Overestimated time-to-collision for quiet vehicles: Evidence from a study using a novel audiovisual virtual-reality system for traffic scenarios

Abstract

To avoid collision, pedestrians intending to cross a road need to estimate the time-to-collision (TTC) of an approaching vehicle. Here, we present a novel interactive audiovisual virtual-reality system for investigating how the acoustic characteristics (loudness and engine type) of vehicles influence the TTC estimation. Using acoustic recordings of real vehicles as source signals, the dynamic spatial sound fields corresponding to a vehicle approaching in an urban setting are generated based on physical modeling of the sound propagation between vehicle and pedestrian and are presented via sound field synthesis. We studied TTC estimation for vehicles with internal combustion engine and for loudness-matched electric vehicles. The vehicle sound levels were varied by 10 dB, independently of the speed, presented TTC, and vehicle type. In an auditory-only condition, the cars were not visible, and lower loudness of the cars resulted in considerably longer TTC estimates. Importantly, the loudness of the cars also had a significant effect in the same direction on the TTC estimates in an audiovisual condition, where the cars were additionally visually presented via interactive virtual-reality simulations. Thus, pedestrians use auditory information when estimating TTC, even when full visual information is available. At equal loudness, the TTC judgments for electric and conventional vehicles were virtually identical, indicating that loudness has a stronger effect than spectral differences. Because TTC overestimations can result in risky road crossing decisions, the results imply that vehicle loudness should be considered as an important factor in pedestrian safety.