Increased cognitive control after task conflict? Investigating the N-3 effect in task switching

Abstract

Task inhibition is considered to facilitate switching to a new task and is assumed to decay slowly over time. Hence, more persisting inhibition needs to be overcome when returning to a task after one intermediary trial (ABA task sequence) than when returning after two or more intermediary trials (CBA task sequence). Schuch and Grange (J Exp Psychol Learn Mem Cogn 41:760-767, 2015) put forward the hypothesis that there is higher task conflict in ABA than CBA sequences, leading to increased cognitive control in the subsequent trial. They provided evidence that performance is better in trials following ABA than following CBA task sequences. Here, this effect of the previous task sequence ("N-3 effect") is further investigated by varying the cue-stimulus interval (CSI), allowing for short (100 ms) or long (900 ms) preparation time for the upcoming task. If increased cognitive control after ABA involves a better preparation for the upcoming task, the N-3 effect should be larger with long than short CSI. The results clearly show that this is not the case. In Experiment 1, the N-3 effect was smaller with long than short CSI; in Experiment 2, the N-3 effect was not affected by CSI. Diffusion model analysis confirmed previous results in the literature (regarding the effect of CSI and of the ABA-CBA difference); however, the N-3 effect was not unequivocally associated with any of the diffusion model parameters. In exploratory analysis, we also tested the alternative hypothesis that the N-3 effect involves more effective task shielding, which would be reflected in reduced congruency effects in trials following ABA, relative to trials following CBA; congruency effects did not differ between these conditions. Taken together, we can rule out two potential explanations of the N-3 effect: Neither is this effect due to enhanced task preparation, nor to more effective task shielding.

Cognitive Multitasking: Inhibition in Task Switching Depends on Stimulus Complexity

“…might still have occurred, which might have increased the influence of episodic interference on N-2 repetition costs (see Grange, Kowalczyk, & O’Loughlin, 2017 ). To address these issues in Experiment 2, we used a paradigm that had produced robust N-2 repetition costs in several previous studies (e.g., Schuch & Grange, 2019 ; Schuch & Konrad, 2017 ; Schuch, 2016 ). We manipulated cue presentation time and number of stimuli, in order to explore whether these features could explain the absence of N-2 repetition costs in Experiment 1.…”

“…Experiment 2 was structurally identical to Experiment 1, in that participants switched between three different categorization tasks, the relevant task was indicated by a task cue preceding the stimulus, and the stimuli and responses were multivalent. Experiment 2 differed from Experiment 1 with respect to stimulus material and categorization tasks: A face categorization paradigm with abstract visual cues was applied in Experiment 2 that had produced robust N-2 repetition costs in several previous studies (e.g., Schuch & Grange, 2019 ; Schuch & Konrad, 2017 ; Schuch, 2016 ). Across three experimental conditions (between-subjects), we varied cue presentation time and stimulus set size, in order to explore whether these features modulated N-2 repetition costs.…”

“…As in our previous studies ( Schuch & Grange, 2019 ; Schuch & Konrad, 2017 ; Schuch, 2016 ), each portrait was presented inside a colored frame, with red frame indicating the “age” task, blue frame indicating the “gender” task, and yellow frame indicating the “emotion” task. The inside of the frames measured 10.5 cm in width and 14.0 cm in height (i.e., just the size of the face pictures); the thickness of the frame lines was 0.3 cm.…”

“…In Condition 2a, the task cue (colored frame) stayed on the screen while the stimulus (face picture) was presented, and disappeared together with the stimulus as soon as a response key was pressed. This procedure was adopted from previous work with this paradigm ( Schuch & Grange, 2019 ; Schuch & Konrad, 2017 ; Schuch, 2016 ).…”

Conflict adaptation during multilingual language production as evidenced by the n-3 effect

“…Declerck et al, 2017). This was done to assure that our setup would allow for the observation of the n-3 effect in a non-multilingual context, similar to the findings of Schuch and Grange (2015; 2019).…”