Augmented Reality in Cultural Heritage: A Narrative Review of Design, Development and Evaluation Approaches

Abstract

Cultural heritage is important for every group and society, as it represents a part of their identity and helps preserve traditions, along with significant tangible and intangible elements. These elements are not only passed on to future generations but also play a role in education today. Many studies have examined digital applications that promote cultural heritage, particularly those that use Augmented Reality (AR) technology. However, few have systematically examined the methodologies used in the design, development and evaluation of AR applications for cultural heritage. This narrative review addresses that gap by analyzing 45 papers published between 2008 and 2024 from the Scopus and Web of Science databases. The review identifies the main motivations for AR applications, such as enhancing the learning experience, promoting tourism and the use of gamification. The majority of these applications concentrate on tangible cultural heritage, such as historical buildings and cultural routes, while intangible heritage remains less explored. In most cases, AR applications were developed with a focus on user experience, using various development tools and showing different levels of technical complexity. Although many applications initiated evaluation processes, few completed them thoroughly. These evaluations vary widely, ranging from usability tests to pre-/post-tests. The results generally report positive impacts in terms of learning and user experience. This review offers a structured framework and useful insights for researchers regarding the design and evaluation of AR applications and helps identify research gaps and limitations for future work.

1. Introduction

Cultural heritage includes all the tangible and intangible elements of a society inherited from previous generations and preserved for future generations, as defined in EU policy for cultural heritage. Tangible cultural heritage includes natural, built and archeological sites, museums, monuments, artworks and historic cities. Intangible cultural heritage includes elements that cannot be “touched”, such as knowledge, myths, customs, oral traditions, dances, music and songs, skills and techniques that represent traditional folk culture [1].

Cultural heritage enriches the lives of people and serves as a driving force for cultural and creative sectors. In recent years, digital transformation has clearly impacted the education sector, leading to increased demands for educators to address new challenges. Within this context, the culture and heritage of local communities play an important role, with growing efforts to promote them [2]. Various digital technologies have been employed to present and promote cultural heritage to new generations, including Augmented Reality (AR), virtual reality (VR), geographical information systems (GIS), 3D modeling, digital storytelling, serious games and social media platforms [3]. These technologies provide immersive, rich experiences, increase accessibility for visitors and offer valuable learning opportunities for all ages [3].

Several review studies have examined digital applications in cultural heritage with particular emphasis on those using AR technology [4,5,6]. These studies have typically focused on the importance of AR for cultural heritage, its advantages and disadvantages and the identification of application areas such as education, museums and cultural tourism. However, there is a lack of review analysis of the methodologies used to design, develop and evaluate AR applications for cultural heritage. This gap limits our understanding of how to create effective, educationally meaningful and culturally sensitive AR experiences.

The purpose of our study is to address this gap by analyzing the current state of AR applications for cultural heritage in order to understand their motivations and target cultural contexts and the methodologies used in AR design, development and evaluation. This research, compared to previous reviews that concentrated on technologies, application domains or case studies, highlights the methodological aspect, specifically the design, development and evaluation of AR applications in cultural heritage. Previous reviews have addressed some methodological aspects, yet without a systematic comparison. The contribution of this work is a systematic framework that offers insights to guide researchers when designing and evaluating future AR applications. Additionally, this paper extends and revises our preliminary conference paper presented in 2023 [7]. It presents and systematizes the findings in a unified way, incorporating new studies and AR applications published between 2023 and 2024. The findings of this review can provide a useful framework and insights for researchers regarding the design and evaluation of AR applications and help identify research gaps and limitations for future research.

This paper is organized as follows: Section 2 reviews related work. Section 3 presents the methodology and research questions. Section 4 presents the results of the research questions, organized into categories such as motivation, origin, design and development methodologies, technical features and evaluation approaches. Section 5 discusses the findings. Section 6 provides suggestions for further research, while Section 7 concludes with limitations and final remarks.

2. Conceptual Background and Related Work

In this section, we present the key concepts relevant to our research and review the existing literature on AR applications in the cultural heritage field.

2.1. Basic Concepts

This subsection covers technologies commonly used to present and promote cultural heritage. The main technologies include immersive technology (or extended reality) and serious games. Immersive technology is a general term that includes augmented, virtual and mixed reality technologies. The following paragraphs provide definitions of these technologies.

Augmented Reality (AR). Augmented Reality overlays simulated elements onto the physical world such as videos, graphics or location-based data to enhance users’ perception of their environment [3].

Virtual Reality (VR). Virtual Reality has been used in the field of cultural heritage, describing a variety of applications involving visual, 3D environments where the user is “immersed” in an entire simulated world [3].

Mixed Reality (MR). Mixed Reality combines the user’s real-world environment with digital content, allowing both to coexist. Virtual objects behave as if they are in the real world, e.g., they are obstructed by physical objects, their lighting is proportional to real light sources in the environment, and they sound like they are in the same space. As the user interacts with the real and virtual objects, the virtual objects change in space as they would if they were physical [8].

Serious Games. Serious Games are designed with educational goals and have been applied in various cultural heritage contexts and platforms, following different design philosophies. They are designed for both formal learning environments, such as the classroom for students of various ages and levels, and informal learning environments, such as at home, on tourist trips or during heritage visits.

These games include trivia questions, puzzles, mini-games, interactive exhibitions and mobile apps in museums or tourist attractions using reward mechanisms. They also feature simulations of past events (such as the Battle of Waterloo) in adventure games and role-playing games set in realistic digital reconstructions of actual locations [9].

2.2. Related Work

Previous reviews have studied the application of AR technology in cultural heritage from different perspectives. These reviews provide useful insights but often focus on high-level thematic trends without addressing the practical aspects of how AR applications are designed, developed and evaluated.

Table 1. Comparison of related review studies in AR and cultural heritage.

Study/Year	Focus Area	Methodology	Contribution Type	Limitations
Boboc et al. (2022) [4]	AR research trends and topics in CH	Bibliometric analysis	Trend mapping	No analysis of implementation, user groups, or outcomes
Vargas et al. (2020) [5]	Motivation in AR for CH education	Thematic review of studies	Educational impact	Focused only on student motivation, not design process
Bekele et al. (2018) [6]	AR/VR/MR technologies in CH	Technical review	Hardware/architecture	Technological focus; limited design/evaluation aspects
Our study	Design, implementation and evaluation of AR in CH	Narrative review	Practical methodology guide	Focus on tools, usability and evaluation for practitioners

summarizes the review studies in AR and cultural heritage.

Boboc et al. provided an overview of AR use in cultural heritage between 2012 and 2021, analyzing papers published in Scopus and Web of Science databases. They performed a bibliometric analysis using tools such as ScientoPy and VOS Viewer to identify major research trends, including 3D reconstruction, digital heritage, virtual museums, user experience, education, tourism, intangible cultural heritage and gamification [4]. This mapping of the scope and evolution of the field is valuable, but this study is statistical in nature and does not examine how AR applications are actually designed, used, or evaluated in real-world educational or cultural settings.

Vargas et al. focused on the use of AR technology in cultural heritage education and its potential to enhance student motivation. They also examined whether AR can motivate students by providing stimuli that encourage engagement in activities without concern for duration or difficulty. Above all, their paper addresses the advantages, disadvantages, limitations and challenges of AR technologies for cultural heritage, investigates whether AR improves learning motivation and explores the characteristics such applications should have to support learning [5]. However, the scope is narrow, concentrating mainly on student engagement, without exploring broader methodological questions such as AR content creation, technical deployment or evaluation processes.

Finally, Bekele et al. examined essential aspects and the state-of-the-art in augmented, virtual and mixed reality from a cultural heritage perspective, while also describing research on the development of applications and systems. Furthermore, they summarized the adopted technologies and application areas of these studies, highlighting the limitations of existing technology, and aimed to identify which solutions are most appropriate for a given application [6]. However, their review provides limited practical guidance on educational design methodologies for non-technical developers.

In contrast to these reviews, our paper focuses on a less-explored yet practically critical dimension: the methodologies used to design, implement and evaluate AR applications in cultural heritage, particularly for non-expert users such as educators and local communities. It explores not only what has been developed but also how it was created, for whom, using which tools and under what constraints. In doing so, it provides a framework for future developers of AR applications in cultural heritage, particularly those without advanced technical expertise.

3. Methodology

The research methodology used in this paper was a narrative review to examine research on AR applications for promoting cultural heritage. We chose this type of review to conduct a deeper qualitative, rather than quantitative, analysis. Moreover, we would like to highlight details related to the design, development and evaluation of the applications, as well as their pedagogical and social dimensions. Thus, the article is not limited to a statistical depiction of trends, but produces a framework of guidelines useful to researchers and professionals who do not necessarily have technical expertise. The review workflow involved four key steps: formulating research questions, conducting a literature search, describing the selected studies and extracting, analyzing and synthesizing the findings.

3.1. Research Questions

The goal of this review is to investigate the motivations and types of cultural content emphasized, whether they are artistic forms like music and songs or architectural forms, such as churches, mansions, traditional settlements, etc. It also examines the design, implementation and evaluation methodologies used in AR applications for cultural heritage. Therefore, this review formulates the following research questions:

RQ1: Why are AR applications for cultural heritage being designed? What is their purpose and motivation?

RQ2: Which countries and cultural contexts do the applications address?

RQ3: What are the current design methodologies for AR applications for cultural heritage?

RQ4: What are the current methodologies for developing and implementing AR applications for cultural heritage?

RQ5: What technical characteristics (e.g., device, operating system, tracking and registration technique, etc.) are required to run AR applications for cultural heritage?

RQ6: How are current AR applications for cultural heritage evaluated? What are the main research findings?

3.2. Literature Search

To answer the research questions, we retrieved the relevant articles from bibliographic databases such as Scopus and Web of Science. The keywords used were as follows: ((augmented Reality OR AR) AND cultural heritage) in the topic (that is, title, abstract and keywords). These keywords were selected to identify papers specifically related to AR applications for cultural heritage. We limited the search to English-language articles, document types (journal and conference papers) and the time period up to 2024. This expands on our previous review [7], which included articles only up to early 2022.

Papers were selected based on their relevance to mobile and AR applications for educational purposes in cultural heritage, and the inclusion of design and/or evaluation elements.

In the databases Scopus and Web of Science, by using the keywords and time period that we mentioned before, we found 1567 articles in the Scopus database and 636 articles in Web of Science for the time period 2008–2024. Out of a total of 2203 articles, we removed the duplicated studies, of which there were 285. The main exclusion criteria included language (non-English language) and document type (anything other than article and conference paper). Based on these criteria, we discarded 322 articles from Scopus and 121 articles from Web of Science. In total, 443 articles were excluded from the study. Then, 1475 articles were sought for retrieval. By reviewing the titles, abstracts and full texts of these articles, we excluded 1430 studies that are not about creating mobile and AR applications for educational purposes of cultural heritage, resulting in 45 articles.

In total, 1475 papers were examined and 45 were selected for further analysis, as they met the selection criteria.

3.3. Literature Studies Description

This review includes 45 papers for analysis. These papers cover the period from 2008 to 2024 (representing the number of whole year’s publications). Figure 1 shows the distribution of papers published each year.

The findings in Figure 1 show that the number of AR applications was relatively low in earlier years, with few papers published before 2016. In contrast, recent years have seen a significant increase in published papers, especially from 2020 to 2024. This suggests a growing interest in AR applications for cultural heritage, which in turn has spurred increased research activity.

Finally, most papers were published in journals (27), with the remaining 18 being from conferences.

3.4. Data Extraction, Analysis and Synthesis

In this phase, data were extracted from the selected papers on the motivations for AR applications, cultural content and origin, design methodologies, development platforms, technical features and evaluation methodologies. The extracted data were then analyzed and synthesized through content analysis to identify the main findings for each research question, as well as research gaps and limitations. In addition to qualitative synthesis, quantitative descriptions (frequencies and percentages) were used for some research questions.

4. Results

In the following subsections, findings are presented and interpreted based on the defined research questions. For reasons of clarity, the percentages of each category are presented out of a total of 45 studies.

4.1. Motivation of AR Applications

The first research question addresses the purpose and motivations for which AR applications were developed in the context of cultural heritage. Based on this criterion, the studies were classified into seven categories, discussed below:

Promotion of cultural stock for public information and tourism development: A primary motivation for developing AR applications in cultural heritage is to promote cultural resources for public information and tourism development [10,11,12,13]. Many applications aim to promote hard-to-access heritage areas [14,15,16], such as underwater cultural heritage [14], in almost 35.5% of the studies reviewed. These target both specialized tourists (e.g., divers) and non-specialists, aiming to highlight cultural resources and support tourism development in coastal areas and islands. For example, at Borobudur Temple in Indonesia, an app was designed to inform tourists about the stories depicted on the temple panels [17]. Similarly, another application promotes the archeological area of Museo Diffuso Castello d’ Alceste in Lecce, Italy [18].

Educational enhancement and social interaction: Another important motivation is to support innovative approaches in education [10,12,19,20] (28.9%). Specifically, Hincapie et al. [19] used AR and GPS technology to revive cultural heritage around Medellín’s Cisnero Square (also called the “Park of Lights”), studying the positive effects that new technologies have on the process of learning and assimilating new knowledge [19]. Ch’ng et al. [20] designed a mobile AR application enabling multi-user interaction, suggesting that discussions about heritage objects can extend beyond the initial group. Their study also investigates how perceived realism and artificial objects influence communication, and identifies which Social AR features are most effective [20].

Preservation of tradition and cultural identity: Some applications aim to promote the traditions and cultural identity of a group or a nation, representing 17.8%. For example, Weking et al. [21] developed an AR mobile phone application to promote Indonesian traditional food, helping preserve traditions against western influences and pass them on to future generations. Aziz et al. [22] designed an application to promote the multicultural heritage of regions like Pattani, Thailand and Terengganu, Malaysia, encouraging users to visit and engage with these places.

In the field of intangible cultural heritage, Hongning et al. [23] created three mini-games about the history of Peking Opera for young children, combined with some VR elements to teach them the importance of Peking Opera and its characters, along with learning characters’ movements and dancing skills, which make them feel as if they are participating in an opera performance. Tan et al. [13] developed a prototype story game app, “The Story of Praya Lane”, that combines the use of mobile sensing, machine learning and Augmented Reality (AR) to engage users in physical mobile activities as part of a gamified, non-linear storytelling application. This supports constructivist learning of the Kristang culture, from requirement gathering through to implementation.

Environmental awareness and sustainability: Some studies emphasize environmental education and sustainability (8.9%). They chose to use AR applications to raise awareness about the environment, while they emphasize how these applications can be helpful in learning about nature and its flora and fauna. Postolache et al. [24] and Kleftodimos et al. [25] have an environmental orientation, as they have applications about the environment. More specifically, Postolache et al. [24] developed an AR application about the Botanical Gardens in Lisbon, Portugal, by organizing four tours in the gardens, depending on what the visitor wants to see. In the other case, Kleftodimos et al. [25] designed location-based Augmented Reality (AR) games to guide users to visit and explore the destinations, become informed, gather points and prizes by accomplishing specific tasks, and meet virtual characters that tell stories. This approach helps visitors identify destinations in the mountainous regions of Western Macedonia, supporting tourism promotion.

Innovative engagement with cultural resources: Kratchanov et al. [26] propose a design for a serious game based on an AR representation of a mural artwork in the National Library “Ivan Vazov”, in Plovdiv, Bulgaria. In the center of the mural, we find the famous Bulgarian author Ivan Vazov, who is surrounded by fictional characters that exist in his literary works. A head-mounted display would be used in front of the mural by the users and they would be able to see the characters moving, while they would have the opportunity to interact with them in a gamified AR environment [26]. Although still under development, this project could offer a novel way for library visitors to engage with literature and art, and will be used for the promotion of library resources and cultural heritage. In this way, it would combine intangible cultural heritage (Ivan Vazov’s works) and tangible cultural heritage (mural artwork). Similarly, Tian et al. [27] designed “PoeticAR”, an AR application that presents traditional Chinese poems based on physical scenery of the Jichang Garden, in Jiangsu Province, China, to enhance tourists’ cultural and esthetic experience. A total of 8.9% of the studies adopt this innovative approach.

In conclusion, these motivations show an increased recognition of AR not just as a technological advancement but as a tool for transforming cultural communication and engagement. It should be noted that many of the papers we have mentioned do not belong to only one group, such as the case of Kurniawardhani, A. et al. [17], which has been motivated by the promoting of cultural heritage to tourists and visitors of the site, but also tries to educate them about this specific cultural identity. Moreover, it is obvious that the development of AR applications with goals such as preserving intangible heritage, promoting environmental sensitivity and facilitating social interaction is also gaining momentum.

We can identify two main strategies that are followed in order to support the basic motivations, like education, tourist promotion and environmental awareness; the gamification used for cultural preservation and the virtual representation both used often in the designing of the AR applications.

Gamification for cultural preservation: Only a small number of papers focus on gamification applications, but several notable examples are discussed here. Koutsabasis et al. [11] present a mobile learning game to promote and preserve the industrial production of oil in Lesvos as cultural heritage. Koutromanos et al. [28] describe a game with AR and GPS technology that presents traditional old buildings (tobacco warehouses) of Agrinio, a city in Western Macedonia, motivating primary school students to explore these buildings from different perspectives: historical, architectural and cultural value, as well as their relationship with the economic and cultural development of the city. Staying within the same area, West Macedonia, we can see one more study that stands out. Kleftodimos et al. [29] used TaleBlazer to create two serious games about the Dispilio settlement, tested with both school and university students.

Virtual representation: Many mobile applications mix real-world settings with virtual representations of buildings or routes. For example, Flores et al. [16] used AR technology to virtually reconstruct war-damaged and unrepaired buildings on Corregidor Island, Philippines. Similarly, the “Walk1916” app offers a mobile walking tour of Easter Rising sites in Dublin, Ireland [30]. It integrates digital surrogates from archival collections, with audio and text descriptions delivered via AR and GPS technology. De Luca et al. [31] proposed a virtual reconstruction of the church of Sant’ Elia in Ruggiano, in the southern province of Lecce, demonstrating the value of Mixed Reality in expanding their results from their previous research Kurniawardhani, A. et al. [17] designed an application that can facilitate tourists accessing the relief story directly when they see the panel, during their visit to the Borobudur temple. Virtual representation also facilitates visualization of lost or damaged heritage elements, exemplified by the Tang Dynasty tomb murals [32].

Consequently, it is obvious that most of the designed applications are motivated by the promotion of cultural stock for public information and tourism development, while the second group of applications are motivated by the idea of educational enhancement and social interaction. On the other hand, it is quite promising that more and more applications are designed with a different perspective, like environmental awareness and innovative engagement with cultural resources, even though their percentages are still relatively low.

4.2. Origin and Cultural Stock of AR Applications

The second research question examines the regions and countries producing AR applications, as well as their cultural focus. Figure 2 summarizes the countries of origin for the reviewed AR applications. Figure 2 shows that most applications were created in Italy (11 applications), promoting heritage elements and buildings from antiquity and Roman times [10,31,33,34,35,36,37,38,39,40,41].

According to Figure 2, Greece is the next country with a similarly large number of applications, having 10 articles out of the 45 we have studied [11,14,18,25,28,29,42,43,44,45].

The Greek applications focus entirely on tangible cultural heritage [11,42] and cultural reserve [14,18,25,28,43,44] with the aim of promoting it to the general public. In the last two years, there has been increased effort to promote intangible cultural heritage in Greece, as seen in papers [18,45], and in work combining flora and fauna with the history and culture of Greece’s mountainous regions [25,29]. China follows with six applications (application in paper [46,47] is the same, but they describe the whole procedure in more detail in [47]), Malaysia with five applications, Indonesia with two applications and the rest of the countries that mentioned have one application each. Beyond these regions, the Cultural Heritage Project involves a collaboration between Belgium and Greece, exploring AR’s role in public history and cross-European heritage communication [48].

In addition to the country origin of AR applications, the applications were also classified according to the focus of the cultural stock. Figure 3 shows that 86.7% (39 applications) focus on tangible cultural heritage, while 13.3% (6 applications) address intangible cultural heritage.

Tangible cultural heritage: The reviewed applications focused on six categories of tangible cultural heritage, such as big cities and areas, buildings, museums, environment, archeological sites and paintings. These categories are discussed below. In the first category of tangible cultural heritage, one of the two largest groups covers territories and entire cities, comprising 23.25% of applications: 10 out of 43. Note that applications in [25,46] are repeated in [29,47], respectively. We can find applications for whole cities, such as the history and cultural heritage of Nis in Serbia [49], the reactivation of cultural heritage near the square “Medellin’s Cisnero” in Columbia [19] and the promotion and conservation of cultural heritage in Bosnia-Herzegovina [50]. In addition, we can observe the promotion of whole cities, like Melaka in Malaysia [51] and Chania in Greece [43], while Tan et al. created a gamified application about historical sites in Malaysia, aimed mainly at young students [52]. Pietroni [36] presents an application about Matera, a city identified as a cultural heritage monument by UNESCO since 1993, and Garoufallou et al. designed a digital cultural map for the city of Veria in Greece, which contains information about the history and culture of the city and the surrounding area [44]. In this category, we can see the application of the serious game GPS-guided Augmented Reality, which promotes the Renaissance village of Caldana, Italy [40], and Tan S.N. et al.’s paper designed a prototype story game app, “The Story of Praya Lane” that combines the use of mobile sensing, machine learning and Augmented Reality (AR) to engage users in physical mobile activities as part of a gamified, non-linear storytelling application [13].

The second big group has to do with buildings with 18.6% of all applications. This group includes applications about fortresses, like this one in Melaka, Malaysia [15] and Hwaseong Fortress in South Korea [53], a tourist guide application for Corregidor Island, allowing visitors to see virtually restored buildings that were destroyed during the war [16]. Another example is “Walk1916” as a navigator app, which presents Easter Rising suites in Dublin, Ireland [30], while the application of Koutromanos et al. talks about old tobacco warehouses in Agrinio [28]. In the application of Cancianni M. et al. they refer to villa Adriana in Tivoli, Italy [35] and finally, in De Luca V. et al., we can see the visual reconstruction of church Sant’ Elia, in Ruggiano, Italy [31]. Dutra et al. [48] created two applications to promote cultural heritage in different cities [48], featuring two distinct contents: one has to do with “La Grand Place”, an official World Heritage Monument in Brussels, and the second has to do the Greek mythology content regarding the region, such as Jason and the Argonauts and The Centaurs of Pelion, among others.

A smaller category (11.6%) focuses on larger areas and territories with an environmental theme. These include applications for tangible underwater cultural heritage and protection of the water environment [14], the botanical gardens in Lisbon [24], applications and serious games that promote the mountainous areas of Western Macedonia, Greece, and want to educate visitors on various aspects of these destinations, such as their history and cultural heritage [25], and finally, an application and a serious game that combine the natural environment and archeological site to enhance the educational process [29]. The AR mobile application in [22] aims to enhance the tourist experience by providing immersive and informative content specific to the culture and heritage, promoting tourism in Southeast Asia, specifically Thailand and Malaysia.

Another group is archeological sites with four applications (9.3%). We can observe the promotion of cultural heritage in Italy, by creating a database about the catacombs [34] and the promotion of the Anfiteatro Campano in Santa Maria Capua Vetere [37]. Also, Cutri et al. promote and enhance archeological visits in Locri, one of the most important cities in Calabria [38], while Xu et al. created an exploration game about the ruins of Arhat Monastery and the Twin Pagoda (Shuangta) [46,47].

There are also applications about drawings, frescoes and mural artworks, like papers [10,39] which are about guiding visitors in the interpretation of the frescoes inside the Basilica of Saint Catherina of Alexandria in Galatina in Italy. Kratchanov et al. [26] propose a serious game concept (still under development), based on an Augmented Reality representation of a mural artwork located in the National Library “Ivan Vazov” in Plovdiv, Bulgaria. In [40], they have designed a “treasure hunt” serious game in the Renaissance village of Caldana, in order to engage the alpha generation in urban cultural heritage, while in [32] the designed application is used to restore damaged murals in Zhao Yigong Tomb.

In the final group of tangible cultural heritage applications and serious games (11.1%), we can also find applications about objects exhibited in museums like astronomical tools in the Museo Astronomico di Brera located in Milan [33], the “greek type” of watermill in Lesvos, Greece [42] and the King’s Chinese Cabinet (Museum of King Jan III’s Palace at Wilanów, Warsaw, Poland) [54]. The application by Ch’ng et al. [20] is different, as it allows multiuser interaction and proposes that discussions that occur around cultural heritage objects can be replicated beyond groups. It uses examples from various museums, such as a gilt bronze bull from Ningxia Museum, a Vajrasattva bronze statue from Sichuan Museum and Queen from the Lewis Chessmen, from the British Museum [20]. In the case of [41], they have designed a “diachronic” application for the enhancement of the archeological site Museo Diffuso Castello d’ Alceste in Italy.

Intangible cultural heritage: Intangible cultural heritage has fewer applications; however, they are extremely interesting. This category contains applications for the promotion of traditional practices, such as the industrial oil production in Lesvos, Greece [11]. In addition, Galani et al. combine architecture with intangible cultural heritage by promoting the stages of leather tanning process, adopting a novel approach that augments 3D content upon a physical scale model of an old tannery. This approach aims to convey the cultural value of traditional craftsmanship to visitors of the building and associate its architectural elements to its history and use [18]. Furthermore, we can see the promotion of dance, in the case of Xi Shi dance [12] and the traditional food of Indonesia [21]. In [45], they created an application about poetry. More recently, Tian J. et al. [27] combine tangible and intangible cultural heritage by enhancing visits to Jichang Garden with traditional Chinese poems. Finally, Hongning et al. present two applications that promote the cultural heritage of Peking Opera and its basic characters [23]. The applications about intangible cultural heritage constitute 16.2% of the applications reviewed.

This analysis reveals a rich and diverse geographical distribution of AR applications. Countries such as Italy and Greece stand out for their high number of projects, reflecting strong academic interest in heritage technologies. The results also indicate that most projects focus on tangible cultural heritage features, such as historical buildings, archeological sites and urban environments. It is worth mentioning that insufficient attention has been given to intangible cultural heritage, including traditional practices, food culture and oral traditions, revealing a research gap. By contrast, many Asian projects emphasize reviving intangible traditions and fostering intercultural understanding.

4.3. AR Application Design Methodologies

The third research question focuses on the design methodologies used by researchers in the development of AR applications. Design is a systematic process aimed at developing AR applications that offer effective interactions and immersive experiences to end users. In design methodologies, the steps followed by most papers start with requirements analysis, cultural content selection and development platform selection. It should be noted that most papers did not explicitly describe the design methodology they followed. However, the design methodologies extracted from the papers can be classified into three main approaches based on their design description: the user-centered design, the design thinking and the social-centered design.

User-centered design: The user-centered design methodology focuses on designing AR applications based on end user interaction requirements or preliminary study by organizing the cultural content appropriately. It also involves an iterative design process with usability tests in order to provide a rich and interactive end user experience. Most studies follow this approach [13,17,18,19,24,25,27,32,33,40,41,47,50]. Some studies of this approach are analyzed below.

In paper [19], the authors begin by reviewing other applications designed for promotion and then introduce their own idea for promotion, by explaining the designing of their application. In Hincapie et al., the main reason for selecting this cultural content is the fact that the geographical center of Medellin (the Cisneros market) has been chosen, since it marked an era of successive changes in the life of the city and acted as the backbone of transformation and change in the bourgeoisie. Then, they identified the key elements that should be included in the app: first of all, characterizing and visualizing the architectural elements of Cisneros Market Square using Augmented Reality technology. Secondly, the ability to describe how important the city’s architectural heritage is in the case study and finally, to propose a technology and cultural heritage experience scenario, as a contribution to the spread of digital through ICT. Based on the above, the authors of the paper [19] created an application called Vitica that uses GPS and Augmented Reality, as well as various formats to transmit information and knowledge related to the historical sites of this place. The application design defines the structure for presenting the content in two dimensions, space and time, which are both necessary for cultural heritage representations [19]. A pilot implementation of the Vitica application was conducted with 48 students, during which the authors observed user interaction and collected usability data. The findings showed that the application was easy to use and supported cultural heritage learning, while future work is suggested to further evaluate individual functionalities of the system (e.g., GPS-guided navigation, content organization) [19].

In another designed application, we observe that during their review of other works, they choose to pay more attention to those that involve promoting botanic gardens [24]. Having the research on those gardens as an inspiration, they continue with the description of their own application, highlighting the different way they set up the app: although they use the separation into themed trails, which was also used by other botanic garden apps, the content of their proposing trails is different, resulting in these four trails: “Must See Trees”, “Garden with History”, “Birds” and “Biosensors”, while they are giving the opportunity to everyone to browse the app: schoolchildren, students, tourists, specialist researchers, families with children. Although the application is based on thematic trails designed for diverse audiences, the study does not explicitly follow a structured user-centered design methodology. Nevertheless, the design approach suggests an implicit awareness of varied user needs.

One more application related to gardens is the “PoeticAR”, an application designed to promote traditional Chinese poems combined with a walk around the Jichang Garden, in Jiangsu Province, China [27]. “PoeticAR” allows users to interactively learn knowledge about traditional Chinese poems within CH sites and enhance the cultural and especially the esthetic experience in tourism through a verse riddle game, multimedia presentation of poems, poem imagery interaction, poem words’ annotation, a brief background introduction to the poems and other relevant poem explorations [27]. “PoeticAR” was built using a user-centered design methodology, which included user research through interviews and workshops with garden visitors, the creation of user personas and interaction demands and the utilization of iterative feedback during prototype testing. These steps ensured that the application’s features and content were aligned with the expectations and preferences of users, thus reinforcing the cultural experience of the application [27].

Another application that has to do with nature surrounding us is [25], where MR applications with gamification features were designed for the promotion of the mountainous areas of the district of Western Macedonia in Greece. The aim of these applications is to promote education among visitors and tourists of the area in subjects such as history and cultural heritage. These applications are designed to guide users in exploring several destinations, learn new information about them and gather points and prizes when they accomplish specific tasks, and they also have the opportunity to meet virtual characters. Τhe authors conducted pilot tests with students and visitors, gathering feedback on usability and interaction. The applications were refined based on these observations, incorporating user responses to improve interaction. Therefore, the design reflects an implicit user-centered perspective, particularly through post-development evaluation and adaptation.

Finally, in the case of the application MARSS (MusAB in Augmented Reality from Science to Society), which was created for use in the Milan Science Museum in Italy, their references are about apps that other museums have developed to showcase some of their exhibits, identifying the shortcoming that many studies do not have quantitative data on the economic impact of using these apps in museums [33]. Then, they set the creation of a “talking” museum collection as the primary goal of the project, which will lead the visitor to the discovery of astronomy, its history and the role that the “Osservatorio” plays for the city of Milan. They choose the Astronomical Instruments gallery, making an AR application (MARSS) where they have created an AR clip for each astronomical instrument in the exhibition in order to explain its history and use. The exhibition needed to be modernized and improved and by keeping in mind the basic question “what does an astronomer do”, the MARSS project aims to design and develop a digital journey inside the museum, allowing different categories of visitors to enjoy the exhibition in an interactive way. The MARSS was created using an iterative method (Research–Design–Develop–Test) that focused on the needs of different types of visitors, such as children, non-expert adults and expert users. User tests were conducted in the museum to evaluate usability and engagement, and the feedback was used to refine the AR clips and ensure accessibility for diverse audiences.

A more recent approach that is similar to the one with the oil production is [18], where they promote the leather tanning process. The application has the following learning objectives for the visitors: (a) to recognize the basic stages that the leather tanning process consists of; (b) to learn about the conditions, materials and tools used in each of these stages; and (c) to acquire an experiential understanding of the process [18]. Their work focuses on the connection between architectural and intangible heritage, using a hybrid physical-digital set up. They examine the approach of presenting the building, “Kornilakis Tanning Factory” in Syros Island, at a smaller scale where animated digital characters demonstrate craftsmanship of the past and communicate the story and heritage of the environment [18]. The application adopted a user-centered design approach, including the collection of user feedback and usability evaluation. They conducted a System Usability Scale (SUS) questionnaire and an AttrakDiff-based user experience evaluation. The results informed refinements of the application, highlighting the integration of user needs into the development process.

Design thinking: Design thinking is a human-centered approach to innovation that considers the needs of people, the capabilities of technology and the requirements for business success. It involves a creative and iterative process that aims to generate innovative solutions by empathizing with users, defining the problem, ideating solutions, prototyping and testing [11,22]. This approach fosters problem-solving rooted in local contexts and ensures solutions are feasible. Previously, we have briefly referred to this application about the oil production in the island of Lesvos, Greece. They aimed to place greater emphasis on the users’ intentions and needs, and the people involved in the project. Finally, they decided that the application’s design should incorporate: (a) the aspect of the exploratory learning, with emphasis on the respect of the museum and its exhibits, as well as settlement landmarks; (b) digital characters who would be used in the storytelling and narrative part of the application; and (c) prizes which would be given to users when they viewed traditional agricultural tools in 3D (digitized models from museum exhibits) [11]. The design thinking was discussed in terms of: (a) action, where they explained the methods and main outcomes, and (b) reflection, where they discussed lessons learnt and made further suggestions [11]. The main method to empathize with the design space was field visits or field trips, which increased team motivation and established cooperation among members and local experts. They reflected on the resource materials in regular meetings with several interwoven themes, such as the possible digitization of museum exhibits (for possible AR functionality), identifying challenges for the mobile games and noting down ideas for UI/UX design [11]. After the identification of user groups and personas that reflected particular user goals and needs, they decided that the design of the mobile game should emphasize: (a) exploratory learning with respect to particular museum exhibits and settlement landmarks; (b) storytelling and narratives about CH, including digital characters; and (c) rewards in Augmented Reality (AR), where users view traditional tools in 3D (digitized models from museum exhibits) [11].

Social-centered design: Another design methodology is the social-oriented approach, which involves designing AR applications based on interactions within a community of users (or social interaction) to democratize and enrich user experiences at cultural heritage sites. A form of social-oriented design can be considered: Social Augmented Reality (Social AR). Social AR refers to the use of AR effects within social media platforms such as Facebook, TikTok, Instagram and Snapchat (https://rbkavin.studio/whatsSocialAR/what-is-social-ar.php (accessed on 12 March 2024)). An example of social media integration can be found in Grammatikopoulou et al. [45], where they proposed a social AR application called ARtful to foster the development of an online learning community for museum staff and visitors. This application also encourages the exchange of information and interactions among its users in order to enhance social learning in museums. This tool was developed using Unity 3D for the creation of 2D and 3D video games, as well as non-game interactive simulations and visualizations. ARtful was also used in Unity’s Vuforia SDK, to recognize images and objects and enable people to interact with virtual and real-life objects overlaid on the real environment [45]. In addition, another application that has to do with Social AR is this of Ch’ng et al., in which they have designed a mobile AR app that supports up to six synchronous users and is designed for real-world use in various locations, with a minimum space of 1 × 1 m for context-aware object placement. This application includes functionalities such as selecting, positioning, scaling, rotating, lifting and deleting objects. Users could double tap on the surface of an object to indicate the area of interest to other users and communicate via real-time voice calls [20].

This analysis shows that there is a variety of design methodologies in AR applications for cultural heritage, each emphasizing different aspects of the user experience. User-centered design focuses on usability and iterative improvement, design thinking encourages innovation through empathy and contextual problem-solving and social-centered design fosters engagement and shared experiences among users. The above examples also show that some design methodologies can be combined with complementary techniques such as gamification [25] or storytelling [17].

4.4. AR Application Development and Implementation Methodologies

This subsection addresses the fourth research question, which focuses on the software development platforms and authoring tools employed by researchers to implement AR applications. Based on data extracted from reviewed studies, there are two main software platforms for developing AR applications: low-level frameworks and Graphical User Interface (GUI)-based environments.

Low-level frameworks: These frameworks are software libraries that offer a wide range of capabilities (i.e., use of 3D models to manipulate virtual objects), but they require advanced programming skills [29]. Most papers utilize widely adopted platforms such as Unity 3D and Vuforia, which they used to create an interactive 3D book where the user solves puzzles that include historical events, specific words from the legend and other details from the stories [49]. Also, they are also used in the studies for the creation of 3D models [10,15,53]. More specifically, 23 applications used Unity [10,12,15,16,17,18,19,20,21,22,23,26,31,36,40,41,42,43,45,47,49,50,53], while 11 used Vuforia [12,15,18,21,23,39,41,42,45,47,51]. Other work has used Easy AR SDK [50], DynaMus [50], where it uses Google and Europeana web services to create 3D objects in the designing of the time travel trip in Bosnia and Herzegovina. Also, in papers [15,43,51], Unreal and AR Markers platforms have been used [13]. In the case of the Renaissance village of Caldana in Tuscany, Italy, the two 2D graphics were created in Photoshop and the virtual assistants were generated with the help of artificial intelligence [40]. The artificial intelligence was specifically utilized to analyze and adapt the characters’ features and attire, ensuring they were appropriate and inclusive. The three historical characters were modeled in 3D using the freeware “MakeHuman” (http://www.makehumancommunity.org/) and imported into Blender to associate with a skeleton and bland-shapes, in order to have gestural animations from Mixamo (https://www.mixamo.com/) and reproduce lip movement during speech. The Unity Rhubarb Lip Syncer plugin (https://github.com/DanielSWolf/rhubarb-lip-sync (accessed on 12 March 2024)) was used for the lip sync animation [40].

GUI-based environments: These environments are graphical and high-level authoring tools used to develop AR applications easily, without requiring programming skills. At this level, we can see that Koutromanos et al. in paper [28] used a high-level authoring tool called the FreshAir Layar Creator to create 3D objects, which were for the depiction of the old tobacco warehouses in the city of Agrinio. Of particular interest are several cases that use TaleBlazer [25,29]. Specifically, in [29], the authors explain how they decided to choose TaleBlazer, among others, such as ARIS and Metaverse Studio. TaleBlazer is an open-source platform developed by MIT (Massachusetts Institute of Technology). It also offers a user-friendly visual block-based programming environment for the development of AR applications using interactive and storytelling experiences by incorporating elements such as multimedia content and gamification. These applications can take place in outdoor environments, using GPS technology [25]. In [29], they wanted the players to be focused on the game and the learning process without losing the sense of the real world. TaleBlazer enabled the development of a game that could be played by users who are in the game’s location, but also it can be played by users who cannot visit this location at that time. When they were designing the application, they decided to make it available for users who cannot visit the site and, in that way, they would have a typical mobile educational gaming experience, without the location-based elements. Also, they used custom maps in the designing in order to avoid data consumption from the users, and to have the ability to add more information to the map, such as icons [29].

Finally, the environment of Taleblazer allows the use of the “bump” or “tap to bump” settings. With this design, users would be able to “bump” into a GPS location when they are near to this location. The content would be activated automatically, or by tapping on the GPS location icons on the map. How close someone has to be to the location is an option that can be configured in the Taleblazer programming. This is very helpful and important for cultural heritage applications, because the creator of the application can adjust the distance: if the designer wants users to observe something very small, the distance could be close to the site/sign, or if he wants the user to see something bigger, he could change the distance from the location to a further one. This feature was used in the application for the archeological site of Dispilio, because the designers wanted the users to observe a few things from up close (replica items, trees, etc.) [29].

In addition to these two main platforms, there were a few studies that used complementary software tools for the purposes of mapping and 3D modeling. For the creation of maps, the Map Box Library and Easy AR [21], as well as Maxon Cinema 4D [10] are used, but also Google Maps; in study [30], we can see how a map is created with different places marked with a special color and when the user is close to the location, he can select the location to display a fragment of the digital surrogate, seeing the contemporary location along with the digital surrogate, mixing historical figures with the modern setting of the historical site. While Autocad Map 3D has been used in paper [35] along with ArcGIS online platform by Canciani, M., Chiappetta, F. et al. to depict a 3D map of Villa Andriana in Tivoli, Italy, studying the ruins and ancient paths.

Furthermore, Tan et al., in paper [52], have used 3D modeling software like Adobe Fuse C platform and Mixamo platform in order to give movements to their characters. To design 3D objects, they used Sketch Up [52], as they have used in [50] for the scale model based on the building’s plans, and printed them using FDM technology, while in the paper of Kleftodimos et al., Heygen, an AI video generator platform, was used for the creation of virtual characters and animated video [25]. Finally, Pietroni used 3D Studio Max to design 3D objects about the city of Matera in Italy [36]. For 3D models, they have also used Auto Maya [32], Blenden and Cinema 4D [41] and RealityCapture for the 3D reconstruction of buildings in [47].

This analysis covers a wide variety of development tools, from low-level, code-based frameworks to high-level, visual authoring tools. The choice of tool depends on the technical skill of the development team and the amount of interactivity or realism. The professional tools (such as Unity and Unreal Engine) have more control and flexibility for design and performance, whereas GUI-based tools aim to be more accessible for the educators and heritage practitioners who do not have a programming background.

4.5. Technical Features of AR Applications

The fifth research question addresses the devices and technical characteristics that are required to run AR applications. Figure 4 shows the distribution of devices used by reviewed AR applications. The majority of applications, according to Figure 4, can be used on smartphones or tablets (mainly Android), making them accessible to a wide audience. In some cases, applications may also be accessed via desktop platforms.

Computer devices: One of the applications is only for use by computer: the virtual cultural map of the city of Veria [44]. It is an interactive application that includes various functional or thematic areas such as an interactive digital map of Veria, image gallery, videoclips, panoramic site photos and general information about the city of Veria. The VeriaGrid system is a platform based on digital cartography, which supports a vector map of the city of Veria, organized by layers and linked to multimedia objects. It was supposed to be used by personal computers, personal digital assistants (PDAs) and mobile devices. However, because it was made in 2006, we can only visit the website.

Mobile or tracking devices: 44 out of 45 applications are designed to be used by mobile devices, such as smartphones or tablets. We observe that most of them contain GPS geolocation technology, as it helps the user to know exactly where the monument/cultural reserve that they are visiting is located, and most of those we studied use it. Also, in combination with QR codes that can be used, the user’s location determines which image data will be displayed to the user [28,35].

The 34 applications that used GPS are as follows: [10,11,13,14,15,16,17,19,24,25,27,28,29,30,33,35,36,37,38,39,40,41,43,44,46,47,48,49,50,51,52,53], while only 6 used QR codes: [22,49,51,53], along with [28,35] that we have mentioned above. We should also mention that in [26], they have not designed the application yet, but we estimate that they would probably use GPS, since it will combine a mural in a library about the poet Ivan Vazov with his own poems. However, it is not included as an application that uses GPS.

The rest of the applications do not use GPS [12,18,20,21,22,23,32,34,42,54], since they do not need it; for example, [34] is about the designing of a database for the catacombs in Italy, which has not been finished yet and in [12], the application presents the traditional Xi Shi dance in an illustrated book format and they focus mostly on the design of the characters, in order to reproduce the characteristics of clothing to the greatest extent using 3D modeling software. The application of paper [51] is about the traditional food in Malaysia, where the presentation of the information used is through displaying 2D images and/or 3D traditional foods of Gudeg, Lotek and Bose, and there are also icons, menus and text colors and other elements that would be used and displayed in the AR application [21]. Furthermore, the application of paper [42] talks about the traditional type of watermill in Greece and it is based on marker-less tracking, is activated by image recognition from the interior of the watermills and has to be pre-installed because access to Internet is not always available due to the remote position of the watermills, often away from inhabited areas and network coverage [42].

Moreover, the application of paper [54] is for restoring digital old objects in their original form, visualizing their actual past state. In this case, because the objects (here: old cabinet and paintings) are sensitive to light and the Museum allows using only the visible light spectrum, the solution came with the choice of mobile devices such as tablets or smartphones that usually have only a monocular camera and thus, they decided to limit their application to only one camera as their input device and develop an algorithm to estimate its position in real time, keeping in mind that the registered point clouds were scanned using a structural light technique instead of typical state-of-the-art methods: SfM or laser scanners [54]. Finally, in [23], the application is about promoting the traditional Peking Opera and its characters; therefore, GPS is not needed. But, in this application, it is worth mentioning that motion-capture and pose-detection technologies were used to evaluate Peking Opera movements [23]. In [20], the application has features that target real-world use in all types of locations that have at least a 1 × 1 m space for context-aware placement of objects, in order to promote co-viewing in several artifacts from different museums around the world. Finally, in [18], the application re-enacts the stages of leather tanning process in an old tannery building, so GPS is not needed. Likewise, in the application [32], with the murals in the Zhao Yigong Tomb, and in [22], since the application promotes Thailand and Malaysia’s culture remotely.

Overall, the technical features of AR applications for cultural heritage range widely based on user context. Mobile platforms dominate due to their accessibility and GPS support, particularly for outdoor experiences. On the other hand, indoor applications’ image recognition or marker-based tracking is commonly used for higher accuracy. Some publications apply advanced techniques like motion capture or monocular tracking, which require special hardware or domain knowledge. The ideal composition of technical features is according to the desired level of immersion, usability and cultural relevance of the experience.

4.6. Evaluation Methodologies of AR Applications and Research Results

The final research question explores the methodologies used to evaluate AR applications. Among the 45 reviewed studies, not all reported conducting a formal evaluation of their AR applications. According to the articles we studied and Figure 5, the findings show that 36 studies conducted evaluations, of which five are technical evaluations and they study whether the application works properly and how users handle it [16,21,43,54] while in paper [43], both technical evaluation and evaluation from users have been used, but not in a quantitative way. Study [21] employed black-box testing to assess the application’s functionality and operational behavior and the work [22] was evaluated with a heuristic evaluation by ten experts in human–computer interaction (HCI).

Nine applications were not evaluated [12,14,26,35,36,37,38,41,49], whereas five applications do not mention the evaluation sample explicitly. In the paper about Ivan Vazov’s work [26], since the application is still under development, no evaluation has yet been conducted. On the other hand, the evaluations performed in [15,24,28,34,48] did not provide information on sample size or data analysis procedures.

A total of 27 studies reported complete evaluations (as [10,11,18,19,20,22,23,24,25,29,30,31,32,33,39,42,44,45,46,47,50,52,53]). A total of 15 studies out of 26 used quantitative research methods: 2 of 26 used qualitative methods and 9 used mixed methods. As we pointed out previously, the app described in the paper of Cisternino et al. [10] and the one in De Paolis et al. [39] is the same, as both describe the same app that promotes the understanding of cultural heritage of the church “Basilica of Saint Catherina of Alexandria in Galatina”. However, in the second paper [39] they aim to analyze the relationships between usability, user experience and mental workload factors in AR-based digital storytelling [39]. Consequently, we examined the 19 complete evaluations in terms of examined variables, data collection tools and sample size and data analysis methods as well as the findings of the evaluations. Xu N. et al. [47] have explained the process of evaluation for the application that has also been introduced briefly in [46].

Examined variables:

Table 2. Complete research evaluations.

Apps/Variables	Easy to Use	Easy Access to Info	Help in Learning	Promote Tourism	Promote New Knowledge	Update the App	Content Quality
Cisternino D. et al. [10]	✔	✔	✔		✔	✔	✔
Koutsabasis P. et al. [11]	✔	✔	✔	✔	✔	✔
Tan S.N. et al. [13]	✔		✔	✔	✔
Kurniawardhani, A. et al. [17]	✔			✔	✔		✔
Galani S. et al. [18]	✔	✔		✔	✔	✔	✔
Hinciape M. et al. [19]	✔	✔	✔		✔		✔
Ch’ng E. et al. [20]	✔	✔	✔		✔	✔	✔
Aziz N. et al. [22]	✔	✔	✔	✔	✔	✔
Hongning S. et al. [23]	✔		✔	✔	✔	✔	✔
Kleftodimos A. et al. [25]	✔	✔	✔		✔	✔
Tian J. et al. [27]	✔		✔	✔	✔
Kleftodimos A. et al. [29]	✔		✔	✔	✔	✔
Cushing A.L. et al. [30]	✔	✔	✔	✔	✔
De Luca V. et al. [31]	✔	✔				✔	✔
Zheng S. [32]	✔	✔	✔		✔		✔
Spandoni et al. [33]	✔	✔	✔		✔
De Paolis L.T. et al. [39]	✔	✔	✔		✔	✔	✔
Capecchi I. et al. [40]	✔	✔		✔	✔	✔	✔
Tzima S. et al. [42]	✔	✔	✔		✔
Garoufallou E. et al. [44]	✔	✔		✔	✔	✔
Grammatikopoulou A. et al. [45]	✔	✔	✔	✔			✔
Xu N. et al. [46]		✔	✔	✔
Xu N. et al. [47]		✔	✔	✔	✔	✔
Rizvic S. et al. [50]	✔	✔			✔		✔
Siang T.G et al. [51]	✔	✔	✔	✔		✔	✔
Tan K.L. et al. [52]		✔	✔	✔	✔	✔
Koo S. et al. [53]	✔	✔		✔	✔	✔	✔

shows a summary of examined variables used in the complete research evaluations. In this table, we set as variables the ability for the users to navigate, explore and use the application with ease (users’ experience); if the information given can be accessed without barriers, which are very important, the application and the new knowledge they want to pass on could be accessible to all. Moreover, we want to focus on the way these applications are used to promote new knowledge and thus, they are often linked to educational processes and/or aim to promote tourism in specific regions. Another relevant variable to include in the table is the ability to aid in the educational process (or learning new things) and if they are used to enhance a subject about cultural heritage. The last two variables we want to observe are if the creators of the applications are concerned with making their designs even better and whether the volunteers are asked; the second variable has to do with the quality of its content and if the users explicitly evaluate the quality of the app’s graphics or esthetics.

We can see in the evaluation of Spandoni et al. [33] that the digital interactive experience of users was evaluated as successful, since it managed to “translate” content of high scientific value with simpler facts, while in Hincapie et al. [19], students who used the application during their tour in “Medellin’s Cisnero” square scored higher on the unannounced post-tour knowledge test, compared with those who did the tour, but without the app. Koo et al. [53] mention positive results and suggestions for improvement of the app.

Furthermore, there are some studies that mentioned the quality content factor, which is to evaluate if they are satisfied by the graphics or esthetics of the applications and some of them are [53], where they have evaluated the artwork of the app, whether it uses rich and engaging graphics and if this is beneficial for the whole application. In [31], they evaluated the activity as being well structured or not, and it felt so authentic that it looked real, while in [20] they evaluated the app as attractive and esthetically appealing.

Another parameter mentioned within the evaluated applications is the promotion of tourism and the new knowledge that one can gain with the help of AR on cultural heritage issues [11,42]. In Koutsabasis et al. [11] questionnaires show that this application can promote easy learning, while the visit to the museum becomes more entertaining. In Tzima et al. [42], the results of the research show that users were motivated to look closer and more carefully at the mechanism of watermill, while some visitors were emotional, recalling memories of their childhood and the way their parents and grandparents lived.

Data collection tools and sample sizes: In this section, we present the research tools that have been used to evaluate the applications. We can divide them into three groups: questionnaires, interviews and semi-constructed interviews, and finally, those that used pre-/post-tests. Before seeing the groups in detail, we should mention that not all papers refer to their sample size. More specifically, [24] does not refer to their sample size and that is the reason we have not included this in the table above with all the complete and fully evaluated applications. Sample sizes in most surveys were not very large, since they ranged between 15 and 70 in most of them. The exception in this is [51], with 120 people being questioned, who were chosen randomly, and [53] with 169 complete responses. In [53], which is a World Heritage Site AR Tour guide app (protected by UNESCO) for Hwaseong Fortress in Suwon of S. Korea, questionnaires were given to 201 people, also randomly, but 169 complete responses were received that evaluated its ease of use and satisfaction with it.

Evaluations through questionnaires: Our first group established that most studies used questionnaires and in Spandoni et al., 23 volunteers participated in testing the application, to see if the experience of using the application imparted valuable knowledge to the user about the objects (astronomical tools) that were “silent” until then [33]. In the next study about the Basilica of Saint Catherine of Alexandria in Galatina [10], the research sample was 41 people, who filled in a questionnaire with prepared questions using standard tools (such as SUS, NASA-TLX and UEQ), where they evaluated the application (how useful it was, how easy it was to operate, to what extent they controlled the interaction, how creative the design was). It is not mentioned if they were volunteers or they were chosen randomly.

In the applications that were created for the Prehistoric Settlement of Dispilio in Kastoria, Greece, the users who tested the applications were mainly secondary school students and a group of university students. The visits from the school classes were according to an educational program of the Environmental Education Center in Kastoria. When they arrived at the settlement, the students were divided into groups. Some of them played the game on their own, while others were playing with the advice and help of an educator from the Center. The students were provided tablets for the Center, but they could use their own smartphone, if they wanted to. The sample that was used for the evaluation is of 71 university students and 58 school students, who volunteered. The university students employed for the evaluation were attending courses that had to do with digital education: “Digital Media and Education” and “Digital Storytelling”. They used the applications away from the settlement and they evaluated only this specific experience/use. The designers of the applications used the same questionnaire to evaluate both of them [29]. The questionnaires had 22 Likert-scale questions, divided into seven main categories. The results show that the applications had high scores in aspects like easy to use, fulfillment, challenge and entertainment in the evaluations that were performed by university students. In the evaluations performed by school students, the scores were even higher in almost all the categories. This could have happened due to the fact that they used and experienced them in the field, during their visit to the Prehistoric Settlement [29].

In [24], the application itself collects demographic information about the visitors who used it, which can show us who chooses to use each route in the Lisbon Botanical Gardens, but it does not provide us with the number of visitors. Accordingly, in [50] they tell us that they used questionnaires with closed- and open-ended questions for the evaluation, but they do not give either the questions or the sample that answered them.

In the application of paper [51], we can see that they used random interviews of 120 museum visitors, while in the application about the Greek type of watermill [42] a sample of 25 people was used. They organized visits in groups of four people, because they had limited devices in order to use the application. At the end of each tour of the site and use of the application, they filled out a questionnaire about their experience and a discussion followed between the creator of the application and the participants.

Evaluations through interviews: The second group has to do with evaluations that were performed through interviews and semi-structured interviews. In the application “Walk1916”, the walking tour of Easter Rising sites in Dublin [30], the evaluation was preceded with 15 semi-structured interviews with users of the app. The main research question explored by the creators was how the integration of digital surrogates with mobile technology features affected the participants’ perceptions and experience. The study does not specify whether the evaluation participants were selected randomly. It should also be noted that interviews have been used in some papers that utilized mixed methods [13,27,44,46,50,51]. In the oldest application in our research, the digital cultural map of the city of Veria in Greece [44], only 10 users volunteered (Veria residents and University students) to evaluate the usability of the application by semi-structured interview and a questionnaire. They were given 14 tasks to do using the VeriaGrid system and participants were invited to freely express their thoughts and feelings while they were completing the tasks and the semi-structured interview and questionnaire followed it. The variables that are given to examine are frustrating–satisfying, dull–stimulating and difficult–easy, with values based on a one to nine semantic differential scale [44]. In [47], a semi-structured interview was conducted and analyzed using theme-based content analysis in ten users. For experts (three volunteers), they further asked their attitudes about the HeritageSite AR from the perspective of its potential values.

Evaluations through pre- and post-tests: The final group contains applications that evaluated with pre-/post-tests and in the application of promoting the Cisneros Market [19], 48 students from the University of Medellin were used, in the context of the cultural heritage course. A total of 22 of them visited the historical sites in Cisneros Market, only with the information they received verbally from a specialized tour guide (control group). The rest of them (26 students—experimental group) followed the exact same route, receiving the same information verbally from the expert tour guide but they also had the opportunity to use the app to view, analyze and interact with the content it provides: photos, audio material and information in the text. Both groups had to complete an introductory questionnaire of general questions (mainly about demographics) and at the end of the tour, they took a short knowledge test (10 questions) to assess their knowledge of Cisneros Market, without being informed beforehand. One more application with pre- and post-tests, is the one about raising awareness of learning and getting to know historical parts of Malaysia through gamification [52], where a sample of 50 people was used and they completed a pre- and post-test to evaluate the knowledge they already had, but also the new knowledge they acquired through the game. They also filled out a questionnaire about its use and design.

Another paper that stands out for its experimental method is this of Ch’ng et al. [20], where they wanted to find out which features of Social AR can assist the communication, in case of a “virtual museum”, since they have created an AR application that allows many users to interact and discuss cultural heritage objects. Their experimental evaluation involved 30 participants grouped into 10 sets of three to simulate social interactions. Participants engaged in experimental scenarios, progressing from individual learning to group interactions in an AR application. The participants of each group completed the pre-/post-tests and questionnaires to evaluate user experiences and perceptions. Their research showed the importance of the sense of social presence and the perception that the objects are real into the place-context of the users, in terms of intention, focus and reciprocity, while the experience of communicating through Social AR was unique. They have proved that co-viewing is of high importance when we talk about cultural heritage objects, and they propose using this kind of technology in order to enhance a visit to a real museum [20].

Data analysis methods: The data analysis techniques were employed in most studies to analyze the data collected from participants during the evaluation of the AR applications. They help in assessing the effectiveness, usability and user experience of the AR, as well as the impact of AR on learning or knowledge. The quantitative data analysis methods used in quantitative or mixed studies were descriptive statistics, inferential statistics or multivariate statistics. Descriptive statistical methods were used in most evaluations to summarize the demographic data of the participants or to summarize the answers of participants in questions about the test of knowledge or the user experience tests in the format of tables, graphs or measures (i.e., means or standard deviations).

Inferential methods used in the studies were the t-tests in [19] to analyze the difference in means between the control and experimental groups regarding the testing of knowledge about Cisneros Market, the ANOVA tests in [44] to compare mean differences between the two categories of users (expert and novice) based on performance data of usability such as time needed, errors performed and user satisfaction. Pearson’s correlation was used in [25] to examine relationships between three dimensions of the virtual experiences (immersion, entertainment and knowledge) and participants’ satisfaction and future intentions with the virtual experience and the destinations of Western Macedonia. Spearman’s rank correlation was used in [20] to evaluate the relationships between different variables related to user experiences and social interactions in AR environments around cultural heritage objects and the chi-square test of independence was used in [20] to examine the relationship between gender and user preferences, either alone or accompanied when visiting physical museums or using XR museums.

A few studies used multivariate statistical methods such as regression or cluster analysis. A regression analysis was used in [51] to examine the factors affecting user acceptance of the AR application with a specific focus on performance expectancy, effort expectancy, social influence, playfulness expectancy and content relevance expectancy. Furthermore, an ordinal logistic regression was conducted in [20] to explore the impact of factors affecting user experiences in AR apps in the context of cultural heritage communication through AR apps. In [31,39], they performed a PCA and cluster analysis to explore the relationships between usability, user experience and mental workload factors in their AR application.

The qualitative data analysis methods included content analysis [16], keyword analysis, word clouds or descriptive analysis in [29,33,44,46,50,51], to analyze the answers from some open-ended questions in questionnaires or interviews about the emotional experiences of AR apps.

Findings of evaluations: To summarize the results of the evaluations, keeping in mind that not all applications have been evaluated in the same way, we come to the conclusion that most of those who have used AR applications are satisfied, and that it really helps to better understand and assimilate new knowledge [11,13,18,19,20,22,25,28,29,30,32,33,42,45,46,47,49,52,53] at young [13,19,23,28,29,40,46] and older ages [15,17,18,22,27,29,31,32,42,45,47]. In [33], they claim that this app can “translate” content of high scientific value by using simple vocabulary and everyday facts. In [49], the research for information and gaining new knowledge about a place’s history (Nis, Serbia) is easier, while in [11] the app will be given to tourists and more visitors of the museum, so they will have more interesting results to share with us. The evaluation until now is about finding ways to make the application even better [11]. In [19] the results of the evaluation show that AR technology can help the process of learning, while making it easier to assimilate and understand new knowledge.

Moreover, in [28], their evaluation shows that serious games are an important aid in learning about the cultural heritage of one place, especially in classes with younger students in primary school. In [46], the volunteers who used and evaluated the application for Shuangta found it to have made the cultural heritage visits more meaningful and playful, and [23] where the application helps young children learn more about the Peking Opera, it is obvious that they emphasize in the education of younger children. On the contrary, in [29], the focus is equally distributed on the learning process in both younger and older groups. Applications in [13,23,29,40,46] share a common goal of making the learning process more enjoyable and engaging when learning about cultural heritage.

In [15], the application about the Famosa Fortress and [42], with the application that shows a Greek type watermill, we can see that they focus more on older users and how they can use the app, but they do not exclude younger learners from the learning process. In [31,45] the applications appear to be easy to use and they enhance the enjoyment of cultural heritage.

Overall, these applications enhance learning by making it more interactive, enjoyable and contextually meaningful, but it is shown more clearly in papers [11,13,18,19,23,24,27,29,30,32,40,46,47,52]. We can conclude that the applications we studied could achieve their main goal: awareness and education on cultural heritage issues and monuments, like the visit to the Osservatorio in Milan, Italy and the explanation of highly scientific instruments and theories [33], the promotion of the Basilica of Saint Catherine of Alexandria in Galatina, Italy [10] and the promotion and reactivation of the cultural heritage of Cisneros Market in Columbia [19]. There are also the monuments in Dublin that are connected with the Easter Rising [30], the exploration of historical places in Malaysia [52], the promotion of the church Sant’ Elia in Ruggiano, Italy [31], the exploration of a museum with a self-guided tour tool that could enhance the public’s participation and inclusion by making visitors co-creators of the offered cultural experience and help museums transform into more inclusive spaces [45], the Relics of the Arhat Monastery and Twin Pagoda (Shuangta) in [29] and the promotion of history of the Peking Opera and its characters in [23].

On the other hand, many of them are helping the tourism development of the respective area, such as the promotion and the protection of the underwater cultural heritage [14]; the promotion of the Italian catacombs in Rome [34]; the promotion of the Famosa Fortress [15]; the promotion of Indonesian food [53]; the city of Melaka in Indonesia [24]; the sites of the Easter Rising in Dublin [30]; the city of Matera in Italy [36]; the city of Veria in Greece [44]; the promotion of mountainous areas in Western Macedonia in Greece [25] and the promotion of traditional tannery [18]; the promotion of the Relics of the Arhat Monastery and Twin Pagoda (Shuangta) in China [46,47]; the promotion of the Borobudur temple in Indonesia [17]; the promotion of the Renaissance village of Caldana, Italy, to raise awareness of the importance of urban heritage and to stimulate its visit by tourists, and especially younger people [40]; the promotion of history and way of life at Praya Lane, in Malacca, Malaysia [13]; and finally, the enhancement of the Jichang Garden in China with traditional Chinese poems to educate its visitors [27]; the promotion of the murals in the Zhao Yigong Tomb [32] and the remote enhance of the tourist experience in Southeast Asia [22]. All the applications that are mentioned above are trying to promote those places to tourists, while they are distributing new knowledge about those sites.

5. Discussion

This review examined the use of AR applications for cultural heritage, with an emphasis on education and accessibility. In this section, we discuss the key findings of our research.

5.1. Tangible Heritage and Regional Distribution

Most of the reviewed applications were developed to present tangible cultural heritage, such as monuments, historic buildings and archeological sites. Only a few addressed intangible heritages, including oral traditions, poetry, music or traditional crafts. For instance, PoeticAR by Tian et al. aimed to promote the appreciation of Chinese poetry [27], while the application by Galani et al. focused on traditional leather tanning [18]. These efforts are valuable, yet they represent exceptions.

Moreover, we observed regional clustering, with many applications emerging from Europe (especially Greece and Italy) and Asia. European countries primarily focused on tangible heritage, while countries such as Indonesia and China explored intangible cultural heritage content, such as music and traditional crafts. Other regions such as the Middle East or Balkan countries were underrepresented. This issue raises questions about global accessibility and the extent of cultural diversity represented in AR development.

5.2. Methodological Approaches to Design and Development

Our analysis identified a range of design and development approaches. Most studies adopted user-centered design, emphasizing usability and iterative prototyping. Fewer used design thinking to align user needs with creative problem-solving or social-centered design, promoting social interaction in AR environments.

In terms of development tools, most applications used low-level programming environments such as Unity, AR Foundation or C#, providing flexibility and realism but requiring technical expertise and longer development [23,31,45]. Few used graphical tools like TaleBlazer [25,29] or Blueprint Visual Scripting [46], making AR creation more accessible but sometimes limiting customizability. The use of low-level tools creates barriers for educators or stakeholders without a technical background, while the lack of accessible authoring tools limits the democratic production of AR content.

Additionally, the incorporation of AI-based features, such as the use of Mediapipe to detect body movement in educational contexts [23] or AI content generative tools such as Heygen for creating virtual characters [25], shows that newer technologies are being adopted in novel ways. However, the field still lacks interoperable, open-source and easy-to-use authoring environments for heritage practitioners and educators.

5.3. Educational Enhancement Through AR

It is evident that most of the applications studied had a positive educational impact. From the applications we have studied, 31 of them are being used to help and assert the educational process, among students and other visitors of the sites. Some of them are the application of Hincapie et al. that presents the way to revive the cultural heritage in Medellin’s square [19] and the application of Grammatikopoulou et al.’s [45] “Artful” for the art exhibition “Poetry meets AR”, which keeps a user-friendly character, and users are able to interact with the exhibits. Similarly, Tian et al. [27] enhanced visits to the Jichang Garden by exploring traditional Chinese poetry, while Galani et al. [18] re-enacted the stages of the leather tanning process, adopting a novel approach that augments 3D content upon a physical scale model of an old tannery.

A common feature of these studies is the intention to allow users to interact not only with the application itself but also with their physical environment, whether it is a square in the city, a famous garden or an old building as the center of a tanning factory from the previous century. We can also detect that they want to enhance old practices (leather tanning process) or forgotten parts of the city that used to be the trade center. By using them, the participants gain new knowledge, while they discover parts of their own history that were forgotten or considered less important, like poetry.

5.4. Use of Gamification and Serious Games

Only a limited number of applications employed gamification to promote education, with an emphasis on interactivity and immersive engagement. Only 12 of them used gamification [12,13,23,25,28,29,40,46,47,52,53] and one more is currently in progress [26]. Some of those applications are presented briefly below: in Yan et al.’s research, their application was designed for the myth of Xi Shi and consists of an interactive book, while it also requires solving puzzles [12], while Koo et al. have included gamification activities in their main application that promotes an important tourist attraction in South Korea: the Hwaseong Fortress [53]. In this application, they have included two types of AR construction games: simple and sandbox; while the simple construction game is intended for users to follow the construction process as it was already designed, the sandbox AR construction game provides the opportunity to build a structure more freely within a limited time, using various construction materials. Another application with a designed serious game is the application of Tan et al., which was created to emphasize historical places in Perak (Malaysia) [52].

In Greece, two cases stand out: Koutromanos et al. with an educational (serious) game for young students who attend elementary school to learn about old tobacco warehouses in the city of Agrinio, in western Greece, and their impact on the city’s history [28]. The other case is the one in Kleftodimos et al.’s study, which presents two location-based AR educational applications for a prehistoric settlement near the lake. The second application has more gamification elements and is called “Crime in the lake settlement”. Its purpose is to offer more knowledge about the prehistoric settlement in the lake of Kastoria, with the assistance of a game that includes a story unfolding in this place [29].

Gamification generally leads to higher user acceptance and greater engagement, as confirmed by the studies discussed above.

5.5. Evaluation Practices and Methodological Gaps

Regarding the evaluation studies, we reached the conclusion that not only are there few applications that use gamification activities, but they also have not been tested or had results collected about their use. Two of them [28,52] have been tested properly and one of them [42] has collected pre- and post-tests about using the application and its design.

From recent studies (2023–2024), we observe that most of them have made an attempt at a complete evaluation [13,17,18,20,22,23,25,27,29,31,32,40,45,46,47] by giving the methodology of evaluation, the sample surveyed and some of the results. However, many applications are still untested or rely on small, non-generalizable user groups. In some cases, evaluations are pending or incomplete [12,26,41]. It should be noted that most applications present some form of evaluation, but they often lack standardized tools such as the SUS, UEQ, or pre-/post-tests. This limitation affects the generalizability of results and comparative evaluations of the effectiveness of AR applications.

Having this in mind, the findings suggest that there are some gaps in most studies concerning the results of using applications. In addition, more emphasis should have been placed on the impact they have on the educational process, while highlighting the respective cultural reserves that have been selected.

Studying their evaluations [20,23,28,29,52,53] through questionnaires and mini-tests, the findings suggest that, indeed, those applications have helped users to learn new things about particular cultural heritage reserves [13,17,18,27,31,32,40,45,46,47,52,53]. Moreover, users reported that these applications motivated them to visit heritage sites and were generally easy to use. Nevertheless, they suggested improvements to enhance usability and user-friendliness [20,23,25,28,52].

5.6. Cultural Heritage Types and Their Relation to Design and Evaluation Practices

Table 3. Evaluated applications.

Article	Title of App	Type of Cultural Heritage	Design Methodology	Development Methodology	Evaluation Methodology	Evaluation Framework
Cisternino D. et al. [10]	Basilica of Saint Catherine of Alexandria in Galatina	tangible	user-centered	low-level	quantitative	tour on the site
Koutsabasis P. et al. [11]	Mouseion Topos Lesvos	intangible	design thinking	low-level	mixed	museum/tour
Tan S.N. et al. [13]	The Story of Praya Lane	intangible	gamification, user-centered	low-level	mixed	tour on the site
Kurniawardhani, A. et al. [17]	Temple Relief Storytelling	tangible	user-centered	low-level	quantitative	tour on the site
Galani S. et al. [18]	The Tanning Process of Leather	intangible	user-centered	low-level	quantitative	on the site (museum)
Hinciape M. et al. [19]	Vitica	intangible	user-centered	high-level	quantitative	tour on the site
Ch’ng E. et al. [20]	Social AR app	tangible	social-centered	low-level	mixed	educational use (in class/at home)
Aziz N. et al. [22]	Pattani heritage	tangible	design thinking	low-level	quantitative	in class/office
Hongning S. et al. [23]	OperAR	intangible	gamification	low-level	mixed	educational use (in class)
Kleftodimos A. et al. [25]	“Once upon a time in Dispilio” and “Crime in the Lake Settlement”	tangible	user-centered gamification	high-level	quantitative	educational tour (students)
Tian J. et al. [27]	Poetic AR	intangible	user-centered	low-level	mixed	tour on the site
Kleftodimos A. et al. [29]	“Once upon a time in Dispilio” and “Crime in the Lake Settlement”	tangible	user-centered gamification	high-level	quantitative	educational tour (students)
Cushing A.L. et al. [30]	Walk1916	intangible	user-centered	high-level	qualitative	tour
De Luca V. et al. [31]	CumeRa	tangible	user-centered	low-level	quantitative	tour on the site
Zheng S. [32]	Zhao Yigong Tomb murals	intangible	user-centered	low-level	quantitative	tour on the site
Spandoni et al. [33]	MusAB/ MARSS	tangible	user-centered	low-level	quantitative	museum
De Paolis L.T. et al. [39]	Basilica of Saint Catherine of Alexandria in Galatina	tangible	user-centered	low-level	quantitative	tour on the site
Capecchi I. et al. [40]	Village of Caldana	tangible	gamification	low-level	quantitative	tour on the site (students)
Tzima S. et al. [42]	The Greek Type Watermills	tangible	user-centered	low-level	mixed	on the site (museum)
Garoufallou E. et al. [44]	The Veria Grid	intangible	user-centered	high level	quantitative	tour
Grammatikopoulou A. et al. [45]	Artful	intangible	social-centered	low-level	mixed	on the site
Xu N. et al. [46]	Heritage Site AR	tangible	user-centered gamification	low-level	qualitative	tour
Xu N. et al. [47]	Heritage Site AR	tangible	user-centered gamification	low-level	qualitative	tour
Rizvic S. et al. [50]	Sarajevo 5D	tangible	user-centered	low- level	mixed	tour on the site
Siang T.G et al. [51]	When History Comes Alive	intangible	social centered	low-level	qualitative	tour
Tan K.L. et al. [52]	MIGHT	intangible	gamification	low-level	mixed	on the site
Koo S. et al. [53]	Hwaseong Fortress	tangible	user-centered user-centered	low-level	quantitative	tour on the site

presents the 27 applications that have been tested and evaluated, organized according to the type of cultural heritage they address. Additional columns show the design and development methodologies applied and the evaluation setting.

The distinct trend that emerges is that, depending on the nature of the cultural content (tangible or intangible cultural heritage), it determines the methodological choices for design, development, and evaluation.

For applications centered on tangible heritage (e.g., MARSS—Spandoni et al. [33]; Basilica of Saint Catherine—Cisternino et al. [10]; CumeRa—De Luca et al. [31]; Village of Caldana—Capecchi et al. [40]), user-centered design dominates, often combined with elements of gamification (e.g., Dispilio—Kleftodimos et al. [25,29]; Heritage Site AR—Xu N. et al. [46,47]). An exception is the application “Pattani heritage”—Aziz N. et al. [22], which followed the design thinking, even though it promotes tangible cultural heritage. Most of these applications are developed using low-level approaches and are often deployed as on-site tours or museum experiences, with only a few high-level implementations in educational contexts. Evaluation is predominantly quantitative, typically relying on questionnaires or structured metrics. Although this evaluation approach provides clear conclusions on user satisfaction and comprehension, it rarely captures the deeper dimensions of learning or cultural engagement. In this domain, AR tends to operate primarily as a tool of representation and visualization.

By contrast, applications devoted to intangible heritage (e.g., Mouseion Topos Lesvos—Koutsabasis et al. [11]; Sarajevo 5D—Rizvic et al. [50]; Vitica—Hinciape et al. [19]; PoeticAR -Tian J. et al. [27]; OperAR—Hongning S. et al. [23]; Artful—Grammatikopoulou et al. [45]; Praya Lane—Tan S.N. et al. [13]) employ a more diverse methodological palette, including design thinking, social-centered approaches and gamification. Although these applications also tend to remain at a low level of technical development, their evaluation methods are more often mixed or qualitative, reflecting attempts to account for experiential, esthetic and social dimensions of engagement. For example, OperAR used gamification in classroom contexts using mixed-method evaluation, while Praya Lane combined social and gamification-based approaches to foster collaborative learning and storytelling.

Another pattern emerges when comparing educational applications with tours. Applications designed for students or classroom settings (e.g., Dispilio games—Kleftodimos et al. [25,29]; OperAR—Hongning S. et al. [23]) frequently combine gamification with user- or social-centered design and adopt more comprehensive evaluation frameworks that assess not only usability but also learning effectiveness. By contrast, tourist-focused AR tours (e.g., Temple Relief Storytelling— Kurniawardhani et al. [17]; Basilica of Saint Catherine—De Paolis et al. [39]) typically rely on quantitative methods and site-based surveys.

The data show that no single methodological approach is efficient across different contexts. Instead, effectiveness depends on the cultural setting. User-centered design, often supported by gamification, proves most effective for tangible heritage in tours, as it enhances visitor engagement and satisfaction. In contrast, for intangible heritage, social-centered and gamification approaches show greater effectiveness in educational and community contexts, where mixed evaluations confirm stronger learning outcomes and collaborative participation.

The findings of this review enrich the theory around AR in cultural heritage, since they contribute to the understanding of the relationship between technology–learning–culture. They advance the following principle: the type of heritage (tangible vs. intangible) and the context of use (tourism vs. educational/community-oriented) systematically determine (i) the design model (user-centered vs. social-centered), (ii) the development stack (low-level frameworks vs. high-level authoring tools) and (iii) the evaluation method (quantitative vs. mixed methods). This principle explains why no single methodology is universally optimal and provides a conceptual basis for comparing and theorizing the effectiveness of Augmented Reality in cultural heritage research. It also offers more insights into design and evaluation processes, which can be especially valuable for scholars who do not have a strong technical background.

6. Limitations and Future Research Directions

Based on the analysis of the papers, we identify some research gaps and future research directions:

Promoting of cultural stock in Greece: Our study showed that there is a limited number of papers on AR applications for enhancing cultural heritage in Greece compared it with the archeological, historical and cultural wealth that the country has, both tangible and intangible. It is important to note that a few new studies and applications have been made about the region of Western Macedonia [25,29] during the last two years. This shows that there is indeed a vast cultural reserve in this region, waiting to be used in such interesting and useful applications. Furthermore, our study showed that there is a lack of AR applications for promoting intangible cultural heritage in general. Therefore, there is a clear opportunity for future development of applications in this direction.

Development of AR applications using user-friendly platforms: The findings show that most AR applications were implemented using low-level software libraries (like Vuforia, Unity, etc.) which demand strong programming skills. On the other hand, a limited number of AR applications were developed using graphical programming platforms (i.e., TaleBlazer), for which no coding expertise is required. Hence, there is potential for developing AR cultural heritage applications in a fast and easy way using high level authoring tools. For example, web-based platforms like TaleBlazer, ARIS and Metaverse Studio are easier for non-specialized users to use, to create mobile applications. These tools provide educators with various capabilities, such as creating AR applications using gamification elements (i.e., scavenger hunts), informative tours and storytelling experiences. They also enable educators to place GPS locations on a digital map and associate multimedia elements with these locations [29].

Generative AI in AR design: Artificial intelligence (AΙ) provides important help for designing AR applications for cultural heritage. More specifically, generative AI tools can provide ideas and narratives in order to enhance the flow of experiences and foster engaging AR experiences and interactions. AI tools can also generate multimedia content generation, such as virtual objects, characters, environments and other elements seamlessly integrated into AR experiences. Although there has been limited research in this field, such as the work in [25], it could still serve as a valuable tool for flow experience and content generation in future AR applications.

Evaluation of AR applications: It is evident that there is a limited number of works that have evaluated AR applications integrated in the Greek education system or tourist audience and even fewer that have been reported on World Heritage sites protected by UNESCO [21,53]. Moreover, we observed that the majority of papers were focused on evaluating AR applications based on usability, visual content and their effect on the learning process, using observation studies. However, only a few studies, like [19], have used experimental and control groups to investigate the impact of learning or knowledge with an AR application compared with a traditional approach (i.e., without AR application). Consequently, there is an opportunity for further quantitative evaluation regarding the effectiveness of AR applications with gamification elements. For instance, several experimental designs can be used (such as a tour approach in cultural sites with three levels: one tour without an AR application, one tour using AR gamification with points and one tour using AR gamification with puzzles, etc.) in order to compare the effect of several gamification elements of applications in terms of user experience and other variables such as knowledge and learning. Also, there can be a qualitative evaluation of the AR applications in terms of the sentiments and experiences, to obtain more in-depth information on the use of AR by the users.

Including people with disabilities and behaving with respect for each culture: In the applications we studied, there are no AR applications that take into account the needs of people with disabilities. AR applications are very easy to be use for everyone, but we should think in advance to make them easier to use for people who may have a disability (e.g., limited or no vision).

Another limitation that was found is the fact that even though almost every researcher state that their AR applications are used to promote and enhance their cultural heritage, the context in which they approach the space is not clear enough; the respect for the sanctity of the intangible and tangible cultural heritage should be clearly stated, as they are samples of the respective cultures.

Learning theories: Almost none of the studies (85.2%, 23 out of 27) systematically utilized learning or technology adoption theories, except for three papers (11.2%). For instance, paper [45] used social constructivism, Communities of Practice (CoP) and connectivism in the design of the “Artful” app. The “MIGHT” application [52] drew on a generalization of task-based theory for the SandBox serious game. Finally, paper [13] applied transformative learning theory through a deductive thematic analysis of evaluation data in the “Story of Praya Lane” application. Consequently, we suggest that the integration of such theories (constructivism, presence, multimodalities, cognitive load, TAM) could provide better guidance for future research.

This review has some limitations that should be considered. First, the search was conducted in only two databases (Scopus, Web of Science), which may have excluded relevant works from other sources such as Google Scholar. Second, the selection and categorization of motivations and methodologies was based on qualitative analysis, which involves a degree of subjectivity. Third, the sample is limited to publications up to 2024 and cannot capture future developments.

7. Conclusions

In this research, we investigated AR applications that were created to highlight and enhance cultural heritage, either tangible or intangible, focusing on promoting new knowledge in an educational context or enhancing the touristic experience. Based on 45 reviewed papers from the Scopus and Web of Science databases, we can summarize that most applications are used to highlight tangible cultural heritage, particularly buildings and cultural routes that can be followed within cities or specific historical or cultural areas. Fewer applications deal with intangible cultural heritage, mainly focusing on traditional food, books and fictional characters or industrial production practices.

In addition, it is evident that AR technology in the context of cultural heritage is widespread and many creators choose it to promote cultural heritage elements. We also found that most AR applications were designed with a user-centered approach. The developers of these applications are technically skilled and their creations are often quite complex. Furthermore, we observed that AR applications using GPS are the preferred type of mobile app, as they can be easily implemented for promoting cultural sites or artifacts in outdoors settings.

However, the biggest deficiency we identified concerns evaluation. Although many applications initiated an evaluation process, not all of them have completed it. Most evaluations were quantitative or used mixed methods. They were typically conducted using close-ended questionnaires, although interviews and pre- and post-tests were also used, particularly in cases involving knowledge assessment in educational applications. It is of great importance that data be published regarding how each application contributes to the promotion of cultural heritage and its effectiveness in educational contexts.

Furthermore, the findings of this review show that the type of cultural heritage (tangible or intangible) affects methodological choices in design, development and evaluation. This suggests that there is no single approach; rather, methodological choices need to be carefully adapted to the type and context of heritage content.

Our future research will focus on the design of AR applications, aiming to address as many research gaps as possible. An application that combines gamification elements, interactive storytelling and AR technology will be our starting point, while we will also strive to find and select a user-friendly platform, given that not all designers are experts in application development. Finally, it is essential to include an evaluation of the application, with a specific sample of users who will test and assess our work. This will enable us to present the evaluation results to better understand how helpful AR technology can be in the educational process, particularly within the Greek educational system, while also motivating other researchers and educators to design and develop similar applications using AR technology and gamification.