LAUSR

A disproportionate number of road deaths occur at intersections where one vehicle is a motorcycle. Previous research has not systematically varied the type of vehicles presented in a controlled environment. We compared drivers’ (n=54) gap acceptance when either a car or motorcycle was approaching a junction. We used a QUEST adaptive staircase to estimate gap acceptance thresholds for cars and motorcycles separately. On each trial, drivers saw a car approaching from the left and a vehicle (car or motorcycle) approaching from the right. The driver had to stop for the car from the left, but could choose whether to pull out in front of the vehicle from the right, or to wait for it to pass. Participants completed the task in either a medium-fidelity simulator (steering wheel and pedals, 180-degree screen) or a high-fidelity simulator (fully instrumented car, 360-degree screen). Participants accepted significantly smaller (riskier) gaps in front of motorcycles than in front of cars, particularly in the high-fidelity simulator. The speed of crossing the junction did not differ between vehicle types, meaning that drivers were closer to the motorcycle than the car during the manoeuvre. There was one instance that appeared to replicate a ‘Look But Fail To See’ error, where a participant pulled out in front of an oncoming motorcycle resulting in a crash. This suggests that drivers accept riskier gaps around motorcycles than cars, which may be due to a difference in attitude towards different vehicles or differences in optic flow properties. These results help to explain the disproportionate involvement of motorcycles in real junction crashes.