Andrea Kiesel

Who they are

The review lists seven authors and their affiliations as printed: Kiesel at the University of Wuerzburg, Steinhauser at the University of Konstanz, Wendt at Helmut Schmidt University, Falkenstein at the Leibniz Research Centre for Working Environment and Human Factors in Dortmund, Jost at the University of Marburg and RWTH Aachen University, and Philipp and Koch at RWTH Aachen University. Kiesel is the corresponding author, and Koch is given as a second correspondence address.

We do not state her present post or career history, because the sources read for this page give only the 2010 affiliation.

Key papers

Kiesel, A., Steinhauser, M., Wendt, M., Falkenstein, M., Jost, K., Philipp, A. M., & Koch, I. (2010). Control and interference in task switching: A review. Psychological Bulletin, 136(5), 849-874. doi:10.1037/a0019842.

The abstract describes a survey of task-switching methods, then preparation for upcoming tasks and theories of how preparation works, then interference from recently performed tasks, stimuli and responses, and closes with future research directions. The review notes that a 2003 review by Monsell had 314 citations and that a search returned 2,513 hits in Medline and PsycINFO, which gives a sense of the field's size.

Key ideas

The review defines the core measure. A switch cost is the performance penalty on trials where the task changes compared with trials where it repeats, and the basic finding is a robust switch cost in both reaction time and error rates (p. 849). Mixing costs, the global slowdown in mixed-task blocks compared with single-task blocks, are kept apart from the local switch cost (pp. 850, 852). It places Jersild (1927) and Spector and Biederman (1976) among the earliest studies, using fixed sequences in which the task changed on every trial.

On preparation, it reports residual switch costs that remain even with ample time for preparation (p. 855), and gives the main explanations: two-stage models with a second reconfiguration process that can start only after the stimulus appears, and accounts based on interference from the other task (p. 860). It records an example of size, a residual of 58 milliseconds at an 800 millisecond preparation interval in the work of Schneider and Logan (p. 858).

The review surveys more than one way of running the experiment. In the alternating-runs paradigm, costs were substantial even though the tasks were rather simple (p. 852). In the task-cuing paradigm, where a cue announces the next task, the review reports robust switch costs as well, and robust costs also emerge when participants choose the task themselves (p. 852). Monsell and Mizon (2006) observed large switch costs that declined with the cue-stimulus interval, and Arrington and colleagues (2007) obtained substantial task-switch costs with double registration (p. 859). The point of listing the paradigms is that the cost appears in all of them, whatever the way the switch is signalled.

It also records the cases that strain a tidy account. Brief cues decreased the residual substantially and in some cases completely in work by Verbruggen and colleagues, while Steinhauser and colleagues found substantial residual costs despite brief cues, and Lien and colleagues found large costs for some stimulus-response rules with more than two seconds of preparation (p. 856). Strong incentives raised the proportion of fully prepared trials only marginally in work by Nieuwenhuis and Monsell.

The review's own summary is that there are usually substantial true task-switch costs, and that they decrease if participants prepare (p. 860). It does not translate those costs into time lost at work, and this site's articles take care not to do so on its behalf.

---