Changes in Risky Behavior in Long-Term Head-Down Bed Rest and Relation to Psychological Status
INTRODUCTION: The present study aimed to investigate changes in risky behavior in a sample of 36 healthy men during a 90-d head-down bed rest (HDBR) experiment and examined whether psychological factors—general self-efficacy, stress, and recovery—could influence these changes. METHODS: Subjects completed the Balloon Analog Risk Task (BART) and two psychological scales once during the acclimation period, six times during the HDBR period, and twice during the recovery period. During the HDBR period, subjects were required to maintain a −6° head-down position for most daily activities and only permitted to change positions around the longitudinal axis of their bodies. RESULTS: The results demonstrated that subjects’ risk-taking behaviors were significantly affected by bed rest, with an increased propensity to engage in risky activities during the head-down stage. In addition, BART scores did not return to baseline when subjects entered the recovery stage. In terms of psychological variables, the results indicated that scores of general self-efficacy and recovery were negatively correlated with BART indicators, while stress levels were positively correlated with risky behaviors. Compared to other psychological variables, the perceived physical stress, including fatigue, somatic complaints, and sleep quality, exhibited the strongest correlations with BART indicators. DISCUSSION: The findings of this study implied that prolonged exposure to a simulated microgravity environment and confined isolation conditions may have a sustained impact on risk-taking tendencies, with changes in risky behaviors in the head-down state more closely associated with physiological symptoms. He X, Lei Y, Xu Z, Li K, Nicolas M, Wu R, Li Y. Changes in risky behavior in long-term head-down bed rest and relation to psychological status. Aerosp Med Hum Perform. 2025; 96(4):304–313.
As the duration of manned space missions continues to increase, astronauts are forced to face more complex and uncertain tasks during their space travels. Therefore, the crew’s ability to effectively make risky decisions in extreme environments is crucial in guaranteeing their safety and the successful accomplishment of their missions. Risky decision-making is closely connected to complicated cognitive processes and emotional components. Due to the randomness of multiple choices and outcomes under uncertain conditions, individuals are required to carefully assess and evaluate risk information and potential consequences of decisions. Risky decision-making can be influenced by a variety of physiological and psychological factors. Previous studies have shown that sleep restriction or deprivation can lead to an increase in individuals’ risk-taking behavior when performing the Balloon Analog Risk Task (BART),1 suggesting that sleep duration is related to risk perception and risk tolerance.2 In addition, negative emotions and stress perceived by individuals can also influence their risky decision-making tendencies and strategies.3,4
In the extreme environment of spaceflight, stressors such as weightlessness, noise, hypoxia, circadian rhythm (biological clock) disruption, and confined isolation can cause a series of negative emotions which in turn may affect the decision-making process of the crew.5 As one of the most intense stressors encountered by astronauts, the impact of microgravity on decision-making behavior has also received considerable attention from researchers. Although weightlessness in the space environment is difficult to fully replicate in the laboratory, the −6° head-down bed rest (HDBR) experiment has been proved to be a relatively effective method of inducing the possible changes in various aspects of astronauts during spaceflight.6,7 It can not only simulate the physiological and psychological effects of headward fluid shift caused by weightlessness, but can also affect an individual’s behavior and decisions due to limited social contact with the external environment.8 Previous studies have shown that long-term exposure to HDBR conditions may damage the normal psychological functions, emotions, and decision-making abilities of astronauts.9 In the HDBR experiment, individuals’ behavior and emotional state will alter to a certain extent because of the restriction of physical movement5 and isolation from their familiar living environment.8 However, research on the potential effect of simulated microgravity on risky decision-making is still limited, and no consistent findings have been obtained. For instance, Lipnicki et al. revealed that subjects showed an increased propensity to engage in high-risk activities and had longer reaction time latencies, indicating that long-term bed rest could have negative effects on executive function.10 In contrast, Rao et al. reported no significant changes in the subjects’ behavior on the BART following 45 d of HDBR, suggesting that risky decision-making may not be affected by simulated microgravity.11 Since the length of spaceflight will have a complex effect on astronauts’ psychological and cognitive performance, it is necessary to conduct long-term (more than 6 wk) simulated microgravity experiments to explore the influencing factors and change mechanism of risky decision-making behavior.
Conditions in extreme environments can impose physical, psychological, and social limitations on individuals and may constitute high-stress situations that require adaptation.12 Previous research has demonstrated that the crew will experience different stress responses and negative emotions at distinctive stages in microgravity.13 Although stress has received extensive attention in space missions, recovery as an essential variable associated with individual performance in extreme environments is overlooked in simulation experiments.7 Similar to the effects of excessive stress, lack of recovery may impair astronauts’ performance on work tasks by inducing emotional, cognitive, and behavioral disorders, which may influence their adaptation to the environment.14 Therefore, the crew’s ability to proactively balance the level of stress and recovery in extreme conditions and to counteract the negative effects of stress through their own recovery capabilities is crucial to the safety of spaceflight. A 60-d HDBR experiment examined the effects of a simulated microgravity environment on stress and recovery in 16 healthy women.7 The findings indicated that simulated weightlessness may induce psychological stress and alter the perception of recovery. However, previous studies have not clarified the relationship between stress-recovery status and decision-making ability. In view of this, it is necessary to concentrate on the trend of stress and recovery levels in long-term HDBR experiments, as well as to understand the effects of changes in perceived stress and recovery on risk-taking propensities.
Self-efficacy refers to an individual’s perception or belief in his or her own capability in the face of a challenging environment,15 and it is significantly associated with negative emotions such as depression and anxiety.16 Studies have confirmed a strong correlation between self-efficacy and risk-taking decisions in a variety of domains. It has been demonstrated that a high level of self-efficacy has a positive impact on unpredictable and challenging tasks.17 During HDBR, subjects’ self-efficacy may fluctuate over time due to sudden changes in their life circumstances or enhanced negative emotions caused by stressful tasks or confinement during the experiment. Nonetheless, some researchers have suggested that self-efficacy may remain stable when people are confronted with challenging events or in a stressful period.18 For example, a study showed that individual’s self-efficacy during confinement was relatively stable and positively correlated with positive emotions and adaptive work performance.19 Therefore, further validation is required to determine whether self-efficacy is a stable psychological trait when in a prolonged state of discomfort and confinement during a HDBR experiment, and how it relates to risky decision-making.
Although a few previous studies have investigated the possible changes in risk-taking propensity during spaceflight, the trend in risky decision-making abilities of astronauts in long-term HDBR conditions and its relationship with psychological factors have still not received sufficient attention, and consistent conclusions have not been reached. Therefore, the study of individual changes in risky decision-making in the long-term HDBR experiment is an effective reference for predicting the impact of long-duration spaceflight on decision-making abilities and the risk propensity of the crew. As the −6° HDBR has been demonstrated to be one of the most effective ground-based models for simulating weightlessness, the present study adopted this condition as the experimental setup. The main objectives of this study were to: 1) investigate the changes in risky behavior at different stages of the 90-d −6° HDBR experiment; 2) examine the impact of long-term simulated microgravity and isolation conditions on self-efficacy, perceived stress, and recovery, and test whether these psychological factors are stable traits or fluctuating states in extreme environments; and 3) explore the relationships between risky decision-making and psychological factors at different experimental stages.
The following hypotheses are proposed by combining the results of previous studies with the objectives of the current investigation:
H1: Risk-taking tendencies would vary significantly across the different experimental stages of 90-day HDBR, with a significant increase in risky behavior during the HDBR stage.
H2: Long-term simulated microgravity and isolation conditions will have a negative impact on self-efficacy, perceived stress, and recovery scores over time.
H3: Risky decision-making behavior can be influenced by psychological factors, especially during the head-down state. Specifically, lower self-efficacy and recovery, along with higher perceived stress, will be associated with increased risk-taking tendencies during the HDBR stage.
METHODS
Subjects
The study protocols were approved by the Institutional Review Board of the China Astronaut Research and Training Center according to the principles of the 2013 Declaration of Helsinki (IRB approval number: ACC201904). Subjects were informed of the purpose, risk, and potential adverse effects of this experiment before signing the informed consent form. They were free to withdraw from the study at any time and received financial rewards for their participation.20
A total of 36 healthy adult men, ages between 23–45 yr (Mean = 31.39, SD = ±5.03), were recruited to participate in a 90-d HDBR experiment held in Shenzhen. All subjects had an education level of junior high school or above. The selection process involved physical and psychological examinations to exclude psychiatric history, psychological disorders, substance abuse, organic diseases, and infectious disease history, and subjects did not have any other potential physical and/or psychological disorders which were unsuitable for this experiment. The current study was limited to men in order to avoid potential physical discomforts and inconveniences associated with the menstrual cycles of women of a similar age during the 90-d head-down state. All subjects successfully completed the experiment, including the experimental task and questionnaires.
Materials
The BART is a computer-based measure that captures an individual’s dynamic risky decision-making process and changes in the propensity to take risks by simulating a decision-making task similar to a real-world situation.21 During the task, a simulated balloon and a button labeled “Inflate” are displayed on the computer screen. Subjects can temporarily accumulate ¥0.05 for each pump. Once the balloon is inflated to a point exceeding a certain point, it will explode and all gains on the trial is lost. Subjects can stop pumping the balloon at any point during each trial and click the button labeled “Collect ¥¥¥” to convert the temporary gain into a permanent gain. After each explosion or money collection, a new balloon will appear until all trials are completed. The computer screen also displays additional relevant information, including the temporary income on the current balloon, the total number of trials and the number of the current round, the accumulated inflation times of this trial, and the permanent money earned up to the last balloon. Therefore, the task requires subjects to make successive decisions and consider the two options and possible outcomes at each inflation: 1) keep pumping the balloon, increasing the likelihood of gaining more but also facing a greater risk of losing all the gains (balloon explosion); 2) stop pumping the balloon, decreasing the likelihood of gaining more but preserving the existing money.
The experiment consisted of a total of 30 trials and each balloon could be inflated up to 128 times. Subjects were informed that each balloon would explode randomly between the first and 128th inflations. Actual monetary rewards were adopted to ensure that the task maximized the stimulation of subjects. The indicators commonly used in BART were chosen for the following analysis to represent an individual’s level of risk preference: the total number of pumps, the total and average number of adjusted pumps (only for trials in which subject collected the money before the balloon exploded), and the total number of exploded balloons. Subjects completed a total of nine BART throughout the whole experiment, including one during the acclimation period, six during the HDBR period, and two during the recovery period.
The General Self-Efficacy Scale was designed to assess individuals’ perception and overall confidence in their ability to cope with stress and challenges in different situations on a 4-point Likert scale (1 = Not at All True; 4 = Very True). In this research, the Chinese version of The General Self-Efficacy Scale, which has been shown to have good reliability (α = 0.87), was adopted.22 The score of this scale was expressed as the sum of the values of 10 items, with higher scores indicating greater general self-efficacy. The alpha coefficient indicating the internal consistency of this measure within the current sample was 0.93. Subjects completed the scale nine times throughout the entire experiment, including once during the acclimation period, six times during the HDBR period, and twice during the recovery period.
The Recovery-Stress Questionnaire (RSQ)23 was originally designed to assess the balance between stress and recovery states in athletes, which could reflect an individual’s ability to recover from physical or mental stress in specific situations. In this study, the two general dimensions which have been shown to have good reliability in previous studies24 were adopted. The General Stress dimension score corresponds to the sum of all seven stress subscale scores: general stress, emotional stress, social stress, conflict/pressure, fatigue, lack of energy, and somatic complaints. The General Recovery dimension score corresponds to the sum of all five recovery subscale scores: success, social relaxation, somatic relaxation, general well-being, and sleep quality. Since a Chinese version of RSQ was unavailable, the scale was independently translated by two psychology undergraduates and translated back into English. Then the translated scale was reviewed by two professors to ensure its accuracy. A 6-point Likert scale was used with values ranging from 1 (never) to 6 (always), indicating how frequently the subjects experienced each situation during the last week. The internal consistency for General Stress and General Recovery was examined using Cronbach’s alpha coefficient and was deemed acceptable with 0.89 and 0.91, respectively. Subjects completed the scale nine times throughout the whole experiment, including once during the acclimation period, six times during the HDBR period, and twice during the recovery period.
Procedure
The experiment was divided into three phases: an acclimation period (15 d), a HDBR period (90 d), and a recovery period (33 d). During the HDBR period, subjects were required to maintain a −6° head-down position for the majority of their daily activities. They were only permitted to change positions around the longitudinal axis of their bodies. Social support activities were carried out regularly during the whole experiment and targeted psychological support was provided according to the status of the subjects.
Nine test points were selected to measure the behavioral and psychological indicators of the subjects: day 10 of the acclimation period (P10); days 15, 29, 42, 56, 69, and 83 of the HDBR period (B15, B29, B42, B56, B69, B83); and days 5 and 25 of the recovery period (R5, R25). Fig. 1 depicts the specific arrangements.
Citation: Aerospace Medicine and Human Performance 96, 4; 10.3357/AMHP.6567.2025
Statistical Analysis
Data analysis was conducted using SPSS version 29.0 (IBM, Armonk, NY, United States). Since the results of the Shapiro-Wilk Test indicated that some of the variables in this study did not follow a normal distribution, nonparametric methods were applied for statistical analysis. The Friedman Test and Wilcoxon Signed-Rank Test were used to investigate changes in BART indicators and psychological variables across the nine test points and different experimental stages. Spearman correlation analysis was employed to assess the associations among different variables over time. A value of P < 0.05 was considered to be statistically significant.
RESULTS
Fig. 2 presents the results of the BART experiment. The Friedman Test revealed that there were significant effects of time on the total number of adjusted pumps (χ2 = 21.469, P = 0.006), the average adjusted pumps (χ2 = 16.798, P = 0.032), and the total number of exploded balloons (χ2 = 22.121, P = 0.005). Meanwhile, the results proved there were no significant effects of time on the total number of pumps (χ2 = 11.571, P = 0.171). The post hoc analysis results of the Wilcoxon Signed-Rank Test showed that subjects performed more pumps at B42, B83, R5, and R25 than at P10. Moreover, they had a higher number of average adjusted pumps at B29, B42, B56, and R5 compared to P10. In addition, the number of exploded balloons was significantly higher at B15 compared to P10. Combined with the information in Fig. 2, it can be concluded that subjects showed a higher risk-taking tendency after entering the HDBR period, and that this increased level of risk-taking behavior did not fall back to its initial state during recovery.
Citation: Aerospace Medicine and Human Performance 96, 4; 10.3357/AMHP.6567.2025
The Friedman Test was conducted to further compare the differences in each BART indicator across the three experimental stages, with means and standard deviations summarized in Table I. The results indicated a significant effect of experimental stage on the total number of adjusted pumps (χ2 = 10.889, P = 0.004). Additionally, the total number of pumps showed a marginally significant difference across the stages (χ2 = 5.056, P = 0.080). To be more specific, the total number of adjusted pumps was lower in the acclimation stage than in the HDBR period, while it was lower in the HDBR period than in the recovery stage. Moreover, subjects had lower total pump counts during the acclimation phase than in the recovery phase, with no significant differences observed between the HDBR and recovery phases for this indicator. However, the results demonstrated no significant differences in the number of exploded balloons or average adjusted pumps among the three stages. Overall, these findings suggested that subjects appeared to have a significantly greater propensity to take risks after the start of the head-down state, with this increased risk-taking tendency persisting into the recovery phase.
Fig. 3 presents the trend of psychological indicators over time. The Friedman Test revealed significant effects of time on the score of general self-efficacy (χ2 = 20.222, P = 0.010), General Stress (χ2 = 20.360, P = 0.009), and General Recovery (χ2 = 16.244, P = 0.039). Post hoc analysis indicated that subjects had lower general self-efficacy scores at B15 and B69 compared to P10. Additionally, recovery scores at P10 were significantly higher than those at B15, B29, B42, and B69. However, the post hoc analysis revealed no significant differences in stress levels across the testing points. In order to explore the changing trend of psychological variables, a further analysis was conducted to assess the effects of the three experimental stages, with means and standard deviations summarized in Table II. The results showed that the experiment stage had no significant effects on scores for general self-efficacy, general stress, or general recovery. The post hoc analysis revealed that subjects’ scores of general self-efficacy and recovery during the HDBR stage were lower than those in the acclimation stage. In addition, no significant differences in stress levels were found among the three stages. It can be concluded from the above results that subjects’ psychological states remained relatively stable throughout the three experimental phases. However, participants showed a slight increase in stress levels during the head-down tilt phase, along with a slight decrease in self-efficacy and recovery scores. Furthermore, the levels of the three psychological variables did not fully return to baseline by the end of the experiment.
Citation: Aerospace Medicine and Human Performance 96, 4; 10.3357/AMHP.6567.2025
The correlations among BART indicators and psychological variables in different experimental stages were tested separately by Spearman correlation analysis and presented in Table III. It can be found from the results that there were no significant correlations between BART indicators and the scores of psychological variables in the acclimation and recovery stages. In the HDBR stage, the general self-efficacy score was significantly correlated with the total number of pumps. In addition, the results also revealed that the scores of General Stress and General Recovery were significantly correlated with BART indicators when subjects were in the head-down state, except for the total number of adjusted pumps. These significant relationships between BART indicators and psychological variables in the HDBR stage suggested that subjects with less confidence in their abilities to cope with stress and challenges in a long-term simulated microgravity and isolation environment showed higher risk-taking propensities in behavioral experiments. Meanwhile, they were less able to recover from mental or physical stress in this extreme situation, resulting in stronger feelings of pressure. Compared to the scores of general self-efficacy and General Recovery, the score of General Stress had stronger correlations with BART indicators, which indicated that risky decision-making during simulated microgravity and isolation environment may be more closely associated with the perception of pressure.
As the scores of General Stress and General Recovery were significantly correlated with the total number of pumps, the average number of adjusted pumps, and the number of exploded balloons during the HDBR phase, the correlations between each BART indicator and the subdimensions of the RSQ were further examined to explore the specific factors most strongly associated with risky decision-making. According to the results presented in Table IV, fatigue, somatic complaints, general stress, emotional stress, and sleep quality were significantly correlated with the three indicators. In addition, lack of energy was significantly associated with the total number of pumps and the average number of adjusted pumps. Social stress was significantly associated with the average number of adjusted pumps and the number of exploded balloons. Fatigue, somatic complaints, and sleep quality had the highest correlation coefficients with BART indicators, indicating that the level of risky decision-making in a head-down state was more closely related to physiological symptoms.
DISCUSSION
The long-term HDBR environment had a significant effect on risky decision-making, and subjects exhibited a higher propensity for risk-taking during HDBR period than at baseline. In particular, the total and average number of adjusted pumps and the number of exploded ballons showed significant time effects. Furthermore, the results of the experiment indicated that the level of risk-taking did not return to baseline when the subjects entered the recovery period, suggesting that the long-term HDBR state may have a lasting influence on individuals’ risky decision-making. Meanwhile, there were significant correlations between individuals’ risk propensity during the HDBR phase and psychological indicators, with the physiological stress (fatigue, somatic complaints, and sleep quality) experienced by subjects during bed rest being the most closely related to risky behavior.
Overall, risk-taking propensity showed an upward trend in this experiment, with subjects showing more risky behaviors during the HDBR stage than the acclimation stage. Additionally, the experimental results demonstrated that the BART scores during the recovery stage were higher than during the other two phases, suggesting that individuals continued to make riskier decisions even after they ended the head-down state and returned to normal life.
In terms of psychological indicators, self-efficacy, stress, and recovery levels exhibited significant time effects. Compared to baseline, participants experienced notably lower self-efficacy in the early stages of HDBR. Additionally, recovery scores at baseline were significantly higher than at other testing points, suggesting that extreme conditions may have a sustained adverse impact on individuals’ recovery capacity. Furthermore, these findings indicated that changes in stress in a simulated microgravity and isolation environment were characterized by stages. However, the changes in psychological variables during the whole experimental process were moderate.
During the early stage of HDBR, subjects’ scores on various BART indicators significantly increased, particularly in the average number of adjusted pumps and the number of exploded balloons. As for psychological indicators, the stress score initially decreased and then increased during this stage, though there was no significant difference compared to baseline levels. The score of general self-efficacy significantly decreased at B15, while recovery levels at both B15 and B29 were significantly lower than baseline levels.
Due to the abrupt transition into the head-down state, that strong stressor can result in a significant decrease in individuals’ self-efficacy and their abilities to recover from negative emotions during the initial stage of this experiment. The discomfort and activity restrictions imposed by HDBR may cause negative emotional responses, sleep disturbances, and physiological discomfort, thereby reducing the subjects’ risk perception abilities, enhancing their willingness to engage in risky behaviors, and consequently increasing the probability of actual risk-taking behaviors. In addition, the changes in cognitive resources and psychological needs induced by HDBR and a confined isolation environment may impair individuals’ executive control functions and significantly reduce their positive emotions,9 increasing their propensity to engage in high-risk activities.10 The need to adapt the sudden change in the new environment and concerns about physiological discomfort caused by the experiment may also contribute to fluctuations in stress levels.
During the midstage of HDBR, subjects showed slight fluctuations in their scores on the BART indicators, with overall levels remaining above the baseline. Specifically, the total number of adjusted pumps at B42 was significantly higher than at P10, and the average number of adjusted pumps was significantly higher at both B42 and B56. In terms of psychological indicators, the scores of stress decreased and did not significantly differ from the baseline level. However, their recovery scores at B42 remained significantly lower than at P10.
The trends in experimental indicators during the later stage of HDBR were similar to those observed in the early stage. Subjects showed an increase in the total number of adjusted pumps at B69 and B83, with significantly higher scores at B83 than P10. Additionally, the perceived stress levels of subjects further increased, while the levels of self-efficacy and recovery showed slight fluctuations, with both two scores at B69 still significantly lower than baseline.
In the middle to late stages of HDBR, subjects’ risk-taking propensity remained at a relatively high level. During the entire experiment, subjects were required to be isolated from familiar living environments and social circles, confined to limited spaces, and had to complete monotonous and demanding experimental tasks for more than 3 mo. Therefore, the stimulation brought about by risky behaviors may serve as a countermeasure against the chronically monotonous environment.25 Although subjects were allowed to use mobile devices during the experiment, the prolonged separation from family and friends continued to enhance their sense of boredom in the experimental environment. Ma et al. found that individuals may be more inclined to make decisions in a way that could bring them the most pleasure without compromising the overall safety of the experiment.25 In a decision-making scenario similar to the BART experiment, the focus on negative consequences of the decision can be reduced because the options do not involve responsibility toward others.
Long-term HDBR is characterized by a lack of social contact and isolation from familiar environments, which can lead to stage-specific changes in psychological indicators. According to the four-stage model of emotional changes, after experiencing negative emotions caused by physiological discomfort in the first stage of the experiment due to the sudden entry into a simulated microgravity environment, individuals can gradually adapt to their physiological changes and enter a period of emotional stability in the second stage.26 However, individuals will experience a deterioration of their emotions as a result of prolonged bed rest, isolation from family, and the restrictive and monotonous nature of their lives in the third stage, which was consistent with the changes in stress levels experienced by the subjects in this experiment.
After entering the recovery period, the overall risky decision-making levels of subjects continued to show an increasing trend and had not returned to the initial level. The total and average number of adjusted pumps at R5, as well as the total number of adjusted pumps at R25, remained significantly higher than at P10. Meanwhile, the changing trend of psychological variables during the recovery period was similar to that during the middle of HDBR. The stress scores significantly decreased after ending the bed rest and returned to their pre-experiment levels. However, the overall recovery ability and self-efficacy of individuals at R5 and R25 was still lower than at P10.
Previous studies have shown that the duration of experiments can have complex effects on cognitive functions.9 In the long-term bed rest experiments, subjects’ abilities to adapt to extreme environments gradually declines, and advanced cognitive functions may experience sustained damage.27 In addition, as the subjects’ living environment and work tasks were in a transitional phase during the recovery period, they may need to invest more cognitive resources and emotional efforts to adapt to changing conditions, which may also be one of the reasons why the risky decision-making level did not immediately decline in the recovery stage.
On the other hand, the decrease in stress levels during the recovery period reaffirmed the stage-specific nature of emotional alterations. The four-stage model indicates that when individuals realize the experiment is about to end, they will develop a joyful and optimistic psychological state. Therefore, the changes in stress levels exhibited by individuals during the early, mid, and late stages of HDBR, and the recovery stage followed a “high-low-high-low” pattern. This result confirmed the findings of Qin et al., who observed a trend of “high-low-high-low” cycles in negative emotions among 21 healthy men in a 60-d HDBR period.28
In this experiment, general self-efficacy scores at B15 and B69 showed significant differences from baseline and were negatively correlated with risky decision-making during the HDBR stage. However, statistical results indicated that changes in general self-efficacy throughout the experiment were relatively minor, with only slight fluctuations over time. These findings suggested that general self-efficacy can be viewed as a relatively stable personality trait that supports astronauts in coping with microgravity and confinement during space missions. Researchers proposed that general self-efficacy can serve as a stable resource for individuals facing tense conditions or experiencing a period of stress.19 Higher self-efficacy can be viewed as a protective factor that could help individuals withstand the negative impacts of traumatic events. According to social cognitive theory, self-efficacy can positively influence decision-making behaviors through motivational, emotional, and cognitive factors, and a strong sense of capability can contribute to cognitive processes and the quality of decision-making in various environments.29 Individuals with high self-efficacy will make adventurous decisions within their perceived abilities, without blindly pursuing high-risk options.30 These experimental results have implications for future astronaut selection and task allocation during long-duration space missions. Individuals with high general self-efficacy are more likely to maintain a moderate risk preference in extreme environments, regulate negative emotions and cognitive impairments caused by stress, and better adapt to sudden environmental changes.
Prolonged bed rest and movement restrictions lead to changes in individuals’ psychological states, and the lack of privacy and separation from family members can contribute to the significant stress perceived by subjects.5 Furthermore, the physiological changes caused by simulated microgravity in the head-down state may result in various physiological stress responses such as headaches and sleep disorders.26 Existing research indicated that the cognitive functions of astronauts can be influenced by stress factors such as fatigue, the burden of adapting to the environment, and sleep disturbances.25 The stress perceived by individuals may induce depletion of cognitive resources and an imbalance in psychological requirements, leading to a deterioration of executive functions and an increase in reward seeking. Both changes will not only result in an overvaluation of risky options and a decrease in risk perception, but also adversely affect the choice of decision-making strategies. On the other hand, the increased amygdala activity caused by stress can induce avoidance of regret emotions, prompting individuals to focus more on monetary rewards and neglect the potential negative consequences of risky behaviors, thereby increasing their risk-taking propensity.31
The results of the correlation analysis suggested that the risky decision-making level during the HDBR stage was more closely related to physiological stress (fatigue, somatic complaints, and sleep quality). Previous experiments have proved that poor sleep quality could affect an individual’s risk perception and increase their willingness to take more risks.1 Individuals with poorer sleep quality showed a significant decrease in attention allocation and self-control, and therefore exhibited an increased willingness to engage in risky activities.32 Ma et al. revealed that individuals may experience sleep disorders as a result of changes in their biological rhythms, which in turn reduced their risk perception and performed more risk-taking behavior.25 Furthermore, a 45-d HDBR study found that subjects exhibited higher levels of somatic anxiety during bed rest, while some mood-related anxieties were not affected by the experimental condition.33 Since the experiment is mainly based on the fact that the changes in physiological conditions which HDBR induces are very similar to the effects of microgravity, it may have a more significant effect on the stress and discomfort experienced by individuals at the physiological level.
On the other hand, the results of the current experiment showed that HDBR can result in a significant decrease in individuals’ recovery levels, validating the conclusion that one of the primary manifestations of the effects of microgravity on emotions is a significant decrease in positive emotions.9 In addition, the results of data analysis demonstrated a significant negative correlation between recovery levels and risk-taking propensity during the HDBR phase, and the subjects continued to exhibit diminished recovery abilities after the head-down state ended. Previous research has revealed that insufficient recovery ability can disrupt individuals’ performance and result in emotional, cognitive, and behavioral disturbances.14 Moreover, emotional and cognitive impairments, as well as a decline in decision-making quality can, in turn, affect individuals’ adaptation to extreme environments and the recovery of psychological resources.7 Since subjects need to readjust to changes in their environment and tasks during the recovery period, there is an increased consumption of cognitive and psychological resources, which may prevent recovery abilities from returning to their pre-experiment levels. However, further research is still needed to determine whether these extreme conditions involved in the long-term bed rest experiment cause lasting damage to the recovery abilities of individuals.
In summary, this experiment explored the change of risky decision-making behaviors in long-term HDBR and its relationship with psychological factors. The results of the experiment indicated that subjects exhibited more risk-taking behaviors during the HDBR phase. In addition, risk-taking propensity was negatively correlated with stress, particularly physiological stress, and positively correlated with general self-efficacy and recovery. Although individuals’ general self-efficacy, stress, and recovery levels were influenced by the head-down state, their psychological indicators changed moderately over the course of the whole experiment. This may be because countermeasures such as social and professional psychological support provided during the experiment attenuated the negative effects of the long-term simulated microgravity and isolation-restricted environment on subjects’ psychological states. Exploring the changes and mechanisms of risky decision-making in long-term HDBR as well as their influencing factors is important for future crew selection and task assignment for long-term space missions.
Time arrangement in 90-d HDBR experiment. HDBR: head-down bed rest.
The trend of BART scores during the experiment. *P < 0.05, as compared with P10. BART: Balloon Analog Risk Task.
The trend of psychological variables during the experiment. *P < 0.05, as compared with P10.
Contributor Notes
