Modality-specific preparatory influences on the flexibility of cognitive control in task switching

ABSTRACT

In the current study, we addressed modality-specificity of the flexibility of cognitive control. We compared performance in single-task and mixed-tasks blocks between blocked auditory and visual stimuli assessing alternation costs (single vs. mixed). Mixed blocks comprised task switches only. The tasks consisted of numerical parity, magnitude, and distance judgments about numbers between one and nine without five. A cue indicated the relevant task. The cue–stimulus interval was varied (short vs. long interval) to examine preparation effects. The results indicated higher response times (RTs) and error rates (ERs) in mixed- vs. single-tasks blocks. The alternation costs in ERs were larger for auditory compared to visual stimulus presentation. Moreover, the reduction of RT alternation costs based on increased preparation time was more pronounced for the auditory modality compared to the visual modality. These results suggest a modality-specific influence on processes involved in maintaining and updating task sets in working memory.

KEYWORDS:

• Stimulus modality
• cognitive control
• task switching
• task preparation
• alternation costs

Acknowledgements

We would like to thank Linn Zulka and Helen Overhoff for data collection as well as Sander Los and two anonymous reviewers for valuable comments on an earlier version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 When analyzing the data with Experiment as a between-subject variable, results showed a main effect of experiment in RT, F(1, 46) = 41.58, p < .001,  = .48, and ER, F(1, 46) = 6.87, p = .012,  = .13, in opposite directions: participants were slower but more accurate in Experiment 1B compared to Experiment 1A. This is probably due to the longer response window in Experiment 1B allowing longer responses to be included as valid responses instead of being dismissed as errors. Also, the interaction of block type and experiment was significant in RT, F(1, 46) = 74.84, p < .001,  = .62. Alternation costs were more pronounced in Experiment 1B compared to Experiment 1A. In addition, there was a significant interaction of CSI and experiment in ER, F(1, 46) = 13.53, p = .001,  = .23. A difference between experiments occurred in the short CSI with higher ERs in Experiment 1A. However, there were no significant interactions of experiment with modality, so that we decided to collapse data over experiments.

2 With three tasks in the mixed blocks, it is possible to assess n–2 repetition costs as a measure of persisting inhibition (Koch, Gade, Schuch, & Philipp, 2010; Mayr & Keele, 2000). When analysing the data of the mixed blocks with n–2 task transition as an additional independent variable, we did not observe consistent results, and the n–2 repetition costs were not significant for any condition (see Grange, Juvina, & Houghton, 2013, for a discussion of the meaning of the absence of n–2 repetition costs). Therefore, we do not report these data analyses in more detail.

Additional information

Funding

This research was funded by the DFG (Deutsche Forschungsgemeinschaft [grant number: KO 2045/15–1/2]).