Abstract: The driver’s steering reaction time in the traffic accident is very important significance to drivers’ reliability and traffic accident reasons. The linearity regression is used to construct the model based on the data got from the questionnaire, drivers’ static indexes(deepness aesthesia, attention centrality, attention assignment) and steering reaction time of drivers upon the driving stimulator, and also proved to the validity of the model through data check-up. It is found that it is linear principle among the interrelated factors and reaction time in 95% level; and the preciseness is 79%, so the affecting of other stochastic factors should not be ignored except driver’s psychologic factors ,driving experience and age when driver’s reaction time is studied. This kind of method has offered an effective way for studying driver’s reacting behavior in emergency case.

Key words: Tail-crashing, driver’s steering reaction characteristic, critical reaction time, Linearity regression

摘要： 交通事故中驾驶员反应时间对研究驾驶员可靠性、交通事故致因有着非常重要的意义。本文在调查问卷、驾驶员的三个主要静态指标（深度知觉、注意力集中能力、注意分配能力）的测量以及基于驾驶模拟器的驾驶员反应时间的测定基础上，利用线性回归对驾驶员的转向反应时间建模，并对模型的有效性进行了验证。发现在95﹪水平上，影响因素与因变量反应时间呈线性关系；预测合格率为79﹪，判断出驾驶员反应时间在受到一些心理因素、驾驶经验与年龄的影响外，外界随机因素的影响不可忽略。这种方法为研究驾驶员应急反应行为提供了一条有效途径。

关键词: 追尾碰撞, 驾驶员转向反应特性, 临界反应时间, 线性回归