Motion Sickness from Combined Lateral and Roll Oscillation: Effect of Varying Phase Relationships

Transcription

1 RESEARCH ARTICLE Motion Sickness from Combined Lateral and Roll : Effect of Varying Phase Relationships Judith A. Joseph and Michael J. Griffin J OSEPH JA, G RIFFIN MJ. Motion sickness from combined lateral and roll oscillation: effect of varying phase relationships. Aviat Space Environ Med 2007; 78: Background: Previous studies have investigated motion sickness caused by combined lateral and roll oscillation occurring in phase with each other. In tilting trains there can be a phase difference between the two motions. Hypothesis: It was hypothesized that sickness caused by combined lateral and roll oscillation would depend on the phase between the lateral acceleration and the roll displacement. Method: At intervals of at least 1 wk, 20 subjects were seated in a cabin and exposed to four 30-min exposures of combined 0.2 Hz sinusoidal lateral acceleration ( ms 22 ) and 0.2 Hz roll displacement ( ). The roll oscillation had one of four phases relative to the lateral oscillation: 1) 0 delay (giving 100% compensation of the lateral acceleration); 2) 14.5 delay (75% compensation); 3) 29 delay (50% compensation); and 4) 29 advance (50% compensation). Subjects gave ratings of sickness at 1-min intervals. Results: Sickness was greatest with no delay (100% compensation). Increasing the delay to 14.5 (75% compensation) and to 29 (50% compensation) decreased sickness. Less sickness occurred when the roll displacement led the lateral acceleration by 29 (phase advance) than when the roll displacement followed the lateral acceleration by 29 (phase delay). Conclusions: With combined lateral and roll oscillation, sickness depends on the phase between the two motions. Increasing the delay in the roll motion reduces sickness, but also reduces the compensation. There is less sickness when the roll displacement leads the lateral acceleration than when the roll displacement lags the lateral accel eration. Keywords: motion sickness, oscillation, lateral, roll, compensation. T HERE ARE MANY factors influencing the extent of motion sickness in transport, including the frequency, magnitude, direction, and duration of the motion. Sickness can be caused by low-frequency vertical oscillation, as in some ships ( 14, 16 ), or by low-frequency fore-and-aft and lateral acceleration, as in some forms of road transport ( 12, 20, 21 ). When vehicles turn, they tend to roll. The roll can arise from a combination of the camber of a road or the cant of a track (causing the vehicle to roll in ) and the deflection of the vehicle suspension (causing the vehicle to roll out ). Tilting a vehicle in at a curve reduces the lateral acceleration felt by passengers and can reduce disturbance to their comfort and postural stability. However, fully compensating for the lateral acceleration with roll compensation seems to increase motion sickness ( 1, 2, 4 ). Earth-lateral acceleration, a, is produced by horizontal translational acceleration along the lateral (y-axis) of the body; roll through an angle, u, involves rotational movement around the fore-and-aft (x-axis) of the body. These two motions cause the body to be exposed to lateral forces that are proportional to a and g.sin u, respectively. Lateral acceleration can be combined with roll oscillation so that the forces caused by the lateral acceleration are balanced by the forces caused by the roll oscillation. A person exposed to these combined motions can be made to feel no resultant lateral force (i.e., there is 100% compensation of the forces caused by the lateral acceleration). If the roll is of an inappropriate magnitude, or occurs at the wrong time, the compensation will be other than 100% and the person will experience lateral force. Roll oscillation by itself is not usually highly provocative of motion sickness ( 4, 13, 22 ). At frequencies between and 0.40 Hz, Howarth and Griffin (13) found that 68 of roll oscillation provoked low levels of illness with no significant difference between the five frequencies investigated. The results suggest that if the angular displacement is the same at each frequency, the frequency of roll oscillation has no affect on motion sickness. Roll motion results in acceleration in the plane of the floor (due to rotation through the gravitational vector) that is proportional to the angle of roll. The results, therefore, suggest that motion sickness caused by roll oscillation is roughly independent of the frequency of the angle of roll and independent of the frequency of the translational acceleration due to gravity. It was concluded that the motion sickness produced by roll oscillation differs in its frequency dependence from that of horizontal oscillation ( 3 ) and vertical oscillation ( 15 ). Donohew and Griffin ( 3 ) found that the sickness caused by sinusoidal lateral oscillation with a peak velocity of 61.0 ms2 1 was highly dependent on the frequency of oscillation over the range Hz to 0.20 Hz, with greater sickness at the higher frequencies. With From the Human Factors Research Unit, Institute of Sound and Vibration Research, University of Southampton, Southampton, UK. This manuscript was received for review in December It was accepted for publication in July Address reprint requests to: Professor Michael J. Griffin, Human Factors Research Unit, Institute of Sound and Vibration Research, University of Southampton, Southampton SO17 1BJ, England; M.J.Griffin@soton.ac.uk. Judith Joseph is a Research Student at the University of Southampton. Reprint & Copyright by Aerospace Medical Association, Alexandria, VA. DOI: /ASEM Aviation, Space, and Environmental Medicine x Vol. 78, No. 10 x October 2007

2 a peak velocity of 60.5 ms 21, sinusoidal lateral oscillation produced similar sickness over the frequency range 0.2 to 0.8 Hz, although there was an increased risk of mild nausea around to 0.40 Hz ( 10 ). Griffin and Mills ( 11 ) found that sickness caused by Hz lateral oscillation increased when the magnitude of oscillation increased from 0.28 ms 22 rms to 1.11 ms 22 rms. Studies investigating the effect of the frequency of fore-and-aft oscillation have yielded similar findings to those with lateral oscillation ( 6 8 ). Studies in the laboratory have found that roll compensation of the accelerations caused by lateral oscillation can increase motion sickness. Förstberg et al. ( 4 ) reported that full compensation (i.e., 100% compensation) of lateral oscillation by roll oscillation caused more sickness than no compensation (0% compensation), and more sickness than 56% and 75% compensation. With 0.2 Hz lateral oscillation at an acceleration of ms 22, Donohew ( 1 ) found that 50% compensation resulted in less illness than 100% compensation and marginally less illness than uncompensated lateral oscillation (i.e., 0% compensation). With 0.1 Hz lateral oscillation at ms, 75% and 100% compensation 22 caused more motion sickness than 25% compensation ( 2 ). These studies show that the sickness caused by combined lateral and roll oscillation does not increase monotonically with increasing roll, and sickness is not least when the lateral acceleration measured in the plane of the floor is zero, so sickness cannot be predicted solely from the lateral acceleration or solely from the roll motion there is some interaction between the lateral acceleration and the roll displacement that influences sickness. In previous laboratory studies, the lateral acceleration and roll compensation have been applied with no phase lag between the motions. In practice, there can be a lag between the lateral acceleration and the compensatory roll, such as in a tilting train with a compensation system having a delay. The delay will reduce the effectiveness of the compensation so that the lateral accelerations experienced by passengers increase, but it is not known whether this affects motion sickness. For a combination of lateral acceleration and roll oscillation that produces 100% compensation with no delay, the percentage compensation will reduce as the delay increases. The objective of this study was to investigate whether, during combined lateral and roll oscillation, motion sickness depends on the phase between the lateral acceleration and the roll angle. The study investigated the effects of phase lag and phase lead on motion sickness with 0.2 Hz lateral oscillation compensated by roll oscillation. It was hypothesized that although the two individual motions were unchanged, motion sickness would depend on the phase between the motions. METHODS Apparatus Motions were produced by a simulator capable of 12 m of horizontal oscillation and 10 of roll oscillation. The motions were generated and monitored using an HVLab data acquisition computer system (version 3.81, Human Factors Research Unit, Institute of Sound and Vibration Research, University of Southampton, Southampton, UK). An inductive accelerometer ( 6 12 g, 503 AD/32; S/N: AE 2653/77; Smith Industries, UK ) was mounted at the base of the cabin beneath the roll system to measure the lateral acceleration. Another inductive accelerometer ( 6 12 g, 503 AD/32; S/N: AE 2983/77; Smith Industries) was mounted inside the cabin at the seat surface (the center of roll) to measure the lateral acceleration in the plane of the floor (i.e., the sum of the lateral acceleration and the gravitational component due to roll). The accelerometer signals were recorded so as to be able to confirm that each subject was exposed to the correct combination of lateral and roll oscillation. Exposure Conditions Subjects were exposed to the same 0.2 Hz sinusoidal lateral oscillation combined with the same 0.2 Hz roll oscillation in each of four 30-min exposures. The difference between the sessions was the phase between the lateral acceleration and the roll displacement. The 0.2 Hz lateral acceleration was 0.89 ms 22 rms (a peak acceleration of ms 22, a peak velocity of 61.0 ms 21 and a peak displacement of 60.8 m). The 0.2 Hz roll oscillation had a peak displacement of The roll oscillation was controlled to one of four phases relative to the lateral oscillation: 1) 0 delay (equivalent to 100% compensation); 2) 14.5 delay (75% compensation); 3) 29 delay (50% compensation); or 4) 29 advance (50% compensation) ( Table I ). Waveforms showing the lateral oscillation, roll oscillation, and the lateral acceleration at the seat surface for the four conditions are shown in Fig. 1. The beginning and end of each motion was tapered using cosine tapering over a period of 2.5 cycles so that there was a smooth transition when starting and stopping the motion. Subjects were seated in a closed cabin (2.2 m high m wide m deep), which reduced external cues such as air movement, light, and sound. They sat on a rigid seat 445 mm above the platform of the simulator. The high backrest to the seat was flat and rigid and extended 540 mm above the seat surface. The center of rotation was at the center of the seat surface, between the ischial tuberosities of subjects. A loose lap belt was worn for safety reasons. Subjects were instructed to adopt relaxed upright postures with their hands on their laps TABLE I. PHASE, EQUIVALENT COMPENSATION, AND SUBJECT LATERAL ACCELERATION IN THE FOUR EXPERIMENTAL CONDITIONS. Relative Phase (degrees) Compensation (%) Subject Lateral Peak Acceleration (ms 22 ) Root Mean Square Subject Lateral Acceleration (ms 22 rms) 0 delay delay delay advance Aviation, Space, and Environmental Medicine x Vol. 78, No. 10 x October

3 Environment Headphones producing white noise at db(a) measured using a Knowles Electronics Mannequin for Acoustics Research (KEMAR) were worn by all subjects to mask the noise of the simulator. The light inside the cabin was produced by a 40-W filament bulb mounted in the center of the ceiling of the cabin. The experimenter communicated with subjects via a microphone, which interrupted the white noise input to the headphones. Subjects were under continual observation while in the cabin by means of closed circuit television. A video camera mounted inside the cabin enabled the experimenter to check that subjects maintained the correct posture and kept their eyes open at all times throughout the experiment. Subjects were told to look at a fractal image ( m) during the motion exposures. The fractal was positioned in front of them, at approximately eye level, on the internal wall of the cabin at a distance of 0.7 m. Subjects There were 20 healthy male volunteers, between 18 and 26 yr of age (mean age: 23.3 yr, SD: 62.0 yr; mean stature: 179 cm, SD: 5.8 cm), who took part in the experiment. They were selected from the staff and student populations of the University of Southampton. In the repeated measures design, all subjects were exposed to all conditions, so there was a possibility of carry-over effects occurring where the response of a subject to one condition was affected by exposure to a previous condition. Carryover effects were controlled by the use of a Latin square, so five subjects were exposed to condition 0 delay first, five subjects were exposed to condition 14.5 delay first, and so on, with the orders also balanced across subjects. Each subject was tested at approximately the same time of day to control for the influence of any circadian effects, and at an interval of at least 1 wk to control for effects of habituation. Prior to motion exposure, all subjects completed a motion sickness susceptibility questionnaire, which was used to calculate six measures of motion sickness susceptibility: illness susceptibility in transport in the past year, I susc(yr) ; vomiting susceptibility in transport in the past year, V susc(yr) ; total susceptibility to vomiting, V total ; total susceptibility to motion sickness, M total ; susceptibility to motion sickness in land transport, M land ; and susceptibility to motion sickness in nonland transport, M nland ( 9 ). Subjects gave their informed consent to participate in the experiment and were free to withdraw at anytime. Ethical approval was granted by the Human Experimentation Safety and Ethics Committee of the Institute of Sound and Vibration Research at the University of Southampton. Fig. 1. Waveforms for the four experimental conditions. Lateral oscillation: dotted line; roll oscillation: solid thin line; lateral acceleration at seat surface: solid thick line. and feet separated flat on the floor with their backs in contact with the backrest. Illness Rating Scale Illness was monitored at 1-min intervals from 5 min before the start of motion until 15 min after the motion had ceased (i.e., over a period of 50 min). Subjects gave ratings of their illness using a 7-point illness scale: 0 5 no symptoms; 1 5 any symptoms, however slight; 2 5 mild symptoms; 3 5 mild nausea; 4 5 mild to moderate nausea; 5 5 moderate nausea but can continue; 6 5 moderate nausea and want to stop. A session was terminated if a subject reached a rating of 6 (the subject was removed from the cabin and no longer required to rate their illness every minute) and a rating of 6 was assumed for the remainder of the motion exposure (up to minute 35) for that session. Subjects reporting illness rating 6 did not report their illness during the recovery period (minutes 36 to 50). Symptom Checklist A symptom checklist was completed by each subject after the end of each session. Subjects indicated which of 946 Aviation, Space, and Environmental Medicine x Vol. 78, No. 10 x October 2007

4 10 common symptoms of motion sickness they had experienced during the exposure. The symptoms on the checklist were: yawning, increased salivation, stomach awareness, bodily warmth, headache, nausea, dry mouth, cold sweating, dizziness, and drowsiness. The number of symptoms reported were added together to give a total symptom score for each subject. Statistical Methods The data were not normally distributed, so nonparametric statistics were used to analyze the data. The Kruskal-Wallis test was used to investigate differences between several independent groups; the Mann-Whitney U-test was used to investigate differences between two independent groups (i.e., when investigating possible carry-over effects). Friedman analysis investigated differences between several repeated measures on the same subject and the Wilcoxon matched-pairs signed ranks test was used to test for differences between two repeated measures on the same subject (i.e., when investigating order effects and differences between motion conditions). Spearman s rho was used to test for correlations between accumulated illness and the six measures of self-rated motion sickness susceptibility. Statistical analysis was carried out using SPSS (version 14.0, SPSS, Chicago, IL ). RESULTS When using a repeated measures (i.e., within-subjects) experimental design, the response of a subject to a specific motion may depend on the motions to which the subject has been previously exposed (e.g., whether it is the first or last exposure and which types of motion have been experienced previously). Sequence effects occur when the conditions have different influences on subsequent responses, and will be greater the more the conditions vary. The component motions in the four conditions in this study were the same and every condition followed every other condition the same number of times, so it was assumed that sequence effects would be small. Order effects occur if repeated exposure to motion influences responses (e.g., if a subject habituates so as to become less sick with more experience of motion), so that illness ratings differ according to whether a motion is presented in the first, second, third, or fourth session. Fig. 2 shows the mean illness ratings reported by subjects in each of the four sessions. There was a significant difference in mean illness ratings between the four sessions ( P ; Friedman), with significantly higher illness ratings in the first session than in the other three sessions ( P 0.037; Wilcoxon). The influence of this order effect on the results was investigated by excluding the first session of each subject and performing Wilcoxon analysis on the remaining 10 subjects in each pair-wise comparison. Significant differences in illness ratings were found between 0 delay and 29 delay ( P ), and between 14.5 delay and 29 delay ( P ). These were the same trends as found when including the first session in the analysis, so it was concluded that order effects were not the main influence on the illness ratings reported in this study. Fig. 2. Mean illness ratings reported by the 20 subjects at each minute during each of the four sessions (with exposure to motion between 5 and 35 min), showing a habituation effect. During each of the 30-min exposures to motion, the average illness ratings of the 20 subjects increased ( Fig. 3 ). The mean illness ratings (calculated over the 30-min exposures to motion) differed between the four conditions ( P ; Friedman). Among pairs of conditions, the mean illness ratings differed significantly between 0 delay and 29 advance ( P ), between 14.5 delay and 29 advance ( P ), and between 29 delay and 29 advance ( P ). The number of subjects reaching each illness rating in each of the four conditions is shown in Table II. The percentages of subjects reaching an illness rating of 6 (moderate nausea and want to stop) in each of the four conditions were: 0 delay - 45%; 14.5 delay - 35%; 29 delay - 25%; and 29 advance - 10%. The experiment was terminated before the beginning of the recovery period in 23 out of the 80 sessions when subjects reported an illness rating of 6 (moderate nausea and want to stop). The number of subjects who terminated the experiment before the end of the motion was 9 with 0 delay, 7 with 14.5 delay, 5 with 29 delay, and 2 with 29 advance. Subjects who reported illness rating 6 were not required to report further illness ratings, so the analysis of the mean illness ratings during the recovery period excludes those subjects who terminated the session early by reporting illness rating 6. During the recovery period after the cessation of motion, illness Fig. 3. Mean illness ratings reported by the 20 subjects at each minute during the four conditions (with exposure to motion between 5 and 35 min). Aviation, Space, and Environmental Medicine x Vol. 78, No. 10 x October

5 TABLE II. NUMBER OF SUBJECTS TO REACH EACH ILLNESS RATING IN EACH CONDITION. Condition Illness Rating Delay 29 Delay 29 Advance TABLE III. MEAN ILLNESS RATINGS REPORTED BY SUBJECTS IN EACH CONDITION AFTER MOTION EXPOSURE. Condition Mean Illness Ratings at End of Motion (minute 35) Mean Illness Ratings During Recovery Period Mean Illness Ratings at End of Recovery Period (minute 50) 0 delay delay delay advance ratings decreased. Fig. 4 shows the mean illness ratings reported by subjects from minute 35 (end of motion exposure) to minute 50 (end of recovery period) over all four motion conditions. The mean illness ratings reported by sub jects during the recovery period are shown in Table III. At the end of the motion (i.e., at 35 min), significant differences were found between the illness ratings in the four conditions ( P , Friedman). At this time, Wilcoxon analysis showed significant differences in illness ratings between 0 delay and 29 advance ( P ), between 14.5 delay and 29 advance ( P ), and between 29 delay and 29 advance ( P ). The percentage of subjects who reported an illness rating of 0 at the end of the recovery period in each condition was 82% with 0 delay, 85% with 14.5 delay, 93% with 29 delay, and 94% with 29 advance. The average of the total symptom scores for each subject in each condition were: 4.9 with 0 delay, 4.8 with 14.5 delay, 4.5 with 29 delay, and 3.1 with 29 advance. There were significant differences between the total symptom scores reported by subjects over the four conditions ( P , Friedman). Between pairs of conditions, the total symptoms scores were significantly less with 29 advance than with the other three conditions ( P, 0.010). No significant correlations were found between the accumulated illness ratings in each condition (from 5 min to 35 min) and each of the six measures of motion sickness susceptibility (I susc(yr), V susc(yr), V total, M total, M land, and M nland ) obtained from the motion sickness susceptibility questionnaire. DISCUSSION The illness ratings during and following motion, and the symptom scores, show that motion sickness was influenced by the phase between the lateral acceleration and the roll displacement. The same lateral oscillation and the same roll oscillation were employed in all four conditions, but there was greater motion sickness with no delay than with either a roll delay (either 14.5 or 29 ) or a roll advance (29 ). Motion sickness was least with 29 phase advance, greater with 29 phase delay, somewhat more with 14.5 delay, and most with no delay. The trend has similarities to the effect of varying the percentage compensation by altering the magnitude of the roll motion, where greater sickness has been found with 100% compensation than with 50% compensation ( 1 ). Motion sickness does not occur in humans with nonfunctioning vestibular systems ( 17 ). Table IV s u m m a - rizes the vestibular stimulation arising with four idealized combinations of lateral and roll motion of the head. Without motion of the head, there is no variation in the stimulation of either the semicircular canals or the otoliths and no difference between the interpretations of their sensory information. With pure roll motion of the head around the vestibular system, there is stimulation of the semicircular canals (due to the rotation) and stimulation of the otoliths (due to changes in the gravitational component, e.g., g.sin u ). Body movements may also produce some translation of the otoliths due to the axis of rotation not being through the otoliths, but that will be neglected here for simplicity [the 50 th percentile sitting eye height for British men 19 TABLE IV. COMBINATIONS OF ROLL AND LATERAL OSCILLATION CAUSING SICKNESS. Fig. 4. Mean illness ratings reported by subjects during recovery from each illness rating over all four motion conditions. Otolith stimulation Semicircular canal stimulation Motion sickness A. No B. Roll C. Lateral D. Roll & Lateral (i.e., 100% compensated lateral acceleration) None Yes Yes No None Yes No Yes Low Low High High 948 Aviation, Space, and Environmental Medicine x Vol. 78, No. 10 x October 2007

6 to 45 yr of age is about 795 mm ( 18 )]. If the subjects moved rigidly with the cabin, the peak lateral acceleration of the head due to the head not being at the center of rotation was ms 22. Pure lateral acceleration will result in stimulation of the otoliths, but no stimulation of the semicircular canals. Lateral acceleration with 100% roll compensation will result in stimulation of the semicircular canals with no stimulation of the otoliths. If the otoliths and the semicircular canals are both assumed to indicate roll motion, the first two conditions (a stationary head and roll motion of the head) have compatible patterns of stimulation of the semicircular canals and otoliths, and produce little or no sickness. The second two conditions (lateral oscillation and lateral oscillation with 100% compensation) produce inconsistent stimulation of the semicircular canals and the otoliths, and are associated with motion sickness. From the above analysis, different effects would be expected when, starting with a given lateral acceleration and 100% roll compensation, the magnitude of one or the other motion is reduced. If the lateral acceleration is reduced, so that the roll motion over-compensates for the lateral acceleration, motion sickness would be expected to reduce as the lateral acceleration reduces to near zero and the motion becomes solely a roll oscillation. Roll oscillation causes little sickness as it gives a naturally occurring combination of semicircular canal and otolithic stimulation, which are consistent if both are assumed to reflect roll. If the roll motion is reduced, so that it under-compensates for the lateral acceleration, motion sickness may change, but it will not reduce to zero because pure lateral oscillation causes motion sickness it gives a combination of semicircular and otolithic stimulation that does not occur naturally and results in inconsistent indications from the otoliths and the semicircular canals. With changing magnitudes of roll motion, the extremes (conditions C and D in Table IV ) are unnatural, whereas an intermediary condition with, say 50% compensation, is more normal in that there is otolithic stimulation in the expected direction and at the expected time, but the magnitude of the stimulation is less than expected. It seems that with some of these intermediary compensations, motion sickness can be less than with either 100% or 0% compensation ( 1, 2 ). The center of rotation of the cabin was at the center of the seat surface. In consequence, the percentage compensation was somewhat greater at higher positions (such as the organs of balance) and less at lower positions (such as the feet). Although the difference may be small compared to some other factors that may influence motion sickness, such as active control of body movement, the location of the center of rotation may be of importance in some systems. In the present experiment, the difference between 29 delay and 29 advance is interesting. Both conditions had the same motions and the same percentage compensation, yet there was significantly less sickness with the phase advance. The sensory phenomena caused by these two motions will differ if they are mediated by two sensory mechanisms having different delays (i.e., different phase responses). If sickness arises from inconsistency in the sensory information on translation and rotation after transduction in sensory channels (e.g., otoliths and semicircular canals) with different phase lags, it is reasonable for there to be different sickness despite the same difference in phase between the motions in the two conditions. The most widely known theory of motion sickness, the sensory conflict theory, states that motion sickness arises because of a conflict between the information received from two or more sensory systems: the visual system, vestibular system, and somatosensory system ( 19 ). It is assumed that the greater the conflict between the sensory systems, the greater the motion sickness. The sensory rearrangement theory suggests that motion sickness may arise from a discrepancy or incompatibility between what is expected from previous experiences and what is being experienced. If the phase difference between the different channels of sensory information (assumed here to be the semicircular canals and otoliths) were fixed, it seems likely that some combination of evolution and habituation would compensate for the difference. However, such phase differences depend on the frequency of excitation, so require different phase habituation for different motions. Possibly the subjects in the present experiment were expecting to receive sensory information from the otoliths later than it comes with 0.2 Hz lateral oscillation fully compensated by roll. Hence, they were less disturbed when the roll motion was advanced by a phase lead than when it lagged by the same amount. If this is correct, the effects of phase reported here will be different at other frequencies of oscillation. With the same frequency and magnitude of oscillation employed in the present experiment, Donohew ( 1 ) found a very similar degree of sickness when the lateral acceleration was fully compensated by roll displacement (i.e., 100% compensation), as in the condition with 0 delay in the present study ( Fig. 5 ). Both studies also employed conditions with 50% compensation, but this was achieved by different means. In the present study, both 29 delay and 29 advance correspond to 50% compensation. Donohew achieved 50% compensation by reducing the magnitude of the roll motion; this caused slightly less sickness than with 29 advance and Fig. 5. Comparison of mean illness ratings with 0 delay, 29 delay, and 29 advance (current experiment), and 100% and 50% compensation at 0.2 Hz, 0.89 ms 2 2 rms (from Ref. 1). Aviation, Space, and Environmental Medicine x Vol. 78, No. 10 x October

7 considerably less sickness than with 29 delay as found in the present study ( Fig. 5 ). With 50% compensation, there is the possibility of a near-normal pattern of sensory stimulation that arises with roll oscillation (condition B in Table IV ). It seems that a near normal pattern of sensory information is less likely when the same percentage compensation arises as a result of a phase lag. The visual conditions were unchanged between the four motion conditions investigated. However, it is possible that visual stimulation played a part in the causation of motion sickness and that the role of visual stimulation may have differed between the four conditions. The differing phases between the lateral and roll oscillations may have produced differing movement of the retinal image, but the role of visual stimulation in the causation of motion sickness in conditions with combined lateral and roll oscillation is not yet understood. There are other factors that may influence motion sickness in the conditions studied. For example, subjects may have responded differently to the different combinations of motions they experienced, leading to variations in head movements in each of the four conditions. Furthermore, different subjects may have made different responses, with some moving with the motion and some moving to compensate for the motion. With combined fore-and-aft and pitch oscillation there is a suggestion that pitch motion of the head may have differing nauseogenic potential depending whether the alignment arises from voluntary head movement or as a result of cabin tilting ( 5 ). Increasing the phase lag or increasing the phase lead reduces the effective compensation and reduces the extent to which the roll motion preserves aspects of passenger comfort other than motion sickness. In this experiment, increased phase lag reduced the compensation and reduced the sickness, but it will have increased discomfort due to increased forces acting on the subjects. For practical application to roll compensatory systems in transport, it is desirable to maximize the compensation (so as to minimize discomfort and postural instability) with the minimum increase in motion sickness. With the motions used here, for the same compensation (i.e., conditions assumed to cause similar comfort) a phase advance resulted in less sickness than an equivalent phase lag. With some motions, a phase advance might result in less sickness than a motion having the same compensation (and same comfort) but with no lag or lead. The effect may depend on the frequency, and therefore the waveform, of the motion that is being compensated. Conclusions When lateral oscillation and roll oscillation are combined, the consequent motion sickness depends on the phase between the two motions. With oscillation at 0.2 Hz, motion sickness was greatest when there was no phase delay between the two motions, so that the roll displacement was fully compensated by the lateral acceleration (i.e., 0 delay giving 100% compensation). Increasing the delay between the lateral acceleration and the roll displacement to 14.5 (75% compensation) and to 29 (50% compensation) reduced motion sickness. Less sickness occurred when the roll displacement led the lateral acceleration by 29 (phase advance) than when the roll displacement followed the lateral acceleration by 29 (phase delay). REFERENCES 1. Donohew BE. Effect on motion sickness of the percentage of rollcompensation during lateral oscillation. Proceedings of the 38 th United Kingdom Conference on Human Response to Vibration; September Alverstoke, Gosport, UK : The Institute of Naval Medicine ; 2003 : Donohew BE. Effect on motion sickness of the percentage of roll-compensation of 0.1 Hz lateral oscillation. Proceedings of the 39 th United Kingdom Conference on Human Response to Vibration; September Ludlow, Shropshire, UK : RMS Vibration Test Laboratory ; 2004: Donohew BE, Griffin MJ. Motion sickness: effect of the frequency of lateral oscillation. Aviat Space Environ Med 2004 ; 75 : Förstberg J. Effects from lateral and/or roll motion on nausea on test subjects: studies in a moving vehicle simulator. Proceedings of the 34th UK Group Meeting on Human Response to Vibration; September Dunton, Essex, UK : Ford Motor Company ; 1999 : Golding JF, Bles W, Bos JE, Haynes T, Gresty MA. Motion sickness and tilts of the inertial force environment: active suspension systems vs. active passengers. Aviat Space Environ Med 2003 ; 74 : Golding JF, Finch MI, Stott JRR. Frequency effect of Hz horizontal translational oscillation on motion sickness and the somatogravic illusion. Aviat Space Environ Med 1997 ; 68 : Golding JF, Marky HM. Effect of frequency of horizontal linear oscillation on motion sickness and somatographic illusion. Aviat Space Environ Med 1996 ; 67 : Golding JF, Mueller AG, Gresty MA. A motion sickness maximum around the 0.2 Hz frequency range of horizontal translational oscillation. Aviat Space Environ Med 2001 ; 72 : Griffin MJ, Howarth HVC. Motion sickness history questionnaire. Southampton, England: University of Southampton, Institute of Sound and Vibration Research ; 2000 : Institute of Sound and Vibration Technical Report No Griffin MJ, Mills KL. Effect of frequency and direction of horizontal oscillation on motion sickness. Aviat Space Environ Med 2002 ; 73 : Griffin MJ, Mills KL. Effect of magnitude and direction of horizontal oscillation on motion sickness. Aviat Space Environ Med 2002 ; 73 : Griffin MJ, Newman MM. Visual field effects on motion sickness in cars. Aviat Space Environ Med 2004 ; 75 : Howarth HVC, Griffin MJ. Effect of roll oscillation frequency on motion sickness. Aviat Space Environ Med 2003 ; 74 : Lawther A, Griffin MJ. The motion of a ship at sea and the consequent motion sickness amongst passengers. Ergonomics 1986 ; 29 : Lawther A, Griffin MJ. Prediction of the incidence of motion sickness from the magnitude, frequency and duration of vertical oscillation. J Acoust Soc Am 1987 ; 82 : Lawther A, Griffin MJ. Motion sickness and motion characteristics of vessels at sea. Ergonomics 1988 ; 31 : Money KE. Motion sickness. Physiol Rev 1970 ; 50 : Pheasant S. Bodyspace: anthropometry, ergonomics and the design of work, second ed. London : Taylor and Francis ; Reason JT, Brand JJ. Motion sickness. London : Academic Press ; Turner M, Griffin MJ. Motion sickness in public road transport: the relative importance of motion, vision and individual differences. Br J Psychol 1999 ; 90 : Turner M, Griffin MJ. Motion sickness in public road transport: the effect of driver, route and vehicle. Ergonomics 1999 ; 42 : Wertheim AH, Bos JE, Bles W. Contributions of roll and pitch to sea sickness. Brain Res Bull 1998 ; 47 : Aviation, Space, and Environmental Medicine x Vol. 78, No. 10 x October 2007