Task switching with a 2:1 cue-to-task mapping: separating cue disambiguation from task-rule retrieval.

An experiment is reported that investigated switching among two numerical judgment tasks with a factorial variation of the cue-to-task mapping (1:1 vs. 2:1) for each of the tasks. In addition, the precuing interval (CSI) was varied. The results suggest that with a long CSI of 1,100 ms, switching performance is almost completely determined by the task-specific conditional probability of a task switch given a cue switch. This effect probably reflects the complexity of task-rule retrieval. Without preparation (CSI = 0 ms), the complexity of cue disambiguation as a function of the number of cues across tasks seems to account for most part of the additional variance observed in this condition. The latter observation is in line with suggestions that increasing the number of cues per task from one to two introduces additional demands on the level of cue processing that reflect the transition from an isomorphic to a homomorphic mapping function.