Postural Instability and Motion Sickness in a Fixed-Base Flight Simulator.
Motion sickness is common in virtual environment (VE) systems that present optical depictions of inertial motion of the user. This visually induced motion sickness (VIMS) has been reported in fixed-base flight and automobile simulation (Frank, Casali, & Wierwille, 1988; Regan & Price, 1994; Yoo, Lee, & Jones, 1997) and in a variety of non-vehicular virtual environments (DiZio & Lackner, 1992; Ellis, 1991). Particularly frustrating for designers and users is the positive correlation between sickness incidence and simulation fidelity: Improvements in simulation fidelity seem to increase the likelihood of sickness (Crowley, 1987; McGuinness, Bouwman, & Forbes, 1981; Miller & Goodson, 1960). The effectiveness of simulation and VE systems, and their acceptance by users, can be severely limited if they produce motion sickness (Biocca, 1992), especially if sickness in simulations occurs when it does not occur in the system being simulated. This provides a practical motivation for understanding the malady. Preventio n of motion sickness would be facilitated if objective measures could be developed to predict it and if the factors that cause it could be identified and eliminated.
Traditional explanations of motion sickness are grounded in the concept of sensory conflict (e.g., Oman, 1982; Reason, 1978). The general premise is that discrepancies between sensory "inputs" and expectations based on past experience constitute "conflict," which must be resolved through processing (i.e., processing changes discrepancy into agreement). The magnitude and/or duration of discrepancy-related conflict are believed to determine the severity and duration of motion sickness. Although the idea of a relation between sensory discrepancies and motion sickness has intuitive appeal, the sensory conflict theory of motion sickness has a number of inadequacies. One of these is its relatively low level of predictive validity (Draper, Viire, Gawron, & Furness, in press; Stoffregen & Riccio, 1991), which reduces the extent to which conflict theory can guide the design of simulators and other virtual environments. This inadequacy is related to a lack of precision in the definition of sensory conflict, which resu lts, in part, from the fact that it is difficult and sometimes impossible to know the internal expectations against which current sensory stimuli may be compared (Stoffregen & Riccio, 1991). This problem has lead Ebenholtz, Cohen, and Linder (1994, p. 1034), to suggest of the sensory conflict theory that "in its present form, it may be untestable."
Postural Sway and Motion Frequency
The incidence of nausea is strongly influenced by the frequency of imposed periodic motion. In vehicles and laboratory whole-body motion devices, motion sickness occurs in the presence of imposed periodic motion at frequencies from 0.08 to 0.4 Hz. Vibration or oscillation at these frequencies is characteristic of ships, trains, and aircraft (Guignard & McCauley, 1991; Lawther & Griffin, 1988). Motion at other frequencies produces little or no sickness, even with exposure durations of up to 12 h (Guignard & McCauley, 1991). This result might suggest that motion sickness is caused by motion in the 0.08-0.4 Hz range, but such a hypothesis cannot be true because the spectral power of spontaneous (unperturbed) standing sway is concentrated between 0.1 Hz and 0.4 Hz (Bensel & Dzendolet, 1968). Given that we are not sickened by our own postural sway, it cannot be the case that vibration in this frequency range is inherently nauseogenic.
The absence of sickness in ordinary stance might be accounted for by positing an ad hoc threshold for nauseogenic conflict (e.g., Oman, 1982). Such a threshold could work if the amplitude of postural sway were low relative to the magnitudes of imposed vibration that are associated with sickness (Stoffregen & Riccio, 1991). This does not appear to be the case: Nausea has been caused by imposed optical oscillations that mimic the amplitude and frequency of postural sway (Lishman & Lee, 1973; Stoffregen, 1985; Stoffregen & Riccio, 1991; Stoffregen & Smart, 1998). Thus the occurrence of visually induced motion sickness in the presence of low-amplitude imposed optical oscillation between 0.08 Hz and 0.4 Hz is a problem for the sensory conflict theory of motion sickness.
Although the relation between motion frequency and motion sickness is well documented in vehicles, it has received little attention in the context of noninertial (fixed-base) flight simulators. One purpose of the present study was to begin to investigate possible relations between motion sickness and the frequency of imposed optical oscillation.
Riccio and Stoffregen (1991) hypothesized that motion sickness results from instability in control of the posture of the body or its segments. Loss of postural stability often is associated with a frank loss of control, such as falling, but this is not the type of instability that Riccio and Stoffregen (1991) discussed in relation to motion sickness. They defined postural stability as "the state in which uncontrolled movements of the perception and action systems are minimized" (p. 202). This definition means that stability may be degraded rather than lost outright. In this definition there can be variation in the magnitude of instability, and instability can persist over long periods without necessarily leading to frank loss of control.
What could cause postural stability to be degraded rather than lost? One possibility is that instability might result from an attempt to control posture in the presence of imposed oscillations whose frequency is between 0.08 Hz and 0.4 Hz. The interaction of the imposed oscillations with the body's natural oscillation frequencies could lead to a form of wave interference (Tipler, 1987). When independent waveforms interact, the results are a function of their relative frequencies. When two systems oscillate at very different frequencies, the resulting waveforms will pass through each other with little effect. However, if the systems oscillate at similar frequencies, the interaction of the waveforms can lead to dramatic instabilities in both frequency and amplitude. Thus imposed oscillations in the frequency range of spontaneous sway may destabilize the postural control system in frequency and/or amplitude through a wave interference effect. This hypothesis would explain why sickness is associated with imposed motion in the narrow band of frequencies that is spontaneously produced by postural sway.
The postural instability theory predicts that postural instability should exist before the onset of subjective symptoms of motion sickness. This prediction was evaluated by Stoffregen and Smart (1998). Standing participants were exposed to optical flow produced by a "moving room" (e.g., Stoffregen, 1985). The amplitude and frequency of room motion resembled the amplitude and frequency of body sway in an unperturbed stance. Recordings of postural sway during exposure to the imposed optical flow revealed increases in postural sway prior to the onset of subjective symptoms of motion sickness, as predicted. Significant increases were observed in the amplitude of postural motion (operationalized as the variability of head position) and in the velocity and range of postural motion.
Extension to Operational Systems
Stoffregen and Smart (1998) studied upright stance and exposed participants to the very small amplitudes of motion that characterize spontaneous postural sway. For this reason their study is of limited relevance to motion sickness in operational VE systems in general and in flight simulation in particular. In the present study we sought to evaluate the postural instability theory of motion sickness in a situation that is more representative of operational flight simulators. The experiment was conducted using a fixed-base flight simulator. Participants were seated during exposure to imposed optical motion. In addition, the visual scene oscillated in the roll axis, so that it resembled aircraft roll resulting from air turbulence. We also increased the amplitude of imposed motion to make it more representative of oscillations observed in flight. Flight simulators are widely considered to be a class of virtual environment. However, we do not assume that the results of the present study would generalize to other types of virtual environments; that is open to further empirical evaluation.
In the present experiment, we regard humans as adaptive nonlinear systems (Riccio & Stoffregen, 1991). Adaptive refers to the general tendency of living things to adopt functional changes in behavior on the basis of experience. An important consequence of adaptive systems is that their behavior is characterized by variability. In nonlinear systems, there is no linear relation between inputs (e.g., perceptual information) and outputs (e.g., action). Because of the complexity and high variability of adaptive nonlinear systems, instability in such systems is not yet well defined (Canudas de Wit, 1988). One consequence of this is that there is no clear operationalization of the postural instability theory. Riccio and Stoffregen (1991) discussed a variety of parameters of postural motion in which instability might be observed. In the present study we evaluated a subset of these parameters: the variability, velocity, and range of postural motion.
A total of 14 persons (8 males and 6 females) participated. All reported normal or corrected-to-normal vision. None had eaten for at least 1 hr before exposure time in the simulator. Participants ranged in age from 20 to 42 years, with a mean of 28 years and were paid $7.50/hr. They gave informed consent for their participation and were aware that the focus of the study was motion sickness.
The experiment was conducted at the Synthesized Immersion Research Environment (SIRE) facility, located at Wright-Patterson Air Force Base, Ohio. We employed the fixed-base, spherical dome, visual flight simulator in SIRE (Figure 1). The dome has a radius of 6.1 m and surrounds an F-16 cockpit. For ingress and egress the cockpit rested on the floor of the laboratory. For operations it was raised approximately 2.3 m, so that the participant's head was at the geometric center of the dome.
The visual imagery was created using a Silicon Graphics Onyx computer image generator (Silicon Graphics, Mountain View, CA). The Onyx had eight 150-MHz R4400 microprocessors and three Reality Engine II graphics pipelines, each running in a multichannel option. The nominal resolution of the dome display was approximately 25 pixels/degree both horizontally and vertically. The visual imagery was portrayed on the spherical dome using a SEOS six-projector system covering an area of 150[degrees] horizontally and 70[degrees] vertically (SEOS Displays Limited, Burgess Hill, England).
The visual stimulus consisted of a space vehicle target (similar to a Tie fighter from Star Wars) against a starfield background. The dimensions of the space vehicle were 24.4 m wide x 24.4 m high x 27.4 m deep. The space vehicle target was projected to simulate a distance of 91.4 m from the participant, resulting in a visual angle of approximately 15[degrees] X 15[degrees]. During experimental trials the starfield and the space vehicle oscillated together in the roll axis. In the 0.2 Hz display the display oscillated [+ or -] 30[degrees] with a frequency of 0.2 Hz and a duration of 60 s. In the sum-of-sines display, the display oscillated as a sum of 10 sine waves, with component frequencies of 0.0167, 0.0416, 0.0783, 0.1050, 0.1670, 0.1800, 0.1900, 0.2200, 0.2600, and 0.3 100 Hz, each having an amplitude of 30[degrees]. These were combined to create a nonrepeating waveform with a duration of 600 s. The phase and amplitude of the component sine waves were adjusted so that the combined waveform had a maximum amplitude of [+ or -] 45[degrees]. The motion stimuli are illustrated in Figure 2.
Head movement was measured using a magnetic tracking system (Flock of Birds, Ascension Technology, Inc., Burlington, VT). The system consisted of an emitter that created a low-intensity magnetic field of known strength, extent, and orientation. Receivers moved within this field. The system could detect the position of receivers relative to the emitter in 6 degrees of freedom (anterior-posterior or AP, lateral and vertical translation, and pitch, roll, and yaw rotation) with position accuracy of 1 mm linear and 0.1[degrees] angular. One receiver was attached to headgear worn by the participant, and the magnetic emitter was placed on a stand immediately behind the participant's head. The receiver was sampled at 20 Hz. Signals from the receiver were stored on disk for data analysis.
A within-participants design was used. Each participant was exposed to the same number and type of trials, in the same order. The duration of exposure for each participant was approximately 2 hr. The full sequence of trials is given in Table 1.
In addition to the trials featuring display motion (stimulus trials), we included several trials in which we collected data on participants' spontaneous head motion in the absence of any imposed optical motion (Trials 1, 2, 10, and 11). In these spontaneous sway trials, participants were seated in the SIRE cockpit. In Trials 2 and 11, their eyes were closed. In Trials 1 and 10, participants fixated the nose of the space vehicle. The 0.2-Hz trials and sum-of-sines trials were run as indicated in Table 1. The single, 10-mm, sum-of-sines display was repeated on each of the four sum-of-sines trials. The trials and their order were identical to those of Stoffregen and Smart (1998).
At the beginning of the experimental session the participant was briefed on the experiment and asked to read and sign the consent form. Participants were asked to stand on one leg and were accepted only if they could do so for 30 s. Participants provided information about their state of fitness, recent illnesses, chemical and alcohol consumption, sleep habits, flight experience, previous motion sickness symptomology in moving environments and simulators, history of symptoms of nausea and dizziness, and willingness to participate in experiments on motion sickness.
The potential risk of motion sickness was explained, and participants were encouraged to discontinue participation if they became symptomatic. They were instructed to notify the experimenter at the immediate onset of symptoms, even if onset occurred during a trial, at which time the trials would be stopped with no financial penalty. Between trials they were questioned about their well-being. Participants were informed that the SIRE facility is a fixed-base simulation environment, and all participants understood that the cockpit would not move during trials.
The receiver from the Flock of Birds was attached to a headset worn by the participant, and the emitter was securely mounted on the cockpit superstructure behind the participant's head. The participant was seated in the cockpit, which was then raised to the design eye point (the position at which the display geometry had its greatest fidelity). In this position the participant was approximately 6.1 m from the display dome. Participants were instructed to fixate the nose of the space vehicle in the display. During trials participants were instructed to hold their head and body in a natural position while focusing on the nose of the space vehicle. They were asked to refrain from talking unless they became uncomfortable. Communication with the experimenter was achieved via the headset. The experimenter remained in the dome area during data collection, in a position that allowed him or her to observe the participant.
Determination of Motion Sickness Incidence
Participants were classified as being either sick or well, based on oral self-reports of symptoms and on the experimenter's judgment based on observable symptoms. Visible changes in symptomology were noted by the experimenter and recorded between experimental trials. Experimenter judgments of participant sickness were based on established procedures in use in the SIRE facility and focused on self-reports such as dizziness and stomach awareness, together with readily observable characteristics such as pallor, unsteadiness, and fear or confusion. Participants were instructed to inform the experimenter immediately if they became incapacitated at any time and that they would be permitted to immediately discontinue the experiment without penalty of any kind. Laboratory safety procedures required that participation be contingent on a self-report and on the experimenter's judgment that the participant was well enough to begin and to continue.
The two experimenters combined had more than three years' experience collecting data in the SIRE laboratory, principally in studies of alternative control technologies, in which motion sickness was common. Experimenter judgments were made prior to and independent of any analysis of postural motion data. Participants were permitted to leave the lab only when they appeared to be free of symptoms and ataxia (this determination was not made on the basis of Simulator Sickness Questionnaire [SSQ] scores; Kennedy & Lane, 1993). The longest delay after leaving the cockpit was 1 hr.
Motion sickness symptoms were quantified using the SSQ. Each participant filled it out before beginning the experiment, immediately after the stance duration test at the termination of the final sum-of-sines trial, and 1 hr later. They were also given a printed version of the SSQ to be completed at 2, 6, 12-18, and 24 hr after leaving the laboratory and asked to return it by mail.
Motion Sickness History
Of the 14 participants, 3 reported having experienced motion sickness in the past (2 developed sickness in the experiment, 1 did not). Participants reported experiencing sickness in a variety of situations, including on amusement park rides and while reading in a car. Of the remaining participants, 8 stated that they were not susceptible to motion sickness (2 later became sick). The remaining participants reported minimal susceptibility to motion sickness. No participants had any prior simulator experience. One participant had 320 hr of fixed-wing flight experience; this participant did not become sick.
Motion Sickness Incidence
The sick group consisted of 6 participants (43%), leaving 8 in the well group. In the sick group, 2 participants stated that they were motion sick (P11 and P14), whereas 4 others were classified as sick by the experimenter. Prior to the experimenter's viewing of the posttest SSQ scores, 5 participants (including P11 and P14) were judged to be "highly at risk for motion sickness," and 1 was judged to be "at risk for motion sickness"; all 6 were included in the sick group. Each participant in the well group was judged by the experimenter as being "not at risk." In each case the posttest SSQ scores were consistent with the experimenter's judgments.
A total of 12 participants completed all experimental trials; the remaining 2 requested discontinuation because of the onset of motion sickness. Participant P11 discontinued after 16 s of the first sum-of-sines trial (Trial 5), and P14 discontinued after completing the first sum-of-sines trial. Among those classified as being sick, only one (P15) overtly denied being motion sick. However, this participant acknowledged the presence of stomach awareness and dizziness and was observed to have heavy, rapid breathing and distorted speech. Other participants in the sick group did not specifically report motion sickness but did report feelings of light-headedness, aberrant motion of the stimulus, uneasiness, drowsiness, and eye strain, and they were visibly unstable or shaky.
As noted earlier, participants were asked to complete the SSQ prior to, immediately after, and 1 hr after exposure to the simulation. The SSQ exists in both long and short forms (Kennedy & Lane, 1993), which consist of 16 and 25 items, respectively. Because of an experimenter error, prior to exposure to the experimental stimuli the long form of the SSQ was administered to some participants and the short form to others (only the short form was used for postexposure evaluations). The short form of the SSQ used a four-point scale (0=none, 1 =slight, 2=moderate, 4=severe). The long form of the SSQ used both the four-point and a yes-(1)/no-(0) scale on different items. Because of the use of different sickness questionnaires, group mean scores were calculated as follows (as recommended by R. S. Kennedy, personal communication, April 1999): Items that were exact matches in the two questionnaires were used as is. For items that used a four-point scale in the short version but a yes/no response in the long version, ye s responses in the long version were scored as 1 (i.e., slight). The remaining items in the long version of the questionnaire were not used.
Kennedy and Lane demonstrated that the extra items did not account for a significant amount of variance independent of the items that were used in the short form. However, care must be taken in interpreting these scores, as participants' responses may have been influenced in some manner by the additional items in the long form. Thus we analyzed only the total severity scores. The SSQ data are summarized in Table 2.
Total severity scores on the SSQ were evaluated using a Kruskal-Wallis test. Analysis of the pretest questionnaires showed that the sick and well groups did not differ prior to exposure to the simulator, [[chi].sup.2](1) = 1.56, p [greater than] .05. Immediately after completion of the experiment, questionnaire scores were higher for the sick group, [[chi].sup.2](1) = 9.45, p [less than] .05. This was also true 1 hr after completion, [[chi].sup.2](1) = 7.81, p [less than] .05.
Motion Sickness Aftereffects
Participants were instructed to fill out the motion sickness questionnaire at 2, 6, 12-18, and 24 hr after exposure to the simulation. Each period was analyzed separately using the Kruskal-Wallis test. For questionnaires filled out 2 hr after leaving the laboratory, there was a significant effect of group, [[chi].sup.2](1) = 6.59, p [less than] .05. Total severity scores were higher for the sick participants (M = 13.36, Mdn = 9.35) than for those in the well group (M = 0.53, Mdn = 0.0). No significant effects were obtained in the questionnaires filled out 6, 12-18, or 24 hr following the experiment.
Pre-exposure spontaneous sway. The dependent variables are discussed separately. We first analyzed the variability of postural motion. Separate one-within (vision), one-between (group) analyses of variance (ANOVAs) were performed for each variable in each axis. There were no effects of vision (eyes open vs. eyes closed) in any of the 6 degrees of freedom. In addition there were no group effects in any of the 6 degrees of freedom.
We next analyzed the velocity of postural motion. There were no significant effects of vision. There were significant effects of group in the yaw axis, F(1, 12) = 5.14, MS = 0.91, p [less than] .05, and in the roll axis, F(1, 12) = 4.98, MS = 1.12, p [less than] .05, accounting for 30% and 32% of the variance, respectively. However, these variables were strongly correlated (r = .76), so it is unlikely that the two effects account for independent portions of the variance (Pedhazur, 1997). A conservative estimate would be that the two variables jointly accounted for 32% of the variance. This is comparable with the variance accounted for by tests of perceptual style and by surveys of motion sickness history; it is much larger than the variance accounted for by physiological measures (Kennedy, Dunlap, & Fowlkes, 1990). In each case velocity was greater in the sick group (yaw, M = 1.33[degrees]/s, SE = 0.12[degrees]/s; roll, M = 1.15[degrees]/s, SE = 0.16[degrees]/s) than in the well group (yaw, M = 0.97[degrees] /s, SE = 0.11[degrees]/s; roll, M = 0.74[degrees]/s, SE = 0.03[degrees]/s). Representative trials are illustrated in Figure 3. There were no effects of group or vision on the range of sway.
0.2-Hz trials. Representative trials are illustrated in Figure 4. Separate one-within (vision), one-between (group) ANOVAs were performed for each variable in each axis. We first analyzed the variability of postural motion. There were effects of vision in three axes; lateral, F(1, 12) = 6.09, MS = 0.85, p [less than] .05; roll, F(1, 12) = 5.95, MS = 17.27, p [less than] .05; and yaw, F(1, 12) = 6.28, MS = 2.64, p [less than] .05, accounting for 36%, 35%, and 34% of the variance, respectively. Motion in the three axes was highly correlated, [r.sub.Roll/Lateral] = .91, [r.sub.Roll/Yaw] = .81, [r.sub.Lateral/Yaw] = .74, so a conservative estimate would be that the vision factor accounted for approximately 35% of the variance overall.
In each of the three axes, variability was greater when the eyes were open: lateral, [M.sub.open] = 0.41 cm, SE = 0.14 cm, [M.sub.closed] = 0.06 cm, SE = 0.01 cm; roll, [M.sub.open] = 1.94[degrees], SE = 0.63[degrees], [M.sub.closed] = 0.37[degrees], SE = 0.09[degrees]; yaw, [M.sub.open] = 1.06[degrees], SE = 0.23[degrees], [M.sub.closed] = 0.45[degrees], SE = 0.12[degrees]. The effect in roll replicates similar effects in studies on postural control in the presence of imposed optical flow (e.g., Stoffregen, 1985), and confirms that control of seated posture was influenced by the displays.
There was a group effect on variability in the vertical axis, F(1, 12) = 5.06, MS = 0.011, p [less than] .05, accounting for 30% of the variance. Variability was greater in participants who later reported motion sickness, [M.sub.Sick] = 0.06 cm, SE = 0.02 cm; ME = 0.02 cm, SE = 0.01 cm. There were no group effects in any of the other 5 degrees of freedom. Stoffregen and Smart (1998, Experiment 1) found group effects in the AP and lateral axes, with increases in motion among participants who later reported motion sickness.
We next analyzed the velocity of postural motion. There was a significant effect of vision in the lateral axis F(1, 12) = 5.50, MS = 5.41, p [less than] .05 (accounting for 31% of the variance), with greater velocity in the eyes-open condition, M = 1.15 cm/s, SE = 0.37 cm/s, than when the eyes were closed, M = 0.26 cm/s, SE = 0.02 cm/s. There were no other significant effects and no effects of group. Stoffregen and Smart (1998, Experiment 1) found significant increases in the sick group in the velocity of both AP and lateral motion.
Our next analysis was of the range of postural motion. There was a significant effect of vision for vertical motion, F(1, 12) = 7.88, p [less than] .05 (accounting for 40% of the variance), with greater range when the eyes were open, [M.sub.open] = 2.38[degrees], SE = 0.63[degrees]; [M.sub.closed] = 0.56[degrees], SE = 0.14[degrees]. There was also a significant effect of vision on pitch motion, F(1, 12) = 6.96, p [less than] .05 (accounting for 37% of the variance), with greater range when the eyes were open, [M.sub.open] = 2.01[degrees], SE = 0.59[degrees]; [M.sub.closed] = 0.40[degrees], SE = 0.12[degrees]. Range in the pitch and vertical axes was highly correlated (r = .91), so that a conservative estimate of the overall variance in range accounted for by the vision factor would be approximately 38%.
There was a significant effect of group for range in the roll axis F(1, 12) = 5.18, p [less than] .05 (accounting for 30% of the variance). Participants who became sick (M = 2.8[degrees], SE = 0.06[degrees]) exhibited greater motion than did the well participants (M = 1.0[degrees], SE = 0.05[degrees]).
Our final analysis was of the cross-correlation between postural motion and motion of the display. Mean cross-correlations were very low ([r.sub.Sick] = -.072, [r.sub.Well] = -.067), and there were no significant effects of vision or group.
Sum-of-sines trials. Our hypothesis was that postural instability should precede the onset of subjective symptoms of motion sickness. In evaluating this hypothesis, it was critically important to ensure that we excluded from the analysis all postural motion that occurred after the onset of subjective symptoms. To avoid analyzing postural motion that occurred after the onset of subjective symptoms of motion sickness, we analyzed only the first sum-of-sines trial (Trial 5). Because participant P11 discontinued early in Trial 5, his data were deleted from this analysis. Thus the analyses of postural motion for the sum-of-sines trials included only five of the six participants in the sick group.
Our first analysis was of the variability of postural motion. It revealed no significant effects of group in any of the 6 degrees of freedom. However, there were group effects on the velocity of postural motion. There were group effects on velocity in the roll axis, [t.sub.11] = 2.62, p [less than] .05, and in the lateral axis, [t.sub.11] = 2.86, p [less than] .05. In both cases motion was greater for the sick group: roll, [M.sub.Sick] = 1 .65[degrees]/s, [M.sub.Well] = 0.81[degrees]/s; lateral, [M.sub.Sick] = 0.38 cm/s, [M.sub.Well] = 0.28 cm/s. Representative trials are illustrated in Figure 5. There were no significant effects of group on the range of postural motion or on the simple correlation between postural motion and motion of the display.
In this experiment visually induced motion sickness was produced in a fixed-base flight simulator. Sickness was observed in response to imposed optical oscillations in the roll axis. During exposure to the sum-of-sines optical oscillation, increases were observed in the axis of stimulation (roll) but also propagated beyond this to the lateral axis (Figure 5); these differences were observed prior to the onset of subjective symptoms of motion sickness. In addition, differences between participants who later reported sickness and those who did not were observed prior to exposure to the sum-of-sines stimulus. These differences occurred during spontaneous sway (Figure 3), and during exposure to a simple sinusoidal optical motion (Figure 4). Overall the data indicate that postural instability preceded motion sickness. The data also suggest that there may be general differences in postural motion between persons who are susceptible to motion sickness and those who are not.
Motion Sickness Preceded by Postural Instability
Prior to the onset of motion sickness, there were significant differences between sick and well participants on a number of parameters of postural motion, as predicted by Riccio and Stoffregen (1991). This replicates a finding of Stoffregen and Smart (1998) and provides further support for the postural instability theory of motion sickness. The present results extend the earlier findings to the control of the head during seated posture, to nauseogenic stimuli in the roll axis, to amplitudes of imposed motion that characterize operational systems that are associated with motion sickness, and to computer-generated imagery.
There were also significant differences between eyes-open and eyes-closed conditions. These have no significance for theories of motion sickness but simply verify that posture was influenced by the visual displays.
Postural motion predicted motion sickness during exposure to the sum-of-sines oscillation, as evidenced by the significant increases in the velocity of head motion among those who later became motion sick. However, the power of postural motion in predicting later motion sickness was not limited to the period of exposure to the nauseogenic stimulus. Motion sickness was predicted by postural motion in response to a relatively nonnauseogenic optical stimulus (the 0.2-Hz oscillation) and also by spontaneous sway prior to exposure to any experimental motion. Stoffregen and Smart (1998) reported both of these effects. The replicated finding of differences in spontaneous sway suggests that there may be a priori differences in postural control between individuals who are susceptible to motion sickness and those who are not. We would interpret this as a relation between characteristics of spontaneous sway and the ability of individuals to adapt to destabilizing imposed motion. This possibility could be evaluated by ex amining spontaneous postural sway prior to exposure to a wide variety of nauseogenic situations.
The exact pattern of instabilities reported by Stoffregen and Smart (1998) was not fully replicated. Most notably, in the present experiment motion sickness was not predicted by changes in the variability of postural motion during exposure to the sum-of-sines stimulus. This is not problematic for the postural instability theory because Riccio and Stoffregen (1991) did not predict that instability would always be confined to a single parameter or set of parameters of postural motion. Riccio and Stoffregen pointed out that the concept of instability in adaptive, nonlinear systems (e.g., humans) is not well developed. They suggested that instability could exist in a wide variety of parameters of postural motion (they gave 13 specific examples), any one of which might predict the onset of motion sickness. The different pattern of instabilities observed in the present study and in Stoffregen and Smart should help to motivate the search for a wider range of variables in which instability might predict motion sickne ss.
One participant developed motion sickness symptoms and discontinued after only a few seconds of exposure to the sum-of-sines stimulus.
Sickness might have been induced by this brief exposure, but it might also have resulted from exposure to the 0.02-Hz trials. This is consistent with several reports of motion sickness induced in standing participants by imposed optical flow limited to a simple sine wave oscillation (e.g., Lestienne, Soechting, & Berthoz, 1977; Lishman & Lee, 1973; Stoffregen, 1985).
Other theories make no predictions about postural motion prior to symptom onset (e.g., Ebenholtz, Cohen, & Linder, 1994; Oman, 1982). Thus our findings pose a challenge to such theories. For example, on what basis might sensory conflict be predicted to elicit postural instability prior to the onset of motion sickness?
Implications for Design
The present findings, and the postural instability theory of motion sickness that motivated the study, may have implications for the design and use of flight simulators. It may be possible to incorporate into such systems hardware and software that monitor postural control and use this information to predict and/or prevent motion sickness in users. This might be done in at least three ways.
First, measures of unperturbed posture might be used to predict susceptibility before an individual begins to use a given simulator. Such information could be used to screen out individuals who would be most likely to develop sickness as a consequence of simulator use, or to advise them that in using the system, they would be at risk for motion sickness.
Second, the postural control of users could be monitored online; deterioration in postural control (increases in instability) could be used to terminate use before the onset of motion sickness, or to advise users of the imminent threat of sickness.
Finally, on-line monitoring of users' postural control might be used to prevent sickness while allowing uninterrupted use of the system. This might be possible if information about developing instability were used to modify the dynamics of the simulator, online, so as to promote stable posture. For example, instability might be reduced by enabling a band-pass filter that would suppress oscillation in the 0.08-0.4 Hz range. One potential negative impact of such a modification is that it might significantly reduce the fidelity or realism of the simulator. Another potential negative impact might be the suppression of perceptual-motor learning that normally leads to adaptation to nauseogenic stimuli.
Implementation of each of these design changes would depend on further research on the exact nature of the simulated motions that give rise to postural instability, and on the number and identity of parameters of postural motion in which instability leads to motion sickness.
Thomas A. Stoffregen received a Ph.D. in human experimental psychology from Cornell University in 1984. He is an associate professor in the Department of Psychology at the University of Cincinnati and director of the Postural Stability Laboratory, which is part of the Human Factors Research Focus.
Lawrence J. Hettinger received a Ph.D. in psychology from Ohio State University in 1987. He is director of human factors and ergonomics at Logicon Technical Services, Inc.
Michael W. Haas received a Ph.D. in engineering and applied science from the University of Southampton, United Kingdom, in 1996. He is the technical director of the Human Interface Technology Branch, Human Effectiveness Directorate, Air Force Research Laboratory.
Merry M. Roe received a B.S. in psychology from Wright State University in 1985. She is a human factors psychologist in the SIRE facility at Wright-Patterson Air Force Base, where she is employed by Logicon Technical Services, Inc.
L. James Smart received a Ph.D. in psychology from the University of Cincinnati in 2000. He is an assistant professor of psychology at Miami University, Oxford, OH.
Bensel, C. K., & Dzendolet, E. (1968). Power spectral density analysis of the standing sway of males. Perception & Psychophysics, 4, 285-288.
Biocca, F. (1992), Will simulation sickness slow down the diffusion of virtual environment technology? Presence, 1, 334-343.
Canudas de Wit, C. A. (1988). Adaptive control for partially known systems: Theory and applications. Amsterdam: Elsevier.
Crowley, J. S. (1987). Simulator sickness: A problem for Army aviation. Aviation, Space and Environmental Medicine, 58, 355-357.
DiZio, P., & Lackner, J. R. (1992). Spatial orientation, adaptation, and motion sickness in real and virtual environments. Presence, 1, 319-328.
Draper, M. H., Viire, E. S., Gawron, V. J., & Furnedd, T. A. (in press). The effects of virtual image scale and system delay on simulator sickness within head-coupled virtual environments. Human Factors.
Ebenholtz, S. M., Cohen, M. M., & Linder, B. J. (1994). The possible role of nystagmus in motion sickness: A hypothesis. Aviation, Space and Environmental Medicine, 65, 1032-1035.
Ellis, S. R. (1991), Nature and origins of virtual environments: A bibliographic essay. Computing Systems in Engineering, 2, 321-347.
Frank, L. H., Casali, J. G., & Wierwille, W. W. (1988). Effects of visual display and motion system delays on operator performance and uneasiness in a driving simulator. Human Factors, 30, 201-217.
Guignard, J. C., & McCauley, M. E. (1991). The accelerative stimulus for motion sickness. In G. H. Crampton (Ed.), Motion and space sickness (pp. 123-152). Boca Raton, FL: CRC Press.
Kennedy, R. S., Dunlap, W. P., & Fowlkes, J. E. (1990). Prediction of motion sickness susceptibility. In G. H. Crampton (Ed.), Motion and space sickness (pp. 179-216). Boca Raton, FL: CRC Press.
Kennedy, R. S., & Lane, N. E. (1993). Simulator sickness questionnaire: An enhanced method for quantifying simulator sickness. International Journal of Aviation Psychology, 3, 203-220.
Lawther, A., & Griffin, M. J. (1988). Motion sickness and motion characteristics of vessels at sea. Ergonomics, 31, 1373-1394.
Lestienne, F., Soechting, J., & Berthoz, A. (1977). Postural readjustments induced by linear motion of visual scenes. Experimental Brain Research, 28, 363-384.
Lishman, J. R., & Lee, D. N. (1973). The autonomy of visual kineasethesis. Perception, 2, 287-294.
McGuinness, J., Bouwman, J. H., & Forbes, J. M. (1981). Simulator sickness occurrences in the 2E6 Air Combat Maneuvering Simulator (ACMS; NAVTRAEQUIPCEN 80-C-0135-4500-1). Orlando, FL: Naval Training Equipment Center.
Miller, J. W., & Goodson, J. E. (1960). Motion sickness in a helicopter simulator. Aerospace Medicine, 31, 204-212.
Oman, C. M. (1982). A heuristic mathematical model for the dynamics of sensory conflict and motion sickness. Acta Otolaryagologica, 44 (Suppl. 392).
Pedhazur, E. J. (1997). Multiple regression in behavioral research (3rd ed.), Fort Worth, TX: Harcourt Brace College.
Reason, J. T. (1978). Motion sickness adaptation: A neural mismatch model. Proceedings of the Royal Society of Medicine, 71, 819-829.
Regan, E. C., & Price, K. R. (1994). The frequency of occurrence and severity of side-effects of immersion in virtual reality. Aviation. Space, and Environmental Medicine. 65, 527-530.
Riccio, G. E., & Stoffregcn, T. A. (1991). An ecological theory of motion sickness and postural instability. Ecological Psychology, 3, 195-240.
Stoffregen, T. A. (1985). Flow structure versus retinal location in the optical control of stance. Journal of Experimental Psychology: Human Perception and Performance, 11, 554-565.
Stoffregen, T. A., & Riccio, G.E. (1991). An ecological critique of the sensory conflict theory of motion sickness. Ecological Psychology, 3, 159-194.
Stoffregen, T. A., & Smart, L. J. (1998). Postural instability precedes motion sickness. Brain Research Bulletin, 47, 437-448.
Tipler, P. A. (1987), College physics. New York: Worth.
Yoo, Y., Lee, G. C. H., & Jones. S. A. (1997). Vection, compensatory sway, and simulator sickness. In B. Dos and W. Karwowski (Eds.), Advances in occupational ergonomics and safety (pp. 589-592). Amsterdam: IOS Press.
Sequence of Trials Trial Condition 1 20 s, eyes open, no imposed motion 2 20 s, eyes closed, no imposed motion 3 1 min, eyes open; display motion at 0.2 Hz, [+ or -]30[degrees] amplitude (Figure 2, top) 4 1 min, eyes closed, 0.2 Hz, [+ or -]30[degrees]amplitude 5-8 each was 10 min; eyes open, sum of 10 sines, [+ or -]45[degrees] amplitude (Figure 2, bottom) 9 1 min, eyes open; 0.2 Hz, [+ or -]30[degrees] amplitude 10 20 s, eyes open, no imposed motion 11 20 s, eyes closed, no imposed motion Mean (Median) Total Severity Scores on the Sickness Questionnaire Pre-Exposure Immediate Post 1 hr Post Sick 6.95 (7.48) 34.75 (37.40) 14.96 (11.22) Well 3.21 (0.0) 4.27 (3.74) 0.0 (0.0)
|Printer friendly Cite/link Email Feedback|
|Author:||Stoffregen, Thomas A.; Hettinger, Lawrence J.; Haas, Michael W.; Roe, Merry M.; Smart, L. James|
|Date:||Sep 22, 2000|
|Previous Article:||The Potential for Misinterpretation Considered More Globally: A Response to Vicente and Ethier.|
|Next Article:||The Framing of Drivers' Route Choices when Travel Time Information Is Provided under Varying Degrees of Cognitive Load.|