About the Project
The decision-making behind the simple but vital act of crossing the street is hardly understood. On the one hand, it is hard to investigate the behaviour in real-life, while on the other hand, laboratory testing lacks realism. Prior experimental research relied heavily on surrogate judgements (e.g. giving a verbal estimate of when to cross) and on impoverished visual simulations displayed on computer screens (e.g. Hall, 2012; Thomson et al., 2005). Consequently, those findings may not translate to the real world. Particularly problematic is the well-known segregation in the human brain of visual information processing leading to an action and leading to perception. It means that responses in action (crossing) and perceptual judgements are not necessarily correlated. An important aspect of the present project it that it will be testing crossing judgements in action by asking participants to cross a street in immersive virtual reality. We will investigate whether well-known speed illusions impact the safety of road crossing by leading pedestrians to underestimate a vehicle’s speed.
Collisions with vehicles have long been attributed to perceptual illusions (Hall, 2012), but the evidence of their impact on pedestrians decision-making, such as the decision to cross the road, remains anecdotal. In particular, the Brown effect (Brown, 1931) refers to the underestimation of an object’s speed depending on its size. A possibly explanation for this misestimation is the influence of prior knowledge on perceptual inference, given that large objects tend to be slower than small ones. Other biases include the well-known underestimation of an object’s speed with lowered visual contrast.
We will test the relationship between decision the confidence of crossing decisions and speed perception. The effect of reduced visibility is likely to increase pedestrian’s safety margins given a lower confidence for judgements under low visibility, because perceptual uncertainty accompanies perceptual biases. For instance, drivers are more cautious at night, having less confidence about their ability to see other cars. However, this is not necessarily the case for the Brown effect, because car speed is misperceived without affecting perceptual variability. This could lead to normal confidence but poor accuracy. A dissociation between accuracy and confidence could explain why many collisions (e.g. car and train) are attributed to the Brown illusion. We will test those hypothesis by linking perceptual time-to-contact judgements, road crossing actions and confidence judgements.
Essential to the success of this interdisciplinary project is the immersive virtual simulation of road crossing. The participant will cross a virtual road by walking across a large lab-room. By tracking the participant’s head and body movements we will ensure that the virtual simulation mirrors movement in the real world. We will also have the possibility to provide body-awareness in VR by wearing a suit full of position sensors, perfecting the immersive experience. We will also have the possibility to use an eye tracker to monitor gaze, allowing to investigate how information sampling correlates with crossing actions and perceptual estimates.
References
Brown, J. F. (1931). The visual perception of velocity. Psychological Research, 14(1), 199–232.
Hall, R. (2012). Handbook of transportation science (Vol. 23). Springer Science & Business Media.
Pretto, P., Bresciani, J. P., Rainer, G., & Bülthoff, H. H. (2012). Foggy perception slows us down. eLife, 2012(1), 1–12. https://doi.org/10.7554/eLife.00031
Thomson, J. A., Tolmie, A. K., Foot, H. C., Whelan, K. M., Sarvary, P. A., & Morrison, S. (2005). Influence of virtual reality training on the roadside crossing judgements of child pedestrians. Journal of Experimental Psychology: Applied, 11(3), 175–186. https://doi.org/10.1016/j.buildenv.2006.10.027