Broadly, we have observed that digital systems for behavioral and mental health management in space and space-analog settings (eg, ICE environments) comprise 3 categories: (1) systems for individual and team monitoring; (2) digital countermeasures; and (3) integrated applications that combine monitoring and countermeasure functions (see Table 1 below for examples). We will discuss research in these areas in the following sections.

Numerous studies have focused on monitoring the health of astronauts and analogous populations [13]. Methods that require active user input include digitized diaries [18], cognition tests [19], and optical stress and fatigue detection systems [20,21]. Due to the additional burden imposed by active input requirements, efforts have focused on developing passive monitoring tools for space. Passive monitoring methods include detecting signs of stress in speech [22,23] and nonverbal behavior [24], wearable devices to monitor sleep [25,26] and team cohesion [27,28], and techniques for the real-time sampling of stress biomarkers [29].

With varying levels of granularity, these systems have the potential to provide time-series data capturing intraindividual variability across markers of behavioral and mental health. At present, such monitoring systems are primarily used to collect research data. However, they could also convey information back to the user (or another decision-maker, eg, a crew commander) to inform timely decisions to prevent decrements in health and well-being and to maintain or improve the function of the astronaut and crew [13]. The utility of monitoring tools, especially those that collect data passively, for capturing dynamics in psychological and social functions is still undetermined. Although metrics derived from passive sensors may offer a degree of objective data, important questions remain about what can be inferred from that data about the subjective state of the person or group being monitored [30].

In addition to monitoring, assessment, and diagnosis, several digital countermeasures to prevent deterioration in health and well-being during LDSE missions have been developed. For instance, the Dartmouth PATH program has produced interactive media that provides training on reducing stress, enhancing mood, and resolving conflict [12,31]. Self-directed resources, such as the Expedition Application for Peak Psychological Performance [31] or Virtual Space Station [32], may be part of a repertoire of tools made available during long voyages [33].

Virtual reality (VR) and augmented reality are also promising methods for mitigating some of the demands posed by living and working under conditions of long-term isolation and confinement. For example, projects have explored how VR might be used to manage sensory deprivation [34], provide a platform for interaction and social consistency [10,16], and optimize mental health and well-being in space [35]. Although there has been some initial evaluation of these methods, much more work is needed to determine when, how, and why such countermeasures are effective.

Beyond digital countermeasures, a range of other health and well-being countermeasures for space are being developed and tested, such as mindfulness training courses [36]. For LDSE, these programs may eventually be translated in their entirety or at the very least augmented with some kind of digital element (eg, refresher training) [37].

Advances in monitoring and digital countermeasures create the possibility for closed-loop integrated applications [13]. Our research group is currently exploring how individuals can use a relatively simple, private tool to self-manage their well-being [38]. Equipping individuals to monitor key markers of behavior, health, and performance can guide personalized interventions and optimize well-being to improve functioning in extreme environments like space.

More sophisticated systems include those like the Crew Interactive Mobile Companion (CIMON), a robotic teammate that has flown on the ISS and has supported astronauts in completing work tasks. A stated future aim is to attempt to imbue CIMON (or at least one of its offspring) with some degree of “emotional intelligence” (ie, the capacity to detect and respond to emotional states) so that it might provide effective psychosocial and health and well-being support to astronauts [39]. Similarly, the Translational Research Institute for Space Health has recently funded a project by Ejenta, which aims to develop a conversational artificial intelligence (AI) that could provide emotional support to crew members during LDSE missions. Like CIMON, this work draws upon advances in natural language processing, speech recognition, and machine learning.

Although these integrated projects are in their infancy and may be some way off from being realized in the ways imagined, the trend toward using real-time monitoring data to drive smart, personalized behavioral and mental health solutions is likely to continue. With this level of automation, there is a need to ensure that systems are trustworthy, do not cause inadvertent harm, and positively contribute to their users' well-being.

While previous work suggests many positive opportunities, caution is warranted when considering technology-driven mental health management solutions. Using the earlier example of CIMON, several issues were highlighted when the robot was flight-tested on the ISS. For example, it would not turn off on voice command, requiring the astronaut running the experiment to repeat himself. Later, CIMON responded to requests in a way that may be perceived as controlling by the user (eg, after the astronaut asked CIMON to stop playing music, it replied using guilt-inducing language by saying “Don't you like it here with me?”). Although the prior example led to amusement amongst observers, if CIMON was truly relied upon as an integrated member of a deep-space crew, this could lead to problems.

More broadly, and to illustrate this point further, imagine what if instead of promoting positive mental health, systems like VR exacerbated feelings of separation, isolation, and confinement by continually reminding astronauts what they cannot have. Or, if intelligent conversational agents (like CIMON), drawing upon poorly validated monitoring data, suggest an individual coping strategy that undermines crew health and safety. Or, maybe the algorithms underpinning autonomous systems include design and training data biases that make them less useful (or indeed more harmful) to crew members from different cultural, ethnic, or neurodiverse backgrounds. Biases might come in many shapes and forms and range from poor accuracy in emotion recognition (eg, mistaking one emotion for another) and inappropriate attributions of a facial or linguistic expression to a particular internal state (eg, assuming a jovial comment corresponds to an individual being happy) to culturally inappropriate suggestions for stress mitigation (eg, attempting to self-manage an issue where seeking support might be most appropriate).

Clearly, the end goal of these systems is to maintain and improve health and well-being. However, to date, most of the work in this area has been driven by technological opportunity rather than cogent psychological (and broader social science) theory. One issue with such a technology focus, often a theoretical approach to design, is that the human user's diverse needs can be underestimated. Further, in the absence of a guiding theory, if the technology does not influence the psychosocial experience in an expected way, it can be difficult to explain why [40]. Even if the technology does influence the psychosocial experience as expected, absent theorizing about the mechanisms makes it challenging to identify further opportunities and risks or potential boundary conditions.

Using psychological theory to inform careful design should reduce the likelihood of health technologies having an adverse impact on individuals and crews. Both psychology and human factors research have a rich set of frameworks that could help guide design decisions to prioritize well-being [41]. According to positive computing principles in human factors research, there are 3 ways that digital systems can be designed for well-being [42]. Measures can be preventative, active, or dedicated. The preventative design looks for obstacles to well-being within the system, which, if identified, triggers a redesign. The active design integrates well-being promotive features into an application that is not directly targeted at improving mental health. Dedicated design deals with technologies that are intentionally built for well-being and human flourishing. Although at different stages of maturity, the applied work discussed earlier would benefit from adopting a dedicated design approach. Moreover, all technologies designed for human spaceflight, well-being focused or otherwise, would benefit from taking, at minimum, a preventative approach.

A second issue concerns design for engagement. As astronauts will be responsible for their own self-care and self-management on LDSE missions [9], it is crucial that they feel motivated to use technologies designed to support their well-being. Timely engagement is critical, as interaction with these systems may be most salient precisely when they are psychologically vulnerable (eg, experiencing stress, fatigue, maintaining a high workload, having low motivation, or having interpersonal difficulties) and, thus, less motivated to engage with these systems. Both system engagement and well-being could be promoted by applying coherent psychological frameworks to design. One such approach would be to use a model like the Motivation, Engagement, and Thriving in User Experience (METUX) framework, which encourages designers to consider how technology influences health and well-being at multiple levels of influence. For instance, by considering the impact of interactions with the interface and technology-enabled tasks, and the influence of the technology on behavior and life impact outside of technology use [43]. METUX is theoretically grounded and draws specifically upon tenets within basic psychological needs theory [44], which suggests that human beings function optimally when their needs for autonomy (ie, a sense of volition), competence (ie, a sense of effectiveness), and relatedness (ie, a sense of connection) are met [45].

In support of the METUX model, emerging evidence suggests that consideration of basic psychological needs can foster user engagement [46], enjoyment, and motivation [47], improve learning [48], and contribute to mental health and well-being [49]. In keeping with these findings, recent evidence from Mars isolation and confinement experiments indicates that when basic psychological needs are fulfilled, crew members report better quality motivation and happiness, are less stressed, work more cooperatively, engage in less oppositional defiance, and perform to a higher standard [50]. Such results underline why designing digital technologies with users' basic psychological needs in mind is beneficial for astronaut health, well-being, and function in space.

Although crew members will operate more autonomously during LDSE missions than they do in LEO, this does not imply that they will experience autonomy in the way that the basic psychological needs theory and the METUX models describe [40]. Enforced independence and self-reliance may actually undermine feelings of autonomy and pose risks to competence and relatedness, especially if individuals feel ineffective at managing demands, are understimulated, and feel disconnected from other crew members, or abandoned by those on Earth [1]. Digital health technologies for LDSE could minimize these effects by considering METUX design principles.

In Figure 1, we have provided a basic visual guide for designing off-world applications using the METUX principles. In essence, we suggest that the type of digital technology being developed will directly shape design decisions in each of the spheres of influence. At each level, designers are prompted to think about how users experience the technology via the impact the interactions with the technology have upon their basic psychological needs for autonomy, competence, and relatedness. If users (astronauts in this case) experience need fulfillment both in and beyond the technology use, this should result in motivation and engagement in later interactions with the technology and have broader benefits for their behavior, performance, and health.