Relationship between motion sickness susceptibility and vestibulo-ocular reflex gain and phase

Abstract

BACKGROUND:

It has been proposed that individual susceptibility to motion sickness is related to the vestibulo-ocular reflex (VOR) through the activation of the velocity storage mechanism.

OBJECTIVE:

We investigated whether motion sickness level was related to the gain and phase of the VOR.

METHODS:

VOR gain and phase were measured in 214 subjects while they rotated in yaw at 0.01 Hz, 0.02 Hz, 0.04 Hz, 0.08 Hz, and 0.16 Hz in darkness, and results were compared to the severity of symptoms the subjects experienced during subsequent tests to provoke motion sickness. These tests included cross-coupled angular accelerations, sudden stops in light or in dark, off-vertical axis rotation, and parabolic flight. The subjects were grouped according to the motion sickness level reached during these tests (none, low, medium, or high).

RESULTS:

No correlation was found between the horizontal VOR gain and motion sickness level. However, for the subjects with high motion sickness level, the VOR phase lead was significantly lower during rotation at frequencies ranging from 0.04 Hz to 0.16 Hz (i.e. the VOR time constant was longer) than the other motion sickness groups.

CONCLUSION:

These results support the theory that the longer the time constant for velocity storage, the more severe the motion sickness.

Keywords

Motion sickness sensory conflict vestibular nystagmus velocity storage

1 Introduction

Pitching or rolling the head while the body is rotating at a constant velocity can provoke motion sickness symptoms. Such head movements stimulate the semicircular canals and the otoliths and change the position of the otoliths relative to gravity, which produces vertical and horizontal nystagmus, disorientation, vertigo, and nausea. With repeated exposure, however, subjects habituate and can make more head movements before they start experiencing motion sickness [6, 17]. In a recent review, Lackner [18] identified several studies that suggest individual susceptibility to motion sickness is related to the spatial-temporal properties of the vestibulo-ocular reflex (VOR) through the activation of the velocity storage mechanism.

The semicircular canals are stimulated by angular acceleration of the head, but the deflection of the cupula is approximately proportional to head velocity when rotating at frequencies ranging from 0.1 Hz to 1 Hz [12]. At lower rotation frequency the velocity storage mechanism maintains 3D spatial orientation by prolonging the nystagmus response and orienting the axis of eye velocity toward gravity [25]. It is believed that this velocity storagemechanism increases the sensory conflict between actual and expected motion by increasing low-frequency vestibular inputs, and consequently triggers motion sickness [7]. Adaptation procedures have shown that increased resistance to motion sickness reduces the time constant of the VOR, and this has provided evidence for a correlation between the properties of the velocity storage mechanism and motion sickness [8 , 28]. The results of these adaptation studies also demonstrated that the severity of the motion sickness was associated with the 3D properties of velocity storage: the larger the deviation of the eye velocity axis from gravity, the more severe the motion sickness.

In an effort to reduce the occurrence of space motion sickness in astronauts during short-duration missions on board the Space Shuttle, NASA conducted a series of ground-based studies in the early 1980s to evaluate tests that could predict susceptibility to motion sickness [26]. These studies concluded that space motion sickness could not be reliably predicted using ground-based provocative tests [10, 19]. However, in light of the recent studies mentioned above regarding the potential role of the velocity storage mechanism, we reexamined the data collected during NASA’s ground-based studies to verify whether there was a correlation between an individual’s susceptibility to motion sickness and the gain and phase of his/her VOR.

A subset of the data reported in this manuscript has been published previously in a book chapter [26]. This previous study used data collected on 164 subjects to test various logistic models for predicting motion sickness during parabolic flight. The investigators chose to predict motion sickness in parabolic flight because for a few seconds it induces weightlessness similar to space flight. This earlier report did not include the VOR characteristics of subgroups of subjects with various levels of motion sickness, which is the main objective of the present manuscript.

2 Material and methods

2.1 Subjects

A total of 214 healthy human subjects (165 male, 49 female) participated in these studies, which were conducted at the NASA Johnson Space Center from 1982 to 1984. The subjects’ ages ranged from 19 to 60 years. The age distribution was as follows: 112 subjects were 19 to 29 years (SD = 8.5); 74 subjects were 30 to 39 years (SD = 9.2); 19 subjects were 40 to 49 years (SD = 9.9); and 8 subjects were 50 to 59 years (SD = 9.6). All subjects passed a US Air Force Class III medical examination and completed a course in physiological training for high altitude survival, which were required before participation in parabolic flight. The study protocol was approved in advance by NASA’s Institutional Review Board. Informed consent was obtained from each subject before they participated in the study.

2.2 Vestibulo-ocular reflex test

Subjects sat in a rotating chair with the head ventro-flexed 30° and held in place with a bite-board that was attached to the chair. The chair was rotated at 5 separate frequencies—0.01 Hz, 0.02 Hz, 0.04 Hz, 0.08 Hz, and 0.16 Hz—for 12 cycles each. Peak velocity was 60°/s at all frequencies, with acceleration ranging from 3.8°/s² at 0.01 Hz to 60.8°/s² at 0.16 Hz. Subjects kept their eyes open during the test and wore light-occluding goggles and white noise-emitting headphones, which masked orientation auditory cues. Subjects were instructed to gaze straight ahead while doing a mental exercise task to enhance alertness. Before and after completing the rotation test, the subjects donned red-filtered goggles for 20 min to adapt to dark. The subjects’ eye movements were then calibrated with alternating flashing LEDs located 1.5 m from the forehead and subtending a visual angle of 10°. Horizontal eye movements were recorded with non-polarizing ECG electrodes positioned at the outer canthus of each eye and a reference electrode located on the forehead.

2.3 Provocative tests

A series of 6 provocative tests for motion sickness were used to assess motion sickness level of each individual. During each of these tests, motion sickness was graded using the diagnostic criteria developed by Graybiel et al. [15]. The provocative tests were performed after the VOR evaluation and the order of the tests was randomized across subjects. To limit vestibular habituation effects, the test sessions were separated by several weeks.

2.3.1 Coriolis Sickness Susceptibility Index (CSSI)

Subjects sat in a chair that rotated in the dark at 180°/s about the vertical axis. While rotating, the subjects performed standardized head movements in pitch and roll. Head movements were performed in sets, with each set consisting of 5 movements (pitch forward, pitch backward, roll right, roll left, pitch forward). Each set was separated by a 20-s period with no head motions, during which subjects were questioned about signs and symptoms indicative of motion sickness. The test was terminated after 150 head movements (30 sets) or when the subjects reached the Malaise III level (8 symptom points) on Graybiel’s scale [15]. The motion sickness level was then expressed on a scale of 0 to 100, with respect to the rotation velocity and the number of head movements performed. A score of 100 denotes extreme resistance to the CSSI test [20] (Table 1).

Table 1
Criteria used to categorize the motion sickness levels of the subjects during the provocative tests. Points were used to grade the symptoms according to Graybiel’s scale [15]. CSSI: Coriolis Sickness Susceptibility Index; SVMT: Staircase Velocity Motion Test; SST-D: Sudden Stop Test in the Dark; SST-L; Sudden Stop Test in the Light; OVAR: Off-Axis Vertical Rotation; PF: Parabolic Flight

Test Level Criteria

CSSI None >90

Low 40–90

Medium 9–39

High 0–8

SVMT None >700 head movements

Low 400–700 head movements

Medium 200–399 head movements

High <200 head movements

SST-D None 40 trials (0 point)

Low 23–39 trials (>0 point)

Medium 13–22 trials

High 0–12 trials

SST-L None 40 trials (0 point)

Low 23–39 trials (>0 point)

Medium 13–22 trials

High 0–12 trials

OVAR None 35 min (<5 points)

Low 30–34 min (>5 points)

Medium 16–29 min

High 0–15 min

PF None 41 parabolas (<8 points)

Low 24–40 parabolas (>8 points)

Medium 15–23 parabolas

High 0–14 parabolas

Test	Level	Criteria
CSSI	None	>90
	Low	40–90
	Medium	9–39
	High	0–8
SVMT	None	>700 head movements
	Low	400–700 head movements
	Medium	200–399 head movements
	High	<200 head movements
SST-D	None	40 trials (0 point)
	Low	23–39 trials (>0 point)
	Medium	13–22 trials
	High	0–12 trials
SST-L	None	40 trials (0 point)
	Low	23–39 trials (>0 point)
	Medium	13–22 trials
	High	0–12 trials
OVAR	None	35 min (<5 points)
	Low	30–34 min (>5 points)
	Medium	16–29 min
	High	0–15 min
PF	None	41 parabolas (<8 points)
	Low	24–40 parabolas (>8 points)
	Medium	15–23 parabolas
	High	0–14 parabolas

2.3.2 Staircase Velocity Motion Test (SVMT)

This test is a modified CSSI test procedure in which the cross-coupled angular acceleration progressed from low-level to maximum-stress stimulation. Subjects were rotated in the dark at 6°/s around a vertical axis while they pitched and rolled their head 40 times in the sequence described above for the CSSI test. After 20 s of no head motion, the chair velocity increased in 12°/s steps with an acceleration of 6°/s². The subjects executed 40 head movements at each velocity step until the velocity reached 210°/s (720 total head movements) or until the subjects reached the Malaise III level. Motion sickness level was determined from the number of head movements performed during the test (Table 1).

2.3.3 Sudden Stop Test in the Dark (SST-D)

The chair was accelerated at 20°/s² up to a velocity of 300°/s that was maintained for 30 s. At the end of 30 s, the chair was decelerated at 150°/s² to a complete stop, and zero velocity was maintained for 30 s. This profile was repeated 40 times. The chair was rotated in one direction for 20 trials, and in the opposite direction for the subsequent 20 trials. The subjects wore light-occluding goggles. After each trial, the subjects were questioned about symptoms indicative of motion sickness. The test was terminated after 40 trials or when the subjects reached the Malaise III level. Motion sickness level was determined from the number of trials performed (Table 1).

2.3.4 Sudden Stop Test in the Light (SST-L)

The conditions of this test were the same as the one above, except instead of wearing light-occluding goggles, the subjects viewed a dark blue cloth drum with vertical stripes (1.74° wide, placed at 64° intervals) that encircled the chair (Table 1).

2.3.5 Off-Vertical Axis Rotation (OVAR)

Subjects sat in a chair that rotated in the dark at 120°/s about the vertical axis for 5 min. The axis of rotation was then tilted in 5° increments at 5-min intervals. During the test, subjects were questioned about symptoms indicative of motion sickness. The test was terminated after a 5-min rotation at 30° tilt (off-vertical) or when the subjects reached the Malaise III level. Motion sickness level was determined from the duration of the test (Table 1).

2.3.6 Parabolic Flight (PF)

Subjects were secured in a seat aboard the NASA KC-135 aircraft, which was flown through a series of 40 parabolic maneuvers with each parabola comprising 24 s of weightlessness and 30–40 s of a 1.8-g pull-up. The subjects’ heads were immobilized using a soft neck brace, and light-occluding goggles were used to eliminate visual cues. The subjects were questioned about symptoms indicative of motion sickness after each parabola. Due to the nature of this test, subjects were not removed from the provocative stimulus until the flight was over. Motion sickness level was determined by the number of parabolas flown before the Malaise III endpoint was reached (Table 1).

2.4 Eye movement analysis

Eye position data was differentiated to calculate eye velocity. Fast phases of the nystagmus were removed manually using an interactive graphical program to derive the slow-phase eye velocity from which the response parameters were estimated. The data for the initial rotation cycle at each frequency were discarded to allow for response stabilization. The eye movement responses were then measured using a sinusoidal curve fit of the remaining slow-phase eye velocity data. Peak slow-phase eye velocity and phase relative to chair velocity were calculated from the sinusoidal curve fit using the method described by Peterka et al. [23].

VOR gain was defined as the ratio of peak slow-phase eye velocity and peak chair velocity. VOR phase was defined as the difference in phase between slow-phase velocity and chair velocity plus 180°. A perfect compensatory VOR response would actually have a gain equal to unity and be 180° out of phase with the stimulus, so the addition of 180° conveniently expresses a response with no phase erroras 0°.

2.5 Motion sickness

For this study we have analyzed retrospective data that were collected more than 30 years ago. The data at our disposal included the VOR phase and gain for each of the 5 rotation frequencies (0.01 Hz, 0.02 Hz, 0.04 Hz, 0.08 Hz, 0.16 Hz) that were averaged according to the level of motion sickness experienced during each of the 6 subsequent provocative tests. The mean and SD of VOR phase and gain for the subgroups of subjects who experienced no, low, medium, and high levels of motion sickness during each of the 6 provocative tests are given in Table 2. In a previous report of these data, Reschke [26] found that the motion sickness levels collected for the CSSI, SST-L, SST-D, OVAR, and PF tests were highly correlated for each other. Unfortunately, the retrospective data did not include the VOR measurements and the corresponding motion sickness levels for individual subjects. Some subjects might have experienced a greater level of motion sickness during one provocative test and no motion sickness during another test [26]. In addition, not all subjects were tested with all 6 provocative tests.

Table 2
Mean and standard deviation of VOR phase and gain at each of the 5 frequencies used during the baseline VOR test. Subjects are grouped according to their motion sickness level (none, low, medium, high) during the 6 subsequent provocative tests. The number of subjects per motion sickness level is shown in parenthesis

CSSI 0.01 Hz 0.02 Hz 0.04 Hz 0.08 Hz 0.16 Hz

Phase Gain Phase Gain Phase Gain Phase Gain Phase Gain

None (N = 4)

Mean 55.40 0.24 29.60 0.25 15.87 0.42 13.17 0.34 7.57 0.37

SD 3.64 0.15 4.10 0.18 4.39 0.26 5.27 0.20 2.57 0.26

Low (N = 12)

Mean 50.99 0.28 29.26 0.36 17.74 0.40 6.70 0.41 2.06 0.47

SD 9.92 0.07 8.20 0.08 8.41 0.11 10.40 0.11 13.30 0.15

Medium (N = 94)

Mean 47.25 0.27 27.72 0.34 15.06 0.38 5.80 0.40 2.93 0.40

SD 8.59 0.12 7.91 0.14 6.70 0.16 6.91 0.16 7.45 0.17

High (N = 97)

Mean 46.26 0.28 26.70 0.35 13.81 0.41 6.58 0.42 1.22 0.52

SD 8.45 0.11 7.13 0.12 6.22 0.15 4.44 0.15 6.65 0.68

SVMT

0.01 Hz 0.02 Hz 0.04 Hz 0.08 Hz 0.16 Hz

Phase Gain Phase Gain Phase Gain Phase Gain Phase Gain

None (N = 9)

Mean 48.91 0.30 26.16 0.31 16.85 0.46 9.28 0.39 4.69 0.42

SD 7.01 0.11 5.80 0.14 4.24 0.18 4.68 0.15 4.48 0.19

Low (N = 22)

Mean 46.44 0.28 27.54 0.37 15.95 0.41 5.98 0.43 2.20 0.44

SD 11.25 0.10 8.09 0.12 7.25 0.14 7.63 0.14 5.07 0.14

Medium (N = 104)

Mean 48.61 0.27 28.30 0.33 15.82 0.38 6.34 0.41 1.86 0.40

SD 7.80 0.12 7.37 0.13 5.69 0.17 6.63 0.16 8.80 0.17

High (N = 48)

Mean 43.30 0.30 25.13 0.38 11.71 0.42 4.80 0.42 0.72 0.60

SD 7.75 0.11 7.15 0.11 7.12 0.12 4.40 0.13 7.35 0.95

SST-D

0.01 Hz 0.02 Hz 0.04 Hz 0.08 Hz 0.16 Hz

Phase Gain Phase Gain Phase Gain Phase Gain Phase Gain

None (N = 48)

Mean 46.43 0.29 27.32 0.35 16.03 0.41 7.12 0.41 2.28 0.43

SD 8.80 0.12 5.54 0.13 5.82 0.17 5.55 0.16 5.63 0.16

Low (N = 49)

Mean 46.88 0.28 26.64 0.34 14.99 0.39 5.62 0.43 1.28 0.51

SD 8.32 0.10 8.76 0.12 6.72 0.14 6.66 0.14 9.62 0.78

Medium (N = 73)

Mean 47.88 0.27 28.24 0.34 13.93 0.38 5.80 0.40 1.46 0.45

SD 9.10 0.11 8.45 0.13 7.02 0.16 6.76 0.16 7.85 0.45

High (N = 44)

Mean 46.67 0.28 27.59 0.35 13.82 0.38 5.27 0.40 1.99 0.48

SD 9.54 0.12 8.77 0.13 6.34 0.14 7.69 0.15 6.84 0.56

SST-L

0.01 Hz 0.02 Hz 0.04 Hz 0.08 Hz 0.16 Hz

Phase Gain Phase Gain Phase Gain Phase Gain Phase Gain

None (N = 3)

Mean 46.20 0.33 24.77 0.44 16.10 0.48 7.57 0.49 5.30 0.53

SD 4.30 0.07 2.64 0.10 3.54 0.09 3.60 0.09 3.28 0.02

Low (N = 21)

Mean 48.04 0.27 27.18 0.33 15.34 0.38 6.03 0.36 1.55 0.40

SD 7.94 0.12 5.04 0.14 5.01 0.19 5.59 0.19 6.04 0.20

Medium (N = 20)

Mean 50.55 0.28 30.01 0.34 17.68 0.44 7.75 0.44 2.49 0.44

SD 8.15 0.12 4.94 0.14 6.52 0.18 4.65 0.17 8.73 0.19

High (N = 150)

Mean 46.59 0.28 27.16 0.34 14.40 0.39 6.03 0.41 1.73 0.47

SD 8.88 0.11 8.18 0.12 6.62 0.15 6.70 0.14 8.01 0.55

OVAR

0.01 Hz 0.02 Hz 0.04 Hz 0.08 Hz 0.16 Hz

Phase Gain Phase Gain Phase Gain Phase Gain Phase Gain

None (N = 67)

Mean 47.35 0.28 26.91 0.34 15.27 0.39 5.91 0.41 2.21 0.41

SD 7.94 0.12 7.12 0.13 6.45 0.17 6.72 0.16 7.76 0.15

Low (N = 37)

Mean 48.59 0.25 27.61 0.30 15.30 0.35 5.64 0.38 1.49 0.38

SD 8.93 0.11 7.55 0.11 6.55 0.15 7.39 0.16 6.84 0.16

Medium (N = 69)

Mean 46.84 0.28 27.46 0.36 15.27 0.41 6.45 0.42 1.88 0.51

SD 9.41 0.10 7.79 0.13 5.10 0.15 5.78 0.14 7.45 0.67

High (N = 26)

Mean 46.66 0.28 28.93 0.36 12.30 0.39 6.94 0.39 0.76 0.55

SD 8.52 0.12 8.15 0.12 8.76 0.11 4.17 0.12 8.86 0.71

PF

0.01 Hz 0.02 Hz 0.04 Hz 0.08 Hz 0.16 Hz

Phase Gain Phase Gain Phase Gain Phase Gain Phase Gain

None (N = 75)

Mean 47.25 0.28 27.93 0.345 16.06 0.40 6.56 0.41 2.98 0.48

SD 9.38 0.12 7.41 0.14 6.32 0.15 6.43 0.16 8.73 0.65

Low (N = 24)

Mean 45.80 0.29 25.59 0.35 14.46 0.43 5.68 0.43 0.11 0.43

SD 5.86 0.10 7.07 0.14 3.87 0.16 6.35 0.16 6.38 0.18

Medium (N = 8)

Mean 45.21 0.29 25.95 0.36 13.73 0.47 3.65 0.46 1.69 0.45

SD 7.52 0.12 7.33 0.12 7.13 0.19 2.83 0.12 3.70 0.14

High (N = 40)

Mean 47.81 0.29 28.71 0.38 14.06 0.42 6.49 0.44 0.91 0.44

SD 6.50 0.10 7.23 0.11 7.81 0.14 5.71 0.14 7.42 0.15

CSSI		0.01 Hz	0.02 Hz	0.04 Hz	0.08 Hz	0.16 Hz
None (N = 4)
	Mean	55.40	0.24	29.60	0.25	15.87	0.42	13.17	0.34	7.57	0.37
	SD	3.64	0.15	4.10	0.18	4.39	0.26	5.27	0.20	2.57	0.26
Low (N = 12)
	Mean	50.99	0.28	29.26	0.36	17.74	0.40	6.70	0.41	2.06	0.47
	SD	9.92	0.07	8.20	0.08	8.41	0.11	10.40	0.11	13.30	0.15
Medium (N = 94)
	Mean	47.25	0.27	27.72	0.34	15.06	0.38	5.80	0.40	2.93	0.40
	SD	8.59	0.12	7.91	0.14	6.70	0.16	6.91	0.16	7.45	0.17
High (N = 97)
	Mean	46.26	0.28	26.70	0.35	13.81	0.41	6.58	0.42	1.22	0.52
	SD	8.45	0.11	7.13	0.12	6.22	0.15	4.44	0.15	6.65	0.68
SVMT
		0.01 Hz	0.02 Hz	0.04 Hz	0.08 Hz	0.16 Hz
		Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain
None (N = 9)
	Mean	48.91	0.30	26.16	0.31	16.85	0.46	9.28	0.39	4.69	0.42
	SD	7.01	0.11	5.80	0.14	4.24	0.18	4.68	0.15	4.48	0.19
Low (N = 22)
	Mean	46.44	0.28	27.54	0.37	15.95	0.41	5.98	0.43	2.20	0.44
	SD	11.25	0.10	8.09	0.12	7.25	0.14	7.63	0.14	5.07	0.14
Medium (N = 104)
	Mean	48.61	0.27	28.30	0.33	15.82	0.38	6.34	0.41	1.86	0.40
	SD	7.80	0.12	7.37	0.13	5.69	0.17	6.63	0.16	8.80	0.17
High (N = 48)
	Mean	43.30	0.30	25.13	0.38	11.71	0.42	4.80	0.42	0.72	0.60
	SD	7.75	0.11	7.15	0.11	7.12	0.12	4.40	0.13	7.35	0.95
SST-D
		0.01 Hz	0.02 Hz	0.04 Hz	0.08 Hz	0.16 Hz
		Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain
None (N = 48)
	Mean	46.43	0.29	27.32	0.35	16.03	0.41	7.12	0.41	2.28	0.43
	SD	8.80	0.12	5.54	0.13	5.82	0.17	5.55	0.16	5.63	0.16
Low (N = 49)
	Mean	46.88	0.28	26.64	0.34	14.99	0.39	5.62	0.43	1.28	0.51
	SD	8.32	0.10	8.76	0.12	6.72	0.14	6.66	0.14	9.62	0.78
Medium (N = 73)
	Mean	47.88	0.27	28.24	0.34	13.93	0.38	5.80	0.40	1.46	0.45
	SD	9.10	0.11	8.45	0.13	7.02	0.16	6.76	0.16	7.85	0.45
High (N = 44)
	Mean	46.67	0.28	27.59	0.35	13.82	0.38	5.27	0.40	1.99	0.48
	SD	9.54	0.12	8.77	0.13	6.34	0.14	7.69	0.15	6.84	0.56
SST-L
		0.01 Hz	0.02 Hz	0.04 Hz	0.08 Hz	0.16 Hz
		Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain
None (N = 3)
	Mean	46.20	0.33	24.77	0.44	16.10	0.48	7.57	0.49	5.30	0.53
	SD	4.30	0.07	2.64	0.10	3.54	0.09	3.60	0.09	3.28	0.02
Low (N = 21)
	Mean	48.04	0.27	27.18	0.33	15.34	0.38	6.03	0.36	1.55	0.40
	SD	7.94	0.12	5.04	0.14	5.01	0.19	5.59	0.19	6.04	0.20
Medium (N = 20)
	Mean	50.55	0.28	30.01	0.34	17.68	0.44	7.75	0.44	2.49	0.44
	SD	8.15	0.12	4.94	0.14	6.52	0.18	4.65	0.17	8.73	0.19
High (N = 150)
	Mean	46.59	0.28	27.16	0.34	14.40	0.39	6.03	0.41	1.73	0.47
	SD	8.88	0.11	8.18	0.12	6.62	0.15	6.70	0.14	8.01	0.55
OVAR
		0.01 Hz	0.02 Hz	0.04 Hz	0.08 Hz	0.16 Hz
		Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain
None (N = 67)
	Mean	47.35	0.28	26.91	0.34	15.27	0.39	5.91	0.41	2.21	0.41
	SD	7.94	0.12	7.12	0.13	6.45	0.17	6.72	0.16	7.76	0.15
Low (N = 37)
	Mean	48.59	0.25	27.61	0.30	15.30	0.35	5.64	0.38	1.49	0.38
	SD	8.93	0.11	7.55	0.11	6.55	0.15	7.39	0.16	6.84	0.16
Medium (N = 69)
	Mean	46.84	0.28	27.46	0.36	15.27	0.41	6.45	0.42	1.88	0.51
	SD	9.41	0.10	7.79	0.13	5.10	0.15	5.78	0.14	7.45	0.67
High (N = 26)
	Mean	46.66	0.28	28.93	0.36	12.30	0.39	6.94	0.39	0.76	0.55
	SD	8.52	0.12	8.15	0.12	8.76	0.11	4.17	0.12	8.86	0.71
PF
		0.01 Hz	0.02 Hz	0.04 Hz	0.08 Hz	0.16 Hz
		Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain	Phase	Gain
None (N = 75)
	Mean	47.25	0.28	27.93	0.345	16.06	0.40	6.56	0.41	2.98	0.48
	SD	9.38	0.12	7.41	0.14	6.32	0.15	6.43	0.16	8.73	0.65
Low (N = 24)
	Mean	45.80	0.29	25.59	0.35	14.46	0.43	5.68	0.43	0.11	0.43
	SD	5.86	0.10	7.07	0.14	3.87	0.16	6.35	0.16	6.38	0.18
Medium (N = 8)
	Mean	45.21	0.29	25.95	0.36	13.73	0.47	3.65	0.46	1.69	0.45
	SD	7.52	0.12	7.33	0.12	7.13	0.19	2.83	0.12	3.70	0.14
High (N = 40)
	Mean	47.81	0.29	28.71	0.38	14.06	0.42	6.49	0.44	0.91	0.44
	SD	6.50	0.10	7.23	0.11	7.81	0.14	5.71	0.14	7.42	0.15

Despite these limitations, analyses of variance (ANOVAs) were used to determine whether there were significant differences between the VOR characteristics and the motion sickness level experienced during the provocative tests. The Bonferroni correction was used to reduce type I error for comparison of the various subgroups of subjects.

3 Results

The eye movement responses to different sinusoidal frequencies of rotation about a vertical axis are shown in Fig. 1. At the lowest frequency tested, the VOR gain was small; however, the gain increased as stimulus frequency increased. At the lower stimulus frequencies, the VOR slow-phase eye velocity led the chair velocity, and was therefore better correlated with head acceleration. As the rotation frequency increased, the VOR phase lead decreased, and became more correlated with head velocity. There was no significant difference between male and female subjects in VOR gain or phase (Fig. 1A). However, a between-groups ANOVA indicated a significant effect of age on the VOR gain (F(3,16) = 3.38, p = 0.04), with the VOR gain being smaller in older subjects (Fig. 1B).

Fig.1

Average horizontal VOR gain and phase measured during rotation frequencies ranging from 0.01 Hz to 0.16 Hz. Mean data±SE are shown for male and female subjects (A) and different age groups (B). Average horizontal VOR gain and phase as a function of motion sickness levels experienced during all 6 provocative tests (C) combined for subjects of all ages and both sexes (mean±SD). A positive value for phase means eye velocity leads head velocity.

Figure 1C shows the average gain and phase of VOR responses, at the 5 frequencies used, for the 4 subgroups of subjects who had no, low, medium, or high levels of motion sickness during all provocative tests combined. The relationships between the gain and phase of the VOR and motion sickness level were analyzed using two-factor ANOVAs (5 frequencies×4 motion sickness levels). Frequency had a significant effect on VOR gain (F(4,100) = 91.18, p < 0.001) and phase (F(4,100) = 3308, p < 0.001). A significant effect of VOR phase on motion sickness level was observed (F(3,100) = 8.38, p < 0.001), but VOR gain had no significant effect on motion sickness level (F(3,100) = 2.25, p = 0.09). Post hoc tests indicated that VOR phase of subjects who experienced no motion sickness was significantly different from that of subjects whose motion sickness level was low (p = 0.004), medium (p = 0.04), or high (p < 0.001), and there was also a significant difference in VOR phase between subjects with low and high motion sickness (p = 0.03) and between subjects with medium and high motion sickness (p = 0.01).

Sudden stops in the light (SST-L) provoked the most motion sickness, followed by head movements that generated cross-coupled angular acceleration during constant velocity rotation (CSSI and SVMT). Motion sickness was generally less severe during sudden stops in the dark (SST-D), OVAR, and parabolic flight (Fig. 2).

Fig.2

Number of subjects who experienced no motion sickness and high levels of motion sickness during each of the 6 provocative tests.

Figures 3 and 4 show the relationship between motion sickness level and VOR gain and phase measured for each rotation frequency. When subjects were groups by motion sickness level, single-factor ANOVAs indicated no significant between-group differences in VOR gain (Fig. 3). However, there was a significant between-group difference in VOR phase at 0.04 Hz (F(3,20) = 7.12, p = 0.002), 0.08 Hz (F(3,20) = 3.25, p = 0.04), and 0.16 Hz (F(3,20) = 7.86, p = 0.001). For these three frequencies, the mean VOR phase for the group who had no motion sickness was significantly different (p < 0.05) from the mean VOR phase for the group who had high motion sickness (Fig. 4). In addition, the mean VOR phase during 0.04 Hz rotation was significantly different for the groups with low and high motion sickness (p = 0.02). The mean VOR phase during 0.16 Hz rotation was also significantly different for the groups with no and low motion sickness (p = 0.01), for groups with no and medium (p = 0.02), and groups with medium and high motion sickness (p = 0.04).

Fig.3

Average VOR phase and gain during rotation at 0.01 Hz and 0.02 Hz during the vestibular test for the subjects who experienced no motion sickness and high levels of motion sickness during each of the provocative tests. Mean±SE.

Fig.4

Average VOR phase and gain during rotation at 0.04 Hz, 0.08 Hz, and 0.16 Hz during the vestibular test for the subjects who experienced no motion sickness and high levels of motion sickness during each of the provocative tests. Mean±SE.

4 Discussion

The values of VOR gain and phase in our subjects during sinusoidal stimulation in the dark are comparable to values reported in earlier studies of a smaller population. Our results confirm that VOR gain and phase are similar for male and female subjects, and that VOR response decreases as a function of age [23]. Although our measurements did not include motion perception, the decrease in VOR response with increasing age complements the observations that vestibular perception thresholds increase with age [2]. Recent studies of healthy subjects and individuals with vestibular disorders found that motion sickness declines with age [13, 22]. Unfortunately, we were not able to verify this finding because the database at our disposal did not include the age of the subjects in the various motion sickness groups.

In previous studies, investigators measured VOR during pre- and post-rotatory nystagmus elicited by angular velocity steps [7 , 28], and they found a correlation between an individual’s VOR time constant and their motion sickness level, which was assessed using a questionnaire. Our study compared VOR during sinusoidal oscillations in the dark with the severity of motion sickness during subsequent provocative tests, and involved a large number of subjects. Our results indicate a greater severity of motion sickness in subjects who had a shorter VOR phase lead during oscillations at 0.4 Hz, 0.08 Hz and 0.16 Hz. The shorter the VOR phase lead during sinusoidal oscillation, the longer the VOR time constant. Therefore, these results confirm the earlier evidence of a relationship between the VOR time constant and motion sickness. Also, in agreement with previous studies, our results show that motion sickness level is not correlated with VOR gain [24], indicating that motion sickness susceptibility is linked to the dynamic properties of the VOR rather than its sensitivity.

Shupak et al. [27] also used sinusoidal stimulations to study relationships between VOR and seasickness. They found significant increase in the VOR phase leads (decreased time constant) in response to yaw rotations at 0.01, 0.02 and 0.04 Hz after repeated exposure to rough sea conditions while habituation to seasickness was significantly correlated with the increase phase lead at 0.02 Hz. Our study complements these previous research results as the dynamic range of phase lead increment (and shortening the time constant of the VOR) would be larger in individuals with lower baseline phase leads.

Evidence of a relationship between velocity storage and motion sickness comes from the observation that labyrinthine defective patients with short vestibular time constants are immune to motion sickness [9]; whereas healthy subjects with longer VOR time constants are more susceptible to seasickness [14], airsickness [4], or motion sickness elicited by off-vertical axis rotation [28]. Studies using repeated exposure to rotating environments in the same individuals have shown that the severity of motion sickness was progressively reduced [6 , 29]. The VOR time constant also decreased as a result of this vestibular habituation [5]. Therefore, it has been proposed that the decrease in semicircular canal velocity storage was responsible for the reduction in motion sickness symptoms [18].

Motion sickness typically occurs during unusual body movements when there is a conflict between sensory and motor signals. Such conflict occurs when movement inputs coming from the visual, vestibular, and proprioceptive sensory systems, and the efferent copy coming from the motor system, do not match with the other sensory inputs and/or an internal representation of what motion is acceptable [18]. According to models of vestibular function, the sensory conflict is low-pass filtered before acting on the emetic center [21]. Thus, only low frequencies trigger motion sickness. A reduction in the time constant of the velocity storage, by decreasing the response to low frequencies, would eliminate the part of the sensory conflict that is the most likely to provoke motion sickness [24].

It has also been proposed that this reduction in time constant of velocity storage accounts for the absence of motion sickness sensitivity to cross-coupling stimulation in the weightless conditions of orbital [16] and parabolic [11] flight. The otolith organs are sensitive to linear acceleration and head tilt relative to gravity. Because the static spatial orientation function related to head tilt relative to gravity is absent in weightlessness, the velocity storage would be less efficient, thus resulting in less motion sickness [18, 28].

Our findings suggest that motion sickness might be reduced by decreasing the angular VOR time constant. The VOR time constant during sinusoidal rotation can be measured using the formula t = 1/(2πf tan φ) described in ref. [12], where f is the frequency at which the phase angle φ reaches 45°. Using this formula, we determined the time constant of the VOR in our study was 13.9 s (phase lead was about 45° at 0.01 Hz, see Fig. 1C) in subjects who had no symptoms of motion sickness and 15.3 s in the subjects who showed a high level of motion sickness. Studies of vestibular habituation of the VOR in humans have shown that the VOR time constant decreased with repeated testing. However, the VOR time constant never fell below 10 s [5], suggesting that only part of the component of the vestibular response produced by velocity storage can be habituated. This might explain why vestibular habituation of the VOR alone is not sufficient for desensitizing individuals who experience motion sickness [1].

Conflicts of interest

The authors declare that no conflicts of interest exist.

Footnotes

Acknowledgments

The authors thank Kerry George for editing the manuscript. This study was supported by the National Aeronautics and Space Administration (NASA).

References

Bagshaw

and Stott

J.R.

, The desensitization of chronically motion sick aircrew in the Royal Air Force, Aviat Space Environ Med 56 (1985), 1144–1985.

Bermúdez Rey

M.C.

, Clark

T.K.

, Wang

, Leeder

, Bian

, and Merfeld

D.M.

, Vestibular perceptual thresholdsincrease above the age of 40, Front Neurol 7 (2016), 162. doi: 10.3389/fneur.2016.00162

Bortolami

S.B.

, Rocca

, Daros

, DiZio

and Lackner

J.R.

, Mechanisms of human static spatial orientation, Exp Brain Res 173 (2006), 374–388.

Bos

J.E.

, Bles

and De Graaf

, Eye movements to yaw, pitch, and roll about vertical and horizontal axes: Adaptation and motion sickness, Aviat Space Environ Med 73 (2002), 436–444.

Clément

, Tilikete

and Courjon

J.H.

, Retention of habituation of vestibulo-ocular reflex and sensationof rotation in humans, Exp Brain Res 190 (2008), 307–315.

Clément

, Deguine

, Parant

, Costes-Salon

M.C.

, Vasseur-Clausen

and Pavy-LeTraon

, Effects ofcosmonaut vestibular training on vestibular function prior to space flight, Eur J Appl Physiol 85(2001), 539–545.

Cohen

, Dai

and Raphan

, The critical role of velocity storage in production of motion sickness, Ann N Y Acad Sci 1004 (2003), 359–376.

Dai

, Kunin

, Raphan

and Cohen

, The relation of motion sickness to the spatial-temporal properties of velocity storage, Exp Brain Res 151 (2003), 173–189.

Dai

, Raphan

and Cohen

, Labyrinthine lesions and motion sickness susceptibility, Exp Brain Res 178 (2007), 477–487.

10.

Davis

J.R.

, Vanderploeg

J.M.

, Santy

P.A.

, Jennings

R.T.

and Stewart

D.F.

, Space motion sickness during 24 flights of the space shuttle, Aviat Space Environ Med 59 (1988), 1185–1189.

11.

DiZio

and Lackner

J.R.

, Motion sickness susceptibility in parabolic flight and velocity storage activity, Aviat Space Environ Med 62 (1991), 300–307.

12.

Fernandez

and Goldberg

J.M.

, Physiology of peripheral neurons innervating semicircular canals of the squirrel monkey. II. Response to sinusoidal stimulation and dynamics of peripheral vestibular system, J Neurophysiol 34 (1971), 661–675 .

13.

Golding

J.F.

, Motion sickness susceptibility, Auton Neurosci 129 (2006), 67–76.

14.

Gordon

C.R.

, Spitzer

, Doweck

, Shupak

and Gadoth

, The vestibulo-ocular reflex and seasickness susceptibility, J Vestib Res 6 (1996), 229–233.

15.

Graybiel

, Wood

C.D.

, Miller

E.F.

and Cramer

D.B.

, Diagnostic criteria for grading the severity of acute motion sickness, Aerospace Med 39 (1968), 453–455.

16.

Graybiel

, Miller

E.F.

, Homick

J.L.

, Experiment M131: Human vestibular function, in: Biomedical Results from Skylab, NASA SP-377, Johnston

R.S.

and Dietlein

L.F.

eds., NASA, Washington DC, 1977, pp. 74–133

17.

Guedry

F.E.

and Graybiel

, Compensatory nystagmus conditioned during adaptation to living in a rotating room, J Appl Physiol 17 (1962), 398–404.

18.

Lackner

J.R.

, Motion sickness: More than nausea and vomiting, Exp Brain Res 232 (2014), 2493–2510.

19.

Lackner

J.R.

and DiZio

, Space motion sickness, Exp Brain Res 175 (2006), 377–399.

20.

Miller

E.F.

and Graybiel

, A provocative test for grading susceptibility to motion sickness yielding a single numerical score, Acta Otolaryngol 274 (1970), 1–20.

21.

Oman

C.M.

, A heuristic mathematical model for the dynamics of sensory conflict and motion sickness, Acta Otolaryngol (Suppl) 392 (1982), 4–44.

22.

Paillard

A.C.

, Quarck

, Paolino

, Denise

, Paolino

, Golding

J.F.

and Ghulyan-Bedikian

, Motion sickness susceptibility in healthy subjects and vestibular patients: Effects of gender, age and trait-anxiety, J Vestib Res 23 (2013), 203–209.

23.

Peterka

R.J.

, Black

F.O.

and Schoenhoff

M.B.

, Age-related changes in human vestibulo-ocular reflexes: Sinusoidal rotation and caloric tests, J Vestib Res 1 (1990), 49–59.

24.

Quarck

, Etard

, Darlot

and Denise

, Motion sickness susceptibility correlates with otolith- and canal-ocular reflexes, NeuroReport 9 (1998), 2253–2256.

25.

Raphan

, Matsuo

and Cohen

, Velocity storage in the vestibulo-ocular reflex arc (VOR), Exp Brain Res 35 (1979), 229–248.

26.

Reschke

M.F.

, Statistical prediction of space motion sickness, in: Motion and Space Sickness, Cramton

G.H.

, ed., CRC Press, Boca Raton FL, 1990, pp. 263–316

27.

Shupak

, Kerem

, Gordon

, Spitzer

, Mendelowitz

and Melamed.

, Vestibulo-ocular reflex as a parameter of seasickness susceptibility, Ann Otol Rhinol Laryngol 99 (1990), 131–136.

28.

Ventre-Dominey

, Luyat

, Denise

and Darlot

, Motion sickness induced by otolith stimulation is correlated with otolith-induced eye movements, Neuroscience 155 (2008), 771–779.

29.

Young

L.R.

, Sienko

K.H.

, Lyne

L.E.

, Hecht

and Natapoff

, Adaptation of the vestibulo-ocular reflex, subjective tilt, and motion sickness to head movements during short-radius centrifugation, J Vestib Res 13 (2003), 65–77.