In the previous project SleepEYE, which aimed at developing tools for driver sleepiness detection, two experiments were carried out. In the first experiment, 18 drivers conducted two driving sessions on a motorway: one during daytime (alert) and one at night (sleep deprived). Sixteen of the drivers participated in the second experiment, which was similar to the first one, except that the driving sessions were conducted in an advanced moving base driving simulator. In the present study, the validity of the simulator was assessed using data that was collected in the two experiments. The validation analysis had three focuses: 1) general driver and driving behaviour related to driver sleepiness, 2) speed profile at change in posted speed limit, and 3) drivers’ subjective experience of the vehicle, the driving environment and the own driving performance. The results showed that there were significant differences in mean speed, blink duration, percentage road centre (gaze), radius of 95 per cent cone (gaze) and KSS between the simulator and the real road. No such differences were found for left and right line crossings. For all indicators except blink duration, there was no interaction effect between platform (simulator vs. real road) and time on task, which indicates that the signs of sleepiness vary in a similar way regardless of platform. In conclusion, differences between the simulator and the real road with respect to driving and driver behaviour as well as subjective experiences were found, mainly in terms of speed, level of sleepiness and subjective experience of the vehicle. However, the signs of sleepiness and the development of sleepiness over time are similar in the simulator and on the real road, although the level is higher in the simulator. Thus, there is a relative but not absolute validity with respect to driver sleepiness. Similarly, there seems to be a relative validity for speed and changes in speed.