An explanation of how the placement of traffic signs affects drivers' deceleration on curves

Driving performance is one of the most important areas in the curve safety research because nearly all crashes on curves are associated with inappropriate driving behavior. This study developed a model to illustrate the deceleration behavior in response to the traffic sign on curves, and a simulator experiment was conducted to empirically test the model. The experiment involved three independent variables: one primary variable is the placement of the traffic sign, and two auxiliary variables are curve radius and trial number (the number of trials that the participants have conducted in the experiment). The dependent variable is the first position of releasing the accelerator pedal (FPRA). The results of the experiment showed as the same as the model indicated: when the traffic sign was placed far enough (e.g., 100 m or more) away from a curve, the FPRA was positively correlated with the placement of the traffic sign; however, when the traffic sign was placed near to the curve (e.g., 50 m or less from the beginning of the curve), the FPRA was positively correlated with the curve radius instead of the placement of the traffic sign. In addition, the more times the participants had driven in the scenario, the closer the FPRA to the beginning of the curve. These results imply that deceleration behavior is not only dependent on whether the drivers acquired the information, but also on the confidence level of such information. Moreover, the trial number is also related to the information perception, and influences the deceleration behavior.