Can Simulated Driving-Based Tasks Reliably Assess Fitness to Drive?

In this paper we presented SimFit2Drive – a solution for gathering and processing of all important cognitive, sensory, and motor abilities needed for a reliable and accurate stratification based on driving-based tests, which can be performed using a driving simulator, and compare it to a reference system based on standardized battery of psychological tests for assessment of cognitive skills necessary for driving. The goal of the study was the assessment of the SimFit2Drive driving-based tests, specifically focusing on exploring the sensitivity of SimFit2Drive for detection of degradation of cognitive skills. The assessment was completed with a user study in which a test (potentially unfit to drive) and control (fit to drive) group were included in the study. The results showed that 9 out of 21 SimFit2Drive metrics correlated moderately with the observed psychological metrics (above |0.35|), 4 of which showed statistically significant difference between unfit and fit to drive participants. These results revel that the SimFit2Drive metrics could be used to obtain an insight into fitness to drive, and that some of them could be used to detect cognitive impairment. The obtained results also revealed which cognitive abilities necessary for driving are not captured with the SimFit2Drive tests, hence dictating the next steps for their enhancement.