Dissociating restart cost and mixing cost in task switching
DOI: 10.1007/s00426-008-0151-9
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study investigates the cognitive mechanisms underlying two performance costs in task-switching paradigms: restart cost and mixing cost. Restart cost refers to the response delay on cued trials compared to non-cued trials within a task run, while mixing cost reflects slower responses on repetition trials under mixed-task conditions compared to single-task conditions. The authors aimed to dissociate these costs by manipulating task predictability and interference across three experiments using a multiple-trial paradigm with bivalent stimuli (color and shape matching). In Experiment 1, participants performed tasks with an unpredictable task order. Results showed a significant restart cost that decreased with successive task repetitions, suggesting that cue-based task activation benefits from repeated execution. Experiment 2 introduced a predictable task order, where participants were informed of the task sequence in advance. While restart cost persisted, it was significantly reduced compared to Experiment 1. Crucially, comparing performance between Experiment 1 (unpredictable) and Experiment 2 (predictable) revealed a mixing cost in the unpredictable condition, indicating that mixing cost arises from the interference associated with unpredictable task orders. Experiment 3 employed a single-task condition with predictable order to serve as a baseline. Comparing Experiment 3 to Experiment 2 confirmed that mixing cost was present only when tasks were mixed and unpredictable, not when they were mixed but predictable or single-task. The findings demonstrate that restart cost and mixing cost are based on dissociable mechanisms. Restart cost arises from processes involved in cue-based task activation needed to resolve task interference, which can be optimized through repeated task execution. In contrast, mixing cost stems from limited preparation on repetition trials when the identity of the approaching task is unpredictable, leading to transient competition between tasks. The study concludes that optimal task execution on repetition trials depends heavily on the predictability of the upcoming task, with unpredictable orders imposing additional cognitive costs due to interference management. These results refine theoretical accounts of task switching by distinguishing between costs related to cue processing and those related to task interference.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-11 |
| archive | success | canonical_url | — | — | 1 | 2026-06-25 |
| extract | success | cached | — | — | 2 | 2026-06-25 |
| clean | success | clean | — | — | 1 | 2026-06-20 |
| chunk | success | chunk | — | — | 1 | 2026-06-20 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-20 |
| enrich | success | openalex | — | — | 1 | 2026-06-20 |
| promote | success | — | — | — | 1 | 2026-06-11 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-25 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-20 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-25; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.