The (in)flexibility of updating a mental task representation: On the origins of costs when shifting from a task-switching to a single-task context.

Task switching requires flexibly engaging in one of two task sets in each trial. Curiously, when one task suddenly becomes irrelevant (fade out), performance in the remaining task is worse than when performed as a single task. This fade-out cost demonstrates that the mental task model of task switching has to be reconfigured to a single-task representation (Mayr & Liebscher, 2001). This study implemented list-wide proportion manipulations during task switching to investigate how global processing adjustments affect fade-out costs. Experiment 1 manipulated the proportion of task switches at the level of task representation: High switch frequency was expected to increase the accessibility of both task sets in working memory, predicting increased fade-out costs. Experiment 2 varied the proportion of task-rule congruency at the level of response selection, predicting no significant effect. Results from 160 German University students showed larger fade-out costs for the high switch frequency group, whereas the proportion of task-rule congruency did not affect fade-out costs. These findings suggest that global adjustments at the task representation level uniquely influence fade-out costs and hereby the reconfiguration of the task model, whereas adjustments at the response level do not. Implications for the mental representation of task models are discussed. (PsycInfo Database Record (c) 2025 APA, all rights reserved).