Abstract
The rapid integration of Web 2.0 technologies and generative artificial intelligence (GenAI) is transforming digital content production in e-learning environments, raising new questions about coherence, pedagogical intent, and design accountability. Within this evolving ecosystem, digital storytelling has emerged not merely as an engagement technique, but as a structural design strategy that organizes interaction, progression, and meaning-making across multimodal learning materials. This study adopts a qualitative, design-oriented approach grounded in artefact-based analysis to examine how digital storytelling is operationalized within contemporary content production platforms. Rather than focusing on learner outcomes or user perceptions, the study systematically analyzes publicly accessible Web 2.0, AI-supported, game-based, and immersive digital artefacts to identify recurring narrative design logics and affordances. Drawing on thematic analysis and theoretical synthesis, the study proposes the AI-Supported Digital Storytelling Design Model (AIDSTM), which conceptualizes storytelling across four interrelated dimensions: narrative generation, narrative structuring, narrative enactment, and narrative immersion. The model also introduces an optional governance layer to address provenance, ethical responsibility, and quality assurance in AI-supported content creation. The findings contribute design-oriented knowledge for e-learning by offering an analytical framework that supports intentional storytelling-based design, critical evaluation of digital tools, and informed integration of generative AI into online learning environments.
Keywords
Introduction
Digital content production is evolving from the ‘participatory, collaborative and interactive’ production logic of Web 2.0 to a more ‘agentic’ ecosystem that can rapidly prototype, automatically generate multimedia components (text–image–audio–video) and produce personalised content streams with generative artificial intelligence (GenAI). In an educational context, this transformation lowers content production barriers for instructional designers and teachers while simultaneously highlighting new requirements such as originality, academic integrity, copyright/provenance, alignment with learning objectives, and critical evaluation (Khan and Saunderson, 2024). From a digital media perspective, this shift also transforms storytelling into a mode of meaning-making that reorganises how educational narratives are produced, circulated, and experienced across platforms.
In this new production ecosystem, storytelling is not merely ‘more entertaining narration’; it emerges as a design principle that supports learners’ meaning-making processes, integrates multimedia components into a pedagogical whole, and structures interactive content design (e.g., branching scenarios, task-based flows, gamified narratives). Digital storytelling has been associated with cognitive, affective, and social benefits in education for years; it has been reported as an approach that supports learner motivation, self-efficacy, identity development, and discipline-specific thinking skills (Robin, 2016; Wu and Chen, 2020).
On the other hand, Web 2.0-based content development tools (e.g., H5P, Genially/Vyond-like interactive/animation production environments) with GenAI-based production tools (e.g., text-to-video/audio/image production) within the same workflow both accelerates ‘storytelling-based’ content design and increase the need for quality assurance and ethical governance (Park, 2015; Saravi et al., 2025).
This article focuses on the following question without collecting participant data (without obtaining data from humans). (1) How can storytelling be systematised in digital content production using Web 2.0 content tools and generative artificial intelligence, and through which theoretical components can it be evaluated?
To this end, the study: (i) synthesises the current literature, (ii) grounds the artefact analysis approach through the real tool/product ecosystem, and (iii) proposes a conceptual model/framework that can be used in the article.
This study is positioned as a design-oriented conceptual analysis rather than an empirical investigation of learner outcomes. Its contribution lies in theorizing how digital storytelling is structurally embedded within contemporary content production environments through tools, templates, and design affordances. By adopting an artefact-based analytical approach, the study aims to generate transferable design knowledge and a conceptual framework to support intentional e-learning design in AI-supported ecosystems, rather than to test instructional effectiveness.
Literature review
Digital storytelling: A pedagogical and design ‘framework’ in education
Digital storytelling (DST) is a multimedia-based production/learning practice that supports learners’ meaning-making by integrating narrative with visual, auditory and textual components. Systematic reviews on the use of DST in education show that it is becoming widespread at different levels and in different disciplines; it has been associated with outcomes such as language development, motivation, reflection, and identity/belonging (Gómez-Martín et al., 2025; Wu and Chen, 2020).
PRISMA-based reviews conducted particularly in the 2024–2025 period emphasise that the forms of DST application have diversified (classical linear narrative → interactive/game-like narrative → AI-supported narrative) and that evaluation criteria have shifted more towards the axes of ‘design quality’ and ‘learning outcome alignment’ (Deslis et al., 2025; Gómez-Martín, 2025). These developments indicate that digital storytelling has increasingly shifted from a learner-produced artefact toward a design-oriented instructional strategy, a transition that has been significantly accelerated by the affordances of Web 2.0 content creation tools.
Interactive storytelling with web 2.0 tools: ‘The democratisation of design’
Web 2.0 content tools have facilitated the scaling of DST in classroom and distance learning contexts by enabling the creation of interactive experiences without requiring coding. Recent applied studies further illustrate how Web 2.0 tools such as Genially enable educators to design interactive, game-based storytelling experiences without advanced technical expertise. For instance, Kuncoro (2025) documents the development of Genially-based interactive learning media within a game-based learning framework, highlighting how design responsibilities shift from professional developers to teachers as content designers. While such studies primarily focus on implementation outcomes, they exemplify the broader democratisation of design facilitated by Web 2.0 storytelling tools.
Similarly, platforms such as H5P offer active learning support by transforming the ‘story flow’ into learning tasks using formats such as interactive video, quizzes, drag-and-drop, and branching scenarios. There are applied studies showing that content developed with H5P can increase student participation and performance (Martín-Alguacil et al., 2025; Sharmin et al., 2024).
Narrative creation with generative artificial intelligence: Speed, personalisation, and the need for ‘critical evaluation’
GenAI enables the rapid production of certain components of a narrative (draft text, characters, scenes, dialogue, visual storyboards, voice-over text, even videos). However, the literature emphasises that this speed also increases the risk of ‘facilitation’ in education; therefore, the use of GenAI must be addressed alongside literacy, ethics, and critical verification competencies (Khan and Saunderson, 2024; Park, 2015).
A systematic review focusing specifically on GenAI literacy reports that in 2023–2024 studies, students’ understanding of GenAI remained at an intermediate level; they struggled with prompt design and critical evaluation of outputs; and ethical issues (privacy, data security, academic integrity) became central (Chandel and Lim, 2024; Fang and Du, 2026).
There are also experimental findings showing that GenAI-supported storytelling applications can produce advantages in terms of interaction and retention in language learning (Namaziandost and Çakmak, 2025).
The (optional) contribution of Web3: Ownership, provenance and trust
The question you asked, ‘Can Web 3.0 be added to the AI-supported part?’, has a logical positioning in the literature: Web3 brings discussions of content ownership, source/provenance traceability, and trust into the field of education with components such as decentralised identity (DID), smart contracts, and distributed storage (Duarte et al., 2025).
Recent conceptual and prototypical work illustrates how Web3 technologies may be integrated into educational platforms to address issues of ownership, provenance, and trust. For example, da Costa et al. (2025) present ALEX, a reimagined LMS architecture that combines Web3, peer-to-peer infrastructures, and active learning principles. While such approaches remain exploratory, they demonstrate how decentralised systems can support transparent content ownership, traceable learning artefacts, and trust-building mechanisms without relying on centralised control. In the context of AI-supported digital storytelling, these developments suggest a potential governance layer rather than a mandatory technological shift.
In this article, it is more academic and safer to position Web3 not as a mandatory technology, but as an optional layer for managing the source, version, licence, and production traces (metadata) of narrative content generated by GenAI.
Theoretical framework
Constructing the theoretical backbone of this article in three layers both grounds ‘narration’ in a pedagogical foundation and makes the Web2 + GenAI (+Web3) combination analytical.
Narrative-based learning and multimedia composition
Narrative-based learning conceptualizes learning as a meaning-making process in which experiences are structured through story. Within this perspective, digital storytelling (DST) operates not merely as a representational technique, but as a cognitive and pedagogical scaffold that integrates content, emotion, and agency into a coherent learning trajectory. By coordinating text, visuals, audio, and temporal sequencing, DST organizes fragmented information into structured narrative wholes (Robin, 2016).
Rather than foregrounding technological novelty, Robin (2016) emphasizes purpose-driven storytelling, audience awareness, reflective narration, and multimodal composition as core design principles. In this sense, narrative coherence functions as a design constraint: multimedia elements are not simply assembled, but deliberately selected and sequenced to support instructional intent.
This position resonates with multimedia learning theory, which underscores the importance of integrating verbal and visual representations into cognitively manageable structures. Narrative provides an organizing schema that guides attention, reduces fragmentation, and situates new information within temporal and causal relations. Story therefore acts as a compositional logic that binds multimedia components into an intelligible learning pathway rather than a collection of discrete assets.
Within the StoryCenter tradition, Lambert’s “cookbook” approach further operationalizes narrative-based learning as a structured production workflow. By foregrounding purpose, emotional resonance, voice, visual language, and sharing context, this framework embeds pedagogical intentionality throughout the storytelling process. Importantly, storytelling is conceptualized as a design workflow rather than a post-production embellishment.
For the present study, these traditions are significant because they frame storytelling as a design logic rather than a learner outcome. As Web 2.0 platforms, AI-supported tools, and immersive environments increasingly automate multimedia production, narrative-based learning theories provide a critical lens for examining whether pedagogical coherence is maintained within technologically mediated storytelling processes.
Learner activity and interaction: Story → scenario → task
While narrative-based learning emphasizes coherence and composition, interactive digital environments extend storytelling into structured learner activity. Web 2.0 tools—such as H5P’s branching scenarios and interactive videos, Genially’s layered navigation, and animation-based narrative flows—transform storytelling from a viewed narrative into a decision-oriented scenario.
This transformation aligns with Moore’s theory of interaction, in which learning is shaped by structured learner–content, learner–system, and learner–learner interactions (Kumtepe et al., 2019; Moore, 1989). In interactive storytelling environments, narrative progression unfolds through learner choices and system feedback, positioning interaction as a constitutive mechanism of meaning-making rather than an auxiliary engagement feature.
Empirical studies of Web 2.0 instructional designs indicate that tools such as H5P enhance learner participation and perceived learning experience when narrative flow is integrated with feedback and conditional progression (Martín-Alguacil et al., 2025; Sharmin et al., 2024). Similarly, research on AI-supported language learning environments shows that scenario-based and dialogic AI applications foster active engagement and learner agency when AI functions as a responsive narrative partner (Sisman-Ugur et al., 2026).
From a design perspective, this shift marks a movement from narrative as representation to narrative as enactment. Story becomes a structuring mechanism for tasks, decisions, and procedural progression. Digital storytelling thus operates not only as a pedagogical frame, but as an architectural principle that organizes learner activity across digital environments in alignment with active and experiential learning paradigms.
GenAI literacy, ethics, and quality assurance
GenAI integration accelerates narrative production while simultaneously increasing the designer’s responsibility for evaluation, verification, and ethical alignment. For this reason, dimensions of GenAI literacy—such as understanding, critical use, evaluation, and ethical awareness—must be linked to design criteria rather than treated solely as learner outcomes (Süner-Pla-Cerdà et al., 2025).
Recent scholarship frames GenAI not as an autonomous content producer but as a co-creator that supports active and experiential learning processes (Conaldi and De Vita, 2025), reinforcing the view that pedagogical intent and quality assurance remain human responsibilities. Within this study, GenAI literacy is therefore conceptualized as a design-oriented competence guiding the analysis of AI-supported storytelling artefacts.
Web3 layer (optional): Provenance and content governance
Web3 may be positioned as a governance-oriented layer that addresses provenance, licensing, versioning, and ownership of GenAI-generated storytelling components. In educational contexts, decentralized identity and traceability mechanisms have been discussed in relation to user sovereignty, transparency, and trust (Calzada et al., 2025). Within this framework, Web3 is not treated as a mandatory technological shift, but as a conceptual layer that supports responsible management and verification of AI-generated educational artefacts.
However, this governance layer should not be interpreted as a deterministic technological solution. Decentralised infrastructures introduce their own complexities, including scalability, accessibility, and regulatory ambiguities. Accordingly, Web3 is framed here as a conceptual lens for examining provenance and accountability rather than as a prescriptive technological upgrade.
Methodology
Research design
This study employs a qualitative, design-oriented research design based on artefact analysis and theoretical synthesis. The purpose is not to evaluate learning outcomes or user perceptions, but to examine how digital storytelling is embedded and operationalized through the designable features of contemporary content production environments integrating Web 2.0 tools and generative AI.
Artefacts are treated as valid units of analysis when the research interest concerns design logic, representational structures, and observable interface/workflow features rather than participant behaviour (Krippendorff, 2011, 2018). This orientation is consistent with design research traditions in educational technology that prioritize the development of transferable design knowledge and conceptual models (Bell et al., 2013; Reeves, 2006).
Focusing on platforms as artefacts is also aligned with work in networked learning and learning design which distinguishes between elements that can be deliberately designed (e.g., tools, resources, environments) and processes that emerge through use; in this study, the analytical emphasis is placed on the former—i.e., the designed structures that resource and guide storytelling activity (Goodyear et al., 2016). In addition, scholarship on multimodal digital artefacts emphasizes that such artefacts can function as analyzable learning tools whose meaning-making potential is shaped through multimodal composition and observable design decisions (Hellwig, 2022).
Data sources: Selection of digital artefacts
Rationale for artefact selection
The data sources consist of real, publicly accessible digital content production platforms that support storytelling-oriented design across Web 2.0, AI-supported, game-based, and immersive environments. Platforms were examined as design artefacts rather than as collections of user-generated products.
Artefacts were selected using a purposeful, theory-driven sampling strategy based on the following criteria. (1) Authenticity and public accessibility, (2) Explicit storytelling and pedagogical affordances, (3) Transparency and observability of design features, (4) Representativeness of contemporary digital design paradigms, (5) Suitability for artefact-based analysis without reliance on learner data.
These criteria align with established principles of document and artefact analysis emphasizing transparency, accessibility, and analytical relevance (Bowen, 2009; De Villiers, 2005; Vermeir et al., 2017).
Artefact selection criteria and their justifications.
This selection strategy follows established recommendations for document and artefact analysis, which emphasize transparency, accessibility, and analytical relevance when artefacts constitute the primary units of analysis (Bowen, 2009).
Selected artefacts and their design characteristics
Artefacts were grouped according to the primary mode through which storytelling is operationalized.
Web 2.0 platforms (H5P, Genially, Padlet) were selected to represent environments where storytelling is structured through interaction, multimedia composition, and pedagogical sequencing. H5P enables explicit narrative structuring through interactive video and branching scenarios; Genially supports non-linear and exploratory storytelling through visual layering and interaction; Padlet facilitates collaborative and spatial narrative construction.
AI-supported tools were included as narrative generators and design accelerators rather than pedagogical agents. These environments enable rapid production of narrative drafts and multimodal components, shifting the designer’s role from authorship to orchestration, curation, and evaluative judgment (Khan and Saunderson, 2024; Park, 2015). Analysis focused on narrative generation capacity, prompt–output structure, and flexibility for pedagogical alignment.
Game-based environments (Twine, Scratch) were selected to examine storytelling enacted through choice, progression, and rule-based interaction. Twine exemplifies branching narrative logic, while Scratch externalizes narrative flow through block-based computational structures. Unity and similar engines were considered conceptually in terms of their capacity to embed narrative within spatial and task-based environments.
Immersive platforms (Spatial.io) were included to represent storytelling embedded within spatial and experiential contexts, highlighting the transition from narrative representation to narrative experience and aligning with theories of presence and embodied meaning-making (Makransky and Petersen, 2019).
Data collection procedure
Data collection in this study did not involve human participants or human-generated data. Instead, it consisted of a systematic examination of digital artefacts as primary data sources. Specifically, data collection included. (1) Examination of platform interfaces and content creation workflows, (2) Analysis of storytelling-related templates, formats, and design features, (3) Review of publicly available documentation and exemplar structures, (4) Identification of narrative-related functions such as sequencing, branching, interaction, and immersion.
No user-created content, learning analytics, or performance data were collected. The unit of analysis was the design affordance of each artefact rather than its use or impact.
This procedure aligns with established document and artefact analysis approaches, in which texts, interfaces, and representational structures constitute the primary data corpus (Bowen, 2009).
Data analysis strategy
Analytical framework
Data were analyzed using qualitative thematic analysis informed by content analysis principles (Krippendorff, 2018). An a priori coding scheme was developed based on the theoretical framework and literature on digital storytelling and instructional design.
The main analytical dimensions included. (1) Narrative Generation: How narratives are created (human-authored, AI-supported, agentic). (2) Narrative Structuring: Sequencing, branching, interaction, and pedagogical alignment. (3) Narrative Enactment: Game mechanics, choice, progression, and task-based storytelling. (4) Narrative Immersion: Spatiality, presence, and experiential storytelling. (5) Design Affordances: Flexibility, scalability, and integration potential.
These analytical dimensions collectively informed the construction of the AIDSTM, which synthesises recurring design logics observed across artefact categories.
Analytical procedure
The analysis followed four iterative steps. (1) Artefact familiarization: Examination of each platform’s storytelling-related features. (2) Coding: Mapping observed features onto the predefined analytical dimensions. (3) Pattern identification: Identifying recurring design logics across artefact categories. (4) Model synthesis: Integrating findings into the AIDSTM.
This iterative process reflects established practices in qualitative synthesis and design-based inquiry (Reeves, 2006).
Trustworthiness and methodological rigor
To ensure rigor and transparency. (1) All artefacts are real and publicly accessible. (2) Analytical categories are grounded in established literature. (3) Interpretations are limited to observable design features, avoiding speculative claims about learning outcomes.
Such strategies are recommended to enhance credibility and dependability in qualitative and document-based research.
Methodological limitations
The study deliberately excludes learner data, performance measures, and user perceptions. Consequently, findings are limited to design affordances and conceptual implications, not empirical effectiveness. This limitation is acknowledged as a methodological choice aligned with the study’s aim to develop a design-oriented conceptual model.
In AI-supported content production environments, key pedagogical decisions are increasingly embedded in tools, templates, and automated workflows rather than in observable learner behavior alone. For this reason, artefact-based analysis offers a methodologically appropriate lens for examining design logic in contemporary e-learning systems, where instructional intent may precede or even bypass traditional learner-facing interventions.
Design findings from artefact-based analysis
This section presents the findings of the artefact-based analysis, focusing on how digital storytelling is operationalized as a design strategy across Web 2.0, AI-supported, game-based, and immersive content production environments. The findings are derived from observable design patterns and narrative structures embedded within the selected digital artefacts, rather than from learner outcomes or performance measures.
The analysis identified four interrelated storytelling dimensions—narrative generation, narrative structuring, narrative enactment, and narrative immersion—which collectively provide the empirical grounding for the proposed AIDSTM.
Narrative generation: From authorship to orchestration
Across AI-supported artefacts, storytelling is primarily enacted through generative mechanisms rather than direct human authorship. The analysis shows that narrative elements—such as plot outlines, character descriptions, dialogues, and visual scenes—are produced through prompt-driven or rule-based generation processes.
In these environments, the role of the designer shifts from composing narratives linearly to orchestrating narrative production by specifying constraints, prompts, and evaluative criteria. This design logic is observable in interfaces that foreground prompt input, iterative regeneration, and selective refinement of outputs (see Figure 1 for an illustrative interface example). Example interface illustrating prompt-driven narrative generation and iterative refinement (screenshot from an AI-supported content generation environment).
Importantly, narrative generation in AI-supported tools is modular and provisional. Generated story elements are not fixed narratives but components that can be recombined, revised, or discarded. This modularity distinguishes AI-supported storytelling from traditional digital storytelling practices and positions narrative generation as an iterative design process rather than a finished product.
The analysis of AI-supported storytelling artefacts indicates a clear shift from traditional notions of authorship toward narrative orchestration. Generative AI systems function as design accelerators that produce provisional narrative drafts, visual storyboards, and multimodal components, which are subsequently curated and refined by human designers. This shift aligns with recent reviews on generative AI literacy, which emphasise that critical competencies in AI-supported content creation involve prompt design, output evaluation, verification, and ethical judgment rather than automated production (Khan and Saunderson, 2024; Park, 2015).
The observed prompt–output–revision cycle supports emerging arguments in the literature that generative AI reshapes educational content design workflows by repositioning educators and instructional designers as orchestrators and quality controllers, rather than sole authors (Park, 2015). Consequently, narrative generation in AI-supported environments must be evaluated not by speed or volume of production, but by the designer’s capacity to align generated narratives with pedagogical intent and ethical constraints.
Narrative structuring in web 2.0 environments
In Web 2.0–based platforms, storytelling is operationalized through explicit structural mechanisms that organize narrative flow, interaction, and pedagogical sequencing. The analysis identified recurring design features that transform stories into navigable and interactive learning experiences.
Platforms such as H5P enable narrative structuring through branching scenarios, interactive videos, and embedded feedback mechanisms. Figure 2 shows an example of a branching scenario generated with h5p. Screenshot from h5p.
The analysis of H5P illustrates how digital storytelling can be operationalized through explicit pedagogical sequencing. Branching scenarios, interactive video, and embedded feedback mechanisms render narrative flow observable as a sequence of learner choices and instructional responses, positioning storytelling as an instructional design strategy rather than a representational format.
This observation is consistent with studies positioning H5P as a tool for interactive and active learning design, emphasizing learner choice, conditional progression, and formative feedback rather than linear content delivery (Martín-Alguacil et al., 2025; Sharmin et al., 2024). In this context, storytelling functions as a scaffold for decision-making and reflection, aligning with task-based and scenario-based learning principles widely discussed in the educational technology literature. Importantly, the visibility of branching paths and conditional progression in H5P makes pedagogical decision-making structurally transparent. The narrative does not merely present content but encodes instructional contingencies within its architecture. In this sense, the artefact functions as a pedagogical script materialised through interface design.
These formats make narrative progression observable as a sequence of learner choices and instructional responses (Figure 3). Similarly, Genially supports non-linear narrative paths through layered visual elements and user-triggered interactions, allowing multiple narrative trajectories to coexist within a single artefact. Figure 3 shows the Genially content creation screen. Genially content creation screen.
Genially exemplifies a Web 2.0 environment where storytelling is structured through non-linear navigation, visual layering, and user-triggered interactions. Genially exemplifies a Web 2.0 environment in which narrative meaning emerges through non-linear navigation, visual layering, and user-triggered interactions. This design logic aligns with contemporary perspectives that conceptualise interactive narratives as exploratory spaces rather than linear texts (Mariani and Ciancia, 2023). By enabling multiple entry points and interaction paths, Genially supports a design logic in which storytelling is enacted through discovery, reinforcing contemporary approaches to learner-centered and inquiry-oriented content design.
Padlet, by contrast, operationalizes narrative structure through spatial and chronological organization. Figure 4 shows the Padlet content screen. Padlet content screen.
Padlet operationalizes storytelling as a socially constructed and spatially organized artefact, where narrative elements are collaboratively arranged through card-based, timeline-based, or spatial layouts. This flexibility allows narrative structure to remain open, iterative, and negotiable rather than fixed, highlighting storytelling as a configurable design process.
This observation aligns with studies highlighting Padlet’s role in collaborative knowledge construction and shared meaning-making in digital learning environments (e.g., collaborative reflection and co-creation practices). Within the context of digital storytelling, Padlet illustrates how narrative structure can remain open, iterative, and socially negotiated rather than fixed.
Narrative enactment in game-based environments
Game-based artefacts operationalize storytelling through enactment, where narratives unfold as a consequence of user actions, choices, and progression rules. In these environments, stories are not primarily told but performed. Twine exemplifies this logic through its explicit choice–consequence structures.
Figure 5 shows Google’s game development and content creation platform. Google’s game development platform.
Narrative meaning emerges as users navigate branching paths, making decisions that determine subsequent narrative states. In Figure 6, Scratch similarly externalizes narrative logic through event-driven programming blocks, rendering story progression visible and inspectable. Scratch’s screen.
In more advanced engines such as Unity (see Figure 7) (examined conceptually), narrative enactment is embedded within spatial scenes, tasks, and progression mechanics. Although programming complexity varies, the shared design pattern is the transformation of narrative from representational content into procedural experience. Unity screen capture.
Key narrative enactment mechanisms observed across game-based storytelling artefacts.
This finding is consistent with literature on game-based and interactive narratives, which conceptualizes storytelling as an enacted process emerging from player actions and system rules (Aylett, 2022; Young and Cardona-Rivera, 2011). Twine, in particular, exemplifies branching narrative logic, where meaning is shaped by cumulative choices rather than linear plot development. Similarly, Scratch externalizes narrative flow through block-based conditional logic, making story structure explicit and inspectable.
Such environments align with theoretical perspectives that frame learning and meaning-making as processes embedded in interaction, agency, and procedural engagement, reinforcing the relevance of enactment-focused analysis in artefact-based research.
Table 2 summarizes key enactment mechanisms observed across game-based storytelling artefacts.
Rather than representing isolated features, these enactment mechanisms collectively illustrate how narrative meaning is operationalized through interaction, rule structures, and learner choice, reinforcing storytelling as an experiential and procedural design logic in e-learning environments.
Narrative immersion and spatial storytelling
Immersive platforms extend digital storytelling into spatial and experiential domains, where narrative meaning is constructed through presence, navigation, and interaction within digital environments.
In platforms such as Spatial.io, storytelling is embedded within virtual spaces rather than presented through linear sequences. Narrative cues are conveyed through spatial arrangement, object placement, and embodied movement, positioning users as situated participants within the storyworld (Figure 8). Screenshot illustrating spatial narrative cues and embodied navigation in an immersive storytelling environment.
This design pattern reflects a shift from narrative representation to narrative experience, where meaning emerges through exploration and embodied engagement rather than through explicit exposition.
The examination of immersive storytelling platforms, represented by Spatial. io, highlights a transition from narrative representation to narrative experience. In these environments, storytelling is embedded within spatial navigation, presence, and embodied interaction, positioning users inside the narrative space rather than as external observers.
This finding corresponds with immersive learning literature emphasizing the roles of presence, embodiment, and spatiality in meaning-making processes (Makransky and Petersen, 2019). From a design perspective, immersive platforms shift storytelling from sequential narration to experiential configuration, where narrative coherence is achieved through environmental cues, movement, and situated interaction.
Importantly, this study does not claim enhanced learning outcomes associated with immersion. Instead, it identifies immersion as a distinct narrative enactment mechanism, expanding the analytical scope of digital storytelling beyond screen-based and interaction-driven paradigms.
Cross-dimensional synthesis of findings
The analysis demonstrates that digital storytelling is not confined to a single design layer but is distributed across multiple environments and mechanisms.
Across artefact categories, the findings demonstrate that digital storytelling in contemporary content production environments is not a singular method but a multi-layered design strategy encompassing narrative generation, structuring, enactment, and immersion. These layers correspond to distinct but interrelated design logics that are shaped by platform affordances rather than by learner data or instructional interventions.
By grounding these observations in current literature, the study reinforces the position that artefact-based analysis offers a valid and necessary lens for understanding how storytelling is operationalized within AI-supported and Web-enabled educational design ecosystems.
Mapping of artefact categories to storytelling design dimensions.
The indicators in Table 3 represent dominant design orientations rather than exclusive capabilities, acknowledging that narrative dimensions may overlap across artefact categories.
Table 3 highlights that narrative enactment in game-based and immersive storytelling environments is primarily realized through mechanisms of choice, progression, and rule-based interaction rather than through linear content delivery. These mechanisms transform storytelling from a representational structure into an experiential process, where meaning emerges through action, decision-making, and consequence. In e-learning contexts, such enactment mechanisms serve as functional substitutes for instructor-led guidance, enabling learners to actively construct narrative coherence and learning trajectories within digital environments.
This synthesis highlights storytelling as a layered and complementary design strategy, rather than a single method or tool-specific feature.
Relation of findings to the proposed model
The design patterns identified across artefact categories provided the direct analytical foundation for the development of the AIDSTM. Each storytelling dimension corresponds to a distinct yet interconnected layer within the model, providing empirical grounding for its conceptual structure (see Figure 9). AIDSTM.
This study deliberately excludes learner data, performance measures, and user perceptions in order to focus on digital storytelling as a design phenomenon rather than an instructional outcome. Consequently, the findings do not make claims regarding learning effectiveness, engagement levels, or cognitive gains. This limitation is not a methodological weakness but a theoretical positioning aligned with design-oriented research traditions in educational technology.
By prioritizing artefacts, interfaces, and design affordances as primary units of analysis, the study foregrounds how narrative meaning is encoded within tools, templates, and workflows—an aspect that is often obscured in participant-based empirical studies. Such a focus enables the identification of transferable design logics that operate across platforms and technological paradigms.
However, this approach also constrains the interpretive scope of the findings. The proposed AIDSTM should therefore be understood as a conceptual and analytical framework, not as an empirically validated instructional model. Its applicability lies in supporting instructional designers, educators, and researchers in analyzing, comparing, and intentionally structuring storytelling-oriented digital content rather than predicting learner outcomes.
Future research may extend this work by empirically examining how the identified narrative generation, structuring, enactment, and immersion mechanisms interact with learner characteristics, disciplinary contexts, and instructional goals. Mixed-method and design-based research studies could test the model in authentic learning settings, integrating artefact analysis with learner data to build cumulative design knowledge.
Discussion
While existing digital storytelling frameworks tend to foreground narrative coherence, multimedia integration, or learner engagement, the AIDSTM advances the field by repositioning storytelling within contemporary AI-mediated design ecologies. Specifically, the model contributes in three ways: (1) by theorizing the shift from narrative authorship to orchestration under generative AI conditions; (2) by conceptualizing interaction as narrative enactment rather than supplementary engagement; and (3) by introducing a governance-sensitive layer addressing provenance and accountability in AI-generated educational artefacts.
Reframing digital storytelling as a design strategy
The findings suggest that digital storytelling in AI-supported environments should be understood not merely as a pedagogical technique but as a design epistemology—an organizing logic through which meaning, progression, and instructional intention are structured within digital systems.
In earlier digital storytelling traditions, narrative coherence emerged from human-authored linear compositions. In contrast, contemporary ecosystems distribute narrative construction across platforms, templates, algorithms, and interaction mechanisms. Storytelling therefore shifts from expressive production to infrastructural organization.
This reframing extends prior conceptualizations of storytelling as a scaffold (Robin, 2016; Wu and Chen, 2020) by demonstrating that narrative logic is increasingly embedded in design affordances rather than solely in instructional scripts. In AI-mediated environments, narrative coherence is partially delegated to system constraints, regeneration cycles, and interface architectures. The key analytical question thus becomes not whether storytelling “improves learning,” but how narrative structure is materially encoded within digital production systems.
From narrative authorship to narrative orchestration in AI-Supported design
One of the most significant theoretical shifts identified in this study concerns the transformation of authorship. In generative AI environments, storytelling is no longer exclusively authored but orchestrated.
Generative systems operate as proposal engines—producing drafts, multimodal fragments, and narrative variations. However, meaning-making authority remains human. Pedagogical alignment, contextual adaptation, ethical judgment, and epistemic responsibility are not automated functions.
This distinction is crucial in avoiding technological determinism. By positioning AI as a design partner rather than an epistemic agent, the study aligns with emerging scholarship cautioning against conflating automation with pedagogical intelligence. Narrative orchestration thus becomes a form of design literacy requiring prompt precision, evaluative discernment, and provenance awareness.
Importantly, this shift also redistributes responsibility: designers become curators of machine-generated narrative possibilities. The educational significance lies not in AI’s productivity, but in the human capacity to regulate and contextualize generated narratives within coherent learning trajectories.
Beyond workflow transformation, this shift toward orchestration also raises epistemic and infrastructural questions. When narrative components are generated within proprietary AI systems, pedagogical intent becomes partially mediated by opaque algorithmic logics and platform architectures. From this perspective, orchestration is not only a design practice but also a negotiation with socio-technical infrastructures that shape what kinds of narratives can be generated, prioritized, or constrained. Recognising this mediation is essential for maintaining pedagogical intentionality in AI-supported e-learning environments.
Narrative structuring as pedagogical visibility in web 2.0 environments
The Web 2.0 artefacts analysed in this study reveal that storytelling functions as a mechanism of pedagogical visibility. Branching logic, interactive sequencing, and conditional progression render instructional structure observable within the narrative flow.
This structural transparency resonates with interaction-oriented theories of distance education, where instructional design compensates for limited real-time presence by embedding guidance into learning materials. Narrative sequencing, feedback loops, and decision points effectively materialize instructional intent within the artefact itself.
At the same time, collaborative platforms demonstrate that storytelling need not be fully predetermined. In socially negotiated environments, narrative coherence emerges through shared arrangement rather than hierarchical sequencing. These dual patterns—scaffolded structuring and emergent configuration—illustrate storytelling as a flexible design orientation rather than a fixed format.
Interaction as a core design mechanism in digital storytelling
Across artefact categories, interaction operates as the mechanism through which narrative meaning is actualized. Rather than functioning as decorative engagement, interaction constitutes the procedural layer of storytelling.
In structured Web 2.0 tools, interaction links narrative elements to conditional progression and feedback. In game-based environments, storytelling unfolds through rule-governed systems where meaning emerges from cumulative choices. In immersive platforms, narrative is enacted spatially through navigation and presence.
This proceduralization of narrative reflects a broader shift from representational storytelling to enacted storytelling. Meaning is not merely communicated but performed through system–user interaction. Importantly, the study refrains from asserting that increased interaction guarantees improved learning. Instead, interaction is conceptualized as a design affordance that externalizes narrative logic and distributes agency across human–system configurations.
Digital storytelling as a structural strategy in E-Learning design
Within asynchronous e-learning environments, storytelling assumes a structural function. Narrative sequencing, progression cues, and embedded interaction may operate as substitutes for continuous instructor presence, reducing transactional distance by embedding guidance within the artefact.
Fragmentation and disengagement in online learning are often linked to disjointed content structures. Storytelling-oriented design addresses this by providing continuity, temporal coherence, and purposeful progression. Rather than serving as motivational decoration, narrative becomes the organizing infrastructure of the learning experience.
AI-supported workflows further complicate this dynamic. While generative tools accelerate content production, they also risk producing narrative incoherence without deliberate orchestration. Thus, scalability and responsibility must coexist. The structural function of storytelling becomes even more critical when production is automated.
Enactment and immersion: Expanding the boundaries of narrative design
Game-based and immersive environments extend storytelling beyond textual and multimedia representation toward environmental configuration. Here, narrative logic is embedded in systems, spatial layouts, and procedural mechanics.
From a theoretical standpoint, this aligns with perspectives framing learning as situated and embodied. However, the study avoids deterministic claims regarding immersion. Enactment and immersion are identified as expanded narrative affordances—new modalities through which storytelling can structure experience.
The significance lies not in technological novelty but in the diversification of narrative design possibilities.
Implications for the AI-Supported Digital Storytelling Design Model (AIDSTM)
The AIDSTM synthesizes these dimensions—generation, structuring, enactment, and immersion—into an analytical vocabulary for examining storytelling across heterogeneous digital environments.
Rather than prescribing instructional methods, the model clarifies how narrative logics operate within AI-mediated infrastructures. The optional governance layer situates the model within emerging debates on authorship, provenance, and accountability in machine-augmented educational production.
Its contribution lies in making design decisions visible and comparable, thereby supporting reflective and ethically grounded e-learning development.
Positioning the contribution within educational technology research
Finally, the study demonstrates the epistemic value of artefact-based inquiry in AI-mediated educational contexts. As digital infrastructures increasingly preconfigure pedagogical possibilities, examining embedded design logics becomes as critical as measuring learner behavior.
Artefact analysis thus complements outcome-based research by revealing the structural conditions under which learning experiences are produced.
Conclusion and implications for E-Learning practice
Conclusion
This study examined how digital storytelling is operationalized as a design strategy within contemporary digital content production environments integrating Web 2.0 platforms, generative artificial intelligence, game-based systems, and immersive environments. Adopting an artefact-based and design-oriented methodology, the study shifted attention from learner outcomes to the structural design logics embedded in tools, interfaces, and workflows.
The findings demonstrate that digital storytelling in AI-mediated ecosystems operates as a multi-layered design configuration encompassing narrative generation, structuring, enactment, and immersion. These dimensions are not technological features per se, but design logics shaped by platform affordances and interaction architectures. As such, storytelling emerges less as a representational technique and more as an infrastructural organizing principle within contemporary e-learning systems.
By synthesizing these patterns, the study proposes the AI-Supported Digital Storytelling Design Model (AIDSTM) as an analytical framework for examining storytelling-oriented design across heterogeneous digital environments. The model is not intended as an empirically validated instructional intervention, but as a conceptual and comparative tool that enables more deliberate, transparent, and ethically informed design decisions in AI-supported content production.
More broadly, the study contributes to educational technology scholarship by demonstrating the methodological relevance of artefact-based inquiry in contexts where pedagogical practice is increasingly mediated by digital infrastructures. Understanding how narrative logic is encoded within systems becomes essential as generative AI reshapes content production processes and redistributes authorship responsibilities.
Implications for E-Learning practice
Several implications follow for e-learning design and development.
First, digital storytelling should be treated as a structural design principle rather than an engagement add-on. In asynchronous environments characterized by fragmented content and limited real-time instructional presence, narrative sequencing and interaction architectures can provide coherence, orientation, and progression.
Second, interaction should be designed as a meaning-making mechanism embedded within narrative flow. Decision points, branching structures, rule-based progression, and spatial navigation are not peripheral features but central mechanisms through which narrative logic becomes actionable and inspectable.
Third, generative AI can enhance scalability and design agility when framed as a partner in narrative orchestration rather than as an autonomous author. Prompt design, output evaluation, and provenance awareness remain critical human responsibilities. Responsible orchestration, rather than automation alone, determines pedagogical integrity.
Fourth, game-based and immersive storytelling environments expand the narrative design space by enabling procedural and spatial enactment. While not universally required, such affordances offer additional layers for scenario-based and experiential e-learning design.
Finally, as storytelling logic becomes increasingly embedded in templates, algorithms, and production workflows, design transparency and governance become central concerns. The AIDSTM framework offers a shared analytical vocabulary that supports critical reflection, comparison, and accountability in AI-supported digital storytelling practices.
Directions for future E-Learning research
Future research may extend this framework by empirically investigating how the identified storytelling dimensions interact with learner characteristics, disciplinary contexts, and instructional goals. Integrating artefact-based analysis with learner-centered and design-based research approaches could further advance cumulative knowledge on storytelling-oriented e-learning design in AI-mediated environments.
Footnotes
Ethical considerations
This study did not involve human participants, personal data, or identifiable human subjects. The research was based exclusively on the analysis of publicly accessible digital artefacts and platform interfaces. No user-generated data, learning analytics, or private materials were collected or examined. Accordingly, ethical approval was not required under institutional or international research ethics guidelines.
Identifying information and anonymity
There are no identifying details concerning the authors, affiliated institutions, funding bodies, or review boards that could compromise anonymity.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.