands would increase on task switch trials when both stimulus sets are the same since set overlap increases task interference. However, it was found that performing two tasks with the same stimulus sets increased task preparation on all trials. Second, transition frequency was manipulated to determine whether switch frequency affects switch preparation. It was found that switch frequency increased general preparation demands on all trial types. Further, it was found that high switch frequency did not increase the expectancy for the switch task type. Third, task strength was manipulated to examine whether transition frequency effects are the same for both the strong and weak task types. It was hypothesized that since the weak task trials already require a high degree of task retrieval to perform, preparation demands would not be further increased by a previous switch trial. It was found that frequent switching increased preparation for the strong task type, but not for the weak tasks. The results from all three experiments support the idea that general task preparation demands are increased on task switch trials.