To be presented at International Workshop on Motion Sickness , Marbella Spain, May 26-28, 1997.
Jerrold D. Prothero, Mark H. Draper, Thomas A. Furness, Donald E. Parker & Maxwell J. Wells
Human Interface Technology Laboratory, University of Washington, Seattle, WA USA 98105
This paper puts forward the "Rest Frame Hypothesis", which addresses how the nervous system performs spatial position, orientation and motion judgments. The hypothesis is quite simple and is implicit in a great deal of existing literature. By making the hypothesis explicit, we suggest that several literatures can be linked in a simple and elegant way. These include the literatures on motion sickness; on the sense of presence in virtual environments; and six classes of visual illusions. The benefits include: the ability to borrow between these literatures; a predicted technique for reducing simulator sickness; Class A ("objective") measures for presence; and a simple display manipulation for increasing presence at a constant field-of-view. Below, we focus on the applications to simulator sickness.
Borrowing from physics, a coordinate system used to define positions, angular orientations and motions is called a "reference frame". The particular reference frame which a given observer takes to be stationary is called the "rest frame" for that observer. (An example of a reference frame which is not a rest frame would be judging the locations of objects with respect to a train which one perceives as moving through the landscape.)
The physical world defines relative motions but not absolute stability. While we usually (but not always) perceive the ground as stationary, in principle our nervous system could equally well create the perception that (for instance) the sun is stationary and that the ground is moving with respect to the sun; or that any arbitrary element of the environment is the stationary point with respect to which the positions and motions of other objects are determined. Likewise, our nervous system could switch on a second-by-second basis our perception of what is and is not stationary.
The above implies that our very strong perception that particular things are stationary is an illusion constructed by our nervous system. The nervous system can simplify its calculations by taking the largest or most important set of consistent spatial cues and defining them to be at rest. It is easier, for instance, to compute the motion of a projectile with respect to the ground than to compute the motion of everything in the environment with respect to the projectile.
We suggest, therefore, that the illusion of stability is quite fundamental to how spatial perception functions. This idea is captured by the "Rest Frame Hypothesis", which states that: "The nervous system has access to many rest frames. Under normal conditions, one of these is selected by the nervous system as the comparator for spatial judgments. We call this the `selected rest frame'. In some cases, the nervous system is not able to select a single rest frame."
Because of the presumed centrality of selected rest frames to perception, we would expect that issues related to selected rest frames would have important consequences. For related work in the measurement of presence (i.e., the sense of "being somewhere", often in a virtual environment), we have put forward the "Presence Hypothesis", which states that: "The sense of presence in an environment reflects the degree to which that environment influences the selected rest frame." (That is, the sense of presence in a virtual environment has to do with switching from using the external environment to define one's sense of position, orientation and motion to using the virtual environment to define one's sense of position, orientation and motion.)
The "Rest Frame" and "Presence" hypotheses together suggest three closely related areas of research, the third of which addresses motion sickness. Area I. Presence reflects selected rest frame decisions. This suggests that "Class A" measures for presence are possible in terms of experiments which create a conflict between real and virtual rest frame cues and gauge their relative effect on the subject's perception (see Prothero et al., 1995A). Area II. What stimuli influence selected rest frame decisions? This leads to techniques for modulating the sense of presence in virtual environments (see Prothero et al., 1995B). Area III. What are the consequences when selected rest frames are inconsistent with some cues (most usually, visual or inertial cues)? This produces at least two interesting subcases. Area III A. Visual cues determine rest frame selection. As discussed briefly below, this appears to lead to a useful way of linking together presence and six classes of visual illusions. Area III B. No functional selected rest frame is formed at all. As a slight refinement to sensory rearrangement theory, we suggest that this condition is the underlying cause of motion sickness.
We briefly discuss Area III A, as it suggests a technique for reducing simulator sickness. We would expect that manipulations to the perceived visual background would be one means to alter the selected rest frame. The visual background generally defines the largest set of coherent cues in the environment. Therefore, calculations performed by the nervous system can be simplified by assigning this set of cues to be the rest frame, and thus as forming the comparator for spatial judgments.
The selected rest frame is hypothesized to underlie our perception of position, angular orientation and motion; both for self and for external objects. Crossing "self" or "external object" with "position", "orientation", or "motion", we would expect that visual background manipulations should produce 6 types of illusions. For 5 of these 6 cases, it is well-known that visual background manipulations can produce the indicated illusion. For orientation, tilting the visual background can produce perceived tilt in the opposite direction, both for self and for objects in the environment. For motion, illusory self-motion induced by visual stimuli is referred to as "vection", whereas illusory motion of objects is referred to as "induced motion". Both kinds of illusions can be created by moving the perceived visual background. For position, the perceived distance to an object can be altered by perspective manipulations in the visual background: for instance, the "corridor illusion". The sixth illusion is self and position. We suggest that all six of these illusions are sub-components of presence. The link between reported presence and the visual background is discussed in (Prothero et al., 1995B).
Traditional sensory rearrangement theory states that "All situations which provoke motion sickness are characterized by a condition of sensory rearrangement in which the motion signals transmitted by the eyes, the vestibular system and the non-vestibular proprioceptors are at variance either with one another or with what is expected form previous experience." (Griffin, 1990; Reason, 1970; Reason, 1978). The Rest Frame Hypothesis suggests that motion sickness does not arise from conflicting motion signals per se, but rather from conflicting rest frames deduced from those motion signals. That is, what is crucial is not the full set of motion cues in the environment, but rather how those motion cues are interpreted to influence one's sense of what is and is not stationary. For instance, if one is seated on a bench watching a flock of birds approaching, one has conflicting motion signals (the birds indicate a relative motion, the inertial cues do not). However, one is very unlikely to become motion sick, because one's perceptual system is unlikely to interpret the flock of birds as defining the stationary rest frame, and, as such, indicating self-motion. Other rest frame cues, from the ground or the sky, are more influential than the flock of birds.
From this point-of-view, the key to avoiding motion sickness is not to remove all conflicting motion cues, but rather to remove those discrepancies which indicate conflicting rest frames. As implied above, visual background manipulations may serve to reduce motion sickness.
A moving visual background (for instance, as provided by a wide field-of-view simulator display) combined with an absence of inertial motion cues is likely to increase motion sickness symptoms. The rest frame indicated by the visual background indicates self-motion, whereas the inertial rest frame cues do not. This suggest that providing a visual background in agreement with inertial cues may serve to reduce motion sickness, even when the foreground cues are not in agreement with inertial cues. The application to simulator design is that providing an "independent visual background" (IVB) which appears behind the simulator's usual content-of-interest (CI) may provide a simple technique for reducing motion sickness. The IVB can be made consistent with the inertial rest frame even if the CI (foreground) is not.
Two experiments were conducted to test the use of an IVB to reduce both reported motion sickness and per-exposure ataxia, using a Virtual i/O head-mounted display in either see-thru or occluded mode. A large effect was found for reduced per-exposure ataxia in the IVB condition in both experiments, and a large effect for reduced reported simulator sickness in one experiment. The results of these experiments, and planned extensions to more sophisticated virtual environments, are presented.
Acknowledgments. Supported by Grant F49620-93-0339 from the Air Force Office of Scientific Research and Grant NAS 0-703 from the National Aeronautics and Space Administration to the University of Washington.
Griffin, MJ (1990). Handbook of Human Vibration. Academic Press.
Prothero, JD, Parker, DE, Furness III, TA & Wells, MJ. (1995A). Towards a robust, quantitative measure for presence. Proceedings of the International Conference on Experimental Analysis and Measurement of Situation Awareness. Embry-Riddle Aeronautical University Press. Available online from http://www.hitl.washington.edu/publications.
Prothero, JD, Hoffman, HG, Parker, DE, Furness III, TA & Wells, MJ (1995B). Foreground/background manipulations affect presence. Proceedings of the Human Factors and Ergonomics Society 39th Annual Meeting. Human Factors Society.
Reason, JT (1970). Motion sickness: a special case of sensory rearrangement. Advancement of Science, 386-393.
Reason, JT (1978). Motion sickness adaptation: a neural mismatch model. Journal of the Royal Society of Medicine, 71, 819-29.