Increased speeds of trains can be achieved by using tilting trains that decrease the lateral acceleration experienced by passengers on curves, thereby allowing trains to run typically 25-30% faster on existing curved track and maintaining good ride comfort. Unfortunately, motion sickness in tilting trains is a major problem for some passengers. To investigate the incidence of motion sickness and the extent to which different tilt compensation strategies influence its occurrence, tests were conducted with a tilting train on a track with a large number of curves. Eighty healthy volunteers were studied, selected partly for their susceptibility. Three different cars were evaluated during 3 test days, with each test ride lasting about 3 h. On four occasions per test ride, the subjects answered a questionnaire concerning activities during the ride, ride comfort, ability to work and read, vegetative symptoms, fatigue, sleepiness, nausea and well-being. Subjects estimation of average ride comfort and ability to work and read was good in all conditions. However, 10% of the test subjects reported various symptoms of motion sickness (SMS). A 55% degree of tilt compensation of the lateral acceleration instead of the normal 70% reduced the symptoms of motion sickness incidence (SMSI) by 25-40%. SMSI correlated poorly with motion doses, which integrates vertical or lateral acceleration but correlated well with roll acceleration motion dose (r2 = 0.43, p < 0.001). For women, riding backward (p < 0.001) minimized SMSI, but men were insensitive to direction. Future railway design will have to optimize tilt systems by both minimizing motion sickness and avoiding excessive lateral acceleration or jerk.