A Framework for Adapting In-Car Touchscreen Interfaces to Driver Behaviors, Perception, and Cognition

Paper Title

A Framework for Adapting In-Car Touchscreen Interfaces to Driver Behaviors, Perception, and Cognition

Publication Info

• Topic area: Adaptive in-vehicle touchscreen interface design for safety and usability.
• Keywords: Adaptive user interfaces, in-car touchscreens, driver cognitive load, human-machine interaction, usability, safety, layout optimization, gaze behavior, co-design, Time-Cost model.

Background and Problem

• Problem / challenge: In-car touchscreens increase functionality but reduce tactile guidance, raising safety concerns due to visual and cognitive demands. Existing adaptive interfaces lack frameworks that balance adaptability with stability, fail to integrate driver-specific profiles and states, and do not provide actionable design guidelines.
• Significance: Addressing these challenges is critical for reducing driver distraction, improving usability, and ensuring safety in increasingly complex in-vehicle systems.
• Motivation and related work: Prior research has explored adaptive interfaces, cognitive load effects, and static design guidelines but has not developed a comprehensive framework that integrates long-term driver profiles with real-time cognitive state adaptations. This paper builds on these gaps to propose a unified framework.

Solution

• Proposed approach: Profile-State Adaptive (PSA) framework.
• Novelty:
  1. Empirical quantification of how cognitive load and interface layout affect touch performance and gaze behavior.
  2. Introduction of the PSA framework, combining long-term profile-based structural adaptations with short-term state-based cosmetic adjustments.
  3. Development of a Time-Cost model for layout optimization based on driver-specific selection probabilities.
  4. Design patterns for real-time cosmetic adaptations that preserve layout stability while enhancing usability.
• Procedure and key techniques:
  1. Observation phase: Driving simulator study to measure the effects of cognitive load and layout configurations on touch performance, gaze behavior, and driving metrics.
  2. Co-design phase: Workshop with automotive UI experts to derive design principles and adaptation policies.
  3. Proposal phase: Operationalization of the PSA framework through a Time-Cost model for profile-based adaptation and design patterns for state-based real-time adjustments.

Results

• Concrete findings:
  • Cognitive load increased touch selection time by 16–20% and led to more frequent but shorter off-road glances.
  • Larger buttons improved selection speed by 0.3 seconds but required more pages, increasing navigation time.
  • The Time-Cost model demonstrated that layouts tailored to driver profiles reduced expected selection times.
• Advantage over baselines:
  • PSA framework balances adaptability with stability, unlike prior systems that either lack real-time adjustments or disrupt spatial predictability.
  • Time-Cost model provides quantitative layout recommendations, improving usability compared to heuristic-based approaches.
• Experiments / evaluation:
  • Driving simulator study with 12 participants under six conditions (three layouts × two cognitive load levels).
  • Co-design workshop with four senior automotive UI/UX designers.
  • Validation of the Time-Cost model using empirical selection-time data and human performance laws (Hick-Hyman and Fitts’s law).
• Limitations and future work:
  • Simplified driving simulator setup; real-world validation in dynamic traffic conditions is needed.
  • Limited co-design iteration due to single-session format; multi-session workshops could refine insights.
  • Future work should explore broader driver states, multimodal feedback, and longitudinal user acceptance studies.

Summary

This paper introduces the Profile-State Adaptive (PSA) framework for in-vehicle touchscreens, integrating long-term profile-based layout optimization with short-term state-based cosmetic adaptations to balance usability and safety. Empirical findings from a driving simulator study informed the development of a Time-Cost model and real-time design patterns, validated through co-design with experts. The framework addresses critical challenges in adaptive interface design by preserving spatial predictability while responding to driver-specific behaviors and cognitive states. Future research will focus on real-world deployment, broader state recognition, and user acceptance.