Is dual-task slowing instruction dependent?

Levy, Jonathan; Pashler, Harold · 2001 · Journal of Experimental Psychology Human Perception & Performance

DOI: 10.1037/0096-1523.27.4.862

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper investigates whether dual-task slowing—the delay in reaction times when performing two tasks concurrently—is caused by structural processing limitations or by instruction-dependent strategies. The authors test the Instruction Dependent Hypothesis (IDH), which posits that dual-task interference can be eliminated if subjects are instructed to assign equal priority to both tasks and are financially incentivized to do so. This contrasts with the Central Bottleneck (CB) theory, which argues that certain mental operations, such as response selection, cannot be performed in parallel regardless of instructions. The study was motivated by prior research (Schumacher et al., 1997) that claimed to eliminate dual-task slowing using equal-priority instructions, thereby challenging the CB model. To test these competing theories, the authors conducted three experiments using a standard dual-task paradigm where subjects responded to visual and auditory stimuli. All experiments employed identical instructions and financial incentives designed to encourage parallel processing. Experiment 1 replicated Schumacher et al.’s design, pairing a visual-manual task (keypress to disk location) with an auditory-vocal task (vocal response to tone pitch). Experiment 2 switched the modality mappings, requiring a manual response to the tone and a vocal response to the disk. Experiment 3 eliminated peripheral conflicts by using a single visual stimulus (a colored disk), requiring subjects to respond to both its color (manual) and location (vocal). The results varied significantly across experiments. Experiment 1 largely replicated the previous findings: vocal responses showed no significant slowing in the dual-task condition compared to single-task conditions, but manual responses did show significant slowing. However, Experiment 2 demonstrated substantial dual-task slowing for both vocal and manual responses when the modality mappings were switched. Experiment 3 also revealed significant slowing for both tasks despite the use of a single stimulus, which theoretically minimized peripheral interference. A control experiment confirmed that the slowing in Experiment 3 was not due to stimulus presentation differences. The authors conclude that equal-priority instructions and incentives are not sufficient to eliminate dual-task interference. The presence of significant slowing in Experiments 2 and 3 contradicts the IDH, which predicts that such instructions should allow for parallel processing. Instead, the findings support the view that dual-task slowing is often due to task-specific features and structural limitations consistent with the Central Bottleneck theory. The study demonstrates that the elimination of interference observed in prior work was likely specific to the particular combination of tasks and modalities used, rather than a general property of human information processing.

Key finding

Equal-priority instructions and financial incentives are not sufficient to eliminate dual-task slowing, as substantial interference persists when task mappings are altered or a single stimulus is used.

Methodology

lab_experiment

Sample size: 64

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. Discovered via openalex_abstract on 2026-05-08.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• dual task performance
• dual task multitasking