Area III Pilot Studies: Motion Sickness

Title: AIIIP1: Undisturbed Postural Stability. Background: Experiment AIIIE1 found an increase in ataxia (measured in terms of stance breaks from the Sharpened Romberg position) from the first to the third minute when exposed to a circular vection stimulus. The current pilot study measured the background level of stance breaks without the vection stimulus. Hypothesis: In the absence of a vection stimulus, few stance breaks occur under the protocol of Experiment AIIIE1. Nor do the number of stance breaks increase significantly over a period of 3 minutes. Methods: Eight participants were asked to follow the protocol of Experiment AIIIE1 for one 3-minute session, while wearing the Virtual i-O HMD but without being exposed to moving visual stimuli. Five of these participants had their eyes open, simulating the see-through condition. Three had their eyes closed, simulating the occluded condition. Results: Seven of the 8 had no stance breaks at all. The 8th, with eyes closed, had 3 stance breaks in the first minute and 1 in the third minute. Conclusions: The stance breaks found in Experiment AIIIE1 were due to the vection stimulus. The increase in stance breaks from the first to third minutes in Experiment AIIIE1 was due to a build-up of effect, rather than to fatigue.

Title: AIIIP2: Independent Visual Background III (high-end). Background: Experiments AIIIE1 and AIIIE2 report initial findings that an IVB can be useful for reducing simulator side-effects for low-end systems. The current pilot study investigated whether the same is true for a high-end driving simulator, in which the nauseogenic stimulus is much stronger. Hypothesis: Providing an IVB consistent with the inertial rest frame may reduce simulator side-effects, even when the simulator's content-of-interest (CI) is not consistent with the inertial rest frame. Methods: Eight subjects drove figure-eights in the Hughes Research Laboratories driving simulator for 10 minutes at simulated speeds of 15-30 miles-per-hour. Inertial cues were not used. In separate conditions, a gray grid with about 2 spacing was either not present, appeared as an independent visual background (only in the sky, with elements from the simulator scene sweeping over and occluding the grid during turns) or (since it was an easy manipulation to try) in the foreground, as a mesh overlaying the entire simulator scene. In all cases, the grid was stationary with respect to the laboratory. The experimental design was between subjects, to avoid demand characteristics. Simulator sickness questionnaire [55] and ataxia data were recorded. Results: The research time available was too short to reach definitive conclusions. Informal observations included the following: 1. The background grid did not appear to reduce simulator sickness. 2. The background grid was often perceived as counter-rotating in the opposite direction during turns, despite the fact that the background grid was stationary with respect to the laboratory. 3. A foreground grid did appear to be somewhat effective in reducing reported simulator sickness, perhaps by breaking up the scene and thus reducing its believability. Conclusions: The ``induced motion'' of the background grid suggests that the selected rest frame was determined by the CI, not the IVB. In accordance with the presence hypothesis (Section 3.3.3), measuring the amount of induced motion of the background grid may provide a useful perceptual measure for the degree of presence in the CI. Unlike the ``cross-over'' measure of Chapter 4, the ``induced motion'' measure is potentially suitable for interactive environments. Further, the ``induce motion'' measure may provide only a minimal extra load on the participant. See Chapter 8 for a discussion.