Abstract: Preview is an important part of the driver model. The accuracy of preview time determines the stability and accuracy of the driver model and the whole model. In order to explore the influence of the concentration degree of the driver's fixation on the accuracy of the preview time, eye tracker was used to collect driver's eye movement data. The fixation point percent was used to express the distribution concentration degree of the driver fixation and the preview time interval length was used to describe the preview time accuracy. The relationship between the degree of concentration of drivers in different sections of the road and the accuracy of preview time was explored. The results show that when the elements of the fixation matrix are greater than a certain value in different section roads, the confidence interval of the preview time fluctuates in a certain range. In the course of driving, the higher concentration of driver's fixation is, the more fixed fixation area and the higher the accuracy of the preview time will be.