Night‐time pedestrian conspicuity: effects of clothing on drivers’ eye movements

Purpose Drivers’ responses and eye movements were assessed as they approached pedestrians at night in order to explore the relative conspicuity benefits of different configurations of retroreflective markings. Methods Eye movements were recorded using an ASL Mobile Eye ( Applied Science Technologies ) from 14 young licensed drivers (M = 24.1 ± 6.4 years) as they drove along a closed-road circuit at night. At two locations, pedestrians walked in place facing either towards or away from the road. Pedestrians wore black clothing with a standard retroreflective vest either alone or with additional retroreflective markers positioned in a configuration conveying biological motion (biomotion). Drivers responded when they recognised that a pedestrian was present and again when they identified the direction the pedestrian was facing. Results Drivers recognised pedestrians from a significantly greater distance (p < 0.05) when the pedestrian's clothing included the biomotion configuration (319.1 m) than when the biomotion markings were absent (184.5 m). Further, the drivers recognised the direction that the pedestrian faced from a longer distance when biomotion markings were present (215.4 m vs 95.6 m). Eye movement data suggested that the biomotion configuration attracted drivers’ attention significantly sooner than the vest (time to first fixation: 1.1 vs 3.5 s), that drivers fixated on pedestrians wearing biomotion for significantly less time prior to responding (3.4 s vs 6.1 s), and the time to first recognise a pedestrian was approximately half that for biomotion compared to vest (6.4 vs 13.9 s). Conclusion Adding biomotion reflectors to the vest facilitated earlier recognition of pedestrians and faster identification of the direction that the pedestrian faced. These findings confirm that the conspicuity advantages of biomotion configurations on pedestrians at night result in part from drivers fixating pedestrians earlier and more efficiently.