Space

NasaArticle.pdf

Home >Business & Finance homework help >Economics homework help >Space

Current Directions in Psychological Science 2015, Vol. 24(3) 200 –207 © The Author(s) 2015 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/0963721414566448 cdps.sagepub.com

Researchers from a variety of disciplines are currently working with NASA to prepare for human exploration of Mars in the next decades. Such exploration will take sci- entific discovery to new heights, providing unprece- dented information about the geology, atmosphere, and potential for life on Mars, including previous life, current life, and perhaps even our own lives in the future. To make these unparalleled discoveries, however, astronauts will need to undertake a novel and unprecedented jour- ney. Moreover, the mission to Mars will require a team of crew members who will have to endure and sustain team performance requirements never seen before.

Imagine living and working in a small, confined space with five other teammates for over a year. Your team needs to complete a series of scientific experiments and perform other rigorous tasks, eventually exploring

a distant location in a dangerous, even life-threatening mission. If you are successful, you will then spend 6 months “commuting” home in the same confined quar- ters and challenging conditions. During this assignment, headquarters cannot provide you with quick advice or coaching, because there is up to a 20-minute communi- cation delay (one-way), but you still need to coordinate as a team with the people back at headquarters. From a personal perspective, during these 2 to 3 years, you can- not see Earth, feel gravity, or spend time with your family. And if you or any of your teammates are having a bad

566448 CDPXXX10.1177/0963721414566448Salas et al.Space Teams research-article2015

Corresponding Author: Department of Psychology, MS-25, Rice University, Sewall Hall 464, Houston, TX 77005 E-mail: esalas@ist.ucf.edu

Teams in Space Exploration: A New Frontier for the Science of Team Effectiveness

Eduardo Salas1, Scott I. Tannenbaum2, Steve W. J. Kozlowski3, Christopher A. Miller4, John E. Mathieu5, and William B. Vessey6

1Department of Psychology, Rice University; 2Group for Organizational Effectiveness, Albany, New York; 3Department of Psychology, Michigan State University; 4Smart Information Flow Technologies, Minneapolis, Minnesota; 5Department of Management, University of Connecticut; and 6Wyle Science, Technology, and Engineering Group, NASA Johnson Space Center, Houston, Texas

Abstract Researchers from a variety of disciplines are currently working with NASA to prepare for human exploration of Mars in the next decades. Such exploration will take scientific discovery to new heights, providing unprecedented information about the geology, atmosphere, and potential for life on Mars, including previous life, current life, and perhaps even our own lives in the future. To make these unparalleled discoveries, however, astronauts will need to undertake a novel and unprecedented journey. Moreover, the mission to Mars will require a team of crew members who will have to endure and sustain team performance requirements never seen before. Multidisciplinary teams of scientists have begun to provide the needed steps to address this challenge. The purpose of this article is (a) to illustrate the kinds of new conceptual frameworks and paradigms needed for teams in space exploration, (b) to delineate promising research paths to ensure that a robust team science can emerge for long-duration space exploration (LDSE), (c) to showcase initial findings and insights from studying astronauts now, and (d) to outline a plan of action for team-effectiveness research in LDSE.

Keywords teams, teamwork, team training, team cohesion

Space Teams 201

day, you cannot simply go out for a walk or call in sick. That is the challenge of long-duration space exploration (LDSE). That is the challenge for NASA.

The aim of the science of team effectiveness is to bet- ter understand how to manage and sustain crew cohe- sion, coordination, and teamwork so as to help crew members succeed, be safe, and thrive during such a mis- sion. To this end, multidisciplinary teams of scientists have begun to provide the needed steps to address this challenge. The purpose of this article is fourfold—first, to illustrate the kinds of new conceptual frameworks and paradigms needed for teams in space exploration; sec- ond, to delineate promising research paths to ensure that a robust team science can emerge for LDSE; third, to showcase initial findings and insights from studying astronauts now; and fourth, to outline a plan of action for team-effectiveness research in LDSE.

The Science-of-Team-Effectiveness Challenge

If we go to Mars, flight crew members and ground con- trol engaging in LDSE will be required to operate under unique conditions that pose both physical and psycho- logical challenges. Specifically, team members will be required to communicate, coordinate, and cooperate for extended periods of time, under complex conditions (e.g., extreme isolation, confinement; Kanas, 2011). Further, such teams will largely operate autonomously, as communication delays will hinder the capacity for imme- diate input from leaders or other outside members (Kanas & Manzey, 2008). Overlaying these issues are the physical demands involved in spaceflight, putting added stress on crew members, and the inherent risk involved in long- duration missions, which leaves little room for perfor- mance lapses (Ball & Evans, 2001; McPhee & Charles, 2009). And finally, add to the mix long durations of bor- ing, monotonous inactivity. All of this creates a high- pressure, high-stress environment that can be harmful for team cohesion, teamwork, and performance (Schmidt, Keeton, Slack, Leveton, & Shea, 2009). LDSE crews thus are not regular teams that can be studied the regular way—the unique conditions they face require us to approach the study of these teams in new and different ways.

The science of team effectiveness has made substan- tial progress in the last decades (for recent reviews, see Kozlowski & Bell, 2003, 2013; Mathieu, Maynard, Rapp, & Gilson, 2008; Salas, Stagl & Burke, 2004). This science has generated evidence-based findings that NASA can draw from to secure mission success (see Salas, Tannenbaum, Cohen, & Latham, 2013; Tannenbaum, Mathieu, Salas, & Cohen, 2012). Furthermore, team science in spaceflight

analog environments such as polar stations, submarines, oil rigs, and other isolated, confined, and extreme envi- ronments has led to important insights into the dynamics of teams operating in these types of environments (e.g., Lugg, 2005; Palinkas, 2003; Schmidt, Wood, & Lugg, 2004).

However, there are significant gaps in our understand- ing of team dynamics under the conditions noted above. For example, while team cohesion has been studied in many settings and several meta-analyses have been pro- duced (e.g., Gully, Devine, & Whitney, 1995), none of these have captured the LDSE conditions under which spaceflight crew members will perform. Even in those environments in which some aspects of LDSE are found (e.g., polar stations, submarines), not all of the environ- mental factors involved in LDSE are present. For exam- ple, while Antarctic stations are physically confining and involve some levels of danger, crews stay at the stations for much shorter durations than those that will be expe- rienced during LDSE. Similarly, naval submariners are exposed to a complex, dangerous, isolated, and confined environment, but crews are rotated regularly.

Moreover, the vast majority of the research foundation is based on cross-sectional data; it is static. We need “intensive” longitudinal research on long-term team func- tioning under isolated, confined, and extreme conditions that are similar to those of LDSE. Team composition, teamwork and team dynamics, individual and team self- regulation, and team adaptation and development are, indeed, challenging to study in these environments, so new ways of studying these teams are of paramount urgency. For example, how can these team members monitor or gauge, unobtrusively, their coordination, com- munication, and cooperation? How can data be captured in real time as they perform? How do these teams self- correct (i.e., get better) without an outside intervention? These are just a sample of key research questions. The good news: Some progress has been made. We elaborate next.

Emerging conceptual paradigms in the science of team effectiveness

Space exploration is really a team sport. It is teams that launch, monitor, operate, and return the space vehicle safely to Earth. Clearly, in LDSE, team cohesion will be paramount to mission success. However, teams are not created equal, and certainly those performing LDSE are not an exception. So, we need a richer and deeper con- ceptualization (and frameworks) of team cohesion, beyond the historical view of it as consisting of (a) com- mitment to the team task and (b) a desire to belong to the team or complete the task and interpersonal

202 Salas et al.

commitment to the team (Festinger, 1950; Forsyth, 2010; Gully et al., 1995). Team cohesion is more than that for LDSE. We submit that team cohesion is a dynamic, multi- faceted, and multilevel phenomenon, with emerging states and stable properties, that is dependent on the task and teammate characteristics.

Our conceptualization of team cohesion as a phenom- enon—a dynamic system, rather than a stable construct— treats it as a complex system of causally linked characteristics that self-reinforce (cf. Borsboom & Cramer, 2013). Cohesion emerges over time, cycles, and fluctuates (Kozlowski, in press; Kozlowski, Chao, Chang, & Fernandez, in press; Schmidt et al., 2009). Team cohesion is affected by variables that are established both before (e.g., team composition, prior experiences) and during the mission (e.g., interpersonal conflicts). What we know today is not enough to help in LDSE. Consider, for exam- ple, what we know about time. There are very few studies that have analyzed team cohesion for weeks and months (Cronin, Weingart, & Todorova, 2011; Kozlowski, in press). Most studies have assessed cohesion for a few hours at best. There is an urgent need for more studies that exam- ine teams for long durations (e.g., months, years), particu- larly when those teams must remain together during that time period.

Our conceptualization of cohesion must also be expanded to include not only aspects of task work and teamwork but psychological factors (i.e., team well- being) as well. LDSE is different from any other job assignment. It is even different (e.g., in its length of time, isolation, and limited space) from all prior space missions (Galarza & Holland, 1999). These differences have impli- cations for the selection of astronauts and the composi- tion of LDSE flight and ground crews, as well as for how to manage the collective performance of all involved in the mission.

Recently, we have also begun exploring the construct of team resilience (Alliger, Cerasoli, Tannenbaum, & Vessey, in press), which refers to the capacity of teams to handle and respond to stressors and challenges. Prior research has examined individual (Friborg, Barlaug, Martinussen, Rosenvinge, & Hjemdal, 2005) and organiza- tional resilience (Somers, 2009), but little work has focused on resilience at the team level. Our initial work has been to define the construct, identify the behaviors resilient teams demonstrate, and establish ways to measure it. We have also been gathering daily measures of team resil- ience from teams in isolated, confined environments, with early results suggesting that team resilience differs from cohesion and other constructs such as psychological safety. The ability to handle both chronic and acute stress- ors will be essential if we want LDSE crews to thrive, so further research related to team resilience is needed.

Measurement of team dynamics

The measurement of team dynamics and effectiveness is challenging. There is a long history of research examin- ing team measurement issues (Brannick, Salas, & Prince, 1997; Rosen, Wildman, Salas, & Rayne, 2012), but there is still no standard measurement approach. There are, how- ever, significant advances in capturing what teams think, feel, and believe (Salas et al., 2012). But for LDSE, the assessment and measurement of team dynamics must look and feel different, yet still be valid and diagnostic. The isolation and delayed communication capabilities that LDSE teams will experience means that they must self-regulate and correct themselves. Team dynamics will need to be continually assessed in an unobtrusive man- ner so that countermeasures (i.e., steps taken to minimize the likelihood of problems or to address existing prob- lems) can be deployed as needed.

Our initial research on the dynamics of team cohesion (and related indicators of this phenomenon) is utilizing experiencing-sampling methods (ESM), which are fre- quent but concise assessments (Beal, in press) of team processes or perceptions experienced by team members. Given the LDSE focus of the research, it is being con- ducted in a range of environments across a variety of dura- tions. For example, we collect ESM data from teams that camp on the ice during the Antarctic summer (6 weeks) and station teams that winter-over (9–12 months). We also collect ESM data in the Human Exploration Research Analog (HERA), a 1- to 2-week asteroid transit simulation; the NASA Extreme Environment Mission Operations (NEEMO) habitat, a 1- to 2-week “weightless” exploration simulation; and the Hawai‘i Space Exploration Analog and Simulation (HI-SEAS), a series of Mars-surface-analog exploration missions with durations of 4, 8, and 12 months. We sample at least daily in all settings and several times daily in some. This research is providing “benchmark” data that is being used to establish expected normative patterns of team-cohesion variation within and between persons and within and between teams. Such data will help to establish standards for detecting anomalies (i.e., when cohesion varies beyond normative patterns). One interesting observation from ongoing work is that each team has unique features and sensitivities with respect to cohesion variability. For example, some teams are highly sensitive to external (e.g., weather, workload) and inter- nal (e.g., interpersonal conflict, unbalanced workload) shocks, whereas others are buffered. Thus, although the benchmark data are useful for establishing general expec- tations about team cohesion, each team is also its own ecosystem. Therefore, it is likely that benchmarks will have to be developed for specific teams prior to the onset of an LDSE. Such “tailored” benchmarks would be used

Space Teams 203

to help teams initiate self-correction episodes or to trig- ger targeted interventions to guide cohesion recovery (Kozlowski et al., in press).

The benchmark data are also useful for validating new technologies to assess cohesion at very high sampling frequencies (Kozlowski et al., in press). There is an emerging array of developing technologies for unobtru- sively assessing physiological indicators and social inter- actions that have the potential to capture the dynamics of team cohesion and psychosocial health (e.g., Olguin, Gloor, & Pentland, 2009; Quwaider & Biswas, 2010). For example, one multidisciplinary research team, composed of organizational psychologists and electrical engineers, is working to develop a high-precision wearable sensor array (Baard, Braun, et al., 2012; Baard, Kozlowski, et al., 2012) that can capture multimodal data indicative of team interaction dynamics, fuse the data to draw inferences about the state of team cohesion, and—when needed— direct interventions to support team-member and team regulation (Kozlowski, Chang, & Biswas, 2013; Kozlowski et al., in press).

In its current state of development, the wearable sen- sor technology is capable of capturing dynamic team- member interactions through several modalities, including face time (face-to-face interaction), physical movement (motion), vocal activity (duration, interval, and intensity of vocalization), and heart rate (beats per minute and variability). These multimodal data streams are sampled at high frequencies and transmitted wirelessly to a com- puter server or to the “cloud.” Real-time data streams can be displayed on a dashboard (e.g., computer display, tab- let, or smartphone). Ultimately, the goal is to develop analytic algorithms that allow the state of team members and the team as a collective to be inferred from the mul- timodal data.

Consider that a collaborative interaction is patterned by a sequence of dynamic indicators. One team member approaches another, with corresponding indicators of physical movement, distance closure, slight elevation in heart rate, face time, vocalization, and finally, disengage- ment by one member or the other. Deviations from estab- lished interaction baselines—large spikes in vocal intensity, heart rate variability, rapid disengagement, and increased social isolation—signal an anomaly in team- member and team functioning. With appropriate algo- rithms applied across interaction patterns, inferences about the status of the team can be drawn. When anoma- lies are detected by the system, countermeasures in the form of feedback, advice, and guidance can then be directed to specific individuals, the team, the team leader, and other relevant parties to support team regulation, cohesion, and effectiveness (Kozlowski et al., 2013). Thus, the approach being advanced is designed to mea- sure and model the unfolding of team process dynamics,

capture reciprocal relations between team processes and team effectiveness over time, and intervene to regulate team functioning (Kozlowski & Chao, 2012; Kozlowski et al., in press).

Most existing measures of team effectiveness, particu- larly those that are survey based, require the interruption of ongoing task work. LDSE crew (and ground-support personnel) are engaged in complex, demanding, high- criticality tasks and are resistant to such interruption, especially if they deem it counter to “productive work” toward their mission. On the other hand, vast quantities of data about human work-based interactions are already available in NASA contexts—in the form of audio, video, and textual records of crew behavior and communica- tions. Since it is through observing interaction and com- munication behaviors that humans naturally infer elements of psychological and social states of their fel- lows, it ought to be possible to automate those inference processes to some extent. Methods for rapidly and easily processing these data and inferring information would be ideal in that they would entail no burden beyond that already acceptable to dedicated astronauts.

Recent research by ourselves and others has suggested that such automated processing has the potential to pro- vide insight into team relationships, cohesion, and per- formance, as well as individual team members’ affective and cognitive states. We have been identifying candidate combinations of assessment techniques, data sources, and psychosocial states of interest to NASA and have focused on automated processing of two types of textual or audio records. First, using variations on techniques pioneered by Pennebaker (Tausczik & Pennebaker, 2010) along with latent-semantic-analysis approaches to senti- ment analysis (Landauer, Foltz, & Laham, 1998; Schmer- Galunder & Sikström, 2007), we have been able to assess many different emotional and attitudinal variations in individual astronauts’ log data, even when these materi- als were prepared for NASA’s publicity or educational purposes. Second, primarily using work we had devel- oped previously based on interactive chat patterns (Miller & Rye, 2012), we are able to assess aspects of team dynamics such as power relationships and their shifts over time and whether the team is experiencing a more or less “comfortable” and “routine” time period. In both cases, these assessments were accomplished simply by automated examination and scoring of team communica- tions during task performance.

In initial promising results from historical archives of NASA data (Wu, Rye, Miller, Schmer-Galunder, & Ott, 2013), we were able to identify and track the emotional states of astronauts keeping blogs or journals as well as find correlations between topics the astronauts wrote about and their positive and negative shifts in emotional states (e.g., one astronaut’s use of terms pertaining to the

204 Salas et al.

crew correlated significantly with his or her use of anxi- ety terms across journal entries, suggesting anxiety about crew members). Our team-interaction analysis techniques were able to detect substantial leadership and comfort/ routine differences between the transcribed radio com- munications of the Apollo 13 mission (with its nearly disastrous oxygen-tank explosion) and the other Apollo missions and to track fluctuations in both parameters over time. In all cases, however, our results could be vali- dated only anecdotally because we lacked any true mea- sure of “ground truth”—that is, an objective measurement of these parameters that fluctuated over the same period. This problem is being addressed in ongoing research comparing automated assessment of crew writing or dia- log with concurrent survey data.

There are, as well, promising approaches to capturing team dynamics unobtrusively and in real time. For exam- ple, Stevens and colleagues (Stevens, Galloway, Wang, & Berka, 2011) have used neurophysiologic indicators to complement communication metrics of team cognition with promising results. Similarly, Guastello, Gorman, Cooke and colleagues (Gorman, Cooke, Amazeen, & Fouse, 2011; Guastello, 2010) have applied nonlinear dynamics and real-time communication pattern analysis to a variety of teams with some success. Taken together, these are encouraging methods for capturing team dynamics—however, better and more robust approaches are needed.

Enhancing and sustaining team effectiveness: Countermeasures

Once we capture and understand team cohesion and dynamics in LDSE, we can do something about it: deploy countermeasures. Considering that space crews will be isolated from other individuals who may serve as media- tors or external problem solvers, it is vital that crew mem- bers be equipped with the means and tools to engage in self-correction and regulation. Team training strategies, then, must be designed, developed, and validated to ameliorate the effects of stress on team cohesion. Current preparation for spaceflight is extensive, with astronauts spending a total of 5 to 10 years training before partici- pating in their first spaceflight mission (Kanas & Manzey, 2008). However, much of this training is done individu- ally or in ad hoc teams, with little training done together by the full team that will engage in the mission (Vessey, 2014). Furthermore, the majority of this training is based around the development of technical skills specific to crew-member roles and tasks, with limited time devoted to the development of more generalizable teamwork skills. There are many promising team-development interventions (see Salas & Frush, 2012; Shuffler, DiazGranados, & Salas, 2011) that can be applied for

these purposes. One that shows much promise is a strat- egy pioneered by Smith-Jentsch and colleagues (Smith- Jentsch, Cannon-Bowers, Tannenbaum, & Salas, 2008) called Team Dimensional Training. This strategy com- prises a structured debriefing guide to focus on team- work processes and to encourage, promote, and reinforce teams’ monitoring and awareness of learned behavior over time. Training based on this approach is beginning to be deployed at NASA.

Clearly, team debriefings, in which teams reflect on their experience and reach agreements about how to work together going forward, are a potentially powerful countermeasure. We conducted a meta-analysis on the efficacy of debriefs that showed that, on average, teams that debrief perform 20% better (Tannenbaum & Cerasoli, 2013). But traditionally, debriefs are often led by a trainer or facilitator. For LDSE, teams will need to be self-reliant, and temporal communication lags with ground control will preclude real-time facilitation from Earth. Therefore, teams must be able to self-debrief and adapt on their own. Automated debrief-guidance countermeasures are required.

We have been experimenting with one such tool with teams living and working together in confined environ- ments, including crews in NASA’s land-based HERA and undersea NEEMO habitats. The tool gathers and analyzes crew input to produce a customized debrief guide for each team, with a focus not only on team- and task work but also on factors that might affect team resilience. Data analysis is underway, but initial findings have been posi- tive, and crews consistently reported that this technique surfaced important teamwork challenges and helped them to self-adjust. A recent study conducted in a more controlled environment showed that teams using this guided debrief approach demonstrated better teamwork behaviors and, subsequently, better team performance and attitudes (Eddy, Tannenbaum, & Mathieu, 2013).

Team composition

During LDSE, crew members cannot take a day off to get away from their coworkers, and no changes can be made to crew membership. Therefore, mission success is con- tingent upon selecting the right team members who can not only perform their respective mission roles effectively but also work well together. Team-composition research- ers have shown that the traditional personnel selection model, which focuses on selecting the most qualified per- son to perform each designated role, will not necessarily yield the most effective team when collaboration and teamwork are at a premium. Therefore, we have been developing a more complete framework for understand- ing team composition (Mathieu, Tannenbaum, Donsbach, & Alliger, 2014). NASA-funded research is examining the

Space Teams 205

mix or complementarity of crew-member attributes, including social team-role preferences and readiness, backup abilities, collective orientation, and living-style preferences, in an effort to optimize crew composition.

Fortunately, there is a very large applicant pool of individuals who aspire to be astronauts, so it is possible to select highly competent individuals. However, although the applicant pool is large, the number of individuals selected to be part of the astronaut program is quite small, and it takes many years to prepare them to be mis- sion ready. So, for any given mission, there are very few potential crew members to choose from, greatly limiting the number of possible team configurations. In addition, when international partners in missions can each choose their own crew members, the flexibility to compose a crew with an optimal, complementary team profile is reduced even further.

Recently, researchers have begun to emphasize that team composition should be considered as a foundational context for understanding how a team is likely to work together, even when membership is unchangeable (Mathieu, Tannenbaum, Donsbach, & Alliger, 2013). Team composition provides the crew with certain strengths and advantages, while also presenting certain liabilities or areas of vulnerability. Therefore, NASA research is begin- ning to examine how composition can predict specific teamwork challenges and how to use those insights to trigger targeted team countermeasures.

The Way Forward

LDSE presents formidable challenges for the science of team effectiveness, but they can be overcome as oppor- tunities for testing and validating findings and counter- measures emerge.

Members of NASA’s research community are moving forward on several fronts to help prepare for future mis- sions. For example, researchers are conducting studies in realistic analog environments such as HERA, NEEMO, and the Antarctic to learn more about team dynamics in isolated, confined, and extreme environments (Stuster, 2011; Vessey, Palinkas, & Leveton, 2013). In addition, sig- nificant work is ongoing to determine how best to con- ceptualize and measure team cohesiveness in a meaningful way and capture affective states over time (e.g., Kozlowski, 2012; Wu et al., 2013). Further, work is going on to clarify the “social roles” that crew members need to fulfill in addition to their technical roles (e.g., Mathieu et al., 2014; Roma et al., 2013).

In terms of team development interventions, some researchers are studying how various countermeasures such as team training and debriefing can enhance team effectiveness, with an emphasis on enhancing team

self-sufficiency, adaptiveness, and resilience (e.g., Eddy et al., 2013; Salas et al., in press; Smith-Jentsch et al., 2008; Tannenbaum & Cerasoli, 2013). Team-effectiveness scientists are also exploring how unobtrusive measures such as team-member proximity, naturally occurring communications, and even videotaped team interactions can reveal insights about team dynamics (e.g., Kozlowski, DeShon, Biswas, & Chang, 2012). Still others are seeking to understand the risks associated with a team’s composi- tion profile so that team-specific countermeasures can be evoked, akin to “personalized medicine” for teams (e.g., Tannenbaum, Mathieu, Alliger, & Donsbach, 2013; Wu et al., 2013).

Progress in these areas will help NASA prepare for future space missions, and it is also likely to yield tools and techniques of interest to other team researchers, as well as insights that can be applied to teams in various settings. Space is indeed a new frontier for energizing the science of team effectiveness.