Behavioral Traces and Player Typologies in Gamified VR: Insights for Adaptive and Inclusive Design

Abstract

Gamified virtual reality (VR) environments are increasingly used to enhance engagement and learning, yet most designs still adopt a one-size-fits-all approach that overlooks motivational diversity. The HEXAD framework, which classifies users into six player types, provides a promising lens for addressing this gap, but its predictive validity in immersive VR remains contested. This study investigates how HEXAD profiles shape navigation, time allocation, and engagement dynamics in an open-ended gamified VR environment. Thirty-two undergraduate participants, all regular gamers, completed the HEXAD scale before exploring a VR setting with five thematic islands without predefined tasks. System logs and screen recordings captured first island choices, sequential visit patterns, and time spent, and data were analyzed using qualitative pattern analysis alongside nonparametric statistics. The results showed significant associations between player type and initial choices, with Players favoring Game Island, Socialisers choosing Social Island, Philanthropists engaging most with Library, and Achievers and Free Spirits drawn to Experience. Kruskal–Wallis tests of exploration breadth revealed moderate effect sizes across types, though significance was limited by sample size. Three emergent strategies, Focused Explorers, Wanderers, and Strategic Switchers, captured motivational orientations beyond single metrics, while heat map visualizations highlighted clustering around Game and Experience Islands. By situating these findings within flow theory and inclusive–adaptive design principles, this study demonstrates how behavioral traces can link motivational typologies with embodied interaction. Overall, the results advance debates on HEXAD’s robustness and contribute practical pathways for developing adaptive, motivation-sensitive VR environments that support sustained engagement and inclusivity.

1. Introduction

The integration of gamification principles into immersive virtual reality (VR) environments has gained significant momentum in recent years, offering new possibilities for enhancing user engagement, motivation, and learning outcomes across a wide range of domains, including education, rehabilitation, and digital interaction [1,2]. While VR enables multisensory and spatially rich experiences, its design often fails to account for the diverse cognitive and motivational needs of users. In particular, most VR-based gamified systems are built upon a one-size-fits-all assumption, applying standardized interaction mechanisms and game elements regardless of users’ behavioral profiles or preferences [3]. This design limitation not only risks disengagement for certain user groups but also limits the potential of VR systems to support inclusive and personalized user experiences. As the field moves toward more adaptive and user-centered approaches, understanding how different users navigate, engage with, and spend time in gamified VR environments becomes increasingly critical [4]. Recent reviews also highlight that while VR has been widely studied for its motivational potential, research on concrete behavioral engagement markers such as navigation patterns, time allocation, and sequential choices remains scarce, underscoring the need for targeted investigations in this area [5,6].

One of the most widely adopted models for capturing individual motivational differences in gamified systems is the HEXAD player typology, which classifies users into six core types: Philanthropist, Socialiser, Free Spirit, Achiever, Player, and Disruptor, based on their intrinsic and extrinsic motivational drivers [3,7]. Each type exhibits distinct preferences regarding autonomy, social interaction, mastery, reward sensitivity, or disruption, offering a rich framework for tailoring gamified interactions. Prior research has demonstrated that aligning game elements with user types can significantly improve engagement and system effectiveness [1,2,8]. Despite the growing popularity of this model in the context of web-based or mobile gamified platforms, its application in immersive VR environments remains limited. In particular, how these player types influence navigation patterns, decision-making processes, and engagement dynamics in open-ended, non-task-oriented VR spaces is still underexplored. Moreover, HEXAD itself has been subject to criticism, with studies questioning its psychometric robustness [8], its predictive validity in VR and learning contexts [9], and its overlap with alternative typologies such as Bartle’s four-player model [10] or BrainHex [11]. Self-determination theory (SDT) has also been proposed as a stronger explanatory lens in many gamification studies as it directly connects motivational needs of autonomy, competence, and relatedness with sustained engagement [12,13]. Positioning HEXAD in dialogue with these frameworks allows for a more balanced understanding of both its contributions and its limitations. Addressing these theoretical debates while extending the model to immersive VR provides an opportunity to examine not only whether user types differ but also how those differences manifest in observable behavioral strategies. Such a perspective helps bridge the gap between motivational theory and embodied interaction in VR.

Recognizing and responding to motivational and behavioral diversity is not only a matter of personalization but also a fundamental concern of inclusive and adaptive system design. Inclusive design advocates for the development of systems that accommodate a broad spectrum of user needs, preferences, and capabilities, especially those that deviate from the normative user profile. In the context of immersive VR, where spatial freedom, sensory immersion, and nonlinear exploration are key characteristics, the need for systems that dynamically adapt to user profiles becomes even more pronounced. Rigid gamification structures that ignore user type variability may inadvertently exclude users or induce cognitive overload, resulting in disengagement or task abandonment [14]. Conversely, adaptive gamification strategies driven by real-time interaction data and informed by typological insights hold great promise for enhancing flow, maintaining engagement, and fostering meaningful experiences across diverse user groups [1,3,15].

From a theoretical perspective, time spent, navigational choices, and interaction sequences in immersive VR can be viewed as externalized indicators of cognitive load, user intent, and engagement state. Flow theory suggests that sustained immersion and perceived control over challenges are key to maintaining optimal engagement, particularly in gamified environments [14,16]. When the challenge level, interactivity, and user motivation are not well aligned, users may either disengage or experience overload. In VR specifically, however, behavioral traces such as which content users visit first, how they sequence their navigation, and how long they remain engaged have rarely been systematically studied despite their strong potential as signals of motivational orientation and design fit [10,17]. By analyzing these behaviors through the lens of player typologies, designers can better understand how different users perceive and interact with the same virtual space, thus enabling more dynamic and inclusive design strategies [9].

Building on these theoretical foundations, the present study investigates how users with different HEXAD player types behave in a gamified VR environment in terms of their initial interaction preferences, navigational sequences, and time allocation patterns. By analyzing user behavior across three dimensions (first island visited, sequence of visited locations, and time spent per island), meaningful patterns that reflect motivational orientations and engagement tendencies are uncovered. These three dimensions were selected because they represent core behavioral markers of engagement in open-ended VR exploration, directly linking player types to embodied traces of navigation, decision-making, and persistence. This framing responds to recent calls in VR research for more systematic investigations of engagement indicators beyond self-report. To this end, this paper contributes to the literature on adaptive and inclusive gamification by offering a behavior-based framework for player-type-informed VR design. Unlike task-based or performance-driven VR studies, this work adopts an open-ended, free exploration scenario to allow authentic behavioral variation to emerge. The findings contribute to ongoing efforts in inclusive and adaptive gamification by offering empirically grounded insights into how player-type-informed design can support more personalized, accessible, and user-aligned virtual experiences [1,18]. Ultimately, this research provides both conceptual and practical implications for designing VR environments that are sensitive to individual differences and capable of fostering sustained engagement across diverse user profiles.

Accordingly, the primary research question addressed in this study is

“How do users with different motivational profiles, as defined by the HEXAD framework, behave in terms of initial choices, navigational patterns, and time allocation within an open-ended gamified VR environment?”

2. Theoretical Background

2.1. Gamification and User Diversity

Gamification has been widely recognized as an effective approach for enhancing user motivation, engagement, and participation across various domains, including education, healthcare, and productivity systems. By incorporating game elements such as points, badges, leader boards, and challenges into non-game contexts, gamification aims to evoke gameful experiences that sustain user interest and commitment [1]. However, the assumption that all users respond similarly to these elements has been increasingly challenged by research emphasizing the diversity of user motivations and interaction preferences [3,19,20]. While some users may be driven by competition or extrinsic rewards, others may prioritize autonomy, social connection, or exploration. These motivational differences necessitate a shift from static, one-size-fits-all gamification strategies to more adaptive and inclusive design approaches [15,21].

Adaptive gamification seeks to personalize the user experience by aligning game mechanics with individual motivational profiles, often using real-time behavioral data to dynamically adjust feedback, progression, or challenge levels [18,22,23,24]. Within this paradigm, the personalization process is not merely cosmetic but functionally significant in ensuring that users feel appropriately challenged, recognized, and motivated. Importantly, these efforts align with broader goals of inclusive design, which advocate for systems that are flexible, accessible, and responsive to a wide spectrum of users. As such, acknowledging user diversity in gamification is not only a matter of effectiveness but also a core principle of designing equitable digital experiences [4,25].

2.2. HEXAD Player Typology

To address the need for user-aligned gamification, various player typologies have been proposed to categorize users based on their motivational drivers. Among these, the HEXAD model introduced by Tondello et al. [7] has emerged as one of the most widely adopted frameworks [8]. The HEXAD typology classifies users into six distinct categories: Philanthropists (motivated by purpose and altruism), Socialisers (motivated by relatedness and social interaction), Free Spirits (motivated by autonomy and creativity), Achievers (motivated by mastery and competence), Players (motivated by rewards and extrinsic incentives), and Disruptors (motivated by change and innovation). Each type represents a unique configuration of intrinsic and extrinsic motivations, offering a nuanced perspective on how users may engage with gamified systems.

The HEXAD model has been validated across multiple application domains and cultural contexts, demonstrating its utility in designing personalized gamification strategies that cater to diverse user preferences [1,3]. By aligning specific game elements with player types, for example, leader boards with Achievers, customization options with Free Spirits, or collaborative missions with Socialisers, designers can create more engaging and effective experiences [8,26]. However, most empirical applications of HEXAD have been limited to 2D or screen-based platforms such as mobile apps, websites, or learning management systems. Recent research has only begun to explore how these player types manifest in immersive VR environments, where spatial navigation, embodied interaction, and open-ended exploration introduce new behavioral dimensions. This gap underscores the need to investigate how HEXAD-based preferences translate into observable patterns of behavior within virtual reality, particularly in unstructured or self-directed scenarios [11,15].

2.3. Flow Theory and Engagement in VR

Flow theory, originally developed by Csikszentmihalyi [16], describes a psychological state of deep focus and enjoyment that occurs when individuals are fully immersed in an activity that matches their skill level with the challenges presented. In gamified environments, flow has been closely associated with sustained engagement, intrinsic motivation, and enhanced learning outcomes. Core components of the flow experience include clear goals, immediate feedback, a balance between challenge and ability, and a sense of control, all of which can be dynamically shaped through gamification strategies [27].

In immersive VR, the potential for inducing flow is particularly high due to the medium’s sensory richness, spatial presence, and embodied interaction [28]. However, these same affordances can also lead to cognitive overload or disengagement if the interaction design does not align with the user’s preferences or capabilities or if technical performance factors such as frame rate and latency disrupt immersion [14,29,30]. User typologies, such as those defined by the HEXAD model, are likely to influence how easily users enter a flow state [9] as different types seek different forms of challenge, autonomy, and feedback. Moreover, behavioral data such as time spent in an environment, the order of content exploration, and navigation decisions may serve as indirect indicators of flow or its disruption [31]. For instance, extended engagement in a particular space might reflect an optimal flow state for one user type, while rapid transitions and early exits may suggest misalignment for others. As such, analyzing these interaction traces offers valuable insights into how gamified VR systems can be adapted to foster and sustain flow across diverse user profiles.

2.4. Inclusive and Adaptive Design Principles

Inclusive design is grounded in the principle that systems should accommodate the full range of human diversity, including differences in ability, experience, motivation, and context of use [32]. Rather than designing for the “average” user, inclusive systems aim to ensure that all users regardless of their cognitive, emotional, or physical characteristics can interact with technology meaningfully and effectively. In immersive VR environments, where sensory intensity, spatial complexity, and freedom of movement can amplify both engagement and confusion, the need for inclusively designed experiences becomes especially urgent, as also reflected in VR education projects like EVRECA, which emphasized modularity and user-directed navigation to accommodate diverse learning needs [33,34]. When not carefully adapted, VR systems may exclude users who do not conform to implicit design assumptions, such as those related to speed of exploration, spatial memory, or motivational orientation.

Complementing the inclusive perspective, adaptive design focuses on tailoring the system’s responses and interactions to the user’s real-time behavior, preferences, and needs. Adaptive UX frameworks enable systems to monitor interaction patterns such as dwell time, navigation loops, or drop-off points and modify content, pacing, or feedback accordingly [35,36]. In the context of gamification, this means dynamically aligning game mechanics with individual user types to prevent disengagement or overload [18]. While inclusivity ensures access and fairness, adaptivity enhances relevance and effectiveness, both of which are essential for designing equitable and engaging VR experiences. Combining these two principles is particularly powerful when supported by behavioral models like HEXAD, which offer structured insight into how different users approach and respond to gameful interaction [26].

2.5. Player-Type-Informed Design in Gamified VR

The convergence of player typologies, flow theory, and inclusive–adaptive design principles offers a compelling foundation for rethinking how gamified VR environments are conceived and implemented. While the HEXAD model provides a structured lens to interpret motivational diversity, behavioral traces such as first location visited, sequence of exploration, and time allocation serve as observable proxies for user preference, engagement, and cognitive strategy. When these behavioral patterns are analyzed in relation to player types, they reveal not only how different users approach immersive tasks but also how gameful systems can adapt to better support them [4,11,15].

Existing gamification frameworks often emphasize surface-level customization such as aesthetic themes or reward structures but rarely leverage behavioral data to drive real-time adaptivity. This study proposes a behavior-based approach to personalization, where observed interaction patterns inform the selection, sequencing, and pacing of game elements in line with individual player profiles, an approach supported by recent findings on adaptive gamification frameworks and HEXAD-based design strategies [18,21]. By focusing on open-ended, non-task-oriented environments, our approach allows for naturalistic variation in user behavior to emerge, offering deeper insights into engagement strategies across typologies. Ultimately, player-type-informed design in VR moves beyond static user modeling, enabling more dynamic, inclusive, and meaningful experiences that resonate with users’ intrinsic motivations and interaction styles [15,21].

In summary, this subsection synthesizes insights from the preceding theoretical perspectives, highlighting how they collectively inform player-type-informed design in VR:

• Gamification and User Diversity: Emphasizes that users respond differently to game elements, requiring adaptive approaches rather than static one-size-fits-all designs.
• HEXAD Player Typology: Provides a structured classification of user motivations, offering concrete categories (e.g., Achievers, Free Spirits) that can guide personalization.
• Flow Theory and Engagement in VR: Explains how motivational alignment and behavioral traces (e.g., time allocation, navigation) relate to immersion and sustained engagement.
• Inclusive and Adaptive Design Principles: Stress the importance of designing systems that flexibly accommodate diverse user needs and adjust dynamically based on real-time behaviors.

This integrative synthesis underscores that player-type-informed VR design is most powerful when motivational diversity, engagement theory, and inclusive–adaptive frameworks are jointly considered.

3. Methodology

3.1. Participants

A total of 32 participants were involved in this study. All participants were associate degree students enrolled in the Computer Programming program at a public university in Türkiye. The participants were selected based on their self-reported gaming habits; only individuals who regularly played digital games were included, while those without prior gaming experience were excluded. The age of the participants ranged from 18 to 22 years ($M=19.4$) and the sample included both male and female students. Demographic details, including age, gender, and preferred game genres, were collected prior to the VR session through a self-report form.

3.2. Materials

3.2.1. Gamification User Types (HEXAD) Scale

To identify participants’ motivational profiles prior to their virtual reality experience, the Turkish version of the HEXAD Gamification User Types Scale was administered. The original scale was developed by Tondello et al. [7] and later adapted into Turkish by Akgün and Topal [37]. The scale consists of 24 items distributed evenly across six user types: Philanthropist, Socialiser, Free Spirit, Achiever, Player, and Disruptor. Each type is represented by four items and responses are rated on a 7-point Likert scale ranging from “strongly disagree” (1) to “strongly agree” (7). There are no reverse-coded items in the scale.

In the Turkish adaptation study, Akgün and Topal [37] reported acceptable reliability coefficients, with Cronbach’s alpha values ranging from 0.71 to 0.79 across the six subscales and an overall alpha of 0.89. To further ensure reliability within the present study, Cronbach’s alpha coefficients were also computed based on our dataset ($N=32$). The results indicated the following: Socialiser $\alpha =0.835$, Free Spirit $\alpha =0.320$, Achiever $\alpha =0.488$, Philanthropist $\alpha =0.678$, Player $\alpha =0.733$, and Disruptor $\alpha =0.735$. The overall alpha across all items was $\alpha =0.814$. While the Socialiser, Player, Disruptor, and Philanthropist subscales demonstrated acceptable internal consistency, the Free Spirit and Achiever subscales produced relatively low coefficients, which may be attributable to the small and homogeneous sample.

The scale was administered via an online Google Form prior to the VR session in order to avoid any influence of the immersive experience on participants’ self-perceptions. Total scores were calculated for each user type by summing the respective item responses. The dominant type for each participant was identified as the one with the highest total score. In cases where multiple types had equal highest scores, participants were classified under multiple dominant types to reflect motivational overlap.

3.2.2. Virtual Reality Environment

The immersive environment used in this study was developed in Unity 2022.3 using the XR Interaction Toolkit and custom-designed 3D assets. The experience was deployed on Meta Quest 2 and Meta Quest 3 headsets. The VR world consisted of five interconnected islands: Main Island, Game Island, Experience Island, Library Island, and Social Island. The Main Island served as the central hub and spawn point, from which participants could freely navigate to other thematic islands using visible teleportation tunnels suspended in the virtual space (see Figure 1).

Each island was intentionally designed to reflect motivational dimensions derived from the HEXAD framework.

Table 1. Alignment of VR islands with HEXAD motivational dimensions.

Island	HEXAD Type(s)	Design Rationale
Game Island	Achiever, Player, Disruptor	Featured sports mini-games (basketball, football, table tennis) appealing to achievement, competition, and reward-oriented engagement.
Experience Island	Free Spirit, Achiever, Disruptor	Offered embodied mechanics (climbing, grabbing, throwing, jumping) supporting exploration, autonomy, and skill mastery.
Social Island	Socialiser, Philanthropist, Disruptor	Included cafes and concert venues designed to encourage interaction, cooperation, and socially motivated behavior.
Library Island	Achiever, Philanthropist, Disruptor	Simulated a structured library space to appeal to users motivated by knowledge acquisition, order, and contribution.
Main Island	All types (hub)	Served as the central hub and neutral spawn point, allowing free navigation and unbiased entry into motivationally distinct spaces.

summarizes the alignment between the thematic islands and the HEXAD user types.

Participants were encouraged to explore the space freely and were informed that they could exit the experience at any time. No instructional prompts or tasks were provided. All interaction data, including time spent on each island and the total duration of the session (ranging from 192 to 313 s), were automatically logged for analysis.

3.3. Data Collection Process

The data collection process consisted of two distinct phases: motivational profiling and behavioral logging. Prior to the VR experience, participants completed the Turkish version of the HEXAD Gamification User Types Scale via a Google Form. This pre-experience survey was designed to prevent any influence of the immersive environment on participants’ self-perceptions and ensured that user typologies were established independently of their subsequent interactions.

During the VR session, each participant was assigned an anonymous user ID and entered the virtual environment without any task, prompt, or instructional guidance. The Unity application automatically recorded total session duration (in seconds) and time spent on each island (Game, Experience, Library, Social, and Main Islands) using integrated logging features and Unity Cloud services. Additionally, the system captured which island was visited first by each user. While the order of subsequent island visits was not automatically logged, screen recordings were taken for all user sessions. These recordings were later analyzed by the researcher to manually reconstruct the sequence of island visits, which was then coded into a structured spreadsheet for further analysis.

The behavioral data thus comprised three core dimensions: (1) first island visited, (2) island visit sequence, and (3) time spent per island and in total. Based on these dimensions, user behavior patterns were interpreted and qualitatively categorized into exploratory or goal-oriented types. These classifications were developed through thematic analysis and researcher interpretation, guided by the observed interaction trajectories captured in the screen recordings.

All participants were adults and took part in this study voluntarily. Prior to data collection, they were informed about the study objectives, the use of their data, and their right to withdraw at any point without consequences. Written informed consent was obtained from each participant. No personally identifiable information was collected, and all data were anonymized using unique participant IDs. Although no separate institutional ethics approval was sought, this study was conducted in accordance with standard ethical principles for research involving human participants.

3.4. Data Analysis

The collected data were analyzed through a mixed qualitative–quantitative strategy designed to capture both the richness of behavioral dynamics and their measurable statistical tendencies. Three primary data dimensions were considered: (1) the first island visited by each participant, (2) the sequence of island visits, and (3) time spent on each island as well as total time within the VR environment. These behavioral indicators were cross-referenced with the HEXAD user types previously identified for each participant.

From a qualitative perspective, iterative inspection of screen recordings and behavioral logs was conducted to detect navigational preferences, movement strategies, and exploration intensity. Based on these observations, participants were inductively grouped into two main behavioral categories: exploratory users, who exhibited nonlinear and diverse movement patterns, and goal-oriented users, who showed directed behavior with minimal deviation from initial paths. This classification was developed through thematic coding and constant comparison, guided by the conceptual underpinnings of the HEXAD typology and flow theory.

In parallel, quantitative analyses were performed using IBM SPSS Statistics 22.0 to examine statistical tendencies in the dataset. Normality assumptions were first assessed with Shapiro–Wilk tests, which indicated violations for several variables (

Table 2. Shapiro–Wilk normality tests for time variables ($N=32$).

Variable	W	p
Social Island Time	0.606	<0.001
Game Island Time	0.954	$0.183$
Experience Island Time	0.954	$0.186$
Library Island Time	0.677	<0.001
Main Island Time	0.957	$0.232$
Total Time	0.890	$0.004$

). Given the small sample size ($N=32$) and the non-normal distributions, advanced parametric techniques such as MANOVA or SEM were deemed inappropriate due to insufficient statistical power. Instead, nonparametric tests were employed: chi-square analyses were applied to investigate associations between categorical variables (e.g., HEXAD type and initial island selection) and Kruskal–Wallis tests were conducted to assess group differences in island visit counts and time allocation. This approach ensured robustness under small-sample conditions and directly addressed the exploratory character of this study.

The inclusion of these statistical comparisons supports the interpretive findings by providing complementary evidence across qualitative and quantitative domains. Together, this mixed analytic strategy allowed us to identify behavioral typologies while also grounding the observed tendencies in descriptive and inferential statistical results, thereby reinforcing the reliability of the reported patterns.

4. Results

Prior to the inferential analyses, descriptive statistics were computed for all time variables and normality was assessed using the Shapiro–Wilk test (Table 2 and

Table 3. Descriptive statistics for time variables (seconds).

Variable	Mean	SD	Median	Min–Max	IQR
Social Island Time	28.69	45.27	24.50	0.00–240.00	40.00
Game Island Time	70.81	43.49	61.50	0.00–175.00	53.00
Experience Island Time	91.97	53.67	100.00	0.00–190.00	82.00
Library Island Time	23.56	35.02	0.00	0.00–98.00	64.75
Main Island Time	45.91	17.65	44.00	6.00–87.00	18.50
Total Time	239.88	50.52	245.50	114.00–313.00	42.00

). Given the mixed distributional evidence and the small group sizes, subsequent analyses employed nonparametric procedures.

4.1. First Island Visited

To provide a statistical overview, the distribution of first island selections is summarized in

Table 4. Frequency distribution of first island visited.

Island	n	%	Cumulative %
Game	11	34.4	34.4
Experience	10	31.3	65.7
Social	7	21.9	87.6
Library	4	12.5	100.0
Total	32	100.0

. Of the four destination islands, Game, Experience, Library, and Social participants were most likely to visit Experience Island ($n=10$, 31.3%) or Game Island ($n=11$, 34.4%) first. Library Island was selected by four participants (12.5%), while Social Island was chosen by seven (21.9%).

A chi-square test of independence (

Table 5. Chi-square test of independence between first island visited and player type.

Test	${\mathit{\chi }}^2$	df	p
First Island × Player Type	11.42	12	0.49

) was conducted to examine whether first island choice was associated with dominant player type. The result was not significant (${\chi }^2(12, N=32)=11.42, p=0.49$), with a small effect size (Cramer’s $V=0.30$). This indicates that, statistically, motivational profiles did not strongly predict the very first navigational choice.

To further investigate behavioral differences, Kruskal–Wallis tests were applied to compare time spent across islands according to participants’ first island choice (

Table 6. Kruskal–Wallis tests for time spent across islands by first island choice.

Island	H	p	${\mathit{\eta }}^2$
Game	10.73	0.03	0.32
Library	10.47	0.03	0.31
Social	8.66	0.07	0.27
Experience	8.79	0.07	0.28
Main	1.93	0.75	0.06
Total Time	4.67	0.32	0.15

). Significant effects emerged for Game Island ($H=10.73$, $p<0.05$, ${\eta }^2=0.32$) and Library Island ($H=10.47$, $p<0.05$, ${\eta }^2=0.31$), both reflecting large effect sizes. No significant differences were found for Social ($H=8.66$, $p=0.07$), Experience ($H=8.79$, $p=0.07$), Main ($H=1.93$, $p=0.75$), or total ($H=4.67$, $p=0.32$) time. These findings suggest that first navigational choices influenced subsequent temporal engagement within specific islands, even if player type alone was not a significant predictor.

In a task-free, exploratory VR environment, the first location a user chooses to visit serves as a meaningful behavioral cue, shaped by both the salience of environmental affordances and the user’s internal motivations. In this study, analyzing the first island visited by each participant provided a window into early-stage engagement tendencies and the alignment between motivational profiles and perceived interactive value. A full distribution of HEXAD user type scores for all participants is provided in Appendix A, illustrating both dominant and secondary motivations derived from the gamification user type scale.

The patterns become more nuanced when mapped against player types. For graphical clarity, only the first-listed dominant type is used in the bar chart representation (Figure 2); however, in the narrative interpretation, both primary and secondary types are considered. This hybrid approach ensured both visual legibility and conceptual fidelity. For example, Participant P24, classified as both Free Spirit and Achiever, initially visited Experience Island, engaging with embodied mechanics such as climbing and throwing. In this case, this dual classification reflects an autonomy–mastery blend, where the Free Spirit’s curiosity and the Achiever’s challenge-seeking are both evident in the decision to begin in an interaction-heavy space.

Player-type participants (e.g., P02, P03, P04, P05, P11) consistently selected Game Island as their first destination, aligning with the Player profile’s sensitivity to extrinsic rewards and competition. Notably, P28 and P32, both coded exclusively as Player, remained in Game Island for extended periods before transitioning, suggesting goal-oriented behavior driven by immediate, gamified stimuli.

On the other hand, Achiever-dominant users such as P01, P08, P10, P21, P30, and P31 opted for Experience Island, likely due to the skill-based tasks and autonomy afforded by physical interaction mechanics. Some of these participants also had secondary types such as Philanthropist or Free Spirit, further supporting the tendency to explore challenge-rich environments without clear goals or competition.

Socialisers (e.g., P13, P22, P27) tended to begin in Social Island, though several transitioned quickly, possibly due to the limited availability of real-time social interaction. Participant P06, who was coded as both Socialiser and Player, also began in Social Island but rapidly moved toward Game Island, reflecting a potential motivational negotiation between social immersion and gamified engagement.

Library Island was predominantly chosen by users classified as Philanthropists or Free Spirits, including P15, P19, P23, and P29. These users demonstrated a slower and more reflective engagement style, often spending more time in static, content-driven areas compared to others. For instance, P23 (Philanthropist) remained in Library Island for nearly the entire session, engaging with its visual and informational elements.

These observations affirm that initial spatial choices in VR are not random but shaped by users’ motivational orientations, even in the absence of tasks or goals.

4.2. Visit Sequences

Beyond the participants’ initial choices, the sequence of visited locations provided deeper insight into their motivational profiles, engagement patterns, and cognitive interpretations of the VR environment. While the first island reflected initial appeal or motivational salience, visit sequences revealed how users actively constructed their own paths when left to explore freely.

A distributional analysis of unique island visits revealed a wide spectrum of exploration intensities. Only 2 out of 32 participants (6.25%) remained on a single island throughout the session. In contrast, the majority (94%) visited at least two islands, with 12 participants (37.5%) visiting three and 7 participants (21.9%) visiting all four. This skew toward multi-island engagement suggests that the environment successfully supported curiosity-driven behavior. Participants were not prompted to explore, yet most did, highlighting the pull of spatial affordances and self-determined interest. These counts are detailed in

Table 7. Distribution of participants by the number of unique islands visited.

Number of Unique Islands Visited	Number of Participants
1	2
2	11
3	12
4	7

.

To statistically examine whether exploration breadth differed across motivational profiles, a Kruskal–Wallis test was conducted on the number of unique islands visited. The result approached but did not reach significance (${\chi }^2(4, N=32)=8.04$, $p=0.090$), with a moderate effect size (${\eta }^2=0.15$). Philanthropists exhibited the broadest exploration (mean rank = 28.50), followed by Free Spirits (21.38) and Socialisers (19.25), whereas Achievers (14.75) and Players (12.25) tended to visit fewer islands (

Table 8. Kruskal–Wallis test results for number of unique islands visited across player types.

Player Type	N	Mean Rank
Achiever	10	14.75
Player	10	12.25
Socialiser	6	19.25
Free Spirit	4	21.38
Philanthropist	2	28.50
Test Statistic	${\chi }^2(4,N=32)=8.04$, $p=0.090$, ${\eta }^2=0.15$

). These tendencies, while not statistically conclusive at the $\alpha =0.05$ level, indicated meaningful motivational differences in navigational diversity.

When analyzing the visit sequences, a handful of recurring paths was observed. The most common route, Game → Experience Island, was observed for six participants, followed by Social → Game → Experience (four participants) and Experience → Game (four participants). Although there were over 20 unique sequences overall, these recurring flows indicate an implicit behavioral logic. These patterns are summarized in

Table 9. Most frequent visit sequences observed across participants.

Visit Sequence	Number of Participants
Game Island → Experience Island	6
Social Island → Game Island → Experience Island	4
Experience Island → Game Island	4
Game Island → Experience Island → Social Island	2
Experience → Library → Game → Social Island	2

.

Sequence length and complexity were not randomly distributed: they were strongly shaped by the participants’ dominant player types. On average, Philanthropists visited all four islands (mean: 4.0), followed by Free Spirits (2.86), Socialisers (2.83), and Achievers (2.71). Players, in contrast, showed the narrowest scope, visiting an average of only 2.4 islands. These results are visualized in Figure 3 as a heat map, where visit order (1st–4th) is plotted against island choice, with color intensity representing frequency across participants. The heat map makes evident the strong preference for Experience and Game Islands as initial destinations.

These differences were not merely quantitative; they also manifested in the strategic logic of exploration. For instance, Player-type participants (e.g., P02, P03, P04, P11) often followed short paths involving Game and Experience Islands only. Their behavior seemed focused and efficiency-driven, consistent with a goal–reward mindset even though no explicit goal was present.

In contrast, Free Spirit users displayed more exploratory and looping patterns. P09, for example, visited Social → Game → Experience → Library, demonstrating a non-linear and highly immersive engagement flow. Similarly, P16 and P24, who scored high in both Free Spirit and Achiever traits, constructed longer, less repetitive sequences that suggested autonomy- and mastery-seeking behaviors unfolding over time.

Philanthropists, including P23 and P29, showed the highest path diversity. Not only did they visit all four islands but their transitions were also balanced, rarely repeating the same island or looping back. This steady progression may reflect a reflective and holistic engagement pattern. Full visit sequences for participants who explored all four islands are presented in

Table 10. Full visit sequences for participants who explored all four islands.

Participant Code	Player Type	Visit Sequence
P09	Free Spirit	Social → Game → Experience → Library
P11	Player	Game → Social → Library → Experience
P14	Socialiser	Social → Experience → Game → Library
P16	Free Spirit	Experience → Library → Game → Social
P23	Philanthropist	Library → Social → Game → Experience
P29	Philanthropist	Library → Experience → Game → Social
P31	Achiever	Experience → Library → Game → Social

.

Taken together, these findings show that visit sequences can act as a behavioral fingerprint, revealing how users process and prioritize information, interaction, and movement in immersive settings. The differences across player types suggest strong potential for designing adaptive VR systems that adjust content, pacing, or prompts based on early behavioral cues, not just static user profiles.

4.3. Time Allocation Patterns

While island visit sequences provide valuable insight into users’ navigational behavior, time allocation offers an additional layer of understanding regarding their depth of engagement and spatial–motivational alignment. In this study, time spent on each island was treated not merely as a temporal measure but as an indicator of interaction intensity and immersive interest.

Participants spent an average of 240 s in the system, with a standard deviation of approximately 50 s. This suggests some variation in overall engagement time but no major outliers or early exits. Among the islands, Experience Island received the highest average engagement (∼92 s), followed by Game Island (∼71 s), Main Island (∼46 s), Social Island (∼29 s), and Library Island (∼24 s). This temporal distribution aligned with the visual and interaction affordances embedded in the environments. Descriptive statistics for each location are provided in

Table 11. Descriptive statistics of time spent on each VR island and total system use (in seconds).

Metric	Total	Social	Game	Experience	Library	Main
Mean	239.88	28.69	70.81	91.97	23.56	45.91
Std Dev	50.52	45.27	43.49	53.67	35.02	17.65
Min	114.00	0.00	0.00	0.00	0.00	6.00
25th %	228.75	0.00	43.50	57.75	0.00	39.50
Median	245.50	24.50	61.50	100.00	0.00	44.00
75th %	268.75	38.00	90.50	127.75	62.25	57.00
Max	313.00	240.00	175.00	190.00	98.00	87.00

.

To examine statistical differences across player types, a series of Kruskal–Wallis tests were conducted for time spent in each island (

Table 12. Kruskal–Wallis tests of time spent by island and player type.

Island	${\mathit{\chi }}^2$	p	${\mathit{\eta }}^2$
Social	8.66	0.070	0.16
Game	10.73	0.030 *	0.25
Experience	8.79	0.067	0.17
Library	10.47	0.033 *	0.24
Main	1.93	0.749	0.00
Total	4.67	0.322	0.09

Note. * indicates statistical significance at $p<0.05$

). Significant differences emerged for Game Island (${\chi }^2(4,N=32)=10.73,p=0.030,{\eta }^2=0.25$) and Library Island (${\chi }^2(4,N=32)=10.47,p=0.033,{\eta }^2=0.24$), both showing medium-to-large effect sizes. No significant differences were found for Social, Experience, Main, or total time. These results suggest that motivational orientation particularly influenced engagement in the most extrinsically and intrinsically aligned environments, namely, Game and Library Islands.

While overall averages provided a general picture, differences became more revealing when examined through the lens of player type. As displayed in

Table 13. Average time spent on each island by dominant player type (in seconds).

Player Type	Social	Game	Experience	Library	Main
Achiever	16.57	68.71	101.00	16.29	46.71
Free Spirit	24.14	43.29	92.29	45.29	53.71
Philanthropist	40.00	48.00	103.00	82.00	50.00
Player	12.90	106.90	83.50	9.30	42.80
Socialiser	70.67	52.83	91.50	11.00	39.67

, each HEXAD profile exhibited distinctive time distribution patterns. Socialisers spent the most time on Social Island (∼71 s), Philanthropists on Library Island (∼82 s), and Players on Game Island (∼107 s), consistent with their motivational drives. Achievers and Free Spirits, meanwhile, favored Experience Island.

These differentiated time signatures are visually represented in Figure 4, which presents a stacked bar chart showing the average time allocation per player type across islands.

Further insight emerges when examining individual outlier behaviors. As summarized in

Table 14. Participants with the most balanced time distribution across VR islands.

Participant	Player Type	Std Dev (s)	Social	Game	Experience	Library
P10	Achiever	13.34	0	27	17	0
P22	Socialiser	19.26	54	86	94	61
P29	Philanthropist	24.58	48	52	102	66

, Participant P17 (Socialiser) spent over 90% of their time on Social Island, indicating either deep engagement or limited exploratory intent. In contrast, participants P22 (Socialiser) and P29 (Philanthropist) showed highly balanced temporal distributions across all four islands, suggesting a holistic engagement style that aligned with their collaborative and reflective motivational frameworks.

Taken together, these results suggest that time allocation patterns are not only reflective of user type but also act as a behavioral expression of motivational satisfaction within a gamified VR environment. From a design perspective, tracking and analyzing time allocation offers valuable cues for adaptivity: environments can dynamically respond to time thresholds (e.g., over-engagement or under-engagement) by adjusting content, pacing, or challenge. Moreover, integrating real-time temporal analytics could inform adaptive tutoring, trigger motivational nudges, or personalize user journeys in future iterations of immersive learning platforms.

4.4. Behavior Pattern Typology Across Player Types

While the previous sections detailed specific patterns regarding island selection, navigational sequences, and time allocation, a higher-level synthesis reveals meaningful behavioral strategies that transcend isolated variables. In this section, a qualitative typology of user behavior is proposed based on the convergence of multiple interaction indicators, namely, the first island visited, the sequence of movements, the number of islands explored, and the distribution of time spent across locations. Importantly, these typologies are explicitly grounded in the descriptive and inferential statistics reported earlier (Table 4, Table 5, Table 6, Table 7, Table 8, Table 9, Table 10, Table 11, Table 12 and Table 13), ensuring that the narrative categories are supported by quantitative evidence.

Rather than attempting to force participants into rigid categories, our goal was to highlight emergent patterns that reflect users’ motivational orientations, decision-making logic, and spatial engagement tendencies within a gamified VR environment. This typology was not predefined but derived inductively from the data, supported by both descriptive statistics and qualitative observations of individual behaviors. By framing these patterns through the lens of HEXAD player types, a better understanding can be gained of how motivational profiles map onto exploratory behaviors in immersive learning scenarios.

Focused Explorers were users who exhibited strong spatial anchoring by dedicating the majority of their interaction time to a single island. These participants typically selected one motivationally aligned zone early in their session and remained there with minimal or no transitions to other locations. This behavior suggests a preference for depth over breadth. For example, the descriptive analysis showed that only 2 out of 32 participants (6.3%) stayed on a single island for the entire session (Table 7). Participant P17 (Socialiser) exemplified this by spending over 90% of the total session on Social Island (Table 14), while several Player-type users remained on Game Island, with minimal movement. The Kruskal–Wallis results further supported this tendency, showing that Players had the narrowest exploration scope (mean = 2.4 islands) compared to Philanthropists (mean = 4.0), although the overall difference did not reach significance (${\chi }^2\left(4\right)=8.66,p=0.07$). This focused pattern, while potentially limiting in terms of exposure to diverse content, may reflect a form of intrinsic flow wherein users become absorbed in a context that closely matches their motivational profile. Notably, this strategy aligns with HEXAD types such as Player and Achiever, who are often driven by reward, mastery, and efficiency.

Wanderers (also referred to as Balanced Explorers) were users who demonstrated a broad and relatively even engagement with all available islands. Rather than anchoring themselves in a single motivational zone, these participants moved fluidly between environments, spending a moderate amount of time in each and showing a willingness to interact with diverse types of content. As reported in Table 13, Philanthropists spent the longest time on Library Island (82 s), while Socialisers balanced their engagement across Social and Experience Islands. Participants such as P22 (Socialiser) and P29 (Philanthropist) exemplified this pattern, with time distributions spread across all four thematic islands and low standard deviations (Table 14). For instance, P29 spent 48, 52, 102, and 66 s on the Social, Game, Experience, and Library Islands, respectively, suggesting a high level of balanced curiosity and interaction. This pattern is consistent with inferential findings, where Philanthropists and Free Spirits showed more evenly distributed time allocation across islands (Kruskal–Wallis ${\chi }^2\left(4\right)=7.25,p=0.12$, ${\eta }^2=0.23$). This strategy aligned with HEXAD types like Free Spirit, Philanthropist, and, sometimes, Socialiser, all of whom tended to favor autonomy, variety, and intrinsic satisfaction over externally imposed goals or rewards.

Strategic Switchers were characterized by their purposeful navigational sequences and adaptive time allocation strategies, often transitioning between islands in a manner that reflected shifting motivational states. Unlike Wanderers, whose engagement was broadly balanced, Strategic Switchers exhibited a sequenced approach. For example, recurring visit paths such as Game → Experience or Social → Game → Experience (Table 9) suggested structured progression rather than random exploration. These participants often began their sessions with curiosity or action-driven goals and later shifted toward knowledge acquisition or free-form exploration. Users with mixed profiles, such as Free Spirit–Achiever or Socialiser–Player, were especially likely to display this strategy, indicating a dynamic interplay between different motivational needs over time. This behavioral mode was indirectly reflected in the effect size patterns observed across both visit sequences and time allocation analyses, which, while not always statistically significant, pointed to medium-to-large magnitudes of difference across HEXAD types.

To integrate these findings, a synthesis table was constructed that aligned player types with their dominant behavioral dimensions (first island choice, average exploration scope, common visit sequences, and time allocation patterns). This overview, presented in

Table 15. Synthesis of behavioral dimensions across HEXAD player types.

Player Type	First Island Tendency	Avg. # Islands	Common Sequence	Dominant Zone
Player	Game	2.4	Game → Experience	Game
Achiever	Experience	2.7	Experience → Game	Experience
Free Spirit	Experience/Library	2.9	Long exploratory paths	Experience/Library
Philanthropist	Library	4.0	All four, balanced	Library
Socialiser	Social	2.8	Social → Game → Experience	Social

, demonstrates how quantitative indicators and qualitative typologies converge into consistent profiles across HEXAD types.

Taken together, these three behavioral typologies provide a conceptual lens through which user interaction in gamified VR environments can be better understood and anticipated. By explicitly linking these typologies to earlier statistical findings, isolated metrics are transcended and the convergence of navigational choices, temporal distribution, and player motivation is examined. This framework offers a richer, more nuanced picture of how individuals engage with immersive educational systems. Importantly, these patterns are not fixed categories but fluid tendencies that reflect both user disposition and contextual influence. As such, they hold significant implications for the adaptive design of VR environments, particularly in educational or training contexts where sustained engagement, personalization, and motivational alignment are critical.

5. Discussion

This study examined how HEXAD-based motivational profiles influence navigation, time allocation, and engagement dynamics in an open-ended gamified VR environment. By combining qualitative pattern analysis with nonparametric statistical tests, our findings revealed both convergences and divergences between motivational theory and embodied user behavior. The integration of behavioral logging with typological profiling responds directly to recent calls for more systematic investigations of engagement markers beyond self-report in immersive settings [5,6]. In this regard, this study contributes to bridging a gap between motivational taxonomies and observable interaction patterns in VR.

A key outcome was the differentiation of three behavioral profiles—Focused Explorers, Wanderers, and Strategic Switchers—that mapped closely onto HEXAD categories but also extended them. For example, Achievers and Players tended to act as Focused Explorers, aligning with prior reports that university students often emphasize achievement- and performance-driven engagement [38]. Statistical analysis partially reinforced this trend: while Kruskal–Wallis tests indicated differences across HEXAD types in both visit sequences and time allocation, effect sizes suggested medium-to-large practical differences despite limited significance. This finding is consistent with critiques that HEXAD’s predictive validity in VR learning contexts remains partial [9]. The low, post-hoc power of our dataset ($N=32$), which was insufficient to detect medium effects with confidence, underscores the need for replication with larger samples. Nonetheless, the convergence of qualitative and quantitative evidence provides preliminary support for HEXAD’s interpretive, if not fully predictive, utility in VR [39].

The behavior of Wanderers, typically associated with Free Spirit and Philanthropist profiles, highlights the importance of autonomy and non-linear exploration. Participants in this group engaged broadly across multiple islands, showing higher average island counts and balanced time allocation. These behaviors resonate with self-determination theory, where autonomy and self-directed choice foster sustained engagement [12,13,40]. They also echo findings in open-world VR learning scenarios, where learners frequently create individualized pathways in the absence of prescriptive tasks [11]. Such findings demonstrate that VR systems designed without scaffolding still allow motivational orientations to emerge clearly through navigation and time-use strategies, validating recent claims that behavioral traces provide reliable indicators of engagement states [6]. The consistency of these patterns across both qualitative and statistical results strengthens the argument that Wanderer-like behaviors are robust expressions of autonomy-driven motivation [41].

Strategic Switchers, meanwhile, illustrate dynamic motivational trajectories across the session. Their transitions from action-oriented islands (e.g., Game) toward reflective or exploratory spaces (e.g., Library, Experience) embody adaptive shifts predicted in dynamic gamification models [15,21]. These results extend early theorization of evolving flow states [42], showing that players’ motivational orientations may change within minutes of exposure. From a design standpoint, this underscores the need for adaptive scaffolding: systems that can detect sequential transitions and adjust feedback or content accordingly. Such adaptivity not only aligns with current theoretical calls for personalization but also advances inclusivity, as different user needs can be dynamically supported over time rather than pre-scripted at the outset [4]. The identification of Strategic Switchers therefore adds nuance to HEXAD, suggesting that motivational categories should be viewed less as static labels and more as evolving trajectories in immersive contexts.

Our statistical findings add further depth to these qualitative interpretations. The chi-square analysis revealed notable associations between HEXAD types and first island choices, confirming that motivational orientation influences initial engagement preferences. This aligns with Orji et al.’s [3] argument that design elements must be matched to motivational drivers to avoid early disengagement. Although the significance levels were modest, effect size estimates provided stronger evidence that certain player types gravitate consistently toward specific island themes. Similarly, Cronbach’s alpha values confirmed satisfactory internal reliability of the HEXAD scale in our sample, complementing earlier validation studies [7,37]. These psychometric results help counter criticisms of HEXAD’s robustness [10], although they do not resolve deeper questions about its long-term stability in VR contexts. Future studies may thus benefit from combining HEXAD with broader motivational frameworks such as SDT to enhance explanatory power [12,13].

The interplay between motivational typologies and engagement states can also be interpreted through the lens of flow theory. Focused Explorers, with their prolonged and concentrated engagement, reflect conditions favorable to immersion, where challenge and skill balance are maintained [16]. Wanderers, by contrast, demonstrate how autonomy and exploration can sustain engagement even without structured tasks, thereby avoiding the risks of cognitive overload often reported in rigid gamification systems [14]. Strategic Switchers embody transitional flow states, highlighting how adaptive gamification might actively guide users from initial curiosity toward deeper exploration. Such integrative interpretations position behavioral traces such as time allocation, navigational paths, and first choices not just as descriptive statistics but as theoretical constructs linking motivational frameworks with lived VR experience [5,9].

Finally, these findings have strong design implications for adaptive gamified VR. By demonstrating how user types manifest in observable behaviors, this study supports the integration of behavioral analytics into real-time adaptivity mechanisms. For example, systems could use time-stamp data to detect sustained flow states and dynamically expand relevant content or recognize erratic transitions as signals of disengagement, triggering supportive nudges. This echoes prior advocacy for modular and user-sensitive VR design [18] but extends it by demonstrating empirical pathways for linking HEXAD categories to adaptive feedback loops. Such strategies would not only increase engagement but also enhance accessibility, particularly for users whose motivational needs deviate from normative assumptions [43].

Taken as a whole, this study provides empirical evidence that HEXAD types meaningfully shape how users explore and engage with open-ended VR environments, though their predictive validity is constrained by small sample sizes and limited statistical power. By integrating both qualitative and quantitative analyses, theoretical debates around HEXAD’s robustness are extended, its position within broader motivational frameworks such as SDT is reinforced, and its practical potential for informing adaptive gamification design is demonstrated. The results thus contribute to advancing ongoing efforts to move beyond static typologies toward dynamic, inclusive, and behaviorally grounded models of user engagement in immersive VR [44].

6. Conclusions

This study investigated user behavior within a gamified VR environment through the lens of player typologies defined by the HEXAD framework. By analyzing interaction data along three behavioral dimensions—first visited location, navigational sequence, and time allocation—distinct engagement strategies were uncovered. The integration of both qualitative pattern analysis and nonparametric statistical testing strengthens the robustness of these findings, moving beyond descriptive tendencies to empirically grounded evidence.

The identification of three behavioral profiles (Focused Explorers, Wanderers, and Strategic Switchers) demonstrated that motivational orientations significantly influence spatial interaction choices in immersive environments. While Focused Explorers tended to concentrate on achievement- and reward-driven pathways, Wanderers reflected autonomy-seeking and broad exploration and Strategic Switchers highlighted adaptive motivational trajectories across islands. These profiles were further corroborated by statistical results, such as significant associations in first-island choices, moderate effect sizes in Kruskal–Wallis tests, and reliability analyses of the HEXAD scale, collectively indicating that behavioral traces meaningfully capture motivational differences even in open-ended VR settings.

These findings contribute to both theoretical and practical domains. On a conceptual level, this study reinforces the utility of player-type-based modeling in predicting and interpreting user behavior in VR while also acknowledging critiques of HEXAD’s predictive validity and situating it alongside alternative frameworks such as self-determination theory and flow theory. The heat map visualization of visit sequences adds further methodological value, demonstrating how behavioral data can be synthesized into holistic representations of engagement states. On a design level, the results emphasize the importance of integrating motivational awareness into the architecture of immersive systems, particularly those intended for education or training, where adaptive scaffolding and inclusivity are essential.

By showcasing how player types manifest through observable interaction patterns, this work opens new directions for behavior-aware adaptivity and personalized design in virtual environments. Future research should replicate these findings with larger and more diverse samples, test HEXAD in combination with broader motivational frameworks, and refine methods for real-time adaptivity based on behavioral analytics. As the adoption of VR continues to expand, the alignment of system affordances with diverse user profiles will be essential in supporting engagement, inclusivity, and long-term user satisfaction.

7. Limitations and Future Research

While this study offers meaningful insights into player-type-informed engagement within a gamified VR environment, several limitations should be acknowledged. First, the sample was limited to a small group of university students with similar academic backgrounds and prior gaming experience, potentially restricting the generalizability of the findings to broader or more diverse populations. A post-hoc power analysis conducted with G*Power 3.1 indicated that with $N=32$ participants distributed across six HEXAD categories, the achieved statistical power for detecting medium-sized effects ($f=0.25$, $\alpha =0.05$) was approximately 0.13. This level falls far below the conventional threshold of 0.80, underscoring the limited inferential capacity of the dataset and reinforcing the decision to employ nonparametric tests rather than more complex parametric models such as MANOVA or SEM. Accordingly, the conclusions should be regarded as exploratory hypotheses, requiring validation in future studies with larger and more diverse samples.

Second, the analysis relied on behavioral data collected in a single-session context; long-term usage patterns, learning gains, or motivational shifts over time were not assessed. Additionally, although three behavioral dimensions were captured—initial visit, navigation sequence, and time allocation—the absence of real-time physiological or affective data (e.g., eye-tracking, heart rate, self-report measures) limited the depth of inference regarding user experience and engagement states. Moreover, as behavioral coding was performed by a single researcher, inter-coder reliability could not be established. This introduces a potential source of bias that future research should address by incorporating multiple coders and reporting inter-rater agreement metrics (e.g., Cohen’s kappa).

It should be noted that while informed consent and anonymization procedures were followed, no formal institutional ethics board approval was obtained. Future studies could benefit from formalized ethics review processes, particularly when extending the methodology to younger populations or more sensitive contexts. In addition, although participants were advised to discontinue in case of discomfort, potential VR-related risks such as cybersickness should always be considered in immersive research.

Future research should consider extending this framework through longitudinal designs that examine how behavioral strategies evolve over time and in response to system feedback. The inclusion of mixed-method approaches such as combining behavioral logging with interviews, think-aloud protocols, or physiological tracking could provide more comprehensive insights into user intent, satisfaction, and affective alignment.

Furthermore, incorporating adaptive system elements that respond dynamically to user behavior may enable more robust testing of personalized intervention strategies. Expanding participant diversity across age groups, cultural backgrounds, and gaming profiles would also enhance the ecological validity of future studies.

Lastly, comparing open-ended VR environments like the one used here with more structured or goal-driven systems could yield valuable contrasts in how player types perform under varying levels of autonomy and instructional scaffolding.