Speed perception and actual speed in a driving simulator and real-world: A validation study

Abstract

Background: Driving simulators have become an effective research tool in traffic safety, but the validity of results obtained in simulated environments remains a debated issue of high importance. Objective: The objective of this study is to validate a fixed-base driving simulator for speed perception and actual speed and to support its application in traffic safety studies. Method: The study consisted of two experiments to test the external and subjective validity of the driving simulator in absolute and relative terms. External validity was framed into two parts i.e. for speed perception and actual speed. In the first part, the external validity was assessed based on the speed perception observations from forty volunteers that participated in the study. Speed estimations for four different requested speeds (50, 70, 80 and 100 kph) were recorded under two conditions: speedometer hidden and speedometer revealed. In the second part, the external validity was assessed based on the comparison of actual speed observations from field and simulator. The subjective validity of the simulator setting was assessed through a questionnaire. Results: Results from both experiments showed correspondence of the driving behavior between the simulator and real-world settings. In general, the profiles for estimated speed and actual speed followed a significantly similar tendency and indicated relative validity in both experiments. Moreover, external absolute validity for speed perception was established on all the requested speeds with speedometer hidden while only for the requested speed of 80 kph with speedometer revealed. Participants' evaluation of the quality and performance of the driving simulator supported the subjective validity of the simulator setting. Conclusion: The fixed-base driving simulator used in this study can be considered as a useful tool for research on actual speed and speed perception.