Deduction may be wrong if we miss the part of human perception. In this paper we design a field experiment to measure the driver's perceptual car width. That car is standing still in front of that driver inside a vehicle cab. We deduce our perceptual car following model based on the results above and also show its connection with the GM's car following model The model we present here is not very like the one based on the general driver's perceptual model. Additionally we establish a car-following simulation model based on our findings. The simulation performances are interpreted in terms of car following stability, shock wave phenomena, and some macroscopic traffic parameters.
