Factors that influence visual attention and their effects on safety in driving : an eye movement tracking approach. Dissertation Louisiana State University.

Statistics show that a high percentage of road related accidents are due to factors that cause impaired driving. Since information extraction in driving is predominantly a visual task, visual distraction and its implications are therefore important safety issues. The main objective of this research is to study some of the implications of demands to human’s attention and perception and how it affects performance of tasks such as driving. Specifically, the study aims to determine the changes that occur in the visual behavior of drivers with different levels of driving experience by tracking the movement of the eye; examine the effects of different levels of task complexity on visual fixation strategies and visual stimulus recognition; investigate the effects of secondary task on attentional and visual focus and its impact on driving performance; and evaluate the implications of the use of information technology device (cellular phone) while driving on road safety. Thirty-eight students participated in the study consisting of two experiments. In the first experiment, the participants performed two driving sessions while wearing a head mounted eye tracking device. The second experiment involved driving while engaging in a cellular phone conversation. Fixation location, frequency, duration and saccadic path, were used to analyze eye movements. The study shows that differences in visual behavior of drivers exist; wherein drivers with infrequent driving per week fixated more on the dashboard area than on the front view (F(3,26) = 3.53, p<0.05), in contrast to the driver with more frequent use of vehicle per week where higher fixations were recorded in the front/center view (F(3,26) = 4.26). The degree of visual distraction contributes to the deterioration of driving resulting to 55% more driving errors committed. Higher time where no fixation was detected was observed when driving with distraction (from 96% to 91% for drivers with less frequency of vehicle use and 55% to 44% for drivers with more frequent use of vehicle). The number of pre-identified errors committed increased from 64 to 81, due to the effect of visual tunneling. This research presents objective data that strengthens the argument on the detrimental effects of distraction in driving. (Author/publisher)