Pattern Complexity Effects on Motion Sickness in an Optokinetic Drum
The experiment reported here was conducted using an optokinetic drum. Under typical optokinetic drum conditions, a stationary observer sits or stands within a large rotating cylinder as it rotates. The observer is instructed to simply view the drum’s interior as it rotates. Commonly used drum interiors consist of vertical stripes or randomly placed dots. For most observers, illusory rotation known as vection occurs in the opposite direction of the drum’s true rotation within 30 seconds. That is, the observer perceives self-motion, even though none is in fact occurring. Extended viewing often results in motion sickness (MS) symptoms. The goal of the current experiment was to test the effect of complexity on MS. It was hypothesized that increasing scene complexity would hasten the onset of MS in an optokinetic drum.
Participants- Twelve Saint Peter’s College undergraduate students and faculty members voluntarily participated in the experiment (5 males, 7 females). The age of participants ranged from 20 to 44 years (mean=25). Potential participants who reported no history of motion sickness susceptibility were not allowed to participate in the experiment. Persons reporting any visual, vestibular, neurological, or gastrointestinal abnormality, or any other general health problem, were not allowed to participate. Participants fasted for at least 2 hours before each experimental trial. The study protocol was approved in advance by the Saint Peter’s College Institutional Review Board. Each subject provided written informed consent before participating.
Apparatus-The optokinetic drum consisted of a synthetic composite cylinder 122 cm in height and 107 cm in diameter. The drum was suspended from a motor attached to a beam directly above the drum with two steel cables. Head position was maintained throughout the experiment by means of an optical chin rest. Viewing took place from the drum’s center resulting in a viewing distance of 53.5 cm when the participant’s line of sight was perpendicular to the drum’s surface. Horizontally positioned baffles attached to the top and bottom of the chin rest restricted the participant’s view so that the only surface seen through the baffles was the interior of the drum. Illumination was provided by two 32-W florescent bulbs positioned directly behind a translucent plastic diffuser panel and 102 cm directly above the top of the drum. The two stimulus displays are shown above. The display for the simple condition consisted of 12 alternating black and white vertical stripes that lined the optokinetic drum. The width of each stripe subtended 30° of visual angle. In the complex condition a black and white checkerboard pattern lined the drum. Each patch was 9° high and 30° wide. Thus both conditions were equal in terms of horizontal spatial frequency and the amount of black and white surface area visible at any given time.
Assessment Scales- Two subjective rating scales were used to assess each participant’s MS symptoms throughout each trial. One scale was a 0-10 overall well-being scale (0=I feel fine, 10=I feel awful as if I am about to vomit). The overall well-being scale was used to comply with the approved human subjects protocol. An assessment consisting of a single number was desirable in order to quickly assess when a trial should be terminated. Human subjects protocol dictated that an individual’s participation in the study be terminated when an overall well-being rating of “5” or more was obtained.
A second scale was used to assess eight subjective symptoms of motion sickness (SSMS). Each symptom was rated by the participant using a 0-3 scale (0=none, 1=mild, 2=moderate, 3=severe). The symptoms rated were spinning, dizziness, bodily warmth, headache, increased salivation, stomach awareness, nausea, and dry mouth. In the present study we wanted to track the development of symptoms throughout the course of each trial. In order to do so, ratings were obtained from each participant every two minutes. Therefore a simplified scale was needed so that participants could provide ratings quickly.
Procedure and Design- The participant sat inside the stationary drum and was familiarized with the two subjective rating scales (overall well-being scale and SSMS). Baseline overall well-being and SSMS ratings were obtained at the beginning of each trial. The participant was instructed to close their eyes until the drum steadily rotated at a speed of 30°/sec (5 RPM). Participants were instructed to open their eyes and overall well-being and SSMS ratings were obtained every two minutes throughout the trial. A trial concluded when the participant’s overall well-being rating was a “5” or higher or 16 minutes had elapsed.
Each participant served in both conditions in counterbalanced order; that is half the participants received 1) the simple condition that consisted of black and white stripes first then, 2) the complex condition that consisted of black and white patches while the other half received the reverse order. At the conclusion of each trial, the participant rested until the severity of their symptoms subsided. The participant was scheduled for a subsequent condition in 48-72 hours. After the participant served in both conditions they were debriefed and asked to describe their overall impressions of the conditions, especially impressions related to the complexity of the patterns that were viewed.
Results- A composite SSMS score was calculated for each participant by adding the subjective ratings (0-3) for each of the eight symptoms probed (spinning, dizziness, bodily warmth, headache, increased salivation, stomach awareness, nausea, and dry mouth). Although eight symptoms were probed, realistically, the highest possible score was a “21” because, it is not possible for a participant to experience increased salivation and dry mouth at the same time. Composite scores attained in the simple and complex conditions at the six-minute mark were analyzed. The six-minute mark was chosen because the data collected from all 12 participants is represented. At later minute marks, participants began dropping out of the experiment, having reached a “five” or more on the overall well-being scale. In accordance with human subjects protocol, participants attaining an overall well-being score of five or higher were not allowed to continue in the experiment. The mean SSMS composite score in the simple and complex conditions at the six-minute mark were 9.12 and 11.75, respectively. A t-test for repeated measures indicated that these SSMS composite scores were significantly different from each other [t(11)=2.1, p<.029, 1-tailed].These results clearly suggest that the addition of spatial complexity significantly hastens the onset of motion sickness symptoms.
Although the SSMS composite scores provide an overall indication of MS symptoms, not all MS symptoms probed were significantly affected by scene complexity. Individual t-tests for repeated measures indicated significant differences for dizziness [t(11)=2.0, p<.034, 1-tailed], headache [t(11)=2.16, p<.027, 1-tailed], and stomach awareness [t(11)=2.24, p<.023, 1-tailed]. Dizziness and headache can be considered to be major MS symptoms because they are particularly unpleasant. Stomach awareness, although not a major symptom per se, can be thought of as the precursor of nausea, which in turn, is the harbinger of vomiting.
Discussion- The results of the present experiment clearly indicate that as the number of visible surfaces in an optokinetic drum is increased, the onset of MS is hastened. In some environments, vection precedes the onset of MS. These environments include vehicle simulators, virtual reality, and optokinetic drums. Because of the correlation between vection and MS (6,17) it is tempting to consider vection as a causal factor in MS onset. Such an assumption however would probably be incorrect. Consider that a significant number of people experience vection but not MS. Also, individuals with nonfunctioning vestibular systems are immune to MS but not vection (11). How then are vection and MS related? The answer may be that in environments where vection can occur there is often a disparity between visual and vestibular sensory inputs. It then stands to reason that the greater the magnitude of visual/vestibular disparity, the faster MS symptoms will become manifest.
When an image moves on the retina there are two possible causes: 1) an object moved relative to the observer who is stationary, or 2) self-motion has occurred. How does the visual system determine which possibility accounts for the retinal image movement? Clues regarding this visual system challenge can be gained by analyzing the visual characteristics of something that is almost always perceived as stationary: the natural environment. One characteristic of our environment is that it is large relative to ourselves. Also, it completely surrounds us. The natural environment also contains many distinct surfaces and many different colors making it visually complex. When an image of our environment moves on our retina we generally perceive self-motion in the form of an eye movement, a head movement, a body movement, or some combination of these three. It then follows that characteristics of the natural environment implemented into an artificial environment could influence how stationary that environment is perceived to be, even if it is fact moving.
Consider the optokinetic drum used in the current experiment. Like our natural environment it completely surrounds the participant. Previous work has shown that if chromaticity is included in the drum pattern, both vection and MS onset are hastened. Chromaticity is a common feature of the natural environment. Results of our complexity vection experiment indicated that increasing spatial complexity by increasing the number of visibly distinct surfaces hastens the onset of vection and increases the magnitude of the self-motion that is experienced. In the current experiment, the same increase in complexity yielded more severe MS symptoms. It seems then that adding natural environment characteristics to an optokinetic drum increases the probability that the drum will be perceived as stationary, what we call field stability. In other words, the more the drum (or any other artificial environment) becomes like the natural environment, the more it will tend to be perceived as stationary. As field stability increases, moving optical flow patterns will be more likely to be interpreted as the result of self-rotation. Because the self-rotation is illusory, sensory input from the visual and vestibular systems will not be in agreement. The more extreme the sensory disparity becomes, the more likely MS symptoms will result.
Application of Results- The benefits of simulator and virtual reality training are often adversely affected by MS symptoms. Optokinetic drum conditions are similar to those occurring in some vehicle simulators (fixed-base) and virtual reality environments in that the self-motion that is experienced is created by moving optical flow patterns. The current results suggest that simulator sickness may be reduced by altering the spatial complexity of the displays that are used. Results of the current experiment suggest that as a display becomes more realistic, that is, more like the real world, the faster the onset of MS symptoms may be. On the surface, this seems to present a dilemma. In order for simulator training to be the most beneficial, the displays used should be as realistic as possible. However, for some training objectives some visual field features may be removed or reduced without compromising training objectives. Further research is needed to more fully understand how spatial complexity affects MS symptoms in simulators and VR environment.