[Paper Review] Designing Touchscreen Menu Interfaces for In-Vehicle Infotainment Systems: the Effect of Depth and Breadth Trade-off and Task Types on Visual-Manual Distraction

Nicolas Louveton, Roderick McCall+1 more|arXiv (Cornell University)|Apr 17, 2024

Human-Automation Interaction and SafetyPsychology3 citations

TL;DR

This study investigates how menu depth and breadth affect visual-manual distraction in in-vehicle touchscreen interfaces during multitasking. Using a driving simulator with eye-tracking and telemetry, it finds that systematic list interaction increases visual demand with depth, while search and memory tasks show an optimal layout favoring breadth over depth, suggesting task-specific design trade-offs are critical for safety.

ABSTRACT

Multitasking with a touch screen user-interface while driving is known to impact negatively driving performance and safety. Literature shows that list scrolling interfaces generate more visual-manual distraction than structured menus and sequential navigation. Depth and breadth trade-offs for structured navigation have been studied. However, little is known on how secondary task characteristics interact with those trade-offs. In this study, we make the hypothesis that both menu's depth and task complexity interact in generating visual-manual distraction. Using a driving simulation setup, we collected telemetry and eye-tracking data to evaluate driving performance. Participants were multitasking with a mobile app, presenting a range of eight depth and breadth trade-offs under three types of secondary tasks, involving different cognitive operations (Systematic reading, Search for an item, Memorize items' state). The results confirm our hypothesis. Systematic interaction with menu items generated a visual demand that increased with menu's depth, while visual demand reach an optimum for Search and Memory tasks. We discuss implications for design: In a multitasking context, display design effectiveness must be assessed while considering menu's layout but also cognitive processes involved.

Motivation & Objective

• To investigate how menu depth and breadth influence visual-manual distraction during in-vehicle touchscreen interactions.
• To examine the interaction between menu structure and cognitive task types (systematic reading, search, memory) on driver distraction.
• To assess driving performance and attentional load using eye-tracking and telemetry data in a simulated driving environment.
• To identify optimal menu layout configurations that minimize visual demand based on task characteristics.
• To inform the design of safer, interruptible in-vehicle infotainment interfaces by integrating cognitive task demands with interface structure.

Proposed method

• Conducted a driving simulation study with 24 participants interacting with eight different menu layout configurations (varying depth and breadth).
• Employed three secondary task types: systematic reading, visual search, and memorization of item states.
• Collected eye-tracking data to measure fixation count, duration, and cumulative fixation time as indicators of visual demand.
• Recorded vehicle telemetry (speed, lateral deviation, inter-vehicle distance) to assess driving performance.
• Calculated a visual demand index combining eye-tracking and usage data to compare layout efficiency across conditions.
• Analyzed data using repeated-measures ANOVAs to assess the effects of menu depth, breadth, and task type on distraction and performance.

Experimental results

Research questions

• RQ1 How does menu depth and breadth affect visual-manual distraction during in-vehicle touchscreen interactions?
• RQ2 How do different secondary task types (systematic reading, search, memory) interact with menu structure to influence visual demand?
• RQ3 Is there an optimal depth-breadth trade-off that minimizes visual distraction for specific task types?
• RQ4 How does driving performance (speed, lateral control, inter-vehicle distance) vary across different menu layouts and task types?
• RQ5 To what extent does task complexity moderate the relationship between menu structure and visual attention?

Key findings

• Systematic reading tasks showed a continuous increase in visual demand with menu depth, with no optimal layout found.
• Visual demand for search tasks reached a minimum at a four-page layout, indicating an optimal balance favoring breadth over depth.
• Memory tasks exhibited an optimal layout at two pages, suggesting that even cognitively demanding tasks benefit from moderate breadth.
• Fixation count, duration, and cumulative fixation time were consistently highest during memory tasks, indicating higher visual attention load.
• Driving performance deteriorated during memory tasks, with reduced speed, increased lateral variability, and shorter inter-vehicle distances.
• Completion time increased with the number of pages for systematic and search tasks, but not for memory tasks, suggesting different cognitive processing patterns.