关键词: 任务性质, 驾驶经验, 驾驶图式, 交通信息, 无意视盲

Abstract:

Drivers’ selective attention causes a significant hazard to traffic safety. Car drivers looking in the direction of other road users do not always perceive the presence of other parties. This is called a “looked-but-failed- to-see” (LBFTS) error. According to the research on attention, task-specific features can make observers form an attention set to guide attention. However, can drivers be prevented from making LBFTS errors by giving them task-specific features? We assumed that task-specific features are only useful for experienced drivers because they have driving schemata. When novice drivers are trained in driving schemata, they can improve their ability to identify traffic information with the help of task-specific features.
In this study, we conducted two experiments. The Tobii T120 eye tracker was used to record participants’ eye movements. Experiment 1 used a driving inattentional blindness task to investigate task-specific features' working conditions, and the drivers' main task was to watch 10 traffic videos and score them. In Dalian, 76 drivers were recruited (mean age = 25.25 years, standard deviation (SD) = 4.41 years). The study adopted a two-factor between-subjects design of 2 (experienced drivers, novice drivers) × 2 (task-specific features: yes, no). The dependent variables were a critical trial’s detection rate, and a total visit duration of traffic signal. Experienced drivers were required to drive a total of 30000 km or more, while novice drivers had to drive a total of 10, 000 km or less. A task-specific feature was observed of whether there were traffic violations. A traffic violation occurred during the critical trial, when the driver was asked whether they had seen a traffic signal after completing the main task. Experiment 2 investigated the effect of driving schema training. The drivers' main task was to watch 30 traffic videos and score them. Sixty-six drivers were recruited (mean age = 24.92 years, SD = 3.85 years). The study adopted a three-factor mixed experimental design of 2 (experienced drivers, novice drivers) × 2 (starting condition of schema training: started, not started) × 2 (event cognitive salience, high: intersection with solid parallel lines; low: unauthorized stop at “No Parking”). The driver type and the starting condition of schema training were inter-subject variables, and the event type was an intra-subject variable. The dependent variable was the driver's correct rate of identifying two kinds of events and the total visit duration of searching for this event.
The results of Experiment 1 showed that the inattentional blindness rate of experienced drivers with task-specific features was significantly lower than that of experienced drivers without task-specific features; however, there was no difference in vision blindness between novice drivers with and without task-specific features. Moreover, the longer the total visit duration, the less likely the inattentional blindness of the drivers. The results of Experiment 2 showed that the recognition performance of the drivers in the schema start group was higher than that in the control group. Total visit duration plays an intermediary role among driving experience, driving schema, and recognition rate of low cognitive salience events.
For drivers, only giving a task feature does not necessarily improve the visual search efficiency during driving, and driving schema plays an important role. Driving schema training can help a task feature make up for the gap brought by driving experience. Driving experience and schemata can improve drivers' performance in identifying traffic information with low cognitive salience by influencing their search time for traffic events.

Key words: task specific feature, driving experience, driving schemata, traffic information, inattentional blindness