Abstract: Hazard detection refers to the ability of detecting potential hazards on the road. Although many studies have found that experienced drivers outperformed novice drivers on hazard detection in computer-based hazard perception test, some studies have not found such differences. Furthermore, some researchers have attempted to find the relationship between drivers’ response time to the hazards with their sensitivity and response criteria, but with mixed results. Some studies found that drivers’ slower response time to the hazards was due to their low sensitivity while others found that it was strict response criteria that resulted in slower response. Given hazard type was a key factor in hazard perception; it might be an underlying factor that accounts for the inconsistency in previous studies. Additionally, the effects of hazard types on drivers’ sensitivity or/and response criteria have not been largely explored. Thus, the present study aimed to examine the effects of driving experience and hazard types on response time, sensitivity and response criteria and to reveal their relationships under different hazard types. A 2×2 mixed experiment design was employed with driving experience as the between- group factor and hazard types as the within-group factor. Thirty-five novice drivers and 35 experienced drivers were asked to complete a hazard perception task where they were instructed to click the left mouse button quickly when they detect a hazard. Then, participants were asked to finish a signal detection task where they were asked to report orally whether the video clips contained a hazard or not before it was blacked out at a certain point. Hazards in the video clips were classified into overt hazards and covert hazards according to the visibility during their materialization. Sixty video clips were used in the present study and 40 video clips contained either a covert hazard or an overt hazard. The results revealed that experience-related advantage was found on hazard detection. Drivers who had more driving experience responded faster to the hazards, and were more likely to distinguish the hazards and identify what they saw as true hazards. Hazard types did play an important role in hazard detection. Drivers detected overt hazards faster than covert hazards. Furthermore, hazard types affected the relationship between drivers’ sensitivities, response criteria and their response time. Negative correlation was found between drivers’ sensitivity on overt hazard and their response time. Such relationship was not found on covert hazards. Drivers’ response criteria correlated positively and significantly with their response time, regardless of hazard types. Regression analysis showed that drivers’ slower response time to the hazards was due to their strict response criteria rather than their low sensitivities. The present study found hazard type partially determined drivers’ detection of the hazards in the video clips. The findings provided a better understanding for the phenomenon of why novice drivers detected some particular hazards slower than experienced drivers and provided more evidence for the application of signal detection theory in explaining experience-related differences in hazard detection.

Key words: hazard detection, hazard type, response time, signal detection theory

摘要： 结合反应时和信号检测方法，考察不同危险类型下驾驶员危险检测的特点。35名新手和35名有经验驾驶员依次完成一个反应时测试和一个信号检测任务。结果发现：新手对明显和隐藏危险的反应时间比有经验驾驶员长。与隐藏危险相比，驾驶员对明显危险的敏感性高、判断标准低。驾驶员对两类危险的反应时间长是因为他们的判断标准严格。研究结果表明，驾驶员危险检测的特点随着危险类型不同而变化。

关键词: 危险检测, 危险类型, 反应时间, 信号检测