Sequential task predictability in task switching

Koch, Iring · 2005 · Crossref

DOI: 10.3758/bf03196354

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 mechanisms of action control in task switching, specifically addressing whether sequential task predictability facilitates a switch-specific reconfiguration process. Traditional theories posit that switching between tasks requires a unique reconfiguration of cognitive processes, whereas repetitions do not. Consequently, manipulations that increase preparation time typically reduce switch costs more than repetition costs. However, previous research using incidental sequence learning suggested that predictability benefits were equal for switches and repetitions, challenging the reconfiguration account. This paper tests whether explicitly instructed sequential predictability yields switch-specific benefits, thereby ruling out methodological confounds such as between-subjects designs or incidental learning limitations. The research comprised three experiments using a within-subjects negative transfer design. Participants performed tasks involving odd/even or greater/smaller judgments of digits presented within square or diamond frames. In Experiments 1 and 2, participants practiced a predictable four-trial sequence (AABB) before being transferred to a random sequence. Experiment 1 used a long cue-stimulus interval (900 msec), while Experiment 2 used a short interval (100 msec) to maximize reliance on sequential predictability. Experiment 3 reversed the design, transferring participants from random to predictable sequences. Reaction times (RTs) and error rates were recorded to measure the impact of predictability on switch and repetition trials. The results consistently demonstrated that the benefit of sequential task predictability was not specific to task switches. In Experiments 1 and 2, transferring from predictable to random sequences increased RTs significantly (by 81 msec in Experiment 1 and 225 msec in Experiment 2), but this increase was equivalent for both switch and repetition trials. Consequently, switch costs remained stable regardless of predictability. Experiment 3 confirmed these findings; transferring from random to predictable sequences improved performance, but the benefit was larger for repetitions than for switches, effectively increasing switch costs in the predictable condition. These results held across different cue-stimulus intervals and transfer directions. The findings challenge the prevailing view that task switching requires a distinct reconfiguration process absent in repetitions. Instead, the data support a general task-updating or goal-setting control process that is necessary for both switches and repetitions. The lack of switch-specific benefits from predictability suggests that the cognitive mechanisms engaged by sequential knowledge are not unique to switching. This implies that switch costs may arise from differences in the duration of component processes, such as response selection, rather than the presence of a switch-specific reconfiguration stage. The study concludes that task preparation based on predictability affects general task control rather than facilitating a specialized switch mechanism.

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.

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.

• task switching