Temporal distinctiveness and repetition benefits in task switching: Disentangling stimulus-related and response-related contributions

Abstract

In cued task switching, decreasing switch costs with increasing response-to-cue interval (RCI) is a typical finding. The traditional account assumes an underlying process of task-set decay. In contrast, we suggest that these RCI effects are due to the influence of temporal distinctiveness on cue-based task retrieval (or reactivation). The present study explored the task-set components that are affected by temporal distinctiveness. According to the idea that a task set consists of response-related components and stimulus-related components, we manipulated response valence (bivalent vs. univalent) in Experiment 1. The results showed that lengthening the RCI leads to a loss of a task-repetition benefit, mainly when the RCI changed from the previous trial to the current trial, but this data pattern did not depend on response valence. In Experiment 2, stimulus valence was manipulated. The results revealed substantially stronger RCI effects with bivalent stimuli than with univalent stimuli. Taken together, the data are inconsistent with task-set decay accounts and suggest that the influence of RCI in task switching modulates cue-based retrieval of stimulus-related task components, such as biasing parameters.

Keywords:

• Cued task switching
• Response–cue interval
• Temporal distinctiveness
• Task-repetition benefit
• Episodic retrieval

Acknowledgments

This project was funded by Deutsche Forschungsgemeinschaft (DFG), Grant KO 2045/6–1. The authors would like to thank Vera Lawo and Liesel Heiermann for their help in conducting the experiments and Kate Arrington, Yoav Kessler, Michael Masson, and an anonymous reviewer for helpful comments on a previous version of this article.

Notes

¹ The “Z” and “Y” keys are reversed in QWERTZ and QWERTY keyboards.

² Further significant interactions were found between task transition and RCI, F(1, 19) = 25.34, MSE = 6,308, p < .001, and between RCI and RCI change, F(1, 19) = 50.99, MSE = 4,295, p < .001. No further interaction reached significance, Fs < 2.58, p ≥ .125. The following main effects were significant: task transition, F(1, 19) = 278.36, MSE = 16,847, p < .001; RCI, F(1, 19) = 11.22, MSE = 12,325, p = .003; and RCI change, F(1, 19) = 20.954, MSE = 65,602, p < .001. The effect of response valence was not significant, F < 1.

³Further significant interactions were found between RCI and RCI change, F(1, 27) = 102.52, MSE = 1,696, p < .001; RCI, RCI change, and stimulus valence, F(1, 27) = 33.82, MSE = 1,365, p < .001; task transition, RCI, and stimulus valence, F(1, 27) = 16.97, MSE = 2,320, p < .001; task transition and stimulus valence, F(1, 27) = 123.33, MSE = 5,135, p < .001; stimulus valence and RCI change, F(1, 27) = 12.25, MSE = 1,365, p = .002; and between task transition and RCI, F(1, 27) = 84.40, MSE = 2,833, p < .001. There were significant main effects of task transition, F(1, 27) = 220.74, MSE = 15,904, p < .001; RCI, F(1, 27) = 4.82, MSE = 36,627, p = .037; RCI change, F(1, 27) = 56.61, MSE = 98,696, p < .001; and stimulus valence, F(1, 27) = 419.80, MSE = 42,135, p < .001, the latter indicating that RTs were higher with bivalent than with univalent stimuli (925 ms vs. 527 ms).

⁴ Also, a schematic of the response mappings was available for participants for the whole experiment, which might have reduced the need for memory retrieval, which in turn might have diminished the effect of temporal distinctiveness.