Although there are advantages to using motion-based simulators in driving, flight, marine, and entertainment contexts, one major disadvantage to their implementation is a common side effect known as simulator sickness (SS). The most common theory for the cause of SS is the discrepancy or mismatch between the illusion of motion created by the virtual world, and the artificial motion cue as detected by the inner ear. SS can come with a range of undesirable symptoms, including: headache, drowsiness, nausea, dizziness, vomiting and sweating. Unfortunately, there are conflicting theories that attempt to justify SS onset, and there is no known method for eliminating the problem. It has been postulated that approximately one in three people are susceptible, even in mild motion environments, and that approximately 80% of individuals exposed to virtual reality (VR) simulations for 20 minutes or more are more likely to report sickness symptoms. This is a continued cause for concern. Studies have been conducted whose primary objective was to investigate the various “determinants” attributed to SS (e.g., transport delay, refresh rates, field of view, duration of exposure). Based on previous findings, it is supposed that the resultant effect is due to the combination of these (and various other) classes of factors. Proposed, but to date not fully realized, was a research strategy based on fractional factorial experimentation, suggesting that engineering psychology requires a program of research, not a miscellany of stand-alone experiments. (Simon, 1973; 1976; 1977). Our primary goal is to specify a factorial-based, holistically-designed, human factors pilot study (N=15 participants, 33 experiments) using Orthogonal Arrays, which is a systematic method of testing the impact of various known or suspected contributing factors. Every vector or “experiment” in the array is defined as “orthogonal”, which means that each is statistically independent of the others. Accordingly, the sequence of trials provides uniformly distributed coverage of the test domain. Ideally, this procedure will allow our research team to systematically isolate design factors (and interaction effects) of motion simulators that contribute most to SS. The desired outcome of our work is to develop design guidelines to minimize/mitigate SS in motion-based simulation environments.
