Content uploaded by Adam J Vanhove
Author content
All content in this area was uploaded by Adam J Vanhove on Mar 31, 2015
Content may be subject to copyright.
NASA/TM-2015-218565
Examining Psychosocial Well-being and Performance
in Isolated, Confined, and Extreme Environments
Final Report
31 July 2014
Adam J. Vanhove
Mitchel N. Herian
P.D. Harms
Fred Luthans
Justin A. DeSimone
National Aeronautics and
Space Administration
Johnson Space Center
Houston, Texas 77058
February 2015
NASA STI Program ... in Profile
Since its founding, NASA has been dedicated
to the advancement of aeronautics and space
science. The NASA scientific and technical
information (STI) program plays a key part in
helping NASA maintain this important role.
The NASA STI program operates under the
auspices of the Agency Chief Information
Officer. It collects, organizes, provides for
archiving, and disseminates NASA’s STI. The
NASA STI program provides access to the NASA
Aeronautics and Space Database and its public
interface, the NASA Technical Report Server,
thus providing one of the largest collections of
aeronautical and space science STI in the world.
Results are published in both non-NASA channels
and by NASA in the NASA STI Report Series,
which includes the following report types:
TECHNICAL PUBLICATION. Reports of
completed research or a major significant
phase of research that present the results of
NASA Programs and include extensive data
or theoretical analysis. Includes compilations
of significant scientific and technical data and
information deemed to be of continuing
reference value. NASA counterpart of peer-
reviewed formal professional papers but has
less stringent limitations on manuscript length
and extent of graphic presentations.
TECHNICAL MEMORANDUM. Scientific
and technical findings that are preliminary or
of specialized interest, e.g., quick release
reports, working papers, and bibliographies
that contain minimal annotation. Does not
contain extensive analysis.
CONTRACTOR REPORT. Scientific and
technical findings by NASA-sponsored
contractors and grantees.
CONFERENCE PUBLICATION. Collected
papers from scientific and technical
conferences, symposia, seminars, or other
meetings sponsored or co-sponsored
by NASA.
SPECIAL PUBLICATION. Scientific,
technical, or historical information from
NASA programs, projects, and missions,
often concerned with subjects having
substantial public interest.
TECHNICAL TRANSLATION. English-
language translations of foreign scientific
and technical material pertinent to
NASA’s mission.
Specialized services also include creating
custom thesauri, building customized databases,
and organizing and publishing research results.
For more information about the NASA STI
program, see the following:
Access the NASA STI program home page
at http://www.sti.nasa.gov
E-mail your question via the Internet to
help@sti.nasa.gov
Fax your question to the NASA STI Help
Desk at 443-757-5803
Phone the NASA STI Help Desk at
443-757-5802
Write to:
NASA Center for AeroSpace Information
7115 Standard Drive
Hanover, MD 21076-1320
NASA/TM-2015-218565
Examining Psychosocial Well-being and Performance
in Isolated, Confined, and Extreme Environments
Final Report
31 July 2014
Adam J. Vanhove
Mitchel N. Herian
P.D. Harms
Fred Luthans
Justin A. DeSimone
National Aeronautics and
Space Administration
Johnson Space Center
Houston, Texas 77058
February 2015
Available from:
NASA Center for AeroSpace Information National Technical Information Service
7115 Standard Drive 5301 Shawnee Road
Hanover, MD 21076-1320 Alexandria, VA 22312
Available in electric form at http://ston.jsc.nasa.gov/collections/TRS
1
Table of Contents
Executive Summary ...................................................................................................................................... 3
Introduction ................................................................................................................................................... 6
Effects of ICE on Psychosocial Health and Performance ............................................................................. 8
Positive Consequences of ICE Experiences ............................................................................................ 10
Well-being in Long-duration Spaceflight ................................................................................................... 12
Definition and Conceptualization of Well-being .................................................................................... 12
Hedonic Well-being ................................................................................................................................ 14
Eudaimonic Well-being .......................................................................................................................... 14
Psychological Capital (PsyCap) .............................................................................................................. 15
Summary ................................................................................................................................................. 16
Relationship between Well-being and Performance ................................................................................... 17
Well-being in Extreme Contexts: Research on Military Personnel ........................................................ 20
Summary ................................................................................................................................................. 23
Temporal Effects on Well-being ................................................................................................................. 24
Operational Tempo and Well-being in the Military ................................................................................ 26
Summary: Research Aims of the Present Report .................................................................................... 28
Method ........................................................................................................................................................ 29
Systematic Review .................................................................................................................................. 29
Inclusion Criteria .................................................................................................................................... 31
Aim 2: Well-being-performance relationship.. ................................................................................... 31
Aim 3: Temporal Change in Psychosocial Factors. ............................................................................ 34
Individual-level and Group-level Change.. ......................................................................................... 35
Standardizing Response Scales and Time Points ................................................................................ 36
Results ......................................................................................................................................................... 37
Well-being-Performance Relationship Sample Descriptions .................................................................. 37
Hedonic Well-being ................................................................................................................................ 38
Hedonic Well-being and Performance and Work Attitude Criteria. ................................................... 38
Hedonic Well-being and Stress and Symptomatology Criteria. ......................................................... 41
Hedonic Well-being Summary ............................................................................................................ 43
Eudaimonic Well-being .......................................................................................................................... 43
Eudaimonic Well-being and Performance and Work Attitude Criteria. ............................................. 43
Eudaimonic Well-being and Stress and Psychosocial Symptomatology Criteria. .............................. 47
2
Eudaimonic Well-being Summary. ..................................................................................................... 50
PsyCap .................................................................................................................................................... 50
PsyCap and Performance and Work Attitude Criteria.. ...................................................................... 51
PsyCap and Stress and Symptomatology Criteria. .............................................................................. 52
Summary. ............................................................................................................................................ 53
General Well-being ................................................................................................................................. 54
Temporal Change in Psychosocial Factors Sample Descriptions ........................................................... 55
Individual-level Temporal Change. .................................................................................................... 55
Group-level Temporal Change. ........................................................................................................... 59
Summary of Individual- and Group-level Temporal Change. ............................................................ 71
Discussion ................................................................................................................................................... 72
Well-being-Performance Relationship (Aim 2) ...................................................................................... 72
Temporal Effects on Well-being ............................................................................................................. 80
Conclusion .................................................................................................................................................. 84
References ................................................................................................................................................... 91
Appendix A. Characteristics of Studies included in Systematic Review of Well-being-Performance
Relationship in ICE Settings. .................................................................................................................... 113
Appendix B. Characteristics of Studies included in Individual-level Temporal Change Analyses. ......... 118
Appendix C. Characteristics of Studies included in Group-level Temporal Change Analyses. ............... 121
3
Executive Summary
The purpose of this document is to provide a systematic review of the literature characterizing
the well-being-performance relationship in isolated, confined, and extreme (ICE) environments.
This report also examines temporal effects on well-being and stress/negative psychosocial
functioning over the duration of ICE missions. In doing so, this report aims to help address
research Gap BMed5, as identified by the Behavioral Health and Performance (BHP) element of
the NASA Human Research Program (HRP): “What individual characteristics predict successful
adaptation and performance in an isolated, confined and extreme environment, especially for
long duration missions?” A brief summary of the findings is as follows:
Well-being and Performance in ICE Settings
Three overarching factors of well-being (hedonic well-being, eudaimonic well-being, and
psychological capital [PsyCap]) were identified as relevant in ICE environments and
were used to organize evidence of the well-being-performance relationship in this report.
A broad set of performance-related criteria that include both direct and indirect (e.g.,
work attitudes) measures of performance and stress/symptomatology was employed for
analyses in this report.
Indicators of eudaimonic well-being have been those most often studied in ICE settings,
followed by indicators of hedonic well-being and PsyCap.
Indicators of PsyCap have been most strongly positively related to direct measures of
performance (e.g., supervisor/peer performance ratings), while indicators of all three
factors of well-being have shown consistently moderate positive relationships with
indirect measures of performance (e.g., self-reported team cohesion and group
functioning).
Positive affect, an indicator of hedonic well-being, has been most consistently
(negatively) related to stress measures; indicators of eudaimonic well-being have been
most consistently (negatively) related to symptomatology measures.
Temporal Trends in Well-being in ICE Settings
Primary research examining temporal effects on psychosocial health within the ICE
literature has not resulted in consensus over patterns of change across mission duration.
Meta-analytic evidence reported here suggests the beginning, and to a lesser extent the
end, of missions to be marked by the greatest changes in both positive (i.e., well-being)
and negative indicators of psychosocial functioning (i.e., stress/symptomatology), while
functioning remained relatively stable during the interim stages.
Moderator analyses of temporal trends suggest similar effects across short- and long-
duration missions. This finding suggests that short-duration analog missions might
4
effectively be used to understand temporal trends in well-being during long-duration
missions.
Evidence regarding crew characteristics (crew nationality and gender) suggests that
homogeneous crews may experience somewhat greater and more stable psychosocial
health throughout mission duration. However, evidence does not suggest that
heterogeneous crews will typically be detrimental to crew psychosocial health, and there
appears to be a fairly clear dichotomy between those who do and do not thrive, at least, in
multi-gender crews.
We offer the following recommendations for the future study of well-being in ICE contexts:
1. Validation of well-being measures specific to ICE settings is needed:
The present research has shown that indicators of well-being are statistically related to
measures of performance. Better understanding the role of well-being in ICE settings can
inform health surveillance efforts in long-duration spaceflight settings and can contribute
to long-duration spaceflight selection and training procedures.
Using the hedonic well-being, eudaimonic well-being, and PsyCap framework, future
research should aim to determine which specific well-being indicators demonstrate
significant relationships with performance and psychosocial health criteria in ICE
settings.
Next, specific well-being predictors should simultaneously be examined using a
regression framework to determine the incremental effects of each on outcomes of
interest. Doing so will identify which factors are most efficient in predicting specific
outcomes deemed most vital to long-duration mission success (e.g., task performance,
group functioning, stress resistance).
Indicators of PsyCap are more strongly related to direct measures of performance than
either hedonic well-being or eudaimonic well-being. Findings suggest that emphasis
should be placed on these indicators of well-being in order to explore the potential
preventive and salutogenic effects of resilience and resilience-based protective factors
during long-duration ICE missions.
HRP is currently undertaking health surveillance efforts to identify the signs, symptoms,
and diagnoses of neurobehavioral health in ICE and spaceflight settings. HRP would
benefit by supplementing this effort or similar future efforts with an examination of the
relationship between indicators of well-being and neurobehavioral health in such settings.
Doing so would provide a comprehensive study assessing the ability of well-being
indicators to predict adverse health and behavioral outcomes.
The research presented in this report suggests that specific indicators of positive and
negative behavioral health likely follow different temporal trends over the course of ICE
missions. Assessing how specific indicators of well-being and symptomatology covary
5
and/or differ over the course of a mission can be used to develop countermeasures to
enhance specific aspects of well-being, to counteract the negative effects of stress and
symptomatology, or both, at various points during a mission.
Studies to validate ICE-specific measures of well-being can be conducted in a variety of
settings. Earth-based analogs provide the most efficient settings for validation studies to
take place initially. These studies could be carried out among individuals and teams in
analog settings such as Polar expeditions; submerged settings such as NASA Extreme
Environment Mission Operations (NEEMO); and analogous Mars settings such as the
Hawai’i Space Exploration Analog and Simulation (HI-SEAS).
Ultimately, low Earth orbit settings such as the International Space Station (ISS) will
provide the most suitable analog for inquiries related to long-duration exploration-class
missions. Examining the relationships between well-being and performance, and well-
being and health in this context will allow researchers to validate a comprehensive
measure of well-being for use in a long-duration mission to Mars.
2. Increase focus on the well-being of ground support and family members:
NASA is currently investigating the applicability of resilience training for ground crews.
It is recommended that NASA continues this effort and also assesses the effects of
maintaining and enhancing well-being among mission controllers. Understanding and
enhancing the well-being of crew-mission control may increase the communication
effectiveness of ground crews, which may ultimately serve to enhance crewmember
performance.
Families serve as a critical point of support for individuals in ICE settings. To date, very
little research has examined the well-being of crew-members’ families. Even less
research has examined how family well-being can impact the well-being and
performance of crew members. Work should be done to examine the effects of
crewmember family well-being on crewmembers’ psychosocial functioning and
performance.
3. Continue to assess well-being of crew members after the mission has ended:
Research is needed to assess lasting effects of ICE missions on crewmember well-being.
This information can be used to inform and develop post-mission support procedures, if
necessary.
Research is needed to identify post-mission factors of healthy psychosocial functioning
and growth to inform crew selection for future long-duration missions.
Assessing the post-mission well-being of astronauts and individuals in ICE settings will
be especially critical to understanding the potential impacts of multiple flights/ICE
missions on well-being.
6
Introduction
The first manned mission to Mars is expected to become a reality within the next 25 years
(Steinberg, Kundrot, & Charles, 2013). Although this may seem a distant goal, a complex set of
hurdles must be overcome for this prospective expedition to be successful. One of the greatest
priorities is to ensure crewmembers’ psychosocial health and effective functioning on such a
mission. However, because a mission of this kind has never been attempted, predictions of what
factors are most relevant need to be grounded in research of individuals in similar environments.
Literature on isolated, confined and extreme (ICE) environments has been quite comprehensive
in identifying the factors of long-duration missions that can induce stress, both in space
(Morphew, 2001) and analogue settings (e.g., polar stations; Palinkas & Suedfeld, 2008).
Considerable concern has been raised over the adverse effects these factors can have on
crewmembers’ health and performance, as well as on overall mission success (Bishop, 2004).
The unique features of a mission to Mars only exacerbate concerns over individual
crewmembers’ health. The initial Mars expedition is expected to exceed two-and-a-half years
(Steinberg et al., 2013). To date, the 18-month Mars500 space simulation, completed by a six-
person international team (Basner et al., 2013; Belavý et al., 2013; Morukov et al., 2013), stands
as the longest-duration ICE crew to be studied, while the longest continuous spaceflight stands at
approximately 14 months, completed by Russian cosmonaut Valeri Poliakov (Manzey, Lorenz,
& Poljakov, 1998). Thus, a Mars expedition represents a mission duration far exceeding that of
anything for which effects on psychosocial health have been previously studied. However,
mission duration is not the only unprecedented characteristic of such an expedition. Given the
physical distance of Mars from Earth and limitations of current technology, a Mars expedition
crew must also cope with significant communication delays (20-45 minutes, one-way)
7
throughout the bulk of the mission, which can lead to miscommunication and frustration (Fischer
et al., 2014). Related, the mission will require crewmembers to manage excessive periods of free
time intermixed with intense periods of work. Finally, crewmembers will be the first humans to
experience the “Earth-out-of-view phenomenon,” as the crew journeys beyond Earth’s orbit on
its way to Mars. The effects of these unique circumstances on crewmembers are currently
unknown (Kanas & Ritsher, 2005), but they will likely test individual limits like never before,
and the importance of maintaining psychosocial health and functioning to mission success is an
issue that cannot be overstated (e.g., Davis, Fogarty, & Richard, 2008; Dawson, 2002; Inoue,
Matsuzaki, & Ohshima, 2004).
The purpose of this report is to contribute to current knowledge regarding crewmember
psychosocial health and functioning in long-duration spaceflight. More specifically, this report
aims to help address Gap BMed 5, as identified by the Behavioral Health Program (BHP)
element of the NASA Human Research Program (HRP): “What individual characteristics predict
successful adaptation and performance in an isolated, confined and extreme environment,
especially for long duration missions?” This report will help address this research gap in
multiple ways. First, the ICE literature has placed far greater emphasis on studying psychosocial
dysfunction in contexts relevant to long-duration spaceflight than on psychosocial well-being.
However, positive psychosocial functioning, as opposed to the simple absence of dysfunction,
may have important consequences for individual and team health and functioning, as well as
overall mission success. Thus, the first purpose of this report is to identify factors of well-being
within the broader psychology literature that are most likely to be relevant to long-duration
spaceflight (Aim 1). Using this framework, this report systematically summarizes the state of
current evidence regarding the effects of well-being on performance in ICE environments (Aim
8
2). Finally, there are important questions that need to be answered regarding the extent to which
and how individuals’ psychosocial health and functioning change across the duration of ICE
missions. A number of primary studies have sought to answer these questions, and a variety of
temporal change patterns have been observed. Thus, the final purpose of this report is to review
existing evidence regarding temporal change in psychosocial health indicators throughout the
duration of ICE missions (Aim 3).
Effects of ICE on Psychosocial Health and Performance
ICE environments present a number of unique challenges which may adversely affect
individuals’ psychosocial health and performance. Scholars have identified an extensive list of
potential stressors associated with spaceflight and analogue ICE settings (e.g., Davis et al., 2008;
Dawson, 2002; Lapierre, Bouchard, Martin, & Perreault, 2009; Mophew, 2001; Nechaev,
Polyakov, & Morukov, 2007), which can be organized succinctly into three categories:
physical/environmental, social, and psychological (Geuna, Brunelli, & Perino, 1996).
Physical/environmental stressors include: ambient noise, hardware/computers, long-duration
confinement, microgravity, radiation exposure, and inability for rescue (e.g., Kelly & Kanas,
1992; Pattyn et al., 2009). Social stressors include: isolation, lack of privacy, demographic
heterogeneity, language barriers, and communication issues/lack of communication with ground
control, loved ones, and Earth (e.g., Boyd et al., 2009; Weiss & Moser, 1998; Stuster, Bachelard,
& Suedfeld, 2000). Psychological stressors include: periods of boredom and monotony and
periods of intense performance demands (e.g., Gushin, Zaprisa, Kolinitchenko, & Efimov, 1997).
Additional sources of stress specific to long-duration spaceflight have also been proposed to
include comparatively extreme mission duration/distance, communication delays, increased
9
levels of autonomy and responsibility, and amount of free time, as well as the Earth-out-of-view
phenomenon (Kanas & Ritsher, 2005).
In general, long-term isolation and confinement has been shown to lead to maladjustment
and decrements in psychosocial health through increased anxiety, depression, fatigue, and stress
(e.g., Dinges et al., 2014; Palinkas, Cravalho, & Browner, 1995; Palinkas & Houseal, 2000;
Stuster et al., 2000). Prolonged isolation and confinement has also been linked to reductions in
cognitive control and physiological reactivity in these contexts, mitigating cognitive (Pattyn et
al., 2009) and physical performance (Belavý et al., 2013).
Negative effects on interpersonal relations have been those most consistently evidenced
within the ICE literature (e.g., Kanas, 1991; Kanas et al., 2001c; Stuster et al., 2000). For
example, interpersonal tension among crewmembers has been found to be driven by individual
differences in crewmember values (Sandal, Bye, & van de Vijver, 2011), personality (Bishop,
Santy, & Faulk, 1998; Inoue et al., 2004), and culture/language (Inoue et al., 2004), as well as
due to competition over group roles (Krins, 2009; Sandal, 2001, 2004). These issues can have
serious consequences on group performance and overall mission success. To demonstrate,
during the 105-day Simulation of the Flight of International Crew and Space Station-1999
(SFINCSS-99) a physical fight broke out among two of the crewmembers, a sexual harassment
incident was reported, and one protesting crewmember withdrew from the study (see Inoue et al.,
2004). In the context of spaceflight, where individual escape or mission termination is rarely an
option, events such as this will certainly place individual psychosocial health and performance,
as well as mission success, in extreme jeopardy. Related to interpersonal issues among
crewmembers, issues regarding leadership have also been evidenced. For example, one
simulation crew leader became increasingly dominant over the mission duration, leading to
10
progressively lower leadership evaluations (e.g., Bergan, Sandal, Warncke, Ursin, & Værnes,
1993; Kansas, 2005). Finally, interpersonal tension has also been observed between crews and
mission control (e.g., Bergan et al., 1993; Inoue et al., 2004; Kanas, 2004, 2005; Kanas,
Salnitskiy, Grund, Gushin, & Weiss, 2002; Kanas et al., 2007a, 2007b, 2007c; Lapierre et al.,
2009; Stuster et al., 2000). Tension between the crew and leader and crew and mission control
have important consequences for the execution of mission tasks, via effective communication,
and mission success (Bergan et al., 1993).
Finally, it is important to note that the adverse effects of isolation and confinement likely
do not end with the mission. As described below, the potential for adverse effects following
experiences in extreme environments has driven much of the military psychology literature (e.g.,
Hoge et al., 2004), and evidence has demonstrated the consequences of long-duration spaceflight
on crewmember health post-mission, both in terms of changes in personality and family
dynamics (see Kanas, 1998). Taken together, evidence suggests the importance of developing a
better understanding of the effects of ICE missions on psychosocial health and performance (e.g.,
Davis et al., 2008) in order to improve selection, training, and countermeasure procedures and
mitigate the potential for adverse effects during such missions.
Positive Consequences of ICE Experiences
The stress associated with long-duration spaceflight is expected to test crewmembers’
psychosocial functioning. However, some scholars have suggested that there exists a bias within
the ICE literature towards overgeneralizing and dramatizing the potential adverse effects of such
missions (Mocellin & Suedfeld, 1991), and a growing body of literature has evidenced the
positive effects ICE experiences can have on individuals’ psychosocial functioning (see Suedfeld
& Steel, 2000). This counterargument appears plausible when one considers the screening
11
process individuals typically must pass prior to being selected for ICE missions. Particularly in
the context of long-duration spaceflight, increasing emphasis is being placed not only on
selecting candidates with the “right stuff” (e.g., Bishop, 2004), but also on preflight training
aimed at improving crewmembers’ ability to effectively manage the stress that will inevitably
arise due to the circumstances of spaceflight (e.g., Rose, 2014). Moreover, advances are being
made in the ability to monitor crewmembers’ psychological and physiological functioning during
spaceflight in order to provide countermeasures when necessary (e.g., Mollicone et al., 2014).
This implies that crewmembers may be better adjusted and equipped to handle the elevated risk
associated with ICE environments, and these individuals may also be more likely to view ICE
conditions as challenges from which they can benefit psychosocially.
A review of polar expedition groups indicates generally high well-being among team
members across missions, as well as generally low anxiety, depression, and stress reactivity
(Leon, 1991). For example, crewmembers completing a seven-week high-Arctic expedition
reported low levels of stress throughout the majority of the expedition, as well as generally high
mood, satisfaction with and enjoyment of the expedition, and camaraderie (Leon, List, & Magor,
2004). Similar effects were found among a two-woman team completing a 97-day Antarctic
expedition, including high positive personality dimension scores (i.e., absorption, work and
positive expressivity) and low verbal aggressiveness scores (Atlis, Leon, Sandal, & Infante,
2004). In addition, interview responses by team members depict how enjoyment and awe of
surroundings increased psychological strength, and one member indicated feelings of a sensed
presence, which was reported to provide motivation during periods of fatigue and
disappointment. Palinkas and Suedfeld (2008) identified a number of positive characteristics
among Antarctic winter-overs, that may be relevant to long-duration spaceflight: excitement over
12
experiencing the unknown; free time to self-improve, exercise, and think; and the opportunity to
remove oneself from daily hassles and negative aspects of life on Earth. The authors went on to
posit a number of salutogenic effects that may result from such missions, including: increased
affiliation with other crewmembers; a sense of achievement, cooperation, excitement and
curiosity; resilience, individual growth, and self-efficacy; as well as changes in worldviews.
Consistent with this, unanimous positive reactions to spaceflight have been reported among
surveyed astronauts and cosmonauts (Ihle, Ritsher, & Kanas, 2007; Ritsher, Ihle, & Kanas,
2005), and in autobiographies, memoirs, interviews, personal diaries, and oral histories of
astronauts of various nationalities (Suedfeld & Brcic, 2011). In addition, astronauts and
cosmonauts taking part in MIR and International Space Station (ISS) missions reported higher
positive mood scores and greater cohesion than was found in a normative sample (Kanas et al.,
2001a), and despite physical performance decrements over the course of the 520 day Mars500
simulation, individuals’ motivation remained consistently high (Belavý et al., 2013). Taken
together, this suggests that experiences in ICE environments can be positively associated with
well-being. However, it is less clear how well-being affects psychosocial functioning within
these environments.
Well-being in Long-duration Spaceflight
Definition and Conceptualization of Well-being
Within psychology, far greater attention has been paid to identifying and treating
psychological dysfunction and disorder than to improving healthy functioning and well-being
(e.g., Myers & Diener, 1995; Ryan & Deci, 2001; Ryff, 1995); research on psychosocial factors
in ICE has been consistent with this trend. In this vein, well-being has often been equated with
the absence of psychosocial symptomatology. However, well-being is not simply the absence of
13
the negative, but instead the presence of positive experiential and emotional characteristics
(Ryan & Deci, 2001; Ryff, 1995; Sin & Lyubomirsky, 2009). Within the field of psychology,
research has periodically emphasized the importance of positive psychosocial functioning and
well-being. For example, humanistic psychology enjoyed popularity during the middle of the
20th century (e.g., Maslow, 1971). The 1980s saw the rise of research exploring the effects of
self-esteem (Rosenberg, 1979) and positive affect (Watson, Clark, & Tellegen, 1988), and with
the 21st century came the field of positive psychology (Seligman & Csikszentmihalyi, 2000),
focusing on positive traits, emotions and experiences.
There is no shortage of descriptions of what it means to be well. For example, well-being
has been defined through indicators such as happiness, positive affect, positive emotions, mental
health, global or domain satisfaction (e.g., job), self-realization, autonomy, and purpose and
meaning in life (see Sin & Lyubomirsky, 2009; see also Frederickson & Joiner, 2002; Wright &
Huang, 2012). Many definitions of well-being include a complex set of these and other relevant
indicators. For example, one recently-proposed model describes well-being as a multi-faceted
construct that includes: competence, emotional stability, engagement, meaning, optimism,
positive emotion, positive relationships, resilience, self-esteem, and vitality (Huppert & So,
2013). To accommodate these various, and often complex, descriptions, well-being has also
been defined as anything affective, behavioral, or cognitive, that is important in an individual’s
life (Rath & Harter, 2010). Research has also aimed to identify the overarching facets of well-
being, and two overarching components—hedonic and eudaimonic well-being—have emerged
throughout much of the literature (see Ryan & Deci, 2001). This has led to differences in
explanations of what it means to be well. Nonetheless, both hedonic and eudaimonic well-being
are likely to be relevant in the context of long-duration spaceflight, and the dichotomy provides a
14
framework for organizing some of the specific indicators described above. In addition, evidence
from the broader psychology literature suggests that both factors relate to performance.
Hedonic Well-being
The hedonic facet represents the pleasure-driven qualities of well-being (e.g., Diener,
1984; Wright & Doherty, 1998). Potential indicators of hedonic well-being include: happiness;
positive affect, emotions and mood; and life, social and work satisfaction. Researchers taking
the hedonic approach have most often assessed well-being, or “happiness”, through the presence
of life satisfaction and positive mood and the absence of negative mood, what Diener and
colleagues describe as subjective well-being (SWB; e.g., Diener, 1984; Diener, Suh, Lucas, &
Smith, 1999). Further, it is theorized that hedonic indicators can be broken down into life
evaluations (i.e., satisfaction) and positive emotions (Kahneman & Deaton, 2010).
Evidence has demonstrated hedonic indicators to be related to a wide range of relevant
outcomes. For example, the robust effects of positive affect/happiness have been demonstrated
meta-analytically, with evidence indicating that happier people experience greater success at
work (r = .27) and in relationships (r = .27) and are healthier (r = .32; Lyubomirsky, King, &
Diener, 2005). In the context of long-duration spaceflight, hedonic indicators are likely most
influential on crewmembers’ day-to-day functioning, such as in maintaining positive
interpersonal relationships and effective task performance.
Eudaimonic Well-being
The eudaimonic facet represents the self-realization-driven qualities of well-being (e.g.,
Deci & Ryan, 2000; Ryff, 1989, 1995; Ryff & Keyes, 1995; Ryff & Singer, 1998). Potential
indicators include autonomy; competence; engagement, meaning and purpose in life and work;
positive relationships; self-realization; spirituality; and vitality. While hedonic well-being is
15
typically assessed through SWB, eudaimonic well-being is most often assessed through
psychological well-being (PWB; Ryff, 1989), often characterized as “self-fulfillment” reflected
in self-acceptance, positive relations and growth, and meaning/purpose in life (Keyes, Shmotkin,
& Ryff, 2002; Ryff, 1989). Self-determination theory (SDT; Deci & Ryan, 2000) represents a
related conceptualization of eudaimonic well-being. That is, SDT holds sufficiently met needs
for autonomy, competence, and relatedness as the core of well-being (Ryan & Deci, 2001).
Eudaimonic indicators have also been shown, in the broader literature, to be related to
outcomes believed to be important in spaceflight. For example, meta-analytic evidence has
demonstrated work engagement, autonomy, and perceived competence to be positively related to
work commitment (Halbesleben, 2010; Mathieu & Zajac, 1990); work engagement, autonomy
and indicators of positive relations have been demonstrated to be negatively related to burnout
(Halbesleben, 2010; Lee & Ashforth, 1996). Thus, in the context of long-duration spaceflight,
eudaimonic indicators likely have the greatest impact from a longer-range perspective through
commitment to the mission and sustained motivation and psychosocial functioning.
Psychological Capital (PsyCap)
There exist in the psychological literature a number of multi-dimensional constructs
indicative of well-being, beyond those captured by the definitions of SWB (i.e., hedonic) and
PWB (i.e., eudaimonic). Due to the work-related nature of ICE missions, one such construct that
may be particularly relevant is psychological capital (PsyCap; Luthans, 2002; Luthans, Youssef
& Avolio., 2007). Positive organizational behavior, an emerging field in the organizational
literature, emphasizes identifying, developing and maintaining indicators of healthy employee
functioning and well-being, specifically in the work context, as a means of optimizing individual
and organizational performance (e.g., Luthans, Avolio, Avey, & Norman, 2007). Much like
16
SWB and PWB, PsyCap has been argued to be a higher-order construct indicative of well-being
in work contexts assessed through four lower-order constructs: hope, efficacy, resilience, and
optimism. Meta-analytic evidence indicates a significant relationship between PsyCap and job
performance (Avey, Reichard, Luthans & Mhatre, 2011). Moreover, research has also suggested
that PsyCap is associated with lower levels of perceived stress (Avey, Luthans, & Jensen, 2009)
and higher levels of psychological well-being (Avey, Luthans, Smith & Palmer, 2010). Meta-
analytic reviews of individual components of PsyCap have demonstrated relationships with both
work-related performance (self-efficacy, r = .38; Stajkovic & Luthans, 1998) and psychological
adjustment (efficacy, ρ = .58; optimism, ρ = .41; Lamp, 2013).
1
Thus, the potential contribution
of PsyCap indicators to overall well-being during long-duration spaceflight, beyond that of SWB
and PWB, is likely shown through the improved ability to adapt to ICE environments and to
buffer the effects of stressors that may arise throughout the mission.
Summary
Overlap certainly exists between the facets described above. For example, evidence
suggests SWB and PWB to be related-but-distinct constructs (Keyes et al., 2002). However, this
trichotomization of higher-order well-being factors provides a parsimonious model by which
lower-order well-being indicators relevant to ICE contexts may be conceptually organized. As
described above, limited research attention has been given to well-being within the ICE
literature. However, where possible, we differentiate between these facets to provide insight as
to which types of indicators have received greater attention in the literature, as well as whether
these facets show differential temporal changes or have differing effects on relevant outcomes.
1
Rho (ρ) represents the meta-analytic correlation (r) corrected for measurement error (see Hunter & Schmidt, 1990).
17
Relationship between Well-being and Performance
Factors that have been identified to positively affect performance in ICE environments
include: age, experience, high emotional stability/maturity, expressiveness, instrumentality, and
motivation, effective social skills and leadership, low extraversion, conscientiousness, and
assertiveness, low need for affection, and an absence of anxiety and depression (Palinkas,
Gunderson, Holland, Miller, & Johnson, 2000; Palinkas, Keeton, Shea, & Leveton, 2011;
Rosnet, Le Scanff, & Sagal, 2000; Sandal, Værnes, Bergan, Warncke, & Ursin, 1996). Many of
these predictors have also been shown to be positively related to well-being in the broader
psychology literature. For example, research employing large national samples indicates age and
emotional stability to be positively related to both PWB and SWB (Keyes et al., 2002; Soto,
2014). Predictive analyses among a similarly large sample demonstrated that individuals high in
eudaimonic well-being were more than two times less likely to be depressed 10 years later, even
when controlling for personality and economic factors, physical and mental health, and history of
depression (Wood & Joseph, 2010). Given the disparities across definitions of well-being, it has
also been argued that many of the factors described above are, themselves at least, distal
indicators of well-being (e.g., Huppert & So, 2013).
However, the practicalities of ICE environments lead to some unique considerations
regarding the expected relationships between psychosocial predictors, indicators of well-being
and performance. For example, meta-analytic evidence from the broader organizational
psychology literature suggests conscientiousness to be a robust predictor of performance across
occupational groups, and extraversion to be an efficient predictor of training proficiency (Barrick
& Mount, 1991). However, these characteristics have been shown to have the opposite effect on
performance in ICE environments (see Palinkas et al., 2011). That is, in ICE settings, highly
18
conscientious individuals are more likely to become frustrated when variable external factors and
limited resources hinder their ability to successfully complete tasks, and extraverted individuals
may find it difficult to adjust to the limited social stimuli associated with ICE settings (Palinkas,
et al., 2000). Further complicating the matter is that evidence from a large national sample has
indicated that conscientiousness and extraversion, along with agreeableness and emotional
stability, positively predicted subsequent SWB scores (Soto, 2014). Thus, it remains unclear
how the unique effects of extraversion and conscientiousness influence the relationship between
well-being and performance in long-duration spaceflight.
Literature examining the effects of well-being on performance in ICE settings remains
relatively scarce. This may be largely the result of methodological limitations of research in ICE
contexts. First, evidence of psychosocial health in ICE environments has thus far focused on
individual ICE teams, with few studies examining multiple teams simultaneously (e.g., Inoue et
al., 2004). This has resulted in a complete absence of evidence comparing between-group
differences in psychosocial health and their effects on performance. Second, the development of,
and access to, valid performance criterion measures has long been an issue within the
psychology literature (see Austin & Villanova, 1992), an issue intensified in ICE psychosocial
research. Objective measures of individual performance may often not be practical in these
contexts, and these measures are noticeably absent in ICE research studying indicators of well-
being, particularly in ICE settings beyond that of Antarctic stations. Unfortunately, the use of
subjective performance measures (e.g., supervisor/peer ratings) has not been common within the
body of literature studying well-being ICE environments either. Taken together, this suggests
that future research is needed that directly examines the well-being-performance relationship,
both at the individual and team levels.
19
Meanwhile, expanding the pool of criteria to proxies more distally indicative of
performance, as well as relying on concurrent self-report measures of well-being and
performance-related evidence may be necessary to begin to understand this relationship. With
regard to indicators of hedonic well-being, correlational evidence from the Mars-105 pilot
mission has demonstrated positive affect to correlate strongly with self-reported problem-solving
(Nicolas, Sandal, Weiss, & Yusupova, 2013), which may be indicative of effective performance.
Among an Antarctic expedition team, those who reported high work satisfaction also reported
high group cohesion (Sarris & Kirby, 2005). Expanding the criteria pool limits our ability to
draw firm conclusions regarding the direct well-being-performance relationship or the nature of
causality among variables, but it may provide important insight moving forward by establishing
which well-being indicators and performance criteria may be most fruitful to pursue.
In addition, it is often assumed that individuals selected to take part in ICE missions are
well-adjusted, motivated individuals. Thus, it may be expected that effective performance (e.g.,
problem-solving, task performance, effective communication and leadership, group performance
and cohesion, and overall mission success) may be at least partially due to individuals’ high
standing on indicators of well-being. In this vein, evidence has shown that well-adjusted
individuals, based on pre-mission assessments, remained motivated to complete a 56-day
expedition to the North Pole, despite extreme physical conditions that resulted in two members
of the team to be flown out of the mission early due to injuries (Leon, McNally, & Ben-Porath,
1989).
Evidence suggesting a relationship between well-being and performance is not
completely absent from the literature. Within ICE environments, indicators of eudaimonic well-
being, specifically autonomy and positive relations, have been most frequently studied and
20
shown to relate to performance criteria (proximal and distal). For example, allowing
crewmembers taking part in a 105-day space simulation the autonomy to plan their own
schedules led to an increase in perceived group cohesion (Gushin et al., 2012; Kanas et al.,
2010). In addition, emotional support and respect for others were identified as critical to mission
success among multiple teams carrying out polar expeditions (Leon & Sandal, 2003). Strong
social networks have also been shown to be related to effective emotional adjustment during
winter-overs in Antarctica (Palinkas & Johnson, 1990), and among surveyed astronauts and
cosmonauts, contact with loved ones has been reported to positively affect mission performance
(Kelly & Kanas, 1993). Finally, evidence demonstrates the positive relationship between leader
support and crew cohesiveness among Shuttle and Mir astronauts and cosmonauts (Kanas &
Ritsher, 2005). Alternatively, the lack of strong positive relations among crewmembers has been
demonstrated to have detrimental effects on crewmember functioning and overall mission
success (e.g., Inoue et al., 2004; Sandal, 2004).
Well-being in Extreme Contexts: Research on Military Personnel
The major stressors associated with military service likely differ from those associated
with isolation and confinement. Nonetheless, the military context and associated combat
operations represents an extreme and unusual environment which may yield useful information
for understanding well-being in spaceflight and ICE contexts. Moreover, considerable efforts
have been made within the military context to understand the range of factors which influence
psychosocial health and performance. These efforts may be best represented by the Millennium
Cohort Study in which over 75,000 U.S. service members enrolled in the initial stages of the
study in 2001, and a substantial number of whom continue to be tracked, at present. Evidence
originating from Millennium Cohort data demonstrates a wide range of relationships between
21
baseline health and a variety of health and behavioral outcomes. For instance, this research has
shown that baseline health can predict vulnerability to post-deployment stress disorders
(Gehrman et al., 2013; LeardMann, Smith, Smith, Wells, & Ryan, 2009). These findings speak
to the importance of selecting individuals who are both physically and psychologically able to
take on critical missions such as deploying to a combat zone. However, it must be noted that one
Millennium Cohort study found psychological resilience (PsyCap indicator) to be quite common
among deployed service members, as over 80% of soldiers indicated no elevated risk levels of
post-traumatic stress following a deployment (Bonnano et al., 2012).
Consistent with the broader psychological literature, researchers studying psychological
health and well-being in the military have typically focused on the negative aspects of health, or
the absence of well-being. This is illustrated by recent Department of Defense (DoD) reports
that have focused on the rates of psychological problems experienced by deployed and non-
deployed service members (MHAT 6, 2009; J-MAT 7, 2011), as well as the broader literature on
stress and its relationship with health and job performance data (e.g. Pflanz & Ogle, 2006).
Recently, however, research has begun to focus on the positive indicators of
psychological health. The U.S. Army’s Comprehensive Soldier and Family Fitness (CSF2)
program (see Casey, 2011; see also Cornum, Matthews, & Seligman, 2011), for example,
focuses on assessing positive psychological attributes, including indicators of hedonic (e.g.,
positive affect) and eudaimonic well-being (e.g., character strengths, engagement, social
connections, spirituality) and PsyCap (e.g., adaptability, optimism), in soldiers. Research
conducted by CSF2 has demonstrated these indicators to be associated with performance-related
outcomes. Specifically, the research has shown soldier adaptability, coping styles, character
strengths, positive affect, optimism, engagement, social connections, and work engagement to
22
negatively relate to adverse outcomes, such as suicide, documented drug use, and violent
criminal offenses (Lester, Harms, Bulling, Herian, & Spain, 2011a). These indicators have also
been related to positive indicators of performance, such as: early promotion and selection for
special assignments (Lester et al., 2011b). However, it should be noted that these effects have
typically been quite small.
A limited amount of additional research has examined indicators of well-being and their
relationships with performance outcomes, and not all evidence has demonstrated a strong well-
being-performance relationship. For example, the relationship between coping strategies and a
variety of job-related stressors, such as role overload, role ambiguity, and lack of stimulation at
work has been examined, with findings suggesting “positive” coping strategies (e.g., problem-
focused coping, positive emotion-focused coping, and seeking information and emotional
support) to be unrelated to health and behavior complaints (Day & Livingstone, 2001). The use
of “negative” coping strategies (venting of emotions and denial/disengagement), however, led to
a greater incidence of complaints. Most importantly, interactions demonstrated that the use of
venting and disengagement had the potential to exacerbate the effects of both acute and chronic
stressors upon physical and behavior health complaints. Taken together, these findings suggest
that avoiding “negative” coping strategies may be more important than using of “positive”
coping strategies to reduce the odds of health complaints among soldiers.
On the other hand, evidence has demonstrated the potential effects of optimism on
mitigating the onset of adverse psychosocial health (Thomas, Britt, Odle-Dusseau, & Bliese,
2011). That is, dispositional optimism was found to reduce post-deployment posttraumatic stress
disorder (PTSD) symptoms, depression symptoms, and work impairment. Dispositional
optimism was also found to interact with warzone stressors such that high levels of dispositional
23
optimism reduced the effect of warzone stress upon PTSD symptoms. Analyses also showed
that dispositional optimism attenuated the relationship between deployment demands and
depression symptoms, and also attenuated the effects of PTSD symptoms and depression
symptoms upon the probability of work impairment. In short, dispositional optimism was found
to be an important buffer against the effects of mission-related stressors.
Finally, studies have also examined perceptions of leadership and social support as
indicators of well-being in military contexts. Recent DoD reports identified leadership and unit
cohesion as two of the primary buffers against the stressors of combat (MHAT 6, 2009; J-MHAT
7, 2011). Organizational support has also been identified as a buffer against combat stress and
depression and anxiety (McFadden, 2013). Other research has also found that perceptions of
social support can enhance well-being and job satisfaction among military personnel (Limbert,
2004).
Summary
There has been a dearth of research examining the relationship between well-being and
performance in either the ICE or military literatures. However, expanding the pool of relevant
criteria to more distal correlates of performance (e.g., psychosocial and behavioral problems)
being studied in ICE settings may provide greater insight into the well-being-performance
relationship. In this regard, research in the military context has the potential to inform the
research initiatives of NASA, as the focus in the military literature has traditionally been to
mitigate the likelihood of soldiers developing pathologies (e.g., depression, posttraumatic stress
disorder), which may subsequently lead to performance decrements.
24
Temporal Effects on Well-being
Primary research in ICE environments has sought to demonstrate the temporal pattern of
psychosocial effects during isolation and confinement. Temporal effects on psychosocial health
have been studied using a wide range of variables representing both well-being and psychosocial
symptomatology. These variables include: mood, morale, personality, positive and negative
affect, satisfaction, sleep quality, and stress (Atlis et al., 2004; Bhargava, Mukerji, & Sachdeva,
2000; Bishop, Kobrick, Battler, & Binstead, 2010; Gushin et al., 2012; Koscheyev, Roschina, &
Makhov, 1994; Palinkas & Houseal, 2000; Steel, 2001; Wood, Lugg, Hysong, & Harm, 1999).
Researchers have often suggested the presence of a third-quarter phenomenon, wherein
adverse effects on psychosocial functioning peak between half and two-thirds of the way through
the duration of missions. The concept was developed based on anecdotal observations and
reports of negative mood in various ICE environments (Bechtel & Berning, 1991), and has
received some empirical support in analogue settings. For example, self-reported stress peaked
at the midway point of the mission among males taking part in a four-month ICE simulation at
the Flashline Mars Arctic Research Station (FMARS). Among female crewmembers, self-
reported stress and avoidant coping strategies also saw an uptick midway through the duration of
the simulation, before progressively receding throughout the remainder of the mission, while the
inverse trend was found for the use of healthy coping strategies (Bishop et al., 2010). Evidence
has also demonstrated interpersonal rapport among individuals wintering-over in the Antarctic
reached its lowest points during the third quarter of the stay (Bhargava et al., 2000), and similar
effects have also been found in relation to morale among an Antarctic research expedition crew
(Wood et al., 1999). Finally, content analyses of diary entries completed by individuals serving
25
at remote stations in the Antarctic and South Indian Ocean provide additional documentation of
this phenomenon (Stuster et al., 2000)
Research has demonstrated a number of alternative temporal change patterns. For
example, stable increases in anxiety and hostility have been observed among individuals
wintering-over in the Antarctic (Evans, Stokols, & Carrere, 1987; Sasaki et al., 1980), as have
ratings of stress among ISS astronauts (Dinges et al., 2014), and ratings of homesickness among
individuals taking part in polar expeditions (Sandal et al., 1996). However, Sandal et al. (1996)
also reported stable increases in coping and decreases in anxiety and depression among
hyperbaric chamber crews. Findings have also demonstrated non-linear effects that deviate from
the temporal pattern described as the third-quarter phenomenon, such as heightened stress at the
onset and end of a 25-day simulation, and low stress during the middle of the simulation (Bergan
et al., 1993; Mocellin & Suedfeld, 1991; Sandal, 2001). Similar effects were found among
individuals wintering-over at two Antarctic stations, who showed decreases in various negative
mood scales (anxiety, depression, anger, confusion, fatigue) throughout the austral winter, only
to report increases in anxiety and fatigue, as well as decreased vigor, during the latter part of the
mission (Palinkas & Houseal, 2000). Evidence from spaceflight missions has demonstrated a
novelty effect among astronauts that declined non-linearly across the duration of the mission
(Kanas et al., 2001a, 2001b, 2001c). Still other winter-over evidence has demonstrated a more
complex circannual rhythm, by which psychological and physiological systems create a feedback
cycle (Palinkas et al., 2001).
What should be made of this evidence suggesting various temporal change patterns in
ICE environments? It is possible differences in temporal patterns are driven by factors other
than time. That is, some research has found support for critical events being more indicative of
26
changes in psychosocial problems than time-based factors (e.g., Inoue et al., 2004; Leon et al.,
2004, Steel & Suedfeld, 1991). Related, researchers have begun to employ multilevel modeling
statistical analyses (Hox, 1998; Singer & Willett, 2003) to establish the amount of variance in
scores due to individual, group, and temporal effects. Using this approach research has
demonstrated that the majority of variance in weekly team climate and perceived social support
scores (57%-71%) among Antarctic expedition crews was due to individual differences, while
far less was due to temporal effects (Schmidt, Wood, & Lugg, 2004; 2005). However, evidence
has shown scores on different psychosocial health and well-being indicators to peak and trough
at different phases of the mission (e.g., Bhargava et al., 2000), as well as different indicators to
be more malleable to temporal change than others (Steel, 2001). Additionally, characteristics
associated with different settings may also influence temporal effects on psychosocial health
indicators. For example, the selection and training requirements, age, and average education
level, likely differ between spaceflight and many analogue settings. Moreover, differences in
mission duration and the physical and environmental characteristics of these settings have also
been described (e.g., Manzey, 2004). Given the variability across observed temporal trends in
psychosocial health, a key research objective is to better understand the underlying sources of
these differences (Aim 3).
Operational Tempo and Well-being in the Military
Given that one objective of the present report is to summarize the temporal effects on
indicators of psychosocial functioning over the course of a mission, it may be useful to briefly
review the literature on the impacts of deployment lengths upon these factors in the military.
Research has shown that simply being sent on a combat deployment can have a detrimental
impact on health and behavioral outcomes, irrespective of what happens during a deployment
27
(Fear et al., 2010; Polusny et al., 2009; Vasterling et al., 2010). Additionally, consistent with the
sensitization perspective of stress reactance, research has shown that longer deployment lengths
are positively related to depression and post-traumatic stress symptoms (Adler, Huffman, Bliese,
& Castro, 2005), and a review of multiple studies showed that deployment lengths are predictive
of decreased physical and psychological well-being among service members deployed to combat
(Buckman et al., 2011).
Only a limited amount of research has focused on the effects of dwell time—the time
between combat deployments—on mental health among soldiers experiencing multiple
deployments. A recent study of Marines showed that a greater ratio of dwell-to-deployment time
served to lower the risk of developing mental health issues following a second deployment
(MacGregor, Han, Dougherty, Galarneau, 2011), presumably because longer dwell times reduce
the number of threat appraisals and allow for service members to replenish the pool of resources
used to effectively cope with stress. In contrast, other research has demonstrated that longer
dwell times may increase the likelihood of developing mental and physical health issues, with
longer dwell times being associated with PTSD, anxiety-related disorders, and depressive
disorders (Armed Forces Health Surveillance Center, 2011). This is, perhaps, because longer
dwell times allow service members time to fully readjust to normal life before having to fully
reestablish the “warrior” mentality necessary for effective psychosocial functioning while on
deployment. It should be noted that there is a lack of evidence within the military psychology
literature tracking temporal changes among soldiers throughout the course of a deployment.
Thus, evidence stemming from military personnel may not provide direct insight into Aim 3.
However, indirectly, evidence from the military literature does suggest that temporal changes in
psychosocial functioning, such as increased stress and anxiety, can appear during the lead-up to
28
long-duration deployments and post-mission, during the period of reassimilation into normal
living conditions (e.g., Bliese, Wright, Adler, Thomas, & Hoge, 2007). This knowledge may be
important to the interpretation of temporal patterns found during ICE missions, particularly at the
beginning and end of missions.
Summary: Research Aims of the Present Report
Primary evidence exists regarding the relationship between well-being and performance
in ICE settings as well as for the temporal effects on well-being and psychosocial functioning
over ICE mission duration. However, no attempt has been made thus far to systematically
review the evidence of either. With regard to the former, the lack of summary evidence may be
due to the absence of a theoretically-supported organizing framework of the range of well-being
indicators that have been studied in ICE environments, and/or the absence of evidence linking
well-being indicators to measures directly reflecting performance. In the present report we
address these issues by organizing well-being indicators into three distinct factors (hedonic well-
being, eudaimonic well-being, and PsyCap) and qualitatively reviewing the effects of each in
relation to a broad set of performance-related criteria. With regard to the latter, various temporal
trends have been observed in relation to indicators of both well-being and stress/symptomatology
within the empirical ICE literature. A quantitative summary of such evidence may contribute
greatly to the ongoing discussion of the temporal effects on these variables. However,
differences in both the measurement scales used and length of mission studied in primary ICE
research of temporal effects has made summarizing this data difficult. Therefore, we use an
innovative approach of converting primary study response scale and mission duration data into
comparable and easily interpretable intervals to provide meta-analytic evidence that maps the
29
temporal trends of indicators of both positive (i.e., well-being) and negative psychosocial
functioning (i.e., stress/symptomatology) across both individuals and crews.
Method
Systematic Review
Again, the systematic review has two primary purposes: 1) to summarize the effects of
well-being indicators on performance-related criteria, and 2) to examine the temporal change in a
broader set of psychosocial factors (both well-being and psychosocial stress/symptomatology)
experienced by individuals across ICE missions. To identify potentially relevant studies we
conducted a systematic search of multiple databases: PsycInfo, Medline (First Search), and
Medline (PubMed). Prior to conducting our search we developed a broad and comprehensive set
of search terms to be used. We identified 21 primary search terms, which we paired with an
additional 23 secondary search terms. Primary and secondary search terms are presented in
Table 1. All systematic searches of key terms were conducted within study abstracts.
Using the terms depicted in Table 1, the PsycInfo search produced a total of 55,951
results. We found multiple primary search terms used in the PsycInfo search to return a vast
number of articles, most of which were not identified as potentially relevant, while other search
terms yielded very few articles, none of which were identified as potentially relevant. In total,
we identified seven primary search terms that were associated with one of the above issues.
Thus, these terms were eliminated prior to our conducting searches in the Medline First Search
and PubMed databases. Search terms retained for these subsequent searches are also noted in
Table 1. Our search of the Medline (First Search) and Medline (PubMed) databases produced an
additional 16,302 articles. This resulted in a total of 72,253 articles. It should be noted that a
number of articles were returned multiple times across our searches. Given the vast number of
30
articles our searches produced the exact number of duplicated articles contributing to this total
estimate is unknown. However, within our initial PsycInfo search we identified 580 potentially
relevant articles. Of these, 306 had already been retained through previous key term searches,
resulting in a total of 274 relevant articles from PsycInfo. Within our Medline searches a total of
303 potentially relevant articles were identified. However, 258 had already been retained
through previous searches, resulting in a total of 45 additional potentially relevant articles. The
total number of unique and potentially relevant articles identified through database searches was
319. In addition, we conducted a secondary search based on a review of the articles obtained via
database searches and identified an additional 45 articles that were potentially relevant. This
resulted in the identification of a total of 364 articles to be examined for possible inclusion.
Table 1. List of Search Terms
PsycInfo search
Retained
PsycInfo search
Retained
Antarctic
x
Adapt
x
Arctic
x
adaptability
x
astronaut
x
adaptive
x
capsule
x
adaptation
x
Concordia
x
Attitude
x
confined
x
conscientiousness
x
cosmonaut
x
citizenship
x
desert
Control
x
enclosed habitat
x
depression
x
extreme
emotion
x
Greenland
Health
x
ICE
individual difference
x
inextremis
x
mindfulness
x
isolated
x
Mood
x
long duration
x
Morale
x
McMurdo
x
performance
x
space
personality
x
spaceflight
x
resilience
x
submarine
x
satisfaction
x
unusual
self-esteem
x
unusual environment
x
self-efficacy
x
Stress
x
well-being
x
31
Inclusion Criteria
Again, this systematic review was designed to summarize the effects of well-being
indicators on performance-related criteria (Aim 2), and to examine the temporal change in a
broader set of psychosocial factors (both well-being and psychosocial symptomatology)
experienced by individuals across ICE missions (Aim 3). The inclusion criteria for studies
differed somewhat across these two purposes. However, as a general inclusion criterion, studies
were required to be conducted either within ICE environments (i.e., Antarctic or Arctic
stations/expeditions, deep sea dives, hyperbaric chambers, submarines, simulated space capsules,
bed rest, underground bunkers, and spaceflight) or through retrospective data collection efforts
regarding previous ICE experiences. Further inclusion criteria were specific to each of the two
purposes of this review, and are discussed below.
Aim 2: Well-being-performance relationship. Studies were required to report
quantitative evidence regarding the effects of indicators of well-being on performance-related
outcomes. As noted above, well-being is a broad construct so indicators were organized under
three separate factors based on the broader existing psychological literature: hedonic well-being,
eudaimonic well-being, and PsyCap. Variables and scales identified in the studies meeting Aim
2 inclusion criteria as indicators of well-being are classified according the three well-being
factors presented in Table 2. Emotional stability was identified as a broad indicator of well-
being that has also been used within the ICE literature, but does not reflect any specific factor
more than the others. Thus, we describe emotional stability as an indicator of general well-
being, and present evidence regarding this indicator separately after our review of evidence
regarding the effects of hedonic well-being, eudaimonic well-being, and PsyCap.
32
Table 2. Primary Study Variables Classified by Each Well-being Factor2
Hedonic
joy in work
perceived fairness
positive affect and negative affect (reverse coded)3
satisfaction (i.e., general or domain-specific)
vigor/activity
Eudaimonic
altruism
autonomy
competence (i.e., mature defense mechanisms, positive
instrumentality, perceptions of challenge)
control
drive
positive social relations (i.e., perceptions of fit, friendship,
group identification)
meaning in work (i.e., job morale; organizational commitment,
perceptions of job importance, job/work involvement)
positive self-image congruent with others’ perceptions of self
sense of coherence
Psychological capital
optimism
self-efficacy (i.e., personal and collective efficacy, self-
confidence)
resilience, ability to adjust successfully, and hardiness
Initially, Aim 2 of this review was to summarize the well-being performance relationship.
However, there were very few studies within the existing ICE literature linking well-being to
performance, as it is traditionally described in the broader literature (see Austin & Villanova,
1992). Thus, we expanded consideration to measures reflecting various work attitudes and
aspects of psychosocial functioning, which either may proximally relate to or subsequently
influence performance. These indicators were categorized according to two broad factors:
performance/work attitudes and stress/psychosocial symptomatology. Table 3 depicts our
2
Other indicators of well-being have been defined as representing hedonic well-being, eudaimonic well-being, and
PsyCap. For example, PsyCap includes a fourth component, hope. However, this table includes only those for
which empirical evidence exists within studies meeting inclusion criteria.
3
As reflected above, our position is that well-being is not simply the absence of negative psychosocial qualities and
states. However, subjective well-being (i.e., hedonic well-being) operationalized as a function of life satisfaction,
positive affect, and (the absence of) negative affect is well-established in the literature (see Diener et al., 1999).
Thus, in this instance we considered the absence of negative affect as an indicator of hedonic well-being. Further, it
should be noted that the absence of negative affect, in terms of indicating well-being, may be far removed from the
absence of, for example, depressive symptomology.
33
classification of relevant indicators that we were able to identify within included studies into
these two factors.
Table 3. Primary Study Variables Operationalized by Each Performance Factor4
Performance/Work
attitudes
amount of crew-mission control communication
adaptation to ICE environment
attitudes towards ICE experience (i.e., intentions to return for
a subsequent mission, motivation to return, whether or not
actually did return, willingness to recommend experience to
others)
crew cohesion and functioning (includes group conflict, group
polarization and alienation from the group)
peer and supervisor dimensional ratings of performance (e.g.,
leadership, task, interpersonal, overall)
performance-related trait characteristics (i.e., instrumentality,
mastery, work orientation, and achievement strivings)
psychomotor task performance
self-reported task performance accuracy and efficiency
team decision-making quality
work attitudes (e.g., role clarity/conflict, job satisfaction)
Stress/Psychosocial
symptomatology
anger/aggression/hostility
anxiety symptoms and sensitivity
confusion
depressive symptoms
fatigue
immature defense mechanisms
perceived stress
poor quality of life
posttraumatic stress disorder symptoms
work pressure
In order for studies to be included in our review satisfying Aim 2 of this report, studies
were required to report quantitative evidence regarding the relationship between at least one
indicator of well-being (Table 2) and one indicator of performance (Table 3). In total, we
identified 24 studies providing such evidence. However, three studies were identified as using
the same sample as a study already included, while other studies provided data sufficient for us
4
As with indicators presented in Table 2, Table 3 reflects only those variables for which evidence exists among the
included studies.
34
to extract multiple unique samples. Thus, the total number of independent samples (k) included
for achieving Aim 2 of this review was 31. Appendix A lists each of the included samples, as
well as depicts study characteristics (sample size, homogeneity of sample nationality, sample
sex, ICE setting, and mission duration) and the indicators of well-being and performance
examined.
Aim 3: Temporal Change in Psychosocial Factors. In order to examine temporal
effects included studies were required to report variable scores at a minimum of three distinct
time points over the duration of an ICE mission. Moreover, included studies reporting temporal
effects needed to do so for one of two types of indicators. The first type of indicator was that of
well-being (i.e., positive psychosocial functioning). The same broad set of indicators was
considered for Aim 3 as was considered as representing hedonic well-being, eudaimonic well-
being, or PsyCap for Aim 2. However, the focus of Aim 3 was on positive psychosocial
functioning, in general. Thus, we did not further separate indicators by well-being factor for
Aim 3 analyses. In addition, there is considerable overlap in the well-being indicators included
for Aim 2 (Table 2) and Aim 3. However, some of the indicators employed by studies meeting
the inclusion criteria for Aim 2 were not employed by studies meeting the inclusion criteria for
Aim 3, and vice versa. Thus, well-being indicators differ slightly across Aim 2 and 3. The
second type of indicator was that of stress or psychosocial symptomatology (i.e., negative
psychosocial functioning). The same broad set of indicators was considered for Aim 3 as was
considered as representing the stress/symptomatology factor for Aim 2 (Table 3). Again, there is
considerable overlap in the stress/symptomatology indicators included for Aim 2 and Aim 3, and
again, some indicators employed by studies meeting inclusion criteria for one aim were not
employed by studies meeting inclusion criteria for the other aim of this review. Thus, the
35
specific indicators employed differ slightly across Aim 2 and Aim 3 analyses. Taken together,
for studies to be included they were required to report scores at a minimum of three time points
on at least one indicator reflecting either positive (well-being) or negative psychosocial
functioning (stress/symptomatology). The variables identified among included studies and
included in Aim 3 analyses are reported, by study, in Appendices B and C.
Individual-level and Group-level Change. For Aim 3 we conducted two sets of
analyses. Given that studies of ICE missions often include small crews (e.g., ≤ 6 crewmembers)
there was an opportunity to combine individual scores across a series of k = 15 studies into a
single dataset in order to examine temporal effects on psychosocial functioning throughout the
duration of ICE missions without losing the variability associated group-level data aggregation.
This resulted in a single sample containing data for N = 67 individual ICE crewmembers.
Appendix B lists each of the included samples for individual-level temporal change analyses, as
well as depicts study characteristics (sample size, homogeneity of sample nationality, sample
sex, ICE setting, and mission duration) and the positive and negative psychosocial functioning
indicators examined.
Although the unique sample characteristics of ICE missions allowed for the examination
of individual-level temporal effects, the majority of studies did report temporal effects only at the
mean level. Thus, we subsequently analyzed temporal effects at this level by including mean-
level data from 32 studies providing a total of k = 41 independent samples. Appendix C lists
each of the included samples for group-level temporal change analyses, and indicates study
characteristics (sample size, homogeneity of sample nationality, sample sex, ICE setting, and
mission duration) and the positive and negative psychosocial functioning indicators examined.
36
Standardizing Response Scales and Time Points. Within the included studies, specific
indicators of positive and negative psychosocial functioning were often measured using response
scales with differing numbers of response options (e.g., Measure A = 7-point scale; Measure B =
4-point scale), an issue that became even more pronounced when comparing response scales
across studies. Thus, it was first necessary to convert the various response scales used into an
equivalent scale in order to examine either individual- or group-level temporal effects across
indicators and studies. To convert raw scores into scores on an equivalent scale, we used the
percentage of maximum possible score (POMP; Cohen, Cohen, Aiken, & West, 1999), which
can be used to convert response scales with any number of points into a score on a 0.00-1.00
scale. The formula for standardizing raw scores using POMP is:
POMP = [(observed – min.)/(max. – min.)]
where observed represents the observed raw score on the original response scale, min. represents
the minimum possible score on the original scale (e.g., 1-7 scale = “1”), and max. represents the
maximum possible score on the original scale (e.g., 1-7 scale = “7”). The advantage of POMP
scoring is that, while it preserves the relationships between variables, it also increases the
interpretability of scores by transforming scores into a percentage. For example, using the
formula above with a response scale that ranged from “1” to “7”, a score of “1” would be
transformed into 0.00 (0%), while a “7” would be transformed into 1.00 (100%).
In addition to the existing variability in the response scales used among positive and
negative psychosocial functioning indicators both within and across included studies, there was
also variability in the time points at which indicators were measured. This was typically an issue
observed across studies, but one that created variability in the number of time points at which
indicators were measured, as well as the time, relative to the full mission duration, at which
37
indicators where measured. Three time points was the minimum at which each indicator was
required to be measured in order for temporal trends to be appropriately examined. However,
many studies included many more measurement time points. Even when studies employed the
same number of measurements of relevant variables, the specific time points during the mission
at which those variables were measured often differed across studies. Related to this issue was
that the duration of ICE missions also differed greatly. Given these issues, we converted the
observed measurement time points to reflect the percentage of mission complete, a slight
variation of POMP, as it is described above. For example, if an individual reported well-being at
Month 1 of a four month mission, then that time point was computed as .25 (or 25%) complete.
Results
Well-being-Performance Relationship Sample Descriptions
Appendix A depicts the characteristics of studies included in this review. With respect to
study setting, the vast majority of research examining the well-being-performance relationship
has been conducted among crews in analogue settings. The analogue setting most studied among
included studies was Antarctic winter stations (k = 14), followed by simulation settings (k = 8),
submarines (k = 3), Arctic expeditions (k = 1), and spaceflight (k = 1). Four studies (k = 4) were
conducted retrospectively on ICE experiences, two of which were among those who had
experienced spaceflight (the remaining two were submariners and Antarctic winterers,
respectively). The mean sample size was quite large, M = 75.77 (SD = 122.76), which reflects
the strong influence the large number of Antarctic station studies exhibited on mean sample size.
Average mission length across samples was approximately seven months, M = 207.28 days (SD
= 170.05 days). Seventeen samples consisted exclusively of male crewmembers (k = 17).
Eleven included both males and females (k = 11), although the proportion of females within
38
these samples was typically small. Crew sex information was not reported for k = 3. Finally,
seven studies included multi-national crews (k = 7), the nationality of 20 crews was
homogeneous (k = 20), and nationality information was not reported for k = 4. Among
homogeneous crews, 15 were American (k = 15), two were British (k = 2), one was French (k =
1), one was Norwegian (k = 1), and one was Japanese (k =1).
Examining the frequency with which the three well-being factors are represented across
these samples, eudaimonic indicators were most often linked to relevant outcomes (k = 17),
followed by PsyCap indicators (k = 12) and hedonic indicators (k = 10). In addition, indicators
of general well-being appeared in examinations of five of the included independent samples (k =
5). Among outcomes, performance/work attitudinal indicators appeared more often (k = 24) than
did indicators of stress/psychosocial symptomatology (k = 12).
To present the systematic review of the well-being-performance relationship we have
separated existing evidence into three distinct sections, reflecting each of the three overarching
well-being factors: hedonic, eudaimonic, and PsyCap. Within each section we further separate
evidence according to the broad outcome factor: performance/work attitudes and
stress/psychosocial symptomatology. As described above, a final section reviews the effects of
emotional stability, which was not clearly classifiable under any of the three well-being factors
and is treated as an indicator of general well-being.
Hedonic Well-being
Hedonic Well-being and Performance and Work Attitude Criteria. The most direct
measures of performance available within the included data (i.e., peer and supervisor
performance ratings) include evidence from Antarctic winterers on measures of joy in work and
positive and negative affect. For example, using retrospective survey methodology, Rose, Fogg,
39
Helmreich, and McFadden (1994) found astronaut-reported joy in work to be positively, but
weakly and non-significantly, related to peer ratings of interpersonal (r = .19), technical (r =
.12), and leadership abilities (r = .05), as well as supervisor performance ratings (r = .16). In a
second study, Grant et al. (2007) used scores on a series of measures completed by individuals
prior to wintering-over in Antarctica to predict commanders’ post-mission ratings of whether
individual crewmembers demonstrated exceptional adaptation during the stay. Among these pre-
mission predictors were measures of positive and negative affect, both of which were non-
significant. Importantly, the authors observed reductions in both positive and negative affect
between pre-mission and midwinter, suggesting that winterers experienced an overall flattening
of affect, but also that those reporting smaller reductions in positive affect showed significantly
greater adaptation. The same moderating effect, however, was not observed for negative affect
scores.
Additional evidence exists in relation to hedonic indicators and less-direct measures of
performance. These include: crew cohesion, crew-mission control communication frequency,
perceptions of team decision-making quality, and attitudes towards involvement in future ICE
experiences. In addition, this evidence implicates a somewhat broader set of hedonic indicators
than those described above, including: job satisfaction, perceptions of ingroup positive and
negative affect and outgroup positive and negative affect, vigor and perceived fairness. Finally,
although hedonic well-being-direct performance evidence is constrained to the Antarctic station
setting, evidence examining the relationship between hedonic well-being and a broader set of
outcomes indicative of performance comes from Antarctic stations, Arctic expeditions,
simulations, and spaceflight.
40
Three studies have examined the relationship between hedonic indicators with measures
reflecting crew cohesion. Sarris and Kirby (2005) found job satisfaction to be significantly
related to reported group cohesion (r = .36) among Antarctic winterers. Among a small group (N
= 11) of Soviet and American Arctic expeditioners seeking to cross the Bering Strait land bridge,
perceived fairness was shown to relate strongly to perceived disruption to group cohesiveness (r
= -.83) and interpersonal conflict (r = -.83; Leon, Kanfer, Hoffman, & Dupre, 1994). Finally, in
the first of five studies (Study 1) conducted by Krins (2009), the author examined concurrently-
measured relationships between ingroup (and outgroup) positive affect and group cohesion and
polarization at four different time points during a month-long simulation mission. Reported
ingroup positive affect related moderately to reported levels of group cohesion during the first
week of the mission (r = .27) and strongly during the remaining weeks of the mission (rweek 2 =
.76; rweek 3 = .91; rweek 4 = .99). With respect to the remaining relationships examined, an
inconsistent trend was found in individuals’ reports of outgroup positive affect and cohesion
across time points (r = .00 to .99), and non-significant relationships were found for ingroup and
outgroup positive affect with outgroup polarization.
Moving beyond cohesion criteria, three additional studies examined three unique hedonic
indicator-performance relationships. First, among ISS astronauts, astronaut-reported vigor
related negatively to the amount of communication between ISS crewmembers and mission
control (Kanas, Gushin, & Yusupova, 2008). Initially, this finding may seem somewhat
unexpected. However, Kanas et al. also found astronaut-reported anxiety, depression, hostility,
fatigue, and confusion related positively to communication quantity. This suggests that greater
communication reflects more instances of crewmembers displacing psychosocial problems onto
controllers, a phenomenon observed by Kanas and colleagues (e.g., Kanas, 2002: Kanas et al.,
41
2001d) in earlier work on space crews. Second, Leon et al. (1994) found perceived fairness to be
strongly related to perceptions of team decision making quality (r = .75) among a multinational
Arctic expedition crew. Finally, among a cohort of Antarctic winterers, Sarris and Kirby (2005)
found job satisfaction to be positively, but weakly, related to individuals’ intentions to return (r =
.13), willingness to recommend the experience to others (r = .07), and whether individuals
actually returned for another Antarctic stay (r = .10).
Hedonic Well-being and Stress and Symptomatology Criteria. Much of the evidence
relating hedonic well-being to stress and psychosocial symptomatology comes from the study of
positive and negative affect, with outcomes ranging from stress to quality of life and depressive
symptoms. For example, the positive affect-stress relationship has been examined in a pair of
simulation studies. First, during a month-long simulation (Krins, 2009, Study 1) ingroup and
outgroup positive affect significantly related to perceived stress during the second week of the
simulation mission (r = -.55 and -.91, respectively). The relationship between ingroup and
outgroup positive affect and perceived stress was non-significant during the remaining three
weeks. However, at least in the case of the ingroup positive affect-stress relationship, this was
more an issue of low statistical power, as effects found during the first and third weeks were also
potentially practically significant (r = -.33 and -.44, respectively). Second, among a seven-
member crew taking part in a four-month FMARS simulation mission, Bishop et al. (2010)
found positive affect to be strongly related to stress when measured at baseline (r = -.59). A far
weaker relationship was observed during the first month of the mission (r = -.14), but it increased
in magnitude over the remaining months, returning to near baseline levels (rmonth 2 = -.42; rmonth 3
= -.33; rmonth 4 = -.51). A somewhat different trend was found across the mission duration in the
relationship between negative affect and stress. That is, while a strong relationship was found at
42
baseline (r = .49) and during the first month of the mission (r = .59), only a weak-to-moderate
relationship was found during subsequent months (rmonth 2 = .17; rmonth 3 = .33; rmonth 4 = .15).
Positive and negative affect have also been linked to various symptomatological criteria.
Controlling for the effects of a range of other predictors in a multiple regression model, negative
affect scores remained significantly (positively) related to poor quality of life (β = .20), among a
sample of submariners (Brasher, Dew, Kilminster, & Bridger, 2010). Interestingly, the
researchers observed an almost identical effect for positive affect scores in the same analysis (β =
.18). Given the large number of predictors included in the regression model, however, this
unexpected positive relationship may have resulted from variable suppression and should be
interpreted with caution. Contrary to the findings of Brasher et al., positive affect related
strongly to the use of immature defense mechanisms (r = -.60), across the duration of the 105-
day Mars500 pilot simulation mission (Nicolas et al., 2013). However, no relationship was
found between negative affect and the use of such mechanisms (r = -.03), and both positive
affect and negative affect showed trivial relationships with depressive symptoms (r = .06 and -
.04, respectively).
We identified only a single study linking an indicator of hedonic well-being, beyond that
of positive or negative affect, to stress/symptomatology criteria. In a study of Antarctic
winterers, Palinkas and Browner (1995) found satisfaction with social support to be significantly
related to depressive symptoms at both baseline measurement (r = -.32) and as the end of the
mission (r = -.44). In addition, baseline satisfaction with social support was the only significant
predictor of depressive symptoms at the end of the mission (r = -.34) in their model, beyond that
of baseline depressive symptoms.
43
Hedonic Well-being Summary. Indicators of hedonic well-being appear to be only
weakly but positively related to the most direct measures of performance available in these data
(i.e., peer and supervisor ratings of performance), as well as to individuals’ own attitudes
towards their ICE experiences. Conversely, a stronger relationship has been consistently
observed between hedonic well-being and measures of group functioning (e.g., crew cohesion).
In fact, this evidence indicates that, relative to other outcomes reviewed, the role of hedonic
well-being may be most beneficial to improved group functioning. Most existing ICE research
examining the effects of hedonic well-being on stress and symptomatology has employed
measures of positive and negative affect. Evidence suggests affect to be moderately related to
stress, in general. However, both studies reviewed indicate that positive and negative affect
related differently to stress over time, a finding that may be further elaborated on through our
analyses as part of the second purpose of this review. Little evidence was found for the effects
of negative affect on symptomatology, and multiple studies indicated the positive affect-
symptomatology relationship to be complex. As a final note, it should be mentioned that fewer
studies included stress/symptomatology outcomes, relative to those employing
performance/attitudinal outcomes, and only limited evidence was available beyond that relating
outcomes to positive and negative affect.
Eudaimonic Well-being
Eudaimonic Well-being and Performance and Work Attitude Criteria. There exists
some evidence linking eudaimonic well-being indicators to direct measures of performance (i.e.,
subjective performance ratings made by peers and supervisors, objective measures of
psychomotor performance), though the evidence does not suggest this relationship to be strong,
or even consistently positive. For example, among surveyed astronauts, Rose et al. (1994)
44
sought to use astronaut-reported work involvement, job involvement, and individual drive in
predicting peer and supervisor-rated performance. Individual drive and work involvement scores
were shown to be weakly and largely negatively related to peer (i.e., interpersonal, technical, and
leadership ability) and supervisor-rated performance (r = -.17 to -.04 and r = -.11 to .01,
respectively). Individuals’ reported job involvement showed a similar relation with peer ratings
of performance (r = -.04 to .02), with a somewhat stronger positive relationship with supervisor
performance ratings (r = .16).
Two additional studies further examined the effects of eudaimonic indicators on
subjective performance ratings. First, Palinkas et al. (2000) found locus of control to
significantly predict peer ratings of leadership ability (β = .10), but not additional peer-rated
criteria (task ability, emotional stability, social compatibility, and overall performance). Second,
among separate samples of enlisted Navy and civilian personnel wintering in Antarctica, Doll
and Gunderson (1969) examined the relationships between individuals’ reported job morale and
job importance and peer and supervisor performance ratings. Findings showed considerable
differences between Naval and civilian personnel. That is, trivial effects were found regarding to
the relationships between well-being factors and peer and supervisor performance ratings (r = .01
to .05) among Naval participants, while much stronger effects were found across those same
relationships among civilian scientists (r = .27 to.47). This may suggest that differences in either
the nature of the work or the work culture between these two groups contributed to the
differential effects observed.
An additional study provides evidence of the relationship between eudaimonic well-being
and objective performance measures. Rosnet et al. (2000) sought to examine the relationship
between self-image and psychomotor performance tasks. One may expect those with positive
45
self-images and who also perceive that others view them positively to possess the greatest
eudaimonic well-being. However, the researchers found these individuals actually performed
worse than those with incongruent self-images (e.g., negative self-images with perceptions that
others view them even more negatively) on a series of psychomotor tasks. In fact, positive self-
images congruent with one’s perceptions of others’ view of him or her were negatively related to
cognitive performance scales.
Additional evidence comes from studies employing indirect measures of performance
(i.e., crew cohesion/functioning and attitudes towards past ICE experiences). This evidence
yields somewhat more positive evidence regarding the effects of eudaimonic well-being.
Through a series of studies Krins (2009) provides fairly consistent evidence of the effect of
eudaimonic well-being on group functioning. For example, among a simulation crew (Study 1),
the researcher observed a strong relationship between concurrently measured group identification
and group cohesion throughout all four weeks of a month-long mission (r = .79 to .97).
However, concurrent measures of group identification and group polarization were not found to
be significantly related at any of the four measurement time points. In a second study conducted
by Krins (2009; Study 2), among a crew taking part in a two week simulation at the Mars Desert
Research Station, the researcher found that those who showed greater group identification also
showed lower scores on perceived alienation—that is, perceptions of the self or others being
alienated (r = -.56). However, group identification was unrelated to group functioning (r = -.01).
In a final study by Krins (2009; Study 5), which was conducted among winterers at the
Concordia station, reports of group identification, but not friendship among crewmembers, were
found to be significantly related to reports of cooperation with other crewmembers (r = .68 and r
= .59, respectively).
46
In the study of the Bering Strait land-bridge expedition team, Leon et al. (1994) found
altruistic behavior to be strongly negatively related to disruptions in group cohesion (r = -.72).
In retrospective reports of group cohesion during a past Antarctic stay, Sarris and Kirby (2005)
observed comparatively weaker but still practically significant relationships between two
measures capturing individuals’ perceptions of how well they “fit” with the Antarctic station
culture and environment and their perceptions of group cohesion (r = .47 and .39).
Researchers have also examined the relationship between eudaimonic indicators of well-
being and both self-reported performance and retrospective attitudes towards their ICE
experiences. Kanas et al. (2011) tested the effects of increasing the autonomy on self-reported
performance accuracy and efficiency among crewmembers of the Mars105 simulation, finding
little-to-no change in reports of post-manipulation (Cohen’s ds ranged from .00 to .11). Three
studies (two independent samples) have examined retrospective attitudes towards ICE
experiences. Sarris and Kirby (2005) sought to examine the relationship between fit and
measures of attitudes toward past ICE experiences. Their findings indicate that scores on both fit
measures weakly related to individuals’ reported intentions to return (r = .03 and .07) as well as
reports of whether they actually did return (r = .09 and .09). Slightly stronger relationships were
found between the two fit measures and individuals’ reported willingness to recommend the
experience to others (r = .22 and r = .10). The relationships between cultural styles and work
attitude outcomes were further explored among the male respondents within the same data
(Sarris & Kirby, 2007). Of particular interest in those analyses was the effect of what the authors
described as a “satisfaction culture,” which reflects humanistic, affiliative, achievement, and
self-actualizing styles. The presence of a satisfaction culture was significantly associated with
greater perceptions of role clarity (r = .23), job satisfaction (r = .54), and intentions to return (r =
47
.26), as well as weaker perceptions of role conflict (r = -.31); a non-significant relationship was
found with individuals’ willingness to recommend the experience to others (r = .14).
A final study examining work attitude outcomes comes from van Baarsen (2013) who,
among the Mars500 simulation crew, examined the concurrent relationships between multiple
eudaimonic indicators (two measures of autonomy, one of perceived challenge, and one of
altruism) and motivation to participate in future ICE research. Concurrent measurements were
taken at baseline and at five different time points throughout the 520 day simulation. The mean
correlation between concurrently-measured motivation to participate in future ICE missions and
autonomy (r = .66) and altruism (r = .29) represent moderate-to-strong positive relationships.
However, the mean correlation between motivation to participate in the future and motivation
regarding the challenge of the simulation was found to be negative (r = -.19).
Eudaimonic Well-being and Stress and Psychosocial Symptomatology Criteria.
Available evidence generally fails to support a strong relationship between stress and
eudaimonic well-being. For example, using hierarchical regression, Sandal, Endresen, Vaernes,
and Ursin (2003) demonstrated positive instrumentality/expressivity to be negatively, but non-
significantly, related to reported stress due to social factors, leadership/workload, homesickness,
or isolation during the initial stage of multiple submarine missions. Using the same technique to
examine these relationships again during the final stage of the same missions, the researchers
showed positive instrumentality/expressivity to be significantly, negatively related to stress due
to social factors and isolation, but non-significantly (negatively) related to stress due to
leadership/workload and homesickness. Similarly weak evidence has been demonstrated among
individuals completing a month-long simulation mission, wherein group identification was found
48
to be unrelated to concurrently-measured perceptions of group stress at each of the four time
points the variables were measured (Krins, 2009; Study 1).
Other research suggests somewhat more positive evidence of the eudaimonic well-being-
stress relationship, specifically for evidence examining autonomy as a eudaimonic indicator. For
example, although Kanas et al. (2010) found greater autonomy among the Mars105 simulation
crew to result in reports of substantially greater work pressure (Cohen’s d = 1.06), further
evidence of the autonomy-stress relationship presented by Kanas et al. (2010) based on data from
two NASA Extreme Environment Mission Operations (NEEMO) program crews paints a
somewhat different picture. The first crew, the NEEMO 12 crew, was provided low levels of
autonomy throughout the 12-day mission, while the second crew, the NEEMO 13 crew, was
exposed to five days of low autonomy followed by five days of high autonomy. Researchers
measured outcomes midway through and at the end of both missions in order to capture the
between-group effects of the autonomy manipulation. The researchers found an increase in
fatigue among the NEEMO 12 crew (low autonomy) during the second half of the mission, while
a reduction in fatigue was observed among the NEEMO 13 crew (high autonomy). Also, while
the NEEMO 13 crew showed no increase in reported work pressure during the second half of the
mission, a small increase was observed among the NEEMO 12 crew. Taken together, these
findings suggest that greater autonomy may promote resistance to fatigue and work pressure.
However, it should be noted that the researchers did find that high autonomy crewmembers
reported greater confusion than did low autonomy crewmembers during the second half of the
mission, likely the result of issues associated with the transition to greater autonomous
functioning.
49
Only one study each has linked eudaimonic well-being to anxiety symptoms and poor
quality of life. A study conducted among Japanese winterers found individuals’ reported sense
of coherence to correlate negatively with their reported anxiety sensitivity (r = -.43; Weiss,
Suedfeld, Steel, & Tanaka, 2000). Examining quality of life, Brashner et al. (2010) used
regression analyses with submariner data to examine the effects of autonomy and control and
organizational commitment. Results indicated a non-significant effect for autonomy and control
scores, but a significant effect for organizational commitment scores in predicting poor quality of
life (β = .41).
Greater attention has been given to depressive symptoms. Two such studies were
conducted among the Mars105 simulation crew. In the first, Nicolas et al. (2013) demonstrated
intermediate and mature defense mechanism scores, which reflect the characteristics possessed
by emotionally healthy individuals (e.g., altruism, mindfulness, humility, and courage), to only
weakly relate to depression scores (r = -.20 and -.06, respectively). In the second study,
increased autonomous functioning was shown to lead to a considerable drop in individuals’
depression (Cohen’s d = .77) and anger (Cohen’s d = .63). Among a group of almost 100
winterers, Palinkas and Browner (1995) examined the relationship between locus of control and
depressive symptoms, both of which were measured at baseline and the end of the year-long
Antarctic stay. Two points are worth noting regarding their findings. First, the relationship
between locus of control and depressive symptoms was weak, both when examined concurrently
(rbaseline = .13 and rmission end = .14, respectively) and when baseline locus of control scores were
used to predict depressive symptoms at the end of the mission (r = .06). Second, the positive
directionality across effects suggests that those reporting greater locus of control also reported
greater levels of depressive symptoms, a finding counter to what may be expected.
50
Finally, two studies have examined the relationship between eudaimonic well-being and
what might best be described as maladaptive coping strategies. First, Sandal et al. (2003) found
positive instrumentality to be significantly negatively related to problem avoidance (r = -.29)
among submariners. Second, Nicolas et al. (2013) found intermediate and mature defense
mechanism scores differentially related to immature defense mechanism scores among the
Mars105 simulation crew, with mature scores being negatively related (r = -.43) and
intermediate scores being positively related to immature scores (r = .66).
Eudaimonic Well-being Summary. Eudaimonic well-being showed a fairly weak
relationship with direct measures of performance, and a consistently stronger relationship was
shown when examining less direct measures of performance, in particular group cohesion and
functioning. This is a similar trend in the effects observed for hedonic well-being. However,
among the eudaimonic evidence, the weak relationship with direct measures of performance
(e.g., both subjective and objective) was consistently shown to be negative, suggesting greater
eudaimonic well-being to be associated with worse performance ratings. As mentioned,
evidence linking eudaimonic indicators to measures of group cohesion and functioning has been
more encouraging regarding the potential positive contributions of these indicators, and evidence
regarding the eudaimonic well-being-work attitude relationship suggests these two are
moderately related. Evidence has consistently shown eudaimonic well-being to not diminish
perceptions of stress. In addition, its effects on depressive symptoms have been mixed.
Moderate negative relationships were observed between eudaimonic well-being and additional
outcomes: anxiety, poor quality of life, and maladaptive coping strategies. However, there may
be too few studies employing these outcomes to draw any strong conclusions.
PsyCap
51
PsyCap and Performance and Work Attitude Criteria. Two studies have examined
PsyCap indicators along with direct measures of performance. First, Doll, Gunderson, and
Ryman (1969) examined the predictive validity of a series of selection tests in relation to peer
nominated/supervisor rated performance criteria collected at the conclusion of an Antarctic
winter-over. Among these were clinical evaluations of individuals’ likelihood of successfully
adjusting to the conditions associated with wintering-over. The researchers examined the
predictive validity of clinical evaluations separately for Navy construction personnel, Navy
technical and administration staff, and civilian scientists. Clinical evaluations were ineffective at
predicting peer/supervisor rated emotional stability among Navy construction or technical-
administrative personnel (6.2% and 0.0%, respectively), but were the strongest predictor of the
selection tests among civilian scientists, successfully predicting emotional stability among 31.3%
of the sample. Clinical evaluations were ineffective at predicting peer/supervisor task ratings
across all three samples (0.0% to 6.2%). Clinical evaluations were also ineffective in predicting
peer/supervisor rated social compatibility among Navy technical and administrative personnel,
but moderately effective among Navy construction personnel (12.5%) compared to the other
predictive tests, and the most effective predictive test among civilian scientists (37.5%). Clinical
evaluations were most effective test in predicting peer/supervisor leadership ratings among Navy
construction (31.3%) and technical-administrative personnel (68.7%), but not among civilian
scientists (0.0%). In examining the effects of individuals’ self-reported optimism on peer and
supervisor ratings on a series of performance dimensions, Palinkas et al. (2000) found that
optimism significantly predicted peer ratings of emotional stability (β = .18) and social
compatibility (β = .14). Optimism did not significantly predict the remaining three criteria (task
ability, leadership, overall performance).
52
Two additional studies have examined less direct indicators of performance. First, Grant
et al. (2007) found highly optimistic winterers to be 2.6 times more likely than less optimistic
winterers to show exceptional levels of adaptation to Antarctic station winter-over conditions.
Second, personal self-esteem was shown to strongly and consistently relate to concurrently-
reported group cohesion throughout the four week simulation mission (r = .76 to .84; Krins,
2009, Study 1). In the same study, the relationship between perceptions of collective self-esteem
and group cohesion was less consistent, with correlations ranging from r = .46 to .92 across
measurement time points. The relationship between both personal and collective self-esteem and
group polarization was inconsistent in both magnitude and directionality across time points.
PsyCap and Stress and Symptomatology Criteria. Five studies have empirically
examined the relationship between PsyCap indicators and a range of stress and symptomatology
measures. To begin, two studies have examined the relationship between self-esteem and such
outcomes. First, evidence from a four-week simulation study showed the relationship between
personal and collective self-esteem and perceived stress to be inconsistent across measurement
time points both in terms of relationship magnitude and directionality (Krins, 2009, Study 1).
Second, a study among Antarctic winterers demonstrated self-confidence to be moderately
related to concurrently-measured depressive symptoms at baseline (r = -.34) and the end of the
mission (r = -.31), while a slightly weaker relationship was shown between baseline self-
confidence scores and depressive symptoms at the end of the year-long mission (r = -.23;
Palinkas & Browner, 1995).
The remaining studies linking PsyCap indicators to measures of stress and
symptomatology have focused on resilience or the closely-related characteristic of hardiness.
53
For example, Eid, Johnsen, Saus, and Risberg (2004) linked hardiness with PTSD symptoms
and scores on a measure reflecting poor life quality during a week-long disabled submarine
mission. Overall hardiness scores were significantly related to the negative impact of the event
(r = -.52) and poor quality of life (r = -.47) and non-significantly related to PTSD symptoms (r =
-.29). In a separate study, hardiness has also been linked to anxiety sensitivity among Japanese
winterers (Weiss et al., 2000), wherein the two were shown to be moderately related (r = -.43).
In a final study, one which employed three separate bed rest simulation conditions, Zuckerman,
Persky, Link, and Basu (1968) assessed the relationship between self-reported resilience to
isolation conditions and anxiety, depression, hostility, and two measures of stress. The strongest
effects of resilience to isolation scores on psychosocial symptomatology scores were found
among those completing the least extreme bed rest simulation condition, with correlations
ranging from r = -.44 (anxiety) to r = -.25 (stress due to the tediousness of the simulation
conditions). The next strongest effects were found among those completing the median extreme
simulation condition, with correlations ranging from r = -.28 (anxiety) to r = -.05 (hostility).
Interestingly, self-reported resilience to isolation was most weakly related to the majority of the
psychosocial symptomatology measures, among those completing the most extreme simulation
condition, with effects ranging from r = -.02 (hostility) to r = .15 (both stress scales).
Summary. Of the three factors of well-being, the least amount of evidence exists
regarding the effects of indicators of PsyCap on measures of performance, work attitudes, stress,
and symptomatology. However, among studies that have considered these indicators, the
evidence appears to suggest consistently moderate effects of PsyCap indicators across outcomes.
Particularly noteworthy is that among available evidence relating each of the three well-being
factors to direct measures of performance that which exists for indicators of PsyCap may be the
54
most convincing. Moreover, whereas the evidence regarding both hedonic and eudaimonic well-
being has been mixed in relation to measures of stress and/or symptomatology, the literature
reviewed examining the effects of indicators of PsyCap has been quite consistently moderate.
Taken together, we remind the reader that only a limited amount of evidence exists with respect
to the effects of PsyCap indicators in ICE settings, but that which does exist suggests that the
further exploration of these effects may prove fruitful to the psychosocial health and
performance of individuals in ICE settings.
General Well-being
As described above, we identified a set of studies that examined well-being indicators
that could not be clearly identified as hedonic, eudaimonic, or PsyCap. Three such studies
provided evidence examining the well-being-performance relationship, each including a measure
of emotional stability as a well-being indicator. Two of these studies linked survey data to
performance-related outcomes. The first study found emotional stability among astronaut
applicants to be significantly related to a number of self-reported trait characteristics that
potentially contribute to effective performance (instrumentality, r = .39; mastery, r = .24; work
orientation, r = .22; achievement strivings, r = .20). Emotional stability was also shown to be
negatively related to verbal aggressiveness (r = -.27), an indicator of psychosocial
symptomatology (Musson, Sandal, & Helreich, 2004). The second study linked astronaut
emotional stability scores to actual peer and supervisor ratings, finding emotional stability to be
only weakly related to actual performance ratings (r = .09 to .21; Rose et al., 1994). More
encouraging evidence has been shown in the final study we identified as indicative of general
well-being, which was conducted among wintering U.S Naval construction (i.e., Seabees), U.S.
Naval technical-administrative, and civilian samples (Doll & Gunderson, 1970). The researchers
55
found peer-nominated emotional stability to be consistently and fairly strongly related to peer-
nominated task performance (Seabee r = .46; Technical-Administrative r = .53; Civilian r = .47),
social compatibility (Seabee r = .56; Technical-Administrative r = .50; Civilian r = .61),
leadership skills (Seabee r = .38; Technical-Administrative r = .57; Civilian r = .38), and
whether or not individuals would excel in a future Antarctic winter-over (Seabee r = .74;
Technical-Administrative r = .74; Civilian r = .67). Unfortunately, given the small number of
studies and the range of effects that have been found it may not be appropriate to make any
strong conclusions based on these data.
Temporal Change in Psychosocial Factors Sample Descriptions
Individual-level Temporal Change. Appendix B depicts the characteristics of studies
included in these analyses. In total, there were 15 studies reporting individual-level temporal
change data (k = 15). The majority of these studies were conducted among crews either in
simulation settings (k = 7) or during Arctic expeditions (k = 6). Data were also reported from an
Antarctic winter-over (k = 1) and a spaceflight (k = 1). As described above, the total sample size
was N = 67 across the 15 studies (M = 4.47, SD = 2.90). The average mission length across
samples was 142.73 days (SD = 155.83 days). Just over half of the studies examined crews that
were exclusively male (k = 8). Five studies examined crews of heterogeneous sex (k = 5), and
two studies examined exclusively female crews (k = 2). Just under half of the studied crews
were multi-national (k = 7), six were homogeneous (k = 6), and crew nationality was not reported
for k =2. Among homogeneous crews, three were American (k = 3), while one crew each was
reported to be British (k = 1), Russian (k = 1), and Australian (k = 1).
With regard to the frequency with which temporal trends were reported for positive and
negative indicators of psychosocial functioning, 13 of the 15 studies reported evidence regarding
56
positive indicators (30 positive indicators total) and 12 of the 15 studies reported evidence
regarding negative psychosocial functioning (20 negative indicators total). The results of these
analyses are presented in Figures 1 and 2, respectively. Both figures depict psychosocial
functioning (both positive and negative) to be slightly above the middle possible score on
response scales from which data were gathered (0.5 to 0.6; or between 50% and 60% of
maximum possible score).
Overall, Figure 1 (positive psychosocial functioning, k =13, N = 48) suggests that
individuals reported near average positive psychosocial functioning throughout the duration of
the mission. However, some trends are worth noting. Least subtle is the trend over the first 15-
20% of the mission during which positive psychosocial functioning increases by approximately
20%. From there and for the much of the remainder of the mission mean positive psychosocial
functioning appears to regress back towards the beginning level, reaching a low point of just over
60% of the maximum possible score, before the mean score then increases again shortly after
90% of the mission is complete, reaching nearly 70% of the maximum possible score by the end
of the mission.
57
Figure 1. Individual-level Temporal Change: Indicators of Positive Psychosocial Functioning
(Top: scatterplot of data points; Bottom: confidence intervals aggregated at each 5% mission
complete interval)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
58
Figure 2. Individual-level Temporal Change: Indicators of Negative Psychosocial Functioning
(Top: scatterplot of data points; Bottom: confidence intervals aggregated at each 5% mission
complete interval)
Figure 2 (negative psychosocial functioning, k = 12, N = 52) depicts an almost mirror
image of Figure 1, wherein scores on negative functioning drop by approximately 25% of
maximum possible by the time just over 10% of the mission is complete. Negative psychosocial
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
59
functioning then shows a slight increase leveling out at approximately 40% of the maximum
score possible. This suggests that individuals reported mid-to-low levels of psychosocial
functioning throughout the remainder of the mission.
Group-level Temporal Change. The characteristics of studies included in these
analyses are presented in Appendix C. As described above, 32 studies containing 41
independent samples were included in these analyses (k = 41, N = 686). The mean sample size
across studies was n = 16.59 (SD = 17.45). The most common study setting was Antarctic
stations (k = 17), followed by simulations/hyperbaric chambers (k = 14), and Arctic expeditions
(k = 6). An additional two samples each were drawn from submarine/diving (k = 2) and
spaceflight settings (k = 2). Average mission length across samples was 174.06 days (SD =
134.60 days). The majority of crews were exclusively male (k = 23). Twelve crews included
both males and females (k = 12), one crew was exclusively female (k = 1), and crew sex was not
reported for the remaining k = 5. Most crews were nationally homogeneous (k =24), k = 11
crews were multi-national, and crew nationality was not reported for k = 7. Among crews of
homogeneous nationality, nine were American (k = 9), five were French (k = 5), three were
Russian (k = 3), two were Australian (k = 2), two were British (k = 2), one was Indian (k = 1),
and one was Norwegian (k = 1).
Indicators of positive psychosocial functioning were reported among 29 of the 41
samples (52 total positive indicators; N = 421). Indicators of negative psychosocial functioning
were reported among 33 of the 41 samples (57 total negative indicators; N = 616). Results of
group-level temporal trend analyses are presented in Figures 3 and 4, respectively. Contrary to
the almost mirror-image trends in positive and negative psychosocial functioning observed
through the individual-level analysis, the group-level trends presented in Figures 3 and 4 suggest
60
positive and negative indicators function somewhat differently. Mean trend levels in Figure 3
suggest that individuals reported above-average positive psychosocial functioning across mission
duration. Moreover, the trend remains fairly flat throughout the first 90% of mission duration.
At mission start, scores are slightly above average (approximately 60% of the maximum possible
score). Positive functioning scores then show a small (approximately 5%) increase over the
course of the first 20% of the mission. Mean scores then regress back to initial levels by the
half-way point of the mission. At 70% of mission completion positive functioning scores show a
downward trend, dropping nearly 10% by the time 90% of mission duration is complete.
However, the most acute change in positive psychosocial functioning appears during the final
10% of mission duration during which scores increase by approximately 10%, to the
approximate level of positive functioning reported at the beginning of the mission.
Moving next to examine mean trends in negative psychosocial functioning, Figure 4
shows negative psychosocial functioning to be well below average at mission start
(approximately 25% of maximum possible), suggesting a general lack of psychosocial problems
among crewmembers. However, over the course of the first mission quarter, the mean trend in
negative psychosocial functioning scores saw an increase of over 30%. During the second
mission quarter, Figure 4 depicts a reduction in negative functioning scores to approximately
50% of that maximally possible (i.e., the middle point on the response scales for which the data
were originally reported). From there, mean trends suggest that negative functioning scores
remained largely stable throughout the remainder of mission duration.
61
Figure 3: Group-level Temporal Change: Indicators of Positive Psychosocial Functioning (Top:
scatterplot of data points; Bottom: confidence intervals aggregated at each 5% mission complete
interval)
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
62
Figure 4: Group-level Temporal Change: Indicators of Negative Psychosocial Functioning (Top:
scatterplot of data points; Bottom: error bars aggregated at each 5% mission complete interval)
In addition to examining main effect positive and negative psychosocial functioning
trends across studies we sought to examine whether study characteristics moderated these trends.
The potential moderating variables we examined were: mission duration (short = 1-90 days;
intermediate = 91-180 days; long = 181+ days), crew nationality (multinational, homogeneous-
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
63
American, homogeneous-other nationality), and crew gender (heterogeneous crew, male crew).
Given the small number of studies reporting individual-level temporal trends and concern over
spurious findings due to further partitioning these data, we restricted our analyses to group-level
data. First, we assessed whether temporal effects differed due to mission duration (Figures 5a
and 5b). The temporal trends among crews taking part in long- and short-duration missions were
very similar, although the temporal trends observed among crews taking part in intermediate
missions differed. With respect to positive psychosocial functioning, intermediate trends appear
be inverted in comparison to the trends observed across long- and short-duration missions.
Moreover, while the greatest change in positive functioning during long- and short-duration
missions can be observed during the earliest stage of the mission (initial 20% of mission
duration), the greatest change during intermediate-duration missions can be observed during the
final 10% of mission duration. With respect to negative psychosocial functioning, the trend
observed for intermediate-duration missions appears to actually be similar to long- and short-
duration trends, save an initial 10% decrease in negative psychosocial functioning scores.
Importantly, few studies evaluated temporal trends in intermediate-duration missions, which
raises concern over whether these trends truly differ from the trends observed across long- and
short-duration missions. Given the similar findings among shorter- and longer-duration
missions, for which considerably more evidence exists, observed intermediate-duration mission
temporal trends may represent spurious effects.
Next we examined crew nationality as a potential moderator. Specifically we conducted
separate analyses of temporal effects on positive and negative psychosocial functioning among
multi-national crews, homogeneous American crews, and homogeneous crews of other
nationalities (Figures 6a and 6b). Considerable variation across crew make-up was observed
64
across these categories. Examining positive psychosocial functioning, the observed trend among
multi-national crews was highly similar to the overall temporal trend found across all group-level
data reporting positive psychosocial functioning, wherein the greatest fluctuation in scores was a
10-15% increase during the final 10% of mission duration. Among homogeneous American
crews Figure 6a shows a similar trend in the fluctuation in scores as was observed among multi-
national crews. However, American crew data also demonstrates a clear downward trend in
positive psychosocial functioning scores over the course of the mission. Importantly, these
trends are based on only three studies (N = 13), which is likely too few upon which to draw any
reliable conclusions. Finally, evidence indicates positive psychosocial functioning scores were
highest among non-American homogeneous crews (approximately 60-70% of maximum possible
responses) and the trend remained remarkably stable across the entirety of mission duration.
65
Figure 5a. Group-level Temporal Change: Indicators of Positive Psychosocial Functioning by
Mission Duration
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
Long-duration (181+ days;
k = 15, N = 241)
Intermediate-duration (91-
180 days; k = 6, N = 35)
Short-duration (1-90 days;
k = 8, N = 73)
66
Figure 5b. Group-level Temporal Change: Indicators of Negative Psychosocial Functioning by
Mission Duration
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
Short-duration (1-90 days;
k = 14, N = 235)
Intermediate-duration (91-
180 days; k = 5, N = 31)
Long-duration (181+ days;
k = 12, N = 225)
67
Figure 6a. Group-level Temporal Change: Indicators of Positive Psychosocial Functioning by
Crew Nationality
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
Multi-national crew (k =
10, N = 98)
Homogeneous American
crew (k = 3, N = 13)
Other nationality
homogeneous crew (k =
10, N = 122)
68
Figure 6b. Group-level Temporal Change: Indicators of Negative Psychosocial Functioning by
Crew Nationality
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
Multi-national crew (k = 8,
N = 81)
Homogeneous American
crew (k = 8, N = 195)
Other nationality
homogeneous crew (k =
10, N = 148)
69
Multi-national crews showed a 20% drop in scores between 10% and 30% of mission
completion on negative psychosocial functioning. Additionally, multi-national crews showed a
sharp, 20% increase in scores between 70-95% of mission duration, followed by an even sharper
20% decrease during the final 5% of the mission. Despite the downward trend in positive
psychosocial functioning scores observed in the underpowered analyses conducted among
homogeneous American crews, the trend in negative psychosocial functioning scores among
these crews remained largely stable throughout mission duration at an approximately average
level (50% of maximum score possible). The lowest negative psychosocial functioning scores at
mission outset were observed among non-American homogeneous crews (approximately 25% of
maximum score possible), suggesting few psychosocial issues. However, these scores increased
dramatically (an approximate 40% increase to 65% of the maximum possible score) prior to 30%
mission completion, which then decreased to near average (i.e., 40% maximum possible score)
just over halfway through the mission and remained fairly stable thereafter.
The final potential moderator examined was crew gender (multi-gender versus male-
only). The temporal trends for positive psychosocial functioning scores associated with multi-
gender (k = 9, N =221) and male-only crews (k = 14, N = 146) are highly similar (see Figure 7:
top). Conversely, the negative psychosocial functioning scores appear to be higher overall and
fluctuate more among multi-gender crews (k = 11, N = 237), as compared to male-only crews (k
= 17, N = 312), suggesting that the multi-gender crew context may be a source of increased stress
among crewmembers (see Figure 7: bottom). Importantly, the positive psychosocial trend
among multi-gender crews appears to be bimodal. That is, individuals appeared to score
particularly high or particularly low in multi-gender crews. This finding is discussed in greater
depth in the discussion section.
70
Figure 7. Group-level Temporal Change as a Function of Crew Gender
Positive Psychosocial Functioning/Multi-gender crews Positive Psychosocial Functioning/Male crews
Negative Psychosocial Functioning/Multi-gender crews Negative Psychosocial Functioning/Male crews
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
% of Maximum Possible Score
% of Mission Complete
71
Summary of Individual- and Group-level Temporal Change. With few exceptions,
mean trends for both positive and negative psychosocial functioning remained largely stable
across individual-level and group-level analyses. The first exception is during the initial stage of
missions. For example, individual-level analyses suggest both an increase in positive
functioning and a decrease in negative functioning during this period. To a lesser extent, group-
level analyses also show an increase in positive functioning during this period. Taken together,
these trends may reflect individuals’ and crews’ initial excitement regarding their missions.
Conversely, the group-level analysis of negative functioning shows a sharp increase in negative
psychosocial functioning, which suggests crews experienced an initially stressful adjustment
period. Moreover, this initial increase receded only slightly indicating crews reported average
levels of negative psychosocial functioning throughout the remainder of the mission, rather than
returning to their initial level. The second possible exception is during the final phase of
mission. This stage was marked by an increase in positive psychosocial functioning in the
group-level analysis, and to a lesser extent in the individual-level analysis. This likely reflects
crews’ and crewmembers’ excitement over the prospect of returning to normal, non-ICE
environments.
Moderator analyses indicate that both positive and negative psychosocial functioning
follows a similar trend across short- and long-duration missions. Somewhat different trends
were observed among intermediate-duration missions, but those findings were based on a limited
number of studies and a relatively small sample size. Moreover, there exists no theoretical
rationale for this difference, suggesting that intermediate-duration trends were spurious. Greater
heterogeneity was observed among trends based on crew nationality. Taking this evidence
together, it appears that homogeneous crews’ functioning may be more stable across mission
72
duration, but evidence also suggests that multi-national crews are not particularly dysfunctional.
Finally, somewhat mixed evidence was observed through crew gender trend analyses. First,
multi-gender and male-only crews did not appear to differ in temporal effects on positive
psychosocial functioning, but it does appear slightly higher and more volatile levels of negative
psychosocial functioning was associated with multi-gender crews.
Discussion
The current report meets three specific aims. First, it provides a framework for studying
well-being in ICE settings (Aim 1). Second, it summarizes existing primary evidence regarding
the relationship between indicators of well-being and measures reflecting or related to
performance in ICE settings (Aim 2). Third, it summarizes evidence regarding the manner in
which indicators of positive and negative psychosocial functioning have been observed to change
over the course of ICE missions (Aim 3).
Well-being-Performance Relationship (Aim 2)
Although there has been an increased focus on well-being in the ICE literature in recent
decades (e.g., Mocellin & Suedfeld, 1991; Palinkas & Suedfeld, 2008; Suedfeld & Steel, 2000),
the amount of research examining any specific relationship (e.g., between positive affect and task
performance) remains limited. Thus, we presented a theoretically-consistent way of classifying
well-being indicators and performance-related measures into a parsimonious framework and
provided an initial qualitative review of existing evidence linking hedonic well-being,
eudaimonic well-being, and PsyCap with measures of performance/work attitudes and stress and
psychosocial symptomatology.
The evidence included in this review suggests that eudaimonic well-being indicators have
been those most commonly studied in relation to performance-related measures in the empirical
73
ICE literature. However, eudaimonic indicators have typically not had the strongest effects on
performance-related outcomes. For example, PsyCap indicators showed the strongest and most
consistent positive relationship with direct measures of performance (i.e., objective and
subjective measures of task performance). Weaker relationships were consistently found
between indicators of hedonic (positive) and eudaimonic well-being (negative) and direct
measures of performance. In interpreting these findings, it is important for readers to consider
that direct measures of performance have been almost completely absent from research in ICE
settings beyond that of Antarctic stations. This may be due to the comparatively large crews that
take part in these missions, and their organizations’ need to use formal performance appraisal
systems as a means of effectively managing performance in these larger units. Additionally,
multiple reviews have sought to compare and contrast the characteristics of long-duration
spaceflight and various analogue settings (e.g., Bishop, 2004; Manzey, 2004), and although
winter-overs typically represent one of the most similar analogues to long-duration spaceflight in
terms of mission duration, it could also be argued that Antarctic wintering crews are less
confined and dependent upon each other than long-duration spaceflight crews. Introducing peer
or leader appraisals of performance in small and highly co-dependent crews may actually be
deleterious to individual and crew performance, as well as interpersonal relations and trust.
Taken together, it may be difficult for future research to obtain direct evidence of performance,
short of establishing unobtrusive and objective measures.
With this in mind, we sought to expand our criteria to indirect measures of performance.
Examining indirect measures of performance, we found that indicators of all three factors
showed positive relationships with individuals’ perceptions of group functioning and cohesion.
Moreover, this evidence appears to be robust across a wide range of specific well-being
74
indicators and ICE settings. Some evidence also exists regarding the indicators of hedonic and
eudaimonic well-being and individuals’ retrospective perceptions of their ICE experiences.
Among this evidence, well-being indicators have been consistently (positively) related to
retrospective “attitude” criteria. However, the magnitude of these effects has been relatively
weak. Taken together, existing research is fairly definitive in demonstrating the importance of
the relationship between well-being and broader group functioning, as compared to evidence
linking well-being to direct measures of performance. As such, this subgroup of
performance/work attitude outcomes may be those to which well-being holds the greatest utility.
In this report, we also examined the relationship between indicators of well-being and
measures of stress and symptomatology. The most consistent evidence relating well-being to
stress has come from positive affect (hedonic well-being), while the evidence relating stress to
eudaimonic well-being and PsyCap indicators has been inconsistent, but generally weak. Much
of the evidence examining hedonic indicators and symptomatology has come through the use of
positive and (the absence of) negative affect, with findings typically being weak. However, one
study employing another indicator of hedonic well-being, satisfaction with social support
(Palinkas & Browner, 1995), showed a moderate relationship with subsequently-measured
depression. A range of eudaimonic well-being indicators have been linked to symptomatology,
and the evidence generally suggests that these indicators may have practical utility with regard to
a range of symptoms. We identified only four studies linking PsyCap indicators to
symptomatology measures. Among these, the strongest evidence was found for hardiness and
self-efficacy, while inconsistent evidence was found for resilience.
In sum, the reviewed evidence suggests each of the three facets of well-being to
differentially relate to performance-related criteria. Based on our findings we conclude that
75
PsyCap is the factor that has been most strongly related to direct measures of performance,
though all three factors have been consistently related to group cohesion and functioning.
Hedonic well-being, and more specifically positive affect, has been most consistently
(negatively) related to perceptions of stress. The strongest relationships with symptomatology
outcomes were observed across a number of eudaimonic indicators. Less robust evidence was
found across hedonic and PsyCap indicators in relation to symptomatology outcomes. However,
we found evidence of negative relationships with symptomatology for satisfaction with social
support (hedonic), self-efficacy (PsyCap), and hardiness (PsyCap).
There are multiple considerations the reader must make when interpreting the evidence
reviewed in this report. First, much of the evidence reviewed here represents concurrently-
measured well-being and performance-related variables. Therefore, the reviewed evidence
cannot be interpreted as causal in nature. Effective performance, positive interpersonal
relationships, and the absence of stress may be just as likely to influence individuals’ well-being,
as well-being is to influence these factors. In fact, the relationship is likely cyclical (e.g.
Fredrickson, 2003). Nonetheless, the link between well-being and performance-related measures
remains important in understanding how to optimize both, while minimizing the likelihood and
magnitude of negative psychosocial functioning among individual crewmembers. Related, it is
clear from the findings reviewed here that those based on concurrently-measured, same-source
data consistently showed stronger effects than did evidence based on multi-source and/or
prospective data. The problems associated with common method bias have been well-
documented (Podsakoff, MacKenzie, Lee, & Podsakoff, 2003), and readers must strongly
consider research design when interpreting the practical significance of primary evidence
reviewed here.
76
Second, although a reasonable amount of evidence exists regarding the well-being-
performance relationship, the research on relationships between specific indicators of well-being
and specific indicators of performance is quite limited. This creates difficulty in attempting to
conduct summary analyses. Integrating the wide range of well-being and performance indicators
into a more parsimonious set of well-being factors and performance-relevant categories provides
greater opportunity to summarize existing evidence and draw general conclusions. However, it
is also important to note that the conclusions presented here are just that, general. There were a
number of instances within this review in which specific well-being indicators representing any
one broader factor showed differential relationships with particular performance-relevant
categories of outcomes. For example, among eudaimonic well-being indicators, evidence
indicates autonomy to be strongly (negatively) related to depressive symptoms (Nicolas et al.,
2013), but locus of control to be weakly (positively) related to depressive symptoms (Palinkas &
Browner, 1995). This highlights the fact that, although many of the well-being constructs
included in this report may show commonality and be expected to covary to some extent, theory
suggests that these indicators are also somewhat unique (e.g., see, for example, Crawford &
Henry, 2004 for a discussion of the theoretical conceptualization of positive and negative affect
as related but different constructs). Moreover, there are also instances in which multiple studies
evidencing the effects of a specific well-being indicator showed different relationships across
performance-relevant outcomes, even those categorized under the same general performance
factor. For example, job satisfaction was significantly and moderately related to group cohesion
but only weakly and non-significantly to intentions to return for a subsequent ICE mission
(Sarris & Kirby, 2005). Making this issue even more complex is the fact that study
characteristics (e.g., crew size, ICE setting, mission duration) varied greatly across the studies
77
reviewed. This likely further affected the magnitude (and possibly directionality) of the well-
being-performance relationships observed across primary studies. However, given that few
specific well-being-performance relationships (e.g., life satisfaction-depression, optimism-stress)
have been assessed in multiple studies, the extent to which study characteristics influenced
observed effects remains largely unknown.
The conclusions drawn from this review provide a foundation for which future research
can develop more targeted studies regarding relationships between specific well-being indicators
and performance criteria. As described above, there was often only a single study (and a single
performance or symptomatological criterion) that examined the effects of any one of these
indicators, and it is unclear the extent to which the effects of these well-being indicators possess
criterion-related validity across a broader range of relevant criteria. Thus, future research should
aim to determine which well-being indicators demonstrate the most robust effects across
performance and psychosocial health criteria. Developing a greater understanding of the extent
to which specific indicators of well-being predict a wider range of performance and
symptomatological criteria will be important in establishing the validity of such measures for use
in long-duration spaceflight selection and training procedures.
Moreover, additional research is needed to examine the effects of multiple indicators
simultaneously in order to further identify which indicators are most effective in predicting
specific outcomes deemed most vital to long-duration mission success (e.g., task performance,
group functioning, stress resistance). For example, a range of eudaimonic, hedonic, and PsyCap
indicators have been shown to positively affect group functioning and cohesion, but it remains
unclear from the existing evidence which indicators most effectively predict these outcomes
when tested simultaneously. Based on the present evidence, it appears that, broadly, PsyCap
78
indicators may have the greatest utility in predicting performance and eudaimonic well-being
indicators in predicting symptoms. However, beyond the effects of positive affect on perceived
stress, there is little definitive evidence regarding which specific well-being indicators (e.g.,
resilience, autonomy) best predict specific outcomes (e.g., task performance, depressive
symptoms). Such research has important implications for predictive efficiency in selection,
training, and monitoring procedures. For selection-related efforts, there exists a considerable
restriction of range among astronaut candidates, as most candidates are highly qualified from a
knowledge, skills, and abilities standpoint. Indicators of psychosocial health may provide a
viable means for differentiating these individuals, especially in the context of long-duration
missions in which resistance to stress may be as important to mission success as will be technical
proficiency. Given the high stakes of such selection decisions, however, it is vitally important
that indicators of psychosocial health chosen to be included as part of selection are actually those
which will best predict effective and healthy functioning among crewmembers.
Consequently, such research will inform future efforts toward the development of any
well-being measures created specifically for NASA’s purposes. That is, identifying which well-
being indicators have the most robust effects across outcomes, as well as which provide the
greatest incremental validity in predicting specific outcomes deemed as most crucial to mission
success, will guide decisions regarding which indicators are most important to capture in order
for such a scale to be effective.
This review provides a foundation for identifying these more specific indicators. For
example, PsyCap indicators demonstrated arguably the most consistent relationship across
categories of performance-relevant outcomes. This is not particularly surprising given the
psychological, social, and environmental stressors that have been associated with ICE missions
79
(e.g., Geuna et al., 1996) and the research demonstrating the protective power of these constructs
against maladaptive stress responses in high-risk and extreme contexts (e.g., Bartone, 2006;
Lamp, 2013; Lester et al., 2011a). However, PsyCap indicators have been those least studied
within the context of ICE environments. It remains unclear which PsyCap indicators have the
strongest effects with various relevant outcomes, as well as whether the whole of these related
constructs is greater than the sum of their parts. Thus, an important direction for future research
may be to further explore the potential preventive and salutogenic effects of resilience and
resilience-based protective factors during long-duration ICE missions.
In addition, this review identified a number of eudaimonic indicators, as well as one or
more hedonic and PsyCap indicators that related to a wide range of symptomatology criteria.
However, given the lack of consistency with which well-being and symptomatological measures
have been employed, this literature is currently underdeveloped. An important future research
need is surely to better understand these various relationships. As a more immediate focus,
future research may seek to identify the symptomatological criteria most crucial to long-duration
mission success (e.g., aggression/hostility, depression) and assess the incremental validity of
indicators identified through this review as potentially most relevant (e.g., altruism,
instrumentality, mindfulness, resilience, satisfaction with social support, self-efficacy) in
predicting these criteria.
Finally, the emphasis of the present review was on ICE crewmembers, an emphasis that
is reflective of the broader ICE literature (for exceptions see Kanas, 2004; Kanas et al., 2002;
Kanas, Weiss, & Marmar, 1996; Leon & Scheib, 2007). However, it has been pointed out that
mission control personnel and crewmembers’ families also play an important role in the well-
being and performance of crewmembers (Brady, 2005; Buckley & Mortimer, 2004), and the
80
deleterious effects of a lack of well-being and positive relations has been documented within the
empirical ICE literature (e.g., Bergan et al., 1993; Kanas et al., 1996). This suggests that greater
well-being among these secondary units may well have positive effects on the performance and
psychosocial functioning of crews. Support for such a relationship can be seen in the military
psychology literature, where interventions have been developed targeting not just soldiers, but
also those individuals central in soldiers’ lives, in attempts to maximize these individuals’
contribution to soldiers’ health and performance (e.g., Park, 2011; Prevail Health Solutions,
2011). Thus, future research should aim to expand its focus to the effects of maintaining and
enhancing well-being among mission controllers on crew-mission control communication
effectiveness and crewmember performance, as well as the effects of crewmember family well-
being on crewmembers’ psychosocial functioning and performance.
Temporal Effects on Well-being
Overall, the summary evidence presented here does not support the third-quarter
phenomenon, which has been commonly theorized to exist during ICE missions (Bechtel &
Berning, 1991). Instead, the meta-analytic evidence reported here strongly suggests that the
critical points of fluctuation across the broad categories of positive and negative psychosocial
functioning appear at the outset and final phase of missions. As described above, research has
argued for a number of trends in addition to the third-quarter phenomenon, including the trend
observed here (see Bergan et al., 1993; see also Mocellin & Suedfeld, 1991; and Sandal, 2001).
One plausible explanation for the trends is that both positive and negative indicators of “arousal”
are highest at the beginning and end points of the mission. For example, ICE researchers have
commonly discussed the presence of a “novelty” effect among crewmembers during the initial
acclimation phase (e.g., Kanas et al., 2001a), wherein crewmembers experience a mix of
81
excitement and anxiety, as well as increased stress while initially adapting to ICE environments.
When nearing mission completion, crewmembers may experience similarly mixed emotions
reflecting increased arousal as they prepare to return and readjust to normal life, while also
potentially experiencing negative emotions due to the understanding that they will be leaving the
environment and individuals to which they have grown accustomed.
One shortcoming of the present analyses and the subsequent conclusions that can be
drawn from them is the broad nature with which the temporal trends of variables could be
assessed. That is, the relative lack of continuity in the specific indicators used in examining
temporal trends in the ICE literature made partitioning these indicators into specific categories of
psychosocial functioning unfeasible; instead we used the very broad categories of positive and
negative aspects of psychosocial functioning. The lack of consensus in the ICE literature over
the temporal changes experienced by ICE crewmembers may well be a function of the more
specific psychosocial indicators being tested. That is, negative indicators of “arousal” and
“depression” may show different trends. For example, Bhargava et al. (2000) showed reports of
sleep difficulty to peak relatively early in an Antarctic winter-over and subsequently recede,
suggesting that individuals experienced initial difficulty adjusting to their ICE environment,
while dissatisfaction slowly grew and peaked during the third quarter of the mission before
receding during the final phase. Using alternative frameworks in conceptualizing psychosocial
health indicators, such as that based on arousal/depression, may provide greater continuity to
existing theories of temporal change across ICE missions. Further, better understanding how
time differentially affects more specific aspects of psychosocial health may also have important
implications for determining the specific countermeasures most important during different
phases of long-duration missions.
82
The moderator analyses presented above provide additional and potentially important
information. First, despite the somewhat divergent trends associated with intermediate-duration
missions, the meta-analytic evidence presented in this report suggests striking consistency in the
temporal trends in both positive and negative psychosocial functioning across short- and long-
duration missions. The demonstration of highly similar temporal effects among crews taking
part in 1-90 day and 181+ day (typically ranging to upwards of one year in duration) missions
alike has potentially important practical implications because it suggests that similar effects may
appear in even longer-duration missions (two-plus years) for which evidence is either non-
existent or highly limited. From a theoretical standpoint, these findings also suggest that existing
explanations of temporal change (e.g., early/final phase arousal) likely generalize to missions of
a wide range of durations.
Second, moderator analyses controlling for crew characteristics (nationality and gender)
indicate potential differences between homogeneous and heterogeneous crews. Existing theory
from the broader literature suggests that homogeneous teams tend to function better
interpersonally (Haleblian & Finkelstein, 1993; Jackson et al., 1991). This phenomenon is
reflected in the temporal trends reported here, and existing primary research has documented
problems associated with heterogeneous crews (e.g., Inoue et al., 2004). This presents a
particularly important issue, given that long-duration spaceflights will most likely involve multi-
national initiatives and crews will likely be heterogeneous in both cultural and gender make up.
However, multiple points warrant further mention. First, this evidence does not suggest
heterogeneous crews have been particularly dysfunctional, and research within the ICE literature
has described the possible greater importance of crew personality characteristics to long-duration
mission success (Bishop, 2004; Palinkas et al., 2001). Second, analyses depict that while some
83
within multi-gender crews showed relatively low scores on positive psychosocial functioning
throughout mission duration, others appeared to thrive in these social contexts, with high positive
functioning scores throughout mission duration. ICE researchers have provided
recommendations for maximizing functioning in gender-heterogeneous ICE crews (Rosnet,
Jurion, Cazes, & Bachelard, 2004), but an important future research need will be to identify and
possibly select on the characteristics possessed by those who respond positively to heterogeneous
crew make-up. More generally, crew compatibility should be taken into consideration in order to
optimize psychosocial functioning throughout mission duration.
It should be noted that many of the findings stemming from the moderator analyses were
based on relatively small numbers of studies and primary sample sizes. Moderator analyses were
not conducted on the individual-level data available, given a clear lack of power. In addition, we
cannot be sure the reported temporal trends associated with many of the group-level moderator
categories reflect stable estimates. For example, the confidence bands reported for the main
effect temporal trends among positive and negative psychosocial functioning scores are not
reported for moderator analyses. This decision was made to aid readers in comparing and
contrasting the trends observed across multiple categories within each moderator analysis.
However, in all cases the unreported confidence bands were very wide, given the data combined
in the main effect analyses were partitioned into 2-3 categories for each moderator analysis.
Thus, moderator analyses should be interpreted with considerable caution.
Finally, analyses reported here were constrained to temporal changes during ICE
missions. A considerable amount of research in the military psychology literature suggests that
pre- and post-mission temporal trends are also important to comprehensively understanding
individuals’ psychosocial health in extreme environments (e.g., Bliese et al., 2007). As such,
84
future research should expand the time frame with which psychosocial health indicators are
monitored. Doing so may prove important in predicting crewmembers ability to initially adapt to
ICE environments. In addition, pre-mission monitoring may help identify the need for more
careful monitoring of specific crewmembers and potentially improve NASA’s ability to identify
specific crewmember’s countermeasure needs throughout the mission phases.
With regard to expanding measurement time points to the post-mission phase, very high
rates of former ICE crewmembers surveyed have reported intending not to return for subsequent
missions (e.g., Sarris & Kirby, 2007). It is unclear why such high rates have been observed, but
it may be due, at least partially, to the stressors associated with such missions. There exists a
complete absence of research within the literature assessing the longer-term (e.g., six-months-to-
one-year post-mission) effects of ICE mission participation on crewmembers’ psychosocial
health. First and foremost, research is needed in order to assess whether ICE missions have
lasting negative effects on crewmembers and ensure that post-mission support and
countermeasures are provided, if necessary. This is particularly important in the context of long-
duration missions, when crews will experience unprecedented levels of isolation and
confinement. Moving forward, this may represent a major concern regarding the potential
effects of long-duration missions. Second, identifying those who maintained healthy
psychosocial functioning post-mission or even demonstrated post-mission growth (Tedeschi,
Park, & Calhoun, 1998) may prove useful to informing crew selection for future long-duration
missions.
Conclusion
This report had three purposes: to identify a model of well-being factors within the ICE
literature (Aim 1), to systematically summarize existing evidence from the ICE literature
85
regarding the well-being-performance relationship (Aim 2), and to summarize temporal effects
on crewmember well-being over the course of ICE mission duration (Aim 3). Although
sufficient primary evidence from ICE settings exists in order to systematically summarize the
well-being-performance relationship and temporal effects on psychosocial functioning over
mission duration, attempts to do so have not previously been made. With respect to the well-
being-performance relationship, this may be a result of the absence of a theoretically-supported
organizing framework of the broad set of well-being indicators that have been, individually,
infrequently studied in ICE environments and/or the absence of evidence linking well-being
indicators to measures directly reflecting performance. By organizing well-being indicators into
three distinct factors (hedonic well-being, eudaimonic well-being, and PsyCap) and qualitatively
reviewing the effects of each in relation to a broad set of performance-related criteria we were
able to draw a number of conclusions to guide future research on this topic:
Eudaimonic well-being indicators have been those most commonly studied in relation to
performance-related measures in the empirical ICE literature, while evidence regarding
the relationships between indicators of hedonic well-being and PsyCap and performance-
related measures have appeared in the ICE literature relatively less frequently.
PsyCap indicators have most consistently shown the strongest, positive relationships with
direct measures of performance (e.g., objectively measured task performance,
peer/supervisor performance ratings). This suggests that selecting on and developing
characteristics such as optimism, resilience, and self-esteem may have the greatest effects
on individual-level task-related performance.
Relatively consistent and moderate (positive) effects were found for indicators of all three
well-being factors on indirect measures of performance, namely self-reported perceptions
86
of group functioning and cohesion. In addition, a consistently positive, but generally
weak, relationship was found between indicators of hedonic and eudaimonic well-being
and retrospective measures of individuals’ positive attitudes towards their ICE
experience. Thus, although hedonic and eudaimonic well-being may not strongly
influence individual-level task performance, these factors may play an important role in
crew functioning and individuals’ perceptions of ICE experiences.
Evidence suggests that positive affect, an indicator of hedonic well-being, is the most
consistently (negatively) related to perceptions of stress.
In general, eudaimonic well-being has been most consistently (negatively) related to
symptomatological outcomes. Less consistent is evidence regarding hedonic well-being
and PsyCap, although evidence suggesting practically significant effects has been
observed with regard to satisfaction with social support (hedonic), self-efficacy (PsyCap),
and hardiness (PsyCap).
Our analyses provided evidence with both theoretical and practical implications regarding the
temporal effects on indicators of both positive (i.e., well-being) and negative psychosocial
functioning (i.e., stress/symptomatology):
Meta-analytic evidence reported here suggests the beginning, and to a lesser extent the
end, of missions to be marked by the greatest changes in both positive (i.e., well-being)
and negative indicators of psychosocial functioning (i.e., stress/symptomatology), with
relatively stable trends during the interim stages of mission duration.
Across mission duration, individuals and crews have reported slightly above average
levels of positive psychosocial functioning and average levels of negative psychosocial
functioning.
87
Evidence among short- (1-90 days) and long-duration (181+ days) missions shows highly
similar temporal trends, suggesting temporal effects to be largely robust.
Evidence regarding crew characteristics (crew nationality and gender) suggests that
homogeneous crews may experience somewhat greater and more stable psychosocial
health throughout mission duration. However, evidence does not suggest that
heterogeneous crews will typically be detrimental to crew psychosocial health. In
addition, there appears to be a clear difference between those who do and do not thrive, at
least, in multi-gender crews.
Based on the results of this study, the following recommendations are offered for the future study
of well-being in ICE settings, with a specific eye toward preparations for a long-duration mission
to Mars:
1. The validation of well-being measures specific to ICE settings is needed:
The present research has shown that indicators of well-being are statistically related to
measures of performance. Better understanding the role of well-being in ICE settings can
inform health surveillance efforts in long-duration spaceflight settings and can contribute
to long-duration spaceflight selection and training procedures.
Using the hedonic well-being, eudaimonic well-being, and PsyCap framework, future
research should aim to determine which specific well-being indicators demonstrate
significant relationships with performance and psychosocial health criteria in ICE
settings.
Next, specific well-being predictors should simultaneously be examined using a
regression framework to determine the incremental effects of each on outcomes of
interest. Doing so will identify which factors are most efficient in predicting specific
88
outcomes deemed most vital to long-duration mission success (e.g., task performance,
group functioning, stress resistance).
Indicators of PsyCap are more strongly related to direct measures of performance than
either hedonic well-being or eudaimonic well-being. Findings suggest that emphasis
should be placed on these indicators of well-being in order to explore the potential
preventive and salutogenic effects of resilience and resilience-based protective factors
during long-duration ICE missions.
HRP is currently undertaking health surveillance efforts to identify the signs, symptoms,
and diagnoses of neurobehavioral health in ICE and spaceflight settings. HRP would
benefit by supplementing this effort or similar future efforts with an examination of the
relationship between indicators of well-being and neurobehavioral health in such settings.
Doing so would provide a comprehensive study assessing the ability of well-being
indicators to predict adverse health and behavioral outcomes.
The research presented in this report suggests that specific indicators of positive and
negative behavioral health likely follow different temporal trends over the course of ICE
missions. Assessing how specific indicators of well-being and symptomatology covary
and/or differ over the course of a mission can be used to develop countermeasures to
enhance specific aspects of well-being, to counteract the negative effects of stress and
symptomatology, or both, at various points during a mission.
Studies to validate ICE-specific measures of well-being can be conducted in a variety of
settings. Earth-based analogs provide the most efficient settings for validation studies to
take place initially. These studies could be carried out among individuals and teams in
analog settings such as Polar expeditions; submerged settings such as NASA Extreme
89
Environment Mission Operations (NEEMO); and analogous Mars settings such as the
Hawai’i Space Exploration Analog and Simulation (HI-SEAS).
Ultimately, low earth orbit settings such as the International Space Station (ISS) will
provide the most suitable analog for inquiries related to long-duration exploration-class
missions. Examining the relationships between well-being and performance, and well-
being and health in this context will allow researchers to validate a comprehensive
measure of well-being for use in a long-duration mission to Mars.
2. NASA should focus on the well-being of ground support and family members:
NASA is currently investigating the applicability of resilience training for ground crews.
It is recommended that NASA continues this effort and also assesses the effects of
maintaining and enhancing well-being among mission controllers. Understanding and
enhancing the well-being of crew-mission control may increase the communication
effectiveness of ground crews, which may ultimately serve to enhance crewmember
performance.
Families serve as a critical point of support for individuals in ICE settings. To date, very
little research has examined the well-being of crew-members’ families. Even less
research has examined how family well-being can impact the well-being and
performance of crew members. Work should be done to examine the effects of
crewmember family well-being on crewmembers’ psychosocial functioning and
performance.
3. NASA should continue to assess well-being of crew members after the mission has ended:
90
Research is needed to assess lasting effects of ICE missions on crewmember well-being.
This information can be used to inform and develop post-mission support procedures, if
necessary.
Research is needed to identify post-mission factors of healthy psychosocial functioning
and growth to inform crew selection for future long-duration missions.
Assessing the post-mission well-being of astronauts and individuals in ICE settings will
be especially critical to understanding the potential impacts of multiple flights/ICE
missions on well-being.
91
References
*indicates study was included in well-being-performance analyses
†indicates study was included in temporal effects analyses
†Abraini, J. H., Ansseau, M., Bisson, T., Juan de Mendoz, J.-L., & Therme, P. (1998).
Personality patterns of anxiety during occupational deep dives with long-term
confinement in hyperbaric chamber. Journal of Clinical Psychology, 54, 825-830.
Adler, A. B., Huffman, A. H., Bliese, P. D., & Castro, C. A. (2005). The impact of deployment
length and experience on the well-being of male and female soldiers. Journal of
Occupational Health Psychology, 10, 121-137.
Armed Forces Health Surveillance Center. (2011). Associations between repeated deployments
to Iraq (OIF/OND) and Afghanistan (OEF) and post-deployment illnesses and injuries,
active component, U.S. Armed Forces, 2003-2010. MSMR, 18, 2-11.
†Atlis, M. M., Leon, G. R., Sandal, G. M., & Infante, M. G. (2004). Decision processes and
interactions during a two-woman traverse of Antarctica. Environment and Behavior, 36,
402-423.
Austin, J. T., & Villanova, P. (1992). The criterion problem: 1917-1992. Journal of Applied
Psychology, 77, 836-874.
Avey, J. B., Luthans, F., & Jensen, S. M. (2009). Psychological capital: A positive resource for
combining employee stress and turnover. Human Resource Management, 48, 677-693.
Avey, J. B., Luthans, F., Smith, R. M., & Palmer, N. F. (2010). Impact of positive psychological
capital on employee well-being over time. Journal of Occupational Health Psychology,
15, 17-28.
92
Avey, J. B., Reichard, R. J., Luthans, F., & Mhatre, K. H. (2011). Meta-analysis of the impact of
positive psychological capital on employee attitudes, behaviors, and performance.
Human Resource Development Quarterly, 22, 127-152.
Barrick, M. R., & Mount, M. K. (1991). The big five personality dimensions and job
performance: A meta-analysis. Personnel Psychology, 44, 1-26.
Bartone, P. T. (2006). Resilience under military operational stress: Can leaders influence
hardiness? Military Psychology, 18, S131-S148.
Basner, M., Dinges, D. F., Mollicone, D., Ecker, A., Jones, C. W., Hyder, E. C., … Sutton, J. P.
(2013). Mars 520-d mission simulation reveals protracted crew hypokinesis and
alterations of sleep duration and timing. Proceedings of the National Academy of
Sciences, 110, 2635-2640.
Bechtel, R. B., & Berning, A. (1991). The third-quarter phenomenon: Do people experience
discomfort after stress has passed? In A. A. Harrison, Y. A. Clearwater, & C. P. McKay
(Eds.), From Antarctica to outer space: Life in isolation and confinement (pp. 260-265).
New York: Springer-Verlag.
Belavý, D. L., Gast, U., Daumer, M., Fomina, E., Rawer, R., Schießl, H., … Felsenberg, D.
(2013). Progressive adaptation in physical activity and neuromuscular performance
during 520d confinement. PloS one, 8, e60090.
†Bell, J., & Garthwaite, P. H. (1987). The psychological effects of service in British Antarctica:
A study using the General Health Questionnaire. British Journal of Psychiatry, 150, 213-
218.
†Bergan, T., Sandal, G., Warncke, M., Ursin, H., & Værnes, R. J. (1993). Group functioning and
communication. Advances in Space Biology and Medicine, 3, 59-80.
93
†Bhargava, R., Mukerji, S., & Sachdeva, U. (2000). Psychological impact of the Antarctic
Winter on Indian Expeditioners. Environment and Behavior, 32, 111-127
Bishop, S. L. (2004). Evaluating teams in extreme environments: From issues to answers.
Aviation, Space, and Environmental Medicine, 75, C14-21.
*†Bishop, S. L., Kobrick, R., Battler, M., & Binstead, K. (2010). FMARS 2007: Stress and
coping in an arctic Mars simulation. Acta Astronautica, 66, 1353-1367.
Bishop, S. L., Santy, P. A., & Faulk, D. (1998). Team dynamics analysis of the Huautla cave
diving expedition—a case study. Human Performance in Extreme Environments, 3, 37-
41.
Bliese, P. D., Wright, K. M., Adler, A. B., Thomas, J. L., & Hoge, C. W. (2007). Timing of
postcombat mental health assessments. Psychological Services, 4, 141-148.
Bonanno, G. A., Mancini, A. D., Horton, J. L., Powell, T. M., LeardMann, C. A., Boyko, E. J…
& Smith, T. C. (2012). Trajectories of trauma symptoms and resilience in deployed U.S.
military service members: Prospective cohort study. The British Journal of Psychiatry,
200, 317-323.
Boyd, J. E., Kanas, N. A., Salnitskiy, V. P., Gushin, V. I., Saylor, S. A., Weiss, D. S., & Marmar,
C. R. (2009). Cultural differences in crewmembers and mission control personnel during
two space station programs. Aviation, Space, and Environmental Medicine, 80, 532-540.
Brady, J. V. (2005). Behavioral health: The propaedeutic requirement. Aviation, Space, and
Environmental Medicine, 76, B13-B24.
*Brasher, K. S., Dew, A. B. C., Kilminster, S. G., & Bridger, R. S. (2010). Occupational stress in
submariners: The impact of isolated and confined work on psychological well-being.
Ergonomics, 53, 305-313.
94
Buckely, N. D., & Mortimer, A. (2004). Introduction and summary of the International Space
Life Sciences Working Group Workshop on Group Interactions. Aviation, Space, and
Environmental Medicine, 75, C1-C2.
Buckman, J. E., Sundin, J., Greene, T., Fear, N. T., Dandeker, C., Gerrnber, N., & Wesseley, S.
(2011). The impact of deployment length on the health and well-being of military
personnel: A systematic review of the literature. Occupational and Environmental
Medicine, 68, 69-76.
Casey, G. W., Jr. (2011). Comprehensive Soldier Fitness: A vision for psychological resilience
in the U.S. Army. American Psychologist, 66, 1-3.
†Cleveland, S. E., Boyd, I., Sheer, D., & Reitman, E. E. (1963). Effects of fallout shelter
confinement in family adjustment. Archives of General Psychiatry, 8, 54-62.
Cornum, R., Matthews, M. D., & Seligman, M. E. P. (2011). Comprehensive Soldier Fitness:
Building resilience in a challenging institutional context. American Psychologist, 66, 4-9.
Cohen, P., Cohen, J., Aiken, L. S., & West, S. G. (1999). The problem of units and the
circumstance for POMP. Multivariate Behavioral Research, 34, 315-346.
Crawford, J. R., & Henry, J. D. (2004). The positive and negative affect schedule (PANAS):
Construct validity, measurement properties and normative data in a large non-clinical
sample. British Journal of Clinical Psychology, 43, 245-265.
Davis, J. R., Fogarty, J. A., & Richard, E. E. (2008). Human health and performance risk
management—an approach for exploration missions. Acta Astronautica, 63, 988-995.
Dawson, S. J. (2002). Human factors in Mars research: An overview. Proceedings of the Second
Australian Mars Exploration Conference.
95
Day, A. L., & Livingstone, H. A. (2001). Chronic and acute stressors among military personnel:
Do coping styles buffer their negative impact on health? Journal of Occupational Health
Psychology, 6, 348-360.
†Décamps, G., & Rosnet, E. (2005). A longitudinal assessment of psychological adaptation
during a winter-over in Antarctica. Environment and Behavior, 37, 418-435.
Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human needs and
the self-determination of behavior. Psychological Inquiry, 11, 227-268.
Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542-575.
Diener, E., Suh, E. M., Lucas, R. E., & Smith (1999). Subjective well-being: Three decades of
progress. Psychological Bulletin, 125, 276-302.
Dinges, D. F., Basner, M., Mollicone, D. J., Jones, C. W., Ecker, A. J., Bartels, R., & Mott, C.
(2014). Effects of time in mission: ISS astronauts ratings of stress. Proceedings of the
NASA Human Research Program Investigators’ Workshop. Houston, TX, February 11-
14, 2014.
*Doll, R. E., & Gunderson, E. K. E. (1969). Occupational group as a moderator of the job
satisfaction-job performance relationship. Journal of Applied Psychology, 53, 359-361.
*Doll, R. E., & Gunderson, E. K. E. (1970). The relative importance of selected behavioral
characteristics of group members in an extreme environment. The Journal of Psychology,
75, 231-237.
*Doll, R. E., Gunderson, E. K. E., & Ryman, D. H. (1969). Relative predictability of
occupational groups and performance criteria in an extreme environment. Journal of
Clinical Psychology, 25, 399-402.
96
*†Eid, J., Johnsen, B. H., Saus, E.-R., & Risberg, J. (2004). Stress and coping in a week-long
disabled submarine exercise. Aviation, Space, and Environmental Medicine, 75, 616-621.
†Evans, G., Stokols, D., & Carrere, S. (1987). Human adaption to isolated and confined
environments: Preliminary findings of a seven-month Antarctic winter-over human
factors study (NASA Grant No. NAG 2-387). Moffett Field, CA: NASA/Ames Research
Center.
Fear, N. T., Jones, M., Murphy, D., Hull, L., Iversen, A. C., Coker, B., Machell, L… Wessely, S.
(2010). What are the consequences of deployment to Iraq and Afghanistan on the mental
health of the UK armed forces? A cohort study. Lancet. 375, 1783-1797.
Fischer, U. M., Mosier, K. L., Munc, A., Reich, K., Swarts, J., Fox, D., & Manchem, S. (2014).
Communication and performance under time delay conditions. Proceedings of the NASA
Human Research Program Investigators’ Workshop. Houston, TX, February 11-14,
2014.
Frederickson, B. L. (2003). Positive emotions and upward spirals in Organizations. In K. S.
Cameron, J. E. Dutton, & R. E. Quinn (Eds.), Positive organizational scholarship
(pp. 163–175). San Francisco: Berrett-Kohler.
Fredrickson, B. L., & Joiner, T. (2002). Positive emotions trigger upward spirals toward
emotional well-being. Psychological Science, 13, 172-175.
Gehrman, P., Seelig, A. D., Jacobsen, I. G., Boyko, E. J., Hooper, R. I., Gackstetter, G. D…
Smith, T. C. (2013). Predeployment sleep duration and insomnia symptoms as risk
factors for new-onset mental health disorders following military deployment. Sleep, 36,
1009-1018.
97
Geuna, S., Brunelli, F., & Perino, M. A. (1996). Stressors, stress, and stress consequences during
long-duration manned space missions: A descriptive model. Acta Astronautica, 36, 347-
356.
*Grant, I., Eriksen, H. R., Marquis, P., Orre, J. J., Palinkas, L. A., Suedfeld, P., Svensen, E., &
Ursin, H. (2007). Psychological selection of Antarctic personnel: The “SOAP”
instrument. Aviation, Space, and Environmental Medicine, 78, 793-801.
†Gushin, V. I., Kolinitchenko, T. B., Efimov, V. A., & Davies, C. (1996). Psychological
evaluation and support during Exemsi. Advances in Space Biology and Medicine, 5, 283-
295.
†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.
Gushin, V. I., Zaprisa, N. S., Kolinitchenko, T. B., & Efimov, V. A. (1997). Content analysis of
the crew communication with external communicants under prolonged isolation.
Aviation, Space, and Environmental Medicine, 68, 1093-1098.Halbesleben, J. R. (2010).
A meta-analysis of work engagement: Relationships with burnout, demands, resources,
and consequences. In A. B. Baker & Leiter, M. P. (Eds.), Work engagement: A handbook
of essential theory and research (pp. 102-117). New York: US Psychology Press.
Haleblian, J., & Finkelstein, S. (1993). Top management team size, CEO dominance, and firm
performance: The moderating roles of environmental turbulence and discretion.
Academy of Management Journal, 36, 844-863.
98
Hoge, C. W., Castro, C. A., Messer, S. C., McGurk, D., Cotting, D. I., & Koffman, R. L. (2004).
Combat duty in Iraq and Afghanistan, mental health problems, and barriers to care. New
England Journal of Medicine, 351(1), 13-22.
Hox, J. (1998). Multilevel modeling: when and why. In I. Balderjahn, R. Mather, & M. Schader
(Eds.). Classification, data analysis and data highways (pp. 147-154). New York:
Springer.
Huppert, F. A., & So, T. T. C. (2013). Flourishing across Europe: Application of a new
conceptual framework for defining well-being. Social Indicators Research, 110, 837-861.
Hunter, J. E., & Schmidt, F. L. (1990). Methods of meta-analysis. Newbury Park, CA: Sage
Publications.
Ihle, E. C., Ritsher, J. B., & Kanas, N. (2007). Positive psychological outcomes of spaceflight:
An empirical study. Aviation, Space, and Environmental Medicine, 77, 93-101.
†Inoue, N., Matsuzaki, I., & Ohshima, H. (2004). Group interactions in SFINCSS-99: Lessons
for improving behavioral support programs. Aviation, Space, and Environmental
Medicine, 75, C28-35.
Jackson, S. E., Brett, J. F., Sessa, V. I., Cooper, D. M., Julin, J. A., & Peyronnin, K. (1991).
Some differences make the difference: Individual dissimilarity and group heterogeneity
as correlates of recruitment, promotions, and turnover. Journal of Applied Psychology,
76, 675-689.
Joint Mental Health Advisory Team 7 (J-MHAT 7). (2011). Operation Enduring Freedom 2020.
Afghanistan. Office of the Surgeon General United States Army Medical Command,
Office of the Command Surgeon, HQ, USCENTCOM, and Office of the Command
Surgeon US Forces Afghanistan (USFOR-A). Retrieved from www.dtic.mil
99
†Kahn, P. M., & Leon, G. R. (1994). Group climate and individual functioning in an all-women
Antarctic expedition team. Environment and Behavior, 26, 669-697.
Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not emotional
well-being. Proceedings of the National Academy of Sciences, 107, 16489-16493.
Kanas, N. (1991). Psychosocial support for cosmonauts. Aviation, Space, and Environmental
Medicine, 62, 353-355.
Kanas, N. (1998). Psychiatric issues affecting long duration space missions. Aviation, Space, and
Environmental Medicine, 69, 1211-1216.
Kanas, N. (2002). Psychosocial and psychiatric issues in space. Journal of Gravitational
Physiology, 9, 307-310.
Kanas, N. (2004). Group interactions during space missions. Aviation, Space, and Environmental
Medicine, 75, C3-C5.
Kanas, N. (2005). Interpersonal issues in space: Shuttle/Mir and beyond. Aviation, Space, and
Environmental Medicine, 76, B126-B134.
*Kanas, N., Gushin, V., & Yusupova, A. (2008). Problems and possibilities of astronauts—
Ground communication content analysis validity check. Acta Astronautica, 63, 822-827.
*Kanas, N., Harris, M., Neylan, T., Boyd, J., Weiss, D. S., Cook, C., & Saylor, S. (2011). High
versus low crewmember autonomy during a 105-day Mars simulation mission. Acta
Astronautica, 69, 240-244.
Kanas, N., & Ritsher, J. (2005). Leadership issues with multicultural crews on the international
space station: Lessons learned from Shuttle/Mir. Acta Astronautica, 56, 932-936.
Kanas, N. A., Salnitskiy, V. P., Boyd, J. E., Gushin, V. I., Weiss, D. S., Saylor, S. A., …
Marmar, C. R. (2007a). Crewmember and mission control personnel interactions during
100
International Space Station missions. Aviation, Space, and Environmental Medicine, 78,
601-607.
Kanas, N., Salnitskiy, V., Grund, E. M., Gushin, V., & Weiss, D. (2002). Lessons learned from
Shuttle/Mir: Psychosocial countermeasures. Aviation, Space, and Environmental
Medicine, 73, 607-611.
†Kanas, N. Salnitskiy, Grund, E. M., Weiss, D. S., Gushin, V., Kozerenko, O., … Marmar, C. R.
(2001a). Human interactions in space: Results from Shuttle/Mir. Acta Astronautica, 49,
243-260.
Kanas, N. Salnitskiy, Grund, E. M., Weiss, D. S., Gushin, V., Kozerenko, O., … Marmar, C. R.
(2001b). Human interactions during Shuttle/Mir missions. Acta Astronautica, 48, 777-
784.
Kanas, N., Salnitskiy, V., Gushin, V., Weiss, D. S., Grund, E. M., Flynn, C., … Marmar, C. R.
(2001c). Asthenia—does it exist? Psychosomatic Medicine, 63, 874-880.
Kanas, N., Salnitskiy, V., Grund, E. M., Weiss, D. S., Gushin, V., Bostrom, A., ...Marmar, C. R.
(2001d). Psychosocial issues in space: Results from the shuttle/MIR. Gravitational and
Space Biology Bulletin, 14, 35-45.
Kanas, N. A., Salnitskiy, V. P., Ritsher, J. B., Gushin, V. I., Weiss, D. S., Saylor, S. A., …
Marmar, C. R. (2007b). Psychosocial interactions during ISS missions. Acta
Astronautica, 60, 329-335.
Kanas, N. A., Salnitskiy, V. P., Weiss, D. S., Grund, E. M., Gushin, V., Kozerenko, O., …
Marmar, C. R. (2007c). Crewmember and ground personnel interactions over time
during Shuttle/Mir space missions. Aviation, Space, and Environmental Medicine, 72,
453-461.
101
*Kanas, N., Saylor, S., Harris, M., Neylan, T., Boyd, J., Weiss, D. S., Baskin, P., … Marmar, C.
(2010). High versus low crewmember autonomy in space simulation environments. Acta
Astronautica, 67, 731-738.
†Kanas, N., Weiss, D. S., & Marmar, C. R. (1996). Crewmember interactions during a Mir space
station simulation. Aviation, Space, and Environmental Medicine, 67, 969-975.
†Kelly, T. H., Hienz, R. D., Zarcone, T. J., Wurster, R. M., & Brady, J. V. (2005). Crewmember
performance before, during, and after spaceflight. Journal of the Experimental Analysis
of Behavior, 84, 227-241.
Kelly, A. D., & Kanas, N. (1992). Crewmember communication in space: A survey of astronauts
and cosmonauts. Aviation, Space, and Environmental Medicine, 63, 721-726.
Kelly, A. D., & Kanas, N. (1993). Communication between space crews and ground personnel:
A survey of astronauts and cosmonauts. Aviation, Space, and Environmental Medicine,
64, 795-800.
Keyes, C. L. M., Shmotkin, D., & Ryff, C. D. (2002). Optimizing well-being: The empirical
encounter of two traditions. Journal of Personality and Social Psychology, 82, 1007-
1022.
†Koscheyev, V. S., Roschina, N. A., & Makhov, V. V. (1994). Psychophysiological
characteristics related to the functional state of the members of the Soviet-American
Arctic Bering bridge. Environment and Behavior, 26, 166-178.
*†Krins, P. W. (2009). Beyond the “right stuff”: The role of group processes in isolated
confined extreme environments. (Unpublished doctoral thesis). The Australian National
University, Acton, Austraila.
102
Lamp, K. (2013). Personal and contextual resilience factors and their relations to psychological
adjustment outcomes across the lifespan: A meta-analysis. (Unpublished dissertation).
Loyola University, Chicago, IL.
Lapierre, J., Bouchard, S., Martin, T., & Perreault, M. (2009). Transcultural group performance
in extreme environments: Issues, concepts and emerging theory. Acta Astronautica, 64,
1304-1313.
LeardMann, C. A., Smith, T. C., Smith, B., Wells, T.S., & Ryan, M. A. K. (2009). Baseline self
reported functional health and vulnerability to post-traumatic stress after combat
deployment: Prospective US military cohort study. British Medical Journal, 338, no
pagination specified.
Lee, R. T., & Ashforth, B. E. (1996). A meta-analytic examination of the correlates of the three
dimensions of job burnout. Journal of Applied Psychology, 81, 123-133.
Leon, G. R. (1991). Individual and group process characteristics of polar expedition teams.
Environment and Behavior, 23, 723-748.
†Leon, G. R., Atlis, M. M., Ones, D. S., & Magor, G. (2002). A 1-year, three-couple expedition
as a crew analog for a Mars mission. Environment and Behavior, 34, 672-700.
*Leon, G. R., Kanfer, R., Hoffman, R. G., & Dupre, L. (1994). Group processes and task
effectiveness in a Soviet-American expedition team. Environment and Behavior, 26, 149-
165.
†Leon, G. R., List, N., & Magor, G. (2004). Personal experiences and team effectiveness during
a commemorative trek in the high Arctic. Environment and Behavior, 36, 386-401.
103
Leon, G. R., McNally, C., & Ben-Porath, Y. (1989). Personality characteristics, mood, and
coping patterns in a successful north pole expedition team. Journal of Research in
Personality, 23, 162-179.
Leon, G. R., & Sandal, G. M. (2003). Women and couples in isolated extreme environments:
Applications for long-duration missions. Acta Astronautica, 53, 259-267.
†Leon, G. R., Sandal, G. M., Fink, B. A., & Ciofani, P. (2011). Positive experiences and
personal growth in a two-man North Pole expedition team. Environment and Behavior,
43, 710-731.
Leon, G. R., & Scheib, A. (2007). Personality influences on a two-man Arctic expedition, impact
on spouse, and the return home. Aviation, Space, and Environmental Medicine, 78, 526-
529.
Lester, P. B., Harms, P. D., Bulling, D. J., Herian, M. N., & Spain, S. M. (2011a). Evaluation of
relationships between reported resilience and soldier outcomes. Report #1: Negative
outcomes (suicide, drug use, & violent crimes). Washington, DC: Defense Technical
Information Center, Department of the Army. Retrieved from www.dtic.mil
Lester, P. B., Harms, P. D., Bulling, D. J., Herian, M. N., Spain, S. M., & Beal, S. J. (2011b).
Evaluation of relationships between reported resilience and soldier outcomes. Report
Number 2: Positive performance outcomes in officers (promotions, selections, &
professions). Washington, DC: Defense Technical Information Center. Department of the
Army Retrieved from www.dtic.mil
Limbert, C. (2004). Psychological well-being and job satisfaction amongst military personnel on
unaccompanied tours: The impact of perceived social support and coping strategies.
Military Psychology, 16, 37-51.
104
Luthans, F. (2002). The need for and meaning of positive organizational behavior. Journal of
Organizational Behavior, 23, 695-706.
Luthans, F., Avolio, B. J., Avey, J. B., & Norman, S. M. (2007). Positive psychological capital:
Measurement and relationship with performance and satisfaction. Personnel Psychology,
60, 541-572.
Luthans, F., Youssef, C. M., & Avolio, B. J. (2007). Psychological capital. Oxford, UK: Oxford
University Press.
Lyubomirsky, S., King, L., & Diener, E. (2005). The benefits of frequent positive affect: Does
happiness lead to success? Psychological Bulletin, 131, 803-855.
MacGregor, A. J., Han, P. P., Dougherty, A. L. Galarneau, M. R. (2011). Effect of dwell time on
the mental health of US military personnel with multiple combat tours. American Journal
of Public Health, 102, S55-S59.
Manzey, D. (2004). Human missions to Mars: New psychological challenges and research issues.
Acta Astronautica, 55, 781-790.
†Manzey, D., Lorenz, D., & Poljakov, V. (1998). Mental performance in extreme environments:
Results from a performance monitoring study during a 438-day spaceflight. Ergonomics,
41, 537-559.
Maslow, A. H. (1971). The farther reaches of human nature. New York: Viking Press.
Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of the antecedents, correlates,
and consequences of organizational commitment. Psychological Bulletin, 108, 171-194.
McFadden, A. (2013). I get by with a little help from my friends: The buffering effects of unit-
level moderators on the combat exposure-mental health relationship (Doctoral
dissertation). All Theses. Paper 1755.
105
Mental Health Advisory Team (MHAT) 6. (2009). Operation Enduring Freedom 2009.
Afghanistan. Office of the Command Surgeon US Forces (USFOR-A) and Office of the
Surgeon General United States Army Medical Command. Retrieved form www.dtic.mil
†Mocellin, J. S., & Suedfeld, P. (1991). Voices from the ice: Diaries of polar explorers.
Environment and Behavior, 23, 704-722.
Mollicone, D. J., Stubna, M. D., Kan, K. G., Mott, C. G., Basner, M., Dinges, D. F., & Pickard,
S. (2014). Software system for real-time medical operational support on ISS.
Proceedings of the NASA Human Research Program Investigators’ Workshop. Houston,
TX, February 11-14, 2014.
Morphew, M. E. (2001). Psychological and human factors in long duration spaceflight. McGill
Journal of Medicine, 6, 74-80.
Morukov, B. V., Rykova, M. P., Antropova, E. N., Berendeeva, T. A., Morukov, I. B.,
Ponomarev, S. A. (2013). Immunological aspects of a spaceflight to Mars. Human
Physiology, 39, 126-135.
*Musson, D. M., Sandal, G. M., & Helreich, R. L. (2004). Personality characteristics and trait
clusters in final stage astronaut selection. Aviation, Space, and Environmental Medicine,
75, 342-349.
Myers, D. G., & Diener, E. (1995). Who is happy? Psychological Science, 6, 10-19.
Nechaev, A. P., Polyakov, V. V., & Morukov, B. V. (2007). Martian manned mission: What
cosmonauts think about this. Acta Astronautica, 60, 351-353.
*†Nicolas, M., Sandal, G. M., Weiss, K., & Yusupova, A. (2013). Mars-105 study: Time-courses
and relationships between coping, defense mechanisms, emotions and depression.
Journal of Environmental Psychology, 35, 52-58.
106
*Palinkas, L. A., & Browner, D. (1995). Effects of prolonged isolation in extreme environments
on stress, coping, and depression. Journal of Applied Social Psychology, 25, 557-576.
Palinkas, L. A., Cravalho, M., & Browner, D. (1995). Seasonal variation of depressive symptoms
in Antarctica. Acta Psychiatrica Scandinavica, 9, 423-429.
*†Palinkas, L. A., Gunderson, E. K., Holland, A. W., Miller, C., & Johnson, J. C. (2000).
Predictors of behavior and performance in extreme environments: The Antarctic Space
Analogue Program. Aviation, Space, and Environmental Medicine, 71, 619-625.
†Palinkas, L. A., & Houseal, M. (2000). Stages of change in mood and behavior during a winter
in Antarctica. Environment and Behavior, 32, 128-141.
Palinkas, L. A., & Johnson, J. C. (1990). Social relations and individual performance of winter-
over personnel at McMurdo station. Antarctic Journal of the United States, 25, 238-240.
Palinkas, L. A., Keeton, K. E., Shea, C., & Leveton, L. B. (2011). Psychosocial characteristics of
optimum performance in isolated and confined environments. (NASA/TM-2011-
216149). Retrieved from: http://ston.jsc.nasa.gov/collections/trs/_techrep/TM-2011-
216149.pdf
†Palinkas, L. A., Reed, H. L., Reedy, K. R., Van Do, N., Case, H. S., & Finney, N. S. (2001).
Circannual pattern of hypothalamic-pituitary-thyroid (HPT) function and mood during
extended Antarctic residence. Psychoneuroendocrinology, 26, 421-431.
Palinkas, L. A., & Suedfeld, P. (2008). Psychological effects of polar expeditions. Lancet, 371,
153-163.
†Palmai, G. (1963). Psychological observations on an isolated group in Antarctica. British
Journal of Psychiatry, 109, 363-370.
107
Park, N. (2011). Military children and families. Strengths and challenges during peace and war.
American Psychologist, 66, 65-72.
Pattyn, N., Migeotte, P.-F., Morais, J., Soetens, E., Cluydts, R., & Kolinsky, R. (2009). Crew
performance monitoring: Putting some feeling into it. Acta Astronautica, 65, 325-329.
Pflanz, S. E., & Ogle, A. D. (2006). Job stress, depression, work performance, and perceptions of
supervisors in military personnel. Military Medicine, 171, 861-865.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method
biases in behavioral research: A critical review of literature and recommended remedies.
Journal of Applied Psychology, 88, 879-903.
Polusny, M. A., Erbes, C. R., Arbisi, P. A., Thuras, P., Kehle, S. M., Rath, M…, & Duffy, C.
(2009). Impact of prior Operation Enduring Freedom/Operation Iraqi Freedom combat
duty on mental health in a predeployment cohort of National Guard soldiers. Military
Medicine, 174, 353-357.
Prevail Health Solutions, LLC (2011). Warriors Prevail pilot program with the Illinois National
Guard : A discussion of goals and outcomes [White Paper]. Retrieved from
http://www.prevailhs.com/wp-content/uploads/2013/02/Warriors-Prevail-ILNG-Pilot.pdf
Rath, T., & Harter, J. (2010). Wellbeing: The five essential elements. New York: Gallup.
Ritsher, J. B., Ihle, E. C., & Kanas, N. (2005). Positive psychological effects of space missions.
Acta Astronautica, 57, 630-633.
Rose, R. D. (2014). Autonomous multimedia resilience training: User experience. Proceedings
of the NASA Human Research Program Investigators’ Workshop. Houston, TX, February
11-14, 2014.
108
*Rose, R. M., Fogg, L. F., Helmreich, R. L., & McFadden, T. J. (1994). Psychological predictors
of astronaut effectiveness. Aviation, Space, and Environmental Medicine, 64, 910-915.
Rosenberg, M. (1979). Conceiving the self. Malabar, FL: Krieger.
Rosnet, E., Jurion, S., Cazes, G., & Bachelard, C. (2004). Mixed-gender groups: coping
strategies and factors of psychological adaptation in a polar environment. Aviation,
Space, and Environmental Medicine, 75, C10-C13.
*†Rosnet, E., Le Scanff, C., & Sagal, M.-S. (2000). How self-image and personality influence
performance in an isolated environment. Environment and Behavior, 32, 18-31.
Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: A review of research on
hedonic and eudaimonic well-being. Annual Review of Psychology, 52, 141-166.
Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of
psychological well-being. Journal of Personality and Social Psychology, 57, 1069-1081.
Ryff, C. D. (1995). Psychological well-being in adult life. Current Directions in Psychological
Science, 4, 99-104.
Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being revisited.
Journal of Personality and Social Psychology, 69, 719-727.
Ryff, C. D., & Singer, B. (1998). The contours of positive human health. Psychological Inquiry,
9, 1-28.
†Sandal, G. M., (2000). Coping in Antarctica: Is it possible to generalize results across settings?
Aviation, Space, and Environmental Medicine, 71, A37-A43.
Sandal, G. M. (2001). Crew tension during a space station simulation. Environment and
Behavior, 33, 134-150.
109
†Sandal, G. M. (2004). Culture and tension during an International Space Station simulation:
Results from SFINCSS’99. Aviation, Space, and Environmental Medicine, 75, C44-C51.
Sandal, G. M., Bye, H. H., & van de Vijver, F. J. R. (2011). Personal values and crew
compatibility: Results from a 105 days simulated space mission. Acta Astronautica, 69,
141-149.
*Sandal, G. M., Endreson, I. M., Vaernes, R., & Ursin, H. (2003). Personality and coping
strategies during submarine missions. Human Performance in Extreme Environments, 7,
28-42.
†Sandal, G. M., Værnes, R., Bergan, T., Warncke, M., & Ursin, H. (1996). Psychological
reactions during polar expeditions and isolation in hyperbaric chambers. Aviation, Space,
and Environmental Medicine, 67, 227-234.
*Sarris, A., & Kirby, N. (2005). Antarctica: A study of person-culture fit. Australian Journal of
Psychology, 57, 161-169.
*Sarris, A., & Kirby, N. (2007). Behavioral norms and expectations on Antarctic stations.
Environmental & Behavior, 39, 706-723.
Sasaki, D., Saitoh, Y., Narita, N., Ishioka, A., Kawakami, K, & Ogawa, K. (1980). A follow-up
study of the changes of psychological states of the members of the 18th Japanese
Antarctic wintering research expedition. Shishin-Igaku, 20, 277-284.
Schmidt, L. L., Wood, J., & Lugg, D. J. (2004). Team climate at Antarctic research stations
1996-2000: Leadership matters. Aviation, Space, and Environmental Medicine, 75, 681-
687.
Schmidt, L. L., Wood, J., & Lugg, D. J. (2005). Gender differences in leader and follower
perceptions of social support in Antarctica. Acta Astronautica, 56, 923-931.
110
Seligman, M. E. P., & Csikszentmihalyi, M. (2000). Positive psychology: An introduction.
American Psychologist, 55, 5-14.
Sin, N. L., & Lyubomirsky, S. (2009). Enhancing well-being and alleviating depressive
symptoms with positive psychology interventions: A practice-friendly meta-analysis.
Journal of Clinical Psychology: In Session, 65, 467-487.
Singer J. D., & Willett, J. B. (2003). Applied longitudinal data analysis: Modeling change and
event occurrence. New York: Oxford University Press.
Soto, C. J. (2014). Is happiness good for your personality? Concurrent and prospective relations
of the big five with subjective well-being. Journal of Personality. Early online view
retrieved from: http://onlinelibrary.wiley.com/enhanced/doi/10.1111/jopy.12081/
Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A meta-
analysis. Psychological Bulletin, 2, 240-261.
†Steel, G. D. (2001). Polar moods: Third-quarter phenomena in the Antarctic. Environment &
Behavior, 33, 126-133.
†Steel, G. D., & Suedfeld, P. (1991). Temporal patterns of affect in an isolated group.
Environment and Behavior, 23, 749-765.
Steinberg, S., Kundrot, C., & Charles, J. (2013). Human health and performance considerations
for near Earth asteroids (NEA). Acta Astronautica, 92, 119-124.
†Stuster, J., Bachelard, M. D., & Suedfeld, P. (2000). The relative importance of behavioral
issues during long-duration ICE missions. Aviation, Space, and Environmental Medicine,
71, A37-A43.
Suedfeld, P., & Brcic, J. (2011). Resolution of psychosocial crises associated with flying in
space. Acta Astronautica, 69, 24-29.
111
Suedfeld, P., & Steel, G. D. (2000). The environmental psychology of capsule habitats. Annual
Review of Psychology, 51, 227-253.
†Taylor, D. A., Wheeler, L., & Altman, I. (1968). Stress relations in socially isolated groups.
Journal of Personality and Social Psychology, 9, 369-376.
Tedeschi, R. G., Park, C. L., & Calhoun, L. G. (1998). Posttraumatic growth: Positive changes
in the aftermath of crisis. Mahweh, NJ: Erlbaum.
Thomas, J. L., Britt, T. W., Odle-Dusseau, H., & Bliese, P. D. (2011). Dispositional optimism
buffers combat veterans from the negative effects of warzone stress on mental health
symptoms and work impairment. Journal of Clinical Psychology, 67, 866-880.
*†van Baarsen, B. (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.
Vasterling, J. J., Proctor, S. P., Friedman, M. J., Hoge, C. W., Heeren, T., King, L. A., & King,
D. W. (2010). PTSD symptom increases in Iraq‐deployed soldiers: Comparison with
nondeployed soldiers and associations with baseline symptoms, deployment experiences,
and postdeployment stress. Journal of Traumatic Stress, 23, 41-51.
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: the PANAS scales. Journal of Personality and Social
Psychology, 54, 1063-1070.
†Weiss, K., & Gaud, R. (2004). Formation and transformation of relational networks during an
Antarctic winter-over. Journal of Applied Social Psychology, 34, 1563-1586.
Weiss, K., & Moser, G. (1998). Interpersonal relationships in isolation and confinement: Long-
term bed rest in head-down tilt position. Acta Astronautica, 43, 235-248.
112
*Weiss, K., Suedfeld, P., Steel, G. D., & Tanaka, M. (2000). Psychological adjustment during
three Japanese Antarctic research expeditions. Environment and Behavior, 32, 142-156.
Wood, A. M., & Joseph, S. (2010). The absence of positive psychological (eudaimonic) well-
being as a risk factor for depression: A ten year cohort study. Journal of Affective
Disorders, 122, 213-217.
†Wood, J., Lugg, D. J., Hysong, S. J., & Harm, D. L. (1999). Psychological changes in hundred-
day remote Antarctic field groups. Environment and Behavior, 31, 299-337.
Wright, T. A., & Doherty, E. M. (1998). Organizational behavior ‘rediscovers’ the role of
emotional well-being. Journal of Organizational Behavior, 19, 481-485.
Wright, T. A., & Huang, C.-C. (2012). The many benefits of employee well-being in
organizational research. Journal of Organizational Behavior, 33, 1188-1192.
*Zuckerman, M., Persky, H., Link, K. E., & Basu, G. K. (1968). Experimental and subject
factors determining responses to sensory deprivation, social isolation, and confinement.
Journal of Abnormal Psychology, 73, 183-194.
113
Appendix A. Characteristics of Studies included in Systematic Review of Well-being-Performance Relationship in ICE Settings.
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
GWB
H
E
PC
S
P
Bishop et al. (2010)
7
Y
B
Simulation
FMARS (2007)
120
x
x
Brasher et al.
(2010)
219
N (UK)
M
Archival
(submarine)
NR
NA
x
x
x
Doll & Gunderson
(1969) Civilan
Scientists
66
N (US)
M
Antarctic
Winter-over
NR
365
x
x
Doll & Gunderson
(1969) Navy
enlisted
129
N (US)
M
Antarctic
Winter-over
NR
365
x
x
Doll & Gunderson
(1970) Civilians
125
N (US)
M
Antarctic
Winter-over
Multiple US
Naval Antarctic
stations and across
multiple years
(1963-1968)
365
x
x
Doll & Gunderson
(1970) Seabees
121
N (US)
M
Antarctic
Winter-over
Multiple US
Naval Antarctic
stations and across
multiple years
(1963-1968)
365
x
x
(Table continues)
114
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
GWB
H
E
PC
S
P
Doll & Gunderson
(1970) Technical
Administrator
93
N (US)
M
Antarctic
Winter-over
Multiple US
Naval Antarctic
stations and
across multiple
years (1963-1968)
365
x
x
Doll et al. (1969)
Civilians
5
N (US)
M
Antarctic
Winter-over
NR
365
x
Doll et al. (1969)
Navy Seabees
5
N (US)
M
Antarctic
Winter-over
NR
365
x
Doll et al. (1969)
Navy technical-
administrative
5
N (US)
M
Antarctic
Winter-over
NR
365
x
Eid et al. (2004)
18
N (NOR)
M
Submarine
NR
6
x
x
X
Grant et al. (2007)
140
N (UK)
B
Antarctic
Winter-over
1999-2003 British
Antarctic
stationers
270
x
x
X
Kanas et al. (2008)
6
Y
NR
Space
ISS
NR
x
X
(Table continues)
115
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
GWB
H
E
PC
S
P
Kanas et al. (2010)
Neemo
25
NR
NR
Submarine
Neemo 12 and 13
12
x
x
Krins (2009) Study
1
25
NR
B
Simulation
Expedition 2
14.5
x
x
x
x
X
Krins (2009) Study
2
12
NR
B
Simulation
Mars Desert
Research Station
14
x
X
Krins (2009) Study
5
9
Y
B
Antarctic
Winter-over
Concordia
Antarctic Station
215
x
X
Leon et al. (1994)
11
Y
M
Arctic
expedition
Soviet-American
Bering Strait
Expedition
61
x
x
X
Musson et al.
(2004)
147
N (US)
B
Archival
(space)
1990, 1992, and
1994 NASA
applicant pools
NA
x
X
Nicolas et al.
(2013)/Kanas et al.
(2010) Mars 105/
Kanas et al. (2011)
6
Y
M
Simulation
Mars105 (March-
July 2009)
105
x
x
x
(Table continues)
116
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
GWB
H
E
PC
S
P
Palinkas &
Browner (1995)
121
N (US)
NR
Antarctic
Winter-over
U.S. Antarctic
Program 1988-
1989
360
x
x
x
x
X
Palinkas et al.
(2000) Operation
Deep Freeze (1963-
1974)
657
N (US)
M
Antarctic
Winter-over
Operation Deep
Freeze (1963-
1974)
NR
x
x
X
Rose et al. (1994)
65
N (US)
B
Archival
(space)
1990 NASA
astronaut corps
supervisory
assessment
NA
x
x
x
X
Rosnet et al. (2000)
16
N (FRA)
M
Antarctic
Winter-over
Dumont-d'Urville
Station
210
x
X
Sandal et al. (2003)
50
Y
M
Submarine
10-d NATO
exercise
21.4
x
x
X
Sarris & Kirby
(2005, 2007)
117
NR
B
Archival
(Antarctic
station)
Australian
Antarctic stations
(1950-2000)
NR
x
x
X
van Baarson (2013)
6
Y
M
Simulation
Mars500
520
x
X
(Table continues)
117
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
GWB
H
E
PC
S
P
Weiss et al. (2000)
107
N (JAP)
M
Antarctic
Winter-over
JARE 1991-1993
330
x
x
x
X
Zuckerman et al.
(1968) sensory
deprivation
12
N (US)
B
Simulation
(bed rest)
NR
1
x
x
X
Zuckerman et al.
(1968) social
confinement
12
N (US)
B
Simulation
(bed rest)
NR
1
x
x
X
Zuckerman et al.
(1968) social
isolation
12
N (US)
B
Simulation
(bed rest)
NR
1
x
x
X
Notes: N = total sample size; M = male only crew; B = male and female crew; NR = not reported; GWB = general well-being; H =
hedonic; E = eudaimonic; PC = PsyCap; S = stress/psychosocial symptomatology; P = performance.
118
Appendix B. Characteristics of Studies included in Individual-level Temporal Change Analyses.
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Atlis et al.
(2004)
2
Y
F
Arctic
expedition
NR
97
Confidence; Energy;
Positive affect;
Negative affect;
Stress
Bell &
Garthwaite
(1987)
12
N (UK)
M
Antarctic
Winter-over
Rothera Base
(April 1983-
March 1984)
330
NA
General health
symptoms;
Bergan et
al. (1993)
6
Y
M
Simulation
EMSI (Aug.-Oct
1993)
25
Cheerfulness; Energy;
Feeling toward others;
Personal/team mood;
Team closeness;
Team harmony
Complaints of
physical
conditions;
Problems outside
team control;
Tenseness
Bishop et al.
(2010)
7
Y
B
Simulation
FMARS (2007)
120
NA
Stress;
Cleveland et
al. (1963)
4
N (US)
B
Underground
shelter
NR
14
Positive mood; Relief
(Mood);
NA
Gushin et
al. (1996)
1
NR
M
Simulation
EXEMSI
60
Joking;
Negative
statements;
Gushin et
al. (2012)
6
Y
M
Simulation
Mars105
(March-July
2009)
105
Activity; Health;
Mood;
Negation;
(Table continues)
119
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Inoue et al.
(2004)
2
Y
B
Simulation
SFINCSS-99
110
Vigor-activity;
Anger-hostility;
Confusion-
bewilderment;
Depression-
dejection;
Fatigue-interia;
Tension-anxiety;
Kahn &
Leon (1994)
4
N (US)
F
Arctic
expedition
1986 North Pole
expedition
67
Positive affect;
Negative affect;
Leon et al.
(2004)
3
Y
B
Arctic
expedition
Otto Sverdrup
Centennial
Expedition (late
2000)
46
Positive affect;
Satisfaction
Negative affect;
Tension/conflict
Leon et al.
(2011)
2
N (US)
M
Arctic
expedition
55
Positive affect;
Negative affect;
Manzey et
al. (1998)
1
N (RUS)
M
Space
MIR (Jan 1994-
March 1995)
438
Emotional
balance/alertness;
Sadness;
Steel &
Suedfeld
(1991)
5
NR
B
Arctic station
(short)
Isachsen
decommissioned
weather station
49
Arousal; Pleasure;
NA
(Table continues)
120
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
van Baarson
(2013)
6
Y
M
Simulation
Mars500
520
Experienced
challenge as
encouragement;
Freedom of choice;
Perceived
voluntariness;
Willingness to
sacrifice;
NA
Wood et al.
(1999)
6
N (AUS)
M
Arctic
expedition
1993-
1994/1994-1995
Lambert Glacier
Basin traverse
105
Cognitive Readiness;
Emotional State;
Personal Morale;
Group tension;
Notes: N = total sample size; M = male only crew; B = male and female crew; NR = not reported; NA = not applicable.
121
Appendix C. Characteristics of Studies included in Group-level Temporal Change Analyses.
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Abaini et al.
(1998)
16
N (FRA)
M
Diving
Hydra V
Hydra VIII
Hydra IX
43
NA
Anxiety;
Apprehension;
Emotional instability;
Low self-control;
Suspicion; Tension;
Bergan et al.
(1993)
6
Y
M
Simulation
EMSI (Aug.-
Oct 1993)
25
Cheerfulness;
Energy; Feelings
toward others;
Personal/team
mood; Team
closeness; Team
harmony;
Complaints of
physical conditions;
Problems outside
team control;
Tenseness;
Bhargava et al.
(2000)
25
N (IND)
M
Antarctic
Winter-over
11th Indian
Expedition
330
Rapport;
Satisfaction with
work/life;
Alcohol intake;
Interpersonal
hypersensitivity;
Sleep difficulty;
Smoking;
Bishop et al.
(2010)
7
Y
B
Simulation
FMARS
(2007)
120
Positive mood;
Negative mood;
Décamps &
Rosnet (2005)
27
N (FRA)
NR
Antarctic
Winter-over
Dumont-
d'Urville
Station (1997-
1998)
350
NA
Stress reactions;
(Table continues)
122
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Eid et al. (2004)
18
N (NOR)
M
Submarine
NR
6
NA
General health
complaints; Negative
impact of event;
PTSD symptoms;
Evans et al.
(1987)
9
N (US)
M
Antarctic
Winter-over
Palmer station
180
NA
Anxiety; Depression;
Hostility;
Gushin et al.
(1996)
1
N (RUS)
M
Simulation
EXEMSI
60
Joking;
Negative statements;
Gushin et al.
(2012)
6
Y
M
Simulation
Mars105
(March-July
2009)
105
Activity;
Health;
Mood;
Negation;
Kahn & Leon
(1994)
4
N (US)
F
Arctic
expedition
1986 North
Pole expedition
67
Positive affect;
Negative affect;
Stress
Kanas et al.
(1996)
3
N (RUS)
M
Simulation
MIR
simulation
135
NA
Mood disturbance;
Kanas et al.
(2001a) US
astronauts
5
N (US)
NR
Space
Shuttle/Mir
120
Cohesion; Self-
discovery;
NA
Kelly et al.
(2005)
4
N (US)
B
Space
STS-89
10
Arousal;
Fatigue;
(Table continues)
123
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Koscheyev et al.
(1994) Russian
6
N (RUS)
M
Arctic
expedition
Soviet-
American
Bering Strait
Expedition
61
Activity; Positive
mood; Well-being;
Anxiety;
Krins (2009)
Study 5
9
Y
B
Antarctic
Winter-over
Concordia
Antarctic
Station
215
Cooperation;
Friendship; Group
identification;
Socialization;
NA
Leon et al.
(2002)
6
Y
B
Arctic
expedition
NR
365
Positive affect;
Negative affect;
Negative events
Mocellin &
Suedfeld (1991)
Antarctic
7
N (UK)
M
Antarctic
Winter-over
19th-20th
explorer early
explorers
240
Arousal; Pleasure;
NA
Mocellin &
Suedfeld (1991)
Arctic
6
N (UK)
M
Arctic
expedition
19th-20th
explorer early
explorers
240
Arousal; Pleasure;
NA
Nicolas et al.
(2013)
6
Y
M
Simulation
Mars105
(March-July
2009)
105
Positive emotions;
NA
Palinkas &
Houseal (2000)
McMurdo
63
NR
B
Antarctic
Winter-over
1991
McMurdo
Station
360
Vigor;
Mood disturbance;
(Table continues)
124
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Palinkas &
Houseal (2000)
Palmer station
7
NR
B
Antarctic
Winter-over
1991 Palmer
Station
360
Vigor;
Mood disturbance;
Palinkas &
Houseal (2000)
South Pole
station
18
NR
B
Antarctic
Winter-over
1991 South
Pole Station
360
Vigor;
Mood disturbance;
Palinkas et al.
(2000) South
Pole Station
(1991-1994)
83
NR
B
Antarctic
Winter-over
South Pole
Station (1991-
1994)
NR
Vigor;
Mood disturbance;
Palinkas et al.
(2001)
12
N (US)
B
Antarctic
Winter-over
McMurdo
Station (1997-
1998)
300
NA
Mood disturbance;
Palmai (1963)
14
N (AUS)
M
Antarctic
Winter-over
Australian
National
Research
Expedition
(1960)
330
Positive emotional
response;
Negative emotional
response;
Rosnet et al.
(2000)
16
N (FRA)
M
Antarctic
Winter-over
Dumont-
d'Urville
Station
210
Harmonic self-
image;
NA
Sandal (2000)
land base
19
Y
NR
Antarctic
Winter-over
Dronning
Maud's land
90
Optimism;
Well-being;
Aggression; Stress;
Sandal (2004)
Group 1
4
NR
M
Simulation
SFINCSS-99
240
NA
Problems outside of
individual control;
(Table continues)
125
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Sandal (2004)
Group 3
4
Y
B
Simulation
SFINCSS-99
110
NA
Problems outside of
individual control;
Sandal et al.
(1996)
hyperbaric
chambers
18
Y
B
Hyperbaric
chambers
ISEMSI;
EXEMSI;
O2 dive
39
Positive
instrumentality;
Aggression; Anxiety;
Homesickness;
Sandal et al.
(1996) polar
expeditions
15
Y
B
Arctic
expeditions
Polaremsi;
Norwegian
South Pole
expedition;
Greenland
North-South
expedition;
MIE Arctic
expedition
80.5
Positive
instrumentality;
Anxiety; Depression
Steel (2001)
Study 1
8
NR
NR
Antarctic
Winter-over
Scott Base
240
Arousal; Vigor
Mood disturbance;
Steel (2001)
Study 2
9
NR
NR
Antarctic
Winter-over
Scott Base
330
Vigor;
Mood disturbance;
Stuster et al.
(2000)
9
N (FRA)
M
Antarctic
Winter-over
French remote
stations (1993-
1994)
193
Positivity;
NA
(Table continues)
126
Study
N
Mixed
Cultural
Crewmember
Sex
Setting
Mission
Mission
Length
Positive
Negative
Taylor et al.
(1968) 20d No
privacy
41
N (US)
M
Simulation
Great Lakes
Naval Training
graduates
20
NA
Anxiety; Stress
Taylor et al.
(1968) 20d
Privacy