Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Improved feasibility of astronaut short-radius artificial gravity through a 50-day incremental, personalized, vestibular acclimation protocol

Improved feasibility of astronaut short-radius artificial gravity through a 50-day incremental,... www.nature.com/npjmgrav ARTICLE OPEN Improved feasibility of astronaut short-radius artificial gravity through a 50-day incremental, personalized, vestibular acclimation protocol 1 1 Kathrine N. Bretl and Torin K. Clark The “Coriolis” cross-coupled (CC) illusion has historically limited the tolerability of utilizing fast-spin rate, short-radius centrifugation for in-flight artificial gravity. Previous research confirms that humans acclimate to the CC illusion over 10 daily sessions, though the efficacy of additional training is unknown. We investigated human acclimation to the CC illusion over up to 50 daily sessions of personalized, incremental training. During each 25-min session, subjects spun in yaw and performed roll head tilts approximately every 30 s, reporting the presence or absence of the illusion while rating motion sickness every 5 min. Illusion intensity was modulated by altering spin rate based upon subject response, such that the administered stimulus remained near each individual’s instantaneous illusion threshold. Every subject (n = 11) continued to acclimate linearly to the CC illusion during the investigation. Subjects acclimated at an average rate of 1.17 RPM per session (95% CI: 0.63–1.71 RPM per session), with the average tolerable spin rate increasing from 1.4 to 26.2 RPM, corresponding to a reduction in required centrifuge radius from 456.6 to 1.3 m (to produce loading of 1 g at the feet). Subjects reported no more than slight motion sickness throughout their training (mean: 0.92/20, 95% CI: 0.35–1.49/20). We applied survival analysis to determine the probability of individuals reaching various spin rates over a number of training days, providing a tolerability trade parameter for centrifuge design. Results indicate that acclimation to a given, operationally relevant spin rate may be feasible for all subjects if given a sufficient training duration. npj Microgravity (2020) 6:22 ; https://doi.org/10.1038/s41526-020-00112-w INTRODUCTION Previous research has shown that humans are capable of acclimating to the CC illusion, as subjects reported reduced For decades, artificial gravity (AG) has been proposed as the most 8–11 intensity of the illusion after two or three sessions of exposure . comprehensive spaceflight countermeasure, holding the potential These early acclimation studies exposed subjects to the CC illusion to concurrently protect multiple physiological systems from in- at spin rates of 23 RPM or higher. This very provocative stimulus flight, microgravity-induced deconditioning. Several conceptual resulted in substantial motion sickness in subjects; notably, it designs have been proposed, although all face the challenge of caused ~25–35% of subjects to dropout of each study due to budgetary and engineering constraints. The most feasible motion sickness, despite the fact that subjects who were highly approach to AG in the near term is thought to be short-radius, 1 susceptible to motion sickness were often excluded from intermittent centrifugation , though additional challenges surface 8,10,11 participating . with this design. Of these, it appears the human body can 2 3,4 Subsequent studies investigated the possibility of benign acclimate quite quickly to Coriolis forces and gravity gradients , exposure to the CC illusion. It was found that an incremental but acclimation to the disorienting “Coriolis” cross-coupled (CC) protocol provides a less provocative way to expose subjects by illusion is less expeditious and thorough (we note that the CC reducing motion sickness via incremental spin rate exposure (i.e., illusion is not the result of a conventional and substantial Coriolis increasing spin rates over time rather than directly exposing effect and therefore have placed it in quotations to distinguish subjects to a high CC illusion stimulus/fast spin rate). One from traditional Coriolis forces). It remains unclear whether all incremental CC illusion acclimation approach exposed subjects individuals can tolerate and/or acclimate to the moderately high to spin rates of 14 RPM on day 1, 23 RPM on day 2, and 30 RPM on spin rates (e.g., ~15+ rotations per minute (RPM)) that would be day 3 . This study showed that the incremental steps improved required for short-radius centrifugation. subjects’ tolerance of the acclimation protocol, yet it was still too The CC illusion is a provocative tilting or tumbling sensation aggressive for some of the subjects, as 14% (4/28) of the subjects experienced by a subject in a constantly rotating field after he/she dropped out due to motion sickness. Another variant of performs a head tilt out of the plane of constant rotation. The incremental CC illusion acclimation showed that personalized, illusion is highly disorienting and typically leads to motion incremental acclimation over 2 days of training resulted in sickness . The intensity of the illusion is proportional to the spin essentially no cases of motion sickness, and therefore, a 0% rate at which the subject is exposed, the magnitude of the head subject dropout rate (0/8 subjects) . 6,7 tilt, and the velocity of the tilt . Higher spin rates are required for We have recently extended previous efforts to investigate effective short-radius centrifuge designs; thus, the CC illusion personalized, incremental acclimation to the CC illusion over ten becomes increasingly relevant as the potential for disorientation consecutive weekdays . In short, subjects were seated upright and motion sickness becomes more prominent with a faster- and spun in yaw about an Earth-vertical axis. After an introduction spinning centrifuge. to the CC stimulus, each subject began the training protocol University of Colorado Boulder, 3775 Discovery Drive, Boulder, CO 80309, USA. email: kathrine.bretl@colorado.edu Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; K.N. Bretl and T.K. Clark spinning at 1 RPM. While spinning at a constant rate, subjects RESULTS performed roll head tilts (40° right ear down and back to upright) Extended acclimation protocol resulted in continued acclimation approximately every 30 s. After each head tilt, subjects reported if with no evidence of plateau in all subjects they experienced the CC illusion. If the subject did not report the Eight of our 11 subjects completed the investigation by illusion on two consecutive head tilts of one head tilt pair (head reaching one of our first two ending criteria (see “Methods”): tilt down and back upright), the spin rate was increased by 1 RPM; either a beginning threshold (i.e., fastest spin rate at which no otherwise, the spin rate was maintained. Ultimately, the spin rate illusion was experienced at the beginning of the session) of 25 was incremented only based upon subject response, and the RPM (n = 7) or completing a total of 50 testing sessions (n = 1). stimulus was always just barely noticeable (i.e., personalized and Notably, none of the subjects appeared to have reached a threshold based). The stimulus was effective in acclimating all plateau in ending threshold based on our ending criteria 10 subjects, while not being excessively strong to elicit substantial (Fig. 1). The final three subjects who enrolled in the study were motion sickness. Subjects acclimated from an average threshold included in analysis, as they each completed 15 or more testing (i.e., fastest spin rate at which no CC illusion was felt) of 1.8 RPM sessions. However, they did not complete the study protocol; all (range: 1–3 RPM) before any training to 17.7 RPM (range: 3–30 RPM) after 10, 25-min training sessions. Across all sessions and three chose to prematurely leave the study due to challenges subjects, an average motion sickness rating of only 1.06/20 was with scheduling (denoted with a black “X” in Fig. 1b). Including reported. This investigation confirmed that a personalized, all 11 subjects, the average number of training sessions incremental protocol is a highly effective, benign method of completed was 25.5 (range: 10–50 sessions). Subjects began acclimating individuals to the CC illusion. the investigation at an average pre-training threshold (i.e., the In addition to personalized acclimation, we have also investi- fastest spin rate at which no illusion was felt before any gated the efficacy of standardized (i.e., incremental, but non- acclimation training) of 1.4 RPM (range: 0–3 RPM, a threshold of personalized) acclimation and the ability of subjects to retain 0 RPM corresponds to reporting the illusion at 1 RPM). While gained acclimation after a period without training . These studies acclimating over a varying number of sessions, subjects have suggested that non-personalized acclimation is not as reached an average ending threshold (i.e., fastest spin rate effective or tolerable as personalized acclimation; however, that elicited no CC illusion at the conclusion of the session) of acclimation in both studies appears to be mostly retained for up 25.8 RPM (range: 10–38 RPM) on their final day of testing. This to 90 days of unmonitored activity. These previous studies sought resulted in a significant increase in CC illusion acclimation to investigate acclimation after an initial protocol of 10 daily threshold (paired t-test: t(10) = 8.95, diff = 24.4 RPM, Cohen’s sessions. Remarkably, it appeared acclimation continued linearly d = 3.73, p < 0.0005). Subjects’ acclimation can also be quanti- over those 10 days, suggesting further acclimation may be fied by their acclimation rate (i.e., the slope of a linear fittoeach possible. However, potential for acclimation beyond these 10 days subject’s ending threshold as a function of session number). is unknown. Specifically, it is of operational and scientific interest Average acclimation rate was found to be significantly non-zero to determine if acclimation continues with additional exposure, or across subjects (One-Sample t-test: t(10) = 4.82, mean = 1.17 if a plateau is reached in some or all individuals, after which no RPM persession,95% CI:0.63–1.71 RPM per session, p = further acclimation is possible. 0.0007). Further, all 11 subjects exhibited positive acclimation The objective of this investigation is to test the ability of rates (range: 0.26 RPM per session to 2.91 RPM per session), subjects to acclimate to extended CC illusion exposure (i.e., up to showing that all subjects—even those who increased their 50, 25-min sessions), using essentially the same personalized, incremental staircase described above, and inform operational threshold at a slower pace—were able to acclimate to the CC considerations of centrifuge design. illusion. All Acclimation Subjects | First 10 Sessions Extended Acclimation Subjects | All Sessions 40 1.1 40 1.1 10-Day Personalized Subject 6 Subject Complete Subject 1 Subject 7 Subject Drop-Out Subject 2 Subject 8 a b Subject 3 Subject 9 35 1.5 35 1.5 Subject 4 Subject 10 Subject 5 Subject 11 30 2.0 30 2.0 25 2.9 25 2.9 20 4.5 20 4.5 15 8.0 15 8.0 10 17.9 10 17.9 5 71.6 5 71.6 0 0 Pre-Training 1 2 3 4 5 6 7 8 9 10 Pre-Training 5 10 15 20 25 30 35 40 45 50 Session Number Session Number Fig. 1 Extended acclimation findings. CC illusion acclimation pretraining and ending thresholds for (a) 10-day and extended acclimation subjects over the first 10 sessions and (b) extended acclimation subjects over the entire duration of their involvement in the study. In panel a, the previous 10-day investigation subjects are shown in gray, while the current extended acclimation subjects are overlaid in color. In panel b, subjects who completed the study by reaching the investigation’s ending criteria are shown with a black circle at their final data point, while subjects who left the study due to scheduling challenges are denoted with a black “X”. npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; Ending Threshold (RPM) Required Diameter to Produce 1 G (m) Ending Threshold (RPM) Required Diameter to Produce 1 G (m) K.N. Bretl and T.K. Clark Extended acclimation investigation subjects with modified Extended Acclimation Motion Sickness Rating staircase protocol and previous 10-day subjects acclimated Subject Average MSR - Extended similarly over first 10 days 95% Confidence Interval Subject Maximum MSR - Extended These findings are similar to what has been previously reported Group Average MSR - Extended Group Average MSR - 10 Day over just 10 days , although additional confidence has been gained with the extended exposure. Figure 1a shows our original 10-day personalized acclimation subjects in gray with our extended acclimation subjects overlaid in color (for just the first 13 10 days of the investigation). This plot shows that the extended 12 acclimation subjects performed within the same bounds as our previous 10-day acclimation subjects, despite the slight modifica- tions in the staircase protocol with the inclusion of catch trials and the option to increment to slower spin rates if the CC illusion was reported on each head tilt for three consecutive head tilt pairs (see “Methods”). A two-sample t-test investigating potential differences between ending thresholds on day 10 of the two subject groups confirmed the two groups were comparable, with a non- significant result (Two-sample independent t-test: p = 0.46). Additionally, there was no evidence to suggest a difference in acclimation rate between the extended acclimation subjects and the previous 10-day acclimation subjects over the first 10 days of the investigation (Two-sample independent t-test: p = 0.27). All subjects acclimated to the CC illusion, but as displayed in Fig. 1, they acclimated at substantially different rates. As observed in 13–16 previous studies , there is a large degree of inter-individual Fig. 2 Extended acclimation motion sickness results. Subject differences in acclimation. As one measure in the present study, averages are denoted with the light blue bars, maximum reports across all sessions are shown with light blue asterisks, and the dark the acclimation rate averaged 1.17 RPM per session but varied blue bar shows the overall average across all extended duration substantially with a standard deviation of 0.81 RPM per session, acclimation subjects. The gray bar shows the 10-day personalized yielding a coefficient of variation of 0.69. subject group average motion sickness score for comparison. Error bars show 95% confidence intervals about the subject means. Subjects reported low motion sickness throughout study The incremental, personalized protocol was designed to limit were Low Catch Trials (50% slower spin rate, 35 trials which would motion sickness experienced during acclimation. The protocol was have been Low Catch Trials were not performed due to the rules outlined in “Methods”). Overall, subjects almost always responded effective in providing this benign acclimation to the CC illusion, as to the catch trials as we would have expected. Specifically, on only subjects reported generally low motion sickness scores through- 4.5% of the catch trials did the subject respond unexpectedly: out the investigation (Fig. 2). either that they felt the illusion on both head tilts for a Low Catch Most importantly, none of the 11 subjects dropped out of the Trial (2.4%) or that they did not feel the illusion during a High study due to excessive motion sickness, nor was any session terminated early due to subject discomfort or elevated motion Catch Trial (2.0%). This gives us confidence that the subjects were sickness reporting (i.e., 10/20 or higher on the utilized 0–20 scale, reporting reliably and were able to maintain consistent criteria of where 0 corresponds to no motion sickness and 20 corresponds to what qualifies as the presence of the CC illusion throughout the extreme nausea or vomiting). We note that one subject did report duration of the study. an 11/20 on one of the sessions, although this report was communicated in the final moments of the session, therefore the Survival analysis and expected population acclimation complete session was conducted. The average MSR across all With this study and those published previously, we have shown subjects and sessions was 0.92 (95% CI: 0.35–1.49) on the that subjects who undergo the developed training protocol commonly used 0–20 scale. This was not significantly different become more tolerant of the CC illusion over time, though than the average MSR of 1.06/20 (95% CI: 0.37–1.76) from the subjects acclimate at varying rates. To use this research as a tool 10 subjects who completed the previous 10-day personalized for centrifuge design, we have conducted an analysis to estimate protocol (Mann–Whitney–Wilcoxon test, p = 0.78). the probability that individuals will acclimate to a specific spin rate We observed no trend in MSR reporting across sessions or over a number of training days. Since we found no statistical reports within a session; however, we did find a statistically differences between the data collected during the previous 10-day significant correlation between subjects’ average reported MSR acclimation study (n = 10) and the current extended acclimation and their MSSQ percentile (Spearman Rank Correlation: t(9) = 3.35, study (n = 11), we pooled these data to improve the precision of R= 0.75, p = 0.008). This suggests that subjects who are more the analysis. susceptible to motion sickness (based on our screening before the For each spin rate from 1 RPM to 32 RPM (i.e., the spin rate that experiment began) experienced and reported higher motion applies 2 g loading at the feet and approximately 1 g at the rider’s sickness levels during the investigation. center of mass for the shortest feasible centrifuge), we calculated the probability of individuals reaching each spin rate within each High reliability of subject reporting of subjective CC illusion of one to 50 sessions. The entire compilation is provided as a table In order to assess reliability of subject reporting, we randomly in the Supplementary Materials, while the plots in Fig. 3 provide included up to one “catch trial” per session, which was either well the survival analysis staircase for (a) 20 RPM and (b) other example above or well below the subject’s instantaneous CC illusion staircases at 5, 10, and 30 RPM. For each spin rate, as the number threshold (see “Methods”). A total of 245 catch trials were of days increases, there is a subsequent increase in the probability completed across all subjects and sessions; 132 (53.9%) of the of subjects reaching the desired spin rate. For example, by training trials were High Catch Trials (50% faster spin rate), and 113 (46.1%) day 11 there is a 100% expected probability of subjects reaching 5 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2020) 22 Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10 Subject 11 Extended Average 10-Day Average Motion Sickness Rating (MSR) K.N. Bretl and T.K. Clark Survival Analysis Staircase | 20 RPM Survival Analysis - Spin Rate Suite 1 1 0.9 0.9 0.8 0.8 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 5 RPM 0.1 0.1 10 RPM Kaplan-Meier Survival Curve 20 RPM 95% Confidence Interval 30 RPM 0 0 Subject Censored Subject Censored 0 1020304050 0 1020304050 Days of Training Days of Training Fig. 3 Staircase of survival analysis findings. Survival analysis results for (a) 20 RPM and (b) 5, 10, 20, and 30 RPM. In panel a, a 95% confidence interval is plotted with gray dashed lines. The confidence intervals are omitted in panel b. Within both plots, the “X” markers along the staircase show when a subject becomes censored (i.e., when a subject leaves the investigation and additional data are unavailable). RPM, while this level of confidence requires several more training significant differences between the results of these subject days for higher RPMs—about 29 days for 10 RPM and 41 days for groups, we pooled subject data from both studies to complete 20 RPM. After the 50th day, there is a 70% probability that subjects the survival analysis. would acclimate to 25 RPM and a 60% probability of reaching 30 By conducting survival analysis of the data from all 21 subjects RPM, due to the greater CC illusion stimulus associated with exposed to a personalized, incremental training protocol, we higher spin rates. aimed to create a tool to aid in development of centrifuge designs. This tool may provide designers with a human tolerability trade (i.e., allowable spin rate) to accompany existing engineering DISCUSSION trades (e.g., mass, power, and volume). These results specifically The results of this study show that all subjects continued to reflect acclimation rates and performance using the protocol acclimate to the CC illusion with continued incremental exposure. implemented in the current investigation; different acclimation Based on our predetermined criteria, the data show no evidence approaches may result in different survival analysis estimates. of subjects reaching a plateau in ending threshold within the Further, using the outcomes in Fig. 3 and the Supplementary duration investigated, and all subjects exhibited a non-zero Materials table, we note that the survival analysis estimates are (positive) acclimation rate, indicating an upward trend in tolerable likely not the upper limit for tolerability. In our protocol, we threshold over multiple testing sessions. This suggests that there defined subjects to have not fully “acclimated” to a spin rate until exists potential for any individual (even one who acclimates at a they did not feel any CC illusion. Presumably, some presence of slower rate) to tolerably acclimate to a given spin rate of the illusion would still be tolerable for astronauts. This means that operational importance. there may be a higher probability of subjects tolerating a given Acclimation occurred for all subjects while motion sickness was spin rate with fewer days of training, or for higher spin rates to kept to a minimum. Throughout the study, subjects reported become tolerable with a shorter training protocol. Therefore, generally low motion sickness levels, and none of the subjects left designers should note that the analysis presented here likely the study due to nausea or dizziness, even those highly provides a conservative estimate of tolerability. susceptible to the motion sickness that could be induced by the Throughout the study, it was critically important that subjects CC illusion. Notably, an individual’s average motion sickness level remained naïve to the protocol and were able to maintain was correlated with the subject’s pretest MSSQ score. While this consistent criteria for reporting the presence of the CC illusion. We association might be expected, this is our first acclimation used blinded catch trials (presented at 50% or at least 5 RPM experiment in which the correlation was statistically significant. higher or lower than the subject’s current, near-threshold spin We suggest this finding resulted from the extended CC illusion rate) to assess the reliability of subject reports and guard against testing and exposure, which allowed for a more representative the potential for subjects to report what they thought they should average motion sickness rating for each subject. have been feeling or what they thought the test operators were Acclimation findings and motion sickness reports were compar- expecting. Nearly 95.5% of all catch trials resulted in subjects able across our previous 10-day personalized study and the reporting the presence or absence of the illusion as expected. The current extended study (Fig. 1a). This suggests that there was not remaining 4.5% may be explained by factors outside of our a dramatic effect of our modifications to the staircase (including control. Subjects verbally reported that at times, they could sense the catch trials) or a change in provocativeness over the additional acceleration of the centrifuge, potentially from airflow or vibration sessions. Additionally, our largely male subject group in the cues (white noise was played to mask auditory cues and the current study and balanced cohort in the previous investigation acceleration rate was intended to be subthreshold). When subjects performed comparably, further suggesting the absence of a reported feeling a change in spin rate, they were often not sure if gender effect in acclimation training. Because there were no they were spinning up or spinning down. If, for example, a subject npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA Probability of Reaching Ending Threshold of 20 RPM Probability of Reaching Desired Ending Threshold K.N. Bretl and T.K. Clark was exposed to a Low Catch Trial, but the vibration of the HERD was insufficient to overcome scheduling challenges. We note that made them mistakenly feel as though they were spinning up, they the three subjects who left the study before reaching our ending may report an unexpected presence of the illusion due to what criteria (X’s in Fig. 1b) appeared to be acclimating on similar they thought they should feel. Due to the infrequency of these trajectories as those subjects who did reach the ending criteria. unexpected results, we believe subjects were largely reliable in We recommend that future investigations work to test their reporting. additional subjects and further validate the design tool created A more thorough discussion on the limitations of the here, though in more operationally relevant configurations. experimental design and setup can be found elsewhere , but Although we do not anticipate conflicting results, we believe it briefly, it is important to note the most relevant limitations here. is necessary to ensure the efficacy of this protocol during off-axis First, subjects were seated upright while spinning about an Earth- centrifugation, in which the subjects lay supine on an off-axis vertical axis; head tilts were performed in one axis (roll tilt) while rotating bed (i.e., conventional centrifuge configuration) rather spinning in one direction. As this was the first study completed of than spinning about an Earth-vertical axis on an upright chair. this extended duration, the setup allowed for greater confidence Additionally, although we have demonstrated the prevalence of in our design due to reduced confounds. By targeting one head acclimation to head tilts performed in a rotating environment, tilt axis and rotation direction, we were able to isolate the subsequent studies should further investigate the potential for stimulus-response relationship and quantify acclimation in that more the generalized acclimation that will be necessary for single axis. Although the CC illusion acclimation appears to not astronauts—acclimation to complex head movements or acclima- readily transfer to other rotation axes , we suggest these findings tion transfer across head tilt planes. provide a proof-of-concept demonstration for short-radius cen- In this investigation, we sought to better understand human trifuge design, redefining previously accepted tolerable spin rate acclimation to the CC illusion through an extended personalized limits. acclimation protocol, such that we could better inform conceptual Second, we collected only subjects’ verbal reports of the centrifuge design. The results from this study quantify the longest presence or absence of the CC illusion following each head tilt. In CC illusion acclimation investigation completed to date. Although contrast with recording vestibulo-ocular reflex eye movements, limited in sample size, our findings suggest that all individuals we recognize that this is a subjective measure. However, we may have the capacity to acclimate to a spin rate of operational focused on subjective perception, as the associated disorientation interest, if they are given a training protocol of sufficient duration. is likely to be the primary operational concern of the CC illusion for This leads us to believe that the CC illusion and/or individual short-radius centrifugation. Despite the use of this subjective motion sickness susceptibility may no longer be the limitations metric, our catch trial results suggest that subjects did report restricting certain short-radius AG approaches. Instead, future reliably. limitations of minimum radius designs may shift to subject height As an additional limitation, all testing was completed in a (the centrifuge will presumably not be shorter than the subject), ground-based, 1 g environment. It is unknown how CC illusion engineering or power constraints, and/or other physiological acclimation would be altered in microgravity on-orbit. Previous 18–20 concerns. We have shown that tolerable short-radius, fast-rotation parabolic flight experiments producing ~20 s of microgravity 21–23 centrifugation is possible, and have developed a design tool to and on-orbit experiments during Skylab suggest that in quantify the expected acclimation to specific spin rates over up to microgravity, the CC illusion may be less intense, but still present 50 training sessions. This tool aims to assist in the development of in some form. However, it is worth noting that these flight a short-radius, intermittent centrifuge for artificial gravity imple- experiments had the subject’s head aligned with the spin axis (upright yaw rotating chair). Any centrifuge orientation in which mentation to enable superior protection of astronauts during the subject’s head is situated off-axis would induce centrifugal long-duration space exploration. loading to the vestibular system. We speculate that this would create Earth-like sensory conflict as a result of the CC illusion. METHODS Additionally, it may be important to have a centrifuge on the surface of the Moon or Mars to continue to mitigate physiological Subjects deconditioning during long-duration surface stays. Any planetary Eleven healthy subjects (10 M/1 F) volunteered to participate in this body-based centrifuge would still have a gravitational force acting investigation. We did not intentionally recruit a cohort with more males; on the rider. Even if this was a fraction of Earth’s gravity (e.g., 1/6 g however, our previous studies with a balanced cohort suggest there are not significant gender effects. Subjects had an average age of 22.2 years on the Moon or 3/8 g on Mars), it may still be sufficient to induce old (range: 20–25 years) and, on average, scored in the 43rd percentile on disorientation and motion sickness from the CC illusion like in our the Motion Sickness Susceptibility Questionnaire (range: 0–89). Subjects ground-based studies. The overall relevance of the CC illusion in were neither excluded nor included based on susceptibility to motion both ground- and microgravity-based centrifuges drives the need sickness; however, we verbally ensured that all subjects who were enrolled for individuals to become more tolerant of the illusion in order to in the study had no known history of vestibular dysfunction. All subjects increase both the feasibility and tolerability of using AG as a signed a written informed consent, and all protocols were approved by the spaceflight countermeasure. University of Colorado Institutional Review Board. Although we recognize the limitations of this ground-based approach for spaceflight, we also note that the encouraging Equipment results of our approach using an incremental, personalized This experiment utilized the upright chair configuration of the Human protocol to induce acclimation to the CC illusion may have more Eccentric Rotator Device (HERD) within the Bioastronautics Laboratory at broad applications to other vestibular adaptation and rehabilita- 14–16 the University of Colorado Boulder. As in previous studies , subjects tion programs here on Earth. For example, previous research has were positioned in the center of the rotating platform, seated in the chair shown that dynamic incremental training is a beneficial rehabilita- with a 4-point harness, and spun clockwise in yaw about an Earth-vertical tion tool in improving gaze stabilizing reflexes and reducing axis. All testing was completed in the dark. Head tilts were limited by two postural instability . foam blocks on either side of the subject’s head, and earbuds played white As a final limitation of the study, we note the challenge in noise to mask auditory cues from the HERD. Wireless two-way commu- recruiting subjects to participate in a study of such length. Only nication and video surveillance ensured continuous monitoring. Subjects eight of our 11 enrolled subjects completed the investigation by verbally reported the presence of the CC illusion and their motion sickness meeting one of our ending criteria. Subjects were provided ratings when prompted; they also pressed wireless pushbuttons for monetary incentives to finish the study, but in some cases, this redundant reporting. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2020) 22 K.N. Bretl and T.K. Clark Procedure responses, we added catch trials into the protocol, to which the subjects were naïve. On a catch trial (one trial is synonymous with one head tilt pair) All subjects in this investigation were exposed to a threshold-based, we altered the spin rate outside of the standard staircase described above. personalized, incremental acclimation protocol. To accomplish this, we Each catch trial was randomly selected as either a High Catch Trial (spin incremented spin rate based upon subject response to the CC illusion rate was adjusted to 50% higher than the spin rate at which the subject stimulus. Each subject experienced unique stimuli throughout the study, as was exposed immediately prior to the catch trial) or a Low Catch Trial (50% each protocol was individualized to ensure maximal tolerability by lower than preceding spin rate). On High Catch Trials, we would expect exposing subjects to spin rates at or just above their threshold (i.e., barely subjects to report feeling the illusion on one or both head tilts of the head perceivable). Critically, subjects remained naïve to the protocol and tilt pair, since the spin rate stimulus is much greater. Conversely, on a Low staircase as described. Subjects were informed that the investigation was a Catch Trial, we would expect that they would report not experiencing the “CC illusion acclimation study”, and that the HERD may accelerate or illusion on one or both head tilts of the pair. Depending upon the previous decelerate at random times throughout the experiment, but they were to spin rate, it was not always possible to increase or decrease the spin rate simply tilt their head when prompted, then report presence or absence of by exactly 50%, so we applied the following rules for catch trials and spin the CC illusion as a result of the head tilt without focusing much thought rate determination. First, the HERD can only be commanded with whole- on the spin rate at which they were rotating. They were not made aware number RPMs; therefore, if the desired 50% higher or 50% lower rate when the spin rate was changed nor the rules for why they may change. resulted in a decimal, Low Catch Trial spin rates were rounded down to the All subjects began the investigation spinning at a supra-threshold spin nearest whole number, while High Catch Trial spin rates were rounded up. rate of 10 RPM such that they could become familiar with the sensation of Second, to ensure a sufficient adjustment to the illusion intensity, we the CC illusion. Our previous investigations suggest that 10 RPM is a 14,15 required the change in spin rate between the regular protocol and the sufficiently strong stimulus to elicit the illusion in all subjects . After catch trial to be at least 5 RPM. The resulting catch trial spin rate could spinning at 10 RPM for ~30 s (to allow for the equilibration of the neither be slower than 1 RPM nor faster than 30 RPM (due to our safety endolymph in subjects’ inner ears), subjects performed a head tilt 40° (right protocol). If these rules were violated, the catch trial was not performed for ear) down over approximately one second and remained in the tilted that session. One catch trial was administered at a random time point position while reporting the presence or absence of the CC illusion as a within each session (one catch trial of the ~20 trials within each training direct result of that head tilt. Subjects verbally reported “yes I felt the session) to investigate if subjects were reporting as expected. Using catch illusion” or “no, I did not feel anything different from tilting my head in a trials to assess reporting reliability is particularly important given the stationary environment” while also pressing the corresponding pushbut- duration of this study. Because subjects are testing for up to 50 sessions, ton for reporting redundancy. After spinning with their head tilted for subjects must maintain stable decision criteria from their initial distinct, ~30 s, subjects were instructed to tilt their head back to the upright supra-threshold CC illusion exposure at the beginning of the first session to position over approximately one second and again report the presence or the end of their participation in the investigation. absence of the illusion. Following each head tilt pair (head tilt down and back to upright), the experiment operators decided whether to increase, Ending criteria. Subjects completed one 25-min acclimation session on maintain, or decrease the current spin rate, based on both subject every weekday for at least 10 days and up to 50 total days. Unlike our initial response and our predetermined bidirectional staircase rules. personalized acclimation protocol , all subjects did not complete the study after the same number of sessions. Instead, each subject remained in Protocol staircase. We employed an acclimation staircase that used each the experiment until one of three ending criteria were met: subject’s reporting of the CC illusion to determine the spin rate progression as the subject acclimated. The staircase was closely modeled (1) Reaching a beginning threshold of 25 RPM 14 13 after that which we used previously , inspired by Cheung et al . It sought (2) Reaching a plateau in acclimation (not increasing ending threshold to provide subjects with a CC illusion stimulus at or just above their for 10 consecutive sessions) threshold. The intention was to provide sufficient conflict between (3) Completing 50 acclimation sessions expected sensations and actual sensory input to drive acclimation, while The ending criteria, to which the subjects remained naïve, were limiting motion sickness (as compared to previous approaches cited intended to maximize knowledge gained from the experiment while earlier) by not being excessively provocative. minimizing unnecessary subject involvement and/or subject risk. We As in our previous staircase , subjects were introduced to the stimulus hypothesized that if all subjects completed all 50 sessions, some subjects at a supra-threshold spin rate of 10 RPM, administered only on the first day would likely reach spin rates that would exceed those of operational of the experiment. Subjects were accelerated to 10 RPM over 45 s, relevance. Although the optimal loading level is still an unknown design completed one head tilt pair at 10 RPM, then were decelerated to 1 RPM parameter, existing AG conceptual designs recommend loading on the over 60 s to begin the training protocol. At each subsequent spin rate, if order of 1 g at the subject’s center of mass and 2 g at the subject’s feet. subjects reported not feeling the illusion on both head tilts of one head tilt Similarly, an optimized centrifuge size has not yet been determined; pair, the spin rate was increased by 1 RPM over 10 s (acceleration was however, the centrifuge radius would likely be at least 2 m in order to chosen in an effort to maintain naivety of the staircase and protocol in accommodate even the tallest astronauts. Given these bounds, the fastest subjects). Alternatively, if subjects reported that they felt the illusion on an operational centrifuge would be spun is around 30–35 RPM; acclimating either or both head tilts within one head tilt pair, the spin rate was to higher spin rates is likely unnecessary. nominally maintained. However, as an addition to our previous staircase , This rationale drove the development of our ending criteria. We stopped in this study if subjects reported feeling the illusion on each head tilt of subjects after they achieved a sufficiently fast beginning threshold, defined three consecutive head tilt pairs, the spin rate was decreased by 1 RPM as the fastest spin rate at which no illusion was felt at the beginning of the over 10 s. If subjects repeatedly reported feeling the CC illusion (as would session (i.e., without any training during that session). A beginning be true in this scenario), it would indicate that the stimulus was decidedly threshold ending criteria of 25 RPM was selected to correspond to the spin supra-threshold. We added this third option to enable the staircase to be rate required to create at least 1 g loading at the feet of the majority of our bidirectional (i.e., the spin rate could decrease as well as increase). This subjects (if they were to be positioned supine on a centrifuge with their addition is important to quantify extended acclimation and identify a head at the center of the centrifuge). Once subjects reached that plateau if it exists. beginning threshold cutoff, we tested them throughout the rest of that For the entire duration of the training protocol, subjects were session; upon conclusion of the session, their participation in the study was accelerated or decelerated based upon their response to the previous terminated. stimuli, then spun for 30 s at the constant rate before any head tilts were The second ending criteria was included in the event that continued performed. As done previously , each subsequent session’s initial spin acclimation was not possible with continued exposure—that subjects’ rate was 1 RPM less than the spin rate in which the subject first reported ending thresholds (i.e., fastest spin rate at which no illusion was felt at the feeling the illusion on the prior session. Substantial effort was put forth to end of a session) reached a plateau. If this was the case, we did not want to train individual subjects at roughly the same time each day (i.e., within the continue testing individuals when additional acclimation would not be same 2–3 h), although subject availability ultimately dictated scheduling. possible. Finally, if subjects did not reach a beginning threshold of 25 RPM or a Catch trials. As subjects’ responses defined the staircase, a lack of proper plateau in ending threshold, we tested them for 50 total acclimation reporting (in terms of subjects not responding truthfully or reliably sessions. These ending criteria and testing duration allowed us to test our regarding their perception of a presence or absence of the illusion) could hypotheses regarding extended acclimation, working towards a better result in inaccurate or incomplete results. To assess the reliability of subject npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA K.N. Bretl and T.K. Clark understanding of the capacity of subjects to acclimate to the CC illusion, of sessions, where t refers to a time at which at least one subject reached thus informing centrifuge conceptual design. the desired spin rate, d references the total number of subjects reaching the desired rate at that time, and n is the number of subjects who are still being tested and have not yet acclimated to the given spin rate. The Motion sickness monitoring. To verify that our protocol facilitated benign complement of that percentage provides a calculation of the probability acclimation to the CC illusion, we asked subjects to verbally report their that subjects would reach the desired spin rate over t days of training, subjective motion sickness rating (MSR) once every 5 min of each 25-min assuming the adoption of our protocol exposing subjects to one, 25-min acclimation session. Motion sickness scores were to be reported on the 6–9,11–13,17 session per day. For each desired spin rate, a staircase cumulative survival simple but commonly used scale of 0–20 , where a score of 0 is function (i.e., estimated population acclimation curve) can be plotted, used to convey no sense of motion sickness, and a score of 20 represents showing how the probability of reaching the desired ending threshold the subject feeling as though he/she is on the verge of vomiting. increases as the number of training sessions increases. These developed If at any point subjects reported that they were feeling ill or wanted to spin down (for motion sickness reasons or otherwise), the testing operators curves can be used to aid in the conceptual design and development of a spaceflight centrifuge by ensuring tolerability of the selected design. did so. Additionally, if a subject reported an MSR of 10/20 or higher, the day’s session would prematurely conclude. If the same subject reported 10/20 or higher on a subsequent session, he/she would not continue the Reporting Summary study. This motion sickness-based ending criterion was implemented to Further information on research design is available in the Nature Research prevent excessive motion sickness, though as discussed in the Results, no Reporting Summary linked to this article. subject reported motion sickness greater than 10/20 more than once. Data and analysis DATA AVAILABILITY Metrics and variables of interest. We extracted similar metrics as in our The raw minimal datasets for this study have been made publicly available initial 10-day acclimation study , as both investigations sought to evaluate (https://osf.io/zw6xe/). We request citing this paper when using these datasets for the tolerability and feasibility of acclimation. To assess feasibility of further analysis. acclimation, we calculated beginning and ending threshold for each session. From these, we calculated a linear acclimation rate (a measure of Received: 10 April 2020; Accepted: 7 July 2020; the amount of acclimation achieved per session of testing). To quantify tolerability, we calculated the maximum and average motion sickness rating throughout all sessions and subjects. Each individual’s Motion Sickness Susceptibility Questionnaire (MSSQ) percentile was correlated with reported motion sickness levels. With these variables, we were primarily interested in determining the REFERENCES long-term potential of subjects to acclimate to the CC illusion, and if the 1. Clement, G., Bukley, A. & Paloski, W. Artifiical gravity as a countermeasure for protocol changes implemented in the extended acclimation study (i.e., mitigating physiological deconditioning during long-duration space missions. updated staircase and addition of catch trials) had a significant impact on Front. Syst. Neurosci. 9,1–11 (2015). how subjects were acclimating. 2. Lackner, J. R. & DiZio, P. Rapid adaptation to coriolis force perturbations of arm trajectory. J. Neurophysiol. 72, 299–313 (1994). Statistical tests. We performed statistical tests in MATLAB and R/RStudio. 3. Diaz-Artiles, A., Heldt, T. & Young, L. R. Short-term cardiovascular response to The assumption of normality was verified with Anderson–Darling and short-radius centrifugation with and without ergometer exercise. Front. Physiol. 9, Shapiro–Wilks tests, and F-tests were used to verify equality of variance. 1–16 (2018). Comparisons between groups were performed using two-tailed, two- 4. Duda, K. R., Jarchow, T. & Young, L. R. Squat exercise biomechanics during short- sample or paired t-tests with either equal or unequal variances (based on radius centrifugation. Aviat. Space Environ. Med. 83, 102–110 (2012). the result from the associated F-test). In the event that the dataset failed the 5. Guedry, F. E. & Montague, E. K. Quantitative evaluation of the vestibular coriolis normality tests, nonparametric tests were utilized (Mann–Whitney–Wilcoxon reaction. Aerosp. Med. 32, 487–500 (1961). test). Finally, Spearman rank nonparametric correlation tests were used to 6. Sheehan, S. E., Young, L. R. & Jarchow, T. The effect of head turn velocity on cross- measure potential association between subjects’ reported motion sickness coupled stimulation during centrifugation. J. Vestib. Res. 18,1–14 (2008). levels and their pretest MSSQ percentiles. A required level of significance of 7. Jarchow, T. & Young, L. R. Adaptation to head movements during short radius α ¼ 0:05 was used for all statistical tests. centrifugation. Acta Astonautica 61, 881–888 (2006). 8. Young, L. R., Sienko, K., Lyne, L. E., Hecht, H. & Natapoff, A. Adaptation of the Use of survival analysis to develop design tool. To investigate the long- vestibuo-ocular reflex, subjective tilt, and motion sickness to head movements term outlook of acclimation—namely, the expected ability of subjects to during short-radius centrifugation. J. Vestib. Res. 13,65–77 (2003). acclimate to certain operationally relevant spin rates over a given number 9. Young, L. R. et al. Artificial gravity: head movements during short-radius cen- of testing days—we applied survival analysis to the collected data. To trifugation. Acta Astronautica 49, 215–226 (2001). specifically accommodate subjects who did not complete the full 10. Hecht, H., Brown, E. L. & Young, L. R. Adapting to artificial gravity (AG) at high investigation or did not reach a spin rate of operational interest, this rotational speeds. J. Gravitational Physiol. 9,1–5 (2002). statistical method was utilized to project the censored data forward in 11. Brown, E. L., Hecht, H. & Young, L. R. Sensorimotor aspects of high-speed artificial time . Survival analysis works to critically evaluate the time it takes for an gravity: I. Sensory conflict in vestibular adaptation. J. Vestib. Res. 12, 271–282 event to occur. The event of interest in the current investigation is reaching (2002). a given spin rate threshold of operational relevance, such as that required 12. Elias, P., Jarchow, T. & Young, L. R. Incremental adaptation to yaw head turns to reach a desired loading level with a given centrifuge size (e.g., 15 RPM during 30 RPM centrifugation. Exp. Brain Res. 189, 269–277 (2008). for an 8-m diameter centrifuge). The ultimate goal with this analysis is to 13. Cheung, C. C., Hecht, H., Jarchow, T. & Young, L. R. Threshold-based vestibular estimate a population acclimation curve from our sample. adaptation to cross-coupled canal stimulation. J. Vestib. Res. 17, 171–181 (2007). We calculated the Kaplan–Meier survival analysis estimate , which 14. Bretl, K. N. et al. Tolerable acclimation to the cross-coupled illusion through a 10- quantifies an approximation of “survival”, or the time until the event of day, incremental, personalized protocol. J. Vestib. Res. 29,97–110 (2019). interest occurs. 15. Bretl, K. N., Sherman, S. O., Dixon, J. B., Mitchell, T. R. & Clark, T. K. A standardized, incremental protocol to increase human tolerance to the cross-coupled illusion. J. S ¼ 1  Vestib. Res. 29, 229–240 (2019). t (1) tit 16. Bretl, K. N. et al. Retention of cross-coupled illusion training to allow for a short radius space centrifuge. in IEEE Aerospace Conference (2018). 17. Garrick-Bethell, I., Jarchow, T., Hecht, H. & Young, L. R. Vestibular adaptation to The survival rate is expressed with the survival function, S , which is the centrifugation does not transfer across planes of head rotation. J. Vestib. Res. 18, proportion of individuals surviving longer than time t out of the total 25–37 (2008). number of individuals studied at that time. The product limit method, 18. Evanoff, J., DiZio, P. & Lackner, J. R. The influence of gravitoinertial force level on unique to the Kaplan–Meier analysis, can be seen in Eq. 1. For our oculomotor and perceptual responses to coriolis, cross-coupling stimulation. application, the survival function uses collected data to estimate the Aviat. Space Environ. Med. 58, 218–223 (1987). probability of not reaching the desired spin rate of interest over t number Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2020) 22 K.N. Bretl and T.K. Clark 19. Lackner, J. R. & Graybiel, A. The effective intensity of coriolis, cross-coupling completed by K.N.B. and T.K.C. Both authors edited and approved the final stimulation is gravitoinertial force dependent: Implications for space motion manuscript. sickness. Aviat. Space Environ. Med. 57, 229–235 (1986). 20. Lackner, J. R. & Graybiel, A. Influence of gravitoinertial force level on apparent magnitude of coriolis cross-coupled angular accelerations and motion sickness. COMPETING INTERESTS Mech. Predic. Prev. Treat. 22-1–22-7 (1984). The authors declare no competing interests. 21. Graybiel, A., Miller, E. F. & Homick, J. L. Individual differences in susceptibility to motion sickness among six Skylab astronauts. Acta Astronautica 2,155–174 (1975). 22. Lackner, J. R. Motion sickness: more than nausea and vomiting. Exp. Brain Res. ADDITIONAL INFORMATION 232, 2493–2510 (2014). Supplementary information is available for this paper at https://doi.org/10.1038/ 23. Graybiel, A., Miller, E. F. & Homick, J. L. Biomedical Results from Skylab (National s41526-020-00112-w. Aeronautics and Space Administration, 1977). 24. Schubert, M. C. & Migliaccio, A. A. New advances regarding adaptation of the Correspondence and requests for materials should be addressed to K.N.B. vestibulo-ocular reflex. J. Neurophysiol. 122, 644–658 (2019). 25. Cakit, B. D., Saracoglu, M., Genc, H. & Erdem, H. R. The effects of incremental Reprints and permission information is available at http://www.nature.com/ speed-dependent treadmill training on postural instability and fear of falling in reprints Parkinson’s disease. Clin. Rehab. 21, 698–705 (2007). 26. Reason, J. T. Relations between motion sickness susceptibility, the spiral after- Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims effect and loudness estimation. Br. J. Psychol. 59, 385–393 (1968). in published maps and institutional affiliations. 27. Miller, R. G. Survival Analysis. (Wiley-Interscience, 1998). 28. Goel, M. K., Khanna, P. & Kishore, J. Understanding survival analysis: Kaplan–Meier estimate. Int. J. Ayurveda Res. 1, 274–278 (2010). Open Access This article is licensed under a Creative Commons ACKNOWLEDGEMENTS Attribution 4.0 International License, which permits use, sharing, This work was supported by a NASA Space Technology Research Fellowship, grant adaptation, distribution and reproduction in any medium or format, as long as you give number 80NSSC17K0085. We thank all of the subjects who participated in the appropriate credit to the original author(s) and the source, provide a link to the Creative investigation, as well as Carson Brumley, Sebastian Metcalf, Varun Seth, and Marcos Commons license, and indicate if changes were made. The images or other third party Mejia for helping operate test sessions. Preliminary results were presented at the material in this article are included in the article’s Creative Commons license, unless following conferences and scientific meetings: Aerospace Medical Association indicated otherwise in a credit line to the material. If material is not included in the Scientific Meeting, Dallas, TX, 2018; Human Research Program Investigators’ article’s Creative Commons license and your intended use is not permitted by statutory Workshop, Galveston, TX, 2019. regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. AUTHOR CONTRIBUTIONS Experiments were designed and executed by K.N.B. and T.K.C., data analysis was © The Author(s) 2020 conducted by K.N.B., and data interpretation and manuscript preparation were npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png npj Microgravity Springer Journals

Improved feasibility of astronaut short-radius artificial gravity through a 50-day incremental, personalized, vestibular acclimation protocol

Bretl, Kathrine N.; Clark, Torin K.
npj Microgravity , Volume 6 (1) – Aug 26, 2020
Download PDF
8 pages

Loading next page...
 
/lp/springer-journals/improved-feasibility-of-astronaut-short-radius-artificial-gravity-IVRzkxjHE1

References (34)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2020
eISSN
2373-8065
DOI
10.1038/s41526-020-00112-w
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/npjmgrav ARTICLE OPEN Improved feasibility of astronaut short-radius artificial gravity through a 50-day incremental, personalized, vestibular acclimation protocol 1 1 Kathrine N. Bretl and Torin K. Clark The “Coriolis” cross-coupled (CC) illusion has historically limited the tolerability of utilizing fast-spin rate, short-radius centrifugation for in-flight artificial gravity. Previous research confirms that humans acclimate to the CC illusion over 10 daily sessions, though the efficacy of additional training is unknown. We investigated human acclimation to the CC illusion over up to 50 daily sessions of personalized, incremental training. During each 25-min session, subjects spun in yaw and performed roll head tilts approximately every 30 s, reporting the presence or absence of the illusion while rating motion sickness every 5 min. Illusion intensity was modulated by altering spin rate based upon subject response, such that the administered stimulus remained near each individual’s instantaneous illusion threshold. Every subject (n = 11) continued to acclimate linearly to the CC illusion during the investigation. Subjects acclimated at an average rate of 1.17 RPM per session (95% CI: 0.63–1.71 RPM per session), with the average tolerable spin rate increasing from 1.4 to 26.2 RPM, corresponding to a reduction in required centrifuge radius from 456.6 to 1.3 m (to produce loading of 1 g at the feet). Subjects reported no more than slight motion sickness throughout their training (mean: 0.92/20, 95% CI: 0.35–1.49/20). We applied survival analysis to determine the probability of individuals reaching various spin rates over a number of training days, providing a tolerability trade parameter for centrifuge design. Results indicate that acclimation to a given, operationally relevant spin rate may be feasible for all subjects if given a sufficient training duration. npj Microgravity (2020) 6:22 ; https://doi.org/10.1038/s41526-020-00112-w INTRODUCTION Previous research has shown that humans are capable of acclimating to the CC illusion, as subjects reported reduced For decades, artificial gravity (AG) has been proposed as the most 8–11 intensity of the illusion after two or three sessions of exposure . comprehensive spaceflight countermeasure, holding the potential These early acclimation studies exposed subjects to the CC illusion to concurrently protect multiple physiological systems from in- at spin rates of 23 RPM or higher. This very provocative stimulus flight, microgravity-induced deconditioning. Several conceptual resulted in substantial motion sickness in subjects; notably, it designs have been proposed, although all face the challenge of caused ~25–35% of subjects to dropout of each study due to budgetary and engineering constraints. The most feasible motion sickness, despite the fact that subjects who were highly approach to AG in the near term is thought to be short-radius, 1 susceptible to motion sickness were often excluded from intermittent centrifugation , though additional challenges surface 8,10,11 participating . with this design. Of these, it appears the human body can 2 3,4 Subsequent studies investigated the possibility of benign acclimate quite quickly to Coriolis forces and gravity gradients , exposure to the CC illusion. It was found that an incremental but acclimation to the disorienting “Coriolis” cross-coupled (CC) protocol provides a less provocative way to expose subjects by illusion is less expeditious and thorough (we note that the CC reducing motion sickness via incremental spin rate exposure (i.e., illusion is not the result of a conventional and substantial Coriolis increasing spin rates over time rather than directly exposing effect and therefore have placed it in quotations to distinguish subjects to a high CC illusion stimulus/fast spin rate). One from traditional Coriolis forces). It remains unclear whether all incremental CC illusion acclimation approach exposed subjects individuals can tolerate and/or acclimate to the moderately high to spin rates of 14 RPM on day 1, 23 RPM on day 2, and 30 RPM on spin rates (e.g., ~15+ rotations per minute (RPM)) that would be day 3 . This study showed that the incremental steps improved required for short-radius centrifugation. subjects’ tolerance of the acclimation protocol, yet it was still too The CC illusion is a provocative tilting or tumbling sensation aggressive for some of the subjects, as 14% (4/28) of the subjects experienced by a subject in a constantly rotating field after he/she dropped out due to motion sickness. Another variant of performs a head tilt out of the plane of constant rotation. The incremental CC illusion acclimation showed that personalized, illusion is highly disorienting and typically leads to motion incremental acclimation over 2 days of training resulted in sickness . The intensity of the illusion is proportional to the spin essentially no cases of motion sickness, and therefore, a 0% rate at which the subject is exposed, the magnitude of the head subject dropout rate (0/8 subjects) . 6,7 tilt, and the velocity of the tilt . Higher spin rates are required for We have recently extended previous efforts to investigate effective short-radius centrifuge designs; thus, the CC illusion personalized, incremental acclimation to the CC illusion over ten becomes increasingly relevant as the potential for disorientation consecutive weekdays . In short, subjects were seated upright and motion sickness becomes more prominent with a faster- and spun in yaw about an Earth-vertical axis. After an introduction spinning centrifuge. to the CC stimulus, each subject began the training protocol University of Colorado Boulder, 3775 Discovery Drive, Boulder, CO 80309, USA. email: kathrine.bretl@colorado.edu Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; K.N. Bretl and T.K. Clark spinning at 1 RPM. While spinning at a constant rate, subjects RESULTS performed roll head tilts (40° right ear down and back to upright) Extended acclimation protocol resulted in continued acclimation approximately every 30 s. After each head tilt, subjects reported if with no evidence of plateau in all subjects they experienced the CC illusion. If the subject did not report the Eight of our 11 subjects completed the investigation by illusion on two consecutive head tilts of one head tilt pair (head reaching one of our first two ending criteria (see “Methods”): tilt down and back upright), the spin rate was increased by 1 RPM; either a beginning threshold (i.e., fastest spin rate at which no otherwise, the spin rate was maintained. Ultimately, the spin rate illusion was experienced at the beginning of the session) of 25 was incremented only based upon subject response, and the RPM (n = 7) or completing a total of 50 testing sessions (n = 1). stimulus was always just barely noticeable (i.e., personalized and Notably, none of the subjects appeared to have reached a threshold based). The stimulus was effective in acclimating all plateau in ending threshold based on our ending criteria 10 subjects, while not being excessively strong to elicit substantial (Fig. 1). The final three subjects who enrolled in the study were motion sickness. Subjects acclimated from an average threshold included in analysis, as they each completed 15 or more testing (i.e., fastest spin rate at which no CC illusion was felt) of 1.8 RPM sessions. However, they did not complete the study protocol; all (range: 1–3 RPM) before any training to 17.7 RPM (range: 3–30 RPM) after 10, 25-min training sessions. Across all sessions and three chose to prematurely leave the study due to challenges subjects, an average motion sickness rating of only 1.06/20 was with scheduling (denoted with a black “X” in Fig. 1b). Including reported. This investigation confirmed that a personalized, all 11 subjects, the average number of training sessions incremental protocol is a highly effective, benign method of completed was 25.5 (range: 10–50 sessions). Subjects began acclimating individuals to the CC illusion. the investigation at an average pre-training threshold (i.e., the In addition to personalized acclimation, we have also investi- fastest spin rate at which no illusion was felt before any gated the efficacy of standardized (i.e., incremental, but non- acclimation training) of 1.4 RPM (range: 0–3 RPM, a threshold of personalized) acclimation and the ability of subjects to retain 0 RPM corresponds to reporting the illusion at 1 RPM). While gained acclimation after a period without training . These studies acclimating over a varying number of sessions, subjects have suggested that non-personalized acclimation is not as reached an average ending threshold (i.e., fastest spin rate effective or tolerable as personalized acclimation; however, that elicited no CC illusion at the conclusion of the session) of acclimation in both studies appears to be mostly retained for up 25.8 RPM (range: 10–38 RPM) on their final day of testing. This to 90 days of unmonitored activity. These previous studies sought resulted in a significant increase in CC illusion acclimation to investigate acclimation after an initial protocol of 10 daily threshold (paired t-test: t(10) = 8.95, diff = 24.4 RPM, Cohen’s sessions. Remarkably, it appeared acclimation continued linearly d = 3.73, p < 0.0005). Subjects’ acclimation can also be quanti- over those 10 days, suggesting further acclimation may be fied by their acclimation rate (i.e., the slope of a linear fittoeach possible. However, potential for acclimation beyond these 10 days subject’s ending threshold as a function of session number). is unknown. Specifically, it is of operational and scientific interest Average acclimation rate was found to be significantly non-zero to determine if acclimation continues with additional exposure, or across subjects (One-Sample t-test: t(10) = 4.82, mean = 1.17 if a plateau is reached in some or all individuals, after which no RPM persession,95% CI:0.63–1.71 RPM per session, p = further acclimation is possible. 0.0007). Further, all 11 subjects exhibited positive acclimation The objective of this investigation is to test the ability of rates (range: 0.26 RPM per session to 2.91 RPM per session), subjects to acclimate to extended CC illusion exposure (i.e., up to showing that all subjects—even those who increased their 50, 25-min sessions), using essentially the same personalized, incremental staircase described above, and inform operational threshold at a slower pace—were able to acclimate to the CC considerations of centrifuge design. illusion. All Acclimation Subjects | First 10 Sessions Extended Acclimation Subjects | All Sessions 40 1.1 40 1.1 10-Day Personalized Subject 6 Subject Complete Subject 1 Subject 7 Subject Drop-Out Subject 2 Subject 8 a b Subject 3 Subject 9 35 1.5 35 1.5 Subject 4 Subject 10 Subject 5 Subject 11 30 2.0 30 2.0 25 2.9 25 2.9 20 4.5 20 4.5 15 8.0 15 8.0 10 17.9 10 17.9 5 71.6 5 71.6 0 0 Pre-Training 1 2 3 4 5 6 7 8 9 10 Pre-Training 5 10 15 20 25 30 35 40 45 50 Session Number Session Number Fig. 1 Extended acclimation findings. CC illusion acclimation pretraining and ending thresholds for (a) 10-day and extended acclimation subjects over the first 10 sessions and (b) extended acclimation subjects over the entire duration of their involvement in the study. In panel a, the previous 10-day investigation subjects are shown in gray, while the current extended acclimation subjects are overlaid in color. In panel b, subjects who completed the study by reaching the investigation’s ending criteria are shown with a black circle at their final data point, while subjects who left the study due to scheduling challenges are denoted with a black “X”. npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA 1234567890():,; Ending Threshold (RPM) Required Diameter to Produce 1 G (m) Ending Threshold (RPM) Required Diameter to Produce 1 G (m) K.N. Bretl and T.K. Clark Extended acclimation investigation subjects with modified Extended Acclimation Motion Sickness Rating staircase protocol and previous 10-day subjects acclimated Subject Average MSR - Extended similarly over first 10 days 95% Confidence Interval Subject Maximum MSR - Extended These findings are similar to what has been previously reported Group Average MSR - Extended Group Average MSR - 10 Day over just 10 days , although additional confidence has been gained with the extended exposure. Figure 1a shows our original 10-day personalized acclimation subjects in gray with our extended acclimation subjects overlaid in color (for just the first 13 10 days of the investigation). This plot shows that the extended 12 acclimation subjects performed within the same bounds as our previous 10-day acclimation subjects, despite the slight modifica- tions in the staircase protocol with the inclusion of catch trials and the option to increment to slower spin rates if the CC illusion was reported on each head tilt for three consecutive head tilt pairs (see “Methods”). A two-sample t-test investigating potential differences between ending thresholds on day 10 of the two subject groups confirmed the two groups were comparable, with a non- significant result (Two-sample independent t-test: p = 0.46). Additionally, there was no evidence to suggest a difference in acclimation rate between the extended acclimation subjects and the previous 10-day acclimation subjects over the first 10 days of the investigation (Two-sample independent t-test: p = 0.27). All subjects acclimated to the CC illusion, but as displayed in Fig. 1, they acclimated at substantially different rates. As observed in 13–16 previous studies , there is a large degree of inter-individual Fig. 2 Extended acclimation motion sickness results. Subject differences in acclimation. As one measure in the present study, averages are denoted with the light blue bars, maximum reports across all sessions are shown with light blue asterisks, and the dark the acclimation rate averaged 1.17 RPM per session but varied blue bar shows the overall average across all extended duration substantially with a standard deviation of 0.81 RPM per session, acclimation subjects. The gray bar shows the 10-day personalized yielding a coefficient of variation of 0.69. subject group average motion sickness score for comparison. Error bars show 95% confidence intervals about the subject means. Subjects reported low motion sickness throughout study The incremental, personalized protocol was designed to limit were Low Catch Trials (50% slower spin rate, 35 trials which would motion sickness experienced during acclimation. The protocol was have been Low Catch Trials were not performed due to the rules outlined in “Methods”). Overall, subjects almost always responded effective in providing this benign acclimation to the CC illusion, as to the catch trials as we would have expected. Specifically, on only subjects reported generally low motion sickness scores through- 4.5% of the catch trials did the subject respond unexpectedly: out the investigation (Fig. 2). either that they felt the illusion on both head tilts for a Low Catch Most importantly, none of the 11 subjects dropped out of the Trial (2.4%) or that they did not feel the illusion during a High study due to excessive motion sickness, nor was any session terminated early due to subject discomfort or elevated motion Catch Trial (2.0%). This gives us confidence that the subjects were sickness reporting (i.e., 10/20 or higher on the utilized 0–20 scale, reporting reliably and were able to maintain consistent criteria of where 0 corresponds to no motion sickness and 20 corresponds to what qualifies as the presence of the CC illusion throughout the extreme nausea or vomiting). We note that one subject did report duration of the study. an 11/20 on one of the sessions, although this report was communicated in the final moments of the session, therefore the Survival analysis and expected population acclimation complete session was conducted. The average MSR across all With this study and those published previously, we have shown subjects and sessions was 0.92 (95% CI: 0.35–1.49) on the that subjects who undergo the developed training protocol commonly used 0–20 scale. This was not significantly different become more tolerant of the CC illusion over time, though than the average MSR of 1.06/20 (95% CI: 0.37–1.76) from the subjects acclimate at varying rates. To use this research as a tool 10 subjects who completed the previous 10-day personalized for centrifuge design, we have conducted an analysis to estimate protocol (Mann–Whitney–Wilcoxon test, p = 0.78). the probability that individuals will acclimate to a specific spin rate We observed no trend in MSR reporting across sessions or over a number of training days. Since we found no statistical reports within a session; however, we did find a statistically differences between the data collected during the previous 10-day significant correlation between subjects’ average reported MSR acclimation study (n = 10) and the current extended acclimation and their MSSQ percentile (Spearman Rank Correlation: t(9) = 3.35, study (n = 11), we pooled these data to improve the precision of R= 0.75, p = 0.008). This suggests that subjects who are more the analysis. susceptible to motion sickness (based on our screening before the For each spin rate from 1 RPM to 32 RPM (i.e., the spin rate that experiment began) experienced and reported higher motion applies 2 g loading at the feet and approximately 1 g at the rider’s sickness levels during the investigation. center of mass for the shortest feasible centrifuge), we calculated the probability of individuals reaching each spin rate within each High reliability of subject reporting of subjective CC illusion of one to 50 sessions. The entire compilation is provided as a table In order to assess reliability of subject reporting, we randomly in the Supplementary Materials, while the plots in Fig. 3 provide included up to one “catch trial” per session, which was either well the survival analysis staircase for (a) 20 RPM and (b) other example above or well below the subject’s instantaneous CC illusion staircases at 5, 10, and 30 RPM. For each spin rate, as the number threshold (see “Methods”). A total of 245 catch trials were of days increases, there is a subsequent increase in the probability completed across all subjects and sessions; 132 (53.9%) of the of subjects reaching the desired spin rate. For example, by training trials were High Catch Trials (50% faster spin rate), and 113 (46.1%) day 11 there is a 100% expected probability of subjects reaching 5 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2020) 22 Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10 Subject 11 Extended Average 10-Day Average Motion Sickness Rating (MSR) K.N. Bretl and T.K. Clark Survival Analysis Staircase | 20 RPM Survival Analysis - Spin Rate Suite 1 1 0.9 0.9 0.8 0.8 0.7 0.7 0.6 0.6 0.5 0.5 0.4 0.4 0.3 0.3 0.2 0.2 5 RPM 0.1 0.1 10 RPM Kaplan-Meier Survival Curve 20 RPM 95% Confidence Interval 30 RPM 0 0 Subject Censored Subject Censored 0 1020304050 0 1020304050 Days of Training Days of Training Fig. 3 Staircase of survival analysis findings. Survival analysis results for (a) 20 RPM and (b) 5, 10, 20, and 30 RPM. In panel a, a 95% confidence interval is plotted with gray dashed lines. The confidence intervals are omitted in panel b. Within both plots, the “X” markers along the staircase show when a subject becomes censored (i.e., when a subject leaves the investigation and additional data are unavailable). RPM, while this level of confidence requires several more training significant differences between the results of these subject days for higher RPMs—about 29 days for 10 RPM and 41 days for groups, we pooled subject data from both studies to complete 20 RPM. After the 50th day, there is a 70% probability that subjects the survival analysis. would acclimate to 25 RPM and a 60% probability of reaching 30 By conducting survival analysis of the data from all 21 subjects RPM, due to the greater CC illusion stimulus associated with exposed to a personalized, incremental training protocol, we higher spin rates. aimed to create a tool to aid in development of centrifuge designs. This tool may provide designers with a human tolerability trade (i.e., allowable spin rate) to accompany existing engineering DISCUSSION trades (e.g., mass, power, and volume). These results specifically The results of this study show that all subjects continued to reflect acclimation rates and performance using the protocol acclimate to the CC illusion with continued incremental exposure. implemented in the current investigation; different acclimation Based on our predetermined criteria, the data show no evidence approaches may result in different survival analysis estimates. of subjects reaching a plateau in ending threshold within the Further, using the outcomes in Fig. 3 and the Supplementary duration investigated, and all subjects exhibited a non-zero Materials table, we note that the survival analysis estimates are (positive) acclimation rate, indicating an upward trend in tolerable likely not the upper limit for tolerability. In our protocol, we threshold over multiple testing sessions. This suggests that there defined subjects to have not fully “acclimated” to a spin rate until exists potential for any individual (even one who acclimates at a they did not feel any CC illusion. Presumably, some presence of slower rate) to tolerably acclimate to a given spin rate of the illusion would still be tolerable for astronauts. This means that operational importance. there may be a higher probability of subjects tolerating a given Acclimation occurred for all subjects while motion sickness was spin rate with fewer days of training, or for higher spin rates to kept to a minimum. Throughout the study, subjects reported become tolerable with a shorter training protocol. Therefore, generally low motion sickness levels, and none of the subjects left designers should note that the analysis presented here likely the study due to nausea or dizziness, even those highly provides a conservative estimate of tolerability. susceptible to the motion sickness that could be induced by the Throughout the study, it was critically important that subjects CC illusion. Notably, an individual’s average motion sickness level remained naïve to the protocol and were able to maintain was correlated with the subject’s pretest MSSQ score. While this consistent criteria for reporting the presence of the CC illusion. We association might be expected, this is our first acclimation used blinded catch trials (presented at 50% or at least 5 RPM experiment in which the correlation was statistically significant. higher or lower than the subject’s current, near-threshold spin We suggest this finding resulted from the extended CC illusion rate) to assess the reliability of subject reports and guard against testing and exposure, which allowed for a more representative the potential for subjects to report what they thought they should average motion sickness rating for each subject. have been feeling or what they thought the test operators were Acclimation findings and motion sickness reports were compar- expecting. Nearly 95.5% of all catch trials resulted in subjects able across our previous 10-day personalized study and the reporting the presence or absence of the illusion as expected. The current extended study (Fig. 1a). This suggests that there was not remaining 4.5% may be explained by factors outside of our a dramatic effect of our modifications to the staircase (including control. Subjects verbally reported that at times, they could sense the catch trials) or a change in provocativeness over the additional acceleration of the centrifuge, potentially from airflow or vibration sessions. Additionally, our largely male subject group in the cues (white noise was played to mask auditory cues and the current study and balanced cohort in the previous investigation acceleration rate was intended to be subthreshold). When subjects performed comparably, further suggesting the absence of a reported feeling a change in spin rate, they were often not sure if gender effect in acclimation training. Because there were no they were spinning up or spinning down. If, for example, a subject npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA Probability of Reaching Ending Threshold of 20 RPM Probability of Reaching Desired Ending Threshold K.N. Bretl and T.K. Clark was exposed to a Low Catch Trial, but the vibration of the HERD was insufficient to overcome scheduling challenges. We note that made them mistakenly feel as though they were spinning up, they the three subjects who left the study before reaching our ending may report an unexpected presence of the illusion due to what criteria (X’s in Fig. 1b) appeared to be acclimating on similar they thought they should feel. Due to the infrequency of these trajectories as those subjects who did reach the ending criteria. unexpected results, we believe subjects were largely reliable in We recommend that future investigations work to test their reporting. additional subjects and further validate the design tool created A more thorough discussion on the limitations of the here, though in more operationally relevant configurations. experimental design and setup can be found elsewhere , but Although we do not anticipate conflicting results, we believe it briefly, it is important to note the most relevant limitations here. is necessary to ensure the efficacy of this protocol during off-axis First, subjects were seated upright while spinning about an Earth- centrifugation, in which the subjects lay supine on an off-axis vertical axis; head tilts were performed in one axis (roll tilt) while rotating bed (i.e., conventional centrifuge configuration) rather spinning in one direction. As this was the first study completed of than spinning about an Earth-vertical axis on an upright chair. this extended duration, the setup allowed for greater confidence Additionally, although we have demonstrated the prevalence of in our design due to reduced confounds. By targeting one head acclimation to head tilts performed in a rotating environment, tilt axis and rotation direction, we were able to isolate the subsequent studies should further investigate the potential for stimulus-response relationship and quantify acclimation in that more the generalized acclimation that will be necessary for single axis. Although the CC illusion acclimation appears to not astronauts—acclimation to complex head movements or acclima- readily transfer to other rotation axes , we suggest these findings tion transfer across head tilt planes. provide a proof-of-concept demonstration for short-radius cen- In this investigation, we sought to better understand human trifuge design, redefining previously accepted tolerable spin rate acclimation to the CC illusion through an extended personalized limits. acclimation protocol, such that we could better inform conceptual Second, we collected only subjects’ verbal reports of the centrifuge design. The results from this study quantify the longest presence or absence of the CC illusion following each head tilt. In CC illusion acclimation investigation completed to date. Although contrast with recording vestibulo-ocular reflex eye movements, limited in sample size, our findings suggest that all individuals we recognize that this is a subjective measure. However, we may have the capacity to acclimate to a spin rate of operational focused on subjective perception, as the associated disorientation interest, if they are given a training protocol of sufficient duration. is likely to be the primary operational concern of the CC illusion for This leads us to believe that the CC illusion and/or individual short-radius centrifugation. Despite the use of this subjective motion sickness susceptibility may no longer be the limitations metric, our catch trial results suggest that subjects did report restricting certain short-radius AG approaches. Instead, future reliably. limitations of minimum radius designs may shift to subject height As an additional limitation, all testing was completed in a (the centrifuge will presumably not be shorter than the subject), ground-based, 1 g environment. It is unknown how CC illusion engineering or power constraints, and/or other physiological acclimation would be altered in microgravity on-orbit. Previous 18–20 concerns. We have shown that tolerable short-radius, fast-rotation parabolic flight experiments producing ~20 s of microgravity 21–23 centrifugation is possible, and have developed a design tool to and on-orbit experiments during Skylab suggest that in quantify the expected acclimation to specific spin rates over up to microgravity, the CC illusion may be less intense, but still present 50 training sessions. This tool aims to assist in the development of in some form. However, it is worth noting that these flight a short-radius, intermittent centrifuge for artificial gravity imple- experiments had the subject’s head aligned with the spin axis (upright yaw rotating chair). Any centrifuge orientation in which mentation to enable superior protection of astronauts during the subject’s head is situated off-axis would induce centrifugal long-duration space exploration. loading to the vestibular system. We speculate that this would create Earth-like sensory conflict as a result of the CC illusion. METHODS Additionally, it may be important to have a centrifuge on the surface of the Moon or Mars to continue to mitigate physiological Subjects deconditioning during long-duration surface stays. Any planetary Eleven healthy subjects (10 M/1 F) volunteered to participate in this body-based centrifuge would still have a gravitational force acting investigation. We did not intentionally recruit a cohort with more males; on the rider. Even if this was a fraction of Earth’s gravity (e.g., 1/6 g however, our previous studies with a balanced cohort suggest there are not significant gender effects. Subjects had an average age of 22.2 years on the Moon or 3/8 g on Mars), it may still be sufficient to induce old (range: 20–25 years) and, on average, scored in the 43rd percentile on disorientation and motion sickness from the CC illusion like in our the Motion Sickness Susceptibility Questionnaire (range: 0–89). Subjects ground-based studies. The overall relevance of the CC illusion in were neither excluded nor included based on susceptibility to motion both ground- and microgravity-based centrifuges drives the need sickness; however, we verbally ensured that all subjects who were enrolled for individuals to become more tolerant of the illusion in order to in the study had no known history of vestibular dysfunction. All subjects increase both the feasibility and tolerability of using AG as a signed a written informed consent, and all protocols were approved by the spaceflight countermeasure. University of Colorado Institutional Review Board. Although we recognize the limitations of this ground-based approach for spaceflight, we also note that the encouraging Equipment results of our approach using an incremental, personalized This experiment utilized the upright chair configuration of the Human protocol to induce acclimation to the CC illusion may have more Eccentric Rotator Device (HERD) within the Bioastronautics Laboratory at broad applications to other vestibular adaptation and rehabilita- 14–16 the University of Colorado Boulder. As in previous studies , subjects tion programs here on Earth. For example, previous research has were positioned in the center of the rotating platform, seated in the chair shown that dynamic incremental training is a beneficial rehabilita- with a 4-point harness, and spun clockwise in yaw about an Earth-vertical tion tool in improving gaze stabilizing reflexes and reducing axis. All testing was completed in the dark. Head tilts were limited by two postural instability . foam blocks on either side of the subject’s head, and earbuds played white As a final limitation of the study, we note the challenge in noise to mask auditory cues from the HERD. Wireless two-way commu- recruiting subjects to participate in a study of such length. Only nication and video surveillance ensured continuous monitoring. Subjects eight of our 11 enrolled subjects completed the investigation by verbally reported the presence of the CC illusion and their motion sickness meeting one of our ending criteria. Subjects were provided ratings when prompted; they also pressed wireless pushbuttons for monetary incentives to finish the study, but in some cases, this redundant reporting. Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2020) 22 K.N. Bretl and T.K. Clark Procedure responses, we added catch trials into the protocol, to which the subjects were naïve. On a catch trial (one trial is synonymous with one head tilt pair) All subjects in this investigation were exposed to a threshold-based, we altered the spin rate outside of the standard staircase described above. personalized, incremental acclimation protocol. To accomplish this, we Each catch trial was randomly selected as either a High Catch Trial (spin incremented spin rate based upon subject response to the CC illusion rate was adjusted to 50% higher than the spin rate at which the subject stimulus. Each subject experienced unique stimuli throughout the study, as was exposed immediately prior to the catch trial) or a Low Catch Trial (50% each protocol was individualized to ensure maximal tolerability by lower than preceding spin rate). On High Catch Trials, we would expect exposing subjects to spin rates at or just above their threshold (i.e., barely subjects to report feeling the illusion on one or both head tilts of the head perceivable). Critically, subjects remained naïve to the protocol and tilt pair, since the spin rate stimulus is much greater. Conversely, on a Low staircase as described. Subjects were informed that the investigation was a Catch Trial, we would expect that they would report not experiencing the “CC illusion acclimation study”, and that the HERD may accelerate or illusion on one or both head tilts of the pair. Depending upon the previous decelerate at random times throughout the experiment, but they were to spin rate, it was not always possible to increase or decrease the spin rate simply tilt their head when prompted, then report presence or absence of by exactly 50%, so we applied the following rules for catch trials and spin the CC illusion as a result of the head tilt without focusing much thought rate determination. First, the HERD can only be commanded with whole- on the spin rate at which they were rotating. They were not made aware number RPMs; therefore, if the desired 50% higher or 50% lower rate when the spin rate was changed nor the rules for why they may change. resulted in a decimal, Low Catch Trial spin rates were rounded down to the All subjects began the investigation spinning at a supra-threshold spin nearest whole number, while High Catch Trial spin rates were rounded up. rate of 10 RPM such that they could become familiar with the sensation of Second, to ensure a sufficient adjustment to the illusion intensity, we the CC illusion. Our previous investigations suggest that 10 RPM is a 14,15 required the change in spin rate between the regular protocol and the sufficiently strong stimulus to elicit the illusion in all subjects . After catch trial to be at least 5 RPM. The resulting catch trial spin rate could spinning at 10 RPM for ~30 s (to allow for the equilibration of the neither be slower than 1 RPM nor faster than 30 RPM (due to our safety endolymph in subjects’ inner ears), subjects performed a head tilt 40° (right protocol). If these rules were violated, the catch trial was not performed for ear) down over approximately one second and remained in the tilted that session. One catch trial was administered at a random time point position while reporting the presence or absence of the CC illusion as a within each session (one catch trial of the ~20 trials within each training direct result of that head tilt. Subjects verbally reported “yes I felt the session) to investigate if subjects were reporting as expected. Using catch illusion” or “no, I did not feel anything different from tilting my head in a trials to assess reporting reliability is particularly important given the stationary environment” while also pressing the corresponding pushbut- duration of this study. Because subjects are testing for up to 50 sessions, ton for reporting redundancy. After spinning with their head tilted for subjects must maintain stable decision criteria from their initial distinct, ~30 s, subjects were instructed to tilt their head back to the upright supra-threshold CC illusion exposure at the beginning of the first session to position over approximately one second and again report the presence or the end of their participation in the investigation. absence of the illusion. Following each head tilt pair (head tilt down and back to upright), the experiment operators decided whether to increase, Ending criteria. Subjects completed one 25-min acclimation session on maintain, or decrease the current spin rate, based on both subject every weekday for at least 10 days and up to 50 total days. Unlike our initial response and our predetermined bidirectional staircase rules. personalized acclimation protocol , all subjects did not complete the study after the same number of sessions. Instead, each subject remained in Protocol staircase. We employed an acclimation staircase that used each the experiment until one of three ending criteria were met: subject’s reporting of the CC illusion to determine the spin rate progression as the subject acclimated. The staircase was closely modeled (1) Reaching a beginning threshold of 25 RPM 14 13 after that which we used previously , inspired by Cheung et al . It sought (2) Reaching a plateau in acclimation (not increasing ending threshold to provide subjects with a CC illusion stimulus at or just above their for 10 consecutive sessions) threshold. The intention was to provide sufficient conflict between (3) Completing 50 acclimation sessions expected sensations and actual sensory input to drive acclimation, while The ending criteria, to which the subjects remained naïve, were limiting motion sickness (as compared to previous approaches cited intended to maximize knowledge gained from the experiment while earlier) by not being excessively provocative. minimizing unnecessary subject involvement and/or subject risk. We As in our previous staircase , subjects were introduced to the stimulus hypothesized that if all subjects completed all 50 sessions, some subjects at a supra-threshold spin rate of 10 RPM, administered only on the first day would likely reach spin rates that would exceed those of operational of the experiment. Subjects were accelerated to 10 RPM over 45 s, relevance. Although the optimal loading level is still an unknown design completed one head tilt pair at 10 RPM, then were decelerated to 1 RPM parameter, existing AG conceptual designs recommend loading on the over 60 s to begin the training protocol. At each subsequent spin rate, if order of 1 g at the subject’s center of mass and 2 g at the subject’s feet. subjects reported not feeling the illusion on both head tilts of one head tilt Similarly, an optimized centrifuge size has not yet been determined; pair, the spin rate was increased by 1 RPM over 10 s (acceleration was however, the centrifuge radius would likely be at least 2 m in order to chosen in an effort to maintain naivety of the staircase and protocol in accommodate even the tallest astronauts. Given these bounds, the fastest subjects). Alternatively, if subjects reported that they felt the illusion on an operational centrifuge would be spun is around 30–35 RPM; acclimating either or both head tilts within one head tilt pair, the spin rate was to higher spin rates is likely unnecessary. nominally maintained. However, as an addition to our previous staircase , This rationale drove the development of our ending criteria. We stopped in this study if subjects reported feeling the illusion on each head tilt of subjects after they achieved a sufficiently fast beginning threshold, defined three consecutive head tilt pairs, the spin rate was decreased by 1 RPM as the fastest spin rate at which no illusion was felt at the beginning of the over 10 s. If subjects repeatedly reported feeling the CC illusion (as would session (i.e., without any training during that session). A beginning be true in this scenario), it would indicate that the stimulus was decidedly threshold ending criteria of 25 RPM was selected to correspond to the spin supra-threshold. We added this third option to enable the staircase to be rate required to create at least 1 g loading at the feet of the majority of our bidirectional (i.e., the spin rate could decrease as well as increase). This subjects (if they were to be positioned supine on a centrifuge with their addition is important to quantify extended acclimation and identify a head at the center of the centrifuge). Once subjects reached that plateau if it exists. beginning threshold cutoff, we tested them throughout the rest of that For the entire duration of the training protocol, subjects were session; upon conclusion of the session, their participation in the study was accelerated or decelerated based upon their response to the previous terminated. stimuli, then spun for 30 s at the constant rate before any head tilts were The second ending criteria was included in the event that continued performed. As done previously , each subsequent session’s initial spin acclimation was not possible with continued exposure—that subjects’ rate was 1 RPM less than the spin rate in which the subject first reported ending thresholds (i.e., fastest spin rate at which no illusion was felt at the feeling the illusion on the prior session. Substantial effort was put forth to end of a session) reached a plateau. If this was the case, we did not want to train individual subjects at roughly the same time each day (i.e., within the continue testing individuals when additional acclimation would not be same 2–3 h), although subject availability ultimately dictated scheduling. possible. Finally, if subjects did not reach a beginning threshold of 25 RPM or a Catch trials. As subjects’ responses defined the staircase, a lack of proper plateau in ending threshold, we tested them for 50 total acclimation reporting (in terms of subjects not responding truthfully or reliably sessions. These ending criteria and testing duration allowed us to test our regarding their perception of a presence or absence of the illusion) could hypotheses regarding extended acclimation, working towards a better result in inaccurate or incomplete results. To assess the reliability of subject npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA K.N. Bretl and T.K. Clark understanding of the capacity of subjects to acclimate to the CC illusion, of sessions, where t refers to a time at which at least one subject reached thus informing centrifuge conceptual design. the desired spin rate, d references the total number of subjects reaching the desired rate at that time, and n is the number of subjects who are still being tested and have not yet acclimated to the given spin rate. The Motion sickness monitoring. To verify that our protocol facilitated benign complement of that percentage provides a calculation of the probability acclimation to the CC illusion, we asked subjects to verbally report their that subjects would reach the desired spin rate over t days of training, subjective motion sickness rating (MSR) once every 5 min of each 25-min assuming the adoption of our protocol exposing subjects to one, 25-min acclimation session. Motion sickness scores were to be reported on the 6–9,11–13,17 session per day. For each desired spin rate, a staircase cumulative survival simple but commonly used scale of 0–20 , where a score of 0 is function (i.e., estimated population acclimation curve) can be plotted, used to convey no sense of motion sickness, and a score of 20 represents showing how the probability of reaching the desired ending threshold the subject feeling as though he/she is on the verge of vomiting. increases as the number of training sessions increases. These developed If at any point subjects reported that they were feeling ill or wanted to spin down (for motion sickness reasons or otherwise), the testing operators curves can be used to aid in the conceptual design and development of a spaceflight centrifuge by ensuring tolerability of the selected design. did so. Additionally, if a subject reported an MSR of 10/20 or higher, the day’s session would prematurely conclude. If the same subject reported 10/20 or higher on a subsequent session, he/she would not continue the Reporting Summary study. This motion sickness-based ending criterion was implemented to Further information on research design is available in the Nature Research prevent excessive motion sickness, though as discussed in the Results, no Reporting Summary linked to this article. subject reported motion sickness greater than 10/20 more than once. Data and analysis DATA AVAILABILITY Metrics and variables of interest. We extracted similar metrics as in our The raw minimal datasets for this study have been made publicly available initial 10-day acclimation study , as both investigations sought to evaluate (https://osf.io/zw6xe/). We request citing this paper when using these datasets for the tolerability and feasibility of acclimation. To assess feasibility of further analysis. acclimation, we calculated beginning and ending threshold for each session. From these, we calculated a linear acclimation rate (a measure of Received: 10 April 2020; Accepted: 7 July 2020; the amount of acclimation achieved per session of testing). To quantify tolerability, we calculated the maximum and average motion sickness rating throughout all sessions and subjects. Each individual’s Motion Sickness Susceptibility Questionnaire (MSSQ) percentile was correlated with reported motion sickness levels. With these variables, we were primarily interested in determining the REFERENCES long-term potential of subjects to acclimate to the CC illusion, and if the 1. Clement, G., Bukley, A. & Paloski, W. Artifiical gravity as a countermeasure for protocol changes implemented in the extended acclimation study (i.e., mitigating physiological deconditioning during long-duration space missions. updated staircase and addition of catch trials) had a significant impact on Front. Syst. Neurosci. 9,1–11 (2015). how subjects were acclimating. 2. Lackner, J. R. & DiZio, P. Rapid adaptation to coriolis force perturbations of arm trajectory. J. Neurophysiol. 72, 299–313 (1994). Statistical tests. We performed statistical tests in MATLAB and R/RStudio. 3. Diaz-Artiles, A., Heldt, T. & Young, L. R. Short-term cardiovascular response to The assumption of normality was verified with Anderson–Darling and short-radius centrifugation with and without ergometer exercise. Front. Physiol. 9, Shapiro–Wilks tests, and F-tests were used to verify equality of variance. 1–16 (2018). Comparisons between groups were performed using two-tailed, two- 4. Duda, K. R., Jarchow, T. & Young, L. R. Squat exercise biomechanics during short- sample or paired t-tests with either equal or unequal variances (based on radius centrifugation. Aviat. Space Environ. Med. 83, 102–110 (2012). the result from the associated F-test). In the event that the dataset failed the 5. Guedry, F. E. & Montague, E. K. Quantitative evaluation of the vestibular coriolis normality tests, nonparametric tests were utilized (Mann–Whitney–Wilcoxon reaction. Aerosp. Med. 32, 487–500 (1961). test). Finally, Spearman rank nonparametric correlation tests were used to 6. Sheehan, S. E., Young, L. R. & Jarchow, T. The effect of head turn velocity on cross- measure potential association between subjects’ reported motion sickness coupled stimulation during centrifugation. J. Vestib. Res. 18,1–14 (2008). levels and their pretest MSSQ percentiles. A required level of significance of 7. Jarchow, T. & Young, L. R. Adaptation to head movements during short radius α ¼ 0:05 was used for all statistical tests. centrifugation. Acta Astonautica 61, 881–888 (2006). 8. Young, L. R., Sienko, K., Lyne, L. E., Hecht, H. & Natapoff, A. Adaptation of the Use of survival analysis to develop design tool. To investigate the long- vestibuo-ocular reflex, subjective tilt, and motion sickness to head movements term outlook of acclimation—namely, the expected ability of subjects to during short-radius centrifugation. J. Vestib. Res. 13,65–77 (2003). acclimate to certain operationally relevant spin rates over a given number 9. Young, L. R. et al. Artificial gravity: head movements during short-radius cen- of testing days—we applied survival analysis to the collected data. To trifugation. Acta Astronautica 49, 215–226 (2001). specifically accommodate subjects who did not complete the full 10. Hecht, H., Brown, E. L. & Young, L. R. Adapting to artificial gravity (AG) at high investigation or did not reach a spin rate of operational interest, this rotational speeds. J. Gravitational Physiol. 9,1–5 (2002). statistical method was utilized to project the censored data forward in 11. Brown, E. L., Hecht, H. & Young, L. R. Sensorimotor aspects of high-speed artificial time . Survival analysis works to critically evaluate the time it takes for an gravity: I. Sensory conflict in vestibular adaptation. J. Vestib. Res. 12, 271–282 event to occur. The event of interest in the current investigation is reaching (2002). a given spin rate threshold of operational relevance, such as that required 12. Elias, P., Jarchow, T. & Young, L. R. Incremental adaptation to yaw head turns to reach a desired loading level with a given centrifuge size (e.g., 15 RPM during 30 RPM centrifugation. Exp. Brain Res. 189, 269–277 (2008). for an 8-m diameter centrifuge). The ultimate goal with this analysis is to 13. Cheung, C. C., Hecht, H., Jarchow, T. & Young, L. R. Threshold-based vestibular estimate a population acclimation curve from our sample. adaptation to cross-coupled canal stimulation. J. Vestib. Res. 17, 171–181 (2007). We calculated the Kaplan–Meier survival analysis estimate , which 14. Bretl, K. N. et al. Tolerable acclimation to the cross-coupled illusion through a 10- quantifies an approximation of “survival”, or the time until the event of day, incremental, personalized protocol. J. Vestib. Res. 29,97–110 (2019). interest occurs. 15. Bretl, K. N., Sherman, S. O., Dixon, J. B., Mitchell, T. R. & Clark, T. K. A standardized, incremental protocol to increase human tolerance to the cross-coupled illusion. J. S ¼ 1  Vestib. Res. 29, 229–240 (2019). t (1) tit 16. Bretl, K. N. et al. Retention of cross-coupled illusion training to allow for a short radius space centrifuge. in IEEE Aerospace Conference (2018). 17. Garrick-Bethell, I., Jarchow, T., Hecht, H. & Young, L. R. Vestibular adaptation to The survival rate is expressed with the survival function, S , which is the centrifugation does not transfer across planes of head rotation. J. Vestib. Res. 18, proportion of individuals surviving longer than time t out of the total 25–37 (2008). number of individuals studied at that time. The product limit method, 18. Evanoff, J., DiZio, P. & Lackner, J. R. The influence of gravitoinertial force level on unique to the Kaplan–Meier analysis, can be seen in Eq. 1. For our oculomotor and perceptual responses to coriolis, cross-coupling stimulation. application, the survival function uses collected data to estimate the Aviat. Space Environ. Med. 58, 218–223 (1987). probability of not reaching the desired spin rate of interest over t number Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA npj Microgravity (2020) 22 K.N. Bretl and T.K. Clark 19. Lackner, J. R. & Graybiel, A. The effective intensity of coriolis, cross-coupling completed by K.N.B. and T.K.C. Both authors edited and approved the final stimulation is gravitoinertial force dependent: Implications for space motion manuscript. sickness. Aviat. Space Environ. Med. 57, 229–235 (1986). 20. Lackner, J. R. & Graybiel, A. Influence of gravitoinertial force level on apparent magnitude of coriolis cross-coupled angular accelerations and motion sickness. COMPETING INTERESTS Mech. Predic. Prev. Treat. 22-1–22-7 (1984). The authors declare no competing interests. 21. Graybiel, A., Miller, E. F. & Homick, J. L. Individual differences in susceptibility to motion sickness among six Skylab astronauts. Acta Astronautica 2,155–174 (1975). 22. Lackner, J. R. Motion sickness: more than nausea and vomiting. Exp. Brain Res. ADDITIONAL INFORMATION 232, 2493–2510 (2014). Supplementary information is available for this paper at https://doi.org/10.1038/ 23. Graybiel, A., Miller, E. F. & Homick, J. L. Biomedical Results from Skylab (National s41526-020-00112-w. Aeronautics and Space Administration, 1977). 24. Schubert, M. C. & Migliaccio, A. A. New advances regarding adaptation of the Correspondence and requests for materials should be addressed to K.N.B. vestibulo-ocular reflex. J. Neurophysiol. 122, 644–658 (2019). 25. Cakit, B. D., Saracoglu, M., Genc, H. & Erdem, H. R. The effects of incremental Reprints and permission information is available at http://www.nature.com/ speed-dependent treadmill training on postural instability and fear of falling in reprints Parkinson’s disease. Clin. Rehab. 21, 698–705 (2007). 26. Reason, J. T. Relations between motion sickness susceptibility, the spiral after- Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims effect and loudness estimation. Br. J. Psychol. 59, 385–393 (1968). in published maps and institutional affiliations. 27. Miller, R. G. Survival Analysis. (Wiley-Interscience, 1998). 28. Goel, M. K., Khanna, P. & Kishore, J. Understanding survival analysis: Kaplan–Meier estimate. Int. J. Ayurveda Res. 1, 274–278 (2010). Open Access This article is licensed under a Creative Commons ACKNOWLEDGEMENTS Attribution 4.0 International License, which permits use, sharing, This work was supported by a NASA Space Technology Research Fellowship, grant adaptation, distribution and reproduction in any medium or format, as long as you give number 80NSSC17K0085. We thank all of the subjects who participated in the appropriate credit to the original author(s) and the source, provide a link to the Creative investigation, as well as Carson Brumley, Sebastian Metcalf, Varun Seth, and Marcos Commons license, and indicate if changes were made. The images or other third party Mejia for helping operate test sessions. Preliminary results were presented at the material in this article are included in the article’s Creative Commons license, unless following conferences and scientific meetings: Aerospace Medical Association indicated otherwise in a credit line to the material. If material is not included in the Scientific Meeting, Dallas, TX, 2018; Human Research Program Investigators’ article’s Creative Commons license and your intended use is not permitted by statutory Workshop, Galveston, TX, 2019. regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons. org/licenses/by/4.0/. AUTHOR CONTRIBUTIONS Experiments were designed and executed by K.N.B. and T.K.C., data analysis was © The Author(s) 2020 conducted by K.N.B., and data interpretation and manuscript preparation were npj Microgravity (2020) 22 Published in cooperation with the Biodesign Institute at Arizona State University, with the support of NASA

Journal

npj Microgravity – Springer Journals

Published: Aug 26, 2020

There are no references for this article.