Evaluating Cultural Heritage Preservation Through Augmented Reality: Insights from the Kaisareia-AR Application

Abstract

This study investigates how augmented-reality (AR) technology can enhance the presentation and preservation of cultural heritage, using Kayseri Castle as a case study. Although previous studies have explored AR applications in heritage contexts, few have addressed their role in representing multi-layered architectural histories of complex sites. The research focuses on the development and evaluation of the KAISAREIA-AR application, which integrates historical, architectural, and cultural narratives into an interactive AR platform. By reconstructing the castle’s distinct historical layers—spanning the Roman, Seljuk, Ottoman, and Republic periods—the study seeks to assess AR’s effectiveness in providing immersive visitor experiences while maintaining the authenticity of heritage sites. Three-dimensional models of the castle were created using 3ds Max, enriched with visual and auditory information, and deployed via Unity software on wearable AR devices. The study employed the Unified Theory of Acceptance and Use of Technology (UTAUT) framework and Structural Equation Modeling (SEM) to evaluate the application’s usability and impact on user engagement. The findings indicate that AR significantly enhances the accessibility, understanding, and appreciation of cultural heritage by providing dynamic, immersive experiences. The KAISAREIA-AR application demonstrated its potential to bridge historical authenticity with modern technology, offering a replicable model for integrating AR into cultural heritage conservation and education.

1. Introduction

Cultural heritage serves as both a tangible and intangible asset of human civilization, embodying traditions, values, architectural knowledge, collective memory, and identity. The preservation of cultural heritage has evolved beyond static conservation practices into a multidimensional, participatory, and technologically mediated endeavor. As tourism becomes an increasingly important platform for cultural interaction, the value of cultural heritage now extends beyond its intrinsic significance to encompass economic, educational, and experiential dimensions [1].

Within this dynamic context, cultural tourism ecosystems have emerged, transforming heritage sites from passive objects of consumption into active spaces of co-creation and interpretation. Tourists, site managers, local communities, and institutions now form a network of stakeholders who engage with heritage through increasingly digital means. In such ecosystems, the concepts of cultural landscape, intangible heritage, stakeholder engagement, and experiential authenticity are central. A cultural landscape reflects the interaction between humans and their environment, shaping identity through spatial practices. Intangible heritage includes oral traditions, social practices, rituals, and expressions that live within and around physical structures. Furthermore, experiential authenticity—the sense of encountering something meaningful, real, and context-rich—has become a key metric for visitor satisfaction and learning. Stakeholder engagement ensures that preservation remains inclusive, ethical, and socially embedded.

Amid these developments, Augmented Reality (AR) has emerged as a transformative tool for experiencing, interpreting, and preserving cultural heritage. AR overlays digital information—images, audio, reconstructions—onto real environments in real time, enabling users to interact with multi-layered narratives in situ [2]. In cultural tourism, AR supports experiential learning, dynamic storytelling, and co-creation of meaning, especially when delivered through wearable devices that integrate seamlessly with users’ spatial experiences [3].

The growing use of AR in heritage aligns with contemporary trends in digital preservation, which emphasize accessibility, interactivity, and sustainability. From large-scale initiatives like Google Arts & Culture to site-specific mobile applications and mixed-reality museum exhibits, AR is used not only to attract diverse audiences but also to deepen public understanding through immersive, user-centered experiences [4]. Despite this momentum, questions remain regarding AR’s ability to represent multi-period historical sites, maintain authenticity, and support stakeholder-informed heritage narratives, especially in less globally prominent yet culturally significant locations.

In this regard, the city of Kayseri in central Anatolia presents a compelling yet underrepresented context for exploring AR applications in cultural heritage. A historically significant settlement, Kayseri has hosted numerous civilizations, including the Romans, Byzantines, Seljuks, Ottomans, and the modern Turkish Republic. The Kayseri Castle, a central monument within this urban fabric, embodies a multi-layered architectural history that remains largely inaccessible to the public in its full historical depth. Although each period left physical traces—walls, towers, and reinforcements—these transformations are not readily visible to the naked eye, resulting in a disconnection between the site’s tangible form and its intangible historical layers.

To bridge this gap, the KAISAREIA-AR application was developed AR tool that visualizes the architectural evolution of Kayseri Castle. By integrating digital reconstructions of the Roman, Seljuk, Ottoman, and Republican periods, the application enables visitors to explore the site through wearable devices such as Magic Leap 1. This interactive AR experience allows users to perceive the castle’s transformation in real time while remaining physically present at the site. The KAISAREIA-AR model was developed using Unity, 3ds Max (version Autodesk 3ds Max 2022), and The Lab, and features an interactive interface optimized for field use.

Beyond enhancing public engagement, this project contributes to the growing discourse on digital heritage, co-creation, and technology-assisted preservation. The study is guided by the following research question:

• How can augmented-reality (AR) technologies enhance visitor experiences while preserving the authenticity of cultural heritage sites?

Additionally, the study explores the following sub-questions:

• In what ways can wearable AR devices increase public awareness of architectural conservation and cultural heritage?
• What strategies can maximize AR’s impact on preserving and promoting cultural heritage?
• How can the implementation process for Kayseri Castle serve as a replicable model for similar heritage sites?

Methodologically, this research combines historical data analysis, 3D modeling, and wearable AR deployment within an integrated digital workflow. It also empirically examines user perceptions through a quantitative evaluation based on the UTAUT model, which is well-suited for assessing technology acceptance, combined with structural equation modeling. In doing so, the KAISAREIA-AR project addresses a critical gap in AR heritage practices, offering both a functional prototype and a conceptual framework for strengthening authenticity, interpretation, and stakeholder engagement at culturally layered heritage sites.

Ultimately, this study demonstrates how AR can facilitate both on-site and remote engagement with cultural heritage, providing a practical framework for incorporating AR into preservation efforts. The KAISAREIA-AR application thus serves as an innovative example of how AR technologies can deepen our understanding, experience, and appreciation of historically layered sites such as Kayseri Castle.

2. Literature Review

2.1. Cultural Asset Preservation and Digitization

Cultural heritage is now widely understood as a complex construct that encompasses both tangible and intangible elements, ranging from monuments and artifacts to oral traditions, rituals, and collective memory. As [1] notes, preservation efforts today extend far beyond the conservation of material remnants, aiming instead to safeguard the cultural meanings and identities embedded within spatial and social practices. The notion of cultural landscapes, introduced by [5], underscores the dynamic relationship between human agency and the built environment, positioning heritage as an evolving process rather than a static product.

Recent scholarship has emphasized the need for inclusive and participatory models of heritage preservation. According to [6], the competitiveness and sustainability of heritage are strongly influenced by stakeholder engagement. This perspective expands the circle of responsibility to include not only heritage professionals and policymakers but also local communities, tourists, and technologists, each playing a role in shaping the interpretive narratives that surround cultural assets. Integrating diverse stakeholder perspectives not only democratizes access to heritage but also enriches interpretation and strengthens long-term resilience.

Furthermore, the concept of experiential authenticity has reshaped how heritage value is understood. Rather than focusing solely on objective historical accuracy, scholars such as [7] highlight the importance of emotional resonance, narrative richness, and situational relevance in shaping a visitor’s sense of authenticity. These frameworks help explain how technologies like AR can enhance heritage interpretation by deepening engagement and meaning-making.

2.2. AR/VR Use in Heritage Interpretation

Augmented Reality (AR) has emerged as a transformative tool within digital heritage practices, enabling users to interact with contextually embedded digital content in real-world settings. As [2] points out, AR represents a disruptive innovation within the tourism sector, fundamentally reshaping how destinations are mediated and experienced. In heritage contexts, AR serves not only as a visualization tool but also as a medium for immersive storytelling, simulation, and memory activation.

The role of AR in promoting experiential learning has been widely acknowledged. For example, Ref. [8] emphasizes that AR supports embodied cognition by aligning digital narratives with physical exploration, thereby enhancing knowledge retention and emotional engagement.

However, the implementation of AR in heritage contexts remains inconsistent, often constrained by technical limitations, curatorial concerns, and the lack of a robust theoretical foundation. As the number of applied AR projects grows, it is increasingly important to critically assess how these technologies influence meaning-making, authority, and inclusivity in heritage interpretation.

2.3. Visitor Engagement and Co-Creation

The adoption of AR in cultural tourism aligns with broader theoretical shifts toward interactivity, personalization, and co-creation in heritage interpretation. As ref. [3] describes, this shift represents a move away from passive service delivery toward real-time co-creation, where visitors actively shape their experiences by engaging with digital content in context-sensitive ways. AR technologies—particularly when implemented through wearable devices—support this paradigm by enabling multisensory, location-aware, and responsive interaction.

The role of stakeholder co-creation has also attracted growing scholarly attention. According to [4], meaningful digital heritage experiences emerge when users, designers, and institutions collaborate in content development and feedback processes. Wearable AR devices such as Magic Leap 2 or HoloLens not only enhance immersion but also blur the boundaries between observation and participation, transforming visitors into situated actors within the heritage narrative.

To systematically evaluate these dynamics, theoretical frameworks such as the Unified Theory of Acceptance and Use of Technology (UTAUT) provide valuable insights. Originally proposed by [9], UTAUT identifies key determinants—performance expectancy, effort expectancy, social influence, and facilitating conditions—that predict technology adoption. When applied to heritage contexts, these dimensions offer a useful lens for assessing how users perceive and interact with AR-based applications.

2.4. Comparative AR Applications in Heritage Contexts

Several case studies demonstrate AR’s potential to enrich the understanding and experience of heritage sites.

Archeoguide Application: Developed by [10], this mobile outdoor application allows users to experience damaged historical sites and environments, aiming to improve the perception of historical spaces.

Heijo-Palace in Japan: Ref. [11] recreated the historical Heijo-Palace area in Nara, Japan, using mixed reality. Models of the palace buildings were placed in scenes derived from global aerial imagery, allowing for visualization in their original locations.

Aurelian Wall Reconstruction: Ref. [12] conducted a photogrammetric study of the Aurelian Wall in Rome, focusing on its stratigraphy and construction phases. The findings were used to create a virtual reconstruction of the northern gate.

Selimiye Mosque in Edirne: Ref. [13] the Selimiye Mosque, creating a 3D model that could be experienced using HTC Vive wearable technology in a virtual environment through the Unity software.

Mixed-Reality Heritage Guide: Ref. [14] developed a mixed-reality guide using HoloLens wearable AR devices to provide information about historical structures. The system is connected to the cloud to offer photos, written data, and videos, enriching the user experience.

Dhaka Heritage Sites: Ref. [15] developed 3D models of the Lalbagh Mosque, the Greek Monument, and Divan-i Aam in Dhaka, Bangladesh. These models were integrated into a virtual reality environment, enabling users to explore these heritage sites interactively.

Rumeli Fortress in Istanbul: Ref. [16] scanned Rumeli Hisarı using a laser scanner and created a 3D model in 3ds Max. The resulting model was made accessible to users through HTC Vive wearable technology.

Warsaw Royal Castle: Ref. [17] used AR to depict the historical changes in the Royal Castle and surrounding structures in Warsaw, Poland, a UNESCO World Heritage Site. Different periods and restoration phases were modeled and transferred to the Unity platform, later deployed on HoloLens 1 wearable AR devices for user interaction.

Alexandria Troas Project: Ref. [18] implemented a location-based AR application for the Alexandria Troas Temple in Çanakkale, Turkey. Unity ARKit AR software was used, and user evaluations were conducted via QR codes and surveys on iOS devices.

Yavuz Battleship: In the study by [19], the historically significant Yavuz Battleship was modeled using the 3ds Max program. The Unity program enhanced Three-dimensional models with auditory and visual historical data, and the work was transferred to the HoloLens mixed-reality headset.

Arkon: In the study prepared by [20], Kapanca Street was scanned using a laser scanner, and the data were transferred to the 3ds Max program to create three-dimensional models. Subsequently, the study was transferred to the Unity program. This application, providing information about the historical buildings on Kapanca Street, was transferred to the HoloLens headset and made available for user experience. The reality environments, devices used, and modeling approaches of these works are illustrated in Figure 1.

These examples highlight the significant role of AR in preserving and presenting cultural heritage, offering interactive and immersive experiences that bridge the gap between historical authenticity and modern technology.

In contrast, the KAISAREIA-AR system introduced in this study is distinguished by its integration of multiple historical periods, location-aware interaction, and wearable device deployment. Rather than simply projecting 3D visuals, it allows users to transition between temporal phases—Roman, Seljuk, Ottoman, Republican—thus constructing a multi-temporal narrative of architectural transformation. This aligns with [8] ’s argument that temporal storytelling enhances user memory, contextual learning, and perceived authenticity.

3. Methodology

This study follows a five-stage methodological process integrating historical documentation, digital modeling, AR application development, empirical testing, and data analysis. Each step has been structured to support the research aim of evaluating the potential of augmented reality (AR) to enhance cultural heritage interpretation and visitor experience at historically layered sites.

1.

Historical Data Collection and Archive Analysis

The first phase of the study includes a comprehensive literature review and archive study in order to analyze the historical continuity of Kayseri Castle and its transformation in different periods in detail. In this context, written and visual documents (engravings, drawings, photographs, etc.) compiled by archaeologist and restitution specialist Mahmut Akok and French architect and art historian Albert-Louis Gabriel were examined. In addition, in order to document the cadastral status of Kayseri Castle, existing plans, maps, academic articles, and thesis sources were scanned, and detailed information about its architectural development was obtained [21]:

2.

3D Model Development

In light of the findings obtained from the historical studies on Kayseri Castle, it was concluded that the castle should be evaluated by dividing it into different historical layers. In this analysis process, which covers different periods including Roman, Seljuk, Ottoman, Republic, and today, various parameters such as the plan features of the structure, material use, function, and restoration processes were taken into consideration.

In this context, models were created in the 3ds Max program for each period. The modeling process consists of the following stages:

• Transfer of historical plans and material textures to digital media
• Determining transformations in the architectural structure by analyzing functional changes
• Three-dimensional modeling of architectural features of Kayseri Castle from different periods using 3ds Max software
• Transferring the created three-dimensional models to Unity software to integrate them into the augmented-reality environment

3.

AR Application Integration

The three-dimensional models were transferred to the Unity program.

The augmented-reality environment containing written, visual (jpeg), and audio (mp3) technical information about Kayseri Castle was created using the Unity program and Visual Studio 2019. During the AR integration process, architectural changes from each period of the castle were presented with a layered structure, allowing users to experience historical transformations by switching between different periods.

The application was tested by integrating it into the Magic Leap 1 wearable augmented-reality glasses in order to increase users’ spatial interactions and experiences. This process constitutes an important stage in terms of examining the contribution of AR technology to the digital presentation of historical structures.

4.

User Testing and Evaluation

In the application phase of the study, case study, historical narrative, qualitative research, and data collection methods were used. An evaluation process was carried out within the framework of the Unified Theory of Acceptance and Use of Technology (UTAUT) [9] to measure user experience and evaluate the effectiveness of the application. A 5-point Likert scale was employed to ensure user comprehension and response consistency, given the limited participant pool and the exploratory nature of the study.

The testing phase of the application was carried out with 100 students of Erciyes University Faculty of Architecture who had previously experienced Kayseri Castle. Users were given the opportunity to experience Kayseri Castle in an augmented-reality environment, and a survey was applied at the end of the process. All procedures involving human participants were reviewed and approved by the Kocaeli University Faculty of Science and Engineering Ethics Committee. No generative AI tools were used during the design, implementation, or evaluation stages.

5.

Data Analysis and Evaluation of Results

The survey results were tested with reliability analysis in order to increase the reliability of the study. After determining that the data were reliable, the hypotheses developed in accordance with the Unified Theory of Acceptance and Use of Technology (UTAUT) were empirically tested, and the study was completed.

4. Case Study and Results

4.1. Data Collection on the Historical Evolution of Kayseri City and Castle

Urban landscapes undergo continuous transformation, preserving traces of their cultural and historical evolution. Kayseri, owing to its strategic location and political significance, has been home to various civilizations, including the Assyrians, Hittites, Phrygians, Medes, Persians, Romans, Danishmends, Seljuks, Mongols, Eretnaids, and Ottomans [22]. The earliest evidence of Kayseri Castle appears on coins minted during the reign of Emperor Gordian III in the 3rd century CE.

The castle comprises two main components: the inner and outer fortifications. While originally constructed during the Roman period, subsequent rulers, including the Byzantines, Danishmends, Seljuks, Dulkadirids, Karamanids, and Ottomans, undertook modifications and expansions of the structure. The most extensive renovations were carried out in the 14th century [23]. Although much of the outer walls have been lost over time, significant portions of the inner fortifications remain well-preserved. A restitution study by [24], presented in Figure 2, features detailed drawings of both the inner and outer castle walls, offering valuable documentation of their historical configuration.

To investigate the historical evolution of Kayseri Castle, archival sources such as engravings, maps, architectural plans, and photographs were systematically reviewed. Primary sources, including works by [21,24], were analyzed to trace the castle’s architectural and cultural history across five major historical periods: Roman, Seljuk, Ottoman, Republican, and modern. These documents were essential in reconstructing the architectural features, material usage, structural changes, and restoration efforts over time. This phase also involved the categorization and stratification of the castle’s historical data to support the subsequent modeling process.

4.1.1. Kayseri Castle in the Roman Period

The origins of Kayseri Castle trace back to the Roman era. Extensive walls were built during Emperor Gordian III’s reign (242 CE) but were later narrowed under Byzantine Emperor Justinian (527–565 CE) [25]. Archaeological findings from the nearby Gültepe mound suggest the possibility of earlier structures [21]. Variations in construction techniques and architectural plans indicate the castle underwent functional and structural changes over time. Ref. [21] reconstructions reveal Roman inner castle was primarily defensive in design, which is depicted in Figure 3.

4.1.2. Kayseri Castle in the Seljuk Period

During the Seljuk period, Kayseri Castle served as a central point for defense and administration. Repairs and additions included the reuse of Roman walls and the construction of new bastions, such as the Yogunburç towers, to fortify the southern walls. A wide moat was added around the outer walls for enhanced defense. Ref. [24] noted the presence of gardens surrounding the castle, likely irrigated by the moat’s water supply. (Figure 4).

4.1.3. Kayseri Castle in the Ottoman Period

Following the Seljuk state’s decline in the 14th century, Kayseri transitioned through the Beyliks period before becoming part of the Ottoman Empire in 1467. Many Seljuk structures were retained, while urban life centered around the castle (Figure 5). Matrakçı Nasuh’s 16th-century miniatures depict the castle as a focal point for religious and civic activities.

In the 17th century, travelers such as Polish Simeon and Evliya Çelebi documented the castle, noting its robust construction and strategic layout. Ref. [24] later illustrated the inner castle, highlighting its religious and residential structures (Figure 6). Notably, a mosque and fountain were constructed in the inner castle during the Ottoman period under the supervision of Gedik Ahmet Pasha [29].

During the Ottoman period, Gedik Ahmet Pasha supervised the construction of a mosque and fountain, replacing an earlier mosque. This structure, located in the northwest of the castle, is referred to as both the Fatih Mosque and the Castle Mosque. (Figure 7). In the early 20th century, plans to convert the inner castle into a prison led to expropriation and the relocation of residents, but unsuitable conditions halted this use [30].

4.1.4. Kayseri Castle in the Republican Period

Between 1927 and 1929, Gabriel documented extensive deterioration of Kayseri’s historical structures, including mosques adorned with colored plaster, madrasahs converted into barns, and decaying tombs [24]. (Figure 8). By 1936, observed that houses within the inner fortress, deemed lacking in historical significance, had been demolished during the war, and the site was subsequently utilized as a municipal warehouse. By 1939, the area had been transformed into a marketplace, which evolved into a vegetable market in the 1950s, accompanied by the construction of small shops to facilitate local commerce [21]. The transformation of the fortress into a marketplace is depicted in Figure 9.

In the 1970s, large sections of the outer walls were demolished, although the inner walls were preserved. In the 1980s, a new gate was added to improve access, but this addition faced criticism for disrupting the castle’s visual and structural integrity. The High Council of Immovable Antiquities and Monuments designated Kayseri as a protected area in 1975, followed by the inner castle and its surroundings being declared an urban protected area in 1979. Despite these protections, makeshift structures appeared near the walls, complicating preservation efforts [21].

4.1.5. Kayseri Castle in the Contemporary Period

Since 1994, the inner castle has served as a public space. In 2008, a national competition titled “Conversion of Kayseri Inner Castle into a Cultural and Artistic Environment by Preserving It” aimed to redefine the site while safeguarding its historical significance. Following the competition, the marketplace within the castle was removed, initiating extensive restoration efforts. As depicted in Figure 10, the structures within the castle were demolished to pave the way for new construction.

In 2020, the Kayseri Inner Castle Archaeological Museum was inaugurated, prioritizing the preservation of historical elements while promoting public engagement. The restoration project integrated terraces and streets that connected prominent features such as the mosque and gates. This approach highlighted the seamless blending of historical and recreational elements, emphasizing the castle’s role as a cultural hub [34]. The winning project’s plan and visualization are presented in Figure 11.

Figure 10. (a,b) Demolition of the bazaar complex within the castle [35].

Figure 10. (a,b) Demolition of the bazaar complex within the castle [35].

Figure 11. (a) Kayseri Castle Archaeological Museum Plan [36], (b) General View of Kayseri Castle Archaeological Museum [33].

Figure 11. (a) Kayseri Castle Archaeological Museum Plan [36], (b) General View of Kayseri Castle Archaeological Museum [33].

The extensive dataset collected to analyze the historical evolution of Kayseri Castle has enabled a thorough examination of its architectural and structural characteristics across various historical periods. This information has been instrumental in creating detailed three-dimensional models and integrating them into augmented-reality (AR) technologies. While Akok’s documentation provides partial insights into the Roman fortifications, the reconstructions of the elevations remain largely hypothetical due to limited surviving architectural evidence. These reconstructions were based primarily on comparative typologies and assumptions informed by regional Roman military architecture, which introduces inherent uncertainties. Acknowledging these limitations is essential to contextualizing the visual outputs within an interpretative framework rather than as definitive restorations. In this context, the subsequent section delves into the methodology for digitizing the historical layers of Kayseri Castle and their seamless integration into the augmented-reality application.

4.2. 3D Modeling and AR Development: Integrating Kayseri Castle into Augmented-Reality Applications

This section provides a detailed overview of the integration of augmented-reality (AR) technology into the KAISAREIA-AR application, specifically developed for use with wearable AR devices. Among the available software platforms, Unity was selected for its extensive compatibility with a range of AR devices and its advanced functional capabilities.

Unity was chosen for the KAISAREIA-AR application due to its high versatility and flexibility, enabling compatibility with various AR devices. The AR Foundation framework within Unity facilitates seamless integration with platforms such as iOS (ARKit) and Android (ARCore), thereby enhancing the application’s potential to reach a broader and more diverse user base in the future. Additionally, Unity provides an extensive asset library, comprehensive documentation, and a robust community support system. These features contribute to accelerating the development timeline and enhancing the overall quality of the application. Furthermore, Unity’s real-time 3D visualization capabilities are essential for accurately rendering the historical and cultural sites of Kaisareia, thus ensuring a high degree of precision in the representation of these spaces.

The Magic Leap wearable AR device was chosen for its advanced features, such as high-resolution imagery and environmental sensing, which enable a highly interactive and immersive user experience. With high-resolution imagery and advanced environmental sensing capabilities, these devices enable the realistic reconstruction and presentation of historical sites. Magic Leap’s wide field of view and support for natural user interactions are consistent in optimizing user engagement within the KAISAREIA application. Consequently, Unity and Magic Leap were chosen for their complementary strengths in ensuring the application’s performance and enhancing the user experience.

4.2.1. Modeling Kayseri Castle in 3ds Max

Three-dimensional models of Kayseri Castle were created in 3ds Max, depicting its Byzantine, Seljuk, and Ottoman periods. The modeling process was based on detailed historical drawings by [21,24]. (as seen in Figure 12). These models were exported in FBX format to ensure compatibility and subsequently transferred to Unity for further development.

4.2.2. Transferring Models to Unity AR Application Development

The Unity platform was configured to support wearable AR devices by installing specific tools. The “Lab” application was used to set up the development environment on both Windows and Mac devices. Unity and the Zero Iteration package were installed using the Unity Package Manager. The Unity version used was 2020.3, with the Lumin SDK version 0.26. Compatibility details for these versions are summarized in the Unity Hub application was configured with features such as “Lumin OS (Magic Leap) Build Support” and “Visual Studio 2019.” A sample Unity project file compatible with SDK 0.24.0 was used to facilitate development. Furthermore, the setup included enabling Lumin compatibility and adding the required SDK from the preferences menu. The Magic Leap XR Plugin (version 6.2.2) was installed to manage AR interactions, and API settings were optimized for the Lumin platform. Unity project settings were updated to include API-level adjustments, with color modes set to linear for improved rendering quality. Platform-specific optimizations were made for both Mac and Windows systems. A developer certificate was generated in the Magic Leap interface to enable project deployment. The certificate was added to Unity’s Player-Publish Settings section. Once technical configurations were completed, AR functionalities were added, beginning with replacing the default Unity camera with the Magic Leap Main Camera.

• Adding Historical Information layers

As illustrated in Figure 13, historical information cards were created in Photoshop, combining visual and textual details about the historical phases of Kayseri Castle. These cards were then imported into Unity as JPEG files to enrich the application’s educational features.

• Adding Audio Features

Audio narration describing the history of Kayseri Castle was recorded and integrated into the Unity project. Audio Source and MSA Listener components were employed to create an immersive 3D sound experience, aligning with the immersive nature of AR technology (Figure 14).

• Creating a User Interface

A user-friendly menu was developed to simplify navigation (See Figure 15). The menu included tabs for the Byzantine, Seljuk, and Ottoman periods, allowing users to access corresponding models, images, and audio. Buttons were programmed for switching between periods and performing model manipulations, such as enlargement or rotation.

In the wearable AR device interface, the trigger button is used to select the exhibition section, with all selection actions performed using the same command. Within the period views, the trigger button enables model enlargement, the bumper button allows rotation, and the back button restores the model to its original state (Figure 16).

5. Discussion

100 undergraduate and graduate students from the Faculty of Architecture at Erciyes University evaluated the KAISAREIA-AR application. The Unified Theory of Acceptance and Use of Technology (UTAUT) model and Structural Equation Modeling (SEM) were employed to analyze the application’s usability and effectiveness.

5.1. Model and Hypotheses

The UTAUT framework [9] was used to examine the relationships between key constructs: Performance Expectancy (PE), Effort Expectancy (EE), Facilitating Conditions (FC), Hedonic Motivation (HM), Behavioral Intention (BI), and Usage Behavior (UB). Five hypotheses were proposed to assess how these constructs influenced technology adoption. (

Table 1. Hypotheses of the Study.

Hypothesis	Statement
H1	Performance Expectancy positively influences Behavioral Intention
H2	Effort Expectancy positively influences Behavioral Intention
H3	Facilitating Conditions positively influence Usage Behavior
H4	Hedonic Motivation positively influences Behavioral Intention
H5	Behavioral Intention positively influences Usage Behavior

). Path analysis using SEM was conducted to determine the relationships among these variables. (Figure 17).

Path analysis using Structural Equation Modeling (SEM) was employed to test these hypotheses and evaluate the relationships between the key constructs.

Participants rated the constructs using a five-point Likert scale ranging from “Strongly Disagree” to “Strongly Agree”. The survey included 23 items distributed across the constructs as follows: Performance Expectancy (4 items), Effort Expectancy (4 items), Facilitating Conditions (4 items), Hedonic Motivation (3 items), Behavioral Intention (3 items), and Usage Behavior (2 items). The list of survey items is included in Appendix A.

5.2. Participants and Procedure

The study involved a sample of 100 participants, comprising undergraduate and graduate students from the Department of Architecture at Erciyes University, Kayseri.

Table 2. Participant Demographics.

Category	Subcategory	Count
Gender	Female	59
	Male	41
Age	18–25	68
	24–30	27
	30–35	3
	35+	2
Education Level	High School Graduate	71
	Bachelor’s Graduate	25
	Master’s Graduate	4
Previous AR Experience in Heritage	Yes	19
	No	81
Previous Use of Wearable AR Devices	Yes	10
	No	90

summarizes the participants’ demographics.

Before administering the survey, participants received comprehensive information regarding AR technology, the KAISAREIA-AR application, and the functionality of a wearable AR device. Subsequently, participants engaged with the application (depicted in Figure 18), after which their responses were collected via an online survey.

5.3. Reliability Analysis

Reliability analysis of the survey responses was conducted using Cronbach’s Alpha, a common statistical measure for internal consistency. The overall Cronbach’s Alpha value for the study was 0.935, indicating a high level of reliability [38] (

Table 3. Reliability Analysis Results.

Construct	Cronbach’s Alpha
Performance Expectancy	0.892
Effort Expectancy	0.783
Facilitating Conditions	0.775
Hedonic Motivation	0.854
Behavioral Intention	0.835
Overall	0.935

). This suggests that the survey items were consistent in measuring the constructs.

Corrected Item–Total Correlations ranged from 0.537 to 0.819, indicating strong contributions of individual survey items to their respective constructs. Cronbach’s Alpha values remained stable even when individual items were removed, further validating the instrument’s reliability.

These high values suggest that the survey items were consistent and reliable in measuring the intended constructs.

5.4. Structural Equation Modeling (SEM)

Path analysis was conducted using SEM in SPSS v27. AMOS to test the hypotheses. The model fit indices, including Chi-square (X²), Degrees of Freedom (df), RMSEA, and SRMR, indicated acceptable fit according to standard criteria [39].

The results of the Structural Equation Modeling (SEM) analysis are presented in Figure 19 to visualize the model’s capacity to evaluate the relationships between key hypotheses. The figure summarizes the relationships between Performance Expectancy (PE), Effort Expectancy (EE), Facilitating Conditions (FC), Hedonic Motivation (HM), Behavioral Intention (BI), and Usage Behavior (UB), as well as the direct and indirect effects among these variables. Notably, the path coefficients and the direction of the relationships provide significant visual evidence supporting the study’s main findings. This path diagram, along with the model fit indices, offers a detailed overview of the accuracy of the hypotheses and the strength of the connections between the variables.

In the path analysis, the path coefficients and their significance levels (p-values) between the factors were examined. The Chi-square (X²) value of the model was calculated as 285.6, with 156 degrees of freedom (df). This X² value falls within the acceptable range of 0 to 2 times the degrees of freedom (0 ≤ X² ≤ 2df), meeting the criteria for validity as established in the literature [38]. Additionally, several fit indices commonly used in structural models were evaluated alongside the path coefficients and their significance. These indices, which reflect the structural model’s alignment with established benchmarks, are presented in detail in

Table 4. Fit Indices of the Model.

Fit Index	Acceptable Range	Model Value
X2	0 ≤ X2 ≤ 2df	285.6
df	—	156
X2/df	0 ≤ X2/df ≤ 2	1.831
RMSEA	≤0.05	0.092
SRMR	0.05 ≤ SRMR ≤ 0.10	0.0801
NFI	0.87 ≤ NFI ≤ 1	0.841
CFI	0.95 ≤ CFI ≤ 1	0.892
GFI	0.90 ≤ GFI ≤ 1	0.803

.

These results suggest that the model fits the data well, with most fit indices falling within acceptable ranges.

5.5. Results

The application of Structural Equation Modeling (SEM) allowed for rigorous hypothesis testing and an in-depth analysis of the interrelationships among variables. The results of the analysis, including path coefficients, p-values, and the outcomes for the five proposed hypotheses, are detailed in

Table 5. Hypothesis Testing Results.

Hypothesis	Path	Path Coefficient	p-Value	Result
H1	PE → BI	0.49	<0.05	Supported
H2	EE → BI	0.80	>0.05	Not Supported
H3	FC → UB	0.429	<0.05	Supported
H4	HM → BI	0.46	<0.05	Supported
H5	BI → UB	0.583	<0.05	Supported

. The findings indicate that Performance Expectancy (H1) and Hedonic Motivation (H4) significantly positively influence Behavioral Intention, whereas Facilitating Conditions (H3) and Behavioral Intention (H5) exert significant positive effects on Usage Behavior. Conversely, Effort Expectancy (H2) was found to have no statistically significant impact on Behavioral Intention. These results highlight the critical importance of performance-related factors and user enjoyment in driving the adoption of augmented-reality (AR) applications, while ease-of-use perceptions appear to play a less influential role in this context.

Of the five hypotheses tested, four were supported. Performance Expectancy (H1) and Hedonic Motivation (H4) emerged as significant predictors of Behavioral Intention, with path coefficients of 0.49 and 0.46, respectively (p < 0.05). Additionally, Facilitating Conditions (H3) and Behavioral Intention (H5) demonstrated strong positive influences on Usage Behavior, with path coefficients of 0.429 and 0.583, respectively (p < 0.05). However, Effort Expectancy (H2) had no statistically significant effect on Behavioral Intention (p > 0.05).

The findings suggest that the adoption of the KAISAREIA-AR application is primarily driven by Performance Expectancy, Hedonic Motivation, and Behavioral Intention, with Facilitating Conditions playing a crucial role in shaping Usage Behavior. Conversely, the minimal impact of Effort Expectancy indicates that perceived ease of use is not a decisive factor for user intention in this specific context.

• SEM Analysis Findings by Hypothesis

• H1: Performance Expectancy (PE) → Behavioral Intention (BI): Performance Expectancy significantly impacts Behavioral Intention (β = 0.49, p < 0.05). Users are more likely to adopt the application if they perceive it will enhance their experience.
• H2: Effort Expectancy (EE) → Behavioral Intention (BI): Contrary to initial assumptions, Effort Expectancy does not significantly influence Behavioral Intention (β = 0.80, p > 0.05), suggesting that ease of use is not a critical adoption factor.
• H3: Facilitating Conditions (FC) → Usage Behavior (UB): Facilitating Conditions, such as resource availability and support, positively influence Usage Behavior (β = 0.429, p < 0.05), emphasizing the need for a supportive technological environment.
• H4: Hedonic Motivation (HM) → Behavioral Intention (BI): Hedonic Motivation strongly predicts Behavioral Intention (β = 0.46, p < 0.05). Users’ enjoyment and the pleasurable aspects of the AR experience play a key role in their intention to use the application.
• H5: Behavioral Intention (BI) → Usage Behavior (UB): Behavioral Intention significantly affects Usage Behavior (β = 0.583, p < 0.05). Higher user intention to adopt the application correlates with increased engagement.

• Key Insights and Implications

The SEM analysis based on the UTAUT framework underscores the factors influencing the adoption of AR technologies in cultural heritage. The main findings are in

Table 6. SEM-based evaluation of user responses according to the UTAUT model.

UTAUT Construct	Empirical Finding	Interpretation
Performance Expectancy	Strong positive effect on Behavioral Intention	Users adopt AR tools when they perceive clear usefulness
Hedonic Motivation	Significant positive influence	Enjoyment enhances the willingness to engage with AR
Facilitating Conditions	Positive effect on actual use	Institutional and technical support is essential
Effort Expectancy	Statistically insignificant in this context	Ease of use is not a primary factor in this implementation

.

Despite participants’ architectural background, post-application responses indicated a notable enhancement in their understanding of Kayseri Castle’s specific historical transformations. The immersive nature of AR allowed users to experience not only structural features but also contextual narratives of different periods, fostering a deeper awareness of the castle’s evolving cultural identity.

To enhance adoption, future research could explore strategies for increasing the relevance of Effort Expectancy by improving perceived ease of use. Addressing these factors could optimize user acceptance and engagement with AR applications in similar cultural heritage settings.

5.6. Constraints and Assumptions

The sample included 100 architecture students from Erciyes University, chosen for their familiarity with cultural heritage. While the homogeneity of this sample ensured well-informed responses, it inherently limited the generalizability of the findings to broader, more diverse populations. Future studies should include participants from diverse demographic and professional backgrounds to ensure broader generalizability and external validity of the findings.

5.7. Characteristics and Limitations of the Sample

Technical Proficiency: The architecture students demonstrated high familiarity with technical terminology and concepts related to cultural heritage and AR technologies. This technical competence likely facilitated their adoption of the application and positively influenced outcomes related to performance expectancy and user experience.

Emotional and Aesthetic Dimensions: Broader user groups may exhibit diverse cognitive and emotional responses, particularly in the context of hedonic (pleasure-driven) motivation, potentially leading to variance in the observed results.

Perceived Ease of Use: The students’ prior exposure to AR technologies and 3D modeling may have positively affected their perception of ease of use. In contrast, less technically experienced user groups may report lower ease-of-use perceptions, influencing behavioral intention and adoption rates.

These limitations suggest that the findings primarily reflect a specialized user demographic and may not be directly applicable to other populations. Comprehensive testing involving varied demographic groups is essential to validate and generalize the results. For instance, individuals with limited technical expertise may perceive ease of use differently, potentially altering their behavioral intentions and adoption patterns.

5.8. Methodological Reflections

The study adopted a quantitative methodology underpinned by the UTAUT framework, enabling the systematic analysis of variables and relationships. While this approach provided robust and structured data, integrating qualitative methods in future research could yield deeper insights into the nuanced emotional and aesthetic dimensions of user experiences. The inclusion of control groups would further enhance the rigor of comparative analyses, offering clearer evaluations of AR’s efficacy relative to traditional cultural heritage presentation methods.

5.9. Assumptions and Data Validity

Key assumptions included the accuracy of participants’ self-reported data and their comprehension of survey instruments. These factors were critical to ensuring the validity, reliability, and interpretability of the findings. Addressing these assumptions rigorously in future studies will further strengthen the robustness of the research outcomes.

6. Conclusions

The advancement of technology has introduced new interactive applications across various fields, enabling the creation, documentation, and efficient dissemination of digital information networks. This progress has also paved the way for the integration of such applications into cultural heritage sites. Augmented-reality (AR) technologies, in particular, offer a unique opportunity to revive historical structures that have been lost, altered, or transformed, while simultaneously enabling real-time interaction with users.

In this study, Kayseri Castle, a site of great historical and cultural significance to the city of Kayseri, was modeled in 3ds Max for its Byzantine, Seljuk, and Ottoman periods. The models, developed based on restitution drawings by Mahmut Akok and Albert Gabriel, were enriched with visual and auditory historical information in Unity and subsequently transferred to an AR environment. The resulting KAISAREIA-AR application, accessible via wearable AR devices, provided users with an innovative spatial experience.

For evaluation, the study utilized the Unified Theory of Acceptance and Use of Technology (UTAUT) and Structural Equation Modeling (SEM), both widely employed in disciplines such as information technology and sociology. These frameworks analyze the factors influencing technology adoption, allowing for a structured evaluation of the application’s parameters. Reliability testing using Cronbach’s Alpha yielded a value of 0.935, indicating high reliability (above the 0.80 threshold). Hypotheses were tested through path analysis in SEM, with four out of five hypotheses supported. Fit indices, including X², df, p, RMSEA, and SRMR, were within acceptable ranges, while NFI, CFI, and GFI values were close to acceptable thresholds, indicating overall model validity.

This study was guided by three key research questions. The findings enable us to answer them as follows:

RQ1: How can augmented-reality (AR) technologies enhance visitor experiences while preserving the authenticity of cultural heritage sites?

The KAISAREIA-AR application provided an immersive and layered representation of Kayseri Castle’s architectural evolution. Participants reported a deeper understanding of the site’s cultural and historical significance, while the digital format maintained the physical integrity of the monument, fulfilling both experiential and preservation objectives.

RQ2: In what ways can wearable AR devices increase public awareness of architectural conservation and cultural heritage?

User testing revealed that wearable AR devices offered intuitive access to complex historical data. The multisensory interaction enabled users to connect emotionally and cognitively with the site, thus enhancing their awareness of architectural conservation principles and the importance of heritage continuity.

RQ3: How can the implementation process for Kayseri Castle serve as a replicable model for similar heritage sites?

The interdisciplinary methodology—combining archival research, 3D modeling, and UTAUT-based user evaluation—offers a replicable framework for other multi-layered heritage sites. The project demonstrated that localized cultural narratives can be effectively translated into digital experiences without compromising scholarly rigor or authenticity.

Compared to widely used AR heritage applications such as Google Arts & Culture or Nomad, which offer generalized, often image-based or curated experiences, Kaisareia-AR provides an immersive, site-specific interaction that is spatially and temporally layered. Unlike TimeLooper or StreetMuseum, which present discrete reconstructions of historical scenes, our tool enables users to navigate through evolving architectural phases (Roman to Ottoman) in a continuous spatial interface. Its pedagogical design—developed in collaboration with architecture educators—further differentiates it by focusing on historical stratification, material transitions, and conservation awareness in heritage interpretation.

The KAISAREIA-AR application, designed for integrating cultural heritage sites into AR environments, was tested with a limited user group. Future studies should expand the application to include diverse demographic groups, broadening its accessibility and impact. Showcasing this application at Kayseri Castle, now functioning as an archaeological museum, would contribute to its dissemination and further engage visitors in understanding the castle’s historical transformation and architectural evolution. Such efforts are expected to enhance the transmission of historical information to visitors and emphasize the importance of architectural conservation.

Future research should focus on enhancing the application’s usability and interactivity, including features such as multilingual support and broader device compatibility. Furthermore, exploring the adaptation of AR technology in different geographical and cultural contexts, with a focus on cultural diversity and local customization, would provide valuable insights. Ultimately, preserving and promoting cultural heritage demands a multifaceted approach, incorporating innovative tools like AR technology, continuous research, educational programs, and effective policy development.

The integration of AR into cultural heritage preservation represents a promising step forward in making history accessible, engaging, and educational for future generations. By leveraging innovative technologies, such as wearable AR devices and 3D modeling, this research contributes to the broader goals of cultural conservation and public education. The KAISAREIA-AR application not only enhances the preservation of Kayseri Castle but also sets a precedent for future applications of AR in the cultural heritage sector.

As technology continues to evolve, it is essential to explore and refine these tools to ensure that cultural heritage remains relevant and accessible to diverse audiences worldwide. Through continued research, development, and collaboration, AR can play a pivotal role in the future of cultural heritage preservation.