Frequency dependence of allowable differences in visual and vestibular motion cues in a simulator

In the real world in which we move around, inertial and visual motion are usually equal; what you see is what you feel. In a simulator, however, this is usually not the case. On the contrary, due to the relatively small motion space of even the largest simulators, the inertial motion cues must be filtered and scaled down considerably. Typically, inertial motion cues are high-pass filtered and scaled down by as much as fifty to seventy percent in a flight simulator. This does not necessarily mean that the motion simulation is unnatural or unconvincing, since certain differences between inertial and visual motion cannot be detected by the human perceptual system. Especially, when the visual scene has rich content with a lot of detail, high-contrast and a wide field-of-view. The maximum allowable difference between visual and inertial motion that goes undetected is defined as the visual-vestibular coherence zone. Knowledge of these coherence zones is very valuable for the development of effective simulator motion cueing. The results of the yaw motion perception experiment described in this paper provide strong indications that the coherence zone between visual and inertial yaw motion is not simply determined by a simple gain or threshold, but is frequency dependent. From a perception modeling point-of-view one can argue that the frequency dependency is related to the inverse dynamics of the semi-circular canals. The experimental findings support this argument, although further research is required to determine a more precise dynamic coherence zone and to test other degrees-of-freedom. Copyright © 2009 by Mark Wentink, TNO.