A Study on the Effects of the Dynamic Features of Light-Based eHMI on Pedestrians' Crossing Behavior.

Higher animation speed of dynamic light eHMIs led to more crossing behaviors among pedestrians.

• Animation speed was identified as the primary dynamic feature influencing pedestrians' objective crossing behavior.
• The effect of animation speed on crossing decisions was more pronounced than that of animation pattern or light-emitting area.
• Participants viewed video simulations of an approaching AV displaying headlight designs combining multiple dynamic features.
• A factorial design framework was employed to systematically investigate the effects of dynamic lighting elements.

Animation pattern and light-emitting area did not play an important role in pedestrian crossing decision-making but influenced visual attention.

• Proper design of animation pattern and light-emitting area can evoke higher visual attention despite not significantly affecting crossing decisions.
• Objective measures of cueing effects were complemented by subjective ratings and user preference study via questionnaires.
• These two features were evaluated through both objective crossing behavior measures and subjective ratings.

Vehicle-pedestrian distance moderated the effect of dynamic lighting features on pedestrians' willingness to cross.

• When vehicle-pedestrian distance was longer, the dynamic features of lighting more strongly affected people's willingness to cross.
• Vehicle-pedestrian distance was integrated as a variable to examine its interaction effect with lighting features.
• Distance was examined across different vehicle motion states (e.g., constant speed vs. other motion conditions).

The effects of light eHMIs were more significant when the AV was travelling at constant speed compared to other motion states.

• Different vehicle motion states were used as conditions in the factorial design framework.
• The interaction between vehicle motion state and eHMI dynamic features was examined in the study.
• This finding suggests that eHMI design effectiveness may depend on the kinematic context of the vehicle.

The study employed a video simulation-based factorial design to evaluate multiple dynamic lighting features simultaneously.

• Participants viewed video simulations of an approaching AV displaying headlight designs combining multiple dynamic features.
• Dynamic elements investigated included animation patterns, animation speed, and light-emitting area.
• Both objective measures (crossing behavior) and subjective measures (ratings and user preference via questionnaires) were collected.
• Vehicle-pedestrian distance was integrated as an additional variable to examine interaction effects with lighting features.