Small scale driving simulators: A source of realistic data ?

Driving simulators are a common way to assess the impact of car driver behaviour in situations that cannot be safely recreated on real roads, or where it is required to put a range of participants (age, gender, driving experience, etc.) in exactly the same scenario to assess variation in behaviour. Because the experiment participants know that they are not driving a real car however there remains an open question of how realistically drivers are behaving and therefore how representative of true conditions such data can be. Several full scale driving simulators exist where the driver sits inside a real car body and uses the actual steering wheel, pedals, etc. as the controls to maximise the sense of realism and encourage realistic driving. The high development and operating costs of full scale driving simulators however make these inappropriate for smaller studies or where the simulator needs to be quickly moved between locations.This paper therefore reports on a study of driver 'immersion' (the feeling of being present in the simulated rather than the real world) in small scale driving simulators. Participants were seated in a standard office style environment and used combinations of off-the-shelf hardware for both vehicle control (from a keyboard and mouse to a mini steering wheel) and visualisation (from a large screen TV to a Virtual Reality headset) to drive around a simulated suburban and motorway scenario. Results show that even simple combinations of control/display can induce a feeling of immersion and realistic driving within participants, provided that fine levels of vehicle control and vision are possible. This confirms the potential of smaller scale driving simulators to be a valid source of data for quantifying car driver behaviour.