Multisensory Digital Heritage Spaces as Smart Environments in Sustainable Architectural Design

Abstract

In the context of sustainable architecture, buildings are no longer isolated entities but are integral components of a broader built environment that shapes and responds to human life. As part of this evolving architectural landscape, immersive digital cultural heritage spaces—such as virtual museums—are emerging as dynamic environments that contribute not only to cultural preservation but also to human well-being. This study examines how multisensory spatial design in virtual heritage environments can meet the physical, psychological, and emotional needs of users, aligning with the principles of smart, responsive architecture. A total of 325 participants experienced three immersive VR scenarios integrating different sensory inputs: visual–auditory, visual–auditory–tactile, and visual–auditory–olfactory. Through factor analyses, a three-dimensional model of user experience was identified, encompassing immersion, cultural engagement, and personalization. Structural equation modeling revealed that informational clarity significantly enhanced immersion (β = 0.617, p < 0.001), while emotional resonance was central to personalization (β = 0.571, p < 0.001). Moreover, ANOVA results indicated significant experiential differences among sensory conditions (F = 4.324, p = 0.014), with the visual–auditory modality receiving the highest user ratings. These findings demonstrate how digital cultural spaces—when designed with human sensory systems in mind—can foster emotionally rich, informative, and sustainable environments. By extending the role of architecture into the digital domain, this study offers insight into how technology, when guided by human-centered design, can create smart environments that support both ecological responsibility and enhanced human experience.

1. Introduction

In the context of the global transition towards sustainable cultural preservation and the mounting demand for immersive digital experiences, digital museums have emerged as a transformative solution for harmonizing heritage communication with ecological responsibility. The field of cultural heritage preservation and communication is undergoing unparalleled innovation and transformation [1]. These platforms transcend the limitations of physical infrastructure, allowing global audiences to virtually explore cultural artifacts while reducing the environmental footprint associated with mass tourism [2]. This evolution not only meets today’s expectations for personalized and immersive engagement but also achieves sustainable development goals by minimizing resource consumption, reducing carbon emissions from travel, and alleviating overcrowding at fragile heritage sites [3].

As a significant avenue of evolution for museums, digital museums have transcended the constraints of conventional physical structures through the integration of cutting-edge technologies, such as virtual reality (VR) and augmented reality (AR). This has enabled audiences to engage with exhibits in a virtual domain, facilitating a comprehensive exploration of the cultural nuances embedded within [4]. Especially in environments like China, digital museums are rapidly expanding as a low-carbon alternative to traditional cultural tourism [5].

Research suggests that multisensory stimulation can amplify emotional resonance and enhance memory retention, fostering stronger emotional connections and cultural identity during these experiences [6]. Within the context of immersive digital tourism, multisensory interaction plays a key role in enabling authentic emotional engagement with the cultural significance of virtually presented exhibits. This highlights the potential for digital museums to transcend their role as information repositories, instead serving as platforms that evoke emotional responses and facilitate cultural understanding [7].

Some scholars have attempted to integrate technologies, such as sound and tactile devices, into digital museums to enhance the depth of user interaction. However, these studies have primarily focused on a single sense or simple combinations and have lacked a systematic methodological framework [8]. Additionally, balancing ecological sustainability with user experience is challenging due to the high energy consumption and technical compatibility issues inherent in immersive systems [9].

Current academic practices exhibit the following gaps:

(1)

There is a lack of quantitative analysis of the specific ways in which different sensory components contribute to the formation of experiences.

(2)

There is a lack of an overall evaluation system applicable to ecological responsibility goals. Existing measurement methods rely primarily on subjective scales and interviews and are unable to scientifically measure the effects of multisensory experiences on immersion, emotions, and cognition.

(3)

There is a lack of design recommendations applicable to practice. Exploratory practices have not provided design pathways suitable for different exhibition conditions or addressed how to leverage sensory elements for experience innovation and ecological communication.

In response to shortcomings in existing research regarding methodological frameworks, evaluation systems, and the balance of ecological responsibility, this paper explores ways to enhance audience emotional responses [10] and information perception in digital museums through multisensory interactive design. Figure 1 shows the research roadmap. The specific research objectives include the following points:

• Describe the overall experience of visitors in digital museums and construct a multidimensional evaluation index system.
• Use structural equation modeling (SEM) to verify the impact of multisensory interaction on visitor experience and understand the role of emotional responses and information perception of visitors in digital museums.
• Propose design strategy recommendations based on multi-sensory experience optimization to promote the sustainable development of digital museums.

2. Literature Review and Research Hypotheses

Before proceeding to the specific literature review, it is necessary to clarify the relationship between the research questions and hypotheses proposed in this paper to enhance the clarity of the research structure and consistency of theoretical logic. The paper centers its research questions on multi-sensory interactive experiences in the context of digital museums and constructs three hypotheses based on these questions. However, to avoid fragmenting the analytical framework, the causal pathways between the various elements must be theoretically clarified. Specifically, the paper begins with technology-driven sensory stimulation as the starting point and sequentially explores its influence on visitors’ emotional states, sense of presence, and information cognition. This process constructs a theoretical model of multi-sensory interactive experiences. Each research question corresponds to key variable relationships within the model, and the corresponding hypotheses serve as empirical tests of these relationships. This approach establishes a close connection between the theoretical framework and the empirical pathways, enhancing the research’s overall coherence and conceptual cohesion.

2.1. Technology Innovations in Digital Museums and Tourism Experiences

The accelerated pace of technological advancement has had a profound impact on virtually every aspect of life, introducing a plethora of novel products, services, and experiences [11]. Digital museums represent the latest technological innovation to significantly alter the travel experience [12]. These museums employ a range of tools, including AR, VR, media technologies, and web-based platforms, to increase visitor engagement, enable richer interactions with collections, and provide customized information and tailored services [13]. Heritage museums are increasingly employing the use of 3D displays to create engaging and interactive exhibits [14]. The exhibition, entitled “Synesthesia—Crossing the senses with art—” by Akihito Okunaka also incorporates multisensory elements, combining visual, auditory, olfactory, spatial, and other environmental cues to immerse visitors in the experience [15]. The advent of multisensory digital technologies, including AR and VR, has greatly expanded the capabilities of museums, enabling experiences that transcend the limitations of traditional exhibit viewing [16]. This shift has paved the way for visitor-centered experiential journeys. Concurrently, visitor motivations are evolving, with a growing preference for experiences that are enjoyable, personalized, interactive, and destination-relevant [17]. To meet these demands, museums are undergoing a transformation from collection-focused institutions into community-centered spaces, incorporating personalized virtual tours to enhance engagement [18].

Museums serve as pivotal sites for experiential consumption and tourism, integrating elements of education, learning, recreation, and social interaction to offer visitors a multifaceted and immersive experience [19]. As the objectives of museums have become more expansive, the importance of experience design has correspondingly increased. In order to gain a deeper insight into the multifaceted experiences of museum visitors, researchers have developed a range of measurement scales designed to evaluate the multisensory aspects of museum experiences. The majority of traditional evaluation scales concentrate on specific dimensions, including entertainment, socialization, learning, aesthetics, and celebration. These scales are designed to assess the cultural, emotional, and social aspects of visitors’ experiences [20]. However, the rapid advancement of emerging technologies has transformed museum experiences, making multisensory interaction design an essential component of experience creation. Technologies such as VR and AR now allow visitors to deeply engage with exhibition content through multisensory participation [21]. Consequently, traditional evaluation dimensions are no longer sufficient for capturing the requirements of modern multisensory experiences, underscoring the need for updated evaluation frameworks that better align with these rich, sensory-driven interactions [22].

This study integrates the typical museum experience scale dimensions with the requirements of multisensory interaction design, resulting in six primary dimensions that comprehensively capture the impact of visitors’ multisensory experiences [8]. The initial dimension, entertainment, assesses the enjoyment and pleasure derived by visitors from the museum, particularly in highly interactive exhibitions. The learning dimension concentrates on visitors’ comprehension and assimilation of knowledge and cultural information, which is crucial in educational contexts. The aesthetic dimension emphasizes the sensory appeal of visual, tactile, and auditory experiences, serving as a fundamental element of multisensory design. Finally, the affective dimension gauges visitors’ emotional engagement, including responses such as empathy, pleasure, and surprise. This dimension is crucial for fostering emotional connections. The social interaction dimension focuses on visitors’ interactions with others and with exhibits and technology, reflecting the museum’s role as a social space. Finally, the immersion dimension is particularly suitable for evaluating the immersive experience of multisensory technologies, measuring whether visitors are fully immersed in the context of the exhibition and thus feel “in the moment” [20].

2.2. Immersive Travel Experiences and Multisensory Interactions

In recent years, there has been a growing interest in immersive travel experiences as a novel form of digital tourism. The incorporation of multisensory interaction is a pivotal aspect of immersive experiences, facilitating tourists’ ability to perceive and engage with virtual travel scenarios in a manner that is akin to being physically present [10]. This is achieved through the comprehensive application of multiple senses, including sight, hearing, and touch. This multisensory design not only enhances tourists’ emotional response and sense of presence but also stimulates deeper emotional resonance, thereby facilitating the development of a more profound sense of identity with the culture of the destination. Recent studies on immersive travel experiences indicate that technologies such as VR and AR can significantly enhance the emotional impact and memory retention associated with travel [23]. Researchers are investigating the integration of multisensory interaction with neuroscientific methods to measure emotional and physiological responses, with the objective of strengthening emotional connections with travel destinations [24]. These technologies not only enhance the immersive quality of tourist experiences but also enable real-time monitoring and optimization, advancing the development of smart tourism [25].

In the contemporary era, digital museums employ immersive technologies to facilitate highly interactive cultural experiences. To illustrate, The Met Unframed, a joint initiative between the Metropolitan Museum of Art and Verizon, employs AR technology to facilitate the exploration of art collections via smart devices, enhancing visual details and creating a realistic virtual display [26]. In Asia, the Story of the Forest exhibition at the National Museum of Korea employs 3D animation and omnidirectional projection to immerse visitors in a virtual forest while showcasing the evolution of natural history [27]. The exhibition, designed by TeamLab, melds digital art with traditional culture, thereby amplifying both immersion and engagement [28]. Similarly, Culturespaces in France has created the Lumières series of immersive art exhibitions at venues like L’Atelier des Lumières [29] in Paris and Théâtre des Lumières in Seoul [30]. These exhibitions employ high-definition projections and 3D sound to present renowned artworks, thereby offering dynamic and engaging experiences for visitors.

The incorporation of multisensory technologies in digital museums can facilitate a more profound and immersive experience through a skillful integration of these modalities. With regard to visual stimuli, it is of paramount importance to incorporate visual stimulation into immersive experiences. The integration of AR and VR technologies enables the creation of highly realistic environments, which can be further enhanced by incorporating other sensory experiences [31]. With regard to hearing, the efficacy of auditory perception in enhancing experiences has been substantiated in a multitude of disciplines [32]. It has been demonstrated that audio design extends beyond the mere provision of background music. Instead, it employs alterations in the direction and volume of the sound source to augment user immersion. The synchronization of sound with visuals has been demonstrated to facilitate the enhancement of emotional responses [33]. Haptic: The latest developments in haptic technology provide users with tangible physical feedback, enhancing the immersive experience. The latest generation of haptic feedback technology is capable of conveying pressure, vibration, or temperature changes to the user via a wearable device or sensor and is particularly suited for applications in virtual environments [34]. Olfactory: The sense of olfactory is distinctive in its capacity to evoke emotional and memory-related responses. New olfactory technologies facilitate the release of specific odors in real time during the experience, thereby enabling visitors to more naturally associate the exhibition theme with relevant situations. The sense of taste is unique in its capacity to elicit emotional responses and memories. New technologies are emerging that are able to release specific odors in real time during the experience, allowing visitors to more naturally associate the exhibition theme with relevant situations [35]. Technologies that facilitate taste experiences are a relatively recent addition to the domain of immersive experiences, yet they have already demonstrated considerable advancement [36]. As illustrated in Figure 2.

The creation of an immersive experience is successfully achieved through the utilization of a multisensory combination of sight, sound, and touch, which is characteristic of immersive art experiences. Recent global case studies on immersive experience projects are summarized in

Table 1. An overview of immersive experience programs in digital museums around the world.

Location	Project Name	Technology and Experience Overview	Sensory Modalities Implemented	Reference
United States	Met Unframed	The Metropolitan Museum of Art, in collaboration with Verizon, uses AR technology to allow viewers to use smart devices to browse art collections at home, enhancing visual detail and realism.	Visual and auditory	[26]
Korea	Story of the Forest	The National Museum of Korea has collaborated with teamLab to use 3D animation and omnidirectional projection technology to immerse visitors in a virtual forest and experience the evolution of natural history.	Visual and motion-interactive	[27]
France	(L’Atelier des Lumières)	Culturespaces produces immersive art exhibitions in cities such as Paris and Seoul, using high-definition projections and 3D sound effects to showcase artists’ works and provide a dynamic artistic experience.	Visual and auditory	[29]
Japan	MORI Building Digital Art Museum (teamLab Borderless)	Interactive digital installations blending physical movement with sound, light, and touch-responsive walls.	Visual, auditory, and tactile	[37]
China	Palace Museum VR Experience	Virtual reconstruction of the Forbidden City using VR headsets and ambient soundscapes; includes haptic controllers.	Visual, auditory, and tactile	[38]
Germany	Invisible Worlds (Deutsches Museum)	A multisensory science-art exhibition combining VR, scent release, and tactile interfaces to explore microscopic environments.	Visual, auditory, tactile, and olfactory	[39]

.

The implementation of multisensory elements in digital museums is not a standalone phenomenon; rather, it is an interactive combination of multiple sensory modalities that collectively enhance the overall experience of the audience. The incorporation of visual, auditory, tactile, and other sensory modalities facilitates a multidimensional experience for the audience. The combination of the visual and auditory senses creates a compelling atmosphere with dynamic imagery and surround sound, which enhances the emotional engagement and immersion of the audience. This synergy not only deepens emotional responses but also makes exhibit content more captivating and impactful. The incorporation of additional sensory modalities, such as touch and olfactory, further enriches the experience. Tactile feedback provides direct physical interaction, thereby deepening the audience’s perception of the exhibit, while olfactory cues evoke emotions and memories, fostering a natural connection with the content [40]. Multisensory interaction thus enhances emotional responses, strengthens information retention, and significantly improves the dissemination of exhibit content.

This study is based on existing research on multisensory interaction in digital museums and the sensory design commonly used in digital tourism. It focuses on three different combinations of multisensory interaction: visual–auditory, visual–auditory–tactile, and visual–auditory–olfactory. These combinations are based on the positive impact of multisensory interactions on immersive experiences and aim to further explore their role in enhancing the value of virtual experiences.

2.3. Multisensory Design Principle

The senses are not merely interdependent, but rather, they exist independently of one another while also complementing each other. The effect of multiple senses acting in concert is greater than that of a single sense. Accordingly, the synesthetic experience is of particular significance in the context of multisensory design. Hiro Ponti previously put forth the notion of “synesthetic experience”, postulating that the establishment of a synesthetic perception environment is a fundamental prerequisite for the creation of an immersive experience [41]. James Gibson [42] posited that multisensory perception activities collectively regulate human movement and behavior in the environment. This illustrates that in multisensory design, the integration of different senses is essential for optimal effect, rather than their operation in isolation [43] (Figure 3). At the material level, information is conveyed directly through the senses of sight and hearing. At the behavioral level, the interactive design arouses user habits and stimulates emotional resonance. At the spiritual level, cultural information and emotional elements are incorporated. The elements that have been meticulously restored in the virtual scene resonate with the user’s memories and past experiences. This multisensory design permeates the entire cognitive process, from the initial attraction of attention to the reconstruction of memories, and offers users an unprecedented immersive cultural experience [44].

2.4. Concepts Related to Museum Experience, Emotional Response, Information Perception, and Multisensory Cues

2.4.1. Eight-Dimensional Experience of Museums

Based on Pine and Gilmore’s [20] experience economy theory, combined with multisensory interaction and digital technology characteristics, this study proposes an eight-dimensional experience model for digital museums:

-

Entertainment experience enhances the fun and duration of participation through interactive exhibits such as AR games [13].

-

Educational experience promotes the internalization of knowledge through VR scene reconstruction and multisensory interpretation [21].

-

Esthetic experience relies on high-precision 3D modeling and light and shadow sound effects to enhance visual art perception [8].

-

Serendipity experience triggers emotional resonance through associative design [41].

-

Communities experience an enhanced sense of collective participation through multiplayer collaborative tasks and virtual community platforms [17].

-

Escapism experience relies on VR/AR to break the boundary between the real and the virtual to enhance cultural communication [16].

-

Localness experience deepens the recognition of heritage values through localized narratives [22].

-

The personalized experience dynamically recommends content based on user data to enhance engagement.

This study integrates the model with traditional theories and technological logic to provide a framework for systematic evaluation.

2.4.2. Measurement of Emotional Responses

Measuring emotion is of key importance in tourism research, with a particular focus on its role in driving stimulus exposure and tourist experience. In this study, we adopted the three-dimensional pleasure–arousal–dominance (PAD) model [45], which systematically quantifies emotional states through the three dimensions of pleasure-displeasure (e.g., happy and hopeful), arousal-nonarousal (e.g., excited and calm), and dominance-neglect (e.g., sense of control and helplessness). Among them, the pleasure dimension reflects individuals’ positive emotions (e.g., satisfaction and joy) in specific situations, the arousal dimension characterizes the degree of emotional activation (e.g., stimulation and inspiration) [46], and the dominance dimension measures the ability of emotions to regulate behavior [47]. This study captures the dynamic emotional feedback triggered by multisensory interactions with a validated measurement tool, providing empirical support for analyzing the role of emotions in immersion experiences.

2.4.3. Dimensions of Information Cognition

Information cognition is defined in this study as the depth of understanding and persistence of memory of visitors to digital museum exhibits [12], which consists of two core dimensions: cultural understanding and memory persistence. Cultural understanding focuses on visitors’ ability to interpret the historical background [48], cultural connotations, and narrative logic of the exhibits (e.g., “It is easier for me to understand the content of the museum”), and its assessment relies on optimizing information delivery through multisensory interactions (e.g., simultaneous audio-visual narration), while memory retention measures the durability of information after the experience (e.g., “I was impressed by the details of the display”), which is enhanced by contextualized memory encoding through immersive technologies (e.g., VR scene reconstruction) [49]. This study quantifies cognitive mechanisms in terms of information reception (cultural understanding) and information integration (memory retention) to provide a structured framework for analyzing the effects of multisensory design on knowledge internalization.

2.4.4. Multisensory Types and Experiences Multisensory Interaction Design

Multisensory interaction design enhances immersion and emotional engagement in digital museums by integrating visual, auditory, tactile, and olfactory sensory channels with a synergetic effect [40]. Visual–auditory combinations have been shown to significantly enhance environmental realism and emotional arousal [8]; haptic feedback deepens cognitive engagement by enhancing interactive realism and behavioral control [34]; and olfactory cues enhance cultural ambience perception by triggering emotional memory [22]. This study focuses on three types of multisensory combinations (visual–auditory, visual–auditory–tactile, and visual–auditory–olfactory), and the multisensory integration strategy provides a technical path and theoretical basis for immersive experience design.

2.5. The Assumption of Multisensory Design for Immersive Cultural Experiences

Multisensory design represents a fundamental strategy for the creation of immersive experiences. The synergistic effects of multiple senses, including sight, hearing, and touch, not only enhance users’ emotional responses but also promote the depth and breadth of information perception. In immersive cultural experiences, the interactive synesthesia of multiple senses can effectively stimulate users’ emotional resonance and reshape cultural understanding and memory with rich sensory information. In this context, this study proposes the following research framework (Figure 4) to explore the mechanisms of multisensory design in immersive cultural experiences:

Hypothesis 1 (H1). 

Emotional responses have a significant positive impact on the digital museum visiting experience.

Hypothesis 2 (H2). 

Information perception has a significant positive impact on the digital museum visit experience.

Hypothesis 3 (H3). 

Different types of multisensory combinations have a significant positive impact on the digital museum visiting experience.

Due to the significance of multisensory combinations in virtual environments and their influence on emotional responses and perception of information, this study poses the following research questions.

Research Question 1: How can multisensory interaction techniques enhance the audience immersion experience in digital museums? What are the key aspects of the audience experience in digital museums?

Research Question 2: Do the emotional response and information perception of the audience impact the immersive experience of multisensory technology in digital museums?

To ensure clarity and scientific rigor, this study pairs specific hypotheses with corresponding research questions. This approach intuitively connects the theoretical basis to the hypotheses while facilitating the orderly verification of each pair during subsequent data analysis (

Table 2. Mapping of research questions and corresponding hypotheses.

Research Question	Hypothesis	Explanation
RQ1. How can multisensory interaction techniques enhance the audience immersion experience in digital museums? What are the key aspects of the audience experience in digital museums?	H3—Different types of multisensory combinations have a significant positive impact on the digital museum visiting experience.	Examines the effect of different sensory modality combinations (e.g., visual–auditory, visual–auditory–tactile, visual–auditory–olfactory) on visitor immersion and experience.
RQ2. Do the emotional response and information perception of the audience impact the immersive experience of multisensory technology in digital museums?	H1—Emotional responses have a significant positive impact on the digital museum visiting experience.	Investigates the influence of emotional dimensions (e.g., pleasure, arousal, and dominance) on audience engagement and immersion.
	H2—Information perception has a significant positive impact on the digital museum visiting experience.	Assesses the role of cognitive aspects such as cultural understanding and memory retention in shaping immersive digital museum experiences.

).

3. Materials and Methods

3.1. Research Design and Procedures

In order to respond to the research questions that have been posed, this study has devised and implemented a scenario-based experimental study. The experimental design enables the testing of research questions and hypotheses in a controlled environment, thus facilitating the accurate capture of differences between different conditions. Moreover, experimental studies are well-suited for investigating causal relationships, which enhances the persuasiveness of the research findings. The participants engaged with the immersive project by visiting the Xi’an Metaverse Digital Center (Xi’an, China), an offline venue. Following their participation in the aforementioned scenario-based experimental study, participants were provided with an online link to access the content they had just experienced and complete the relevant questionnaire.

Figure 5 shows the research steps flowchart to further clarify the process and logical path of the research implementation.

To systematically investigate the impact of different multisensory configurations on immersive experiences, this study employed three experimental scenarios.

Table 3. Experimental conditions and sensory input differences across the three scenarios.

Scenario	Sensory Modalities	Description of Immersive Setting	Sensory Input Characteristics
Scenario 1 (Visual + Auditory)	Visual and auditory	Simulated scientific laboratory: participants see detailed large-scale equipment and hear mechanical sounds and team conversations.	Visual: High-resolution display of instruments and environment Auditory: Synchronized machine and dialogue sounds
Scenario 2 (Visual + Auditory + Tactile)	Visual, auditory, and tactile	Prehistoric animal encounter: participants see lifelike ancient beasts, hear growls, and touch the simulated texture of animal fur.	Visual: Realistic animal models Auditory: Environmental and creature sounds Tactile: Haptic feedback of fur
Scenario 3 (Visual + Auditory + Olfactory)	Visual, auditory, and olfactory	Ancient temple experience: participants see a restored temple scene, hear ambient temple sounds, and smell a faint sandalwood fragrance.	Visual: Cultural architectural visuals Auditory: Temple ambience Olfactory: Ambient incense scent

lists the combinations of sensory modalities and environmental cues used in each scenario.

Participants are randomly assigned to one of three multisensory cue scenarios. Examples are shown in Figure 6, Figure 7 and Figure 8.

3.2. Instruments

The measurement methods employed in this study were adapted from those previously described in the literature. The 11-item emotional response scale was adapted from the previous research literature [44], including the study by Russell (1980) [45], who proposed three basic dimensions of emotional states: pleasure, arousal, and dominance (PAD) [50]. In particular, the following combinations are illustrated: “annoyed-happy”, “unhappy-happy”, “dissatisfied-satisfied”, “desperate-hopeful”, “depressed-cheerful”, and “calm-excited.” The six-item information perception scale [51] includes the item “Using immersive technology when visiting a museum will facilitate my comprehension of the museum.” The eight-dimensional environmental experience encompasses educational experiences, entertainment experiences, aesthetic experiences, experiences of escape, accidental experiences, local experiences, community experiences, and personalized experiences. The scale is employed for the purpose of evaluating the digital museum audience experience. All of the aforementioned items were measured using a 5-point Likert scale. (1 = completely disagree, 5 = completely agree)

This study developed a specific sampling plan and sample size design based on the questionnaire design to ensure a scientific and reasonable data collection process.

(1)

Sampling Strategy and Rationale: A stratified random sampling method was used to recruit participants, ensuring demographic diversity in terms of age, gender, and visit time. The sample included visitors on weekdays and weekends to reduce time bias. The inclusion criteria ensured that participants had no sensory impairments and had not previously visited the same exhibition. Each participant was randomly assigned to one of three sensory scenarios to ensure internal validity.

(2)

Sample Validity Considerations: This sampling method was chosen to enhance the representativeness of the museum-visiting public and reduce self-selection bias. The sample size exceeded the minimum recommended for SEM analysis, ensuring sufficient statistical power for factor structure validation and path modeling.

To ensure the robustness of the data analysis in this study, based on the questionnaire design of 18 questions and the grouping requirements of the experimental situation, and according to the minimum requirements of factor analysis, each question needs a minimum of 5–10 samples. A total sample size of at least 90–180 is required to combine the 18 questions in the questionnaire. In order to conduct a meaningful comparison of the three multisensory scenarios (visual–auditory, visual–auditory–tactile, and visual–auditory–olfactory) within the experimental design, each group must comprise approximately 60 participants, resulting in a total sample size of 180. Furthermore, in order to guarantee the recovery rate of the questionnaires and to address any invalid responses, it is advised to supplement the sample size with a certain number of additional questionnaires. In consideration of the actual number of individuals who viewed the material, 360 questionnaires were prepared for distribution. This calculation method accounts for the complexity of the questionnaire, the experimental design, and the necessity of grouping, thereby providing effective support for the statistical requirements of factor analysis, group comparison, and regression analysis. It furnishes sufficient data for this study. Subsequently, incomplete answers and questionnaires with evident logical inconsistencies were excluded. After deleting incomplete answers and questionnaires with evident logical inconsistencies, 325 valid questionnaires were returned, representing an effective rate of 90.2%.

3.3. Participant Basic Information

A total of 325 samples were analyzed to gain insight into the demographic distribution of the research sample. Following the sorting of the questionnaire samples, it was determined that 181 males and 144 females participated in the survey, representing 55.69% and 44.31%, respectively, of the total number of respondents. This suggests that the number of male participants was slightly higher than that of female participants. With regard to age, the majority of respondents were aged 25–34, representing 36.62% of the total, followed by 105 individuals aged 18–24, accounting for 32.31%. The survey reveals distinct demographic patterns among museum attendees. The visitor base is predominantly composed of younger demographics, with academic qualifications showing significant elevation—nearly half (48.2%) of participants hold bachelor’s degrees, while an additional 15.4% possess graduate-level education or higher, indicating over 60% of respondents have attained tertiary education. Regarding visitation habits, 46.2% reported annual museum visits, complemented by 25.3% attending biannually, constituting a combined majority (71.5%) who engage with cultural institutions on an intermittent basis. This attendance pattern demonstrates that most visitors maintain periodic rather than frequent engagement with museums. The sample encompasses substantial demographic variability across gender distribution, age cohorts, educational backgrounds, and cultural consumption frequencies, ensuring representative diversity in respondent profiles.

3.4. Structural Equation Modeling (SEM) Analysis

To validate the scientific and rational nature of the questionnaire design, variable settings, and sampling methods, this study uses structural equation modeling (SEM) as its primary statistical tool. SEM combines factor and path analyses to address multiple hypotheses about causal relationships and model unobservable latent variables. This allows us to assess the structural integrity and data fit of the research model. This study selected SEM to analyze the relationships between variables while testing the validity of the data structure to ensure a reliable foundation for hypothesis testing.

3.4.1. Principles and Applicability Analysis of SEM Methods

This study used SEM to construct measurement and structural models. In the measurement model, latent variables, such as emotional responses and information cognition, are derived through multiple observable indicators. In the structural model, path relationships between latent variables are established based on theoretical assumptions in order to analyze the causal pathways of emotional states, information cognition, and experiential perceptions within contexts of multi-sensory interaction. SEM is employed to present the relationships among variables systematically under different experimental conditions and to validate the rationality of the model structure and the validity of the scales.

3.4.2. Structural Model Construction and Validation

In this study, we entered all observed variables into AMOS version 24.0 (IBM Corp., Armonk, NY, USA) to create the structural model. Emotional responses and information cognition were the independent variables, while immersive experience, cultural experience, and personalized experience were the dependent variables. We used the maximum likelihood estimation (ML) method to estimate path coefficients and calculated fit indices, such as chi-square, RMSEA, CFI, and TLI, to assess model fit. Standardized path coefficients were used to determine the direction and significance of the relationships between the variables, thereby testing the model’s validity.

3.5. Data Collection and Analysis Procedure

To address the research questions and to test the proposed hypotheses, this study used SPSS Statistics version 26.0 (IBM Corp., Armonk, NY, USA) and AMOS version 24.0 (IBM Corp., Armonk, NY, USA) for comprehensive statistical analysis. The analysis process consisted of several successive stages:

3.5.1. Descriptive Statistics Methods

Descriptive statistics include mean (M) and standard deviation (SD), which are used to summarize the central tendency and variability of the measured variables. For Likert scale items, the mean was interpreted as an indicator of the level of agreement, with higher values indicating greater agreement with the statement.

3.5.2. Reliability Analysis Methods

Cronbach’s alpha was calculated as follows:

$$\mathsf{\alpha }={\displaystyle \frac{\mathrm{N}}{\mathrm{N}-1}}\left(1-{\displaystyle \frac{{\sum }_{\mathrm{i}=1}^{\mathrm{N}}{\mathrm{V}}_{\mathrm{i}}}{{\mathrm{V}}_{\mathrm{t}}}}\right)$$

(1)

where $\mathrm{N}$ is the number of items, ${\mathrm{V}}_{\mathrm{i}}$ is the variance of item $\mathrm{i}$, and ${\mathrm{V}}_{\mathrm{t}}$ is the variance of the total score.

3.5.3. Exploratory Factor Analysis (EFA)

An EFA analysis was conducted to identify potential dimensions of the digital museum experience. Principal component analysis (PCA) and variable axis rotation were used, guided by the Kaiser–Meyer–Olkin (KMO) measure and Bartlett’s test of sphericity. Eigenvalue criteria (>1) and cumulative explained variance (>50%) determined factor extraction.

The Kaiser–Meyer–Olkin (KMO) measure was expressed as follows:

$$\mathrm{K}\mathrm{M}\mathrm{O}={\displaystyle \frac{\sum _{\mathrm{i}\ne \mathrm{j}}{\mathrm{r}}_{\mathrm{i}\mathrm{j}}^2}{\sum _{\mathrm{i}\ne \mathrm{j}}{\mathrm{r}}_{\mathrm{i}\mathrm{j}}^2+\sum _{\mathrm{i}\ne \mathrm{j}}{\mathrm{p}}_{\mathrm{i}\mathrm{j}}^2}}$$

(2)

where ${\mathrm{r}}_{\mathrm{i}\mathrm{j}}$ is the correlation coefficient between items $\mathrm{i}$ and $\mathrm{j}$, and ${\mathrm{p}}_{\mathrm{i}\mathrm{j}}$ is the corresponding partial correlation.

Bartlett’s test of sphericity was denoted as follows:

$${\mathsf{\chi }}^2=-\left(\mathrm{n}-1-{\displaystyle \frac{2\mathrm{p}+5}{6}}\right)\mathrm{l}\mathrm{n}\left|\mathrm{R}\right|$$

(3)

where $\mathrm{n}$ is the sample size, $\mathrm{p}$ is the number of variables, and $\mathrm{R}$ is the correlation matrix determinant.

Principal component analysis and varimax rotation were used to determine the basic structure of the empirical scale. The number of factors was determined based on eigenvalues greater than 1.0. The total variance explained by the extracted factors was calculated to assess how well the factor solutions captured the variability of the data.

This is the factor loading matrix calculated as follows:

$$\mathsf{\Lambda }=\mathrm{V}\cdot {\mathrm{D}}^{{\displaystyle \frac{1}{2}}}$$

(4)

where $\mathrm{V}$ is the eigenvector matrix and $\mathrm{D}$ is the eigenvalue diagonal matrix.

The Promax rotation is expressed as follows:

• First perform varimax orthogonal rotation to get the initial load matrix.
• Generate the diagonal target matrix by power transformation (usually to the power).
• Calculate the diagonal load matrix.

$${\mathsf{\Lambda }}_{\mathrm{promax}}={\mathsf{\Lambda }}_{\mathrm{varimax}}\cdot \mathrm{T}$$

(5)

3.5.4. Confirmatory Factor Analysis (CFA)

CFA analyses were performed to validate the factor structure identified by the EFA and to assess the fit of the measurement model. Several fit metrics were examined, including the model fit index and its calculation:

$$\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}\mathrm{A}=\sqrt{{\displaystyle \frac{\mathrm{m}\mathrm{a}\mathrm{x}\left({\mathsf{\chi }}^2-\mathrm{d}\mathrm{f},0\right)}{\mathrm{d}\mathrm{f}\left(\mathrm{N}-1\right)}}}$$

(6)

$$\mathrm{C}\mathrm{F}\mathrm{I}={\displaystyle \frac{{\mathsf{\chi }}_{\mathrm{n}\mathrm{u}\mathrm{l}\mathrm{l}}^2-{\mathsf{\chi }}_{\mathrm{m}\mathrm{o}\mathrm{d}\mathrm{e}\mathrm{l}}^2}{{\mathsf{\chi }}_{\mathrm{n}\mathrm{u}\mathrm{l}\mathrm{l}}^2}}$$

(7)

where ${\mathsf{\chi }}^2$ is Chi square statistic of the model, $\mathrm{d}\mathrm{f}$ is degrees of freedom of the model, and $\mathrm{N}$ is sample size.

GFI, TLI, and SRMR are as per AMOS defaults.

For composite reliability (CR) and average variance extraction (AVE):

Composite reliability:

$$\mathrm{C}\mathrm{R}={\displaystyle \frac{{\left(\sum {\mathsf{\lambda }}_{\mathrm{i}}\right)}^2}{{\left(\sum {\mathsf{\lambda }}_{\mathrm{i}}\right)}^2+\sum {\mathsf{\theta }}_{\mathrm{i}}}}$$

(8)

AVE:

$$\mathrm{A}\mathrm{V}\mathrm{E}={\displaystyle \frac{\sum {\mathsf{\lambda }}_{\mathrm{i}}^2}{\sum {\mathsf{\lambda }}_{\mathrm{i}}^2+\sum {\mathsf{\theta }}_{\mathrm{i}}}}$$

(9)

where ${\mathsf{\lambda }}_{\mathrm{i}}$ is the standardized factor loading of item $\mathrm{i}$, and ${\mathsf{\theta }}_{\mathrm{i}}$ is its error variance.

3.5.5. Analysis of Variance

A one-way analysis of variance (ANOVA) was conducted to compare the means of the three multisensory scenarios (visual–auditory, visual–auditory–tactile, and visual–auditory–smell) on the dimensions of visitor experience. The F-statistic was calculated to determine if there was a significant difference between the means of the groups. This was a one-way ANOVA and post hoc comparison:

ANOVA F-statistic:

$$\mathrm{F}={\displaystyle \frac{\mathrm{M}{\mathrm{S}}_{\mathrm{b}\mathrm{e}\mathrm{t}\mathrm{w}\mathrm{e}\mathrm{e}\mathrm{n}}}{\mathrm{M}{\mathrm{S}}_{\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n}}}}={\displaystyle \frac{\frac{\mathrm{S}{\mathrm{S}}_{\mathrm{b}\mathrm{e}\mathrm{t}\mathrm{w}\mathrm{e}\mathrm{e}\mathrm{n}}}{\mathrm{d}{\mathrm{f}}_{\mathrm{b}\mathrm{e}\mathrm{t}\mathrm{w}\mathrm{e}\mathrm{e}\mathrm{n}}}}{\frac{\mathrm{S}{\mathrm{S}}_{\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n}}}{\mathrm{d}{\mathrm{f}}_{\mathrm{w}\mathrm{i}\mathrm{t}\mathrm{h}\mathrm{i}\mathrm{n}}}}}$$

(10)

Dunn’s post hoc test (rank-based):

$${\mathrm{Z}}_{\mathrm{i}\mathrm{j}}={\displaystyle \frac{{\mathrm{R}}_{\mathrm{i}}-{\mathrm{R}}_{\mathrm{j}}}{\sqrt{\frac{\mathrm{N}\left(\mathrm{N}+1\right)}{12}\left(\frac{1}{{\mathrm{n}}_{\mathrm{i}}}+\frac{1}{{\mathrm{n}}_{\mathrm{j}}}\right)}}}$$

(11)

where ${\mathrm{R}}_{\mathrm{i}}$ and ${\mathrm{R}}_{\mathrm{j}}$ are the mean ranks for groups $\mathrm{i}$ and $\mathrm{j}$, ${\mathrm{n}}_{\mathrm{i}}$, ${\mathrm{n}}_{\mathrm{j}}$ are their sample sizes while $\mathrm{N}$ is the total sample size.

This integrated statistical approach provides an in-depth study of the multi-sensory dimensions of the digital museum experience and provides a solid foundation for addressing the research questions and testing the proposed hypotheses.

4. Results

4.1. Digital Museum Visitor Experience

4.1.1. Reliability Analysis

Reliability analyses were conducted to examine the reliable accuracy of responses to quantitative data (especially questions on the attitude scale). The 24 items adapted from the eight-dimensional experience scale [52] were used to assess the digital museum visitor experience, and for the experience scale, the reliability coefficient value was 0.869, which is greater than 0.8, indicating that the reliability of the research data is of high quality. For the “CITC value”, the CITC value of the analyzed items is basically greater than 0.4, which indicates that there is a good correlation between the analyzed items and also indicates a good level of reliability (

Table 4. Cronbach’s reliability analysis.

Items	CITC	Cronbach’s α Coefficient
1. I gained new knowledge from my visit to the digital museum.	0.413	0.869
2. This experience broadened my understanding.	0.398
3. It aroused my interest in exploring unfamiliar topics.	0.447
4. I found the digital museum visit to be highly engaging.	0.404
5. The exhibits in the digital museum demonstrated meticulous attention to detail.	0.478
6. I thoroughly enjoyed my time at the digital museum.	0.429
7. The digital museum visit was entertaining.	0.466
8. I was deeply engrossed in the digital museum experience.	0.425
9. The digital museum provided a delightful experience.	0.436
10. I felt as though I assumed a different identity during the visit.	0.409
11. I experienced a sense of being transported to another era or place.	0.467
12. I felt entirely disconnected from the real world.	0.426
13. The digital museum visit gave me a sense of spontaneous excitement.	0.400
14. I unexpectedly engaged in experiences I had never imagined before.	0.511
15. I encountered surprising and delightful moments while visiting.	0.341
16. The digital museum enabled interaction with local individuals.	0.443
17. Through the displays, I could immerse myself in the local culture.	0.464
18. I gained insight into the daily lives of local people.	0.450
19. The experience fostered connections that turned strangers into acquaintances.	0.469
20. I felt a strong sense of belonging to the location I was visiting.	0.368
21. I felt included in a unique travel-oriented community during the visit.	0.476
22. The experience felt customized to suit my preferences.	0.483
23. The digital museum was adapted to meet my specific interests.	0.348
24. The interaction made me feel personally valued as a visitor.	0.436

).

4.1.2. Validity Analysis

As shown in

Table 5. Validity analysis.

KMO and Bartlett’s Tests
KMO		0.930
Bartlett’s sphericity test	Approximate chi-square test	1310.195
	df	276
	p-value	0.000

, the KMO value is 0.930, which exceeds the threshold of 0.6. This indicates that the data can effectively extract information. Bartlett’s sphericity test shows that the chi-square value is approximately 1310.195 with 276 degrees of freedom (p < 0.001), which confirms the suitability of factor analysis.

4.1.3. Principal Component Analysis Results

A principal component analysis yielded a total of three principal components, with all eigenvalues exceeding 1. These three principal components were designated as immersion, culture, and personalization. The variance explained by these three principal components was 46.972%, 4.903%, and 4.289%, respectively, with a cumulative variance explained of 56.163%. Furthermore, the analysis yielded a total of three principal components, with their corresponding weighted explained variances equaling 83.63% (46.972/56.163), 8.73% (4.903/56.163), and 7.64% (4.289/56.163), respectively. The results are presented in

Table 6. Principal component analysis.

Linear Combination Coefficient Matrix
	Component
	F1	F2	F3
Immersion
I1. I gained new knowledge from my visit to the digital museum.	0.200	0.112	−0.373
I2. This experience broadened my understanding.	0.193	0.106	0.139
I3. It aroused my interest in exploring unfamiliar topics.	0.185	−0.118	−0.034
I4. I found the digital museum visit to be highly engaging.	0.212	0.094	−0.047
I5. The exhibits in the digital museum demonstrated meticulous attention to detail.	0.218	0.228	−0.039
I6. I thoroughly enjoyed my time at the digital museum.	0.208	−0.169	0.170
I7. The digital museum visit was entertaining.	0.212	−0.216	−0.300
I8. I was deeply engrossed in the digital museum experience.	0.204	−0.219	0.067
I9. The digital museum provided a delightful experience.	0.232	0.204	−0.114
I10. I felt as though I assumed a different identity during the visit.	0.197	0.063	0.134
I11. I experienced a sense of being transported to another era or place.	0.203	−0.269	−0.042
Culture
C1. I felt entirely disconnected from the real world.	0.186	−0.025	0.406
C2. The digital museum visit gave me a sense of spontaneous excitement.	0.185	0.209	−0.197
C3. I unexpectedly engaged in experiences I had never imagined before.	0.225	0.147	−0.192
C4. I encountered surprising and delightful moments while visiting.	0.178	0.411	0.258
C5. The digital museum enabled interaction with local individuals.	0.193	0.274	0.022
C6. Through the displays, I could immerse myself in the local culture.	0.222	−0.225	−0.193
C7. I gained insight into the daily lives of local people.	0.204	−0.175	0.126
C8. The experience fostered connections that turned strangers into acquaintances.	0.225	0.079	−0.000
C9. I felt a strong sense of belonging to the location I was visiting.	0.190	0.104	0.075
C10. I felt included in a unique travel-oriented community during the visit.	0.219	0.031	−0.329
Personalization
P1. The experience felt customized to suit my preferences.	0.217	−0.299	0.100
P2. The digital museum was adapted to meet my specific interests.	0.177	−0.397	0.081
P3. The interaction made me feel personally valued as a visitor.	0.203	0.066	0.436

.

Figure 9 shows the PCA scatter plot, which reveals the differences between the three dimensions. The immersion dimension (PC1) exhibits minimal variation in user feedback, suggesting its stability across different users. In contrast, the cultural dimension (PC2) and the personalization dimension (PC3) demonstrate a greater range of variation, suggesting that these dimensions are the primary sources of variation in the user experience.

4.1.4. Age and Gender Effects on Experience Scores

An age-based analysis shows that average experience scores decline slightly with age. The 18–24 age group had the highest average score (3.85), and the 55+ age group had the lowest average score (3.25) [53]. Younger participants rated novelty and uniqueness higher. However, conventional experiences did not vary significantly across age groups (Figure 10).

A gender analysis (Figure 11) revealed subtle differences. Men rated spontaneity and interactivity slightly higher, while women favored educational and cultural elements. These results imply that slight gender preference patterns exist in digital museum participation [54].

4.1.5. Confirmatory Factor Analysis Results

The results of the model fit indicators indicate that the overall model exhibits a good fit. The chi-square value (χ²) is 249.059, the degree of freedom (df) is 249, and the p-value is 0.487. These values indicate that the discrepancy between the model and the data is not statistically significant, thereby supporting the conclusion that the model exhibits an excellent fit. The chi-square degree of freedom ratio is 1.000, which is close to the ideal value, thereby further verifying the suitability of the model. The GFI is 0.918, the CFI is 0.996, and the TLI is 0.995, all of which meet the established criteria, indicating a high degree of model fit. The RMSEA was 0.001, with a 90% confidence interval of 0.001 to 0.027, indicating minimal model error and an excellent fit. The RMR was 0.080, which is slightly elevated but still within an acceptable range. It is noteworthy that the NFI value of 0.142 does not meet the criterion for a good fit, which is greater than 0.9. This may be attributed to the data distribution or the complex relationship between variables. Moreover, the PGFI and PCFI values of 0.762 and 0.898, respectively, are within the acceptable ranges (

Table 7. CFA model fitting index.

χ2	df	p	χ2/df	GFI	RMSEA	RMR	CFI	NFI	NNFI
249.059	249	0.487	1.000	0.918	0.001	0.080	0.996	0.142	0.995
TLI	AGFI	IFI	PGFI	PNFI	PCFI	SRMR	RMSEA 90% CI
0.995	0.901	0.999	0.762	0.128	0.898	0.059	0.001~0.027

).

In light of the preliminary analysis, which indicated an optimal fit for the model, the subsequent phase is to substantiate the three-factor structure further. The analysis was conducted to assess the discriminant validity. In terms of immersion, the square root of the AVE value is 0.220, which is greater than the maximum absolute value of the inter-factor correlation coefficient of 0.117. This indicates that the model has good discriminant validity. With regard to the construct of culture, the root mean square of the AVE is 0.323, which is greater than the maximum absolute value of the inter-factor correlation coefficient of 0.165. This indicates that the construct of culture exhibits good discriminant validity. In regard to personalization, the root mean square of the AVE is 0.285, which is greater than the maximum absolute value of the inter-factor correlation coefficient of 0.165, indicating good discriminant validity [55]. Please refer to

Table 8. Pearson correlation and discriminant validity of the square root of AVE.

Factors	Immersion	Culture	Personalization
Immersion	0.220
Culture	0.068	0.323
Personalization	0.117	0.165	0.285

Note: The bold diagonal line is the square root of AVE.

for details.

4.2. Multisensory Experience Analysis of Digital Museums Based on SEM

This section uses maximum likelihood estimation to construct an SEM to investigate the relationships among key psychological and experiential variables. This model examines how two core predictive factors—information perception (including cultural understanding and memory retention) and emotional response (including pleasure, arousal, and dominance)—affect three key dimensions of digital museum experiences: immersion, culture, and personalization. Reliability and validity analyses demonstrated the robustness of the measurement model, and model fit indices confirmed that the model structure was satisfactory. Additionally, one-way analysis of variance and multiple comparisons revealed significant statistical differences in the overall experience across the three multisensory scenarios, confirming Hypothesis 3 and highlighting the distinct contributions of different sensory modalities to immersive museum experiences.

4.2.1. Descriptive Statistics

For a detailed examination of the observed variables within the structural equation model, please refer to

Table 9. Descriptive analysis of measurement variables.

Variables	M	SD
Cultural understanding	3.718	1.321
CU1	3.708	1.311
CU2	3.705	1.367
CU3	3.742	1.287
Memory retention	3.716	1.262
MR1	3.760	1.266
MR2	3.674	1.266
MR3	3.714	1.255
Pleasure	3.770	1.274
PL1	3.726	1.313
PL2	3.822	1.302
PL3	3.766	1.245
PL4	3.766	1.233
Arousal	3.774	1.285
AR1	3.785	1.278
AR2	3.763	1.292
Dominance	3.744	1.274
DO1	3.766	1.252
DO2	3.782	1.252
DO3	3.748	1.307
DO4	3.680	1.280
Immersion	4.825	2.455
Culture	2.792	1.076
Personalization	3.506	1.067

. As shown in the table, the cultural understanding and memory retention variables are as follows: The ratings of the two types of variables were moderate, with a mean value of approximately 3.7.

Emotional response: Scores for pleasure, arousal, and dominance were high, averaging around 3.75, with pleasure scoring the highest. The principal components of the experience are as follows: The score for the immersive experience was the highest (4.825).

4.2.2. Reliability Analysis for Information Perception and Emotional Response

Subsequently, Cronbach’s Alpha, composite reliability, and average variance extracted were employed to ascertain the reliability. The outcomes of the independent variable validation are presented in

Table 10. Cronbach’s alpha reliability analysis results for information perception.

Items	CITC	The Deleted Alpha Coefficient	Cronbach’s Alpha
Information perception			0.600
CU1	0.366	0.542
CU2	0.372	0.539
CU3	0.315	0.564
MR1	0.324	0.560
MR2	0.283	0.576
MR3	0.341	0.553

and

Table 11. Cronbach’s alpha reliability analysis results for emotional response.

Items	CITC	The Deleted Alpha Coefficient	Cronbach’s Alpha
Emotional response			0.701
PL1	0.258	0.696
PL2	0.393	0.672
PL3	0.329	0.683
PL4	0.346	0.680
AR1	0.427	0.666
AR2	0.319	0.685
DO1	0.391	0.673
DO2	0.421	0.667
DO3	0.340	0.682
DO4	0.372	0.676

.

4.2.3. Structural Equation Model Results

As shown in

Table 12. Summary table of model regression coefficients.

X	→	Y	SE	CR	p	Standardized Estimates
Information perception	→	Cultural understanding	-	-	-	0.961
Information perception	→	Memory retention	0.158	6.040	0.000	1.000
Emotional response	→	Pleasure	-	-	-	1.000
Emotional response	→	Arousal	0.222	5.502	0.000	0.887
Emotional response	→	Dominance	0.212	5.429	0.000	0.902
Cultural understanding	→	CU1	-	-	-	0.474
Cultural understanding	→	CU2	0.169	6.242	0.000	0.480
Cultural understanding	→	CU3	0.156	6.040	0.000	0.456
Memory retention	→	MR1	-	-	-	0.450
Memory retention	→	MR2	0.163	5.775	0.000	0.423
Memory retention	→	MR3	0.165	5.985	0.000	0.447
Pleasure	→	PL1	-	-	-	0.368
Pleasure	→	PL2	0.231	5.603	0.000	0.481
Pleasure	→	PL3	0.208	5.308	0.000	0.429
Pleasure	→	PL4	0.208	5.357	0.000	0.437
Arousal	→	AR1	-	-	-	0.521
Arousal	→	AR2	0.142	5.657	0.000	0.414
Dominance	→	DO1	-	-	-	0.493
Dominance	→	DO2	0.163	6.911	0.000	0.554
Dominance	→	DO3	0.156	6.004	0.000	0.444
Dominance	→	DO4	0.155	6.115	0.000	0.456
Information perception	→	Immersion	-	-	-	0.617
Information perception	→	Culture	-	-	-	−0.031
Information perception	→	Personalization	-	-	-	0.004
Emotional response	→	Immersion	-	-	-	−0.620
Emotional response	→	Culture	-	-	-	−0.568
Emotional response	→	Personalization	-	-	-	0.571

, after analyzing the relationships between cultural understanding, memory retention, pleasure, arousal, dominance, immersion, culture, and personalization, the following results were obtained.

The structural equation model path analysis diagram illustrates the influence mechanism of information perception and emotional response on the three-dimensional experience structure of the digital museum in the context of multisensory interaction design. The construct of information perception is composed of two dimensions: cultural understanding and memory retention. The standardized path coefficients for these two dimensions are 0.961 and 1.000. Information perception has a significant positive impact on the immersive experience (path coefficient 0.617), but a weaker or even insignificant impact on the cultural experience (path coefficient 0.004) and the personalized experience (path coefficient −0.031). The construct of emotional response is composed of three dimensions: pleasure, arousal, and sense of dominance. The standardized factor loadings of arousal and sense of dominance are 0.887 and 0.902. Emotional response has a notable positive impact on the personalized experience (path coefficient: 0.571), but a weaker or even negative impact on the immersive experience and cultural experience (respectively, −0.620 and −0.031). These findings are illustrated in Figure 12.

Given the limited scope of this study, the relationship between information perception, cultural experience, and personalized experience is not addressed. The results of the updated model are presented in Figure 13.

A comparison of the model fitting indices of the measured model and the structural model indicates that the measured items utilized in the structural model are appropriate. The model fitting indices are presented in

Table 13. Structural equation model fitting index.

χ2	df	p	χ2/df	GFI	RMSEA	RMR	CFI	NFI	NNFI
173.033	141	0.034	1.227	0.947	0.026	0.069	0.968	0.852	0.961
TLI	AGFI	IFI	PGFI	PNFI	PCFI	SRMR	RMSEA 90% CI
0.961	0.929	0.969	0.703	0.703	0.798	0.043	0.008~0.039

.

4.2.4. Multisensory Format Effects

To elucidate the impact of the multisensory format, this study employed one-way ANOVA to ascertain the influence of the multisensory format on the experience. To ascertain whether there were differences in the means of the three scenarios, ANOVA and multiple comparisons were utilized for further verification. This study employed Dunn’s t-test as the multiple comparison method to compare the means of the three scenarios. The results are presented in

Table 14. ANOVA interim process value.

Source	Difference	Sum of Squares	Freedom	Mean Square	F	p
Overall experience	Group	13.061	2	6.531	4.324	0.014
	Intragroup	486.336	322	1.510
	Total	499.397	324

.

As can be seen from the above table, the overall experience shows significance (p < 0.05) in all cases, meaning that there are differences in the overall experience. Specific analysis shows that the overall experience shows significance at the 0.05 level (F = 4.324, p = 0.014), and specific comparative differences show that the mean values of the groups with more obvious differences are compared, and the results are “scene 1 > scene 3; scene 2 > scene 3”. Figure 14 shows the mean values of the experience under the three scenarios.

The results of the multiple comparisons are presented in

Table 15. Dunn’s t-test for pairwise comparisons.

	(I) Item	(J) Item	(I) M	(J) M	Difference (I–J)	p
Overall experience	Scenario 1	Scenario 2	4.000	4.000	0.000	0.700
	Scenario 1	Scenario 3	4.000	4.000	0.000	0.017 *
	Scenario 2	Scenario 3	4.000	4.000	0.000	0.045 *

* p < 0.05.

. The results demonstrate that there is no statistically significant difference between Scene 1 and Scene 2 (p = 0.700). This indicates that the overall experience is not significantly different between the two scenes. The results demonstrate a statistically significant difference between Scene 1 and Scene 3 (p = 0.017), indicating that the overall experience differs between the two scenes. Similarly, the comparison between Scene 2 and Scene 3 (p = 0.045) also indicates a statistically significant difference in the overall experience.

As illustrated in Figure 14, scenario 1 (visual and auditory) exhibited the highest mean experience score of 4.90. The mean score for Scenario 2 (visual, auditory, and tactile) was 4.45, which was slightly lower than that of Scenario 1. The lowest mean score was observed for Scenario 3 (visual, auditory, and olfactory).

5. Discussion

Building upon the empirical analysis results from Section 4, this chapter integrates multi-sensory interactive experience theory with survey data to explore the key influencing factors and characteristics of the three core dimensions of “immersion”, “culture”, and “personalization.” Additionally, the chapter analyzes how different demographic variables, such as gender and age, affect the perception of multi-sensory experiences. Based on these findings, the chapter proposes strategies to optimize multi-sensory interactive design in digital museums and suggests future development directions.

5.1. Identification and Structural Analysis of Multisensory Interaction Experience Dimensions

Through principal component analysis, three main components were identified: “immersive experience”, “cultural experience”, and “personalized experience.” These form the core framework of multisensory interaction experiences in digital museums. The “immersive experience” dimension encompasses the immersive sense created by the environment, the immersive nature of the interaction process, and emotional resonance and immersion. The “cultural experience” dimension focuses on visitors’ cultural cognition, cultural participation, and cultural internalization processes. The “personalized experience” dimension emphasizes users’ autonomy, content relevance, and sense of participation during the experience. This dimensional classification reveals the structural characteristics of multisensory interactive experiences and provides a theoretical foundation for subsequent design optimization.

5.2. Exploring the Mechanism of “Immersion-Culture-Personalization” in Multi-Sensory Interactive Experiences

Multisensory interactive technology enhances audience immersion by engaging multiple sensory channels, such as vision, hearing, and touch. Structural equation modeling analysis reveals that immersion significantly influences how visitors perceive information and experience emotions, thereby deepening their cultural experiences. Cultural experiences enhance users’ understanding of exhibits and facilitate the construction of cultural memory and identity. Personalized experiences, through customized content recommendations and diverse interaction pathways, enhance user engagement and sense of belonging, improving overall satisfaction with multi-sensory experiences. This mechanism illustrates that immersion serves as the foundation, cultural elements form the core, and personalization acts as the driving force. Together, they construct a three-dimensional support system for multi-sensory interactive experiences in digital museums.

5.3. Comparison with Existing Research and Theoretical Contributions

(1)

High Consistency with Existing Literature:

This study found that “immersive experiences” significantly impact the construction of cultural memory. This finding aligns closely with the conclusions of Anwar [56], who discovered that immersive experiences promote cultural internalization. This indicates that immersive experiences are an important pathway for promoting a deep understanding of cultural content among visitors, aligning closely with existing theoretical frameworks. Additionally, “cultural experiences” serve as a bridge between cognition and emotion, aligning closely with Psomadaki’s [57] findings that cultural experiences can transform sensory experiences into deep cultural understanding. This high degree of alignment enhances our confidence in the role of immersive and cultural experiences in explaining how multisensory technologies empower cultural dissemination.

(2)

Innovations and Differences from the Existing Literature

Unlike the existing literature, this study reveals that “personalized experiences” significantly and stably affect the overall visit experience rather than relying solely on immersion and cultural experiences. This finding expands the theoretical understanding of the pathways through which multisensory experiences exert their effects. It fills the gap where “personalization” was viewed as subordinate to the overall experience. It suggests that providing highly customized experiences at the right time can significantly enhance the overall experience, aligning closely with Jiménez-Barreto’s [58] assertion that experiences must adapt to different groups’ values.

(3)

Differences from the Existing Literature and Reasons

This study also found significant differences in how different age groups respond to multisensory experiences. These differences can be explained by variations in life experiences, values, and the theory that information processing methods change with age. This finding provides theoretical support for practitioners to develop differentiated experiences for different groups and fills a gap in the existing literature, which has not sufficiently addressed the issue of “group heterogeneity.” This focus on group differences closely aligns with the recommendation of Liu and Sutunyarak [59] to design tailored experiences for different target groups while providing concrete operational pathways for practice.

6. Conclusions

Taking multi-sensory interaction design as its starting point, this study explores the pathways through which emotional responses and cognitive processing of information influence the visitor experience in digital museums. The study aims to provide theoretical support and practical guidance for the immersive dissemination of cultural heritage. The main conclusions are as follows:

(1)

Both emotional responses and information perception significantly impact the visitor experience positively. Emotional responses primarily influence personalized experiences, while information perception primarily enhances immersive and cultural experiences.

(2)

The synergistic effects of different sensory combinations on the experience vary. Experiences combining visual and auditory elements are the most prominent, while tactile and olfactory elements have room for improvement.

(3)

Different groups have different preferences for experience modes. Younger groups prioritize highly interactive and personalized experiences, while middle-aged and older groups place greater emphasis on cultural depth and authoritative information. These findings suggest that multi-sensory interactions can be designed to better accommodate the needs of different visitor groups.

7. Limitations and Implications

7.1. Limitations

This study constructed a relatively comprehensive model of multi-sensory interaction experiences and validated its applicability in digital museum settings through empirical analysis. However, the following limitations remain and require further exploration and refinement in future research:

(1)

Limitations in sample region and sensory types. This study primarily focused on digital museum visitors from specific regions, which may introduce regional cultural biases and affect the generalizability of the conclusions. Additionally, while this study involves multisensory cues, such as visual, auditory, tactile, and olfactory stimuli, exploration of gustatory stimuli is lacking due to technical limitations and constraints of the experimental environment. The interactive effects between weaker senses have not been thoroughly analyzed. Future research could expand the geographical scope of the sample and incorporate more sensory elements to enhance the model’s applicability.

(2)

Lack of tracking of dynamic behavior and long-term impacts. This study primarily focuses on visitors’ emotional responses and cognitive changes during short-term visits without addressing long-term behavioral changes after visits, such as cultural dissemination behaviors, cognitive retention, and social interactions. Future research could combine behavioral tracking technologies with longitudinal data collection to explore how multi-sensory experiences enhance cultural identity and promote sustainable actions in the long term.

7.2. Implications

(1)

Government Level: Strengthen policy support and cross-sectoral cooperation mechanisms. Governments can establish application standards for multisensory interactive technologies in the cultural dissemination sector to promote the integration of culture and technology. They should also support collaborative innovation among universities, technology companies, and cultural institutions. They should encourage technical pilot projects and application conversions in digital museum settings. Finally, they should create a favorable policy and technological environment for the high-quality development of the digital cultural industry.

(2)

Enterprise level: Promote the integration of the multi-sensory content industry chain. Cultural and technological enterprises should analyze audience sensory preferences and behavioral patterns, develop more immersive multi-sensory content products, and increase user engagement. They can also explore joint operations with cultural tourism sites and cultural and creative platforms to promote the integration of “virtual + physical” experiences and expand commercial scenarios for multi-sensory interaction.

(3)

Museum management level: Optimize audience-centric sensory experience pathways. Managers should use the “immersive, cultural, and personalized” three-dimensional experience structure to guide the layout of sensory devices in exhibition spaces. This will strengthen the synergy between visual and auditory elements and enhance exhibition engagement and memorability. They should combine audience data to make dynamic adjustments and achieve precise sensory experience design, thereby improving visitor satisfaction and the effectiveness of cultural dissemination.