Driving experience and the acquisition of visual information

The research presented in this thesis was initially motivated by the excessive accident rates for inexperienced drivers. Researchers have previously attempted to discover what type of experience is gained during driving, and how this reduces accident liability. This research was primarily concerned with the visual acquisition of information during driving, and how this ability varies with driving experience. The first experiment was conducted to assess which of two methods was the better suited to the assess the hypothesis. The results favoured eye tracking drivers in both the laboratory and while actually driving in the real world. On this basis experiments 2 and 3 were conducted. Experiment 2 required participants to drive along a set route while being eye tracked, while experiment 3 measured the eye movements of participants as they watched driving videos in a laboratory hazard perception test. The former experiment revealed experiential differences that extended the findings in the literature. The latter experiment revealed very few experiential differences however. The failure of the hazard perception test to evoke such differences was discussed in regard to the limitations of eye tracking methodology. If experienced drivers have less accidents than their inexperienced counterparts, then one would expect differences to occur in their search strategies. However, if the differences between drivers of varying experience lie within peripheral rather than foveal vision, the straightforward measuring of eye movements may not reveal the true differences. On the basis of the results so far and the literature, it was suggested that experience may allow greater deployment of attention in the peripheral field. Three artificial experiments were undertaken to assess the relationship between foveal demand and eccentricity, before returning to the driving context. In the two final experiments participants of varying driving experience watched the same hazard perception clips previously used in experiment 3. The primary task was either to rate each clip along the dimensions of danger and difficulty, or to press a foot pedal in response to the appearance of a dangerous event. The secondary task required participants to press a button whenever they saw a peripheral target light. Peripheral detection ability was found to degrade with increases in foveal demand (the appearance of a hazard in the hazard perception clips) and eccentricity. Of most importance however was the effect of experience. As drivers gain experience they are able to devote more attention to the peripheral visual field, though the appearance of a hazard degraded peripheral attention across all eccentricities and levels of experience. A detailed analysis of the time line of degradation revealed that though the experienced drivers suffered a greater degradation of peripheral attention with the appearance of a hazard than the less experienced participants, this degradation occurred for only a split second. Learner drivers however suffered the effects of this demand-modulated degradation of peripheral attention for over two seconds. Together these results provide evidence for an attentional skill that modifies the timing and magnitude of attention focusing due to an increase in foveal demand. This is a skill that seems to be learned with driving experience. The implications of these results to pure attention research and driving research are considered