Di Hatice Kırmacı, Women Science Teachers Global Network, ENOVA Training Consultancy
Introduzione editoriale
Il campo come metodo
Questo contributo arriva dalla Türkiye, ed è il primo di un dialogo internazionale che comincia adesso.
Hatice Kırmacı è fondatrice e coordinatrice di Women Science Teachers, una rete internazionale di docenti di area scientifica con cui #DiCultHer ha aperto quest’anno un dialogo. Il modello che presenta in queste pagine — il Digital Heritage Learning Model — nasce da anni di lavoro con studenti su patrimoni locali, e propone otto fasi che vanno dall’inquadramento del bene fino alla valutazione, passando per l’indagine storica, la validazione delle fonti, la co-progettazione STEAM, la produzione digitale e la narrazione.
Ma non è l’articolazione in fasi la ragione per cui lo pubblichiamo.
Il modello colloca l’osservazione sul campo come componente strutturale, non come momento accessorio. Non si tratta di una visita che precede il lavoro vero: è il punto in cui una ricostruzione digitale viene messa a confronto con ciò che esiste, e in cui gli studenti scoprono lo scarto fra quello che uno strumento produce e quello che il luogo dice.
Chi legge questo numero per intero riconoscerà quel principio. In un altro contributo — un’analisi di centonove progetti realizzati da docenti italiani — il risultato più solido è che i lavori migliori sono quelli in cui il patrimonio locale non fa da sfondo decorativo, ma diventa il criterio con cui si giudica se un output generato è plausibile o falsato. Due percorsi indipendenti, in due paesi diversi, che arrivano alla stessa conclusione: il territorio è ciò che permette di accorgersi di un errore che nessuno segnala.
L’autrice lo dice con una formula che vale la pena isolare, perché è la più utile del saggio: la sfida centrale non è l’accesso alla tecnologia, ma la coerenza pedagogica. Immagini generate, modelli tridimensionali e ambienti immersivi possono apparire storicamente autorevoli anche quando poggiano su prove incomplete o inesatte. Il problema non è che siano falsi: è che sono plausibili, e la plausibilità è più difficile da smontare della falsità.
Va segnalato, infine, ciò che questo articolo dichiara di non essere.
L’abstract stabilisce con precisione i propri limiti: il gruppo di partecipanti è ridotto e in parte sovrapposto fra le due esperienze, uno dei due casi è analizzato retrospettivamente, mancano misure standardizzate dell’apprendimento — e da tutto questo l’autrice conclude che non è possibile avanzare affermazioni causali o generalizzabili. Il modello viene presentato come una struttura da mettere alla prova, non come un risultato acquisito.
È una scelta rara, e in un dibattito dove abbondano i modelli presentati come validati merita di essere nominata. Un lettore si fida di un autore che gli dice per primo che cosa il proprio lavoro non dimostra.
Il contributo esce in inglese, nella lingua in cui è stato pensato. È una scelta editoriale coerente con un principio che questa rivista ha assunto: che ciascuno debba poter essere letto nella lingua in cui ha ragionato, e che la traduzione sia strumento di mediazione, non sostituzione delle voci. In questo numero convivono tre lingue.
Carmine Marinucci
Student contributors to the pilot implementation: Elif Kırmacı and Ata Kırmacı
Abstract
The purpose of this study is to introduce a pedagogical model for developing digital cultural heritage projects. The Digital Heritage Learning Model (DHLM), distinguishes itself from other educational approaches by incorporating fieldwork, alongside digital tools, as a core component. Digital heritage projects increasingly employ artificial intelligence (AI), immersive technologies, and multimedia; however, the use of multiple tools alone does not constitute a coherent pedagogy.
DHLM is an emerging eight-phase model connecting heritage framing and learning goals, historical inquiry, field observation, evidence validation, STEAM-based co-design, responsible digital production, digital storytelling and public communication, and assessment, preservation, and reuse
DHLM was developed through a qualitative developmental multiple-case study informed by design-based research principles. Two related projects were examined: City of Troy, initiated in 2023 and retrospectively analysed as an antecedent case, and Echoes of the Hippodrome: A Digital Journey Through Time and Motion, which was treated as the principal pilot case. Both projects were selected for Europeana programmes, publicly presented, and externally evaluated.
Historical inquiry, field observation, collaborative production, 3D-based spatial reconstruction, and public communication were documented in both projects, whereas AI-assisted production and digital storytelling were more prominent in the principal pilot project.
The findings suggest that evidence- and place-based inquiry has the potential to support responsible digital production and digital storytelling. However, the small and partially overlapping participant group, consisting of students aged 14–18, the retrospective examination of the City of Troy project, and the absence of standardised learning measures preclude causal or generalisable claims regarding the model’s effectiveness.
DHLM is therefore presented as an emerging interdisciplinary pedagogical model with potential for adaptation across K–12 education and for connecting the humanities, arts, sciences, mathematics, and technology through cultural heritage inquiry and digital production. However, its adaptation to different subjects, age groups, heritage sites, and educational settings requires further development and empirical validation.
Keywords: cultural heritage education; digital competence; artificial intelligence; immersive technologies; digital storytelling; design-based research; STEAM education; place-based learning.
1. Introduction
Digital transformation has changed how learners access information, communicate, collaborate, create media, and participate in public culture. Education must therefore develop not only operational skills but also information literacy, critical evaluation, digital content creation, problem-solving, safety, and ethical judgement. DigComp 2.2 brings these competencies together and emphasises critical, safe, responsible, and human-centred engagement with artificial intelligence (AI) systems (Vuorikari et al., 2022).
Cultural heritage offers an authentic and interdisciplinary context for developing such competencies. Heritage learning requires students to connect material traces with historical sources, compare interpretations, recognise uncertainty and missing perspectives, and communicate conclusions to wider audiences. It also enables them to examine relationships among the past, place, collective memory, identity, and community. Digital technologies can support contextualised and place-based heritage learning (Mendoza et al., 2015), while cultural heritage education can promote critical thinking, experiential learning, collaboration, identity formation, and digital-transformation competencies (Orphanidou et al., 2024). Digital cultural heritage education has expanded across heritage-related disciplines; however, the integration of digital literacy, heritage knowledge, and pedagogical design remains an important educational challenge (Münster et al., 2021).
The central challenge, however, is not access to technology but pedagogical coherence. AI-generated images, three-dimensional models, animations, and virtual environments may appear historically authoritative even when based on incomplete or inaccurate evidence. Generative AI further blurs the distinction among historical sources, scholarly interpretations, digital reconstructions, and creative invention. Digital heritage education therefore requires a structured process that begins with evidence and place, treats digital production as interpretation, and makes accuracy, uncertainty, authorship, copyright, ethics, accessibility, and audience awareness visible.
Existing approaches provide important foundations for such a process. Project-based learning supports authentic inquiry, collaboration, and sustained production (Blumenfeld et al., 1991; Markula & Aksela, 2022). Digital storytelling connects research, scripting, multimodal creation, revision, sharing, and reflection (Kaeophanuek et al., 2019). Immersive and spatial technologies can integrate history and archaeology with science, engineering, design, and place-based experience (Liritzis et al., 2021; Dordio et al., 2024). Three-dimensional modelling can also be combined with AI-assisted contextual narratives (Yu & Hu, 2025). Nevertheless, these components are often implemented separately or organised around particular technologies. A more integrated pedagogical framework is needed to connect historical inquiry, field observation, evidence validation, STEAM interpretation, responsible AI use, digital production, public communication, assessment, and continuity.
DHLM is an emerging eight-phase model connecting heritage framing and learning goals, historical inquiry, field observation, evidence validation, STEAM-based co-design, responsible digital production, digital storytelling and public communication, and assessment, preservation, and reuse. These phases are iterative rather than strictly linear. New evidence, expert feedback, technical challenges, user responses, or reflective evaluation may require learners to revisit earlier decisions. The model’s distinctive contribution lies not in the individual technologies employed but in the pedagogical connections established among evidence, place, interpretation, production, communication, and reflection.
DHLM was developed through the examination of two related digital cultural heritage projects. City of Troy, initiated in 2023, is retrospectively analysed as an antecedent case that anticipated several principles later formalised in the model. Echoes of the Hippodrome: A Digital Journey Through Time and Motion is examined as the principal pilot case in which the developing model became more explicit. The two cases provide a documentary basis for identifying recurring design principles and examining how these principles evolved into the eight-phase DHLM.
This study does not seek to establish the statistical effectiveness of DHLM or make causal claims about learning outcomes. Instead, it adopts a qualitative developmental multiple-case design informed by design-based research principles. Its contribution is developmental and conceptual: to make the pedagogical logic underlying the projects explicit, examine its representation across two related cases, and identify aspects requiring further refinement and empirical validation.
Accordingly, the study addresses the following research questions:
- What recurring pedagogical design principles can be identified across the two cases, and how did they inform the development of DHLM?
- How were the eight phases of DHLM represented across the antecedent and principal pilot cases?
- What opportunities, challenges, and refinement needs emerged from the two cases?
By addressing these questions, the paper presents DHLM as a developing pedagogical model intended for adaptation, systematic refinement, and future validation across different cultural heritage sites, learner groups, and educational settings.
2. Theoretical Foundations
2.1 Design-Based Research as a Model-Development Orientation
Design-based research (DBR) connects empirical educational research, theory-informed design, and implementation in authentic learning environments (Design-Based Research Collective, 2003). Rather than isolating variables under controlled conditions, it examines how educational designs function within the complexity of practice. DBR is therefore appropriate for developing a usable pedagogical model while generating principles that can inform subsequent applications.
Wang and Hannafin (2005) describe DBR as pragmatic, theoretically grounded, interactive, iterative, integrative, and contextual. These characteristics inform DHLM, in which evidence, learner decisions, field observations, technological constraints, and reflection shape the developing design. However, the present study represents an early DBR-informed cycle rather than completed iterative validation. It offers an emerging model, two related cases, digital artefacts, and reflective documentation. Further cycles are required to collect systematic learning evidence, compare adaptations, and refine the model.
2.2 Project-Based and Collaborative Learning
DHLM organises learning around a cultural heritage question rather than isolated software exercises. Digital tools are used only when they support documented historical interpretation and communication. Research on K–12 project-based learning indicates that PBL can promote collaboration, artefact creation, problem-centred inquiry, research, presentation, and reflection. However, aligning student activity, inquiry questions, and learning goals may remain challenging (Markula & Aksela, 2022). Consequently, an attractive product should not conceal weak historical inquiry. DHLM makes the research question, source trail, design rationale, and reflection visible alongside the final artefact.
Collaboration involves differentiated contributions to a shared interpretive product. Roles may include research, field documentation, modelling, scripting, AI prompting, editing, and licensing review.
2.3 Heritage Education, Place, and Field Observation
Heritage education becomes more meaningful when learners connect historical information with the physical, social, and symbolic qualities of place. Emerging technologies can support situated, customised, and contextualised heritage learning (Mendoza et al., 2015). Nevertheless, DHLM retains the importance of direct or carefully mediated engagement with the heritage site, even when the final product is virtual or cinematic.
Field observation draws attention to scale, orientation, surviving monuments, missing structures, changed urban features, present-day uses, sound, and movement. Examining these elements helps learners distinguish what survives, what is historically documented, what is inferred, and what is imaginatively reconstructed. (Orphanidou et al., 2024).
2.4 Digital Storytelling as Interpretation and Public Communication
Digital storytelling connects research, concept development, scriptwriting, storyboarding, media production, revision, sharing, and reflection (Kaeophanuek et al., 2019). This process is particularly relevant to heritage education because it makes the transformation of evidence into a public narrative visible.
2.5 Immersive Technologies and Spatial Historical Thinking
Three-dimensional modelling, augmented and virtual reality, photogrammetry, interactive environments, and code-based visualisation require learners to translate historical evidence into spatial decisions. Even simplified models raise questions concerning proportion, orientation, material, relationships, and detail. Such practices can connect history and archaeology with science, imaging, engineering, and design (Liritzis et al., 2021), while extended reality can function as a didactic resource in cultural heritage education (Dordio et al., 2024).
DHLM treats spatial reconstruction as an epistemic activity because modelling exposes gaps and uncertainties in available knowledge. Learners must seek additional evidence, compare interpretations, state assumptions, and distinguish documented features from plausible reconstruction and creative additions. This position is consistent with the Seville Principles, which emphasise scientific validity, transparency, and documentation in computer-based archaeological visualisation (López-Menchero Bendicho, 2013). Research combining 3D modelling and AI-supported narratives also indicates potential for developing cultural awareness, knowledge, design skills, and user experience (Yu & Hu, 2025).
2.6 Digital and AI Competence
DHLM connects historical inquiry with information literacy; collaborative production with communication; modelling and media creation with digital-content development; technical challenges with problem-solving; and copyright, privacy, attribution, and representation with responsible participation (Vuorikari et al., 2022). UNESCO’s AI competency framework similarly foregrounds human-centred practice, ethics, AI knowledge, pedagogy, and professional learning (Miao & Cukurova, 2024).
Within DHLM, AI is therefore treated as a supervised creative and interpretive tool rather than a historical authority. Learners document significant prompts and revisions, compare outputs with sources, identify fabricated or stereotypical elements, and decide what to revise or reject. Visual realism and linguistic fluency are not historical evidence; responsibility for interpretation remains with the human team.
3. The Proposed Digital Heritage Learning Model
This study proposes the Digital Heritage Learning Model (DHLM), which integrates the investigation, interpretation, digital representation, and public communication of cultural heritage. DHLM is the outcome of an early-stage model-development study informed by design-based research. It is therefore presented not as a universal method whose effectiveness has been demonstrated across different contexts, but as a pedagogical framework grounded in theory and sequential project experiences that requires further development through future research.
The model does not regard digital heritage education solely as the production of digital artefacts. Historical inquiry, field observation, evidence validation, multiple perspectives, STEAM-based co-design, responsible AI use, user testing, assessment, and digital preservation are combined within a single learning cycle.
3.1. Basis for Model Development
DHLM was developed from three sources: the relevant theoretical literature, design experience gained from the earlier City of Troy project, and the research and production processes documented in the principal pilot.
The City of Troy is treated not as an independent validation case but as an antecedent case that contributed to the initial design decisions underlying DHLM. It demonstrated the importance of evaluating historical sources, planning the narrative, and connecting different disciplines before digitally representing cultural heritage.
The principal pilot enabled more systematic examination of the relationships among historical inquiry, field observation, digital production, and reflective assessment. Comparing field data with historical sources, recognising uncertainty in reconstruction, subjecting AI-generated content to human review, and redesigning production in response to technical constraints contributed to the model’s iterative structure.
Not all components of the model are claimed to have been tested equally in the pilot. Evidence validation, accessibility, community participation, and long-term preservation were implemented only to a limited extent or incorporated into the model in response to identified needs. DHLM was developed primarily for learners aged 14–18. Although it may be adapted for other age and learner groups, such adaptations require further testing.
3.2. Theoretical Positioning and Distinctive Contribution
DHLM’s approach to technology is consistent with the AI Readiness Framework, which proposes that technologies should be selected according to educational aims, participant needs, and implementation contexts (Luckin et al., 2022). Digital tools are therefore chosen not for their novelty but for their contribution to cultural heritage inquiry and learning goals.
The co-design dimension supports collaboration among learners, educators, researchers, heritage professionals, local communities, and technology specialists (Luckin & Cukurova, 2019). Consistent with a human–AI hybrid intelligence approach, AI is positioned not as an autonomous producer that replaces learners’ research and creative thinking, but as a tool that supports human decision-making and requires human oversight (Cukurova, 2025).
DHLM’s distinctive contribution lies in combining immersive and generative technologies with critical source analysis, field observation, explicit treatment of uncertainty, multiple perspectives, public communication, and long-term preservation within an eight-phase learning cycle.
3.3. The Eight Phases of the Model
The phases below describe the proposed structure of DHLM rather than components fully implemented in both cases.
Phase 1: Heritage framing.
The cultural heritage element, research problem, learning objectives, target audience, and relevant stakeholders are identified. Before deciding what digital product to create, learners clarify the problem they will investigate and the competencies they are expected to develop. This establishes alignment between technology and pedagogical purpose.
Phase 2: Historical inquiry and critical source analysis.
Learners examine primary and secondary sources and evaluate their reliability, context, and perspectives. The aim is to distinguish among evidence, interpretation, and assumption while identifying voices that may be absent from the available sources.
Phase 3: Field observation and heritage data collection.
Historical information is connected with the heritage site’s present physical and social condition. Photographs, videos, audio, drawings, measurements, or location data may be collected through physical or virtual field visits. Learners examine traces that have survived, changed, or disappeared, as well as current risks affecting the heritage.
Phase 4: Evidence validation and multiple perspectives.
Collected information is compared across sources and examined through expert input and, where appropriate, community dialogue. Evidence may be classified for digital representation as “verified,” “probable,” “hypothetical,” or “unknown.” These categories are not absolute levels of certainty but working tools that make interpretive and reconstruction decisions transparent.
Phase 5: STEAM-based interpretation and co-design.
Evidence is interpreted through the perspectives of science, technology, engineering, the arts, and mathematics. Learners and relevant stakeholders develop design ideas, sketches, storyboards, and production plans. This phase supports interdisciplinary problem-solving, creativity, and collaboration.
Phase 6: Responsible digital production and reconstruction.
Evidence and explicitly stated assumptions are transformed into outputs such as 3D models, augmented or virtual reality applications, animations, interactive maps, digital games, videos, or virtual exhibitions. When AI is used, its purpose and the extent of human intervention should be disclosed. AI outputs should be compared with sources, revised, or rejected when necessary. Copyright, attribution, data protection, and permissions should be addressed throughout production.
Phase 7: Digital storytelling, user testing, and public communication.
The digital product is tested with target users for usability, historical comprehensibility, technical performance, cultural sensitivity, and the clarity of distinctions between evidence and assumption. Following revision, the product is shared through a digital narrative appropriate to the intended audience. Accessibility measures, including subtitles, alternative text, audio description, multilingual options, and clear language, are planned according to the implementation context.
Phase 8: Assessment, reflection, preservation, and reuse.
The learning process and digital product are assessed in terms of historical accuracy, source transparency, STEAM integration, responsible AI use, creativity, collaboration, accessibility, and cultural sensitivity. Rubrics, portfolios, self-assessment, peer assessment, or expert review may be used. Licensing, archiving, updating, and reuse in light of new evidence are also planned.
3.4. Iterative Structure, Scope, and Limitations
DHLM is not a fixed linear sequence. New sources may require a return to historical inquiry, incomplete field data may require further observation, expert feedback may lead to the reclassification of evidence, and user responses may prompt changes to the digital narrative.
Evidence-based inquiry, human oversight, source transparency, responsible AI use, inclusion, cultural diversity, copyright, and sustainability are cross-cutting principles. The extent to which these principles are addressed should be reported transparently in each implementation.
DHLM’s current contribution is not the definitive demonstration of pedagogical effectiveness, but the systematic articulation of relationships among historical evidence, field data, multiple perspectives, STEAM co-design, responsible digital production, public communication, and long-term preservation. Further research across different heritage sites, age groups, and technological contexts is needed to examine the model’s usability and effects on learning.
Figure 1 : The Digital Heritage Learning Model’s Diagram
Figure 2 : Teacher-guided checklist
To facilitate classroom implementation, Figure 2 presents a teacher-guided quality assurance checklist for the eight phases of DHLM. The checklist supports teachers and learners in reviewing evidence, ethics, inclusion, transparency, accessibility, and sustainability throughout the process. It is intended as a flexible framework rather than a prescriptive implementation protocol.
4. Method
4.1. Research Design
This study adopted a qualitative developmental multiple-case design informed by DBR principles. Two related digital cultural heritage projects were examined to assess how recurring design principles contributed to DHLM’s development. Rather than testing causal effectiveness, the study focused on the model’s developmental logic and areas requiring refinement.
4.2. Case Selection and Research Context
Purposive case selection was used. The cases were selected because they were developed by the same project leader, included two students who participated in both projects, focused on digital cultural heritage, and occupied sequential positions in the development of DHLM. However, their research functions and participant scopes differed.
Initiated in 2023, City of Troy was retrospectively examined as an antecedent case developed before DHLM. The project was admitted to the Europeana Low-Code Fest, publicly presented, evaluated by a jury, and received second-place recognition and an honourable mention. Because it was not originally designed to test DHLM, it was not treated as an independent validation case.
Echoes of the Hippodrome: A Digital Journey Through Time and Motion, developed during the fifth edition of the Europeana Build with Bits programme, served as the principal pilot case. It combined historical inquiry and field observation with 3D modelling, AI-assisted media, and digital storytelling. The shift from a planned interactive virtual environment to a cinematic video because of technical and time constraints was examined as a significant design revision.
4.3. Participants and Roles
Both projects were led by Hatice Kırmacı, who was responsible for conceptual design, historical inquiry, pedagogical structuring, coordination, reporting, and the integration of production components.
Six different students participated across the two projects. Elif Kırmacı and Ata Kırmacı also contributed to the digital narrative and content production, whereas the other four students undertook research and presentation tasks.
Elif Kırmacı and Ata Kırmacı subsequently participated in the principal pilot. Ata Kırmacı worked on AI-assisted video production and editing, while Elif Kırmacı developed 3D models of the Hippodrome and the Milion. This partial participant overlap enabled the examination of developmental continuity between the projects but limited their treatment as fully independent cases.
4.4. Data Sources
The dataset comprised project reports, research and field notes, production plans, scripts, storyboards, audiovisual records, 3D models, AI-assisted content, editing materials, final digital artefacts, reflective documents, and project presentations. Programme admission records and available jury and mentor feedback were also examined.
The retrospective analysis of City of Troy was limited to project-level reports, shared research outputs, production records, presentation recordings, external evaluation materials, and final digital artefacts. The individual views, personal information, and educational achievement of the six students were not analysed as research data.
In the principal pilot, the initial knowledge and perceptions of the two student co-creators were explored through oral questions before the project and revisited after completion. Because no standardised instrument was used, responses were not scored, and the interviews were not fully and systematically transcribed, these assessments were treated as exploratory qualitative evidence of students’ experiences and perceived changes rather than as pre-test and post-test data.
4.5. Data Analysis
Data were examined through comparative document and digital artefact analysis using both inductive and deductive approaches. Each case was first analysed separately in terms of activities, evidence, participant roles, design decisions, digital outputs, revisions, challenges, and feedback. Recurring practices formed the initial codes.
During the analysis, the general representation of DHLM’s phases across the two cases was examined through the available project documents and digital outputs. The historical information used in the projects was gathered from various sources and checked for general accuracy by comparison with the available literature. However, during the digital storytelling process, historical information was combined with creative interpretation. Therefore, the resulting narratives were regarded not as complete or definitive reconstructions of the past, but as educational and creative interpretations informed by historical sources.
4.6. Trustworthiness and Researcher Positionality
The interpretation was informed by available project documents, presentations, digital outputs, and external feedback. Programme admission, public presentation, awards, and jury feedback were considered contextual information rather than independent scientific evidence of pedagogical effectiveness. As the researcher was also the designer and leader of the projects, the possibility of interpretive bias was acknowledged. Accordingly, the findings were presented cautiously and without making causal or generalisable claims.
4.7. Ethical Considerations and Methodological Boundaries
Voluntary consent was obtained from all students, and parental permission was secured for research and publication. No sensitive personal data or official educational achievement records were used. The researcher’s role as project leader and parent of two participants was acknowledged as a potential source of bias.
Given the small number of participants, partial overlap between the cases, and absence of standardised measures, no causal or generalisable claims of effectiveness were made. The study is limited to the development of DHLM and the examination of its initial feasibility.
5. Findings
This section examines the development of DHLM through the City of Troy antecedent case and the Echoes of the Hippodrome principal pilot case. The findings draw on a comparative review of project documents, field and production records, digital artefacts, presentations, and available feedback. The analysis considers how the phases of DHLM were represented across the two cases, while recognising differences in the extent and quality of the available documentation. It examines documentary evidence of the model’s development rather than its pedagogical effectiveness.
5.1. City of Troy: Antecedent Case
Conducted in 2023, City of Troy preceded DHLM and was retrospectively examined as an antecedent case. The project combined field observation, historical research, 3D modelling based on a city plan(The city plan was obtained from theRepublic of Türkiye Ministry of Culture and Turism by the participants) , digital storytelling, and public presentation. Its selection for the Europeana Low-Code Fest and subsequent jury recognition provided evidence of external evaluation rather than pedagogical effectiveness. Because the process was not systematically documented according to the DHLM phases, the project was used to identify early design principles contributing to the model, not as its implementation or validation.
5.2. Echoes of the Hippodrome: Principal Pilot Case
5.2.1. Historical and Spatial Focus
Echoes of the Hippodrome examined the Hippodrome of Constantinople, located beneath and around present-day Sultanahmet Square, as a social and political centre. The surviving obelisks and the Serpent Column provided spatial reference points for interpreting the largely lost historical environment.
The narrative connected present-day Istanbul with its Byzantine past through a journey from the Milion to the Hippodrome. Theodora was selected as one of the central characters because of the political leadership she demonstrated during the Nika Revolt. This choice aimed to make women’s roles more visible within political histories that have often been presented through male-dominated narratives and to encourage students to think critically about women’s leadership. The project thereby addressed gender equality and inclusive historical representation alongside cultural heritage education.
5.2.2. Historical Inquiry and Field Observation
The team investigated the Hippodrome’s architecture, public function, racing factions, and the Nika Revolt using written and visual sources. During fieldwork at Sultanahmet Square, the students recorded original photographs and videos and examined the relationship between surviving monuments and lost structures.
Field observation enabled information from historical sources to be compared with the present site and generated original material for the digital narrative. This established a documented connection between historical inquiry and place-based learning.
5.2.3. AI-Assisted Characters and Responsible Digital Production
Theodora and Justinian were interpreted through AI-assisted images and video sequences informed by historical mosaics. The characters were treated not as exact historical reconstructions but as source-informed narrative representations.
Ata Kırmacı worked on the production and editing of six AI-assisted video sequences. The process demonstrated that visual plausibility and historical accuracy must be assessed separately. However, a complete production log covering all prompts, outputs, and revisions was not maintained. This finding highlighted the need for future applications to document the prompt–output–verification–revision chain systematically.
5.2.4. Spatial Reconstruction and STEAM-Based Co-Production
Elif Kırmacı created 3D models of the Hippodrome and the Milion, enabling structures that no longer survive to be incorporated into the digital narrative. The modelling process connected historical and spatial inquiry with geometry, scale, design, and digital production.
The models were not treated as photorealistic or complete historical reconstructions. Their production required decisions balancing available evidence, technical skills, time, and feasibility, illustrating the educational use of simplified models whose assumptions and limitations are explicitly stated.
5.2.5. Digital Storytelling, Public Communication, and Design Revision
An interactive virtual environment was initially planned; however, time and technical constraints led to the production of a cinematic video combining original field footage, 3D models, AI-assisted characters, narration, and sound. The project was publicly presented, and feedback was received from the jury and mentors. However, systematic user testing based on predefined criteria for usability, historical comprehensibility, accessibility, and cultural sensitivity was not conducted.
5.2.6. Assessment, Reflection, and Continuity
Project documentation helped distinguish original field records from student-created and AI-assisted content. However, standardised learning assessment, open licensing, version control, and long-term digital preservation were not fully implemented. Future plans include adding new narratives and AR/VR applications and expanding student participation.
5.3. Comparison of the Two Cases Across the Eight DHLM Phases
Table 1 compares the two cases across the eight DHLM phases and summarises the extent to which each phase was documented in the available project evidence.
| DHLM phase | City of Troy | Echoes of the Hippodrome | Comparative finding |
| 1. Heritage framing, learning objectives, and stakeholders | Partially documented | Documented | Both cases defined the topic and target audience; learning objectives and roles were more explicitly documented in the principal pilot. |
| 2. Historical inquiry and critical source analysis | Documented | Documented | Both cases used historical and visual sources; the principal pilot more explicitly connected these sources with field observations and representational decisions. |
| 3. Field observation and heritage data collection | Partially Documented | Documented | Fieldwork records were limited in the antecedent case; original photographs and videos were collected in the principal pilot. |
| 4. Evidence validation and multiple perspectives | Partially documented | Partially documented | Multiple sources were used, but expert validation, community dialogue, and systematic evidence classification were not implemented. |
| 5. STEAM-based interpretation and co-design | Documented | Documented | Both cases combined historical content, design, and digital production; student roles were more explicitly differentiated in the principal pilot. |
| 6. Responsible digital production and spatial reconstruction | Documented | Documented | Both cases produced digital content; the principal pilot combined AI, 3D models, and field footage, although a complete AI production log was not maintained. |
| 7. User testing, digital storytelling, and public communication | Partially documented | Partially documented | Both projects were publicly presented and received feedback, but systematic user testing was not conducted. |
| 8. Assessment, reflection, preservation, and reuse | Partially documented | Partially documented | Assessment activities and continuity plans were present, but long-term preservation and reuse were not fully tested. |
Table 1. Comparison of the cases across the eight phases of DHLM
The classifications are based on available project documents, digital artefacts, presentations, and feedback. “Documented” indicates the presence of traceable evidence supporting a phase, not demonstrated pedagogical effectiveness.
5.4. Design Development Across the Cases
The comparison indicates that DHLM was developed not from a single project but through the systematisation of implementation experiences, design decisions, and limitations across two sequential projects. Both projects involved historical inquiry, field visits, 3D modelling, interdisciplinary production, public presentation, and online access to project outputs through dedicated websites. The principal pilot extended this structure through AI-assisted digital storytelling and design revisions during implementation. However, structured user testing was not conducted in either project. DHLM is therefore not merely a summary of successful practices but an emerging pedagogical model grounded in documented experiences and identified limitations.
6. Discussion
The comparison of the two cases indicates that the eight phases of DHLM were represented at different levels. Heritage framing directed the projects towards historical and pedagogical purposes; historical inquiry established an evidence-based foundation for digital representation; and field observation connected historical narratives with the present physical environment.
STEAM-based interpretation and co-design brought together knowledge from history, architecture, mathematics, art, technology, and engineering. Students’ differentiated roles in research, 3D modelling, AI-assisted video production, and digital storytelling demonstrated that specialised contributions could be integrated into a shared cultural heritage product.
Responsible digital production and spatial reconstruction demonstrated that digital creation is not merely technical but an epistemic process involving the selection, interpretation, and visual representation of historical evidence. Visually or linguistically persuasive AI outputs are not necessarily historically accurate. They should therefore be reviewed by humans, compared with relevant sources, and documented. Evidence-based reconstruction should also be distinguished clearly from creative completion.
Digital storytelling integrated research, field footage, 3D models, and AI-assisted content into a coherent narrative. Public presentations and external feedback supported communication but did not constitute systematic user testing.
Assessment, reflection, preservation, and reuse were partially represented through project reports, oral assessments, design revisions, and continuity plans. Although the oral assessments provided exploratory evidence of students’ perceived changes, the absence of standardised instruments prevents their interpretation as definitive evidence of learning gains. Similarly, file documentation supported continuity, but open licensing, version control, long-term preservation, and reuse were not fully tested.
DHLM synthesises authentic inquiry, collaboration, and public production from project-based learning (Blumenfeld et al., 1991; Markula & Aksela, 2022); contextualisation from heritage-technology research (Mendoza et al., 2015); research, production, sharing, and reflection from digital storytelling (Kaeophanuek et al., 2019); spatial interpretation from immersive heritage education (Liritzis et al., 2021; Dordio et al., 2024); and critical, responsible, and human-centred AI practices from DigComp and UNESCO (Vuorikari et al., 2022; Miao & Cukurova, 2024).
The model’s distinctive contribution lies not simply in sequencing these elements but in establishing traceable relationships among them. Heritage framing connects technology with learning objectives; historical inquiry defines the evidential boundaries of representation; field observation links knowledge with place; validation makes uncertainty visible; STEAM co-design integrates disciplines and participant roles; responsible production requires scrutiny of AI, copyright, and representational decisions; user testing and public communication examine comprehensibility and accessibility; and assessment and preservation support improvement and reuse.
Another potential contribution of DHLM is its capacity to broaden interdisciplinary participation. Although project-based and STEAM approaches are theoretically open to multiple disciplines, their implementation may concentrate on particular areas of science, technology, or design. Digital cultural heritage projects can bring history, geography, literature and language, art, science, mathematics, technology, and citizenship education into a shared process of inquiry and production. Historical source analysis may support historical thinking; the examination of place and environment may foster geographical thinking; reconstruction may apply scale and proportional reasoning; narrative development may involve language competencies; visual design may encourage artistic expression; and digital production may support technological competence. Realising this potential, however, requires more than allocating separate tasks to different subjects. Explicit connections must be established among learning objectives, production decisions, and assessment criteria.
Digital competence is made visible in this process not only through software use but through purposeful information seeking, source evaluation, distinguishing evidence from assumption, collaborative decision-making, content creation, technical problem-solving, ethical judgement, and product improvement.
Teachers should select technologies that support the heritage learning objectives clearly and responsibly, rather than prioritising technological complexity.
In conclusion, the cases do not demonstrate the pedagogical effectiveness of DHLM or the complete validation of all its phases. Instead, they offer preliminary evidence that workable connections can be established among inquiry, field observation, co-design, digital production, spatial reconstruction, storytelling, and public communication. Evidence validation, community dialogue, systematic user testing, accessibility, standardised assessment, and long-term preservation should be examined more systematically in future DBR cycles. DHLM is therefore positioned not as a completed method but as an emerging pedagogical model developed through sequential project experiences and requiring further empirical refinement.
7. Limitations and Future Research
The study is limited to two related cases developed by the same project leader, with partially overlapping participants and a total of six students. The retrospective analysis of City of Troy, the use of Echoes of the Hippodrome as the only principal pilot, and differences in the scope of student data across the cases limit the transferability of the findings. Although the available documentation supports an examination of DHLM’s development, it is insufficient to establish the model’s educational effectiveness.
Future studies should involve larger and more diverse groups and combine DigComp-aligned rubrics, historical-reasoning tasks, digital artefact analysis, observations, interviews, and, where appropriate, pre-test and post-test measures. Multi-site research should examine the model’s transferability, differentiated student roles, age-appropriate AI transparency, accessibility, and the appropriate level of reconstruction detail.
A three-cycle design-based research programme could focus on classroom feasibility, refinement through student and teacher feedback, and validation across different institutions. Changes to the model and unsuccessful design decisions should be documented in each cycle to strengthen the empirical foundations of DHLM.
8. Conclusion
DHLM connects evidence-based historical inquiry and field observation with interdisciplinary co-design, responsible digital production, public communication, assessment, and preservation within an iterative learning cycle.
City of Troy contributed to the model’s initial design principles as an antecedent case in which relationships among historical inquiry, digital storytelling, spatial representation, and public presentation emerged. Echoes of the Hippodrome extended this structure through field observation, original data collection, differentiated student roles, AI-assisted production, 3D reconstruction, and cinematic storytelling. The design revision undertaken during the principal pilot demonstrated the importance of selecting technological tools according to pedagogical purposes.
DHLM does not assume that technology automatically improves heritage learning. Its central proposition is that educational value depends on evidence-based inquiry, transparent interpretive decisions, learner authorship, ethical practice, public communication, and reflective assessment. The model also has the potential to connect history, geography, language, art, science, mathematics, and technology around a shared cultural heritage problem. However, its usability and effects on learning—including evidence validation, user testing, accessibility, and long-term preservation—should be examined systematically across different educational and heritage settings.
References
Blumenfeld, P. C., Soloway, E., Marx, R. W., Krajcik, J. S., Guzdial, M., & Palincsar, A. (1991). Motivating project-based learning: Sustaining the doing, supporting the learning. Educational Psychologist, 26(3–4), 369–398. https://doi.org/10.1080/00461520.1991.9653139
Cukurova, M. (2025). The interplay of learning, analytics and artificial intelligence in education: A vision for hybrid intelligence. British Journal of Educational Technology, 56, 469–488. https://doi.org/10.1111/bjet.13514
Design-Based Research Collective. (2003). Design-based research: An emerging paradigm for educational inquiry. Educational Researcher, 32(1), 5–8. https://doi.org/10.3102/0013189X032001005
Kaeophanuek, S., Na-Songkhla, J., & Nilsook, P. (2019). Developing web-based learning through digital storytelling to enhance cultural heritage preservation. Journal of Information Science, 37(1), 45–66. https://doi.org/10.14456/jiskku.2019.3
Liritzis, I., Volonakis, P., & Vosinakis, S. (2021). 3D reconstruction of cultural heritage sites as an educational approach: The Sanctuary of Delphi. Applied Sciences, 11(8), 3635. https://doi.org/10.3390/app11083635
López-Menchero Bendicho, V. M. (2013). International guidelines for virtual archaeology: The Seville Principles. In C. Corsi, B. Slapšak, & F. Vermeulen (Eds.), Good practice in archaeological diagnostics (pp. 269–283). Springer. https://doi.org/10.1007/978-3-319-01784-6_16
Luckin, R., & Cukurova, M. (2019). Designing educational technologies in the age of AI: A learning sciences-driven approach. British Journal of Educational Technology, 50(6), 2824–2838. https://doi.org/10.1111/bjet.12861
Luckin, R., Cukurova, M., Kent, C., & du Boulay, B. (2022). Empowering educators to be AI-ready. Computers and Education: Artificial Intelligence, 3, 100076. https://doi.org/10.1016/j.caeai.2022.100076
Markula, A., & Aksela, M. (2022). The key characteristics of project-based learning: How teachers implement projects in K–12 science education. Disciplinary and Interdisciplinary Science Education Research, 4, Article 2. https://doi.org/10.1186/s43031-021-00042-x
Mendoza, R., Baldiris, S., & Fabregat, R. (2015). Framework to heritage education using emerging technologies. Procedia Computer Science, 75, 239–249. https://doi.org/10.1016/j.procs.2015.12.244
Miao, F., & Cukurova, M. (2024). AI competency framework for teachers. UNESCO. https://unesdoc.unesco.org/ark:/48223/pf0000391104
Münster, S., Fritsche, K., Richards-Rissetto, H., Apollonio, F., Aehnlich, B., Schwartze, V., & Smolarski, R. (2021). Teaching digital cultural heritage and digital humanities: The current state and prospects. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XLVI-M-1-2021, 471–478. https://doi.org/10.5194/isprs-archives-XLVI-M-1-2021-471-2021
Orphanidou, Y., Efthymiou, L., & Panayiotou, G. (2024). Cultural heritage for sustainable education amidst digitalisation. Sustainability, 16(4), 1540. https://doi.org/10.3390/su16041540
Vuorikari, R., Kluzer, S., & Punie, Y. (2022). DigComp 2.2: The Digital Competence Framework for Citizens—with new examples of knowledge, skills and attitudes. Publications Office of the European Union. https://doi.org/10.2760/115376
Wang, F., & Hannafin, M. J. (2005). Design-based research and technology-enhanced learning environments. Educational Technology Research and Development, 53(4), 5–23. https://doi.org/10.1007/BF02504682
Yu, Y., & Hu, W. (2025). Three-dimensional modeling and AI-assisted contextual narratives in digital heritage education: Course for enhancing design skill, cultural awareness, and user experience. Heritage, 8(7), 280. https://doi.org/10.3390/heritage8070280