Eye-Tracking-Based Evaluation of Cognitive Style and Driving Task Effects on AR-HUD Navigation Interfaces

Area of Science:

• Human-Computer Interaction
• Automotive Engineering
• Cognitive Psychology

Background:

• Augmented reality head-up displays (AR-HUDs) are increasingly common in intelligent vehicles.
• Poor AR-HUD design can increase driver cognitive load and response times to hazards.

Purpose of the Study:

• To investigate the combined effects of cognitive style, driving task type, and AR-HUD design on driver performance and visual attention.
• To provide evidence for dynamic AR-HUD interface optimization.

Main Methods:

• A driving simulation experiment with 38 participants comparing world-fixed (WF) and screen-fixed (SF) AR-HUD interfaces.
• Tested across goal-directed and stimulus-driven tasks, analyzing reaction times and eye-tracking data.
• Participants were screened for cognitive style using the Group Embedded Figures Test.

Main Results:

• Stimulus-driven tasks, especially rear-vehicle scenarios, significantly increased reaction times.
• WF displays reduced fixation duration and counts during lane-change tasks.
• SF displays enhanced attentional efficiency in pedestrian-warning tasks; field-dependent drivers showed higher cognitive workload.

Conclusions:

• AR-HUD interface design significantly impacts driver behavioral performance and visual attention.
• Dynamic optimization of AR-HUDs based on task context and cognitive workload is recommended.
• Findings support sensor-based adaptive interfaces for intelligent vehicles.