Dynamics of the Central Bottleneck: Dual-Task and Task Uncertainty

Sigman, Mariano; Dehaene, Stanislas · 2006 · OpenAlex-citations

DOI: 10.1371/journal.pbio.0040220

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 addresses a theoretical contradiction between two classical cognitive paradigms: the psychological refractory period (PRP) and task switching. The PRP paradigm posits a passive, structural bottleneck where tasks are processed sequentially on a first-come, first-served basis. Conversely, task-switching research suggests that switching between tasks requires active control by a central executive system. Sigman and Dehaene designed an experiment combining dual-task simultaneity with task uncertainty to reconcile these views, aiming to determine whether task selection and processing involve passive bottlenecks or active executive control. Participants performed a dual-task experiment involving a visual number comparison task and an auditory tone discrimination task. Crucially, the order of stimulus presentation and the stimulus onset asynchrony (SOA) were unpredictable, preventing participants from pre-planning task order. The number task was manipulated via notation (Arabic digits vs. spelled words, affecting perceptual processing) and numerical distance (affecting central decisional processing). This design allowed the authors to isolate the timing of task selection relative to perceptual and central processing stages. The results revealed that task selection is not determined at stimulus onset but rather after perceptual processing is complete. Evidence for this includes a significant shift in the SOA required for equal task selection (50%SOA) based on notation manipulation, but no shift based on numerical distance. Furthermore, response times for the first task were significantly slower at short SOAs (<300 ms) compared to long SOAs, a finding that contradicts the classic passive bottleneck model, which predicts first-task response times should be independent of SOA. This slowing indicates an active control stage involved in preparing for the dual-task context. Additionally, response times for the second task decreased linearly with SOA, confirming the presence of a central bottleneck, but with a slope steeper than -1, reflecting the additive interference from the active control processes affecting the first task. The authors conclude that dual-task performance involves both a passive central bottleneck and active executive processes for task setting and disengagement. They propose a hierarchical model of cognitive architecture that synthesizes PRP and task-switching findings. This model accounts for the observed chronometric relations, suggesting that while central processing stages are serial, the engagement and disengagement of these stages are actively controlled, resolving the apparent contradiction between the two paradigms.

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.

• dual task multitasking
• task switching
• executive function