The aim of this study was to investigate whether drivers' fixation times to objects along the road decrease after numerous exposures to the same road. In an earlier study, this was found in a simulated road environment, and the question was whether this could be reproduced in real driving situations. The second goal was to compare these data in real driving with data received from watching a video of the same road. If these data were found to be similar, showing a video can be used for future purposes since this allows a set-up under more controlled experimental conditions. One group of subjects watched a movie scene 4 times during 3 different days, showing different movies filmed on the same road over different rides (12 rides altogether), taped from the eye position of the driver. Another group of subjects was really driving this road 4 times during 3 different days (12 rides altogether). In this, a comparison has been made between the fixation times of subjects watching a movie scene of driving and subjects who were really driving. When watching the video, subjects also had to steer and brake, trying to copy the behaviour of the 'virtual driver' as good as possible. The amount of time subjects spent fixating traffic signs, information signs or markings on the road decreases as subjects see the environment more often. There were no significant differences in the fixation times between watching a video and real life driving, although there was a tendency for longer fixation times when watching the video compared to real driving. The decrease of fixation time over days was somewhat smaller for the video condition. Showing a video is therefore a valid tool for studying what subjects fixate when driving and how fixation times change with familiarity. The absolute fixations times may be somewhat higher than in real life. (Author/publisher)
Abstract