University of Padova


Eye Tracking on HRT

Project Description

Honda Riding Trainer (HRT) is a compact driving simulator with the commands of a real motorcycle; Honda Motor Europe S.R.L commissioned to the University of Padua the evaluation of the simulator’ efficacy in promoting a safe driving style in different contexts of use (HRT Psychological Evaluation Project).

The project including eye tracking and Honda Riding Trainer is oriented to investigate driving practices using the analysis of eye movements and gaze direction in the virtual environment represented on the Riding Trainer screen; this analysis aims at providing a hint about attentional processes involved in the activity of driving on the road.

Eye movements are often considered indicators of the direction of attention, therefore we can observe and analyse how the abilities of participants in traffic situations change over time and, consequently, we can gain important suggestions on the strategies adopted by drivers immersed in the traffic. The use of the simulator allows us to test these behaviours in a safe setting.

Hazardous situations on the road must be perceived by the driver in time to react appropriately: faults in hazard perception ability are generally considered the major cause of road accidents. Previous researches on hazard perception have tried to define the determinants of this skill in order to develop efficient training programs using the Hazard-Perception Test: this test measures the speed in providing a motor response stimulated by characteristics of the scene that could be interpreted as risky.

Differently from previous studies, we propose oculomotor response (instead of motor response) as a measure of hazard perception ability. When a transient stimulus appears in the periphery of the visual field, the eyes are usually captured and involuntarily oriented to the intervening object (Kean & Lambert, 2003); this response signals that the driver’s attention has been attracted by the approaching vehicle, and therefore driver’s awareness is directed to it (Most & Simons, 2001): For this reason, eye movements can be considered a direct indicator of hazard perception.

Work Description

In this study we tested if and how the ocular response was affected by training. Participants taking part to the experiments were involved in a practise session in order to get acquainted with the commands of the HRT and then they were required to drive in four different environments. The HRT can be set on different kinds of roads and different visibility conditions; however, in our experiment selected routes were limited to urban ones and all data were collected using a daylight condition. During the path, subjects could meet a predefined number of different hazard scenes, different for conditions, causes and type of danger; participants were required to follow the instruction of a guiding voice and to travel avoiding accident and respecting traffic rules.

Results

The analysis of gaze movement experiment have shown that the time taken to perceive an approaching hazard is shorter after the training. This result provides a confirmation on the idea that training improves hazard perception ability. Therefore, the virtual reality environment proved to be a powerful learning instrument: the advantage given by the lack of a real risk can provide an important benefit for people before they face the real situation.

References

– Kean, M. & Lambert, A. (2003). Orienting of Visual Attention Based on Peripheral Information. In – Hyona, J., Radach, R. & Deubel, H. (eds.), The Mind’s Eye: Cognitive and Applied Aspects of Eye Movement Research, 27-47. North-Holland: Elsevier.

– Most, S.B. & Simons, D.J. (2001). Attention capture, orienting, and awareness. In Folk, C. & Gibson, B. (eds.), Attraction, Distraction, and Action: Multiple perspectives on attentional capture. North-Holland: Elsevier.


© 2019 Università di Padova - Tutti i diritti riservati