Driving behavior in (emergency) hazy weather conditions differs strongly from driving behavior in clear weather conditions. Yet the quantitative changes of driving behavior in emergency conditions and the effects of changes in driving simulator fidelity on driving performance are still unknown. In this study, a high-fidelity driving simulator was used to investigate the impact of hazy weather conditions on driving behavior. Four scenarios with clear and hazy weather conditions and two driving simulator fidelities were set up for the investigation; the weather conditions were defined by the visibilities. The driving simulator fidelities were classified by the degree of freedom (df) of motion platforms: static driving simulator with 0 df and dynamic driving simulator with 8 df. The results showed that the impacts of hazy weather conditions on driving behavior were significant, with substantial reductions in the values of the behavioral variables. Dynamic and static driving simulators give different estimations on the relative behavioral changes, which refer to the average behavioral performance and the heterogeneity in driving behavior, caused by hazy weather conditions. This difference implies that in case of an emergency condition, the fidelity of the driving simulator has significant impacts on the research on driving behavior and should be chosen carefully for studying emergency driving behavior.