Symposium on the Role of the Vestibular Organs in Space Exploration (1970)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Effect of Instability During Rotation on Physiologic and Perceptual-Motor Function BERNARD D. NEWSOM ] General Dynamics SUMMARY The requirement for an artificial-gravity space station has not been established because of the limited duration of space missions to date. The eventuality of such a system, however, is generally accepted on the justification of comfort, training, facilitation of mechanical operations, and the more natural environment it offers for prolonged missions. A rotogravity environment has been studied on the basis of radius requirements and rotational velocity limitations for crew habituation and perform- ance. The added parameter that influences design and propellant costs is the stability required for the crew to operate satisfactorily. The effect of perturbation during rotation is the subject of this study. INTRODUCTION This study was completed in three experiments. Experiment 1 investigated the effect of rotation and sinusoidal perturbation on a seated subject performing a battery of perceptual-motor tests representative of the tasks required by an as- tronaut crew. Experiment 2 imposed the added factor of a rapid head motion upon the perturbat- ing environment at high rates of rotation. The criteria were performance of an eye-hand re- sponse test and the time it took for the subject to fixate on a point display following the head turn. Experiment 3 was performed to assess the perturbation effects on the time it took for the habituation of a single perceptual element to rotation. The disorientation that results from a rapid head motion can be evaluated by the oculo- gyral illusion (OGI) that occurs in a darkened room. An illuminated target appears to drift about, and this motion is extinguished as habitua- tion occurs. Results of the three experiments indicate that 1 Assisted by J. F. Brady under contract no. NAS 9-6986 at General Dynamics Convair Division. sinusoidal motions of the magnitude anticipated in space stations should not complicate the per- formance of crew tasks that do not require the translation of a subject in that environment. Perceptual-motor performance was not degraded; rapid head turns caused no more decrement in performance during perturbation than in stable rotation; and the rate of OGI extinction was the same, even though the extent was less, in the perturbating environment. The consistency of results supports the hypoth- esis that perturbation during rotation creates a constant stimulus to the vestibular system and, as such, keeps the receptor in a partial refractor state which, in turn, raises the threshold for sensitivity to cross-coupled acceleration. Certain basic requirements can be predicted for those future vehicle systems requiring multi- man crews to function optimally for long periods in space. A reliable life-support system free from texicological problems, for example, is a well-recognized requirement. A requirement not so well agreed upon is for artificial gravity to maintain the musculoskeletal and cardiovascular systems of the crew close to normal equilibrium for Earth reentry. Much of the thinking in this

308 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION field has been about the immediate problem of relatively short-duration manned space missions, and it appears from available information that man can survive such exposure. However, there are advantages, in addition to physiological support, to be gained from creating an inertial field. Assembly of parts, repair of equipment, and preparation and ingestion of food are all facilitated by an inertial directing force. Per- sonal hygiene can be more closely controlled and sanitation is easier to maintain if dirt and fluids collect on the floor. Artificial gravity also allows the enjoyment of an occasional shower which is usually a priority desire in confinement studies. All of these considerations take on new impor- tance when mission durations are extended to a year or more, provided that such missions are within engineering feasibility and physiological tolerance. ENGINEERING GUIDES To be prepared for possible development of these future artificial-gravity systems, it is nec- essary to start research on the support tech- nologies. The type of information design en- gineers require concerning man's functional tolerances in these environments presently does not exist. Although the response of the otic labyrinth to both weightless and artificial gravity environ- ments certainly requires further elucidation, this study has been directed primarily toward ques- tions concerning man as a functional system in these environments. Engineers choosing op- timal man-machine tradeoffs must know the work potential of man in rotating vehicles of various dimensions and force field characteristics. These guidelines must be realistic —not only for the large orbiting vehicles of the future but also (and more acutely) for experimental systems that could be included in the Apollo Applica- tions Program and as backup concepts for vehicles now in the definition phase. Loret (ref. 1) and Dole (ref. 2) have listed many engineering constraints imposed by the crew during rotation, but neither considered sta- bility of the vehicle as a design limitation. How- ever, in providing a habitable rotating space vehicle, stability could be no less important than angular velocity, radius. #r-level, or rim velocity. In a rotating space vehicle, vehicle precession, as well as head rotation, could stimu- late the crewman's labyrinth. Vehicle instability could be anticipated to lower the permitted angular velocity as it seems reasonable that the stimuli to the labyrinth due to vehicle instability would complement that due to the crewman's active head movements. Vehicle precession predicates caution in as- signing spin-rate ceilings at this time. Investiga- tion of the total dynamic environment in a simu- lated rotating vehicle in relation to habitability and crew performance is necessary. Without such groundwork, design engineers must work in an arbitrary manner, which could be costly and mission limiting. Kurzhals et al. (ref. 3) and Larson (ref. 4) have lent theoretical and empirical consideration to the engineering prob- lem of instability in the manned rotating vehicle. Tests of crew performance as a function of in- stability, as well as of the previously considered parameters, should be coupled to such efforts. Disturbances, such as docking impacts and active or passive changes in crew or hardware mass, may cause many combinations of structural and force-field oscillations which could be detri- mental to crew function. As stability of a rotat- ing space vehicle is related directly to its total mass, the relatively light state-of-the-art vehicles would be particularly susceptible to instability from mass disturbances. In a discoid or toroidal vehicle rotating about its principal Z-axis, a crewman alined with one of the transverse X- or K-axes could be subject not only to the dis- turbing effects resulting from his active head movements relative to the spin plane, but also to a variety of oscillating forces beyond his active control. Any impulsive torque applied about either one of the two transverse axes of the rotating vehicle would result in a wobble (defined as an oscillatory curvilinear movement) about both A and Y trans- verse axes. The amplitudes of these oscilla- tions would be directly proportional to the angular impulse and inversely proportional to the moment of inertia around the transverse axis normal to the axis of torque. A reduction in vehicle size causes a dramatic increase in

EFFECT OF INSTABILITY DURING ROTATION 309 wobble. Several methods of active or passive dampening can be used to increase stability, but they entail weight and power penalties. An inertial unbalance produced by an uncom- pensated mass movement along a transverse axis within the plane of spin will couple with the moment of inertia about the spin axis to produce a disturbance about the transverse axis that is directly proportional to the initial vehicle spin rate and the product of the mo- ments of inertia of the transverse and spin axes. This generated spin coupled with the initial vehicle spin will produce varying angular velocity patterns. A crewman alined with this axis will experience the illusion of complex and ever-varying tilting of the floor as his body perceives the resultant of the linear acceleration oscillating along his longitudinal body axis and the linear acceleration normal to this axis. The linear acceleration normal to this axis would trace the vectorial pattern defined by Larson (ref. 4). Simultaneous dynamic mass unbalances along both transverse axes (the anticipated situ- ation) would complex the vector pattern and the resulting disturbances. APPROACH In a rotating environment, the crew is subject to Coriolis accelerations that may result in dis- orientation, vertigo, and motion sickness (refs. 5 to 7). Even without voluntary movement on the part of the crewman, similar effects could possibly be caused by the vehicular perturbations resulting from internal or external mass unbal- ances in an insufficiently stabilized vehicle. These perturbations —similar in magnitude and effect to the movements of a ship at sea—produce cue conflicts in the orientation triad of vision, inner ear, and deep proprioception. INFORMATION GOAL The problem of stability in a rotating space vehicle has received little study in biologic laboratories, probably because of the relatively sophisticated simulator required. It has not been similarly neglected as an engineering con- sideration and deserves comparable study by workers in bioastronautics. The low-frequency oscillations involved in perturbation have been thought to affect macular sensors, as well as cupular sensors, while the Coriolis accelerations appear to be detected primarily by the cupulae. Those functions found in previous tests to be sensitive to Coriolis accelerations might be expected to degrade even further due to the simultaneous perturba- tion of the rotating environment. MRSSS EXPERIMENTATION On the basis of the studies that have been completed by using the manned revolving space station simulator (MRSSS),2 it has been con- cluded that man can adjust to and resist per- formance decrement in a stable rotogravity environment of 6 rpm (refs. 7 and 8); that the functional adjustments required by his passing into and out of a rotating environment can be eased by graduating such transitions in a step- wise manner; that prior to making the physiologic adjustments required by such transitions, he can adjust his behavior very rapidly to perform optimally (ref. 9); and that proper location and orientation of control and display hardware can maximize his performance (ref. 10), especially during the adjustment to force-field changes (ref. 11). A rotating spacecraft, exposed to either static or dynamic unbalances, would respond gyro- scopically with a tendency to oscillatory preces- sion. Presumably, the spacecraft will have stabilization systems to counteract such disturb- ances, but it is important to know to what extent this control must be effective and what the off- nominal operation tolerance of the crew is in event of system failure. In a rotating environ- ment, this oscillatory precession (wobbling) may create disorientation similar to that resulting from active movements by passively moving the crew. This wobbling can be effectively simu- lated by oscillation of the MRSSS as it rotates. To facilitate this simulation, the hydropneumatic ram system used to incline the MRSSS on the trunnions was modified, thus making it capable 2 The MRSSS is located at Convair, San Diego, Calif., and incorporates the Air Force CEVAT centrifuge complex.

310 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION of executing sustained programs of oscillation. With the capability to simulate an oscillating, rotating environment, it was then necessary to select performance tasks appropriate for deter- mining functional limits within such an environ- ment. Experience from previous testing in the MRSSS has demonstrated that performance adjustment during rotation occurs rapidly. GENERAL PROGRAM PLAN The trunnioned cabin of the Convair MRSSS (fig. 1) can be inclined so that the resultant force field is perpendicular to the center of the floor. It also provides a means to perturb the cabin and to vary the angular velocity so inertial forces within the chamber simulate the vector patterns predicted for various sizes and configurations of spacecraft rotated to produce artificial gravity. No information on how such perturbations affect general performance is available; this study is the preliminary effort to generate such data. The perturbations are expressed as deviations of the inertial vector about the subject's Z-axis, so the results of the study can be applied to vehicles of any radius. Studies by other authors (ref. 12) indicate that the problem is virtually radius independent because of the relative vestibular sensitivities to the motions concerned. The following three studies were completed to assess the effects of perturbation on perform- ance during rotation prior to habituation. FIGURE 1. —Perpr tunl-motor-test consoles. Experiment 1 Baseline measurements were made on 12 subjects under static conditions. Biofunctional efficiency tests were administered repeatedly until a consistent score was obtained. Then, the tests were repeated while the subject was exposed to the following: (1) ±3° perturbation at 0.1 Hz, (2) 6 rpm at a 20-foot radius, and (3) simultaneous perturbation and rotation. Experiment 2 The technique developed on the previous con- tract (NAS 9-5232) was used as a performance measurement to assess the importance of orienta- tion within an inertial force field perturbed at 0.1 Hz. Ten subjects were tested following baseline measurements. Their performance was measured after making F-axis head turns at 0°, 45°, and 90° to the spin plane. Regres- sion slopes were determined at the following: (1) ±3° perturbation at 0.1 Hz, (2) 12.4 rpm. and (3) 12.4 rpm and ±3° perturbation at 0.1 Hz. Experiment 3 Vestibular habituation was compared for a specific head movement while subjects were exposed to rotation and perturbation independ- ently and in combination. Sixteen subjects were exposed to 8 rpm for 4 hours; every 15 minutes, a frontal (A"-axis) head turn was made to the right shoulder, and the duration of OGI was recorded. The extinction rate of the OGI was compared at 0° perturbation and then with ±3° perturbation at 0.1 Hz. The first eight subjects were tested with 0° perturbation first and with ±3° perturbation 2 weeks later. The second eight subjects were tested in reverse order. All subjects received aural caloric tests before and after rotation. PERCEPTUAL-MOTOR RESPONSE TO ROTATION, PERTURBATION, AND COMBINED ROTATION AND PER- TURBATION Experiment 1 A survey of possible tests and test batteries was made to find a series of appropriate per- formance measurements. Both the Air Force

EFFECT OF INSTABILITY DURING ROTATION 311 and NASA (ref. 13) developed "face-value" type consoles for the specific purpose of evaluating man's control performance in the types of en- vironments of interest to this study. Arrangements were made to borrow a unit of the perceptual-motor console model 766:l from the Manned Spacecraft Center (MSC) in con- junction with the performance measurements being made under contract No. NAS 9-5232 and for use in a pilot perturbation study being con- ducted in the MRSSS. Following the pilot study, permission was obtained from MSC to perform this study as part of the experimental effort. The MRSSS is instrumented and equipped to permit continuous rotational studies of four sub- jects for unlimited periods of time and is de- scribed in greater detail in previous reports (ref. 14). For this study, the room was spun at 6 rpm, an angular velocity which previous studies had determined to be realistic for a stable roto- gravity spacecraft of projected dimensions (ref. 8). At this spin rate, and with an effective radius of slightly more than 18 feet, a resultant 1.04 g was imposed on the study participants and the inclination of room vertical (about which per- turbation took place) was 14°. A perturbation profile of 0.1 Hz and ±3° was selected to simulate a reasonable maximum to be encountered in projected rotogravity vehicles (ref. 3). The testing consoles were designed for NASA by J. F. Parker, Jr., et al. (ref. 13) to integrate devices for a battery of tests that measure the primary dimensions of perceptual-motor perform- ance. The tests used are based on well-estab- lished techniques with well-documented inter- pretations of results. Primary dimensions to be measured were chosen to specify abilities under- lying complex perceptual-motor performance and to relate significantly to tasks and duties to be performed by spacecraft personnel. The two consoles (subject and examiner con- soles) are shown in figure 1. Additional test items and accessories included are the manual dexterity test, the stylus and mirror-tracing maze, the mirror-tracing visor, and the finger-dexterity test. The subject and examiner consoles are arranged so the examiner can visually monitor both consoles and the subject (fig. 1). For this test, the subject was seated in the center of the simulator, the position at which the gravito- centrifugal resultant was perpendicular to the floor. The subject faced tangentially (in the direction of spin), as he would if optimally posi- tioned for monitoring a control-display console (ref. 10) in a rotating space vehicle, and in a posi- tion in which he would be perturbated from side to side by vehicular wobbling. Adequacy of identification (ref. 13) indicates the results of the factor-analytic study. For a test to be considered, it had to load 4 at 0.30 or above on the primary factor or ability. If the test was not considered pure (loading only on one factor), secondary factors had to load at 0.30 or above. Previous studies indicated that acceleration produced by the product of angular velocities resulting from head turning and vehicle rotation reduced performance in a rotating environment and that this reduction was partly due to visual location of the task display (ref. 15). Perturba- tion of the subject during rotation can theo- retically elicit similar labyrinthine response from resultant velocity products. Performance could therefore be degraded without movement on the part of the subject (passive motion); however, previous testing had also demonstrated a rapid adjustment to the oscillating force field (rotary pursuit tests) (ref. 9) and fast recovery following forced head turns (ref. 16), so it might be an- ticipated that adaptation would result from sinus- oidal perturbations during rotation. If such adjustment could be determined, the requirement for spacecraft stabilization would be reduced. Procedures The MRSSS was used to test the subject for performance while the following activities were taking place: (1) static, (2) perturbating, (3) ro- tating, and (4) rotating with perturbation. Per- turbation of ± 3° was around the inclined resultant at 0.1 Hz. The rotation rate was 6 rpm. The NASA perceptual-motor test console was 3 Fabricated under contract no. NAS 3-1329 by Biotech- nology, Inc., Arlington. Va. 4 Loading designates the degree of specificity for measure- ment of a particular ability.

312 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION [A—»E order of testing] Subject No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Static A B A B A C B D E A C D B E A D B C E A D C B E A B A B D A C B D E A C D B E A D C B A D B C E A B C D E A B D C E A C Perturbate Rotate C D E D C E C D E B D E P&R . . . C Static E E used, and all 18 tests were performed. The console was located at the simulator center with the subject facing the leading bulkhead. The subject was seated in front of the console with the test conductor immediately behind him. Each subject practiced on the console for a min- imum of three sessions prior to testing on the centrifuge. At the end of each session, a score was obtained to determine the state of learning. No subject was tested that had not reached a learning plateau on each test. Fifteen subjects were scheduled for testing. The results from the iirst 12 tested successfully were used. The testing order was as indicated in the table above. Results The tests were performed on 12 subjects with- out difficulty. Subjects were not distressed by the environment and appeared to enjoy the testing procedure. The scores, shown in figures 2 and 3, are arranged into groups with common test objectives. The test console was quite reliable during the tests; the only problems encountered were as follows. One microampere meter failed to work 70 60 50 40 30 20 10 ROTATION = 6 RPH PERTURBATION 0.1CPS, j TEST ARM-HAND STEADINESS (AHS) WHIST-FINGER SPEED (WFS) FINGER DEXTERITY (FD) MANUAL DEXTERITY (MD) POSITION ESTIMATION (PE) RESPONSE ORIENTATION (RO) CONTROL PRECISION (CP) SPEED OF ARM MOVEMENT (SAM) MULTIL1MB COORDINATION (MLC) POSITION REPRODUCTION (PR) MOVEMENT ANALYSIS (MAI MOVEMENT PREDICTION (MP) RATE CONTROL (RC) ACCELERATION CONTROL (AC) PERCEPTUAL SPEED TIME (PST) PERCEPTUAL SPEED ACCURACY (PSA) TIME SHARING (TS) VISUAL REACTION TIME (VRT) AUDIO REACTION TIME (ART) MIRRIOR TRACING SPEED (MTS) MIRROR TRACING ACCURACY (MTA) I 60 50 40 30 20 PER CENT DECREMENT I 10 20 PER CENT IMPROVEMENT FIGURE 2. —Percent change in perrotatory performance with perturbation. TEST ARM-HAND STEADINESS (AHS) WRIST-FINGER SPEED IWFS) FINGER DEXTERITY (FD) MANUAL DEXTERITY (MD) POSITION ESTIMATION (PE) RESPONSE ORIENTATION (RO) CONTROL PRECISION (CP) SPEED OF ARM MOVEMENT (SAM) MULTILIMB COORDINATION (MLC) POSITION REPRODUCTION (PRt MOVEMENT ANALYSIS (MA) MOVEMENT PREDICTION (MP) IIATE CONTROL (RC) ACCELERATION CONTROL (AC) PERCEPTUAL SPEED TIME (PST) PERCEPTUAL SPEED ACCURACY (PSAi TIME SHARING (TS) VISUAL REACTION TIME (VRT) AUDIO REACTION TIME (ART) MIRROR TRACING SPEED (MTS) MIRROR TRACING ACCURACY (MTA) 1.0 2.0 IMPROVEMENT .001 pOS .02 |0.I| 0.5 .002 .01 .05 0.2 P-VALUE' W///A P-VALUE i0. OS, kXX/J^ TWO-TAILED COMPARISON •STUDENT'S t-TEST. TWO-TAILED COMPARISON FIGURE 3. — Significance of change in perrotatory performance with perturbation.

EFFECT OF INSTABILITY DURING ROTATION 313 for time-sharing and perceptual-speed tests and was replaced. The cam-switch programer was subject to failure unless it was cleaned regularly. Operation of some tests required considerable practice before consistent scores were reached, and even after achieving consistency, it was found that some subjects demonstrated increased capability in the postdynamic test period. Discussion Because some learning was still evident in the more difficult tests at the end of the desig- nated 4.5 hours of training, the better of the two static test scores (predynamic and postdynamic) was selected as the score of 100 percent for normalization of the dynamic data. With the above precaution in mind, the results of an analysis of variance of the normalized data suggest the following: (AHS) static (S) over all three dynamic modes (P < 0.01). (WFS) S over all three dynamic modes (P < 0.05). (FD) S over rotation (R) and rotation and perturbation (RP)(P<0.05). (MD) S over R (P < 0.05). (SAM) S over R and RP (P (MA) S over R (P < 0.01) and RP (P < 0.05). (PST) S over R (P < 0.05). (TS) S over RP(P< 0.05). (VRT) S over all (P < 0.05). (ART) S over P(P< 0.01). (MTS) S over Pi(P< 0.01). (MTA) S over RP (R < 0.01). All other tests failed to demonstrate significant decrements from static scores. The results in some cases show a significant change in score, but only in one test does decrement indicate the perturbation would be a hindrance to that type of operation. As might be expected, the task of maintaining a probe within a hole without touch- ing the sides is difficult in the rocking environ- ment. Such a task would still be possible if there were a contact between a fixture common to that being probed and the carpal portion of the hand. In the administration of these tests, there was little, if any, head motion, so most vestibular stimulation resulted from passive motion of the subject by vehicle perturbation. Quite different results might be expected if a head motion were incorporated into the testing. Figures 2 and 3 indicate the relation between the tests as affected by the imposed variables and the significance of the differences in scores. EFFECT OF PERTURBATION ON PER. FORMANCE FOLLOWING \-AXIS HEAD TURNS DURING ROTATION Experiment 2 A previous study which was reported at the last symposium (ref. 17) demonstrated a progres- sive performance decrement as the angle be- tween the spin plane and the plane of the sub- ject's head turn was increased. Experiment 1 of this study did not indicate that the passive movement of a subject during rotation, due to perturbation of the vehicle, would cause a sig- nificant problem. Those tests, however, re- stricted head motion to a minimum for the test being performed. Motion out of the spin plane was sinusoidal and, therefore, predictable in time and magnitude. Tasks requiring head turns in different planes impose cross-coupled stimuli on the semicircular canals that vary in magnitude and direction, depending on subject orientation, vehicle rotation rate, and motion due to the perturbation. The inertial field of such motion is complex and changing, and it is un- likely that prediction would be possible. Rapid habituation would most likely result only from nonspecific suppression of the vestibular signal. Method The same test arrangement was used for this study as in previous published studies (ref. 16). Recording techniques, however, were extended. In addition to the eye-motion camera (EMC) and the vertical and horizontal electro-oculogram (EOG), a vectoroculogram (VOG) was made. This is an integration of the horizontal and verti- cal EOG and provides a light spot on the cathode- ray tube (CRT) that represents the point of visual fixation. The response analysis tester (RATER) test (as reported in ref. 16) was used, and the head-turn recording, EOG, EMC, frame number, and VOG were "stacked" and recorded on a video tape recorder. This allows simultaneous data presentation for analysis. Only K-axis head turns were made. This limited the antici- pated severity of semicircular canal stimulation

314 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION and reduced the testing per subject so it could be accomplished in 1 day. The primary goal of this study was to acquire information of the effect of perturbation on the process of adjustment to rotation. A secondary purpose was to investigate the use of the VOG as a practical substitute for the EMC for record- ing eye movement. Subjects The test sample consisted of 12 volunteers from San Diego colleges. They were males of 22 ±2 years, 167 ±23 pounds, and 69±4 inches in height. An additional six subjects of compara- ble background and vital statistics began testing but did not finish the required regimen —three because of instrumentation malfunctions and the remaining three because they found the most stressful orientations physiologically disturbing. Before being exposed to the environment of the dynamic test simulator, the subjects were re- quired to pass an airman's third-class medical examination. None of the subjects had histories of undue susceptibility to motion sickness; no special instructions were given regarding diet or rest preceding the day of experimental performance. Apparatus The simulator, subject restraints, data pickups at the subject, and performance tester were unchanged from the previous contract. Rotation of the subject was achieved by using the MRSSS. Orientation and restraint were achieved by using the 45° inclined chair and its associated head restraint. With the centrifuge rotating at 12.2 rpm to produce 1 radial g at the restraint chair position (centrally located within the MRSSS at 20 feet from the centrifuge spin axis), the gravito- centrifugal resultant was perpendicular to the MRSSS floor when the simulator was tilted at a 45° angle. The 45° angle tilt combined with the 45° angle of the restraint chair tilting the subject on his left side permitted (as in the previous contract) orientations of the subject for K-axis head turns at 0°, 45°, and 90° interplanar angles to the centrifuge spin plane by simply rotating the chair (fig. 4). The EMC—2F eye-motion camera (made by Westgate Laboratories) was again used to record direction of gaze. The camera was fixed to the subject by the head- restraint and dental-bite bar as shown in figure 5. The RATER was again used as the perceptual- motor performance device with the display collimated to 1° of visual angle. The RATER tests correct rote responses to lights of four different colors (red, yellow, green, and blue) that are presented in an infinitely random order. The subject depresses one of four console FLOOR PLANE RESULTANT FORCE AT 12.2 RPM SPIN PLANE END VIEW OF MRSSS SIDE VIEW MRSSS FIGURE 4.—Subject orientations.

EFFECT OF INSTABILITY DURING ROTATION 315 FIGURE 5. —Camera fixed by head restraint and dental bite bar. FIGURE 6. —Subject in the inclined chair. buttons, seen in figure 6, that corresponds to each color; when the correct button is pressed, the next color appears. Total responses and correct responses are recorded automatically. The RATER also produces a dc signal that cor- responds to the subject's latency in responding to each color displayed. The RATER response latency was recorded on the polygraph strip- chart. Essential changes in apparatus from the previous contract were made to incorpo- rate the VOG system into the study for addi- tional data recording. The subject's dc EOG signals were amplified by Kintel 114A am- plifiers in the MRSSS before being trans- mitted through the centrifuge sliprings to the main test control room. In the control room, the EOG signals were simultaneously recorded by two systems. The individual horizontal and vertical EOG were written out by two channels of a Sanborn 150 polygraph on the same chart that recorded the subject's and the onboard examiner's ECG, the RATER response latency, the testing event marker, and the head-turn rate. The horizontal and vertical EOG signals were also paralleled to the VOG system. For this system, the two EOG signals were combined by a Tektronix 503 dual-beam oscilloscope into a single two-dimensional eye-movement vector. To provide an integrated display for data records, three readouts (polygraph EOG, oscilloscope VOG, and a digital count of the EMC frame number) were individually photographed by separate television cameras and synchronized into a single picture by a special effects generator for video taping and/or subsequent kinescoping. Experiment Design As in the previous contractual study, the sub- ject's task was to perform a testing sequence in each of the combinations of head-turn orienta- tion and force-field variation. Each sequence consisted of ten 15-second trials separated by 20-second waiting periods at the cue-light posi- tion. All head turns were restricted about the subject's cranial y-axis by the head constraint. The subject would wait for the cue light with his head dorsoflexed 35° back from the hori- zontal. The automatic timing system turned on the cue light and the RATER display and started the EMC simultaneously. The subject imme- diately turned his head 70° downward as rapidly as possible and began responding to the RATER display by pressing the appropriate buttons. At the end of the 15-second scoring period, the timer would shut off the RATER display, which was the subject's signal to turn his head back to the cue-light position and to wait for the light to flash on at the end of the 20-second waiting period. At the end of the 10-trial sequence (150 seconds of scoring), the EMC film was

316 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION changed, and the subject was reoriented by rotat- ing the restraint chair relative to the centrifuge spin axis. Each of the subjects was instructed to perform as well as possible during the 150-second (10- trial) sequence in each orientation. The sub- jects were tested in the same six modes: two each with the head-turn plane at 0°, 45°, and 90° to the MRSSS spin plane; once with the simu- lator rotating at 12.2 rpm without perturbation; and once with it rotating at 12.2 rpm and per- turbating about a tangential axis at ± 3° at a rate of 0.1 Hz. Permutations of the six testing modes were selected on a basis of balance and ran- domly assigned to the 12 subjects. The subject numbers and their assigned orders of testing sequences are shown as follows: 1 S 2 P 3 S 4 P 5 S 6 P 1 S 8 P 9 S 10 P 11 S 12 P S 0°, 45°, and 90° 0°, 45°, and 90° 90°, 45°, and 0° 90°, 45°, and 0° 45°, 0°, and 90° 45°, 0°, and 90° 90°, 0°, and 45° 90°, 0°, and 45° 45°, 90°, and 0° 45°, 90°, and 0° 0°, 90°, and 45° 0°, 90°, and 45° = 12.2 rpm P 0°, 45°, and 90° S 0°, 45°, and 90° P 90°, 45°, and 0° S 90°, 45°, and 0° P 45°, 0°, and 90° S 45°, 0°. and 90° P 90°, 0°, and 45° S 90°, 0°, and 45° P 45°, 90°, and 0° S 45°, 90°, and 0° P 0°, 90°, and 45° S 0°, 90°, and 45° P = 12.2 rpm and ±3° at 0.1 Hz Results Of the 18 subjects who began testing, 12 performed at all six modes and were grouped within a single sample for data reduction and interpretation. For these 12 subjects, two categories of response were considered: (1) the net RATER score (total correct responses minus total incorrect responses) for the 150 seconds of scoring during each modal sequence and for each of the 10 trials making up a sequence; and (2) the parameters of reaction linking the starting signal to the first correct RATER response —the latent period to beginning of head turn, the head turn, end of head turn to eye fixation, and from eye fixation to first correct RATER response. Figure 7 shows the mean net RATER score for all 12 subjects for each of the ten 15-second trials in each of the testing modes. The graph (fig. 7) indicates that for most trials there is a ING Y RDrCATW NGLE t» W«(BES| RETOrTIS HAKI HUUKi »PW PLANE. FIGURE 7. -Net RATER score versus test trial. n „ , o a R ROTATTOK AT II. 1 HP» RP ROTATON AT U.2 RPM • PCRT1 RBATK* OF • :.* AT o.I en SLiiainiPT roLLocr^r. ^ IM*.'*TF» \St,LI .ix I)E(.RH'!*. Hi inir\ PUUVT or HEAD TVHX A>T> , EN rwn « MH-v-J-, M«IS PLASi . FIGURE 8. —RATER score versus subject orientation. decrement in the mean scores as the interplanar angle increases; and for all interplanar angles, there tends to be a reduced performance when perturbation is added to rotation. The excep- tions to the latter are the last four trials of the ¥90 sequences where there is a mean-score decrement without perturbation when compared with the situation of combined dynamics. Figure 8 shows the mean net RATER score for the full sequence as a function of interplanar orientation for each of the dynamic modes. This figure makes the two observations mentioned above more apparent. A decrement of perform- ance is seen as the interplanar angle increases. This decrement is increased with perturbation but becomes obscured during the most stressful interplanar orientation by the cross coupling at YM. To test the statistical significance of the data presented in figure 8, each subject's data were

EFFECT OF INSTABILITY DURING ROTATION 317 normalized on the basis of 100 percent perform- ance value for his best sequence score. Overall means for the total sample then gave performance percentages ranging from 96 percent for Y0 (R only) to 80 percent for Y»0 (RP). An analysis of variance was performed on the normalized data using a P-value equal to or less than 0.05 as being significant. Comparing sequences with one another, only Y90 (R) and Y9o (RP) demonstrated a significant degradation in performance and only when compared to Yo (R). Not included on this graph were the results of perturbation-only sequences run by the last 6 of the 12 subjects. Each of these subjects ran a prerotation and post- rotation perturbation sequence —three subjects with the restraint chair facing tangent to the plane of spin (the orientation for Y0 and Y90 sequences) and three with the chair facing radially (the orientation for Y45 sequences). Their scores for these sequences did not differ significantly from their Yo performance. Table 1 lists the mean scores achieved by each group of three subjects in their perturbation-only sequences. Figure 9 shows initial subject performance dynamics as a function of the six model test sequences. Intervals are in real time (seconds) for means of the entire sample. Although these mean values show a progressive increase in time from starting signal to first correct RATER re- sponse, an analysis of variance of the data did not show these changes to be significant. TABLE I.—RATER Score Versus Perturbation Only Sequence X N 208 3 207 3 212 3 207 3 Data percentage presented in figure 9 was reduced from the polygraph stripchart, the VOG tapes, and the EMC films. In comparing the VOG and EMC eye-motion loci for various trials, it was found that they gave identical measure- ments of eye movement and could be used inter- changeably from that standpoint. Figures 10 and 11 compare oculograms for the same test trial plotted, respectively, from EMC and VOG film. The numbers in both S 5 ..« E » B 5 ...- DJ 0 TIME I How E*f> or HEAD TURN TO EYE HXATKM WITHIN I' of VBUAL ANGLE - AT o.I CM INDICATE tNTERFUUiAJi ANGLE* HCTWFT.N M'HJECTS HLAit IT RN TTBT WW> EKCE FIGURE 9. —Subject response versus test sequence. OCULOGHAM FOR SUBJECT JGM PLOTTED FROM EMC FILM FOR TRIAL NO. 1 OF YM (RP) A12 Att Notes i—mean RATER score for 3 subjects for ISO-second sequence N= sample number P« = subject facing radially P,i = subject facing tangentially TURN DIRECTION 38,39 22 (HEAD STABLE)-- NUMBERS ARE EMC FRAMES FROM START SIGNAL CAMERA SPEED - 8 FRAMES/SECOND 21 "iio O GAZE POSITION WITH HEAD STABLE (TURN COMPLETED). A GAZE POSITION BEFORE TURN IS COMPLETED. • GAZE POSITIONS CORRECTED RELATIVE TO FIELD OF REGARD. ~~] REFERENCE MARK ON FIELD OF REGARD. A19 A18 "116 "115 FIGURE 10. — Eye-motion camera oculogram for subject JGM.

318 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION 12 \ — 3S Q39 "» 22 - - -i — 1 1 1 — \ - 1 "TkJ | > — I 34 \ QL 1 "\l 34 T — I 1 I -1 9- T^K4.^I\ " \m M 1 1 - - -4- ^^ 0 t - I 1 " 1 1 !. iX ' 10 T (L » - \ 40,41 - *>23\ s-T j i ~-24 - 1 S3.43I,; "T ,.r-f - ^tojfj 499, i! P'|45 Ml i • V \ 10' OF VISUAL ANGLE ^T. Tf * ^v .35 ' ;l II 1 Q30,3l| - ^-.J.25,28 . H 29 / \ / \ -i 26" ^20 2' OF VISUAL , ANGLE 30,31 2928 9.19-1/2 1» I I I | 1 I I I --®T\13 10' OF 19' 013-1/2 1 1 T— 1 1 1 1 — T ' 1 1 1 1 '1 11 J-l - t 1 ! ! I i 1 S VISUAL ANGLE \\ 2' OF VISUAL ANGLE 27 26 i A i 1 1 I l . 1 * 1 1 1 T 1 I J i i I 1 1 i i -. i I I I i I 1 1 I 1 i i ' I i PLOT ON RIGHT REPRESENTS ENLARGEMENT OF CENTER SQUAR ' (OUTLINED BY DASHED LINE) IN LEFT-HAND PLOT. OCULOGRAM FOR SUBJECT JGM PLOTTED FROM VOG FILM FOR TRIAL NO. 1 OF Y (RP) i - NUMBERS ARE EMC FRAMES FROM START SIGNAL. EMC SPEED = 8 FRAMES/SECOND. VOG KINESCOPE SPEED - 24 FRAMES/SECON AND ALLOWS INTERPOLATION BETWEEN FULL EMC FRAME POSITIONS. 1 1 ,tii6-i 1 1 i I- I ALL PLOTTED CIRCLES REPRESENT GAZE POSITIONS RELATIVE TO THE SKULL. FIGURE 11. — Vector oculogramfor subject JGM. oculograms refer to the EMC frame numbers from the cue-light start signal. The EMC was run at eight frames per second. The VOG tape originally records 60 fields (frames per second), and when kinescoped, provides a permanent record of 24 frames per second. A comparison of figures 10 and 11 indicates the major difference between the EMC and VOG data-the EMC fixes gaze position in the field of regard independent of head position, whereas the VOG indicates only gaze position relative to the head. In figure 11, prior to frame 22, when the head becomes stable at the end of head turn, gaze positions 18 to 21 could be repositioned relative to the chang- ing field of regard by correcting relative to a reference mark in that field. Positions 10 to 13 could not be so corrected as the reference mark was not yet in view on film. Subsequent to point 49, all gaze positions lay in the area circum- scribed by 45 to 49, and therefore were not plotted. Comparing the two plots, it would seem that for tests involving head immobility, the VOG provides an adequate substitute for the EMC. Discussion This study was the third experimental effort performed by this laboratory involving F-axis head turns and their effect on perceptual-motor performance. The first study used the logical inference tester (LOGIT) as the performance tester, whereas the second study used the RATER as in this study. The first two studies exposed the subjects to Z-axis (side-to-side) head-turn sequences, whereas this study exposed them to y-axis head-turn sequences with per- turbation added to rotation. Apart from that, the formats were quite similar, and the F-axis sequences with rotation and no perturbation are capable of comparison.

EFFECT OF INSTABILITY DURING ROTATION 319 By comparison, the results of this present study are consistent with the results of the first two studies. The previous experiment (ref. 16) demonstrated a 6-percent degradation in LOGIT performance for Y45 and a 12-percent degradation for Y,to- The second study demonstrated a 10- percent degradation in RATER performance for Y45 and a 25-percent degradation for ¥90. This present study demonstrated a 2-percent degrada- tion in RATER performance for ¥45 and a 16- percent degradation for ¥90. The greater degradation witnessed in the first RATER study, in part, must be due to the test regimen including Z-axis head-turn sequences, which have been shown to be significantly more degrading than K-axis head turns of comparable interplanar angle. In the previous RATER study, the Z^ and Z90 sequences produced 25 percent and 37 percent performance degradation, respectively. In that study, as in the present one, permutations of the various stress modalities were balanced, but this only serves to balance the cumulative stress effects among all sequences —not nulling out such effects. Again, comparing these three studies, it is seen that the head-turn times are quite similar in magnitude as well as in proportion. Although the eye-fixation times in the second study are significantly longer than those seen in the pres- ent study, this would be anticipated in view of the substantially greater degradation in perform- ance and, in part, must be due to the exposure to Z-axis head turns as part of the test regimen. In the present study, although perturbation did not produce a significant change in perform- ance, consideration of the mean performance values as a function of perturbation permits some suggestion of possible effect. Table 1 indicated that perturbation alone did not have a degrading effect on performance; therefore, the relative decrement in mean performance values for Yo (RP) and Y45 (RP) compared to the comparable (R) sequences could result from labyrinthine cross-coupled accelerations due to the passive sinusoidal tilting of the subject relative to the MRSSS spin plane. As it is a sinusoidal tilting, it would be anticipated that the effect might be marginal since the resultant stimulus to the cupula-endolymph system would reverse in direction every 5 seconds. When considering the more traumatic Y«o orientation, however, it appears that the minor degradation due to perturbation no longer manifests itself within the context of the major degradation resulting from cross-coupling due to active head turns. This appears to be espe- cially true in the latter trials of the sequence when the cumulation of stress is causing the most significant performance decay. In conclusion, it has been demonstrated that perturbation imposed in this experiment does not have significant effect on perceptual-motor per- formance, either with or without simultaneous rotation. It has also been demonstrated that the VOG is a satisfactory substitute for the EMC in recording two-dimensional eye movements relative to the skull. It is seen, also, that the test results are consistent with previous studies performed of similar format. OCULOGYRAL ILLUSION EXTINCTION IN STABLE AND PERTURBATING ENVIRONMENTS Experiment 3 It has been noted in this and other laboratories (refs. 9, 18, and 19) that a suppression (adapta- tion or habituation) of ocular, perceptual, and somatic responses to Coriolis vestibular stimuli occurs in a rapid and predictable manner when repeated and that the suppression observed is closely specific for the stimulation being used. The transfer of the suppression to other forms of vestibular stimuli, including the unpracticed quadrants of identical Coriolis vestibular stimuli, has not been significant. During combined rotation and perturbation exposures of subjects and examiners in the Convair MRSSS, impres- sions of altered vestibular suppression rates have been consistently reported that indicate the perturbating environment is more easily tolerated. Experiment 3 provides statistical comparison of rate and transference of vestibular suppression of the OGI resulting from cross- coupled angular acceleration as a function of the presence or absence of perturbation. Perturba- tion presents a unique stimulus modality in that it provides passive rotation of the subject's

320 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION labyrinths relative to the spin plane of the simulator but in a form that involves a minimum of conscious involvement. It may thus provide some evidence as to what aspects of the stimulus framework, from the consciousness standpoint, are of primary importance in- establishing end- organ signal suppression. The purpose of experiment 3 was to compare the vestibular suppression due to head rotations in a rotating environment with that due to head rotations in an environment that is simultaneously rotating and perturbating. Method To maximize the use of available centrifuge time and to provide information on the rate of response extinction, experiment 3 was divided into parts A and B; the subjects involved in part A were reexposed 1 month later in part B. Ex- periment 3 consisted of exposing subjects to 4 hours of rotation, with or without perturbation. Prior to and subsequent to the dynamic ex- posures, subjects received vestibular caloric tests—first one ear and then the other. During the dynamic exposures in the simulator, the sub- jects made ,Y-axis head turns (toward one shoulder only) at 15-minute intervals. Responses to both rotational and caloric vestibular stimuli were measured by durations of resultant OGI. As it is of great importance that responses both to the rotational and caloric stimuli be of sufficient initial magnitude to permit measurable decre- ment or suppression of response, subjects and stimuli were chosen to insure such adequacy. Head turns were 45° about the .Y-axis at maximum rate, with the simulator angular velocity at 8 rpm (6 rpm was found to be insufficient for con- sistent response). Caloric stimulation consisted of 50 cc of 25° C water delivered at a rate of 1 cc/sec against the posterior superior wall of the external auditory canal, with the head positioned to place the lateral canals in a vertical position. Only subjects with normal audiometric profiles for the 20- to 30-age span and with active caloric responses were used. A leadtime of 2 days was provided for certification of subjects for use in the study. The bite-bar arrangement that was used to limit the extent of head turn and to insure that the motion was made in a consistent FIGURE 12. — Bite-bar restraint used to limit head-turn motion. plane is shown in figure 12. After all had received their caloric stimulations, subjects were seated facing the leading bulkhead of the MRSSS (upon which the OGI target light is affixed), and the MRSSS was spun up to 8 rpm. Subjects were given a stopwatch and were positioned with the restraining bite bar clenched between their teeth. The ambient illumination was extinguished prior to each head turn (the OGI target consisted of a 3-inch wire cube painted with fluorescent paint and illuminated by a 40-watt ultraviolet (UV) fixture: no light was visible). The onboard examiner commanded, "One, two, three, TURN." At the command "TURN," the subject turned his head as rapidly as possible to the right. The bite-bar restraint maintained the turn in the frontal plane and prevented the turn from passing beyond the designated 49°. The subject started his stop- watch at the same time he began his head turn and stopped it when the resultant OGI appeared to have terminated. He did his best to remember the direction (clock-hour numbers) and magni- tude (units equal to target light side dimension) of the illusory movement of the target for sub- sequent recording on paper. This requirement for memory of illusion was meant to provide a mental task to maintain central arousal of the subject for consistency of that aspect of response threshold. Subjects were instructed to spend an estimated

EFFECT OF INSTABILITY DURING ROTATION 321 TABLE 2.— Schedule for Experiment 3 Caloric testing, dynamic exposure Part A: R ' Part B: R + P2 Part A: R+P Part B: R (1) Left ear Ss 1 to 4 Ss 5 to 8 (2) Right ear (1) Right ear Ss 9 to 12 Ss 13 to 16 3 (2) Left ear 1 R = rotation (8 rpm). 2P = perturbation (±3°, 0.1 Hz). 'Ss 17 to 20 provided backup for data loss. Estimated subject time: 0900-1000: 1 hour, predynamic; 1000-1400: 4 hours, dynamic; 1400-1500: 1 hour, postdynamic. 1 minute (at a consistent turning rate) to return slowly back to center stop. Two minutes after the initial TURN signal, the room lights were turned on, the times recorded, and the stopwatch and polygraph connects transferred to subjects 3 and 4. The testing procedure followed for subjects 1 and 2 was repeated for subjects 3 and 4. followed by a repetition of the complete testing cycle for all four subjects every 15 minutes of the 4 hours of dynamic exposure. Just prior to "spindown," a response to the unexperienced head-turn direction was made. At the end of the 4 hours of dynamic exposure, the subjects received caloric stimulations as during the prespin portion of the test, after which they were released. Hah0 of the samples (eight subjects) were ex- posed to rotation plus perturbation (±3° at 0.1 Hz) during part A and the other half during part B. Half the subjects received caloric irrigations in their left ear first; the other half, in their right ear first. This initial order was maintained throughout both parts of the experiment (A and B). Subject numbers were assigned in groups of fours on a random basis as shown in table 2. Results Twenty subjects were exposed to the first part of the test program. The first four were exposed to 6 rpm, and it was found that the magnitude of OGI produced on the first head turn was not great enough to satisfactorily demon- strate progressive habituation during the 4 hours of exposure. The remaining 16 subjects were ex- posed to 8 rpm; at that velocity, a good OGI response was observed. One subject had a vegetative response as a result of repeated head tilts, and his testing was suspended to avoid possible nausea which would have required the abort of all four test subjects. Eleven of the remaining subjects returned for the second part of testing. Figure 13 shows the results for all subjects tested. From the slope of the regres- sion curves, there is no apparent difference in the rate of adaptation. There is a difference in % 100 90 120 150 MINUTES AT 8 RPM FIGURE l3. — Oculogyral illusion extinction (all subjects).

322 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION O ±3'. 0.1 CFS n • 11 STABLE n - 11 120 150 MINUTES AT 9 RPM FIGURE 1b.—Oculogyral illusion extinction (subjects com- pleting tests). extent of habituation: however, this could be due to the sample size difference. Figure 14 is for the 11 subjects who were tested in both parts A and B. The difference in habituation achieved in the two groups has an even greater significance (points are the mean ± 1 standard error) when the subjects who did not return for the second test are deleted, but the rate of ex- tinction appears the same. Discussion It was hypothesized that habituation during rotation with perturbation would take place faster than in the stable situation. This was based on the supposition that the greater the inter- action of a subject with the cross-coupled accel- erations, the faster the various stimuli would be suppressed centrally. It is clear that the data show a reversed situation. The OGI was ex- tinguished at about the same rate but to a sig- nificantly lesser degree than without perturba- tion. A possible explanation lies in the dynamics involved. The subjects were arranged along a radial line in the MRSSS cabin. The A"-axis head turns were made in a plane perpendicular to the plane of rotation that passed through the center of rotation. The ±3° perturbation of the cabin at 0.1 Hz also took place in this same plane, and the stimuli resulting from the subjects' head turn during perturbation was then de- pendent on the cabin motion at the time of head turn; head turns resulted in a different cross- coupling for each experience and, therefore, could have decreased the rate of habituation. Another important factor is that the perturbation kept a continuous stimulus applied to the semi- circular canals: the cupulae either were being stimulated or recovering and, therefore, could be expected to have a decreased sensitivity— being in a partial refractory state. The raising of the threshold due to a continuous low-level excitation would also explain the feeling ex- pressed by subjects and examiners that rotation seemed to be less disorienting with perturbation. No explanation can be offered for the simulta- neous dip in both curves at 90 minutes of testing. This represents six individual test runs, and the small distribution of data points would indicate a consistent response but no cause has been found to which it could be attributed. Table 3 lists the duration of OGI resulting from caloric stimulation before and after 4 hours of rotation at 8 rpm with and without perturbation. No difference appears to exist between the stable or perturbating conditions, but a very consistent decrease is observed in the time of illusion when prerotation and postrotation values are compared. Six control subjects, tested without rotation but with a 4-hour interval between irrigations, did not show this decrease in OGI response duration. Subjects who did not have strong illusions result- ing from irrigation of both ears prior to testing were not retested after rotation, and caloric data from subjects who did not respond from stimula- tion in both ears after rotation were also discarded. The consistent decrease in response of the remaining subjects could reflect a lack of recovery from the first caloric stimulation. There was a period of 5 hours or more between irriga- tions which should have allowed resensitization. An attractive interpretation of the data would be that they indicate a decrease in receptor sensi- tivity as a result of transference. Such an inter-

EFFECT OF INSTABILITY DURING ROTATION 323 TABLE 3. — Caloric Illusion Duration (Seconds) Left ear Right ear Prerotation Postrotation A Prerotation Postrotation A 8 rpm (steady) 115 61 -54 51 59 8 125 41 -84 136 29 -107 106 74 -32 121 58 -63 206 84 -122 115 83 -32 144 109 -33 139 64 -75 142 46 -96 100 52 -48 82 99 17 85 99 14 69 45 -14 94 36 -58 8 rpm with ±3° perturbation at O.I H/. 114 97 -17 114 65 -49 140 103 -37 156 88 -68 110 147 37 106 110 4 200 42 -138 62 35 -27 109 60 -49 181 108 -73 176 44 032 141 122 -19 49 36 -13 52 34 -18 61 21 140 94 43 -51 Control. 0 rpm (steady) (4-hr repeat) 182 191 9 170 173 3 59 50 -9 83 112 29 189 198 9 124 116 -8 41 54 13 58 60 2 174 195 21 156 165 9 166 186 20 119 114 25 pretation would give function to the efferent nerves of the labyrinth, but other workers have not been able to demonstrate such transference (refs. 16, 19, and 20). Table 4 presents the OGI durations; for example, what lack of extinction occurred when repeated ,Y-axis head turns were made, 45° to the right shoulder, every 15 minutes. The OGI time was recorded for right and left motions at the beginning and again at the end of the test period. This procedure has been used to dem- onstrate the specificity of vestibular habituation to a given motion and the lack of transfer of habituation to nonstimulated receptors. Using nystagmus as a criterion in addition to the subjective OGI response to head turns, Guedry et al. (refs. 18 and 19) have reported on the lack of illusion transfer that was observed when their subjects underwent habituation by repeated right head turns in a stable environment at 7.5 rpm. At the end of their test period, a single head turn to the unpracticed left produced an OGI response in 50 of 64 trials, whereas only 12 of 64 trials (sum of turn and return) had any response at all on a turn to the right. Data from the present study were analyzed in a similar manner; however, because very few subjects had a zero OGI response after 4 hours of rotation, the duration of the illusion was considered the im- portant criterion of habituation. The results reported here on duration of OGI are not nearly so conclusive as the cited study on nystagmus, although the distribution of responses is quite similar.

324 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION TABLE 4. — Transfer of Habituation Illusion Duration (Seconds) Repeated head turns to the right (conditioned) 8 rpm (stable) 8 rpm (±3° at 0.1 Hzl Prerotation Postrotation Prerotation Postrotation C,.-R R-C,. C,.-R R-C, Ct-R R-C,. C,.-R R-C,. X 12.2 8.0 3.5 3.3 12.4 6.7 4.4 5.5 (r 2.7 10 2.8 13 3.1 14 2.7 13 5.4 13 3.9 2.9 3.4 n ' 14 14 14 Total 20.2 66 6.8 20.9 52 9.9 Single head turn to the left (nonconditioned) 8 rpm (stable) 8 rpm (±3° at 0.1 Hz) Prerotation Postrotation Prerotation Postrotation C,.-L L-CL C,-L L-CL C,.-L L-CL C,,-L L-CL £ 6.9 7.4 4.2 4.0 5.2 6.7 5.6 4.7 3.2 13 3.8 13 3.1 14 2.5 14 5.8 3.5 14 3.6 2.9 14 n ' 14 14 Total 14.3 8.2 11.9 10.3 Reduction, percent 41.3 13.5 Number of responses Extent of OGI response declines,2 percent 8 rpm (stable) 8 rpm (±3° at 0.1 Hz) Conditioned Nonconditioned Conditioned Nonconditioned 100-80 12 6 5 6 5 5 1 80-60 3 5 3 11 4 5 2 16 60-40 j 6 5 7 40-20 1 2 20-0' J 1 Durations exceeding 2cr from x were deleted. 2 Total of turn and return. The transfer of habituation found in the stable rotation of the present study does not agree with that of the other authors (refs. 18, 20, and 21) and can only be explained by assuming that the un- practiced head turn was not totally isolated from the environment and that some stimulation was somehow being perceived by the supposedly dormant receptors. These stimulations must have been small as the subjects used a bite bar 3 Includes increases. to limit head motion. There is, however, a de- cided lack of transfer to the unpracticed side, as well as a reduced magnitude achieved by habitua- tion in the perturbated situation. This reduction, in spite of the apparent stimulation, fits the pro- posed hypothesis that perturbation raises the cross-coupled threshold of the semicircular canals. The continuous perturbation, which was in the plane of the head turns, could make the

EFFECT OF INSTABILITY DURING ROTATION 325 system nonresponsive to small stimuli that pro- vided conditioning to the nonpracticed side dur- ing stable rotation. RELEVANCE TO SPACECRAFT DESIGN The study reported in this paper is the first attempt, to the knowledge of the author, to em- pirically assess the problem of instability for a spacecraft employing artificial gravity by cen- trifugation. The conclusions are limited by the sinusoidal nature of the perturbations which may not fully represent the space situation. Sinus- oidal disturbances are sure to exist, but imposed upon these will be random motions. The sinus- oidal pattern is easily anticipated by the subject — consciously or unconsciously. As such, habitua- tion to that motion is probably facilitated. The sinusoidal perturbations used in these studies with a range of vehicle angular velocity were as severe as have been predicted by the vehicle dynamicists. The three types of tests used represent a di- vergent approach that compared the pure passive motion of the subject to that of active head mo- tions. In all cases, the anticipated decrease in performance did not materialize. On hindsight, it can be rationalized that this should have been thought of as a possible result of the constant semicircular canal stimulation due to cross- coupled acceleration produced by the product of vehicle rotation and angular perturbation. The observed insensitivity, however, may be due to more than a simple raising of the threshold of this organ. Cross-coupled acceleration thresholds for the semicircular canal cannot be defined in the same manner as is done in pure angular acceleration. The stimulus is continually varying in vector direction even though the magnitude of the prod- uct of angular velocities is constant. During a head turn, a point in the labyrinth, as it crosses the vehicle plane of spin, has a stimulus vector reduced to zero because the cosine becomes zero in the formula a = o>Xcu cos 9. As that point leaves the plane of spin, the vector quantity increases so there is a continually changing ac- celeration imposed on each of the six canals. The time that each stimulus is applied may be very short —below that of the time constant re- quired for stimulation. In 1960, Carl Clark (personal communications) estimated the thresh- old on himself for cross-coupled illusions on a centrifuge rotating at 10 rpm to be 3.6°/sec2 due to active head turns. Recently, Newsom (ref. 22) has made a more extensive study, passively tilting immersed subjects at various centrifuge speeds. For turns of low angular velocities (6° to 10°/sec), the illusion threshold is very close to 3.6°/sec2, but it varies with the position of the head-turn angle from the plane of spin. How- ever, the same threshold is not reached if the centrifuge speed is reduced and the angular velocity of the head is increased because this decreases the time of stimulus for a given angle of turn, and cross-coupled accelerations of a much higher magnitude are required to reach the threshold. In the same study, exposure of subjects to Coriolis accelerations of 6°/sec2 to 90°/sec2 did not cause nausea or decrease per- formance. In figure 15, an attempt is. made to illustrate the dynamics involved. The MRSSS I-OINT ti( D1 HIM! A TOC POINTS A AND C I'OINT II, DI'RINr. C TO A iIPETAI VECT0f1 CCELERATIIiN VrCTOR DI'E TO MOVEMENT O^ POINT A TO D. MAGXITI I>E DEPENDENT ON I REyI'ENCY AND AMPLITUDE OTATION HATE Of VEHICLE AXIS Tonoi-E TO S(-RJECT NGI I-AR DISPI-ACEMENT DfE To VERTICAL ALIGNMENT MPLITI'DE OF VEHICLE WOBBLE DISPLACEMENT OF VERTICAL (MRSSS SIMULATION) FIGURE 15. — Elements of vestibular disturbance due to vehicle wobble.

326 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION perturbation simulates the displacement of the vertical due to vehicle wobble. It does not properly simulate the cross-coupling, a — u>\ X o>2. where the subject is torqued about his Z-axis (o>2) during rotation about a displaced X- or y-axis (o>]). A second cross-coupling,a=o>i Xo>3, results from vehicle rotation and the movement of the head to aline the body with the changing angle 6. In the MRSSS, this reproduced but is exaggerated by not having the subject's Z-axis close to the plane of spin. The vehicle perturbation of ±3° used in this study is in excess of the mean effective angle through which the man will be realined in roto- gravic space stations. The man displacement is due to the acceleration-vector addition to the centrifugal vector which will be off the normal to the floor, except when the vehicle is at the peak of excursion. The magnitude of this dis- placement of the vertical will be determined by the cycle duration and radius of the vehicle, but it will never be that of total vehicle wobble. This is germane because the simulator used in the reported studies used the maximum angle predicted for vehicle excursion (angle ip in the diagram) of ±3° for angle ij). In addition, the cross-coupling is a function of the angle from the spin plane. In space, as mentioned before, it will vary a few degrees around zero and, there- fore, the cosine factor will be very small. The simulator perturbation was ±3°; when the resultant is used to establish the normal to the floor, that normal is 87.5° at 6 rpm, 68.2° at 8 rpm, and 45° at 12.2 rpm to the spin plane. This means the value by which the cross-product of angular accelerations is being multiplied varies from close to 1 to 0.7 instead of being close to zero as in the space situation. In essence, this means the cross-coupling effect of perturbation used in this study far exceeded that to which the space crews would be exposed. Observations in the study were limited to a maximum of 4 hours, and the subjects did not move around the simulator. The results, there- fore, indicate nothing about how perturbation affects habitability in the rotating environment, which should be the next factor to be investigated. CONCLUDING REMARKS From the results of these few tests, it appears that the sinusoidal perturbation anticipated in a rotogravity space vehicle will not unduly complicate specific task performance in a restrained subject. It is important to extend these observations to a test lasting at least 4 days to determine if the perturbation is equally innocuous for longer periods, for then considerable instability could be allowed in the rotogravity vehicle. REFERENCES 1. LORET, B. J.: Optimization of Space Vehicle Design With Respect to Artificial Gravity. Aerospace Med., vol. 34, 1963, pp. 430-441. 2. DOLE, S. H.: Design Criteria for Rotating Space Vehicles. Rand Research Memorandum RH-2668. Rand Corp., Santa Monica, Calif., Oct. 1960. 3. KURZHALS, P. R.; ET AL.: Space-Station Dynamics and Control. A Report on the Research and Technological Problems of Manned Rotating Spacecraft. Langley Research Center, NASA TN D-1504, 1962. 4. LARSON, C. A.: Space Station Design Parameter Effects on Artificial Gravity Field. AIAA J., Aug. 1964, pp. 1454-1455. 5. GRAYBIEL, A.; CLARK, B.: AND ZARRIELLO, J. J.: Observa- tions on Human Subjects Living in a "Slow Rotation Room" for Periods of Two Days. Arch. Neurol.. vol. 3, 1960. pp. 55-73. 6. MAYO, A. F.: Bioengineering and Bioinstrumentation. Bioastronautics, K. E. Schaefer, ed.. The Macmillan Co., 1964, pp. 227-273. 7. NEWSOM, B. D.; AND BRADY, J. F.: Observations on Sub- jects Exposed to Prolonged Rotation in a Space Station Simulator. The Role of the Vestibular Organs in the Exploration of Space, NASA SP-77, 1965, pp. 279-292. 8. NEWSOM, B. D.: BRADY, J. F.; SHAFER, W. A.: AND FRENCH, R. S.: Adaptation to Prolonged Exposures in the Revolving Space Station Simulator. Aerospace Med., vol. 37, 1966, pp. 778-783. 9. BRADY. J. F.: AND NEWSOM, B. D.: Large Excursion Rotary Tracking of Target and Target Light in a Space Station Simulator Revolving at 7.5, 10.0 and 12.0 rpm. Aerospace Med., vol. 36, 1965, pp. 333-342. 10. O'LAUGHLIN, T. w'.; BRADY, J. F.: AND NEWSOM, B. D.: Reach Effectiveness in a Rotating Environment. Aero- space Med., vol. 39, 1968, pp. 505-508. 11. NEWSOM, B. D.; AND BRADY, J. F.: XIV International

EFFECT OF INSTABILITY DURING ROTATION 327 Congress Aviation and Space Medicine. Prague, Czechoslovakia, 1966. 12. CORMACK, A.; AND CoucHMAN, C. C.: Considerations of Crew Comfort in Relation to the Dynamics of Rotating Space Stations. First AIAA Annual Meeting, Wash- ington, D.C., June 29 to July 2, 1964. 13. PARKER, J. F.; REILLY, R. E.; DILLON, R. F.; ANDREWS, T. G.; AND FLEISHMAN, E. A.: Development of Tests for Measurement of Primary Perceptual-Motor Per- formance. NASA CR-335 (contract NAS 9-2542), Dec. 1965. 14. NEWSOM, B. D.; BRADY, J. F.; AND GOBLE, G. S.: Equi- librium and Walking Changes Observed at 5, 7{, 10, and 12 rpm in the Revolving Space Station Simulator. Aerospace Med., vol. 36,1965, pp. 322-326. 15. NEWSOM, B. D.; BRADY, J. F.; AND O'LAUGHLIN, T. W.: Optokinetic Reflex Responses lo Cross-Coupled Gyro- scopic Stimuli. Final Report NAS 9-5232, 1968. 16. NEWSOM, B. D.; AND BRADY, J. F.: Comparison of Per- formances Involving Head Rotations about Y and Z Cranial Axes in a Revolving Space Station. Aerospace Med., vol. 37, 1966, pp. 1152-1157. 17. NEWSOM, B. D.; AND BRADY, J. F.: Display Monitoring in a Rotating Environment. Third Symposium on the Role of the Vestibular Organs in Space Exploration, NASA SP-152, 1968, pp. 37-46. 18. GUEDRY, F. E.; COLLINS, W. E.; AND GRAYBIEL, A.: Ves- tibular Habituation During Repetitive Complex Stimu- lation: A Study of Transfer Effects. J. Appl. Physiol., vol. 19, 1964, pp. 1005-1115. 19. GUEDRY, F. E.; GRAYBIEL, A.; AND COLLINS, W. E.: Re- duction of Nystagmus and Disorientation in Human Subjects. Aerospace Med., vol. 33, 1962, pp. 1356- 1360. 20. COLLINS, W. E.: Subjective Responses and Nystagmus Following Repeated Unilateral Caloric Simulation. Ann. Otol., vol. 74,1965, pp. 1034-1055. 21. MERTENS, R. A.; AND COLLINS, W. E.: Unilateral Caloric Habituation of Nystagmus in the Cat. Effects on Rotational and Bilateral Caloric Responses. Ada Oto-Laryngol., vol. 64, 1967, pp. 281-297. 22. NEWSOM, B. D.: Feasibility Study of a Centrifuge Ex- periment for the Apollo Applications Program. Vol. IV, NASA CR-66684, 1968.

SESSION X Chairman: LAWRENCE F. DIETLEIN Manned Spacecraft Center, NASA