A statistical examination of three approaches for predicting motion sickness incidence.

This paper examines three approaches which have been suggested for predicting motion sickness incidence (MSI) for actual or simulated broadband ship motion. Although only a small amount of empirical MSI data exists for motion other than that produced by a single sinusoid, this data provides strong statistical evidence that two approaches based on single sinusoid models fail to produce accurate predictions. Statistical judgments about a least-squares weighting approach must be reserved until enough data exists to provide an adequate test. In view of these results, it must be concluded that much empirical research remains to be done in order to develop a reasonably accurate method of predicting motion sickness incidence for broadband motion.