Abstract: To improve the reliability and validity of evaluation results of drivers obtained by using drivers' hazard perception evaluation systems, the effects of evaluation scenarios on evaluation result were discussed. Firstly, the similarities and differences between the evaluation scenarios in various types of evaluation systems were analyzed. Then, as evaluation scenarios are generally dangerous scenarios, the effects of dangerous scenarios on evaluation results were analyzed through systematically reviewing the definition, types and risk degrees of dangerous scenarios. Finally, through analyzing the advantages and disadvantages of the static and dynamic scenarios, the effects of scenarios forms on evaluation results were explored. The results indicate that the higher the consistency between the subjective and objective risks in dangerous scenarios, the better the reliability and validity of the scenarios results, that the static and dynamic scenarios have their advantages and disadvantages, that the static scenarios can effectively avoid the ambiguity of the evaluation results and the driver's tiredness and boredom effect, and that the dynamic scenarios are more conducive to the understanding of dangerous scenarios and are closer to the hazard perception working mechanism when the driver is actually driving.

Key words: hazard perception, evaluation scenarios, driver, evaluation result, reliability and validity