Space Medicine Meets Serious Games: Boosting Engagement with the Medimon Creature Collector

Abstract

Serious games that integrate educational content with engaging gameplay mechanics hold promise for reducing cognitive load and increasing student motivation in STEM and health science education. This preliminary study presents the development and evaluation of the Medimon NASA Demo, a game-based learning prototype designed to teach undergraduate students about the musculoskeletal and visual systems—two critical domains in space medicine. Participants (n = 23) engaged with the game over a two-week self-regulated learning period. The game employed mnemonic-based characters, visual storytelling, and turn-based battle mechanics to reinforce medical concepts. Quantitative results demonstrated significant learning gains, with posttest scores increasing by an average of 23% and a normalized change of c = 0.4. Engagement levels were high across multiple dimensions of situational interest, and 74% of participants preferred the game over traditional formats. Qualitative analysis of open-ended responses revealed themes related to intrinsic appeal, perceived learning efficacy, interaction design, and cognitive resource management. While the game had minimal impact on short-term STEM career interest, its educational potential was clearly supported. These findings suggest that mnemonic-driven serious games like Medimon can effectively enhance engagement and learning in health science education, especially when aligned with real-world contexts such as space medicine.

1. Introduction

Students in STEM and health science disciplines often face cognitive load due to the volume and complexity of required content [1,2,3]. This challenge, combined with decreasing interest in traditional educational approaches, contributes to disengagement and difficulty retaining knowledge [4,5,6,7]. As a result, there is growing interest in innovative teaching strategies that can both reduce cognitive load and increase motivation to learn—particularly among younger learners. Game-based learning (GBL) has emerged as a promising approach that promotes active learning, motivation, and knowledge retention through interactive and immersive experiences [8,9,10,11]. Serious games, which are games whose primary purpose is for education or training rather than entertainment, have been increasingly adopted in health professions education for their ability to blend learning objectives with engaging gameplay [12,13]. Targeted examples underscore the breadth of serious-game applications in health care training. Empiric, a tabletop card game for guideline-based antibiotic prescribing, engaged practicing clinicians at two continuing medical education conferences, who rated the session highly and reported intent to change prescribing behavior [14]. Likewise, AntibioGame^®, a digital case-based simulation on rational antibiotic use, was perceived as attractive, usable, and fun by medical students [15]. Together, these studies illustrate how well-designed GBL interventions can foster both knowledge acquisition and behavior change across the spectrum of medical education.

Another well-established method for supporting learning in high cognitive load fields is the use of mnemonics [16,17]. Mnemonics leverage visual, linguistic, and spatial cues to aid in memory encoding and recall, making them useful for students managing large quantities of information common in STEM and health science fields [18,19,20,21]. Some common examples of visual mnemonics used in medical education are Sketchy Medical and Picmonic [22,23]. While both GBL and mnemonics have been individually validated, few studies meaningfully combine the two into a unified educational experience.

To address this gap, we developed Medimon, an innovative GBL platform that merges mnemonic-based learning with the mechanics of a creature-collector role-playing game [24]. In Medimon, players explore fictional environments, capture characters based on body systems, and engage in turn-based battles using moves and abilities grounded in medical mnemonics. The platform is designed to reduce cognitive load, enhance engagement, and improve knowledge retention through multisensory learning.

Our current preliminary study centers on the Medimon NASA Demo prototype, which focuses on two body systems of critical relevance to NASA and the future of space exploration: the musculoskeletal and visual systems. Astronauts operating in microgravity environments experience significant health challenges, including skeletal tissue degeneration, muscle atrophy, cardiovascular deconditioning, and Spaceflight-Associated Neuro-Ocular Syndrome (SANS) [25,26]. These issues are not only clinically important but also serve as compelling educational topics with real-world relevance that can capture student interest.

The Medimon NASA Demo was designed as a prototype serious game to teach undergraduate students about these systems through a self-regulated learning (SRL) environment [27,28]. We hypothesize that this preliminary demo will increase student knowledge of these body systems, foster engagement with health science education, and potentially boost interest in pursuing STEM-related careers. Though piloted with undergraduates, the demo’s content aligns with advanced high-school biology standards, positioning this as an early-stage feasibility study.

2. Materials and Methods

2.1. Participants

Undergraduate students at the University of Idaho were recruited via institution email invitations and in-person presentations delivered by the authors. Because a central research aim was to examine whether the intervention could stimulate interest in pursuing STEM careers among students not already committed to those fields, we deliberately focused our in-person recruitment efforts on courses outside the traditional STEM disciplines. Brief 15 min study invitation presentations were delivered in two settings that met this criterion yet still showed a natural affinity for game-based learning: (i) students in the Virtual Technology and Design (VTD) program—whose curriculum includes video-game production—and (ii) an entrepreneurship course in the College of Business and Economics. While our primary intent was to engage non-STEM students, we did ultimately enroll a small number (n = 3) of participants majoring in STEM-related fields, including computer science, cybersecurity, medicine, and nutrition science (

Table 1. Participant demographics. Avg: Average, SD: Standard deviation.

Characteristic	Count (Avg)	SD
Age	(21)	2.2
Gender
Male	11
Female	9
Non-binary/third gender	3
Ethnicity
White	23
Hispanic	1
Asian	3
American Indian/Alaskan Native	1
Major
Virtual Technology and Design	12
Creative Writing	1
Conservation Biology	1
Computer Science	1
Cybersecurity	1
Studio Art and Design	1
Medicine	1
Nutrition Science	1
Marketing	2
English	1
Human Resources Management	1
Business Management	1
Finance	1

).

Recruitment occurred on a rolling basis throughout the Spring 2025 semester. Overall, 92 undergraduates provided informed consent to take part in the study (Figure 1). Of these, 83 students (90%) completed the baseline (pre-test) survey. During the subsequent two-week intervention window, 61 students discontinued participation—by never launching the game or by not completing the post-test or the Situational Interest Survey–Multimedia (SIS-M)—leaving 23 students (28% of those who completed the pre-test) who (a) engaged with the game, and (b) submitted both the post-test and the SIS-M. These 23 participants constituted the final analytic sample for all achievement, engagement, and interest analyses reported below. Demographic information is summarized in Table 1. Some participants indicated multiple ethnicities and Majors.

2.2. Design

The prototype was designed as a top-down, role-playing game (RPG) with creature-collector mechanics, drawing inspiration from early-generation titles such as Pokémon Red, Blue, and Yellow for the Game Boy. Players navigate a 2D overworld from a bird’s-eye perspective, encounter and recruit educational creatures (called “Medimon”), and engage in turn-based battles that integrate medical content with gameplay. This familiar format was selected to evoke nostalgia, lower the learning curve for non-gamers, and scaffold engagement through recognizable game systems. The game was developed using the Unity game engine and compiled for both Windows (PC) and macOS platforms. Installation files were distributed to participants via the digital game-hosting platform itch.io, allowing easy and secure access for download.

The game prototype taught concepts from the musculoskeletal and visual systems using mnemonic-based characters and environments. Each character was part of a Medimon family representing a specific body system. Characters were designed across three evolutionary stages—Stage 1 (baby), Stage 2 (adolescent), and Stage 3 (adult)—with both healthy and diseased versions. Healthy Medimon represented normal anatomy and physiology, while diseased Medimon embodied related medical conditions. Diseased versions could be transformed into their healthy counterparts using in-game therapy items modeled after real-world treatments. Families and characters included in the prototype are listed in Appendix

Table A1. Families and characters in the game. 1 = stage 1, 2 = stage 2, 3 = stage 3.

Family	Healthy Characters	Diseased Characters
Bone	1. Osteoclast 2. Osteoblast 3. Skeleton	1. Osteoporosis 2. Postmenopausal Osteoporosis 3. Paget Disease
Muscle	1. Myofibril 2. Muscle Fiber 3. Muscle	1. Myasthenia Gravis 2. Epilepsy 3. Rhabdomyolysis
Eye	1. Rod 2. Cone 3. Eye	1. Retinitis Pigmentosa 2. Color Blindness 3. Glaucoma

.

Each character featured visual mnemonics, which were visual designs that reflect the biological function or pathology of the real-world item they represent (Figure 2). For example, the Osteoclast character visually breaks down bone, symbolizing bone resorption. Locations within the game also supported mnemonic learning, such as the Bone family’s building, “Skeleton Crew Framing,” a hardware store where players accessed visual aids and encountered related wild Medimon (Figure 3).

The turn-based battle system enabled players to battle and capture wild Medimon. Victories awarded experience points that allowed Medimon on the player’s team to level up, enhance stats, and evolve. Key game statistics included:

• HP (Health Points): Measures life; reaching zero causes the Medimon to “flat line,” meaning it cannot be used in battle until healed.
• ATP (Adenosine Triphosphate): Energy required to perform moves in battle, representing cellular energy.
• STS (Stress): Increases with each move during battle; high STS may cause healthy Medimon to evolve into diseased versions or diseased Medimon to self-destruct.
• Attack and Defense: Determine battle effectiveness and resistance to damage.

Moves within battles contained linguistic mnemonics (Figure 4). For example, the Osteoclast’s “Bone Loss” move reinforces its biological role in bone resorption.

Players could also battle non-playable characters (NPCs) to level up their Medimon and visit a hospital to restore HP, ATP, and reduce STS. A comprehensive tutorial guided players through core mechanics, and the in-game menu provided explanations of educational content and mnemonics.

Cognitive Load Theory (CLT) distinguishes between intrinsic load (the complexity inherent to the material), extraneous load (processing demands imposed by the learning environment), and germane load (mental effort devoted to schema construction) [3]. With these principles in mind, we deliberately incorporated several game mechanics that either lighten unnecessary load or channel effort toward meaningful learning. First, visual mnemonics—for example, depicting an Osteoclast physically breaking a bone to signify bone resorption—provide concrete, image-based cues that “chunk” complex physiological information, thereby lowering intrinsic load while simultaneously exploiting dual-coding effects [29]. Second, context-sensitive tool-tips and inventory pop-ups supply just-in-time explanations of battle moves or character functions at the moment players need them; this eliminates the split-attention problem and trims extraneous load. Third, a guided tutorial sequence introduces mechanics in graduated steps rather than all at once, keeping the element interactivity within the learner’s working-memory capacity and managing intrinsic load through scaffolding. Finally, spaced-retrieval Medimon battles require players to revisit earlier content under slightly varied conditions; by prompting effortful recall at optimally spaced intervals, these encounters shift mental resources toward germane load, fostering robust schema formation and long-term retention. Collectively, these intertwined design decisions operationalize CLT within the gameplay loop, ensuring that the cognitive effort students expend is both efficient and educationally productive.

2.3. Intervention

Participants were invited to download and play the Medimon NASA Demo, a serious game prototype integrating health science content related to the musculoskeletal and visual systems. They were instructed to engage with the game over a two-week period in an SRL manner, allowing them to control when and how often they played. The game is freely available online at: https://panacea-interactive.itch.io/medimon-nasa-project (accessed on 3 August 2025).

2.4. Achievement

To evaluate learning, participants completed a short multiple-choice questionnaire before (pretest) and after (posttest) the two-week game period. Questions assessed knowledge of cellular functions, anatomical structures, and associated diseases covered in the game. Participants completed the pre-/posttest through the Qualtrics platform.

2.5. STEM Career Interest

As part of the pretest and posttest, participants rated their interest in pursuing a STEM career using a 5-point Likert scale (1 = Very interested; 5 = Very not interested). Changes in response were used to evaluate whether the game influenced career aspirations.

2.6. Data Collection

After completing the posttest, participants also completed the Situational Interest Survey of Multimedia (SIS-M), a validated instrument for evaluating multimedia engagement [30,31]. The SIS-M evaluates triggered interest, maintained interest, maintained feeling, and value interest using a 5-point Likert scale (1 = Strongly disagree; 5 = Strongly agree) (Appendix

Table A2. SIS items.

SIS Type	Survey Item
SI-triggered	The Medimon NASA demo was interesting.
	The Medimon NASA demo grabbed my attention.
	The Medimon NASA demo was often entertaining.
	The Medimon NASA demo was so exciting, it was easy to pay attention.
SI-maintained-feeling	What I learned from the Medimon NASA demo is fascinating to me.
	I am excited about what I learned from the Medimon NASA demo.
	I like what I learned from the Medimon NASA demo.
	I found the information from the Medimon NASA demo interesting.
SI-maintained-value	What I studied in the Medimon NASA demo is useful for me to know.
	The things I studied in the Medimon NASA demo are important to me.
	What I learned from the Medimon NASA demo can be applied to my major/career.
	I learned valuable things from the Medimon NASA demo.

), and has been used in prior medical education studies [32,33,34,35]. The 12-item SIS-M was administered with two additional items: (1) a preference ranking between different health science learning modalities (Medimon game, traditional materials, or no preference), and (2) an open-ended explanation of that preference. Participants completed the SIS-M through the Qualtrics platform.

2.7. Data Analysis

Quantitative data analysis was conducted using Microsoft Excel, GraphPad Prism (v 9.5.1 (733)), and GPT o3 [36]. Student achievement was assessed by calculating the mean pretest and posttest scores for each participant group, with statistical comparisons conducted using a paired two-tailed t-test (α = 0.5) and a Wilcoxon signed-rank test. To further evaluate learning gains, normalized change calculations were applied, providing a standardized metric for interpreting improvement relative to baseline performance [37].

For the Situational Interest Survey in Multimedia (SIS-M), responses were analyzed across four validated dimensions of situational interest: triggered interest, maintained interest, maintained-feeling, and maintained-value. We classified mean SIS-M scores < 2.5 = low, 2.5–4 = moderate, >4 = high engagement.

The open-ended SIS-M survey responses were analyzed using a large language model (LLM)-driven qualitative workflow informed by recent applications in medical education research [32,33,34,35]. An agentic analysis process was utilized [38,39], which was powered by the “thinking” Google Gemini 2.5 Pro Experimental 03-25 LLM. The process involved three specialized generative artificial intelligent (genAI) agents, each guided by distinct system-level prompts tailored to their roles: Principal Investigator (PI), Qualitative Researcher (QR), and Subject Matter Expert (SME) (Figure 5a). The workflow operated in a sequential, iterative format in which the output from one agent was refined by the next (Figure 5b). Prompt engineering strategies, including Persona Prompting [40,41], were used to optimize agent performance. A detailed account of the full workflow—including all prompts, agent instructions, and resulting outputs—is provided in the Supplemental Materials (Methods S1).

2.8. Ethical Considerations

The study protocol was reviewed and approved as exempt by the Institutional Review Board (23-042) at the University of Idaho. Informed consent was obtained from all participants prior to data collection.

3. Results

3.1. Achievement

All participants completed a multiple-choice questionnaire before and after engaging with the Medimon NASA Demo. Analysis revealed that posttest scores were significantly higher than pretest scores (pretest: 42% ± 13%; posttest: 65% ± 24%; p < 0.001; Wilcoxon signed-rank test: W = 11, n = 20, p = 0.00043, r = 0.78), indicating a learning gain (Figure 6a,b). To account for baseline knowledge variability and ceiling effects, we applied the normalized change method [37]. The average normalized change across participants was c = 0.4 (Figure 6c), representing a moderate improvement in knowledge acquisition during the two-week self-regulated learning period.

3.2. Engagement

The SIS-M provided insights into participants’ levels of interest and engagement with the Medimon NASA Demo. The participants’ had a high average triggered situational interest (M = 4.15, SD = 0.78) and moderate maintained feeling (M = 3.92, SD = 0.79) in the game (Figure 7a).

The results for maintained interest (M = 3.56, SD = 1.02) and maintained value (M = 3.21, SD = 1.10) were lower but still moderate, indicating a more neutral stance with the game. Overall, the majority of participants (74%, 17/23) preferred the game for learning the health sciences, compared to the minority of participants who preferred traditional types of materials such as lectures and textbook readings (9%, 2/23) and those who had no preference in learning material format (17%, 4/23) (Figure 7b).

To understand the underlying reasons for the learning modality preferences reported in the SIS-M, thematic analysis was conducted on participants’ open-ended explanations. This analysis yielded four key themes illuminating the factors driving participants’ choices: (1) Intrinsic Appeal and Affective Engagement, (2) Perceived Learning Efficacy and Mechanisms, (3) Interaction Design and Learner Experience, and (4) Cognitive Factors and Resource Management (

Table 2. Thematic analysis of the SIS-M open-ended responses. Trad: Traditional.

Theme 1: Intrinsic Appeal and Affective Engagement
Description	This theme captures the enjoyment, motivation, and emotional engagement of learners. It reflects how fun, esthetically pleasing, and attention-grabbing each modality was perceived to be.
Prefer Medimon	Participants frequently highlighted the game’s engaging and entertaining nature as central to their preference. Many linked fun to increased focus, emotional reward, and effective learning. Multimodal features (e.g., art, music) were especially impactful.
Prefer Trad	Enjoyment was viewed as secondary. These participants emphasized learning efficacy and saw games as supplemental.
No Preference	Acknowledged the game’s fun and appeal but doubted its learning value. Some felt enjoyment did not equate to educational effectiveness.
Theme 2: Perceived Learning Efficacy and Mechanisms Description
Description	This theme focuses on how well participants believed each method supported understanding and memory, and which features helped or hindered learning.
Prefer Medimon	Reported that the game made content easier to understand and remember. Valued mnemonics, interactivity, and integration of content into gameplay.
Prefer Trad	Believed textbooks provided better depth and clarity. Viewed games as better for review, not initial learning.
No Preference	Questioned the game’s educational depth. Noted retention challenges and issues with recalling in-game content.
Theme 3: Interaction Design and Learner Experience Description
Description	This theme explores how features like interactivity, structure, and user control shaped learner experience.
Prefer Medimon	Valued the game’s interactive nature and familiarity with game mechanics. Highlighted active engagement.
Prefer Trad	Appreciated the control and structure textbooks provided. Emphasized ease of navigation and review.
No Preference	Criticized the game’s lack of structure and goals. Wanted more learning guidance and incentives. Still, some praised its accessibility.
Theme 4: Cognitive Factors and Resource Management Description
Description	This theme involves perceived mental effort and efficiency. It covers how learners managed cognitive demands and time investment.
Prefer Medimon	Some felt the game reduced cognitive load, making complex ideas easier to process through engaging formats.
Prefer Trad	Prioritized time efficiency and directness of textbooks for faster learning.
No Preference	Found the game cognitively demanding without clear structure. Highlighted the need for active attention to extract learning.

).

3.3. STEM Career Interest

To measure the game’s potential impact on career interests toward pursuing a career in a STEM field, participants rated their desire to pursue a STEM career before and after gameplay. Most participants showed no change in their interest in pursuing a STEM career over the two-week period. However, five students reported increased interest in STEM careers, while five reported decreased interest (Figure 8). These results suggest that while the demo had a positive educational impact, its effect on career intentions was limited during the short intervention window. It should also be noted that this data reflects short-term interest, not career commitment.

4. Discussion

The Medimon NASA Demo was designed to merge GBL with mnemonic strategies to address two persistent challenges in STEM and health science education: cognitive overload [1,2] and student disengagement [42]. Our preliminary findings suggest that this approach can improve student knowledge in key biomedical areas while maintaining high levels of learner engagement in an SRL environment.

Participants demonstrated significant learning gains, as evidenced by a 67% average improvement in posttest scores and a normalized change score of c = 0.4. This supports other studies also showing increased efficacy of serious games to promote learning and retention [43]. These results are particularly notable given the brief, two-week exposure to the intervention and the lack of structured instructional support, allowing for higher SRL [27,28]. The results affirm that the game’s specific mechanisms, including its interactive nature and mnemonic-rich design (key elements highlighted in Theme 2: Perceived Learning Efficacy and Mechanisms), can serve as effective learning supports even without traditional scaffolding. The finding lends further support to multimedia-learning theory, which posits that interactive, learner-controlled exploration can substitute for external scaffolds when instructional materials are well-designed [44]. In particular, Medimon’s mnemonic-rich visuals and creature narratives harness dual-coding and imagery principles, mechanisms repeatedly shown to strengthen recall and transfer in biomedical domains [29]. Taken together, the results indicate that Medimon can function as a stand-alone learning tool, a feature that is especially attractive for content areas such as space medicine where traditional classroom resources are limited.

The SIS-M results further reinforce the educational potential of Medimon. Participants reported high levels of triggered situational interest and maintained feeling. These findings align with previous studies in medical and STEM education that have shown serious games can foster motivation and sustained engagement [45]. These quantitative findings align strongly with Theme 1 (Intrinsic Appeal and Affective Engagement) identified in the qualitative analysis, where participants emphasized the game’s fun, engaging nature, and esthetic appeal. Importantly, qualitative feedback highlighted that the integration of visual and linguistic mnemonics helped students grasp and retain difficult concepts, such as the function of osteoclasts in bone resorption or the etiology of conditions like retinitis pigmentosa and myasthenia gravis. This speaks directly to Theme 2 (Perceived Learning Efficacy and Mechanisms), as participants explicitly valued how mnemonics, interactivity, and the integration of content into gameplay made learning easier and more memorable. The presence of spatially located mnemonic-rich environments and NPCs provided contextual reinforcement, turning complex biological and pathological processes into accessible and memorable experiences [46]. Participants reported more moderate levels of maintained interest and value-related interest. This outcome is consistent with the participant demographics, as the majority of students were not enrolled in STEM-related majors and, therefore, may not have perceived a health science educational game as highly relevant or essential to their academic studies.

Interestingly, while the game succeeded in improving knowledge and engagement, it had minimal impact on students’ STEM career aspirations. Only five participants reported increased interest in pursuing a STEM field, while five others reported a decline. This suggests that while GBL tools like Medimon can effectively enhance content understanding and immediate engagement, influencing long-term identity formation and career trajectories may require more prolonged, repeated exposure or the integration of mentorship and narrative career exploration within the game environment [47,48,49,50]. Career interest is often shaped by a broader range of psychosocial factors, including role modeling [51,52], perceived self-efficacy [50,53], and exposure to professional pathways [54,55]. Thus, future iterations of Medimon could include embedded role-play elements, career profiles, or narrative arcs that highlight the real-world applications of biomedical knowledge in various STEM professions, particularly in space medicine.

Evidence suggests that stable personality traits may moderate both engagement with serious games and downstream educational outcomes. For example, individuals high in Openness to Experience tend to prefer novel, visually rich learning environments, whereas those high in Conscientiousness often exhibit greater persistence and goal-directed behavior—qualities that would translate into deeper interaction with game mechanics and higher knowledge gains [56,57]. Recent work using game-based assessment platforms further demonstrates that fine-grained Big Five facets can be reliably inferred from in-game behaviors such as exploration patterns, risk taking, and time-on-task [58]. Although personality was not measured in the present study, future iterations should incorporate brief, validated inventories (e.g., NEO-FFI) and log-data analytics to parse how individual differences shape situational interest, persistence, and ultimately academic achievement. Accounting for these moderating variables will clarify for whom—and under what dispositional conditions—Medimon-style interventions are most effective.

4.1. Limitations

Several limitations of the preliminary current study should be noted. First, the sample size was relatively small (n = 23) and limited to students from a single institution, which may affect the generalizability of the results. One potential contributor to the high participant dropout rate may have been the complexity of the game access process. Participants were required to create an itch.io account, download the itch.io app, save the game to their account, and then install and play the game through the app. Future studies will explore more streamlined and user-friendly methods for distributing and accessing the game to reduce barriers to participation. Second, we did not track in-game behavior metrics such as time played, specific characters interacted with, or the frequency of battles completed. These metrics could provide a deeper understanding of how specific game mechanics influence learning and engagement. Third, the two-week duration may have been too short to observe meaningful changes in career interest. Fourth, the qualitative findings highlight specific design limitations perceived by some participants, particularly the lack of pedagogical structure, clear goals, and review mechanisms (Theme 3). These factors may have dampened learning outcomes or preference for students who thrive on more guided learning, suggesting the current prototype may be more appealing to intrinsically motivated or experienced gamers. Furthermore, concerns about learning depth and recall difficulty (Theme 2) identified by some participants suggest areas where the game’s instructional design could be strengthened. Longitudinal studies are needed to assess how sustained interaction with GBL platforms like Medimon influences educational outcomes and career development over time.

Despite these limitations, this study provides compelling preliminary evidence that serious games integrating mnemonic strategies can enhance learning and engagement in undergraduate health science education. The thematic emphasis on musculoskeletal and visual systems within the context of space medicine also offers a novel interdisciplinary entry point for students who may not initially be drawn to STEM fields but are intrigued by the adventure and problem-solving aspects of space exploration. The self-paced nature of the demo aligns well with modern learner preferences [59,60] and demonstrates the value of autonomy and GBL feedback in promoting knowledge acquisition [61].

4.2. Future Directions

Future research should expand the scope of Medimon to include additional body systems and pathologies, incorporate real-time data analytics, and assess its impact across diverse learner populations. Future iterations of Medimon should directly address the design feedback revealed through the qualitative analysis (Theme 3). This includes incorporating clearer learning objectives, progress tracking, feedback mechanisms, structured challenges or ‘learning checks’, and potentially dedicated review modes to enhance learner control and address recall concerns (Theme 2). These would incorporate problem-based tasks with appropriate feedback to promote extended learning. There is also potential to explore collaborative modes of gameplay and integration into flipped or hybrid classroom models. As space agencies and health care systems prepare the next generation of scientists and clinicians, tools like Medimon offer a scalable, engaging, and pedagogically sound platform to inspire and educate tomorrow’s innovators. Finally, future work could explicitly embed career exploration elements or narratives to potentially influence STEM interest more effectively.

5. Conclusions

Within the limits of a small convenience sample, these findings provide preliminary evidence that the Medimon NASA Demo represents a promising fusion of game-based learning and mnemonic strategies tailored to the unique needs of STEM and health science education. By embedding visually and linguistically rich mnemonics into an engaging, interactive gaming experience, the prototype succeeded in increasing student knowledge of the musculoskeletal and visual systems—two critical domains in space medicine. Participants not only demonstrated significant learning gains but also reported high levels of engagement and enjoyment in the game.

While the demo did not lead to a marked increase in STEM career interest over the short duration of the preliminary study, the strong indicators of cognitive and emotional engagement suggest that platforms like Medimon can serve as effective entry points into complex biomedical content. These results support the growing body of evidence that serious games, when thoughtfully designed, can bridge the gap between passive content delivery and active, meaningful learning.

Future work should focus on expanding content coverage, enhancing career exploration elements, integrating in-game analytics, and testing the platform across diverse learner populations and institutional contexts. With further development and longitudinal evaluation, Medimon has the potential to play a transformative role in health science education, helping cultivate a more scientifically literate and inspired generation of learners—both on Earth and beyond.