Supporting Left-Turn Safety: Evaluating Ecological Augmented Reality Head-Up Display Interfaces in Dynamic Traffic Scenarios

With the increasing adoption of augmented reality head-up displays (AR-HUDs) in vehicles, there is a growing need for effective interface design to support driver performance in complex traffic scenarios. While AR-HUDs offer advantages over traditional heads-up displays by presenting dynamic, context-relevant information, they may also increase cognitive load and introduce safety concerns. This study investigates the effectiveness of an ecological interface design (EID) method applied to AR-HUDs for unprotected left turns. Two experiments were conducted, assessing subjective evaluations, behavioral performance, and eye movement metrics. Experiment 1 compared ecological, virtual traffic light, and no interface conditions, while Experiment 2 explored the consistency of ecological interface effects under varying speeds of oncoming traffic. The results indicate that the ecological interface significantly improved drivers’ subjective experience, decision-making speed, and safe turning rates. In addition, it enhanced search efficiency and reduced distraction. However, an interaction effect between vehicle speed and interface presence revealed that ecological interfaces may amplify the perceptual effect of environmental variables. These findings provide practical guidance for the development of AR-HUD systems, emphasizing the value of EID while highlighting the need for balanced integration with environmental cues.