The nature of residual cost in task switching

Abstract

Two experiments are presented that compare the residual cost found when switching from one task to another one in predictable sequences. In this research we tested Rogers and Monsell's stimulus-cued-completion hypothesis and its variants and alternatives. In our first experiment, two conditions were created: the base line (Stimulus-Response) was a replication of Tornay and Milán (2001, Experiment 3); the second condition (no-go condition) was identical to the first one with one modification: participants had to pay attention to the stimulus but they should not make a response on switch trials. The results indicate that instead of the cost disappearing with the first repetition trial, in the second condition the cost disappears between first- and second-repetition trials. The second experiment was similar to the second condition in experiment one, but participants did not know whether they had to execute the response until the end of the trial, when a go-signal appeared. Again, the cost shifted from switch trials to first repetition trials. So, we can conclude that paying attention to the stimulus or preparing a response to it is not enough to complete reconfiguration. A task-relevant response must be emitted for a full reconfiguration to occur