University of Leicester
Browse
Sudkamp.Souto_.2023.postprint.pdf (926.54 kB)

The effect of contrast on pedestrians’ perception of vehicle speed in different road environments

Download (926.54 kB)
journal contribution
posted on 2023-01-05, 14:27 authored by J Sudkamp, D Souto

It is well known in the psychophysical literature that low visual contrasts can lead observers to misestimate the speed of moving objects. This effect can have important consequences for traffic safety when navigating under low visibility due to adverse weather conditions (e.g., fog) or visual impairments. So far, road traffic research has primarily focused on the perception of self-motion during driving showing that drivers can both under- and overestimate their own driving speed depending on the spatial distribution of contrast. In two experiments, we used a two-interval forced choice discrimination task to investigate whether pedestrians would be subject to similar biases when estimating the speed of approaching vehicles in a simulated traffic scene. We found that the perceived vehicle speed decreased when the contrast of the view was reduced uniformly but increased when contrast was reduced in a distance-dependent manner, simulating more realistically visibility in fog. The increase of the perceived vehicle speed in simulated fog occurred in bare and visually more complex road environments including either roadside trees or road markings. The origins of such misperceptions, specifically in fog, remain unclear. The temporal integration of motion signals in combination with a lack in speed constancy, and the illusion of acceleration due to the dynamic contrast-change of the vehicle constitute potential explanations that need further investigation.

History

Author affiliation

School of Psychology and Vision Sciences, University of Leicester

Version

  • AM (Accepted Manuscript)

Published in

Transportation Research Part F: Traffic Psychology and Behaviour

Volume

92

Pagination

15 - 26

Publisher

Elsevier BV

issn

1369-8478

Copyright date

2022

Available date

2023-11-18

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC