A variety of tools are used to measure the judgment processes which predict the uncertainty. Signal detection theory provides a theoretical framework for forecasting human behavior and decision strategy. The purpose of this study is to apply a practical tool to divide subjects into different sensation seeking groups and examine the effects of risk taking on their driving behaviors. Three experiments are conducted in this study. In the first experiment, subjects are divided into three groups which are risk seekers, risk neutrals, and risk avoiders. Experiment 2 is manipulated to verify the finding of the first experiment. Subjects are requested to complete car following task at a distance chosen by the subjects. Only risk seekers and avoiders are chosen to be the subjects in the third experiment and their driving behaviors are observed and following distance is collected. The results showed that the following distance is closer by the risk seekers than avoiders. Besides, there is a significant difference on decision making in road interaction between risk seekers and avoiders. In this sense, signal detection theory is a reliable tool to predict driving behaviors.