# Effect of modality compatibility on dual-task performance in a more naturalistic environment

## TL;DR

This study explores how task compatibility affects performance in multitasking scenarios, comparing lab settings to a more naturalistic driving simulator.

## Contribution

The study demonstrates that modality compatibility effects are weaker in naturalistic settings compared to classical laboratory environments.

## Key findings

- Dual-task costs were greater for incompatible modality mappings than compatible ones in reaction time.
- The modality compatibility effect was more pronounced in the classical laboratory than in the driving simulator.
- Naturalistic settings showed reduced interference from incompatible modality mappings compared to lab settings.

## Abstract

The modality-compatibility effect describes multitasking-specific performance differences between compatible stimulus-response mappings (e.g., visual-manual and auditory-vocal) and incompatible mappings (e.g., visual-vocal and auditory-manual). The modality compatibility effect has been demonstrated in various studies on dual-tasking with classical laboratory experimental designs. However, it remains unclear whether the effect also exists in everyday-life dual-task situations, such as driving while performing an additional task. Here, we investigated modality compatibility effects under single-task and dual-task conditions and compared a classical laboratory setting with a more naturalistic driving simulator setting, in which participants acted as front seat passengers and were not required to operate the vehicle. We hypothesized larger dual-task costs under modality incompatible mappings compared to modality compatible mappings across settings, and a higher magnitude of the modality compatibility effect on dual-task costs in the classical laboratory setting than in the driving simulator setting. Participants (N = 31, age: 22.86 years ± 2.04, 21 women, 10 men) responded manually or vocally to visual or auditory stimuli (modality compatible mappings or modality incompatible mappings) in both settings. Results revealed greater dual-task costs for modality incompatible mappings than modality compatible mappings for reaction time only. However, overall dual-task costs did not differ between the classical laboratory and the driving simulator settings. Most importantly, there was a significant interaction effect between modality mapping and setting such that the effect of the modality incompatible mapping was more pronounced in the classical laboratory than in the driving simulator. These findings highlight that the magnitude of interference from modality incompatible mappings is weaker in more naturalistic settings compared to the classical laboratory. This suggests that the modality-compatibility effect may exist in real-world contexts, but not in the same way as in the classical laboratory.

The online version contains supplementary material available at 10.1007/s00426-026-02238-0.