A Study of the Key Factors Influencing Young Users’ Continued Use of the Digital Twin-Enhanced Metaverse Museum

Abstract

This research investigates the key factors influencing young users’ continuous use of digital twin-enhanced metaverse museums. Attracting young users to use the metaverse museum for a more extended period consistently contributes to increasing the frequency of visits and content usage and promoting its sustainable development and innovation. However, there is a lack of research on the key factors influencing young users’ continuous use of digital twin-enhanced metaverse museums, which makes the theoretical basis for the in-depth design of user motivation for metaverse museums insufficient. This study constructed a model covering four dimensions—hedonic, utilitarian, social, and technological—based on communication’s uses and gratification theory (UGT). It was validated in the Metaverse Digital Twin Museum (MDTM). Using Spatial.io’s IES Goya Museum as the experimental platform, the research team conducted Partial Least Squares Structural Equation Modeling (PLS-SEM) using SmartPLS 4.0 software through experiments and surveys with 307 participants aged 18 to 30. Quantitative analyses revealed that MDTM’s hedonic gratification (hope), utilitarian gratification (information and self-presentation), social gratification (social interaction and social presence), and technological gratification (immersion) significantly influenced young users’ continued intention. The findings reveal that these six key factors can be the focus of MDTM’s future development to enhance user experience. This study fills the gap in applying UGT in the field of metaverse museums, provides metaverse museum managers with references to the key factors that can prolong users’ continued intention to use, and points out the key factors that need further attention in future research and practice.

1. Introduction

In recent years, leading companies such as NVIDIA, Meta, Tencent Holdings, and Roblox have been investing more in metaverse platforms, and the global metaverse market size is expected to grow from USD 416.02 billion in 2023 to USD 3409.29 billion in 2027 [1]. At the same time, museums are actively exploring metaverse technologies to enhance visitor experiences and expand their influence. For example, the Metropolitan Museum of Art launched the Met Unframed project to provide online virtual tours through AR technology and digital twins [2]; the British Museum partnered with LaCollection to launch non-fungible tokens (NFTs) to allow global players to visit its collections [3]; the Louvre Museum, the National Palace Museum, the National Central Museum of Korea, and the Tokyo National Museum have launched virtual tours and virtual reality projects using 3D scanning and other technologies [4,5,6,7]; Google’s Arts & Culture platform has partnered with museums around the world to create virtual museums and galleries [8]. According to the United Nations Educational, Scientific and Cultural Organization (UNESCO), the total number of museums worldwide as of March 2021 is about 104,000 [9]. Such a large number of museums provides a vast space and great potential for developing metaverse museums.

Young users are the leading user group of metaverse technologies [10]. The entry of museums into the metaverse enables the breaking down of access boundaries. It allows for a more innovative expression of cultural heritage that aligns more with the needs and aspirations of young people [11]. The two representative platforms of the metaverse, Roblox and Zepeto, are both primarily aimed at young users [12,13], with Roblox having created 108 museum projects by 2024 and Zepeto having created 71 museum projects. However, despite the high interest of young users in metaverse museums [14,15], the intention to continue engagement is insufficient [16]. The main problems are as follows: First, the digital version differs significantly from the original artifacts. It is difficult to give users a sense of immersion due to technical limitations that result in a poor experience of contact with the artifacts [17]. Second, excessive replication of the physical space and insufficient variety of experiences are possible in the virtual space [18]. Third, the metaverse museum narrative content fails to keep up with the times, and the timeliness of the information obtained by the users is insufficient [19]. Fourth, operational difficulties are encountered when using VR experiences, such as users’ tendency to touch menus or get too close to virtual objects mistakenly [20]. These issues create barriers to young users’ exploration of metaverse museums and prevent young users’ interest from being positively transformed into sustained engagement.

This study considers the Metaverse Digital Twin Museum (MDTM) as one of the directions to enhance young users’ experience of real artifacts, spatial experience, timeliness of information, and technological interaction. Digital twin technology consists of a physical twin (a real-world entity), a digital twin (a digital representation of the physical twin), and a linking mechanism between the two [21]. A physical twin refers to a physical entity that exists in the real world, and a digital twin is a digital replica used to simulate and analyze physical entities [22]. This study defines the MDTM as a museum platform for realistic real-time digital simulations in a virtual experience space, which allows users to experience and engage with digital twins of cultural heritage in the form of digital avatars and allows for digital interactions with other users [23,24,25,26]. The MDTM operates as follows: a. By high-precision replication or simulation technology to transform real museum artifacts into digital twins. b. Using the metaverse platform to achieve instantaneous and automated data streaming in the museum’s virtual space in both directions. c. Allowing users to carry out real-time manipulation behaviors to transform the digital body.

The International Council of Museums (ICOM) emphasizes that sustained user engagement enables museums to identify and address inequalities in cultural representation, accessibility, educational resource allocation, economic barriers, and community participation while fulfilling their mission to serve society [27]. Some studies have conducted preliminary explorations of ongoing user engagement in metaverse museums; for example, Alba Alabau’s team examined the potential of multi-layer animation and advanced shader technologies within the metaverse platform to enhance user engagement [20]; Chen’s study validated hypotheses related to users’ willingness to revisit the site from a technological and social perspective and highlighted the impact of this behavior on the museum’s economic performance and the conservation and sustainable development of heritage sites [28]; Hyeseung’s study proposes components for the four user experience themes of structure, interactivity, mediation, and interpretation through qualitative experiments with Peaceful Hill, hosted by the National Museum of Korea on the Zepeto meta-universe platform [4]; and Yang’s study proposes enhancing the metaverse’s cultural narrative by boosting its capabilities to improve the user’s interactive experience, thereby increasing engagement [29]. Previous research on improving the experience of young users in metaverse museums remains limited. Therefore, there is much potential for research on the key factors contributing to young users’ continued use of the MDTM. Filling this research gap is necessary for the future development of metaverse museums.

This research team aims to identify and analyze the key factors influencing young users’ continued use of the MDTM and to provide practical recommendations for the design and operation of metaverse museums to facilitate their attraction and retention of young users. In order to achieve these goals, this study explores how to increase users’ willingness to continue using the MDTM from multiple dimensions. Specifically, we set three research questions: (1) What are the key factors that influence users’ continued use of the MDTM? (2) What are the effects of these factors on users’ persistence intentions? (3) Do these factors guide the direction of MDTM development and improvement? The above questions are examined to provide theoretical and practical support for the continued development of the MDTM.

This study is structured as follows: Section 2 presents the theoretical background and literature review. Section 3 presents the research hypotheses. Section 4 and Section 5 outline the research methodology and results, including reliability, validity, and hypothesis testing. Section 6 discusses the findings of this study. Section 7 presents the implications, limitations, and future research directions. Section 8 provides conclusions. The research process adhered to academic standards to ensure the soundness, validity, and replicability of the research and methodology.

2. Theoretical Background and Literature Review

Uses and gratifications theory (UGT) is a user-centered theory that distinguishes Katz’s explanation of the basic assumptions of UGT from the traditional communication theory of passive user acceptance by including characteristics such as user activeness, active choice, and active use [30]. Moreover, the theory helps to explain why and how users actively choose a particular medium to satisfy a specific need [31]. UGT effectively understood users’ motivations for using media such as radio, television, and e-bulletins [32]. As researchers have applied UGT more and more extensively in new media, the theory has evolved into an effective tool for understanding users’ motivations for use and satisfaction [33]. Previous research has demonstrated that UGT directly impacts users’ intention to use [34]. Consequently, using UGT as a theoretical underpinning framework in this study allows for a better understanding and examination of users’ continued intention to use the MDTM.

In recent years, UGT has been widely used in research on social media. This study collates the literature on user motivation research conducted with UGT on different platforms after 2015 (

Table 1. Summary of related studies.

UGT Typology	Platform	Ref.
Hedonic gratification: enjoyment, fantasy, escapism Social gratification: social interaction, social presence Utilitarian gratification: achievement	Social network game	[35]
Hedonic gratification: enjoyment, escapism Social gratification: social interaction, social presence Utilitarian gratification: achievement	AR game	[36]
Technological gratification: intelligent, convenience Hedonic gratification: perceived enjoyment Social gratification: status Utilitarian gratification: achievement Content gratification: education	Liulishuo	[38]
Content gratification: information sharing, self-documentation, self-expression Technological gratification: convenience, media appeal, social presence	Microblogging	[37]
Hedonic gratification: perceived enjoyment Utilitarian gratification: information sharing Technological gratification: media appeal	WeChat	[39]
Utilitarian benefits: information gathering Symbolic benefits: social status	In-home voice assistant	[40]
Hedonic gratification: trust, enjoyment Utilitarian gratification: personalization, service quality, convenience Social gratification: sense of belonging	Online food delivery	[41]
Content gratification: information Process gratification: entertainment Social gratification: companionship, entertainment, social communication	AI smart home assistants	[42]

). Among them, Li’s study explored three aspects of hedonic gratification, social gratification, and utilitarian gratification to unfold the influencing factors of users’ intention to use online games continuously [35]; Bueno focused on the exploration of the same three types of gratification as Li, empirically examining the continued intention to use augmented reality games in terms of gamification as an example to provide theoretical and empirical evidence for attracting and retaining more players [36]. In addition, Liu’s empirical study of microblogging and Gao’s empirical study of smart mobile learning emphasized the dimension of technological gratification [37,38]. While Gan’s study added the dimension of technological gratification to hedonic, social, and utilitarian gratification, his experimental results also showed that technological gratification had the most potent effect on WeChat users’ intention to continue using the game [39].

Prior studies have provided a sufficient theoretical basis for this research to study users’ continuance intention from four gratification dimensions: hedonic gratification, utilitarian gratification, social gratification, and technological gratification. In addition, researchers in different fields have focused on users’ continuance intentions and have repeatedly confirmed that continuance intention plays a crucial role in understanding the behavioral aspects of users [43,44,45,46]. In the field of the metaverse, Oh was the first researcher to apply UGT to identify the motivation of users to use the metaverse, and his findings proved that user motivation has a significant effect on their sustained intention to use the metaverse. However, the research population of his study consisted solely of university students, which somewhat limits the generalizability of the results [47]. Jo explored the determinants of metaverse users’ intention to continue using it within the context of the tourism experience, and based on the UGT, it was confirmed that the three dimensions of gratification (utilitarian, hedonic, and symbolic) had a direct and significant impact on users’ intention to continue using the metaverse; the researcher mentioned the lack of differentiation between metaverse platform types as a limitation of their study [48]. In addition, Yu’s study used a mixed research methodology to explore multidimensional metaverse use motivations and factors influencing metaverse users’ intention to use based on UGT. The study’s results showed that all four factors (communication, new world experience, self-expression, and economic activity) were significantly related to immersion, satisfaction, intention to use, and intention to purchase. Likewise, the study was conducted for the entire metaverse platform and did not provide a targeted discussion of specific domains [49]. Oh, Jo, and Yu’s studies offered vital insights for this study. Previous studies have not explored user motivation and continued willingness from a UGT perspective for the MDTM. This study aims to investigate the key factors affecting the willingness of young users to continue using the MDTM, which is conducive to promoting the dissemination and development of metaverse museums.

3. Hypothesis Development

3.1. Hedonic Gratification

Pursuing pleasure is considered one of the primary ways to achieve happiness [50]. Self-determination theory emphasizes that hedonic motivation often drives user behavior, and researchers have confirmed that hedonic motivation positively impacts user usage intentions in mobile applications [51]. In this study, the hedonic gratification of the MDTM is mainly reflected in entertainment and hope. Entertainment is defined in this study as the pleasure and relaxation users experience when experiencing the MDTM’s realistic digital exhibits and fully immersive virtual environments. Previous research has shown that entertainment factors in information systems (ISs) strongly predict users’ intention to continue using and can attract users’ attention [35,52,53,54]. Balancing entertainment and other factors in museums has become vital [55]. Hope in this study refers to users’ expectations and desires for an excellent future experience and personal feelings in the MDTM. Hope, as a positive emotion [56], can motivate users to participate in activities with the expectation of achieving established goals [57]. The research by Ding et al. is based on the emotional evaluation theory and further confirms that users’ hopeful emotions toward ISs significantly affect their continued use intention [58]. Based on the above discussion, this study aims to empirically examine how entertainment and hope factors in the hedonic gratification dimension affect users’ continuance intention to use the MDTM. Therefore, the following hypotheses are proposed:

H1. 

Entertainment positively influences young users’ continuance intention to use the MDTM.

H2. 

Hope positively influences young users’ continuance intention to use the MDTM.

3.2. Utilitarian Gratification

Self-determination theory emphasizes that utilitarian gratification is the external motivation for user behavior, mainly to seek resources or reduce risks [59]. Previous research pointed out that utilitarian gratification is essential when experiencing and using emerging information technology (IT) [39,60]. This study determines that utilitarian gratification in the MDTM mainly manifests in information and self-presentation. In this study, information refers to how users obtain valuable knowledge and real-time information when accessing the MDTM. In the current context, research on social media and other platforms has attached great importance to the dissemination of information and confirmed several times that information factors are the key driving force in promoting users’ continued use intentions [39,61,62]. This study defines self-presentation as the extent to which users shape personalized digital avatars and personal subjective experiences and feelings in MDTM. Such self-expression mechanisms are commonly found on social media and include posting messages or emoticons to express personal emotions or states, and in virtual experiential environments, they also involve generating specific avatar images and digital identities [63,64]. In Heeseung’s study, self-presentation is a crucial factor in users’ intention to continue [49]. Based on this, this study aims to empirically test the relationship between information and self-presentation factors in the utilitarian gratification dimension and intention to use the MDTM. Therefore, the following hypotheses are proposed:

H3. 

Information positively influences young users’ continuance intention to use the MDTM.

H4. 

Self-presentation positively influences young users’ continuance intention to use the MDTM.

3.3. Social Gratification

Social gratification plays an essential role in influencing user continuance intentions [65,66,67]. The social gratification of the MDTM in this study mainly comprises social interaction and social presence. This dimension defines social interaction as the degree to which users interact and communicate with others on the MDTM platform. Social exchange theory emphasizes the importance of reciprocity [68], which encourages users to revisit and interact with other users more frequently. Previous research has shown that social interaction plays a positive role in enhancing users’ continuance intention [61,66,69,70]. Social presence refers to the degree to which users influence the impressions and behaviors of others by shaping their image on the MDTM platform. Social presence reflects the user’s psychological state of communicating and conversing with other virtual users in a virtual environment as if they were in the real world [71]. Research shows that higher levels of social presence are critical to enhancing users’ continuance intentions [35,72,73]. Additionally, previous research has demonstrated that higher levels of social presence in a metaverse environment influence users’ access intentions [66,67]. Based on these theories and research evidence, we predict that users will be more inclined to continue using an MDTM environment that provides more opportunities to connect with others. Therefore, the following hypotheses are proposed:

H5. 

Social interaction positively influences young users’ continuance intention to use the MDTM.

H6. 

Social presence positively influences young users’ continuance intention to use the MDTM.

3.4. Technological Gratification

Technological gratification has received the attention of more and more researchers in recent years [38]. With the advent of the Industry 4.0 era, emerging technologies have provided an ideal development ecosystem for many fields, among which technology has become an essential dimension of satisfaction [74,75]. The technological gratification of the MDTM in this study is mainly represented through immersion and intelligent interaction. This study defines immersion as the high degree of realism and complete immersion users experience when using the MDTM. Virtual reality technologies such as AR, VR, XR, and MR have been widely used in the museum field, and they provide engaging virtual experiential environments through multisensory interactions [76,77,78,79,80]. Previous studies have shown that a strong sense of immersion can enable users to achieve a state of total concentration, which promotes sustained intent [81,82,83]. In this study, intelligent interaction refers to the efficient operation, fast response, and personalized technological interaction experience users receive when using the MDTM. Intelligent technology creates a more efficient and intuitive user experience by optimizing the human–computer interaction (HCI) environment [84]. Prior research has demonstrated that the sense of gratification from technology in areas such as social media, brilliant feedback, and a sense of user control has a significant positive impact on users’ continued intention to use [38,39]. Given the above theoretical background and research evidence, the following hypotheses are proposed:

H7. 

Immersion positively influences young users’ continuance intention to use the MDTM.

H8. 

Intelligent interaction positively influences young users’ continuance intention to use the MDTM.

3.5. Proposed Research Model

Exploring the factors influencing users’ continued intention to use the MDTM is crucial for developing, optimizing, and promoting the platform [66,85]. Based on the UGT, this study aims to explore the factors contributing to users’ continued intentions and construct a comprehensive model accordingly (Figure 1). This model covers four main dimensions of gratification, and each dimension contains two specific factors:

• Hedonic gratification: entertainment (Ent) and hope (Hop).
• Utilitarian gratification: information (Inf) and self-presentation (SP).
• Social gratification: social interaction (SI) and social presence (SOP).
• Technological gratification: immersion (Imm) and intelligent interaction (Int).

Eight factors under the four gratification dimensions comprise the independent variables of the research model, while continuance intention is the only dependent variable in this study. This study defines continued intention to use the MDTM as the extent to which young users are willing to experience and use the MDTM again. In addition, this research model incorporates a variety of interdisciplinary theories, including motivation theory, self-determination theory, and social exchange theory, to capture different types of user gratification from the user’s perspective in order to provide comprehensive insights into users’ continued use of the MDTM.

4. Research Methods

4.1. Questionnaire Development

The research model consisted of nine variables: entertainment (Ent), hope (Hop), information (Inf), self-presentation (SP), social interaction (SI), social presence (SOP), immersion (Imm), intelligent interaction (Int), and continuance intention (CI). All variables were measured using a multi-item scale adapted from previous literature to ensure that the indicators were adapted to the research context of the MDTM. Items for Ent were modified from Hsu et al.’s study [61]. Items for Hop were modified by Ding et al.’s study [58]. Items for Inf were adapted from Choi et al.’s study [63]. Items for SP were adapted from Li et al.’s study [35]. Items for SI were adapted from Sundar et al.’s study [74]. Items for SOP were adapted from Li et al.’s study [35]. Items for Imm and Int were adapted from Yim et al.’s study [54]. Items for CI were adapted from Choi et al.’s study [63]. The list of research constructs for the research model is presented in table in Section 5.1. All items were measured on a 5-point Likert scale ranging from ‘strongly disagree (1)’ to ‘strongly agree (5)’.

Before distribution, three professional English translation researchers co-translated and proofread the questionnaire several times to minimize translation errors. In addition, three museum industry experts and two academic experts were invited to review the questionnaire to ensure the reliability and validity of its content. The English and Chinese presentations of the questionnaire were also meticulously proofread by experts to eliminate ambiguities.

We designed a survey based on participants’ single immersive experience with the MDTM platform to assess their initial reactions and intentions to use it, thereby inferring their likelihood of continued use [86,87]. The empirical case selected for this study is the IES Goya Museum on the Spatial.io platform (Figure 2). The choice of the experimental platform was informed by the unique characteristics of the MDTM, although selecting the appropriate platform was challenging due to the nascent development stage of the digital twin platform for the Goya Museum. We meticulously considered the following five aspects:

• Advanced Interaction Technology: Spatial.io offers advanced virtual reality features that support complex user interactions, including voice and text communication, which are key to studying the role of social interaction in the metaverse.
• Immersive experience: The IES Goya Museum offers a fully immersive virtual environment that allows users to explore artworks virtually, providing ideal conditions for studying the user experience.
• Social features: The platform emphasizes social interaction and supports multiple users online simultaneously, making it suitable for studying social presence and its impact in digital twin environments.
• Ease of Use: Spatial.io’s user-friendly design ensures broad accessibility and engagement, enabling experiments to appeal to a broader user base.
• Innovative Art Presentation: The IES Goya Museum’s innovative presentation provides a unique opportunity to study how art is perceived and experienced in a virtual environment.

4.2. Data Collection and Analysis

This study aims to explore young users’ gratification when using or experiencing the MDTM and how it affects their intention to continue using the MDTM. We focus specifically on four types of gratification: hedonic gratification, utilitarian gratification, social gratification, and technological gratification. To verify the validity of the research model, we conducted an online survey targeting young user groups in China between 18 and 30 years of age. The ages of 18 to 30 represent a crucial transition period from adolescence to early adulthood [88]. Individuals in this age group are highly receptive to new technologies such as the metaverse and virtual environments. They possess vital learning and adaptability skills, making them more likely to understand and utilize emerging technological platforms [89,90]. Consequently, they are ideal subjects for studying the use and experience of new technologies.

In view of the scope of the research object, this study adopted the snowball sampling method [91]. The first batch of participants were mainly students from three universities in central China. After completing the questionnaire, participants were encouraged to promote the survey to their friends and family. In order to ensure the accuracy of the study, the questionnaire set an age screening at the initial stage, and participants who were not within the age range of the study could not continue. All respondents who completed the questionnaire could participate in a lottery and randomly win red envelope rewards ranging from CNY 5 to CNY 10.

The first part of the questionnaire set up an experiential and usage session about the MDTM platform, which helped the respondents have a more intuitive understanding of the research object’s type, characteristics, and attributes before starting the questionnaire. In this part, we provided a detailed guide with access links to ensure participants could enter the IES Goya Museum. The guide explained how to use the Spatial.io platform on various devices and systems, including Windows, MacOS, iOS, and Android. The platform showcased artworks and paintings and offered interactive information points. Participants could click on the artworks to obtain background information and watch documentaries with other visitors in the Auditorium of the IES Goya. Participants were required to spend at least 15 min in the metaverse platform, exploring different sections and engaging in interactive activities. Regardless of the device used, participants could freely browse as digital avatars and interact with other users in real time. After completing the first part of the experience, participants proceeded to the second part of the questionnaire. Each participant needed to complete both the experience and the questionnaire on the same day to minimize the impact of time intervals on memory and response accuracy.

In the end, 320 responses were collected for this study. All 320 responses were carefully reviewed for content, and we removed 13 questionnaires where all answers were consistent responses. As a result, a valid sample of 307 responses was used for further analyses.

Table 2. Demographics of participants (N = 307).

Measure	Items	Frequency	Percentage (%)
Gender	Male	160	52.12
	Female	147	47.88
Age	18~25	169	55.05
	26~30	138	44.95
Education 1	Less than undergraduate	116	37.78
	Undergraduate	129	42.02
	Post-graduate	46	14.98
	Doctor	16	5.21
	Total of participants	307	100.0

¹ The measure of education includes Graduated and Currently Enrolled.

shows the demographic characteristics of the participants. In this survey group, the number of males was 160 (52.12%), and the number of females was 147 (47.88%). In terms of age group, the majority of the population was aged 18 to 25 years old, amounting to 169 people, accounting for 55.05%. On the other hand, the number of people aged 26–30 years was relatively small, 138 people, accounting for 44.95%. This is highly consistent with the young group-oriented survey objective of this study.

Among the 307 participants aged 18 to 30, 37.78% had an undergraduate degree or lower. This group may have diverse cultural backgrounds, and their acceptance of the metaverse museum platform likely depends on its user-friendliness and content appeal. Participants with a bachelor’s degree comprised 42.02% of the sample, making them the largest group. These participants likely have higher education, cultural literacy, and technological acceptance, which may contribute to a higher intention for continued use of the metaverse museum platform. Those with a master’s degree accounted for 14.98% of the participants. This group might have higher demands for specialized knowledge and cultural depth, finding the platform’s academic and professional content particularly attractive. Doctoral degree holders made up 5.21% of the sample. Although fewer in number, they likely have high expectations for the platform’s content quality and depth, with high-quality content and innovative features being particularly appealing to them.

5. Results

This study utilizes Partial Least Squares Structural Equation Modeling (PLS-SEM) to evaluate the research model and applies SmartPLS 4.0 to adequately validate the measurement and structural models. In studies related to user experience, behavior, and sustained intention within the metaverse, an increasing number of scholars prefer using this method [92,93,94]. PLS is increasingly being used in the social sciences, and the advantages of applying PLS in this study are that it prioritizes the prediction of the explanatory variables on the dependent variable and allows modeling in small-sample-size conditions, suitable for dealing with realistically collected data [95]. Thus, PLS is a powerful analytical tool ideal for this study.

5.1. Measurement Model

In the structural modeling of PLS-SEM, goodness of fit is used to measure the residuals and biases of covariates to assess the fit of the constructed model. We conducted a goodness-of-fit evaluation of the measurement model in this study to ensure the reliability and validity of the research results (

Table 3. Model fit.

Fit Index	Computed Values	Threshold Ref.
SRMR	0.045	[96,97]
NFI	0.821	[98]

). The Standardized Root Mean Square Residual (SRMR) measures the discrepancy between model predictions and observed data. It is generally considered that the SRMR value should be less than 0.08, and values closer to 0 indicate that the model reflects the data structure well [96,97]. In this measurement model, the SRMR value was 0.045, indicating a relatively good model fit. The Normed Fit Index (NFI) compares the fit of the model with the fit of the null model, with values closer to 1 indicating better model fit [98]. In this measurement model, the NFI value was 0.821, indicating a relatively good model fit.

In Partial Least Squares Structural Equation Modeling (PLS-SEM), we focus on the indicator reliability, internal consistency reliability aggregation effects of variables, and discriminant validity of measurement variables. The closer the factor loading coefficients of the measurement variables are to 1, the higher the degree of informativeness of the potential variables detected by the indicator. As shown in

Table 4. Research constructs and factor loading.

Variables	Items/Issue	Factor Loads	Ref.
Entertainment (Ent) (5 items)	Ent1 I enjoy browsing the platform and its pages.	0.815	[54,61,63]
	Ent2 This is very interesting to me.	0.805
	Ent3 It helps me relax.	0.827
	Ent4 While using it, I didn’t realize how much time passed.	0.811
	Ent5 I don’t get bored using it	0.846
Hope (Hop) (3 items)	Hop1 I believe this platform will bring me more benefits in the future.	0.850	[58]
	Hop2 I’m optimistic that it will become increasingly useful in my life.	0.879
	Hop3 I hope it will be more popular than it is today.	0.879
Information (Inf) (3 items)	Inf1 I learned a lot of information through it.	0.850	[61,63]
	Inf2 The information it presents is very helpful to me.	0.860
	Inf3 It can get information about the museum faster than anywhere else.	0.854
Self-presentation (SP) (3 items)	SP1 I want others to think I’m on trend.	0.873	[35,99]
	SP2 I want others to think I have good taste.	0.879
	SP3 I want others to think I am friendly.	0.835
Social interaction (SI) (4 items)	SI1 I can communicate, connect and interact with other people.	0.853	[61,63,74,100]
	SI2 It allows me to expand and enhance my social networking life.	0.828
	SI3 I find comfort in knowing other people’s thoughts and advice.	0.819
	SI4 It’s a platform where I can connect with real-life friends.	0.809
Social presence (SOP) (5 items)	SOP1 I feel connected to other users in that environment.	0.807	[35,101]
	SOP2 During the experience, I felt like I was in a museum.	0.795
	SOP3 In my interactions with other users, I am able to be myself and show who I really am.	0.778
	SOP4 I trust other users in the platform to help me when I need it.	0.810
	SOP5 It was as if my real location had been transferred into a virtual journey through the universe.	0.815
Immersion (Imm) (4 items)	Imm1 It helps me immerse myself in places I wouldn’t be able to experience firsthand.	0.855	[54,61,74]
	Imm2 It creates the experience of being in a remote environment.	0.841
	Imm3 It focused my attention.	0.828
	Imm4 After I stopped using it, I felt like I was back in the “real world.”	0.842
Intelligent interaction (Int) (6 items)	Int1 I expect to interact with the digital twin system.	0.789	[54]
	Int2 There are many tasks that I can perform through the digital twin platform.	0.791
	Int3 It responds well to my action requests.	0.814
	Int4 The process of interacting with it is very intelligent.	0.785
	Int5 Its simulation technology makes interaction smart.	0.778
	Int6 Its immersion technology is very smart.	0.810
Continuance intention (CI) (4 items)	CE1 I feel great when using this platform.	0.809	[35,54,61,63]
	CE2 I would like to continue using the Metaverse Museum Digital Twin platform.	0.820
	CE3 It’s worth using and experiencing this platform.	0.839
	CE4 I will recommend the platform to friends or personal social media.	0.818

and

Table 5. Factor loads and cross-loads.

	CI	Ent	Hop	Imm	Inf	Int	SI	SOP	SP
CI1	0.809	0.303	0.338	0.391	0.346	0.178	0.380	0.333	0.345
CI2	0.820	0.273	0.388	0.364	0.402	0.203	0.432	0.403	0.261
CI3	0.839	0.300	0.398	0.449	0.332	0.221	0.416	0.373	0.297
CI4	0.818	0.298	0.365	0.433	0.386	0.241	0.362	0.414	0.304
Ent1	0.301	0.815	0.363	0.331	0.299	0.357	0.304	0.332	0.246
Ent2	0.292	0.805	0.343	0.281	0.301	0.316	0.288	0.306	0.196
Ent3	0.279	0.827	0.376	0.320	0.324	0.259	0.300	0.290	0.236
Ent4	0.264	0.811	0.364	0.350	0.326	0.303	0.311	0.239	0.206
Ent5	0.324	0.846	0.390	0.364	0.312	0.345	0.329	0.334	0.209
Hop1	0.369	0.354	0.850	0.325	0.371	0.255	0.309	0.246	0.245
Hop2	0.407	0.375	0.879	0.314	0.383	0.277	0.339	0.291	0.274
Hop3	0.405	0.437	0.879	0.376	0.425	0.288	0.360	0.289	0.214
Imm1	0.430	0.359	0.323	0.855	0.364	0.282	0.387	0.282	0.269
Imm2	0.422	0.373	0.357	0.841	0.358	0.285	0.399	0.318	0.275
Imm3	0.416	0.344	0.315	0.828	0.349	0.248	0.313	0.353	0.310
Imm4	0.409	0.271	0.315	0.842	0.370	0.284	0.397	0.327	0.328
Inf1	0.348	0.270	0.378	0.318	0.850	0.221	0.361	0.198	0.222
Inf2	0.393	0.372	0.391	0.377	0.860	0.221	0.370	0.259	0.214
Inf3	0.400	0.326	0.390	0.397	0.854	0.220	0.326	0.234	0.278
Int1	0.197	0.341	0.263	0.265	0.196	0.789	0.282	0.208	0.263
Int2	0.197	0.311	0.229	0.268	0.212	0.791	0.378	0.283	0.149
Int3	0.243	0.349	0.280	0.290	0.222	0.814	0.344	0.232	0.210
Int4	0.203	0.257	0.203	0.223	0.150	0.785	0.274	0.221	0.188
Int5	0.205	0.276	0.305	0.240	0.257	0.778	0.300	0.211	0.200
Int6	0.167	0.304	0.207	0.268	0.186	0.810	0.283	0.271	0.175
SI1	0.412	0.281	0.308	0.363	0.392	0.288	0.853	0.329	0.236
SI2	0.394	0.284	0.301	0.361	0.315	0.356	0.828	0.280	0.297
SI3	0.405	0.334	0.352	0.367	0.337	0.320	0.819	0.358	0.244
SI4	0.391	0.338	0.321	0.382	0.316	0.338	0.809	0.393	0.353
SOP1	0.363	0.338	0.225	0.318	0.185	0.270	0.370	0.807	0.218
SOP2	0.334	0.277	0.280	0.314	0.257	0.175	0.311	0.795	0.185
SOP3	0.357	0.306	0.281	0.317	0.200	0.272	0.300	0.778	0.171
SOP4	0.397	0.283	0.266	0.263	0.184	0.246	0.359	0.810	0.213
SOP5	0.403	0.275	0.225	0.315	0.262	0.226	0.305	0.815	0.223
SP1	0.328	0.224	0.203	0.300	0.213	0.207	0.306	0.200	0.873
SP2	0.338	0.232	0.273	0.319	0.248	0.204	0.278	0.248	0.879
SP3	0.277	0.235	0.252	0.286	0.267	0.241	0.300	0.206	0.835

, the factor loading coefficients of the measured variables are all above the level of 0.7. The range of factor loading coefficients for the constructs is between 0.778 and 0.879.

In the analysis of the construct reliability and validity indexes of the model, Cronbach Alpha value, composite reliability (CR), and Average Extracted Variance (AVE) were mainly used for the measurement (

Table 6. Indicators of the reliability and validity of the concept.

	CA	CR (rho_a)	CR (rho_c)	AVE
CI	0.839	0.840	0.892	0.675
Ent	0.879	0.882	0.912	0.674
Hop	0.839	0.841	0.903	0.756
Imm	0.862	0.863	0.906	0.708
Inf	0.816	0.819	0.890	0.730
Int	0.884	0.888	0.911	0.632
SI	0.847	0.847	0.897	0.685
SOP	0.861	0.863	0.900	0.642
SP	0.828	0.837	0.897	0.744

). The values of Cronbach Alpha, composite reliability (rho_a), and composite reliability (rho_c) for this measurement are above 0.7, so the questionnaire items have good reliability. The Average Extracted Variance (AVE) was above 0.5, and the internal consistency of the questionnaire was good [102].

In the test of convergent validity, the commonly used validation methods include the Fornell–Larcker criterion and cross-loadings. According to the analysis results in

Table 7. Distinguishing validity analyses.

Construct	CE	Ent	Hop	Imm	Inf	Int	SI	SOP	SP
CI	0.821
Ent	0.357	0.821
Hop	0.454	0.448	0.870
Imm	0.499	0.401	0.389	0.841
Inf	0.447	0.380	0.452	0.428	0.855
Int	0.257	0.387	0.315	0.327	0.258	0.795
SI	0.484	0.374	0.387	0.445	0.412	0.392	0.828
SOP	0.464	0.368	0.317	0.380	0.271	0.297	0.411	0.801
SP	0.366	0.266	0.281	0.350	0.280	0.250	0.340	0.253	0.863

The values in bold that form a diagonal are the square roots of the AVEs of the individual constructs; off-diagonal values are the correlations between the constructs.

, it can be seen that in this test of discriminant validity, the standard correlation coefficients between the two of each dimension and the square root of the corresponding AVE value are compared. The correlation coefficients are lower than the square root of the AVE value, so the variables have good discriminant validity.

5.2. Modeling Analysis

In diagnosing the covariance of the measured variables, the covariance VIF value is usually below 10 and the variable paths are considered to be free from serious covariance problems. In this study, the PLS bootstrapping procedure of repeated sampling was used to estimate the model by iterating 5000 new samples. Therefore, none of the supplementary variables have serious covariance between them (

Table 8. Covariance diagnostics.

Paths	VIF
Ent→CI	1.511
Hop→CI	1.507
Imm→CI	1.549
Inf→CI	1.475
Int→CI	1.322
SI→CI	1.585
SOP→CI	1.349
SP→CI	1.233

).

In the analysis of the path relationship of the variables (

Table 9. Analysis of pathway relationships.

Paths	β	SD	t-Value	p-Value	Results
N = 307
H1: Ent→CI	−0.001	0.070	0.016	0.987	Unsupported
H2: Hop→CI	0.164	0.070	2.336	0.020	Supported
H3: Inf→CI	0.143	0.068	2.104	0.035	Supported
H4: SP→CI	0.113	0.047	2.399	0.016	Supported
H5: SI→CI	0.168	0.069	2.433	0.015	Supported
H6: SOP→CI	0.218	0.063	3.469	0.001	Supported
H7: Imm→CI	0.194	0.069	2.820	0.005	Supported
H8: Int→CI	−0.054	0.053	1.005	0.315	Unsupported

), the impact coefficient of H1 is −0.001 (t = 0.016, p > 0.05), so H1 does not have a significant impact, and the hypothesis of H1 is not valid. The impact coefficient of H2 is 0.164 (t = 2.336, p < 0.05), so H2 has a significant positive impact, and the hypothesis of H2 is valid. The impact coefficient of H3 is 0.143 (t = 2.104, p < 0.05); therefore, hypothesis H3 is also valid. The impact coefficient of H4 is 0.113 (t = 2.399, p < 0.05), indicating that when self-presentation improves by one unit, the corresponding continuance intention improves by 0.113 units, so hypothesis H4 is valid. The impact coefficient of H5 is 0.168 (t = 2.433, p < 0.05), and the hypothesis H5 is valid. The impact coefficient of H6 is 0.218 (t = 3.469, p < 0.05), so H6 has a significant positive impact; therefore, hypothesis H6 is established. The impact coefficient of H7 is 0.194 (t = 2.820, p < 0.05); therefore, hypothesis H7 is established. The impact coefficient of H8 is −0.054 (t = 1.005, p > 0.05); therefore, hypothesis H8 is not significant, and hypothesis H8 is not established. The results of the PLS structural model are shown in Figure 3.

5.3. Research Models

Figure 4 labels the results of the eight path coefficients and the corresponding significance levels. The findings show that young users’ continuance intention after experiencing and using the MDTM is influenced by hope, information, self-presentation, social interaction, social presence, and immersion, thus supporting H2, H3, H4, H5, H6, and H7. However, entertainment and intelligent interactions do not significantly affect young users’ continuance intention, thus not supporting H1 or H8.

6. Discussion

Based on UGT, this study explores the key factors influencing young users’ intention to continue using the MDTM. The empirical results show that the MDTM has hedonic gratification (hope), utilitarian gratification (information and self-presentation), social gratification (social interaction and social presence), and technological gratification (immersion), which significantly affected young users’ continuance intention. Surprisingly, the results did not support the hypothesis that the hedonic gratification of entertainment factors and the technological gratification of intelligent interaction influence young users’ intentions to continue using the MDTM. By analyzing the results of this empirical research, we obtained the following key findings:

First, the findings show that social presence (SOP) is one of the key factors influencing young users’ continued use of the MDTM (β = 0.218, p < 0.01). This result is consistent with the findings of Han, Li, and Ying et al. [35,103,104]. This suggests that social presence is crucial in the MDTM, and the extent to which users shape their image (avatar) in order to influence others’ impressions and behaviors on the MDTM platform affects their intention to continue using it. Social presence significantly enhances young users’ trust in the platform and improves user well-being [105,106]. Therefore, by gratifying users’ social presence needs, it not only promotes their continued use of the platform but also expands the impact on potential users through their proactive engagement, co-creation, or sharing behaviors. Furthermore, social presence not only satisfies users’ social needs but also promotes a sense of community in the virtual space [107]. These findings highlight that design solutions that enhance social presence should be prioritized when designing and improving MDTM platforms.

Second, immersion (Imm) significantly influenced young users’ intention to continue using the MDTM in this study (β = 0.194, p < 0.01). This aligns with Yim’s findings on augmented reality technology and Kim’s research on virtual reality tours [54,108]. The high degree of realism and total immersion users feel while using the MDTM affects their intention to continue using it. The salience of immersion emphasizes the role of immersive technologies in enhancing sensory cues for young users [109]. High-quality visuals, interactive elements, and engaging content are essential in MDTM environments. In addition, the quality of the technical execution, such as the clarity of the graphic rendering, the intuitive design of the user interface, and the fluidity of the interactivity, is directly linked to the intensity of immersion [110]. In the museum experience, immersive technological enhancements to storytelling and exhibition content are crucial and contribute to user immersion [111]. Based on these findings, we further infer that there may be a connection between immersion and social presence as a further research direction.

Furthermore, the findings suggest that social interaction (SI) factors positively influence young users’ continued intention to use the MDTM (β = 0.168, p < 0.05). This finding is consistent with Wu’s conclusions in previous metaverse research [66]. The extent to which users interact and communicate with others on the MDTM platform affects their intention to continue using it. Prior research has also pointed out that young users’ motivations for social interaction include not only differentiation in information exchange but also cover diverse forms of sharing and collaboration [112]. In the MDTM environment, this social interaction manifests itself in real-time communication with other users, shared experiences, collaborative problem-solving, or interactions through social networks. These forms of interaction enhance users’ sense of community and engagement and promote information sharing and problem-solving among users, thereby increasing the platform’s attractiveness and user loyalty.

In addition, the results of this study also showed that the hope (Hop) factor was positively correlated with young users’ continued intention to use the MDTM (β = 0.164, p < 0.05). This result is consistent with Ding’s findings [58], emphasizing hope’s importance in driving users’ exploratory usage behavior. That is, the level of users’ expectations and desires for a pleasant future experience, as well as their personal feelings about the MDTM, influences their willingness to continue using the MDTM. High levels of hope motivated users to use the platform more exploratively. Consciously incorporating gamification elements into the MDTM platform by designing engaging tasks and reward mechanisms can motivate users to engage more actively in the exploratory use of the platform. A study by Ahn et al. noted that this gamification design can provide users with a pleasurable experience that satisfies their hedonic needs, enhancing their continued intention to engage with the platform [113]. Hope has a strong relationship with user engagement, and users’ hopes in a museum environment may involve knowledge growth, skill enhancement, or strengthening of social connections. Researchers have also proposed fostering hopeful user emotions by incorporating gamification and other means to increase user retention [114].

This study also found that the information (Inf) and self-presentation (SP) factors significantly influenced young users’ continued intention to use the MDTM. The information factor was supported in this study (β = 0.143, p < 0.05), in line with the findings of Pang et al. [115]. This means the degree of opportunity for users to gain valuable knowledge when accessing the MDTM and real-time information affects their willingness to continue using it. In addition, information quality is proposed to be particularly important for young users. MDTMs extend the cultural heritage function of museums into the virtual space by providing relevant and timely information. Special attention needs to be paid to the high quality and relevance of the information content, including its practicality, educational value, and timely updating and accuracy, in order to promote users’ continued intention to use it. Meanwhile, as a key factor for users to shape and present their identities on social platforms, self-presentation similarly positively affects young users’ persistent willingness in this study (β = 0.113, p < 0.05). This is also evidenced by Chen et al.’s findings [116]. The extent to which users shape a personalized digital image and their personal subjective experiences and feelings in the MDTM affects their willingness to persist. Providing personalization and customization options such as personalized avatars, customizable exhibition spaces, and features that allow users to create and share content not only enables users to express their uniqueness and individuality better but also significantly increases their engagement and satisfaction.

Finally, in this study, we unexpectedly found that the entertainment (Ent) factor under the hedonic gratification dimension and the intelligent interaction (Int) factor under the technological gratification dimension did not significantly impact the continuance intention of young MDTM users as expected. This contrasts with the findings of both Ifinedo and Gao [38,53]. This finding challenges our assumptions and suggests that young users’ expectations of MDTMs may go beyond simple entertainment and intelligent interactions. They may seek more profound value encompassing educational, creative, or personalized aspects [117,118]. In addition, implementing these factors in the platform may not be at a sufficient level of technical sophistication or quality of experience to affect users’ continued intention to use it. For example, if intelligent interaction features are too complex or unintuitive, they may not deliver the expected value to users. Alternatively, the measurement tools used may not accurately capture the impact of entertainment and intelligent interactions on the user experience, or the associated measurement variables may be poorly designed. Furthermore, other factors, such as social presence or immersion, may have a far more significant impact than entertainment and intelligent interactions, resulting in a relatively weak effect of the latter. This insight provides a new theoretical foundation for future researchers. It has essential reference value, suggesting that future research could further explore the interactions between these dimensions and their combined effects on user behavior.

7. Implications and Limitations

7.1. Implications

This study makes the following contributions at the theoretical level. Firstly, based on the use and gratification theory (UGT) of communication and previous studies, this study divides the research dimensions of the MDTM into four dimensions: hedonic gratification, utilitarian gratification, social gratification, and technological gratification. It is the first study that applies UGT to the field of metaverse museums. To a certain extent, it broadens the scope of the application of UGT research and provides valuable references for scholars in the field of metaverse museums. Meanwhile, this study’s emphasis on technological gratification further adapts the model to evaluate technology-oriented platforms and services such as the metaverse, AI, and digital twins, providing insights into user behavior in these high-tech environments. Secondly, this study constructs a multidimensional model of the sustained use of MDTM by young users, which contributes to the theoretical level of understanding of the key factors that prolong the continuous intention of young users to use MDTMs. The model fills the research gap in the existing literature on influencing users’ continued intention in digital twin technology-enhanced metaverse museums. It provides a reliable theoretical basis for enhancing young users’ user experience in metaverse museums. Further, the findings of this study provide scholars with directions and a basis for further research; that is, hedonic gratification (hope), utilitarian gratification (information and self-presentation), social gratification (social interaction and social presence), and technological gratification (immersion), as the key factors significantly influencing the MDTM users’ continued intention to use the museum, are the critical breakthroughs for future theoretical research in the field of metaverse museums, and the further research on them is conducive to contributing more profound value to the theoretical study of metaverse museums. In summary, the theoretical contributions of this study not only provide a new theoretical framework for the study of metaverse museums but also promote the understanding of the factors of users’ continued usage intention in related fields, providing a solid theoretical foundation for future research and practice.

At the practical level, this study provides metaverse museum administrators with key factors that can prolong users’ continued intention to use the platform and identifies critical areas for further attention in future research and practice enhancements. Firstly, social presence was recognized as a key factor influencing young users’ persistence, highlighting the importance of enhancing social features in MDTM platforms and recommending developers design more features to promote user interactions, such as real-time interactive tools and community events, as well as personalized social recommendation systems and community-driven content creation to enhance users’ sense of belonging and social satisfaction significantly. Secondly, to positively impact younger users, immersive experiences should be provided through high-quality visual and audio design and interactive elements, such as deeply explorable virtual exhibitions and incorporating AR or VR elements. This research also suggests the need to re-evaluate product features and focus more on social presence and immersive experiences that drive ongoing user engagement rather than relying solely on entertainment or technical features. On the MDTM platform, real-time pop-ups should be implemented to enhance social presence and data analytics to predict and translate user interests into topics to increase user engagement. At the same time, enhancing the cross-platform compatibility of digital bodies and data analytics of user preferences are also key strategies to improve user experience and platform success. Together, these practical measures have increased user retention and facilitated the continuous upgrading of the platform’s technology and services, ensuring the long-term survival and success of the MDTM platform.

7.2. Limitations and Future Research Directions

This study provides new insights into understanding young users’ willingness to persist in MDTM systems by integrating UGT and user persistence multidimensional modeling. However, there are still limitations. Firstly, the sample size of this study was limited to 307 young users, which may affect the broad applicability of the results and representativeness to users of other age groups or backgrounds. Secondly, the cross-sectional data analysis method used in this study is to some extent unable to track and explain long-term changes in MDTM users’ behaviors and motivations. Moreover, although effective in revealing statistical relationships, the PLS-SEM method could not establish causality.

Concerning these limitations, future research should expand the sample size to increase the generalizability and depth of the findings. At the same time, the fact that the assumptions of entertainment and intelligent interaction were not established challenges the conventional view, and it is necessary to reassess the actual utility of these features in different technological applications in the future. In addition, a longitudinal study design should be considered in future research to understand the usage and feedback of MDTM users over time through regular return visits in order to more comprehensively assess the factors influencing the continued use of the system and to provide insights into trends in user behavior and intentions and their drivers. With these improvements, future research will enhance the theoretical foundation of the existing findings and respond more comprehensively to the needs of a diverse user community, contributing to the long-term development and success of the metaverse museums.

8. Conclusions

This study focuses on answering three core research questions. First, by measuring and testing the constructed multidimensional model of young users’ continuous use of the MDTM platform, this study reveals that social presence, immersion, social interaction, hope, information, and self-presentation are the core factors influencing young users’ continuous use of the digital twin-enhanced metaverse museum. This finding echoes the response to the call for developing metaverse museums in many countries worldwide in the 5G era. It combines various factors such as technology, user experience, social interaction, and personalized services, focusing on meeting the needs and preferences of young users, and provides essential theoretical and practical guidance for the design, operation, and promotion of metaverse museums. Secondly, this study clarifies that the key factors positively influence young users’ willingness to continue to use the museum. This insight can help museum administrators enhance young users’ experience in metaverse museums in a targeted way in their future work, which can effectively encourage young users to transform their interest in the metaverse museum into positive behaviors of continuing to use the museum. Finally, this study delves into how these key factors influence users’ ongoing intentions in the Discussion and Implications sections and distills targeted insights accordingly. This study further identifies the key influences that need to be focused on for future theoretical research and practice improvement of user experience in metaverse museums.