Running head: ISOLATED IN A MOON ANALOG HABITAT
Social isolation in space:
An investigation of LUNARK, the first human mission in an Arctic Moon analog habitat
Paolo Riva1, Patrice Rusconi2, Luca Pancani1, Konstantin Chterev3
1 University of Milano-Bicocca, Italy
2 University of Messina, Italy
3 University of Surrey, UK
Total Word Count: text (10.457 words).
Correspondence concerning this article should be addressed to Paolo Riva, University of
Milano-Bicocca, Department of Psychology, Piazza Ateneo Nuovo, 1, 20126 – Milano (Italy).
Running head: ISOLATED IN A MOON ANALOG HABITAT 1
Research to address the technical challenges of human missions into space is growing. Knowledge
about the social-psychological aspects of individuals’ experiences of confinement within habitats in
space missions or extreme environments is also rapidly expanding. Social isolation is among one of
the best-known risk factors in these environments. This study focuses on the relationship between
time spent in specific activities (e.g., talking about personal matters) and the social-psychological
effects of social isolation and confinement as a part of the LUNARK project, which was aimed at
building and testing the first Moon analog habitat. Two space architects took part in a 61-day
mission in Northern Greenland to simulate human life conditions in the habitat as a prototype of a
human settlement on the Moon. The two crew members independently filled out a time-based diary
with self-report measures on their daily activities and negative emotions, feelings of loneliness,
resignation, desire for social contact, and time perception. First, our results showed that, for either
space architect, desire for social contact increased over time, whereas feelings of resignation did
not. Moreover, the protective role of specific daily activities emerged. Talking about personal
matters and leisure time were associated with a decrease in resignation, whereas talking about
personal topics and physical exercising increased the desire for social contact. Finally, engaging in
leisure activities increased the perceived speed of time. We discussed these results referring to
research on the consequences of long-term social isolation in extreme human expeditions and social
psychological models of social isolation. (249 words)
Keywords: Social Isolation, LUNARK mission, Resignation, Space Psychology, Extreme
Running head: ISOLATED IN A MOON ANALOG HABITAT 2
With the opening to the private market, space travel will likely become more frequent in the
next few decades. NASA has planned the human return to the Moon in 2024 with the Artemis III
mission (GAO, 2021). In addition, several government agencies (e.g., in China, India) are joining
the space travel race, and several private companies and space entrepreneurs are turning their
attention to this possibility, which brings about new challenges (Profitiliotis & Loizidou, 2019).
With the widening of space actors involved, more studies are needed to focus on the mental
consequences of participating in a long-duration mission in an extreme and confined environment
and how to promote psychological adaptation and well-being besides monitoring and fostering
physical health during the mission.
Several authors have highlighted the possible stressors associated with a mission in an
extreme environment, such as working under stress, physical discomfort and physical changes, and
long-term social isolation (Johannes & van Baarsen, 2020). This latter factor is the main focus of
this work. Given the universal human need for belonging (Baumeister & Leary, 1995), how do
individuals psychologically react to the stressors of long-term experiences of social disconnection
that characterize a space mission?
This paper will illustrate the LUNARK project (https://saga.dk/projects/lunark), the first
attempt to design and test a Moon analog habitat in the Arctic, which provided us with a unique
context to investigate the psychological mechanisms triggered by social isolation in an ecologically
valid setting and over time. We will first review the social psychology literature on social isolation
to highlight a gap related to understanding the experiences of individuals who undertake missions
that require social isolation. We will then consider what previous space psychology literature has
revealed on social isolation, mental health, and well-being. Within this theoretical framework, we
will present a study included in the LUNARK project, which allowed us to monitor two space
architects’ changes over time regarding some critical psychological constructs related to the
perception of being isolated during a simulated mission in a confined and extreme environment.
Running head: ISOLATED IN A MOON ANALOG HABITAT 3
1.1. The LUNARK project
Long-term social isolation has holistic effects (Johannes & van Baarsen, 2020). It can induce
psychological changes such as loneliness due to the lack of social alternatives, particularly in small
teams. Loneliness can have adverse effects from a psychological point of view, but it can also harm
individuals’ physical health, including the cardiovascular, immune, and nervous systems (Johannes
& van Baarsen, 2020). Our study addressed the psychological effects of social isolation in a dyad
involved in the LUNARK project, a simulation of a human lunar mission in a confined, extreme
environment. Two space architects (henceforth space architects A and B) lived inside a Moon
analog habitat in Northern Greenland for 61 days between September and December 2020 (the total
expedition duration was 89 days inclusive of the habitat build time). An analog is an environment
on Earth that reproduces the conditions of an astronomical body such as Mars. Analogs are used to
simulate the constraints on a body outside Earth to conduct research, test prototypes and
technology, collect data on human beings’ biological and psychological responses and prepare
human space expeditions. Previous studies on space analogs include the “AustroMars” expedition, a
simulated mission to Mars (Groemer et al., 2010), and the Mars-105 and Mars-500 simulations
(Gushin et al., 2012; Johannes & van Baarsen, 2020; Šolcová et al., 2014; van Baarsen et al., 2009;
van Baarsen et al., 2012; Wang et al., 2014). The LUNARK project is the first Arctic simulation of
a lunar mission.
The LUNARK’s Moon analog habitat was a pod designed to simulate the living conditions
of a lunar environment in a relatively realistic way (Figure 1). The expedition site in Northern
Greenland was chosen due to some features that could mimic those of the lunar South Pole, the
Peak of eternal light, where humans might first settle in future expeditions due to the almost
constant presence of sunlight and the potential presence of water ice deposits (Horneck et al., 2003).
The wind chill could reach -41°C, several weeks of light alternate with several weeks of darkness
throughout the year, and monochrome landscape. This extreme environment is isolated and,
differently from many previous simulations that were conducted in safe settings, it entails real
Running head: ISOLATED IN A MOON ANALOG HABITAT 4
danger due to the hostile presence of polar bears (indeed, the crew members recorded one polar bear
visit during the mission as they noticed fresh tracks in front of the habitat door).
The two space architects were limited in communicating with the outside world because
they had no Internet access. However, they could use a satellite phone to send daily messages (up to
160 characters) to the headquarters in Copenhagen, Denmark
(https://saga.dk/projects/lunark/expedition). They were involved in several activities and 15
research projects to measure their biological and psychological reactions to the stress of a simulated
lunar mission. Several parameters were recorded, including communication data (e.g., frequency),
data about the external environment (e.g., temperature, humidity, light), and biometric data (e.g., the
two space architects’ heart rate, stress levels, and sleep). Our study focused on the self-report
measures filled out by the two space architects on their feelings and perceptions throughout the
Before considering the specificity of the consequences of social isolation in extreme
environments and space missions, we will briefly review the general contribution of social
psychology to forms of isolation in everyday life.
1.2. The contribution of social psychology: Psychological consequences of social isolation
The social psychology literature, building on a long philosophical tradition (Aristotele, 4th
century BC), has highlighted that human beings possess a fundamental drive to form and maintain
at least a minimum number of interpersonal relationships that are lasting, positive (or at least not
negative), and meaningful, that is, they have a fundamental need for belonging (Baumeister &
Leary, 1995). According to these authors, the need for belonging is fundamental as it is not
secondary to any other human need. The need for belonging is universal, not specific to certain
If belonging is a fundamental psychological need, threats to the satisfaction of this need
must produce various adverse effects that involve thoughts, emotions, and behaviors. These threats
refer to multiple forms of social disconnection, including social isolation, loneliness, exclusion,
Running head: ISOLATED IN A MOON ANALOG HABITAT 5
ostracism, and rejection. Tomaka and colleagues (2006) defined social isolation as the objective
physical separation from others. Loneliness has been described as a “distressing feeling that
accompanies the perception that one’s social needs are not being met by the quantity or especially
the quality of one’s social relationships” (Hawkley et al., 2010). Social exclusion has been defined
as the experience of being kept apart from others physically (e.g., social isolation) or emotionally
(e.g., being ignored or told one is not wanted; Riva & Eck, 2016). Social rejection (being explicitly
told one is not wanted) and ostracism (being ignored) can be considered the two core experiences of
social exclusion. Isolation, loneliness, exclusion are typically distinguishable phenomena; however,
to date, it is still unclear whether and how the main psychological reactions experienced by a person
(e.g., negative emotions) differ between one and the other.
Accordingly, similar adverse outcomes are associated with different threats to social
belonging. Isolation and exclusion seem to trigger a wide range of negative emotions, including
anger, sadness, anxiety, guilt, and shame, just to name the main ones. Together with social pain
(e.g., the unpleasant feelings associated with actual and perceived social distance), these negative
emotions can often coexist in the experience of the disconnected individual. Cognitive processing of
non-social and complex stimuli also seems impaired (Baumeister et al., 2002). Isolation can induce
rumination, which likely reduces a person’s ability to think and reason clearly about something else.
When the experience of isolation is prolonged, there might be an increase in the risk of depression
and suicidal thoughts. Moreover, several studies have shown a negative impact of isolation and
exclusion on health outcomes and mortality risk (Cacioppo & Cacioppo, 2018; Hawkley &
Capitanio, 2015; Holt-Lunstad et al., 2010). According to these studies, the human brain is the key
organ for forming, monitoring, maintaining, repairing, and replacing salutary connections with
other people (Holt-Lunstad et al., 2010). Social isolation is associated with stress and anxiety and
increased hypervigilance for social threats (Cacioppo, Balogh, & Cacioppo, 2015).
Scholars have also proposed theoretical models to account for social disconnection based on
the available empirical data. The Temporal Need-Threat Model of ostracism (Williams 2009) is
Running head: ISOLATED IN A MOON ANALOG HABITAT 6
among the most relevant. It focuses on the chain of events that unfolds over time. The beginning of
this process is the detection of ostracism (e.g., feeling ignored), which activates an immediate
response to social pain and lowers the satisfaction level of basic psychological needs (i.e.,
belongingness, self-esteem, control, and meaningful existence). The next phase of the model is
when the individual evaluates the situation and implements behaviors to recover the threatened
needs. However, if the condition of feeling ignored persists over time (for weeks or months), the
individual has no way of recovering adequate levels that satisfy their need to belong. This condition
represents the final stage of the model, called the resignation stage. In this condition, according to
Williams, a chronic state of psychological resignation occurs, which entails four facets: alienation,
depression, helplessness, and unworthiness. An unsatisfied need for belonging for too long leads to
alienation, feeling that there is no place for us in the world. A chronically threatened need for self-
esteem can lead to depressive symptoms based on a permanently devaluing self-perception. The
prolonged lack of control over time risks translating into a sense of pervasive helplessness.
Ultimately, feeling non-existent in the eyes of others, in the long run, can lead to a lack of hope for
To date, the resignation stage is the one that has received the least empirical attention from
scholars, probably due to the impracticality of inducing a condition of chronic exclusion and
isolation in the laboratory. A quasi-experimental study first found that chronic experiences of social
exclusion were associated with higher levels of negative emotions and resignation stage outcomes
compared with other negative conditions (e.g., chronic experiences of physical pain; Riva et al.,
2017; for a theoretical perspective, see Riva, Wesselmann, Wirth, Carter-Sowell, & Williams,
2014). Subsequent studies have then focused their attention on special populations who can be
victims of prolonged experiences of exclusion and social disconnection, such as adolescents with a
migrant background (Mazzoni et al., 2020), prisoners (Aureli et al., 2020), and asylum seekers
(Marinucci & Riva, 2021). However, there is a potentially crucial difference about what is known
for the effects of long-term exclusion and isolation: the groups mentioned above (e.g., adolescents
Running head: ISOLATED IN A MOON ANALOG HABITAT 7
with a migrant background, asylum seekers) typically experience prolonged exclusion and isolation
as victims, while in the case of space missions or extreme environments, people voluntarily choose
to leave and social disconnection is a side effect of their primary goal, their travel. This
motivational aspect could radically change the psychological consequences of prolonged
disconnection over time compared to what is predicted by existing theoretical models (e.g.,
Compared to the commonly studied situations of social exclusion and isolation, not only
motivation but also expectations can be quite different in space or extreme environment missions
that have a known end date. This has been firstly suggested by Taylor, Wheeler, and Altman’s
(1968) classic work, a notable exception to the paucity of studies on prolonged effects (at least in
terms of expectations) of social isolation. Investigating temporal effects, the authors assessed stress
and anxiety reactions in pairs of men confined for eight days in small rooms. The authors
manipulated the participants’ expectations about the mission’s duration, with half of the participants
being told the isolation would have been short (4 days) whereas the other half long (20 days).
Another factor related to privacy, with half of the participants having separate rooms available
while the other half had to live in one room. Results showed that higher stress and anxiety levels
occurred when the length of confinement was anticipated to be long (vs. short). Privacy seemed to
alleviate stress only when isolation was expected to be short.
Therefore, an open empirical question is whether the adverse effects of social isolation (e.g.,
negative feelings, resignation, withdrawal) would inevitably occur in the context of a mission in an
extreme, confined environment or whether the extreme peculiarity of the situation would result in a
different psychological response.
1.3. Long-term social isolation’s effects in the space psychology literature
In recent years, there has been a growing number of studies conducted on crew members
during high-fidelity space simulations, volunteers on a mission in extreme terrestrial environments
(e.g., Antarctica), and retrospective studies on cosmonauts (e.g., Groemer et al., 2010; Šolcová et
Running head: ISOLATED IN A MOON ANALOG HABITAT 8
al., 2014; Wang et al., 2014). The adverse effects reported in space missions include depression,
fatigue, asthenia, suboptimal work performance, interpersonal hostility, and withdrawal (Suedfeld,
2005). Several factors could contribute to stress in an extreme environment mission. These are:
demanding work, the adverse physical condition of the environment, the minimal possibility of
evacuation, boredom, unpredictable events, lack of privacy and personal space, isolation,
confinement, and communication difficulties (De La Torre et al., 2012; Suedfeld, 2005). However,
a review by Kanas (2014) highlighted that among the various psychological and interpersonal
stressors that can occur in a space mission, the first ones are isolation and confinement. Similarly,
the European Space Agency study identifies the psychological risks of prolonged social isolation
and confinement among the principal dangers of a long-term mission to the Moon (Horneck et al.,
Among the scholars who have most forcefully highlighted the possible adverse effects of life
experiences inside enclosed spaces in hostile environments is the work of Tachibana (2019).
According to the author, considering psychological factors and interpersonal dynamics is key to
avoiding the emergence of the Hobbesian state (“all against all”). Psychological and interpersonal
issues due to social isolation, gender, and cultural differences can undermine crews’ functioning,
leading to immoral behaviors such as sexual harassment, problematic social dynamics such as
quarrels and dropouts, and psychological issues including depression and suicidal thoughts. The
longer humans stay in space, the more severe these influences could likely become. The author
argues that behavioral health issues in isolated and confined environments similarly affect well-
trained professionals like astronauts and ordinary untrained people.
The literature on the subject is, however, still highly mixed. Parallel to the theoretical and
empirical works that have underlined the strong stressors, mainly related to social isolation (e.g., the
increased loneliness over time in the Mars-105 pilot study reported by van Baarsen et al., 2009),
some works have shown substantially positive psychological outcomes in relation to life in extreme
environments. As a point in the case, a study by Šolcová, Lačev, and Šolcová (2014) focused on
Running head: ISOLATED IN A MOON ANALOG HABITAT 9
moods and emotions reported by the six male participants in the Mars-500 project, a simulated
mission to Mars for 520 days. A total of 17 measurement sessions of 12 moods and 20 emotions
were taken during the simulated mission. The results showed that the crew members experienced
predominantly positive emotions during their isolation, and the mood changes of the crew members
were not synchronized among the six members during any measurement. In isolation, the crew
members preferred to suppress and neutralize their negative emotions and openly express only
positive emotions. Only a few specific periods were identified in which the level of positive
emotions decreased and negative emotions increased, likely due to the correspondence with
culturally conditioned holidays (i.e., end of December) that perhaps led the members to think about
what they were missing with their family and friends. Similarly, Wang et al. (2014) found that the
same participants in the Mars-500 project tended to rate some unpleasant pictures positively over
time. The authors’ interpretation was that this positive valence bias indicated a defensive strategy to
counteract the extended isolation period’s adverse effects. Reflection on loneliness seems to be a
lever to promote a positive experience (De La Torre et al., 2012; van Baarsen et al., 2009).
Ultimately, the space psychology literature has shown that social isolation is one of the
major risk factors in long-term expeditions. However, it has not yet reached a conclusion on some
specific effects of this isolation on psychological well-being and, most importantly, on the key
factors that could protect or aggravate the effects of social isolation on people’s mental health
during a mission.
2. The present study
The LUNARK project provided us with an ecological setting to test on the one side whether
Williams’ (2009) Temporal Need-Threat model, in particular, the resignation stage brought about
by the prolonged dissatisfaction of fundamental human needs, could be usefully applied to
understand individuals’ feelings and perceptions during long-term experiences of isolation. On the
other side, this project allowed us to relate the activities that can be carried out during a mission of
this type and the psychological well-being of individuals. The first Arctic Moon analog habitat was
Running head: ISOLATED IN A MOON ANALOG HABITAT 10
a unique setting for this test as previous research on the effects of isolation, and its moderators, has
been mainly focused on laboratories studies and relatively mild and short-lived forms of social
isolation (for some recent exceptions, see Aureli et al., 2020; Marinucci & Riva, 2021; Mazzoni et
Accordingly, this study aimed to track changes in feelings and perceptions of two human
beings living in Northern Greenland under conditions that simulated a habitat within a lunar
environment. What happens when isolation is not experienced as an end state but as instrumental to
achieving a meaningful purpose (as it is the case of a simulated space expedition) and it has an end
date known from its onset (i.e., the date of return is known)? These features likely differ from the
episodes of social exclusion typically studied in the social psychological literature (e.g., Aureli et
al., 2020; Marinucci & Riva, 2021; Riva et al., 2017), which are imposed from the outside rather
than being goal-directed, but also from other known forms of social isolation that usually lack a
motivational factor such as that linked to accomplishing a (simulated) space mission.
Previous space psychology research on simulated space missions or missions in extreme
environments has highlighted the individual differences in the valence associated with isolation,
which could bring about positive or negative feelings depending on the individual, their
preferences, coping strategies, and vulnerability (Johannes & van Baarsen, 2020; see also early
work by Taylor, Altman, Wheeler, & Kushner, 1969). Tachibana (2019) has warned against the
possible negative consequences of long-term isolation and confinement on health and social and
moral conduct. In addition, van Baarsen et al. (2009) found that loneliness increased over time
during the Mars-105 pilot study. However, reflection on its meaning seems to be key in producing a
positive experience (De La Torre et al., 2012; van Baarsen et al., 2009). Furthermore, Šolcová et al.
(2014) and Wang et al. (2014) found that isolation led expedition members to express more
positivity in emotions and stimuli assessment during the ground simulated spaceflight to Mars of
the Mars-500 project. Our study contributes to this evolving literature by providing further
empirical data on the effects of long-term isolation in an extreme simulated environment.
Running head: ISOLATED IN A MOON ANALOG HABITAT 11
2.1.1. Participants and Procedure. The two male volunteers (space architect A, aged 23,
and space architect B, aged 25, both Danish) involved in the LUNARK project consented to
participate in the present study after agreeing on the final makeup and length of the questionnaire.
The two space architects have known each other and worked together for long since their
architectural studies. They previously participated in a 14-day stay in the Wadi-Rum desert, which
can be viewed as training for this Arctic mission, which had a higher difficulty level. The actual
Lunark mission within the Moon analog habitat began on 1 October 2020 and ended on 30
November 2020 (61 days of total duration – see https://saga.dk/projects/lunark/expedition). During
the mission, the activities conducted by the two volunteers included performing essential tasks such
as collecting ice from outside the habitat for their water, filming documentaries, conducting studies
and experiments (psychological as well as some in engineering/architecture), working on personal
projects and socializing amongst themselves (e.g., watching films together). For the present
research purposes, for 47 of the 61 days inside the Moon analog habitat, the two volunteers filled
out a paper-and-pencil questionnaire at the end of each day. The questionnaire took about 20
minutes per day to be completed. Data were collected through the same self-report measures every
day to allow for a time-based analysis that dynamically tracks changes in the participants’ mental
states and coping strategies in line with the temporal dimension underlying the Temporal Need-
Threat Model (Williams, 2009; see also Riva, Montali, Wirth, Curioni, & Williams, 2017). This
analysis allowed us to carry out a fine-grained investigation of the consequences of long-term social
isolation as well as the coping strategies learned/developed over time. The study was assessed
through the Self-Assessment for Governance and Ethics (SAGE) procedure (reference number:
514292-514283-64164204), and it did not require a full ethics review by the University of Surrey
Running head: ISOLATED IN A MOON ANALOG HABITAT 12
The daily survey was designed to assess behavioral and psychological variables referring to
the day of survey completion. Except for daily activities (see below for further details), composite
scores were computed by averaging the responses to the items measuring each construct. The
following constructs were included1.
Emotions. Six items from the Rejected-Related Emotion Scale (Buckley et al., 2004) were
used to measure positive and negative emotions felt during the current day (e.g., “I felt angry”). The
items were rated on a 7-point Likert scale (from 1 = not at all to 7 = extremely). Higher scores on
the composite index (Cronbach’s α = .75) indicated higher levels of negative emotions.
Need-threat. Four items (one per dimension) of the Need-Threat Scale (Williams, 2009)
were selected to assess the four fundamental human needs threatened by ostracism as theorized by
Williams (2009): belonging (“I felt disconnected”), self-esteem (“I felt good about myself”), control
(“I felt that I was in control of events”), and meaningful existence (“I felt non-existent”). The items
were rated on a 7-point Likert scale (from 1 = not at all to 7 = extremely). Higher scores on the
composite index indicated a higher threat. However, the low internal consistency (Cronbach’s α
= .40) led us to exclude this construct from the analysis.
Desire for Social Contact. A bipolar ad hoc scale was used to assess the desire for social
contact. The volunteers were asked to position themselves between two opposite extremes on a 7-
point Likert scale, with 1 representing the “staying alone” choice and 7 the “desire for social
contact” choice. The scale included five items (e.g., “Avoid crowded places” – “Spend more time
among people”). Higher scores on the composite index (Cronbach’s α = .96) indicated higher desire
for social contact.
Loneliness. The Three-Items Loneliness Scale (Hughes et al., 2004) was adapted to assess
the daily feeling of loneliness (e.g., “How often did you feel that you lack companionship?”). The
response format was on a 7-point Likert scale (from 1 = hardly ever to 7 = often). Higher scores on
1 The quesonnaire also included 6 of the 7 items of Factor 2 of the State Dicules in Emoon Regulaon Scale (S-
DERS, 21-item, Lavender et al., 2017). For the sake of conciseness, we do not report the results as the responses did
not signi*cantly changed over the considered me.
Running head: ISOLATED IN A MOON ANALOG HABITAT 13
the composite index indicated higher perceived loneliness. However, the low internal consistency
(Cronbach’s α = .46) led us to exclude this construct from the analysis.
Resignation. The feeling of resignation was measured by adapting the scale developed by
Marinucci and Riva (2021; see also Pancani et al., 2021). Twelve items (e.g., “I felt detached from
the world around me”) were selected from the original scale to cover the four facets of resignation
(i.e., depression, unworthiness, alienation, and helplessness) as theorized by Williams (2009). The
response format was on a 7-point Likert scale (from 1 = not at all to 7 = extremely). Higher scores
on the composite index (Cronbach’s α = .83) reflected higher resignation.
Time Perception. Time perception was assessed using items 1, 6, 11, 18, 26 (e.g., “Time is
passing by slower than usual”) of the Time Perception subscale of the Multidimensional State
Boredom Scale (MSBS, Fahlman et al., 2013). They were measured on a 7-point Likert scale (from
1 = strongly disagree to 7 = strongly agree). Higher scores on the composite index (Cronbach’s α
= .97) indicated slower time perception.
Activities. The two volunteers were asked to report the daily time (i.e., amount of time in
minutes) spent in a series of activities. The activities included: (1) talking with your mate about
non-personal/essential work aspects of the mission, (2) talking with your mate about personal/non-
essential work aspects, (3) eating, (4) sleeping, (5) taking care of personal hygiene, (6) exercising,
(7) reading books, (8) watching shows, (9) relaxing, (10) writing or doing other creative activities
non-essential for the mission. Responses to items 3, 4, and 5 were summed up to compute the
composite score of time spent satisfying physiological needs, whereas responses to items 7, 8, 9,
and 10 were summed up to compute the composite score of leisure time.
2.1.3. Analysis plan.
Data analysis was conducted in two steps. First, we ran a series of auto-regressive,
integrated, moving average (ARIMA) models following Tabachnick and Fidell’s (2013)
recommendations. In the context of time-series analysis, ARIMA models allow identifying possible
patterns over time based on multiple observations separated by constant lags. ARIMA models are
Running head: ISOLATED IN A MOON ANALOG HABITAT 14
defined by three main elements, namely p, d, and q, and are usually referred to as ARIMA (p, d, q).
The auto-regressive element p represents the number of parameters used to describe the relationship
among successive observations (i.e., the link between a variable measured at a certain time-point t
and the same variable measured at preceding time-points, such as t-1). The integrated element d
(also called differencing) represents the number of terms needed to make a nonstationary time-
series (e.g., an increasing or decreasing trend over time) stationary, removing the trend: d equal to 0
indicates no trends over time, d equal to 1 shows a linear trend, whereas d higher than 1 indicates
curvilinear trends. The moving average element q represents the number of terms needed to account
for the persistence of a random shock (i.e., presence of residuals once an adequate model is
identified) in the variable level over time. Generally, the maximum number of parameters required
to account for each of these elements does not exceed two. For instance, if a certain variable had an
increasing trend over time (d = 1), its level was predicted by the same variable’s level measured one
(t-1) and two (t-2) lags before (p = 2), and no random shocks were observed (q = 0), the best fitting
model would be an ARIMA (2, 1, 0).
We ran specific ARIMA models for each of five psychological outcomes (i.e., negative
emotions, feeling like being in a jail, resignation, desire for social contact, and time perception) for
a total of 10 models. The ARIMA models allowed us to detect the lag-dependent changes in the five
psychological variables reported by the two space architects. In other words, the ARIMA models
can account for the variability of each of the variables over time. The type and number of ARIMA’s
elements (autoregressive element, p, integrated element, d, and moving average element, q) were
automatically identified by the expert modeler tool of the forecasting module of SPSS, version 27
(IBM Corp., 2020). The expert modeler selects (and estimates) only statistically significant
parameters to describe a time-series, also evaluating the need to include a constant and, if
appropriate, transforming the variable (e.g., using the natural logarithmic function to stabilize
variance over time). Beyond R2 and stationary R2 (which is preferable when there are trends over
time), the following fit indices were computed for each model: (1) the Ljung Box test, which tests
Running head: ISOLATED IN A MOON ANALOG HABITAT 15
the null hypothesis that the model does not show a lack of fit; (2) the Root Mean Square Error
(RMSE), which represents the standard deviations of the unexplained variance; (3) the normalized
Bayesian information criterion (BIC). High R2 and stationary R2, non-significant Ljung Box test,
and lower values of RMSE and normalized BIC indicate better fitting.
In the second step of the analyses, we aimed at identifying which activities (i.e., talking
about non-personal topics, talking about personal topics, exercising, satisfying physiological needs,
and leisure time) were associated with the levels of the five psychological outcomes we considered
in the previous step of the analyses (i.e., negative emotions, feeling like being in a jail, resignation,
desire for social contact, and time perception). We ran a series of mixed models (one per outcome)
estimating the random intercept (i.e., intercept’s variance related to the space architects), fixed
intercept, and fixed regression parameters. Given the small number of observations, we used the
bootstrapping technique with 1,000 resampling, estimating 95% confidence interval for each
parameter. All the predictors were recoded to reflect hours (instead of minutes) spent in the
activities and mean-centered on the space architect’s average. As for time-series, mixed models
were conducted using SPSS, version 27 (IBM Corp., 2020).
3.1. Time-series analysis
Table 1 reports descriptive statistics and correlations of the five outcomes for the two space
architects, separately. We ran a set of paired sample t-tests to compare the outcomes’ means of the
two space architects. According to these initial statistics, space architect A experienced higher
levels of negative emotions, t(46) = 2.92, p = .005, d = 0.43, feeling like in jail, t(42) = 4.92, p
< .001, d = 0.75, resignation, t(46) = 11.85, p < .001, d = 1.73, and time perception, t(45) = 17.33, p
< .001, d = 2.56, compared to space architect B. Conversely, the space architect B showed higher
desire for social contacts than his partner, t(46) = 7.70, p < .001, d = 1.12.
Running head: ISOLATED IN A MOON ANALOG HABITAT 16
The results of the ARIMA models are reported in Table 2 and graphically displayed in
Figure 2. The patterns of the five outcomes over time generally differed between the two space
architects. However, the Ljung Box Q test was always associated with a non-significant probability,
except for space architect B’s desire for social contact, indicating that the estimated elements were
useful in predicting outcomes’ patterns. Negative emotions were predicted by random shocks
occurring at t-1 and t-2 for space architect A, explaining 30% of the outcome’s variance.
Conversely, the constant was the only predictor for space architect B’s emotions, resulting in an
extremely low explained variance. The feeling of being like in jail was predicted by the auto-
regressive element at lag 1 (t-1) for both space architects, indicating positive associations of this
feeling among two consecutive days. However, space architect B showed a higher stationary R2 and
lower RMSE, suggesting a better fit, probably due to this feeling’s highly regular level over time.
This latter result confirms the floor effect of this variable emerged in the descriptive statistics of the
space architect B. Resignation was the only variable for which the same elements (i.e., the constant
and the autoregressive element p at lag 1) were estimated for both space architects. The constant
was higher for space architect A than B, indicating a higher baseline resignation level (consistently
with descriptive statistics). The auto-regressive parameter at lag 1 indicated a positive association of
resignation levels among two consecutive days. The explained variance and RMSE values
suggested a better fit for space architect A than B. The desire for social contact showed an
increasing trend over time for both space architects (accounted by the differencing element d), but
the model did not fit the data for space architect B. Conversely, space architect A showed a positive
association of the desire for social contact among two consecutive days (p at lag 1 becomes 0.55
after raising it to the power of e) and the ARIMA model explained 31% of the outcome’s variance.
Finally, time perception showed strong variability over time for space architect A, but it was
predicted only by the constant, which was above the scale’s midpoint, indicating that time was
perceived as slower than usual. However, the extremely low stationary R2 and the relatively high
RMSE indicated a bad model fit. Concerning space architect B, the constant was close to the lowest
Running head: ISOLATED IN A MOON ANALOG HABITAT 17
possible value (consistently with the strong floor effect observed with descriptive statistics), the
autoregressive parameter at lag 1 was positive, and the model explained 21% of the outcome’s
variance. Therefore, time was generally perceived faster than usual, and time perception was
influenced by itself among two consecutive days.
3.2. Mixed Models
Table 3 reports descriptive statistics and correlations of the five activities for the two space
architects, separately. The results of the mixed models are reported in Table 4. The models ran on
negative emotions and feeling of being like in jail did not yield any significant parameter, except for
fixed and random intercept. Therefore, the space architects’ daily activities did not influence these
two outcomes. Conversely, the resignation was negatively predicted by both the time spent talking
about personal topics (i.e., each hour spent talking about personal topics was associated with a
decrease of 0.13 units on the resignation scale) and leisure time (i.e., each leisure hour was
associated with a decrease of 0.06 units of resignation). The effect of talking about personal topics
also predicted the desire for social contact, but the association was positive (i.e., each additional
hour increased the desire by 0.48 units). Moreover, desire for social contact was positively
predicted by exercising, with an increase of 0.54 units of desire per each additional hour spent
exercising. Finally, the model on time perception yielded only one significant predictor, namely
leisure time. Specifically, each leisure hour increased the perceived speed of time by 0.11 units.
This study investigated the self-reports of two space architects who lived inside the first
Arctic Moon analog habitat in Northern Greenland for 61 days. Specifically, we asked each of the
two space architects to report at the end of each day the type of main activities carried out during
the day and a set of psychological variables related to the prolonged social isolation and
confinement involved in the expedition (i.e., emotions, desire for social contact, resignation, and
time perception). We aimed to follow the evolution of psychological variables over time and detect
associations between them and the type of daily activities carried out during the day. The unique
Running head: ISOLATED IN A MOON ANALOG HABITAT 18
context in which our study took place was the LUNARK mission, a space architectural project
aimed at designing and testing a habitat to be used on the Moon in future missions. Our theoretical
framework was the social-psychological model of Temporal Need-Threat (Williams, 2009), which
predicts that individuals reach a state of resignation (involving alienation, depression, helplessness,
and unworthiness) when experiencing an extended period of social isolation and ostracism.
4.1. Main results: the lingering over effect of resignation feelings and the protective factors for
the psychological well-being
We first conducted a time-series analysis to detect lag-dependent changes of the same
psychological variables (i.e., negative emotions, feeling like being in a jail, resignation, desire for
social contacts, and time perception), confirming the central role of resignation and the desire for
social contact. However, our findings did not reveal a linear or curvilinear relationship of increasing
negative emotions or the psychological state of resignation as the days of isolation increased. This
appears to run counter to some predictions about the effects of social disconnection on
psychological well-being (Tachibana, 2019). The only linear trend that emerged concerned the
desire for social contact; for both space architects, the desire for social contact increased with time.
This result dovetails with the increased loneliness over time found in the Mars-105 pilot study by
van Baarsen et al. (2009). Indeed, increased loneliness can be coupled with an increased desire for
social connections. However, the desire for social contact can also signal that the pernicious
consequences typical of the resignation stage, such as withdrawal, were not predominant, and thus,
a functioning level of well-being was preserved. This could be explained by the nature of the social
isolation the two space architects experience: as functional to achieve a goal rather than as victims
of someone else’s behavior.
We also found that the feelings of resignation were predicted by what was reported on the
previous day on the same variable. In this sense, the feelings of resignation, unlike other
psychological variables, seem to be concatenated in time sequences in which the levels reported on
a given day predict what is reported on the next day. This result suggests the potential risks of
Running head: ISOLATED IN A MOON ANALOG HABITAT 19
experiencing resignation because it might be a non-volatile psychological reaction to prolonged
social isolation. Indeed, even if it did not linearly increase with time, resignation was dependent on
its previously recorded levels (e.g., the resignation levels of the day before), thus suggesting that it
“lingered over.” If so, resignation in this context might be a steady feeling that could be more
difficult to mitigate compared to more volatile psychological reactions that fluctuate day by day.
Finally, the results of this analysis on time perception showed high variability in what was
reported by the two space architects (i.e., slower than usual for space architect A, but faster than
usual for space architect B). Thus, it was not possible to draw any generalizable insight into this
construct. Still, this result suggests that time perception in extreme environment expeditions might
depend on individual differences rather than environmental factors (e.g., lights and brightness in the
habitat). It should be noted that we did not specifically investigate individual differences in the
current study. For example, although it has long been known that personality plays a role in
adjustment to isolation (Taylor et al., 1969), we did not include any personality measures given that
there were only two space architects working together within the LUNARK mission, thus
potentially making any personality-related conclusion not generalizable even to a dyad.
Along with time-series, we conducted mixed model analyses to relate the type of activity
undertaken with the same five psychological variables (i.e., negative emotions, feeling like being in
a jail, resignation, desire for social contact, and time perception). Our data revealed the key role of
feelings of resignation and how they can be mitigated. Indeed, talking about personal matters
predicted lower levels of psychological resignation and a greater desire for social contact.
Furthermore, we found that time spent in leisure activities was associated with lower levels of
resignation and the perception of time running faster. Finally, exercise predicted the desire for
social contact. In summary, protective factors for the psychological well-being of the two space
architects in extreme environments were talking about personal (vs. non-personal) topics, spending
time in leisure activities, and physical exercise. These activities were protective against feelings of
resignation, and they fostered the desire for social contact rather than social withdrawal.
Running head: ISOLATED IN A MOON ANALOG HABITAT 20
4.2. Social isolation in an extreme environment expedition
In line with previous reports (e.g., Šolcová et al., 2014), our study shows that isolation in an
extreme environment might not bear consequences such as affective numbness, passivity, and the
depletion of coping resources (e.g., Williams, 2009) that can characterize other types of chronic
social disconnection, like social exclusion, rejection, or confinement within a prison (Aureli et al.,
2020; Marinucci & Riva, 2021; Mazzoni et al., 2020; Riva et al., 2017). We discuss some factors
that could account for this key difference.
First, there is a highly salient motivational element during a space mission that is typically
missing from other life experiences of isolation. This is because, usually, the experiences of
exclusion and rejection are not chosen by the victim who suffers them but are imposed by the social
environment. Even if the victim chooses to isolate themself, s/he usually does so to protect themself
from further experiences of social pain (Riva et al., 2014). In a mission in space, isolation can be
perceived as a necessary, expected effort for a greater purpose. In line with this idea, analyses of the
Mars-500 study (van Baarsen, 2013) found that the more the respondent thought of their
involvement in the mission as voluntary, the less they perceived a future mission as challenging.
More specifically, the more the crew members felt they were free to decide the more they wanted to
participate in a future mission, the more they were willing to accept the individual sacrifice, and the
less they perceived the first mission to Mars as demanding. Motivation is likely to be relevant to
safe and favorable extraterrestrial human settlements.
Second, isolation in a space mission has typically a precise end date. Several studies of
Antarctic expeditions have shown that the most challenging times for individuals’ affect, cognition
and interactions are those following the midpoint of the expedition (the so-called “third-quarter
phenomenon,” e.g., (Palinkas et al., 2007; Palinkas & Houseal, 2000; Reed et al., 2001; Wang et al.,
2014). This is probably due to the realization that almost half of the expedition has still to be
completed (Wang et al., 2014). As the end of the mission approaches, the individuals know that
they will soon return to social relationships, be it romantic, family, friends, or work. This mindset
Running head: ISOLATED IN A MOON ANALOG HABITAT 21
might protect them from increased conditions of psychological resignation towards the end of the
expedition. Our results seem in line with studies that have investigated the roles of expectations and
social norms in situation of exclusion and isolation. Specifically, research on social exclusion
showed that ostracism and rejection hurt less when they are expected and/or in line with the social
norms of the situation (Greifeneder & Rudert, 2019; Rudert & Greifeneder, 2016; Wesselmann,
Wirth, & Bernstein, 2017). However, other, more well-known forms of social disconnection, on the
contrary, do not have a known end date or lead the person to think that there will never be an end to
their isolation. This can lead to psychological dynamics of self-protection in the form of withdrawal
and passivity. The individual abandons any attempt at social contact and withdraws to avoid the
pain associated with new forms of rejection (e.g., Williams, 2009).
Accordingly, a key finding of this study concerns the protective role of personal
communication. Early studies found that verbal exchanges and self-disclosure tended to be higher in
isolated groups (Altman & Haythorn, 1965; Taylor, Wheeler & Altman, 1973). Good
communication has been highlighted as a buffer promoting friendliness and effectiveness by De La
Torre et al. (2012) in their review of the psychosocial and neurobehavioral challenges involved in
human spaceflight. An analysis of communication among the six crew members of the Mars-105
pilot study showed two different patterns: one of “activation” and the other of “communication
channel closing” (Gushin et al., 2012). The “active” communication strategy was accompanied by
an increase in subjective mood and activity scores. The “psychological closure” strategy was
associated with significantly lower subjective mood and activity scores caused by the accumulation
of negative feelings. Our study adds to this literature by emphasizing the protective role of talking
about personal matters. In this sense, if future studies confirm this preliminary result, space
explorers should be encouraged to talk with the other team members and relate to them to a level
that personal matters can be touched on. Talking about personal matters can counteract the feelings
of resignation. However, it can also increase the crew members’ desire for social contact, which
cannot be completely fulfilled in the habitat. Thus, in planning and training for space missions,
Running head: ISOLATED IN A MOON ANALOG HABITAT 22
space actors should also consider issues related to diminished social connections. The need for
social connections is innate and automatic (Baumeister, & Leary, 1995), and a lack of awareness
about the state of social deprivation of a mission crew can risk undermining the members’
psychological balance and well-being. This should also be considered in light of our result on the
increased desire for social contact as a function of the time spent in physical exercise, an essential
activity in missions that put the body under stress.
The results of our study also show the critical role of leisure activities. In long space
missions, such as those to Mars, where communications between ground and space will be delayed
by up to 40 minutes or will be completely impossible, making available a set of playful, work
unrelated activities (e.g., videogames or other feasible games in the habitat and the context of the
mission) could be a good ally against feelings of resignation and as mental-health promoting
factors. In this study, leisure time also predicted a faster time perception, which has long been
linked to positive experiences (Harton, 1939), while boredom and the mood of fear have been
linked to an overestimation of the time passing by (Droit-Volet et al., 2011; Raffaelli et al., 2018).
This result further corroborates the conclusion that time dedicated to relaxation and creativity could
promote the (analog) astronauts’ well-being holistically.
4.3. Limitations and future directions
Our study results are limited in generalizability, considering that the LUNARK mission
involved only two individuals. This is quite common in space psychology and extreme environment
research (Groemer et al., 2010; Wang et al., 2014). However, in addition to the generalizability
issue, crew size and composition should be considered in future research because they are among
the main psychological aspects to think about in a long-term mission (Horneck et al., 2003). The
number of people in expedition crews changes from mission to mission. They can range from a
single person who remains alone within a module to cases with multiple crew members. The
mission examined in this study involved two crew members. This number means that we cannot
speak of a group, but a dyad and the resulting social dynamics have their own peculiarity.
Running head: ISOLATED IN A MOON ANALOG HABITAT 23
Therefore, future research will have to consider the impact of crew members’ varying sizes and
cultural backgrounds on feelings of alienation (Johannes & van Baarsen, 2020).
Moreover, our observations are limited to 47 of the 61 days the two space architects spent
inside the Moon analog habitat. The missing data are distributed in the following way. The first six
days for Space Architect A and the first seven days for Space Architect B were missing from our
observation. The final two days were also missing for both. In between, Space Architect A did not
complete the survey in October 2020 on the 16th, 24th, and in November 2020 on the 6th, 14th, and
25th. Space Architect B did not complete the survey in October 2020 on the 12th, 13th, 16th, 25th,
and 29th. Overall, there were 14 missing observations for each space architect, concentrated in the
initial setup days of the expedition, in the final ones, and distributed in the middle without apparent
concentrations or clusters. Based on what it was possible to trace during the conversations with the
two volunteers, in the early and late stages, the crew had less time available for surveys, which
resulted in these missing observations for both. In between, on some days, one of the crew members
felt unwell or had a busier day which caused the daily survey of this study not to be completed.
However, the lack of the complete time series throughout the mission days limits the reliability of
our data and future studies should consider ways to maximize compliance with psychological
measures such as those implemented in the present study.
Another important psychological factor to be considered in a space or extreme environment
mission is its duration (Horneck et al., 2003). In relatively short-time missions, it may be easier to
cope with the lack of social connections. Long-term missions could bring about different effects,
such as adaptation and positive loneliness (De La Torre et al., 2012; van Baarsen et al., 2009). For
this reason, the temporal dynamics of psychological reactions to stress in short-term vs. long-term
extreme environment expeditions deserve further investigation.
Future studies should also investigate the interplay between environmental, technological,
and psychological factors. For example, technology could compensate for the psychological strain
experienced by (analog) astronauts during extended social isolation and confinement periods.
Running head: ISOLATED IN A MOON ANALOG HABITAT 24
Indeed, leisure times as well as including game elements in the design of interactive systems, that
is, gamification (Deterding, 2012; Richards et al., 2014; Seaborn & Fels, 2015), could be helpful to
counteract the adverse effects of isolation and confinement, including boredom and loneliness, as
well as increase motivation and performance.
This was the first study of a human simulation of a Moon expedition in an Arctic analog
habitat. We asked the two space architects involved in the mission to self-report their feelings and
perceptions daily via a paper-and-pencil questionnaire. The analysis of their responses on 47 of the
61 days they spent inside the Moon analog habitat revealed associations between the type of daily
activities and psychological variables concerning the prolonged social isolation typically involved
in space missions. Specifically, talking about personal matters predicted lower levels of
psychological resignation. Talking about personal topics and physical exercise predicted a stronger
desire for social contacts, as opposed to social withdrawal. Although we did not find a linear
relationship of increasing negative emotions or the psychological state of resignation as isolation
days increased, we found that the desire for social contact increased with time. These results could
inform future training and the planning of the daily activities during an extreme environment
expedition. For example, talking about personal matters, leisure time, and physical exercise could
be highlighted as activities that, although work unrelated, are relevant because they can protect the
individuals’ psychological well-being against the adverse effects of long-term social isolation.
6. Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
Running head: ISOLATED IN A MOON ANALOG HABITAT 25
We want to thank the two space architects who volunteered to participate in this study, the
LUNARK team, the researchers, and the collaborators who contributed to this international research
Running head: ISOLATED IN A MOON ANALOG HABITAT 26
Altman, I., Taylor, D. A., & Wheeler, L. (1971). Ecological Aspects of Group Behavior in Social
Isolation. Journal of Applied Social Psychology, 1(1), 76-100. https://doi.org/10.1111/j.1559-
Altman, I., & Haythorn, W. W. (1965). Interpersonal exchange in isolation. Sociometry, 28(4), 411-
Aureli, N., Marinucci, M., & Riva, P. (2020). Can the chronic exclusion‐resignation link be
broken? An analysis of support groups within prisons. Journal of Applied Social Psychology,
50(11), 638-650. https://doi.org/10.1111/jasp.12701
Baumeister, Roy, F., & Leary, M, R. (1995). The need To belong: Desire for interpersonal
attachments as a fundamental human-motivation. Psychological Bulletin, 117(3), 497–529.
Baumeister, R. F., Twenge, J. M., & Nuss, C. K. (2002). Effects of Social Exclusion on Cognitive
Processes: Anticipated Aloneness Reduces Intelligent Thought. Journal of Personality and
Social Psychology, 83(4), 817–827. https://doi.org/10.1037//0022-35220.127.116.117
Buckley, K. E., Winkel, R. E., & Leary, M. R. (2004). Reactions to acceptance and rejection:
Effects of level and sequence of relational evaluation. Journal of Experimental Social
Psychology, 40(1), 14–28. https://doi.org/10.1016/S0022-1031(03)00064-7
Cacioppo, J. T., & Cacioppo, S. (2018). The growing problem of loneliness. The Lancet,
391(10119), 426. https://doi.org/10.1016/S0140-6736(18)30142-9
Cacioppo, S., Balogh, S., & Cacioppo, J. T. (2015). Implicit attention to negative social, in contrast
to nonsocial, words in the Stroop task differs between individuals high and low in loneliness:
evidence from event-related brain microstates. Cortex, 70, 213-233.
Running head: ISOLATED IN A MOON ANALOG HABITAT 27
De La Torre, G. G., van Baarsen, B., Ferlazzo, F., Kanas, N., Weiss, K., Schneider, S., & Whiteley,
I. (2012). Future perspectives on space psychology: Recommendations on psychosocial and
neurobehavioural aspects of human spaceflight. Acta Astronautica, 81(December), 587–599.
Deterding, S. (2012). Gamification: Designing for motivation. Interactions, 19(4), 14–17.
Droit-Volet, S., Fayolle, S. L., & Gil, S. (2011). Emotion and time perception: effects of film-
induced mood. Frontiers in Integrative Neuroscience, 5(August), 1–9.
Fahlman, S. A., Mercer-Lynn, K. B., Flora, D. B., & Eastwood, J. D. (2013). Development and
Validation of the Multidimensional State Boredom Scale. Assessment, 20(1), 68–85.
GAO. (2021). NASA LUNAR PROGRAMS: Significant Work Remains, Underscoring Challenges
to Achieving Moon Landing in 2024. In GAO Reports (Issue May).
Greifeneder, R., & Rudert, S. C. (2019). About Flames and Boogeymen: Social norms affect
individuals’ construal of social exclusion. In S. C. Rudert, R. Greifeneder, & K. D. Williams
(Eds.), Current Directions in Ostracism, Social Exclusion and Rejection. Routledge.
Groemer, G., Gruber, V., Bishop, S., Peham, D., Wolf, L., & Birgit, H. (2010). Human performance
data in a high workload environment during the simulated Mars expedition ‘“ AustroMars .”’
Acta Astronautica, 66, 780–787. https://doi.org/10.1016/j.actaastro.2009.08.017
Gushin, V., Shved, D., Vinokhodova, A., Vasylieva, G., Nitchiporuk, I., Ehmann, B., & Balazs, L.
(2012). Some psychophysiological and behavioral aspects of adaptation to simulated
autonomous Mission to Mars. Acta Astronautica, 70, 52–57.
Running head: ISOLATED IN A MOON ANALOG HABITAT 28
Harton, J. J. (1939). An investigation of the influence of success and failure on the estimation of
time. Journal of General Psychology, 21, 51–62.
Hawkley, L. C., & Capitanio, J. P. (2015). Perceived social isolation, evolutionary fitness and
health outcomes: a lifespan approach. Philosophical Transactions: Biological Sciences,
370(1669), 1–12. https://doi.org/10.1098/rstb.2014.0114
Hawkley, L. C., Ph, D., Cacioppo, J. T., & Ph, D. (2010). Loneliness Matters: A Theoretical and
Empirical Review of Consequences and Mechanisms. Annals of Behavioral Medicine, 40(2),
Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social Relationships and Mortality Risk: A
Meta-analytic Review. PLoS Medicine, 7(7), e1000316.
Horneck, G., Facius, R., Reichert, M., Rettberg, P., Seboldt, W., Manz, D., Come, B., & Maille, A.
(2003). HUMEX, a study on the survivability and adaptation of humans to long-duration
exploratory missions, part I: lunar missions. Advances in Space Research, 31(11), 2389–2401.
Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004). A short scale for measuring
loneliness in large surveys. Research on Aging, 26(6), 655–672.
IBM Corp. (2020). Statistical Package for Social Sciences (Version 27.0). IBM Corp.
Johannes, B., & van Baarsen, B. (2020). Psychological monitoring. In A. Choukèr (Ed.), Stress
challenges and immunity in space: From mechanisms to monitoring and preventive strategies
(Second Edi, pp. 421–432). Springer. https://doi.org/https://doi.org/10.1007/978-3-030-16996-
Running head: ISOLATED IN A MOON ANALOG HABITAT 29
Kanas, N. (2014). Psychosocial issues during an expedition to Mars. Acta Astronautica, 103, 73–80.
Lavender, J. M., Tull, M. T., DiLillo, D., Messman-Moore, T., & Gratz, K. L. (2017). Development
and Validation of a State-Based Measure of Emotion Dysregulation: The State Difficulties in
Emotion Regulation Scale (S-DERS). Assessment, 24(2), 197–209.
Marinucci, M, & Riva, P. (2021). Surrendering to social emptiness: Chronic social exclusion
longitudinally predicts resignation in asylum seekers. British Journal of Social Psychology.
Marinucci, Marco, & Riva, P. (2021). How intergroup social connections shape immigrants’
responses to social exclusion. Group Processes & Intergroup Relations, 24(3), 411–435.
Mazzoni, D., Pancani, L., Marinucci, M., & Riva, P. (2020). The dual path of the rejection
(dis)identification model: A study on adolescents with a migrant background. European
Journal of Social Psychology, 50, 799–809. https://doi.org/10.1002/ejsp.2672
Palinkas, L. A., & Houseal, M. (2000). Stages of change in mood and behavior during a winter in
Antarctica. Environment and Behavior, 32(1), 128–141.
Palinkas, L. A., Reedy, K. R., Shepanek, M., Smith, M., Anghel, M., Steel, G. D., Reeves, D., Case,
H. S., Do, N. Van, & Reed, H. L. (2007). Environmental influences on hypothalamic –
pituitary – thyroid function and behavior in Antarctica. Physiology & Behavior, 92, 790–799.
Profitiliotis, G., & Loizidou, M. (2019). Planetary protection issues of private endeavours in
research, exploration, and human access to space: An environmental economics approach to
Running head: ISOLATED IN A MOON ANALOG HABITAT 30
forward contamination. Advances in Space Research, 63(1), 598–605.
Raffaelli, Q., Mills, C., & Christoff, K. (2018). The knowns and unknowns of boredom: a review of
the literature. Experimental Brain Research, 236(9), 2451–2462.
Reed, H. L., Reedy, K. R., Palinkas, L. A., Do, N. V. A. N., Finney, N. S., Case, H. S., Mar, H. J. L.
E., Wright, J., & Thomas, J. (2001). Impairment in Cognitive and Exercise Performance during
Prolonged Antarctic Residence: Effect of Thyroxine Supplementation in the Polar
Triiodothyronine Syndrome. The Journal of Clinical Endocrinology & Metabolism, 86(1),
Richards, C., Thompson, C. W., & Graham, T. C. N. (2014). Beyond designing for motivation: The
importance of context in gamification. CHI PLAY 2014 - Proceedings of the 2014 Annual
Symposium on Computer-Human Interaction in Play, 217–226.
Riva, P., & Eck, J. (Eds.). (2016). Social exclusion: Psychological approaches to understanding
and reducing its impact. Springer. https://doi.org/10.1007/978-3-319-33033-4
Riva, P., Montali, L., Wirth, J. H., Curioni, S., & Williams, K. D. (2017). Chronic social exclusion
and evidence for the resignation stage: An empirical investigation. Journal of Social and
Personal Relationships, 34(4), 541-564. https://doi.org/10.1177/0265407516644348
Riva, P., Wesselmann, E. D., Wirth, J. H., Carter-Sowell, A. R., & Williams, K. D. (2014). When
pain does not heal: The common antecedents and consequences of chronic social and physical
pain. Basic and Applied Social Psychology, 36(4), 329-346.
Rudert, S. C., & Greifeneder, R. (2016). When it’s okay that I don’t play: Social norms and the
Running head: ISOLATED IN A MOON ANALOG HABITAT 31
situated construal of social exclusion. Personality and Social Psychology Bulletin, 42(7), 955-
Seaborn, K., & Fels, D. I. (2015). Gamification in theory and action: A survey. International
Journal of Human-Computer Studies, 74, 14–31. https://doi.org/10.1016/j.ijhcs.2014.09.006
Šolcová, I. P., Lačev, A., & Šolcová, I. (2014). Study of individual and group affective processes in
the crew of a simulated mission to Mars: Positive affectivity as a valuable indicator of changes
in the crew affectivity. Acta Astronautica, 100, 57–67.
Suedfeld, P. (2005). Invulnerability, coping, salutogenesis, integration: four phases of space
psychology. Aviation, Space, and Environmental Medicine, 76(6 Suppl), B61–B66.
Tabachnick, B. G., & Fidell, L. S. (2013). Time-series analysis. In B. G. Tabachnick & L. S. Fidell
(Eds.), Using multivariate statistics (6th Editio, p. Chapter 18). Pearson Internationl Edition.
Tachibana, K. (2019). A Hobbesian qualm with space settlement. Futures, 110(November 2018),
Taylor, D. A., Altman, I., Wheeler, L., & Kushner, E. N. (1969). Personality factors related to
response to social isolation and confinement. Journal of Consulting and Clinical Psychology,
33(4), 411-419. https://doi.org/10.1037/h0027805
Taylor, D. A., Wheeler, L., & Altman, I. (1968). Stress relations in socially isolated groups. Journal
of Personality and Social Psychology, 9(4), 369-376. https://doi.org/10.1037/h0026088
Taylor, D. A., Wheeler, L., & Altman, I. (1973). Self-disclosure in isolated groups. Journal of
Personality and Social Psychology, 26(1), 39–47. https://doi.org/10.1037/h0034233
Tomaka, J., Thompson, S., & Palacios, R. (2006). The Relation of Social Isolation, Loneliness, and
Social Support to Disease Outcomes Among the Elderly. Journal of Aging and Health, 18(3),
Running head: ISOLATED IN A MOON ANALOG HABITAT 32
van Baarsen, B., Ferlazzo, F., Ferravante, D., Di Nocera, F., Jørgensen, J., Smit, J., van Duijn, M.,
Giannini, A.-M., Kuipers, A., & van der Pligt, J. (2009). Digging Into Space Psychology and
Isolation: The Mars500 Lodgead Study.Primary results of the Mars105 pilot study. The 60th
International Astronautical Congress.
van Baarsen, Bernadette. (2013). Person autonomy and voluntariness as important factors in
motivation , decision making , and astronaut safety : First results from the Mars500
LODGEAD study $. Acta Astronautica, 87, 139–146.
van Baarsen, Bernadette, Ferlazzo, F., Ferravante, D., Smit, J., van der Pligt, J., & van Dijn, M. A.
J. (2012). Emotional and cognitive adaptation during 520 days of isolation: Results from the
LODGEAD Mars500 study. The 63th International Astronautical Congress, IAC-
Wang, Y., Jing, X., Lv, K., Wu, B., Bai, Y., Luo, Y., Chen, S., & Li, Y. (2014). During the Long
Way to Mars: Effects of 520 Days of Confinement (Mars500) on the Assessment of Affective
Stimuli and Stage Alteration in Mood and Plasma Hormone Levels. PLoS One, 9(4), e87087.
Wesselmann, E. D., Wirth, J. H., & Bernstein, M. J. (2017). Expectations of Social Inclusion and
Exclusion. Frontiers in Psychology, 8(112). https://doi.org/10.3389/fpsyg.2017.00112
Williams, K. D. (2009). Ostracism: A temporal need-threat model. In M. P. Zanna (Ed.), Advances
in experimental social psychology (Vol. 41, pp. 275–314). Elsevier Academic Press.
Running head: ISOLATED IN A MOON ANALOG HABITAT 33
Table 1 – The five outcomes of the two space architects during the LUNARK mission: means and
standard deviation (in brackets) are reported along with the coefficients of the correlations between
activities for each space architect.
Mean (S.D.) (1) (2) (3) (4) (5)
Space Architect A
(1) Negative Emotions 2.66 (0.82) 1.00
(2) Feeling like in jail 2.51 (1.66) .488* 1.00
(3) Resignation 3.31 (0.69) .756* .489* 1.00
(4) Desire for social contacts 3.75 (1.74) -.308* .028 -.307* 1.00
(5) Time perception 4.36 (1.15) .199 .203 .302* .249 1.00
Space Architect B
(1) Negative Emotions 2.22 (0.59) 1.00
(2) Feeling like in jail 1.20 (0.51) .327* 1.00
(3) Resignation 1.91 (0.42) .799* .127 1.00
(4) Desire for social contacts 5.74 (1.32) -.267 -.116 -.308* 1.00
(5) Time perception 1.10 (0.23) .291* .721* .207 .049 1.00
* p < .05
Running head: ISOLATED IN A MOON ANALOG HABITAT 34
Table 2 – The results of the ARIMA models conducted on the two space architects: estimated parameters and fit indices are reported.
Negative emotions Feeling like in jail Resignation Desire for social contacts Time perception
Estimate (S.E.) p-value Estimate (S.E.) p-value Estimate (S.E.) p-value Estimate (S.E.) p-value Estimate (S.E.) p-value
Space Architect A
Transf. No Natural Log No Natural Log No
Constant 2.68 (0.19) < .001 0.76 (0.14) < .001 3.35 (0.16) < .001 --- 4.36 (0.17) < .001
p - lag 1 --- 0.40 (0.15) .010 0.44 .002 -0.59 (0.13) < .001 ---
d--- --- --- 1 ---
q - lag 1 -0.44 (0.14) .004 --- --- --- ---
q - lag 2 -0.42 (0.15) .006 --- --- --- ---
Stationary R2.30 .13 .18 .31 < .001
R2.30 .11 .18 .46 < .001
Ljung Box Q 12.69 [df = 16] .695 22.33 [df = 17] .172 13.69 [df = 17] .689 10.95 [df = 17] .859 23.64 [df = 18] .167
RMSE 0.70 1.58 0.64 1.29 1.15
Norm. BIC -0.47 1.09 -0.74 0.60 0.37
Space Architect B
Transf. Natural Log Natural Log No No No
Constant 0.77 (0.04) < .001 --- 1.91 (0.08) < .001 0.05 (0.25) .850 1.10 (0.05) < .001
p (lag 1) --- 0.58 (0.12) < .001 0.30 (0.14) .041 --- 0.45 (0.13) .002
d--- --- --- 1 ---
q (lag 1) --- --- --- --- ---
q (lag 2) --- --- --- --- ---
Stationary R2< .001 .26 .09 < .001 .21
R2< .001 .27 .09 < .001 .21
Ljung Box Q 19.83 [df = 18] .343 22.26 [df = 17] .175 20.20 [df = 17] .264 37.08 [df = 18] .005 16.93 [df = 17] .459
RMSE 0.59 0.43 0.40 1.70 0.21
Norm. BIC -0.98 -1.60 -1.66 1.15 -2.99
Note. Transf. = Transformation of the outcome variable; p = autoregressive component(s); d = trend component; q = moving average component(s); RMSE = root
mean square error; Norm. BIC = normalized Bayesian information criterion. Lags represent the number of days preceding the observation that were considered in
predicting the current observation.
Running head: ISOLATED IN A MOON ANALOG HABITAT 35
Table 3 – The daily activities of the two space architects during the LUNARK mission: self-
reported average hours spent (and standard deviation) in and correlations among activities are
shown for each space architect.
Mean (S.D.) (1) (2) (3) (4) (5)
Space Architect A
(1) Talking about non-personal topics 1.85 (1.22) 1.00
(2) Talking about personal topics 1.41 (0.98) .187 1.00
(3) Exercising 0.68 (0.54) -.051 -.028 1.00
(4) Satisfying physiological needs 9.14 (1.18) .013 -.016 .116 1.00
(5) Leisure time 3.36 (1.47) -.177 -.004 -.015 -.110 1.00
Space Architect B
(1) Talking about non-personal topics 1.82 (0.94) 1.00
(2) Talking about personal topics 1.74 (0.95) .300 * 1.00
(3) Exercising 0.61 (0.60) -.009 -.075 1.00
(4) Satisfying physiological needs 8.29 (0.85) ≈.000 -.021 .353 * 1.00
(5) Leisure time 4.29 (1.71) -.116 -.113 -.199 -.126 1.00
* p < .05
Running head: ISOLATED IN A MOON ANALOG HABITAT 36
Table 4 – The results of the mixed models: parameters estimates and 95% bootstrapped confidence intervals (in brackets) are reported.
Negative emotions Feeling like in jail Resignation Desire for social
contacts Time perception
Random intercept 0.085
Fixed intercept 2.443
Talking about non-personal
Talking about personal topics -0.067
Satisfying physiological needs -0.030
Leisure time -0.052
Running head: ISOLATED IN A MOON ANALOG HABITAT 37
Figure 1 – The LUNARK’s Moon analog habitat built by the two space architects who lived in it for
Figure 2 – The results of the ARIMA models conducted on the two space architects: observed and
predicted levels of (a) negative emotions, (b) feeling like in a jail, (c) resignation, (d) desire for
social contact, and (e) time perception over time.
Running head: ISOLATED IN A MOON ANALOG HABITAT 38
Running head: ISOLATED IN A MOON ANALOG HABITAT 39