Do small dual-task costs reflect ideomotor compatibility or the absence of crosstalk?

Halvorson, Kimberly M.; Hazeltine, Eliot · 2015 · Psychonomic Bulletin & Review

DOI: 10.3758/s13423-015-0813-8

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 underlying the reduction or elimination of dual-task costs when tasks utilize ideomotor (IM) compatible stimulus-response (S-R) mappings. Previous research suggested that IM-compatible tasks, where stimuli resemble the sensory consequences of the response, allow for automatic response activation, thereby bypassing central selection processes and minimizing interference. However, alternative explanations propose that reduced costs stem from either the ability to treat two tasks as a single rule-based unit or the segregation of tasks into distinct modality-based systems (e.g., spatial vs. verbal). To distinguish between these accounts, Halvorson and Hazeltine manipulated S-R mappings to test whether automatic activation or modality separation drives the absence of dual-task interference. The researchers employed a between-subjects 2 × 2 design with 80 participants performing an auditory-vocal (AV) shadowing task and a visual-manual (VM) button-press task. Participants were assigned to one of four groups based on instruction type: Group II used IM-compatible "imitate" instructions for both tasks; Group OO used "do the opposite" instructions for both, reversing the S-R mappings; Groups OI and IO used mixed instructions (one imitate, one opposite). This design allowed the authors to test the embodied cognition account, which predicted large dual-task costs for Group OO due to the lack of automatic activation, against the spatial-verbal account, which predicted minimal costs across all groups because the tasks relied on distinct spatial and verbal codes. Performance was measured using reaction times (RTs) and accuracy across single-task, mixed-task (OR), and dual-task (AND) blocks to calculate mixing and dual-task costs. The results revealed no significant dual-task costs across any of the four groups, including those with reversed, incompatible mappings. While mixing costs were present and slightly larger for the fully IM-compatible group, the critical dual-task interference—defined as the RT difference between simultaneous and sequential response requirements—was negligible regardless of the S-R mapping. Accuracy remained high across all conditions, ruling out speed-accuracy trade-offs. These findings contradict the embodied cognition hypothesis that automatic response activation is necessary to eliminate dual-task costs, as the "opposite" tasks, which disrupt such automaticity, performed equally well. Instead, the data support the spatial-verbal account, indicating that the separation of central codes into distinct modality-specific systems (spatial for VM, verbal for AV) prevents crosstalk and interference. The study concludes that the absence of dual-task costs in IM-compatible paradigms is driven by the segregation of tasks into independent modality-based systems rather than the specific compatibility of S-R mappings. This finding aligns with models of working memory and dual-task interference that emphasize the role of distinct input-output channels in minimizing central bottleneck effects. It suggests that response selection can occur nearly simultaneously when tasks engage separate neural resources, even when the stimulus-response relationships are arbitrary or reversed. This challenges theoretical frameworks relying on automatic motor code activation and reinforces the importance of modality-specific processing in understanding human multitasking capabilities.

Key finding

Dual-task costs were negligible across all conditions, demonstrating that the separation of tasks into distinct spatial and verbal modality-based systems prevents interference regardless of stimulus-response compatibility.

Methodology

lab_experiment

Sample size: 80

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 author_sweep_intake on 2026-05-28.

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
• task switching