Wrapped in Anansi's Web: Unweaving the Impacts of Generative-AI Personalization and VR Immersion in Oral Storytelling

Wrapped in Anansi’s Web: Unweaving the Impacts of Generative-AI Personalization and VR Immersion in Oral Storytelling Ka Hei Carrie Lau Technical University of Munich Munich, Germany carrie.lau@tum.de Bhada Yun University of California, Berkeley Berkeley, California, USA bhadayun@berkeley.edu Efe Bozkir Samuel Saruba Technical University of Munich Munich, Germany ge36sev@mytum.de Enkelejda Kasneci Technical University of Munich Munich, Germany efe.bozkir@tum.de Technical University of Munich Munich, Germany enkelejda.kasneci@tum.de Abstract Oral traditions, vital to cultural identity, are losing relevance among youth due to the dominance of modern media. This study addresses the revitalization of these traditions by reconnecting young people with folklore. We introduce Anansi the Spider VR, a novel virtual space that combines first-person virtual reality (VR) with generative artificial intelligence (Gen-AI)-driven narrative personalization. This space immerses users in the story of Anansi the Spider, empowering them to influence the narrative by envisioning themselves as the protagonists, thereby enhancing personal reflection. In a 2 × 2 between-subjects study with 48 participants, we employed a mixed-method approach to measure user engagement and changes in interest, complemented by semi-structured interviews providing qualitative insights into personalization and immersion. Our results indicate that personalization in VR significantly enhances engagement and cultural interest. We recommend that future studies using VR and Gen-AI to revitalize oral storytelling, maintain cultural integrity, and involve cultural communities to ensure authentic, respectful narratives. CCS Concepts • Human-centered computing → User studies; Empirical studies in interaction design; • Applied computing → Interactive learning environments; Arts and humanities; • Computing methodologies → Intelligent agents. Keywords VR, Gen-AI, Intangible Cultural Heritage, Education ACM Reference Format: Ka Hei Carrie Lau, Bhada Yun, Samuel Saruba, Efe Bozkir, and Enkelejda Kasneci. 2025. Wrapped in Anansi’s Web: Unweaving the Impacts of Generative-AI Personalization and VR Immersion in Oral Storytelling. In This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs International 4.0 License. AHs ’25, Masdar City, Abu Dhabi, United Arab Emirates © 2025 Copyright held by the owner/author(s). ACM ISBN 979-8-4007-1566-2/25/03 https://doi.org/10.1145/3745900.3746103 312 The Augmented Humans International Conference 2025 (AHs ’25), March 16– 20, 2025, Masdar City, Abu Dhabi, United Arab Emirates. ACM, New York, NY, USA, 21 pages. https://doi.org/10.1145/3745900.3746103 1 Introduction “Folklore at its best addresses both tradition and innovation and shows how constancy and change are interlinked in the dynamic process of civilization.” Wolfgang Mieder, 1987 [51] Folklore embodies the balance of tradition and innovation, intertwining constancy and change in the evolution of civilization [51]. This is particularly true of oral traditions, which UNESCO classifies as a crucial aspect of intangible cultural heritage (ICH), describing them as “living expressions” passed down through generations via storytelling, music, and shared practices [79, 94, 95]. Oral traditions play a vital role in bridging the past and present, conveying knowledge, values, and beliefs while shaping worldviews through myths, fables, and folklore [17, 56, 89]. Vansina [99] highlights their importance in enabling societies without written records to reconstruct their history. Despite their cultural significance, the fluid and undocumented nature of oral traditions makes them vulnerable to distortion or loss. Challenges such as colonial legacies [39], globalization [45], and declining interest among younger generations [70, 89] further heighten this vulnerability. These factors underscore the need for innovative strategies to preserve and revitalize oral traditions as traditional methods struggle to remain relevant in the digital age. Traditionally, oral traditions were passed down by storytellers who served as cultural guardians, conveying values and knowledge through vivid, participatory narratives [63, 92]. However, the decline in successors and erosion of traditional practices have weakened their transmission [92]. While projects like Digital Folklore [50] and Folk-Computing [8] have digitized narratives for global access, they often reduce oral traditions to static formats, failing to capture their dynamic and performative essence [32]. This limitation is particularly evident for younger, tech-savvy audiences who expect immersive and interactive experiences [15, 70, 86].

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. Throughout history, oral traditions have adapted to new mediums, maintaining their relevance across evolving cultural landscapes [50, 79, 99]. Unlike prior studies that focus solely on VR immersion or static narrative personalization, our research addresses this challenge by reimagining oral traditions as immersive, firstperson experiences. By integrating the sensory engagement of virtual reality (VR) with the real-time adaptability of generative artificial intelligence (Gen-AI) powered by ChatGPT, our approach fosters cultural reflection while bridging generational and cultural divides, making traditional narratives more relatable and engaging for modern audiences. Gen-AI and VR now serve as transformative tools to extend these traditions from specific cultural communities to global audiences. VR fosters a sense of embodiment, enabling users to deeply immerse themselves in culturally rich narratives [58, 62, 75, 76, 91], while Gen-AI dynamically adapts narratives in real-time, offering personalized and resonant storytelling experiences [24, 91, 108]. Together, these technologies broaden the reach and appeal of oral traditions by making narratives more interactive and relatable [13, 34, 55, 74]. By combining Gen-AI and VR, our work demonstrates how oral traditions can extend beyond their origins to engage audiences from diverse cultural backgrounds. This approach not only revitalizes traditional storytelling but also fosters cultural exchange, ensuring these narratives remain relevant in a rapidly digitizing and interconnected world [11, 26, 102]. In this exploratory study, we investigate how VR and Gen-AI may contribute to revitalizing storytelling through the example of Anansi the Spider. Central to Akan traditions, Anansi’s narratives feature themes of kingship, storytelling, and resilience, with the character often depicted as a resourceful trickster and sage figure [20, 47, 67, 98]. Historically, these tales also served as symbols of resistance during the transatlantic slave trade, particularly in Afro-Caribbean communities [35, 48]. Given the enduring relevance of Anansi’s stories, they offer a suitable case for exploring how emerging technologies can preserve and modernize oral traditions. Accordingly, our research addresses the following questions (RQs): • RQ1: How does Gen-AI-driven personalization affect user engagement and interest in cultural learning among younger audiences during storytelling experiences? • RQ2: How does VR immersion influence user engagement and interest in cultural learning among younger audiences during storytelling experiences? • RQ3: How do personalization and immersion together enhance user engagement and interest in cultural learning, thereby supporting the transmission of oral traditions to younger audiences? To answer our RQs, we conducted a 2 × 2 between-subjects design with 48 participants to evaluate the combination of Gen-AI and VR, specifically focusing on the effects of interaction medium (VR vs. Non-VR) and personalized content (Personalized vs. NonPersonalized) on folklore transmission. To the best of our knowledge, this is the first study to explore how the integration of VR immersion and Gen-AI personalization can dynamically adapt and enhance the transmission of culturally significant oral traditions. 313 Our findings demonstrate that VR significantly enhances engagement and interest in cultural learning, while Gen-AI personalization fosters deeper narrative connections. Semi-structured interviews further highlight how personalized VR storytelling enhances immersion and fosters deeper reflection on the story. The contributions of this paper are fourfold and as follows: System We introduce a novel VR experience of Anansi the Spider1 , blending immersive storytelling with Gen-AI personalization to preserve and modernize oral traditions. Conceptual We propose a framework integrating traditional storytelling with adaptive Gen-AI, using modular prompting for culturally authentic NPC behaviors. Empirical A mixed-methods study shows that VR immersion boosts engagement, while Gen-AI personalization fosters self-reflection and cultural connection. Mapping We present a thematic model (Immersion, Personalization, Engagement, Reflection, Cultural Relevance, Technological Intervention), illustrating how VR and Gen-AI enhance cultural storytelling and inform future research. 2 Related Work In this section, we examine the evolution of narrative technology and the digitalization of oral traditions. First, we explore Gen-AI’s advancements in narrative personalization. Second, we discuss VR’s role in preserving oral traditions and transitioning from static archives to immersive platforms. Finally, we highlight the contributions and limitations of these technologies and their combined potential for creating adaptive, culturally resonant storytelling experiences. 2.1 Gen-AI for Personalized Narrative Experiences Narrative technologies have evolved from basic understanding and generation to now narrative intelligence, where AI can craft, tell, and respond to stories [74]. Early systems, such as the Script Applier Mechanism (SAM) and Plan Applier Mechanism (PAM) [78], focused on interpreting narratives through predefined scripts and plans. Tools like the Storytelling Agent Generation Environment (SAGE) [93] and ‘Scheherazade’ [41] introduced personalized storytelling and paved the way for interactive narrative design. Advancements in natural language processing (NLP) later enabled systems like ‘Mimisbrunnur’ [84] to assist authors in creating story outlines via AI planning, showcasing new adaptability in narrative creation. However, these earlier systems were often limited by static responses, constrained narrative flexibility, and reliance on predefined heuristics. One key challenge in interactive storytelling is the narrative paradox. The idea of balancing user agency in shaping the story while maintaining a coherent, pre-authored plot [4, 75, 91]. Murray [54] defines user agency as “the satisfying power to take meaningful action and see the results of our decisions and choices.” Recent advances in Gen-AI, such as OpenAI’s GPT-4, address previous limitations by offering context-aware, real-time storytelling [105]. 1 https://youtu.be/_jiXVDvB_60?si=s2okXFvdocF9OFWg, last accessed on 19 Novem- ber 2024

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates However, challenges like factual inaccuracies and cultural bias persist [25, 72]. Projects like Turing Quest [24] and NarrativePlay [108] demonstrate the potential of Gen-AI in gamified storytelling, using augmented reality markers and role-playing narratives to enhance interactivity. However, these projects largely focus on fictional scenarios and fall short when adapting to culturally rich oral traditions that aim to educate and preserve cultural authenticity. In contrast, our research leverages Gen-AI to provide personalized responses, maintaining the original plot while fostering cocreative storytelling. This approach addresses the narrative paradox by balancing user agency with narrative coherence and preserving the relational and performative essence of oral traditions. By making storytelling adaptive, relatable, and culturally grounded, our work addresses RQ1. 2.2 VR for Revitalizing Oral Traditions in the Digital Era Technological advancements have transformed how we preserve oral traditions, moving from static archives to immersive platforms. Early tools, such as metadata schemas [44] and interactive systems like ‘i-balls’ [8], improved documentation and facilitated copresent interaction, yet lacked the immersive settings and narrative complexity needed to capture the cultural depth and performative essence of oral storytelling. Younger audiences increasingly prefer interactive, visually rich formats, highlighting the need for more engaging approaches [2, 96]. Recognizing this demand, research such as ‘narrative storyliving’ [97] explores how immersive environments deepen engagement by enabling participants to actively contribute to the narrative. VR immersion shows potential for enhancing user agency in a way that fosters cultural appreciation and deeper engagement. VRbased cultural educational projects have demonstrated that flexible, user-centered adaptations can enrich cultural literacy [2], crosscultural understanding [1] and learning traditional knowledge [87], underscoring the importance of responsiveness in immersive cultural experiences. For example, ‘The Book of Distance’ [62] combines historical narratives with interactive storytelling to evoke empathy, but its linear structure limits adaptability. Similarly, ‘The First-Person VR System’ [58] immerses users in folklore through voice and gesture yet lacks the adaptive personalization needed for diverse audiences. Our work builds on these efforts by integrating dynamic VR environments with real-time Gen-AI personalization. This combination responds to users’ actions within the immersive space, delivering context-aware storytelling. Unlike prior approaches that rely on static cultural representations, purely immersive settings, or linear narratives, our approach emphasizes the performative aspects of oral traditions. By enabling real-time adaptation, we enhance user engagement and cultural appreciation, addressing RQ2. 2.3 Combining Gen-AI and VR for Culturally Adaptive Storytelling Kenderdine et al. [36] introduced the concept of embodied museography, emphasizing immersion, interaction, and participation in digitizing ICH. Similarly, Pollock [69] highlights the dynamic and co-creative nature of oral traditions, where storytelling emerges 314 through the interplay of teller, audience, and context. These insights underscore the importance of combining immersive digital affordances with the relational qualities of oral traditions to create culturally adaptive storytelling experiences. Building on this foundation, our work integrates VR and Gen-AI to preserve oral traditions while enhancing their relevance for culturally diverse audiences. Despite their independent potential in real-time personalization and immersive storytelling, the combined use of Gen-AI and VR for revitalizing oral traditions remains underexplored. Recent studies, such as Hirzle et al. [30] and Bozkir et al. [9], have explored the intersection of AI and extended reality (XR) for personalized cultural experiences. While applications like revitalizing traditional sports [40] and dance [66] showcase the potential of AIdriven VR, they primarily focus on usability, offering limited insight into cultural impact. The Baba Yaga project [19], which adapts Eastern European folklore in immersive environments, represents one of the closest parallels to our research. However, it lacks user-driven voice input to interactively shape narratives, a key innovation of our work. Additionally, the project does not empirically evaluate how immersive storytelling fosters deeper cultural engagement, highlighting a gap that our study aims to address. To bridge these gaps, we adopt a mixed-methods approach, combining semi-structured interviews and quantitative surveys to evaluate user engagement and cultural appreciation, drawing on Bahng et al.’s reflexive dimensions [5]. This strategy fosters user reflection within immersive storytelling contexts and builds on prior work that emphasizes the role of qualitative methods in exploring the broader implications of emerging technologies [29, 49]. By prioritizing cultural sensitivity and user perspectives, we ensure responsible, impactful applications that measure both technological effectiveness and cultural significance, addressing RQ3. Furthermore, introducing voice-enabled interactions and modular Gen-AI prompting for real-time narrative adaptation offers a participatory approach to immersive storytelling. In doing so, this integration bridges the gap between technological innovation and cultural preservation, advancing responsible and meaningful applications of emerging technologies. 3 Methods This study explores the use of Gen-AI and VR technologies as tools for cultural preservation and engagement, focusing on Ghanaian Anansi folktales. To achieve this goal, we designed a user study, approved by the Institutional Review Board (IRB) of Technical University of Munich, and evaluated it through our prototype, Anansi the Spider VR. The prototype features a Gen-AI-driven NPC, Onini, enabling participants to interactively shape the traditional tale. The study compares personalized versus non-personalized story versions delivered through VR and non-VR mediums, evaluating their effects on user engagement, reflection, and cultural understanding. This section provides an overview of the technical implementation of Anansi the Spider VR, apparatus, user study design, procedures, participants, measures, and analyses.

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates 3.1 Lau et al. Technical Implementation of Anansi the Spider VR The background of this VR narrative experience is inspired by the Anansi tales, specifically adapted from the story How it came about that the Sky God’s stories came to be known as “Spider-stories” [73], participants take a first-person perspective, joining Anansi in capturing Onini, the Python, to obtain the stories of the world from the Sky God Nyame. The narrative is streamlined for the VR medium and study constraints while preserving its core themes. Central to this VR experience is Gen-AI, which enables natural, personalized interactions with NPCs, particularly with Onini, the Python. Gen-AI refers to systems capable of producing new content, such as text, speech, or images, based on training data [18]. Here, the VR system leverages multiple Gen-AI models to bring the NPCs to life. OpenAI’s Whisper [59] is used for speech-to-text (STT) conversion, transcribing the player’s spoken input. Then, ChatGPT, powered by GPT-4 [61], generates a contextual response, impersonating the NPC, based on the player’s input. ElevenLabs [60] is used throughout the experience to dynamically generate the voices of the NPCs. Together, these technologies allow the NPCs to respond in real-time. We presented the system architecture in Figure 1. Each interaction begins with a 5-second recording window, followed by processing times ranging from 2 to 7 seconds. Latency primarily arises from ChatGPT’s response generation and ElevenLabs’ voice synthesis, varying with response length and complexity. To ensure narrative consistency and personalized engagement, we employ advanced prompt engineering techniques. Building on reusable patterns for ChatGPT [101], the system assigns roles to ChatGPT (e.g., Onini, the Python) using modular prompts inspired by role-prompting and chain-of-thought reasoning [77, 101]. Prompts are divided into five distinct sections, general instructions, character attributes, state descriptions, scene context, and script instructions, shown in Table 1, allowing dynamic adjustments based on game states. During runtime, these sections are combined and updated in real-time, then sent along with the player’s message as input to ChatGPT. This design ensures NPCs remain coherent with their personalities and adapt seamlessly to player interactions. 3.2 Apparatus The study utilized a Varjo VR-3 (Model HS-6) head-mounted display (HMD) powered by a GeForce RTX 4080, Intel Core i7-13700K, and 32 GB of RAM, achieving 90 frames per second. Interaction in the VR condition was facilitated by HTC Vive Controller 2.0 and Steam VR Base Station 2.0, using Unity’s XR Interaction Toolkit, as depicted in Figure 2a. The non-VR setup, illustrated in Figure 2b, included a Schenker desktop with similar specifications, a Dell P2314T monitor (1920 × 1080 resolution), and a Creative Fatal1ty HS-800 Gaming Headset covering a standard frequency range from 20Hz to 20kHz. This setup enhanced overall immersiveness by delivering clear audio that complemented the visual quality. The VR game was developed in Unity 2021.3.33f1 LTS using the Universal Render Pipeline (URP) for optimized performance. Assets were sourced from the Unity Asset Store, TurboSquid, CGTrader, and Sketchfab, including the “Tropical Forest Pack” [85] Anansi’s spider body [16], Onini the Python [81], and Nyame, depicted by an Ashanti moon mask [82]. Custom elements, such as Anansi’s 315 Figure 1: System architecture of the Player-NPC interaction model. Player speech is converted to text using Whisper, and GPT-4 generates dynamic NPC responses. The orange arrows represent communication between Whisper, GPT-4, and ElevenLabs, which processes player input and converts responses into speech. ElevenLabs handles both dynamic and pre-scripted text-to-speech outputs for NPCs. Table 1: NPC Prompting Techniques in the VR Storytelling Experience. Instruction Description Example General instructions Defines NPC role and behavior. You are a character in an Anansi folklore VR game. Respond in character. Character attributes Specifies NPC personality and traits. You are Onini, the prideful python from Anansi folklore. State description Provides game state based on player actions. The player placed a bamboo beside you, challenging your boast. Scene context Defines the setting where interaction occurs. You are coiled around a tree stump in a jungle clearing in Ghana. Script instructions NPC responses and scripted actions. Stay coiled until persuaded, then proclaim: Behold the longest snake in Ghana!” mask and clay huts, were modeled in Blender specifically for this project. ChatGPT integration for dialogue was implemented using an unofficial OpenAI Unity package [83], enabling dynamic interactions and Whisper services for STT. Eleven Labs, a Gen-AI-powered voice synthesis tool, provided text-to-speech (TTS) synthesis via an API [88]. 3.3 User Study Design We designed a 2 × 2 between-subjects experiment resulting in four conditions: Non-VR Non-Personalized (NVNP) (control), NonVR Personalized (NVP), VR Non-Personalized (VNP), and VR

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates (a) Participant using the Varjo (b) Non-VR setup with Creative VR-3 headset with HTC Vive con- Fatal1ty Headset, Schenker i7trollers for the VR condition. 13700K RTX 4080 desktop, and Dell P2314T monitor. Figure 3: Interaction with Onini in the VR environment, where participants use Speech-to-Text options to persuade Onini to stretch alongside a bamboo stick, as part of their task in the ‘Anansi the Spider VR’ narrative. Figure 2: Experimental setup for VR and Non-VR conditions. Personalized (VP). The two independent variables were personalization (Personalized vs. Non-Personalized) and interaction medium (VR vs. Non-VR). A total of 48 participants (𝑛 = 48) were randomly assigned across these four conditions (𝑛 = 12 each). 3.3.1 Personalization. In this research, we apply “personalization” by tailoring story content, such as character references, names, and key dialogue lines, to each participant. In the Personalized conditions, the narrative explicitly addressed participants using their chosen name and pronouns, and ChatGPT (acting as Onini) generated real-time dialogue in response to their negotiation attempts. Figure 9 (in the Appendix) illustrates this interaction with three user-generated narrations. In contrast, the Non-Personalized conditions referred to a generic character, “Anokye” (a common Akan name), and used only third-person references with pre-scripted dialogues. All conditions followed the same core story structure as shown in Figure 4, ensuring consistent plot progression and key events while permitting minor content adaptations (e.g., negotiation) in the personalized versions. 3.3.2 Interaction Medium. To understand the effect of VR, participants in the VR conditions (VNP, VP) experienced the narrative in an immersive environment as shown in Figures 3 and 4, where they could walk around, interact with objects, and negotiate directly with Onini. In VR Personalized (VP), ChatGPT dynamically responded to a participant’s unique input and used their name in dialogue. In contrast, VR Non-Personalized (VNP) used the same interactive mechanics but referred only to “Anokye” and employed pre-scripted dialogue. In the Non-VR conditions (NVNP, NVP), participants progressed through a slide-based interface as shown in Figure 2b with audio narration. They could still advance through the story, and their interactions were pre-scripted chat elements generated by ChatGPT beforehand due to the nature of the interaction medium. Thus, NVP included each participant’s name and minor adaptive text, whereas NVNP retained only generic references to “Anokye.” 316 Figure 4: Interactive Narrative Path in ‘Anansi the Spider VR’: Figure illustrates the five key stages of the VR narrative. Participants start their journey at Stage 1 in the village, walking along a jungle path to Stage 2 where they meet Anansi, climb Anansi’s web at Stage 3 to receive tasks from the Sky God, search for Onini at Stage 4, and conclude the experience at Stage 5. 3.4 Procedures Each session lasted ≈ 50 minutes, with 30 minutes allocated for briefing, experimentation, and post-assessment, and an additional 20 minutes for semi-structured interviews. Data collection took place in a controlled laboratory setting. Eligible participants were

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. pre-screened based on assessment criteria and invited to take part. Upon arrival, participants were briefed on the research goals and signed informed consent forms, they were informed that they could withdraw at any time if they experienced motion sickness. Following a height calibration step, which is crucial for tasks requiring precise spatial awareness in the VR environment [31, 104], participants received a five-minute tutorial covering VR navigation, NPC interactions, and object manipulation. The experiment began after this preparatory phase. During VR sessions, interaction data were automatically recorded in Unity, capturing metrics such as completion time, NPC interactions, and spatial coordinates. VR sessions concluded when participants clicked the “Exit” button after successfully capturing Onini and returning it to the Sky God, while non-VR sessions concluded with a video fade-out, indicating the end of the experience. After the experience, participants completed a post-assessment questionnaire measuring engagement, story comprehension, and changes in cultural appreciation. Semi-structured interviews were conducted with a select group of 20 participants (saturation point), chosen from each of the four experimental conditions. 3.5 Participants A total of 48 participants, aged 22 to 46 years (𝑀 = 26.75, 𝑆𝐷 = 4.66), including 25 males and 23 females, were recruited from a university student population. A pre-assessment questionnaire measured familiarity with digital storytelling (𝑀 = 3.55, 𝑆𝐷 = 0.98), VR (𝑀 = 2.83, 𝑆𝐷 = 1.00), oral traditions (𝑀 = 2.63, 𝑆𝐷 = 0.94), African culture (𝑀 = 1.58, 𝑆𝐷 = 0.84), and interest in other cultures (𝑀 = 3.63, 𝑆𝐷 = 1.00), as shown in Appendix Table 4. The demographic variance was consistent across the experimental conditions, ensuring a balanced representation in the study. Eligibility required participants to be over 18, fluent in English, and have normal or corrected vision. Those reporting severe motion sickness were excluded. Each participant received a €10 voucher for participation. Semi-structured interviews were conducted with 20 participants from the original sample of 48. We recruited volunteers from each of the four experimental conditions to ensure coverage of different experiences. Interviews were conducted incrementally until data saturation was reached, e.g., when no substantially new insights or themes emerged, drawing on qualitative research principles from Longhurst [43], Hennink and Kaiser [28], and Adams [3]. Table 2 provides demographic details of the interviewed participants. These interviews explored personal reflections on the experience, perceived immersion, and emotional impacts, complementing the quantitative findings. 3.6 Measures We conducted a post-assessment to measure participants’ user engagement using the User Engagement Scale and their responses to a cultural interest questionnaire. Interaction log data were also collected during the VR experiment, including gameplay duration, conversation logs with Onini, and walking distance. The User Engagement Scale Long Form (UES-LF) [65] was used to measure the dependent variables: Focused Attention (FA), Perceived Usability (PU), Aesthetic Appeal (AE), and Reward 317 (RW). The scale consists of 30 items rated from 1 (not at all) to 5 (very much), with randomized questions and reverse coding applied to minimize response bias. Details of the UES-LF items used in this study are provided in Table 6 in the Appendix. The UES-LF was selected for its validated reliability across diverse digital contexts [7, 46, 57] and its ability to capture multiple dimensions of engagement, distinguishing it from tools focused solely on user satisfaction. To evaluate the impact of personalization and medium differences on cultural interests, we conducted a pre-assessment to collect data on participants’ cultural backgrounds and prior exposure to storytelling technologies, including familiarity with VR, as shown in Appendix Table 5. A self-designed cultural interest questionnaire, consisting of five Likert-scale questions rated from 1 (not at all interested) to 5 (very much interested), was used to measure changes in attitudes toward cultural learning and curiosity. The specific questions are presented in Table 7 in the Appendix. This approach aligns with prior research [21, 38, 107], which highlights the importance of assessing cultural impact on user knowledge, emotional engagement, and connection to content. 3.7 Analysis 3.7.1 Quantitative Data. Statistical analyses were conducted using Python 3.12.5 [71]. Due to violations of normality and homogeneity of variance, the Kruskal-Wallis test was applied to both UES and Cultural Interest scores, with significant findings leading to post-hoc Mann-Whitney U tests adjusted using Bonferroni corrections. For Cultural Interest scores, the Holm-Bonferroni method, a less conservative alternative, was used to account for multiple comparisons. To further explore the relationships between user engagement and cultural interest, ordinal logistic regression (OLR) analysis was conducted. The Brant test was employed to assess the proportional odds assumption [10], with results indicating no significant violations (𝑝 > 0.05), confirming the suitability of OLR for this dataset. Odds ratios, along with confidence intervals, were reported to quantify the effects of UES scores on various cultural interest outcomes. 3.7.2 Qualitative Data. Qualitative data from the interviews were recorded, transcribed, and analyzed thematically following the methodologies of Braun and Clarke [12] and Campbell et al. [14], resulting in the thematic map shown in Figure 7. To ensure reliability, Cohen’s kappa was calculated for inter-rater reliability between two coders who categorized 269 items into six categories: Immersion, Personalization, Engagement, Reflection, Culture, and Technological Intervention. The kappa score of 0.72, indicating substantial agreement (0.61 to 0.80) as per Viera and Garrett [100] and Eagan et al. [22], supports the consistency and validity of the coding process. Additionally, acknowledging the cultural-technical context of ‘Anansi the Spider VR,’ a positionality statement addressing the author’s race, ethnicity, and potential affinities is provided in Appendix B. This mixed-method approach offered a comprehensive understanding of the relationship between user engagement and the personalized, immersive environments explored in this study.

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Table 2: Summary of Participant Demographics from Semi-Structured Interviews. 4 ID Gender Age Cultural Background VR Experience Occupation VP1 VP2 VP3 VP4 VP5 VP6 VP7 VP8 VP9 VP10 VNP1 VNP2 VNP3 VNP4 VNP5 VNP6 NVP1 NVP2 NVNP1 NVNP2 Female Female Male Female Female Male Female Female Male Male Male Male Female Male Male Female Male Female Male Female 26 23 25 28 29 24 25 23 26 26 22 35 25 24 27 24 31 25 28 24 Indian Turkish Indian Peruvian Uzbek Chinese Romanian Indian, German Brazilian, German Ghanaian, German Russian Kenyan Turkish Vietnamese, American Indian Peruvian Indian Uzbek Indian Italian, German Moderately familiar Very familiar Slightly familiar Slightly familiar Moderately familiar Very familiar Slightly familiar Slightly familiar Slightly familiar Moderately familiar Very familiar Very familiar Moderately familiar Moderately familiar Very familiar Moderately familiar Moderately familiar Moderately familiar Very familiar Slightly familiar Accountant EdTech Researcher Network Planner Consultant Manager VR Engineer Software Engineer Sustainability Consultant Electrical Engineer Business Developer Entrepreneur Manager Kindergarten Teacher ML Engineer Aerospace Engineer Student Product Manager Student Manager Student Results The following sections detail the findings from both the quantitative analysis and the qualitative thematic analysis to comprehensively explore how Gen-AI and VR can enrich cultural storytelling. 4.1 User Engagement Scale (UES-LF) and Participant Interview Our quantitative analysis used the User Engagement Scale Long Form (UES-LF) [65] to evaluate four dimensions of engagement: Focused Attention (FA), Perceived Usability (PU), Aesthetic Appeal (AE), and Reward (RW). Figure 5 shows significant differences in UES scores across conditions (𝑋 2 = 298.59, 𝑝 < .001). Post-hoc analyses with MannWhitney U tests and Bonferroni corrections highlighted the benefits of personalization, particularly in non-VR settings. Both VR conditions (VP and VNP) outperformed non-VR conditions in FA, AE, and RW (𝑝 < .001). For example, in FA, VP (𝑀 = 3.98, 𝑆𝐷 = 0.96) and VNP (𝑀 = 4.05, 𝑆𝐷 = 1.04) scored significantly higher than NVNP (𝑀 = 2.46, 𝑆𝐷 = 1.26). Similarly, in AE, VP (𝑀 = 4.23, 𝑆𝐷 = 0.91) and VNP (𝑀 = 4.40, 𝑆𝐷 = 0.64) outperformed NVNP (𝑀 = 2.52, 𝑆𝐷 = 1.38). In the PU dimension, both VR conditions showed significant advantages. VP (𝑀 = 3.80, 𝑆𝐷 = 1.03) and VNP (𝑀 = 3.93, 𝑆𝐷 = 1.17) outperformed NVNP (𝑀 = 3.02, 𝑆𝐷 = 1.35; 𝑝 = .0013, 𝑈 = 3094; 𝑝 < .001, 𝑈 = 2838, respectively). Interestingly, NVP (𝑀 = 3.75, 𝑆𝐷 = 1.27) also scored higher than NVNP (𝑝 = .0041, 𝑈 = 3196), though it did not differ significantly from VNP (𝑝 = 1.00, 𝑈 = 4244). These results suggest that VR enhances usability and personalization may amplify these effects. In VR settings, qualitative feedback from semi-structured interviews revealed the impact of personalization, which was not 318 fully captured by quantitative results. Participants in VP and VNP conditions provided contrasting experiences. Participant VNP1 remarked, “I would like it more if I was the real character.” In contrast, participant VP1 stated, “Having the name makes you feel like you are really a part and more absorbed into the storytelling experience.” Personalized NPC narration also reduced confusion during gameplay. Participant VNP2 noted, “I wish the NPCs were more personalized to guide me when I was stuck.” Meanwhile, participant VP2 observed, “If the narrator didn’t say my name, I might have been confused about whether it was talking to me.” These insights demonstrate that personalization can be perceived within VR experiences and seamlessly integrated into the narrative. 4.2 Cultural Interest A Cultural Interest Survey was developed to assess differences in cultural engagement, changes in interest, and the applicability of the system to other traditional folklore across different countries. Figure 6 presents the Kruskal-Wallis test revealed significant differences in the scores of five individual cultural interest questions across the conditions (𝑋 2 = 52.34, 𝑝 < .001). Subsequent posthoc pairwise comparisons, conducted using the Mann-Whitney U test with Benjamini-Hochberg correction for multiple testing, are depicted in Figure 6. The results indicated consistent superior performance by VR conditions. For “Interest in Ghanaian Culture,” significant differences were observed between VP (𝑀 = 4.08, 𝑆𝐷 = 0.79) and NVNP (𝑀 = 2.67, 𝑆𝐷 = 1.37, 𝑝 = .013, 𝑈 = 29.5), and between VNP (𝑀 = 4.08, 𝑆𝐷 = 0.79) and NVNP (𝑝 = .013, 𝑈 = 29.5). No other significant differences were noted in this category. For “Relevance of Anansi Story,” VP (𝑀 = 4.58, 𝑆𝐷 = 0.51) demonstrated stronger significance compared to NVNP (𝑀 = 3.17,

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. Response Score Kruskal-Wallis Test: X² = 298.59, p < .001, ² = 0.21 **** **** **** **** **** **** **** **** ** **** ** * **** **** **** **** 5 4 3 2 Conditions NVNP NVP VNP VP 1 FA (Focused Attention) PU (Perceived Usability) AE (Aesthetic Appeal) Factor Groups RW (Reward) Figure 5: User Engagement Scores Across Different Experimental Conditions (NVNP, NVP, VNP, and VP). Each Factor Group represents a set of related items, with details available in Appendix Table 6. Responses were rated on a Likert scale from 1 (Strongly Disagree) to 5 (Strongly Agree). Significance levels are indicated by *, **, ***, and ****, corresponding to 𝑝 < .05, 𝑝 < .01, 𝑝 < .001, and 𝑝 < .0001, respectively. 𝑆𝐷 = 1.34, 𝑝 = .004, 𝑈 = 24.5) than VNP (𝑀 = 4.42, 𝑆𝐷 = 0.67, 𝑝 = .013, 𝑈 = 30.5). For “Interest in African Folklore,” both VP (𝑀 = 3.83, 𝑆𝐷 = 0.94) and VNP (𝑀 = 3.83, 𝑆𝐷 = 1.19) were significantly higher than NVNP (𝑀 = 2.42, 𝑆𝐷 = 1.08, 𝑝 = .005, 𝑈 = 24.5 for VP and 𝑝 = .008, 𝑈 = 26.5 for VNP). In the category of “Interest in Other Anansi Characters,” VP (𝑀 = 4.50, 𝑆𝐷 = 0.67) was significantly higher than NVNP (𝑀 = 3.00, 𝑆𝐷 = 1.48, 𝑝 = .008, 𝑈 = 27.5). For “Interest in Exploring Other Folklore,” significant differences were found for VP (𝑀 = 4.83, 𝑆𝐷 = 0.39) compared to NVNP (𝑀 = 2.92, 𝑆𝐷 = 1.51, 𝑝 = .002, 𝑈 = 22.0), and VNP (𝑀 = 4.50, 𝑆𝐷 = 0.80) compared to NVNP (𝑝 = .009, 𝑈 = 29.0). No significant differences were found between VP and VNP or between NVP and NVNP, indicating that personalization within VR or non-VR mediums did not significantly impact cultural interest metrics in this study. The qualitative insights further highlight these findings, illustrating how personalization makes cultural narratives more relatable for individuals from diverse backgrounds. Participant VP6 commented, “I think African stories are very far away from me. But this game made them so close to me. So I was more interested, even though I am not usually.” This statement highlights the potential of using personalized narratives in VR to bridge cultural gaps and enhance engagement. Similarly, Participant NVP2 reflected, “After the story, I thought to myself: Wow, there are many things about Africa I need to learn about. I will also tell my niece and nephew about this story,” underscoring the potential of personalized content to extend cultural exchange beyond the virtual experience. 319 4.3 The Relationship Between UES and Cultural Interest An Ordinal Logistic Regression (OLR) analysis was conducted to examine the relationship between total User Engagement Scale (UES) scores and various cultural interest outcomes. Results, presented in Table 3, show that higher UES scores significantly predicted greater cultural interest across all measures. For interest in Ghanaian culture, a one-unit increase in UES score increased the odds by approximately eight times (OR = 7.78, 𝑝 < 0.001). Similarly, higher UES scores were associated with a 5-fold increase in the likelihood of finding the Anansi story relevant (OR = 4.88, 𝑝 = 0.001). Interest in African folklore was strongly predicted by UES scores (OR = 10.15, 𝑝 < 0.001), as was interest in other Anansi characters (OR = 13.98, 𝑝 < 0.001). Finally, engagement scores also significantly predicted a nearly 20-fold increase in interest in exploring other folklore (OR = 19.96, 𝑝 < 0.001). 4.4 Qualitative Findings Based on thematic analysis, six key themes emerged: Immersion, Personalization, Engagement, Reflection, Cultural Relevance, and Technological Intervention. These themes delineate how VR and Gen-AI influenced participants’ experiences, providing insights into how immersion and personalization shaped their engagement and understanding of oral storytelling through the Anansi the Spider VR experience. Detailed descriptions of the key themes and subthemes identified are provided in Appendix Tables 10, 11, 12, 13, 14, and 15.

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Response Score Kruskal-Wallis Test: X² = 52.34, p < .001, ² = 0.21 5 4 3 2 1 * * * ** ** ** ** ** ** ** Conditions NVNP NVP VNP VP Interest in Ghanaian Culture Relevance of Anansi Story Interest in African Folklore Interest in Interest in Other Anansi Characters Exploring Other Folklore Cultural Interest Questions Figure 6: Interest Scores Across Different Experimental Conditions in African Folklore and Cultural Learning (NVNP, NVP, VNP, and VP). See Appendix Table 7 for a summary of the Cultural Interest Questionnaire items and the Likert scale used. Responses were rated on a Likert scale from 1 (Not Interested at All) to 5 (Very Interested). Significance levels are indicated by *, **, ***, and ****, corresponding to 𝑝 < .05, 𝑝 < .01, 𝑝 < .001, and 𝑝 < .0001, respectively. Table 3: Results of the Ordinal Logistic Regression analysis with Total UES Score as the predictor for different cultural interest outcomes. Model Summary Predictor: Total UES Score Interest in Ghanaian Culture Relevance of Anansi Story Interest in African Folklore Interest in Other Anansi Characters Interest in Exploring Other Folklore n = 48 Estimate Std. Error t-value p-value Odds Ratio (95% CI) 2.051 1.586 2.317 2.639 2.996 0.478 0.464 0.500 0.558 0.625 4.293 3.420 4.638 4.727 4.795 <0.001 0.001 <0.001 <0.001 <0.001 7.78 (2.46, 24.57) 4.88 (1.94, 12.27) 10.15 (3.63, 28.36) 13.98 (4.50, 43.42) 19.96 (5.36, 74.35) 4.4.1 Immersion. Participants highlighted how VR created a vivid and lifelike environment, enhancing their sense of presence. VNP5 remarked, “I just lost track of time and completely felt the surroundings. Like, I was completely immersed in the jungle.” Similarly, VP1 shared, “When I had to walk through and there were trees, I was bending my head because I felt like they were going to hit me.” The physical sensations of VR also reinforced this realism. VNP3 described, “When I was on top of the mountain, I actually felt a strange, tickling feeling on my feet.” Emotional responses, such as fear and anxiety, further anchored participants in the narrative. 4.4.2 Personalization. Personalization strengthened participants’ connection to the story by tailoring interactions to their identities. VP4 shared, “Hearing my name during the experience made it feel as though it was designed just for me.” VP5 noted, “It pronounced my name correctly, which made the story feel personal and gave me a sense of being the main character.” VP6’s experience underscored the familiarity dimension of personalization, stating, “It felt like I was chatting with my friends or someone who really knows me.” In contrast, non-personalized conditions resulted in detachment. VNP2 explained, “I wish the NPCs were more personalized. They would be much more relevant if they could guide me when I’m 320 stuck.” NVNP1 remarked, “I don’t find the story engaging, there are also many strange names of the characters that I have never come across before.” 4.4.3 Engagement. Participants identified interactivity, active agency, and aesthetic appeal as pivotal factors enhancing their engagement in the Anansi the Spider VR experience. “Games and stories are typically set in stone once told. However, VR experiences are valuable because they showcase how different choices can lead to varied outcomes, illustrating the real-life impact of our decisions.” (VP10) VP4 shared that the interactive nature of the experience made them feel more engaged: “The interaction made me feel more engaged, I was so curious I wished I could interact with everything, though it wasn’t possible... I wanted to explore every part of the world.” Aesthetic appeal also contributed to engagement. VP5 remarked, “The music and sounds were captivating, the sound of the waterfall felt incredibly real, enhancing my enjoyment.” Similarly, VP1 noted, The visuals were engaging and appealing. “From the beginning,

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. think before acting, and you have to make smart moves. The snake is very strong, but you can be stronger with your intelligence.” In the VNP condition, participants also reflected on similar themes but often described them in broader terms. VNP2 remarked, “As a university student in a foreign culture, this story reinforced the importance of relying on intelligence over brute strength, emphasizing its enduring moral relevance.” VNP3 added, “After a tough day at work with hierarchical challenges, tonight’s VR experience was enlightening. It showed me that strength does not matter if you have a good strategy.” Cultural and religious reflections also varied between conditions. In the VP condition, some participants tied the narrative to their family traditions or personal beliefs. For instance, VP4 noted, “Coming from a Catholic background, it reminded me of David and Goliath. My grandfather used to tell it to me like a fairy tale.” Participants in the VNP condition shared broader cultural connections. VNP6 explained, “My family used to tell me stories, and they come from different cities, so they would tell me stories and show me pictures.” These reflections illustrate how participants in different conditions connected the story’s themes to their own lives. Figure 7: The Role of VR and Gen-AI in Enhancing Understanding and Appreciation of African Culture: This thematic map illustrates how VR and Gen-AI influence immersion, personalization, and engagement, deepening reflection and fostering appreciation, while enriching understanding and interest in African culture. it felt like you were really there. The trees and mountains were somewhat intimidating, adding to the interactivity.” Participants also described the experience as requiring less effort to stay engaged compared to traditional storytelling formats. As VP4 explained, “Engaging with VR required less effort compared to traditional oral storytelling or watching a movie, and my interest grew much more easily.” The combination of interactivity and aesthetic appeal supported participants’ engagement, offering a dynamic alternative to traditional storytelling formats. 4.4.4 Reflection. Participants connected the story’s themes to their own lives, drawing parallels between the narrative and personal experiences. These reflections are categorized into four areas: personal experience, daily life situations, cultural and religious connections, and moral lessons and awareness, as shown in Figure 8. Participants in the VP condition frequently described a stronger sense of involvement due to personalized interactions. For example, VP7 shared, “The story reminds me of recent job-seeking efforts where I personalized applications only for positions I truly wanted and mass-sent the others.” Similarly, VP5 reflected, “You have to 321 Figure 8: Table of participant reflections categorized by themes: Personal Experience, Daily Life Situations, Cultural and Religious Connections, and Moral Lessons and Awareness. Each category is accompanied by relevant quotes from participants, illustrating the impact of VR and personalization on their perception and engagement with the narrative. 4.4.5 Cultural Relevance. Participants highlighted how the immersive storytelling experience deepened their understanding of African folklore and oral traditions, making them more relatable. VNP5 remarked, “I appreciate all cultures and find it fascinating to see how different cultures differ in so many activities but share common values of life. It was nice to learn the characters that were represented, like the sky god and the spider.” Similarly, VNP3 noted similarities with their own heritage, stating, “About African culture: I learned that animals represent some emotions and authority figures, and want to share a moral through animal figures. It is also like this in my culture, Turkish culture, because animals were there

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates before humans. They use animals to explain something bigger in the story.” The experience also prompted reflections on the broader significance of oral storytelling. VP8 shared, “Growing up, every time my family members would say a word which we don’t know, they always had a story behind it, so they have to tell us a story to show us what it actually means and why it’s still relevant today.” 4.4.6 Technological Intervention. Participants appreciated how VR and AI made cultural narratives engaging and accessible. VNP6 shared, “This was more interesting than previous VR experiences because I was learning about a new culture.” VP8 highlighted how these technologies appeal to younger, digitally savvy audiences, noting, “The generation right now and the generation which is coming would be more interested in AI and VR because they are already playing games, have mobile phones, etc. It is a way to make them understand their culture.” Participants also valued how these technologies foster understanding and appreciation for other traditions. VP5 emphasized their ability to simulate cultural experiences, making them more relatable. However, VP10 expressed concerns about losing the communal essence of oral storytelling, noting, “Wish the story had started with like a group setting, to then maybe seeing someone that could be my mother... Which story would you like to hear today?” These insights highlight how VR and AI together create immersive and meaningful storytelling experiences, while also revealing opportunities to preserve traditional communal aspects. 5 Discussion In this study, we explored the impact of Gen-AI personalization and VR immersion on oral storytelling through a mixed-method user study (𝑛 = 48), focusing on user engagement and cultural learning. We discuss the results in three directions, examining the individual impacts of VR and Gen-AI, their combined role in rekindling cultural heritage interest, and providing design recommendations to balance authenticity with innovation. 5.1 Impact of Personalization Addressing RQ1, our findings reveal that personalization significantly enhanced user engagement in non-VR conditions, as evidenced by higher PU and RW scores in personalized settings. Qualitative feedback suggests that the integration of personalization, such as name recognition, not only fostered a stronger sense of agency and emotional connection but also added a sense of narrative familiarity as if shared by a close friend. This aligns with prior research showing that correct name pronunciation can enhance users’ perceptions and engagement, demonstrating the role of personalization in deepening user-agent connections [37]. In contrast, non-personalized conditions were often associated with confusion and detachment. However, the subtler impact of personalization in VR conditions may reflect the overshadowing effect of VR’s sensory immersion, as highlighted in prior studies [23, 52]. This underscores the need to optimize personalization in immersive contexts, ensuring it complements rather than competes with VR’s immersive qualities. Additionally, while latency (2–7 seconds) for AI-driven responses was within acceptable ranges for real-time 322 interactions, participants did not perceive this as a hindrance to immersion or flow. Future iterations could focus on further reducing response times to enhance fluidity and maintain seamless storytelling experiences. These findings suggest that personalization can modernize traditional narratives, making them more relatable and engaging for younger audiences. 5.2 Role of VR Immersion Addressing RQ2, VR consistently outperformed non-VR conditions across all engagement dimensions (FA, AE, RW) and cultural interest measures, including interest in exploring other folklore and the relevance of the Anansi story. These findings highlight VR’s ability to evoke curiosity and appreciation for cultural narratives. Qualitative insights supported these results, with participants noting how VR’s sensory and interactive elements transformed traditional storytelling. They emphasized lifelike environments and decision-making opportunities as key factors in creating a dynamic connection to the story. These findings align with prior research on VR’s capacity to enhance emotional engagement and cultural learning [68, 90, 106]. For oral storytelling, VR provides a unique medium to bridge generational and cultural gaps by presenting traditional narratives in formats that resonate with digital-native audiences. The combination of immersive environments and interactive storytelling holds significant potential for modernizing oral traditions. 5.3 Gen-AI and VR for Oral Storytelling Understanding and Preservation Our findings demonstrate the potential of VR immersion and GenAI personalization in supporting the understanding and transmission of oral storytelling traditions. We answer RQ3 by (1) fostering connections to oral storytelling traditions and (2) revitalizing these narratives in the digital era to engage younger audiences. 5.3.1 Connection to Oral Storytelling. The integration of VR immersion and Gen-AI personalization bridges cultural gaps, fostering deeper connections and reflections on traditional oral storytelling. Traditional narratives often struggle to resonate with diverse audiences due to their detachment from individual cultural contexts [6]. Our findings demonstrate that AI-powered personalization, through features like name recognition and adaptive dialogue, helps make these stories more inclusive and relatable. Participants from diverse cultural backgrounds reported stronger connections to the narratives, effectively addressing this critique. Participants reflected on how these tales resonated with their personal experiences and daily lives, as illustrated in Figure 8. GenAI personalization complemented VR by immersing participants in the story’s moral and cultural dimensions, transforming them from passive listeners into active contributors. Notably, Anansi’s role in the experience shifted from the original tale’s storyteller to a facilitated narrative guide, with participants becoming the ones who collected these tales. This change allowed participants to envision themselves as the “Hero of the Day,” aligning with research on “storification,” which emphasizes how first-person engagement deepens emotional resonance, responsibility, and cultural understanding [4, 42].

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. While this narrative shift enhances user engagement and relatability, it raises important questions about preserving the authenticity of the original folklore. Balancing such interactivity with cultural integrity is critical and is further explored in the design recommendations section. 5.3.2 Revitalizing Oral Storytelling in the Digital Era. Our findings illustrate how VR and Gen-AI foster shared cultural understanding for younger audiences. While VR emphasizes individual immersion, participants expressed a strong desire to share the Anansi tale with family members, highlighting storytelling’s intergenerational role in cultural transmission. Gen-AI enhances this by tailoring narratives to participants’ identities, fostering cross-cultural appreciation through shared symbols like animal metaphors, as mentioned during interviews. For participants from immigrant backgrounds, such as VP10, a second-generation Ghanaian living in Germany, these technologies offered a powerful reconnection with cultural roots. VP10 reflected on their limited exposure to cultural narratives growing up and emphasized the importance of incorporating communal elements into immersive storytelling. They envisioned features like multiplayer modes and group settings to evoke the shared atmosphere of traditional storytelling: “It would be better if there were multiplayer scenes where we can be together, like a co-op experience.” (VP10) These insights highlight the potential of VR and Gen-AI to preserve and adapt oral storytelling for younger, digitally native audiences. Traditional oral storytelling methods often struggle to engage second-generation individuals disconnected from their heritage or audiences from other cultural backgrounds unfamiliar with these narratives. VR offers vivid, immediate access to historically and culturally rich stories, while personalized AI provides context, making these tales more relatable to diverse cultural audiences. Together, these technologies address the challenge of making traditional narratives resonate with modern users by bridging gaps in understanding and adapting cultural stories for younger audiences. VP10 also articulated the long-term potential of these technologies for integrating traditional storytelling into future familial practices: “Unfortunately, I did not have as much cultural transmission as I would have loved to have had, which I am trying to make up by reading books, etc., about my own culture. But probably, definitely, for me in the future, these stories are going to be a core part for my family.” (VP10) 5.4 Design Recommendations for Cultural Preservation and Transmission Combining our findings and prior literature, we propose two design recommendations for personalized VR interventions that balance cultural authenticity with fostering cultural reflection and transmission. 5.4.1 Maintain Story Authenticity in Interactive Narrative Structures. When designing immersive narratives based on oral traditions, it is crucial to balance innovation with preserving the core structure of 323 the story. Modern audiences expect interactivity in VR experiences, but these elements should not distort the cultural meaning of the original narrative. In the ‘Anansi the Spider VR’ experience, participants appreciated the balance between personalization and staying true to the original narrative. VP4 commented, “I like when stories are like, you have A and B, and regardless of A or B, the outcome is the same,” expressing a preference for choices that maintain the story’s core outcome. Similarly, VP9 emphasized the importance of preserving the original tale, noting that while it is normal to make changes, it is crucial not to alter the story too much to ensure it can still be passed on in its authentic form. The narrative model we used, referred to as a “yo-yo” structure [27], allowed participants to explore alternative branches while always returning to the main plot. This model preserves cohesion while enabling meaningful interaction. Future immersive storytelling approaches can adopt similar methods to maintain structural integrity while supporting user-driven interaction. 5.4.2 Align VR Experiences with Cultural Origins and Context. VR experiences should reflect the cultural context of oral traditions, which often emerge from communal settings like storytelling around a campfire. VP8 emphasized authenticity, suggesting that visuals should first depict how the story begins before transitioning into the narrative. Grounding users in these cultural and historical roots is crucial. Jones [33] research warns that VR’s visual capabilities may bias storytelling, potentially overshadowing core messages. To counter this, stories must be carefully introduced and framed within culturally significant settings. Following UNESCO ICH guidelines [94], creators can collaborate with cultural custodians such as elders, historians, and storytellers to ensure environments, symbols, and narrative choices remain faithful to the tradition’s narratives. Such collaboration helps maintain authenticity, acknowledges local expertise, and respects the intellectual property of source communities. While modern technology enables interactivity and personalization, designers must prioritize cultural integrity by preserving key moments, settings, and values that have long been transmitted through face-to-face interaction to ensure these immersive experiences stay true to their original context. 6 Limitations and Future Work The discussion highlights key findings that help drive future research in oral storytelling. While this study has certain limitations, they are outlined below to guide more comprehensive evaluations in future work. First, the quantitative results demonstrated limited sensitivity, with high UES scores in VR conditions suggesting a potential ceiling effect that may have masked subtle differences between personalized and non-personalized experiences. While semi-structured interviews provided richer insights, they lacked real-time data on participants’ emotional and cognitive processes during the experience. Future research could address this by employing think-aloud methods during VR sessions [80], enabling participants to articulate their thoughts in real-time. Although these methods may disrupt immersion [103], guided reflections at key narrative moments could uncover hidden dynamics of VR and AI-enhanced storytelling, offering deeper insights into user experiences.

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Second, the modest sample size (𝑛 = 48) and limited diversity of participants in terms of age range and educational backgrounds constrain the generalizability of the findings. While the statistical assumptions for OLR were met, the modest sample size and participant homogeneity require careful interpretation of the results, particularly when extrapolating to broader or more diverse populations. Furthermore, the limited sample size may reduce the power of the analysis to detect subtle effects. Although this study is exploratory and not intended for broad generalization, it offers valuable insights into how emerging technologies can enhance cultural engagement. Future studies could address these limitations by recruiting larger, more diverse participant pools, potentially through partnerships with public cultural institutions. Such efforts would improve inclusivity, strengthen the validity of the findings, and help determine the extent to which these approaches may be adapted for different cultural contexts. Third, the individual-focused nature of VR and Gen-AI in this study diverges from the communal essence of traditional oral storytelling. Participants, such as VP10 and VP8, highlighted the importance of incorporating group settings like storytelling around a campfire to better reflect the shared atmosphere of oral traditions. While this research did not directly involve cultural custodians or local communities in the design process, future work will prioritize collaboration with these stakeholders to ensure authenticity and respect for the cultural origins of these narratives. By aligning closely with traditional practices and exploring co-op storytelling experiences, future research can foster shared cultural connections across generations and diverse backgrounds, addressing both ethical considerations and the richness of communal storytelling traditions. Finally, as emphasized by Otero-Pailos [64], rethinking heritage conservation methods is critical. Collaborations with museums and cultural institutions could position VR and Gen-AI as participatory tools, transforming preservation from an expert-led to a community-driven process. This study, illustrated in Appendix Figure 9, demonstrates how digital storytelling fosters co-creation, allowing individuals to shape and reinterpret traditional tales, thereby promoting a collective cultural identity [53]. Future research could include longitudinal studies and participatory design approaches to assess whether these technologies can foster lasting cultural connections, influence behaviour, and facilitate intercultural dialogue when integrated into public cultural repositories [26, 38]. 7 Conclusion To investigate how VR immersion and Gen-AI-driven personalization can enhance storytelling, particularly in the context of preserving oral traditions, we developed ‘Anansi the Spider VR’, an innovative approach to revitalizing a Ghanaian oral tradition. Through a mixed-methods design with (𝑛 = 48) participants, we explored the impact of VR immersion and Gen-AI-driven personalization on engagement and cultural learning. Quantitative findings showed that both immersion and personalization significantly enhanced user engagement and interest in cultural learning. Qualitative findings revealed that VR and Gen-AI can transform user participation in storytelling, fostering cultural transmission, bridging cultural distances, and promoting self-reflection. Participants not only connected with African folklore but also reflected 324 on their own experiences through the lessons of the stories. These insights highlight the role of immersive technologies in preserving and revitalizing cultural narratives. This study also identified challenges in quantifying personal experiences within VR, underscoring the limitations of standardized tools like the UES. Future research should focus on developing more nuanced measurement methods and collaborating with cultural institutions to further explore and preserve immersive storytelling experiences. Overall, this research underscores the potential of VR and GenAI to maintain the relevance and impact of traditional tales in the digital age, ensuring that stories like Anansi continue to inspire and educate future generations. By bridging cultural gaps and fostering engagement through personalized and immersive technologies, this study provides a foundation for future initiatives aimed at preserving diverse oral traditions, promoting cultural understanding, and innovating heritage education on a global scale. 8 Open Science The source code and assets for Anansi the Spider VR are available at the following link: https://gitlab.lrz.de/hctl/anansispidervr. References [1] Hafiza Abas and Halimah Badioze Zaman. 2010. Digital Storytelling Design with Augmented Reality Technology for Remedial Students in Learning Bahasa Melayu. In Proceedings of Global Learn 2010, Zoraini Wati Abas, Insung Jung, and Joseph Luca (Eds.). Association for the Advancement of Computing in Education (AACE), Penang, Malaysia, 3558–3563. https://www.learntechlib.org/p/34436 [2] Yunus Abidin, Tita Mulyati, Yeni Yuniarti, and Trifalah Nurhuda. 2023. The Effects of Integrating Folklore and Mixed Reality toward Student’s Cultural Literacy. International Journal of Society, Culture & Language 11, 1 (2023), 307– 319. doi:10.22034/ijscl.2023.1995761.2963 [3] William C. Adams. 2015. Conducting Semi-Structured Interviews. John Wiley & Sons, Ltd, Chichester, West Sussex, United Kingdom, Chapter 19, 492–505. doi:10.1002/9781119171386.ch19 arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1002/9781119171386.ch19 [4] Ruth Aylett and Sandy Louchart. 2007. Being There: Participants and Spectators in Interactive Narrative. In Virtual Storytelling. Using Virtual Reality Technologies for Storytelling, Marc Cavazza and Stéphane Donikian (Eds.). Springer Berlin Heidelberg, Berlin, Heidelberg, 117–128. [5] Sojung Bahng, Ryan M. Kelly, and Jon McCormack. 2020. Reflexive VR Storytelling Design Beyond Immersion: Facilitating Self-Reflection on Death and Loneliness. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI ’20). Association for Computing Machinery, New York, NY, USA, 1–13. doi:10.1145/3313831.3376582 [6] Yee Bee Choo, Tina Abdullah, and Abdullah Mohd Nawi. 2020. Digital Storytelling vs. Oral Storytelling: An Analysis of the Art of Telling Stories Now and Then. Universal Journal of Educational Research 8 (05 2020), 46–50. doi:10.13189/ujer.2020.081907 [7] Thomas Bjørner, Andreas Jin Sum, Rune Korsgaard Ludvigsen, Nicolai Lind Bouquin, Frederik Darling Larsen, and Ulrik Kampel. 2022. Making homework fun: The effect of game-based learning on reading engagement. In Proceedings of the 2022 ACM Conference on Information Technology for Social Good (Limassol, Cyprus) (GoodIT ’22). Association for Computing Machinery, New York, NY, USA, 353–359. doi:10.1145/3524458.3547263 [8] Rick Borovoy, Brian Silverman, Tim Gorton, Matt Notowidigdo, Brian Knep, Mitchel Resnick, and Jeff Klann. 2001. Folk computing: revisiting oral tradition as a scaffold for co-present communities. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (Seattle, Washington, USA) (CHI ’01). Association for Computing Machinery, New York, NY, USA, 466–473. doi:10. 1145/365024.365316 [9] Efe Bozkir, Süleyman Özdel, Ka Hei Carrie Lau, Mengdi Wang, Hong Gao, and Enkelejda Kasneci. 2024. Embedding Large Language Models into Extended Reality: Opportunities and Challenges for Inclusion, Engagement, and Privacy. In Proceedings of the 6th ACM Conference on Conversational User Interfaces (Luxembourg, Luxembourg) (CUI ’24). Association for Computing Machinery, New York, NY, USA, Article 38, 7 pages. doi:10.1145/3640794.3665563 [10] Rollin Brant. 1990. Assessing Proportionality in the Proportional Odds Model for Ordinal Logistic Regression. Biometrics 46, 4 (1990), 1171–1178. http:

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. //www.jstor.org/stable/2532457 [11] Tharrenos Bratitsis. 2022. Cultural heritage redesigned through digital storytelling. IGI Global, Hershey, PA, USA, 296–311. [12] Virginia Braun and Victoria Clarke. 2006. Using thematic analysis in psychology. Qualitative Research in Psychology 3 (01 2006), 77–101. doi:10.1191/ 1478088706qp063oa [13] Deidre Brown and George Nicholas. 2012. Protecting indigenous cultural property in the age of digital democracy: Institutional and communal responses to Canadian First Nations and Māori heritage concerns. Journal of Material Culture 17, 3 (2012), 307–324. doi:10.1177/1359183512454065 [14] Karen A. Campbell, Elisabeth Orr, Pamela Durepos, Linda Nguyen, Lin Li, Carly Whitmore, Paige Gehrke, Leslie Pendleton Graham, and Susan M. Jack. 2021. Reflexive Thematic Analysis for Applied Qualitative Health Research. The Qualitative Report 26, 6 (2021), 2011–2028. doi:10.46743/2160-3715/2021.5010 [15] Vanessa Cesário, Sandra Olim, and Valentina Nisi. 2020. A Natural History Museum Experience: Memories of Carvalhal’s Palace – Turning Point. In Interactive Storytelling: 13th International Conference on Interactive Digital Storytelling, ICIDS 2020, Bournemouth, UK, November 3–6, 2020, Proceedings (Bournemouth, United Kingdom). Springer-Verlag, Berlin, Heidelberg, 339–343. doi:10.1007/978-3-030-62516-0_31 [16] cgtrader. 2018. Low Poly Spider Model. https://www.cgtrader.com/free-3dmodels/animals/insect/low-poly-spider-model [17] Kennedy Chinyowa. 2001. The Sarungano and Shona Storytelling: an African Theatrical Paradigm. Studies in Theatre and Performance 21 (04 2001), 18–30. doi:10.1386/stap.21.1.18 [18] Peter Cohan. 2024. What Is Generative AI? Apress, Berkeley, CA, 9–28. doi:10. 1007/979-8-8688-0318-5_2 [19] Larry Cutler, Eric Darnell, Nathaniel Dirksen, Amy Tucker, Scot Stafford, Erick Oh, Anika Nagpal, Eusong Lee, and Nick Ladd. 2022. From quest to quill: pushing the boundaries of VR storytelling in baobab’s Baba Yaga and Namoo. In ACM SIGGRAPH 2021 Production Sessions (SIGGRAPH ’21). Association for Computing Machinery, New York, NY, USA, Article 6, 1 pages. doi:10.1145/3446368.3452126 [20] M. E. Kropp Dakubu. 1990. Why spider is king of stories: the message in the medium of a West African tale. African Languages and Cultures 3, 1 (1990), 33–56. doi:10.1080/09544169008717709 [21] Tales Rebequi Costa Borges de Souza and João Luiz Bernardes. 2016. The Influences of Culture on User Experience. In Cross-Cultural Design, Pei-Luen Patrick Rau (Ed.). Springer International Publishing, Cham, 43–52. [22] Brendan Eagan, Jais Brohinsky, Jingyi Wang, and David Williamson Shaffer. 2020. Testing the reliability of inter-rater reliability. In Proceedings of the Tenth International Conference on Learning Analytics & Knowledge (Frankfurt, Germany) (LAK ’20). Association for Computing Machinery, New York, NY, USA, 454–461. doi:10.1145/3375462.3375508 [23] Aviv Elor, Michael Powell, Evanjelin Mahmoodi, Mircea Teodorescu, and Sri Kurniawan. 2022. Gaming Beyond the Novelty Effect of Immersive Virtual Reality for Physical Rehabilitation. IEEE Transactions on Games 14, 1 (2022), 107–115. doi:10.1109/TG.2021.3069445 [24] Qi Chen Gao and Ali Emami. 2023. The Turing Quest: Can Transformers Make Good NPCs?. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop), Vishakh Padmakumar, Gisela Vallejo, and Yao Fu (Eds.). Association for Computational Linguistics, Toronto, Canada, 93–103. doi:10.18653/v1/2023.acl-srw.17 [25] Xiao Ge, Chunchen Xu, Daigo Misaki, Hazel Rose Markus, and Jeanne L Tsai. 2024. How Culture Shapes What People Want From AI. In Proceedings of the CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI ’24). Association for Computing Machinery, New York, NY, USA, Article 95, 15 pages. doi:10.1145/3613904.3642660 [26] Danilo Giglitto, Luigina Ciolfi, Caroline Claisse, and Eleanor Lockley. 2019. Bridging cultural heritage and communities through digital technologies: Understanding perspectives and challenges. In Proceedings of the 9th International Conference on Communities & Technologies - Transforming Communities. Association for Computing Machinery, New York, NY, USA, 81–91. doi:10.1145/3328320.3328386 [27] Stacey Hand and Duane Varan. 2009. Interactive stories and the audience: Why empathy is important. Comput. Entertain. 7, 3, Article 39 (sep 2009), 14 pages. doi:10.1145/1594943.1594951 [28] Monique Hennink and Bonnie N. Kaiser. 2022. Sample sizes for saturation in qualitative research: A systematic review of empirical tests. Social Science & Medicine 292 (2022), 114523. doi:10.1016/j.socscimed.2021.114523 [29] Alexander Hick and Martina Ziefle. 2022. A Qualitative Approach to the Public Perception of AI. International Journal on Cybernetics & Informatics 11 (08 2022), 1–17. doi:10.5121/ijci.2022.110401 [30] Teresa Hirzle, Florian Müller, Fiona Draxler, Martin Schmitz, Pascal Knierim, and Kasper Hornbæk. 2023. When XR and AI Meet - A Scoping Review on Extended Reality and Artificial Intelligence. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI ’23). Association for Computing Machinery, New York, NY, USA, Article 730, 45 pages. doi:10.1145/3544548.3581072 325 [31] Valentin Holzwarth, Joy Gisler, Christian Hirt, and Andreas Kunz. 2021. Comparing the Accuracy and Precision of SteamVR Tracking 2.0 and Oculus Quest 2 in a Room Scale Setup. In Proceedings of the 2021 5th International Conference on Virtual and Augmented Reality Simulations (ICVARS ’21) (Melbourne, VIC, Australia). Association for Computing Machinery, New York, NY, USA, 42–46. doi:10.1145/3463914.3463921 [32] Emese Ilyefalvi. 2018. The Theoretical, Methodological and Technical Issues of Digital Folklore Databases and Computational Folkloristics. Acta Ethnographica Hungarica 63 (09 2018), 209–258. doi:10.1556/022.2018.63.1.11 [33] Sarah Jones. 2017. Disrupting the narrative: immersive journalism in virtual reality. Journal of Media Practice 18, 2-3 (2017), 171–185. doi:10.1080/14682753. 2017.1374677 arXiv:https://doi.org/10.1080/14682753.2017.1374677 [34] Anna Kasunic and Geoff Kaufman. 2018. Learning to Listen: Critically Considering the Role of AI in Human Storytelling and Character Creation. In Proceedings of the First Workshop on Storytelling, Margaret Mitchell, Ting-Hao Kenneth’ Huang, Francis Ferraro, and Ishan Misra (Eds.). Association for Computational Linguistics, New Orleans, Louisiana, 1–13. doi:10.18653/v1/W18-1501 [35] Lindsey Kelley and Danette DiMarco. 2007. Magic, Myths, and Storytelling in Colonial Caribbean Literature. Final Project Paper, English 598 – Caribbean Texts & Contexts, Indiana University of Pennsylvania. https://www.emscharts. com/pub/images/Carib_Final%20Project%20Paper.pdf [36] Sarah Kenderdine, Leith K. Y. Chan, and Jeffrey Shaw. 2014. Pure Land: Futures for Embodied Museography. J. Comput. Cult. Herit. 7, 2, Article 8 (jun 2014), 15 pages. doi:10.1145/2614567 [37] James Kennedy, Naveen Kumar, and Maike Paetzel-Prüsmann. 2024. Name Pronunciation Extraction and Reuse in Human-Robot Conversations. In Companion of the 2024 ACM/IEEE International Conference on Human-Robot Interaction (Boulder, CO, USA) (HRI ’24). Association for Computing Machinery, New York, NY, USA, 593–597. doi:10.1145/3610978.3640689 [38] Markos Konstantakis, John Aliprantis, Alexandros Teneketzis, and George Caridakis. 2018. Understanding user experience aspects in cultural heritage interaction. In Proceedings of the 22nd Pan-Hellenic Conference on Informatics (Athens, Greece) (PCI ’18). Association for Computing Machinery, New York, NY, USA, 267–271. doi:10.1145/3291533.3291580 [39] Lindah Kotut, Neelma Bhatti, Taha Hassan, Derek Haqq, and Morva Saaty. 2024. Griot-Style Methodology: Longitudinal Study of Navigating Design With Unwritten Stories. In Proceedings of the CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI ’24). Association for Computing Machinery, New York, NY, USA, Article 615, 14 pages. doi:10.1145/3613904. 3642682 [40] Ka Hei Carrie Lau, Efe Bozkir, Hong Gao, and Enkelejda Kasneci. 2025. Evaluating Usability and Engagement of Large Language Models in Virtual Reality for Traditional Scottish Curling. In Computer Vision – ECCV 2024 Workshops, Alessio Del Bue, Cristian Canton, Jordi Pont-Tuset, and Tatiana Tommasi (Eds.). Springer Nature Switzerland, Cham, 177–195. [41] Boyang Li, Stephen Lee-Urban, George Johnston, and Mark O. Riedl. 2013. Story generation with crowdsourced plot graphs. In Proceedings of the Twenty-Seventh AAAI Conference on Artificial Intelligence (AAAI’13). AAAI Press, Bellevue, Washington, 598–604. [42] Zixiao Liu, Shuo Yan, Yu Lu, and Yuetong Zhao. 2022. Generating Embodied Storytelling and Interactive Experience of China Intangible Cultural Heritage “Hua’er” in Virtual Reality. In Extended Abstracts of the 2022 CHI Conference on Human Factors in Computing Systems (New Orleans, LA, USA) (CHI EA ’22). Association for Computing Machinery, New York, NY, USA, Article 439, 7 pages. doi:10.1145/3491101.3519761 [43] Robyn Longhurst. 2003. Semi-structured Interviews and Focus Groups. Sage Publications Ltd, London, United Kingdom, 117–132. [44] Irene Lourdi, Christos Papatheodorou, and Mara Nikolaidou. 2007. A Multi-layer Metadata Schema for Digital Folklore Collections. J. Information Science 33 (02 2007), 197–213. doi:10.1177/0165551506070711 [45] Tasnim Lubis, Achdial Farhan Abus, Nanda Saputra, and Nurul Adilla Alatas Abus. 2023. Educate students through their folklore: Environmental education. AIP Conference Proceedings 2805, 1 (08 2023), 060002. doi:10.1063/5.0148076 arXiv:https://pubs.aip.org/aip/acp/articlepdf/doi/10.1063/5.0148076/18087990/060002_1_5.0148076.pdf [46] Camila Loiola Brito Maia and Elizabeth Sucupira Furtado. 2019. An Approach to Analyze User’s Emotion in HCI Experiments Using Psychophysiological Measures. IEEE Access 7 (2019), 36471–36480. doi:10.1109/ACCESS.2019.2904977 [47] Emily Zobel Marshall. 2007. Liminal Anansi: Symbol of Order and Chaos An Exploration of Anansi’s Roots Amongst the Asante of Ghana. Caribbean Quarterly 53, 3 (2007), 30–40. https://www.jstor.org/stable/40654609 [48] EJZ Zobel Marshall. 2019. This is Not a Fairytale: Anansi and the Web of Narrative Power. In The Fairy Tale World, A Teverson (Ed.). Routledge, London, United Kingdom. https://eprints.leedsbeckett.ac.uk/id/eprint/4451/ [49] Maria J. Mayan. 2023. Essentials of Qualitative Inquiry (2 ed.). Routledge, London, United Kingdom. doi:10.4324/b23331 [50] Olga Konstantinovna Melnikova, Marat Aidarovich Yahin, and Khanif Fakhretdinovich Makayev. 2020. Internet communication as a medium for digital folklore.

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates In Proceedings of the III International Scientific and Practical Conference (Saint Petersburg, Russia) (DEFIN ’20). Association for Computing Machinery, New York, NY, USA, Article 4, 4 pages. doi:10.1145/3388984.3389061 [51] Wolfgang Mieder. 1987. Tradition and Innovation in Folk Literature (1 ed.). Routledge, London, United Kingdom. doi:10.4324/9781315673677 [52] Ines Miguel-Alonso, David Checa, Henar Guillen-Sanz, and Andres Bustillo. 2024. Evaluation of the novelty effect in immersive Virtual Reality learning experiences. Virtual Reality 28, 1 (2024), 27. doi:10.1007/s10055-023-00926-5 [53] Cristina Milano, Elisabetta Falchetti, Pascuala Migone, and Valentina Nisi. 2023. Digital storytelling, cultural heritage, and social inclusion: the MEMEX project. In Digital Approaches to Inclusion and Participation in Cultural Heritage: Insights from Research and Practice in Europe, Danilo Giglitto, Luigina Ciolfi, Eleanor Lockley, and Eirini Kaldeli (Eds.). Routledge, London, United Kingdom, 8–26. doi:10.4324/9781003277606-2 [54] Janet Horowitz Murray. 1997. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. The Free Press, USA. [55] Valentina Nisi, Paulo Bala, Vanessa Cesário, Stuart James, Alessio Del Bue, and Nuno Jardim Nunes. 2023. "Connected to the people": Social Inclusion & Cohesion in Action through a Cultural Heritage Digital Tool. Proc. ACM Hum.-Comput. Interact. 7, CSCW2, Article 319 (Oct. 2023), 37 pages. doi:10.1145/ 3610168 [56] Dorothy Noyes. 2012. The Social Base of Folklore. John Wiley & Sons, Ltd, Chichester, West Sussex, United Kingdom, Chapter 1, 13–39. doi:10.1002/ 9781118379936.ch1 [57] Heather O’Brien and Elaine Toms. 2010. The Development and Evaluation of a Survey to Measure User Engagement. JASIST 61 (01 2010), 50–69. doi:10.1002/ asi.21229 [58] Tokio Ogawa, Kei Kobayashi, and Junichi Hoshino. 2021. The First-Person VR System Augmenting Folklore Experiences. In 2021 Nicograph International (NicoInt). IEEE, Piscataway, NJ, USA, 82–85. doi:10.1109/NICOINT52941.2021. 00022 [59] OpenAI. 2022. Introducing Whisper. https://openai.com/index/whisper/ [60] OpenAI. 2023. Free Text to Speech & AI Voice Generator | ElevenLabs. https: //elevenlabs.io/ [61] OpenAI. 2023. Introducing GPT-4. https://openai.com/index/gpt-4/ [62] David Oppenheim and Randall Lloyd Okita. 2020. The Book of Distance: Personal Storytelling in VR. In ACM SIGGRAPH 2020 Immersive Pavilion (Virtual Event, USA) (SIGGRAPH ’20). Association for Computing Machinery, New York, NY, USA, Article 5, 2 pages. doi:10.1145/3388536.3407896 [63] Araba A Z Osei-Tutu. 2023. Developing African oral traditional storytelling as a framework for studying with African peoples. Qualitative Research 23, 6 (2023), 1497–1514. doi:10.1177/14687941221082263 [64] Jorge Otero-Pailos. 2016. Experimental Preservation. Places Journal 28, 3 (September 2016), 1 pages. doi:10.22269/160913 [65] Heather L. O’Brien, Paul Cairns, and Mark Hall. 2018. A practical approach to measuring user engagement with the refined user engagement scale (UES) and new UES short form. International Journal of Human-Computer Studies 112 (2018), 28–39. doi:10.1016/j.ijhcs.2018.01.004 [66] Pat Pataranutaporn, Phoomparin Mano, Piyaporn Bhongse-Tong, Tas Chongchadklang, Chayapatr Archiwaranguprok, Lamtharn Hantrakul, Jirach Eaimsa-ard, Pattie Maes, and Pichet Klunchun. 2024. Human-AI Co-Dancing: Evolving Cultural Heritage through Collaborative Choreography with Generative Virtual Characters. In Proceedings of the 9th International Conference on Movement and Computing (Utrecht, Netherlands) (MOCO ’24). Association for Computing Machinery, New York, NY, USA, Article 14, 10 pages. doi:10.1145/3658852.3661317 [67] Robert D Pelton. 1989. The trickster in West Africa: A study of mythic irony and sacred delight. Number 8 in Hermeneutics: Studies in the History of Religions. Univ of California Press, Berkeley, CA, USA. [68] Jayesh S. Pillai and Manvi Verma. 2019. Grammar of VR Storytelling: Narrative Immersion and Experiential Fidelity in VR Cinema. In Proceedings of the 17th ACM SIGGRAPH International Conference on Virtual-Reality Continuum and Its Applications in Industry (Brisbane, QLD, Australia) (VRCAI ’19). Association for Computing Machinery, New York, NY, USA, Article 34, 6 pages. doi:10.1145/ 3359997.3365680 [69] Della Pollock. 2004. Oral Traditions in Performance. Oral Tradition 18 (01 2004), 263–265. doi:10.1353/ort.2004.0081 [70] Andru Baskara Putra, Alief Kukuh Nurkusuma, Gregorious Juan Khawarga, Meiliana, and Muhamad Fajar. 2023. Development of Timun Mas Game Platformer for Increasing Generation Z Interest to Indonesian Folklore. In 2023 International Conference on Artificial Intelligence in Information and Communication (ICAIIC). IEEE, Piscataway, NJ, USA, 046–052. doi:10.1109/ICAIIC57133.2023.10067133 [71] Python. 2024. Python 3.12.5 documentation. https://docs.python.org/release/3. 12.5/ [72] Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. 2018. Improving Language Understanding by Generative Pre-Training. https://cdn.openai.com/research-covers/language-unsupervised/language_ understanding_paper.pdf 326 [73] R.S. Rattray. 2023. Akan-Ashanti Folktales: A Collection of 75 Ashanti Folktales. Prince Sarfo-Adu, London, United Kingdom. https://books.google.de/books? id=AHrSEAAAQBAJ [74] Mark O. Riedl. 2016. Computational Narrative Intelligence: A Human-Centered Goal for Artificial Intelligence. arXiv:1602.06484 [cs.AI] https://arxiv.org/abs/ 1602.06484 [75] Selma Rizvic, Nermin Djapo, Fatmir Alispahic, Bojan Hadzihalilovic, Fahira Fejzic Cengic, Ahmed Imamovic, Vensada Okanovic, and Dusanka Boskovic. 2017. Guidelines for interactive digital storytelling presentations of cultural heritage . In 2017 9th International Conference on Virtual Worlds and Games for Serious Applications (VS-Games). IEEE Computer Society, Los Alamitos, CA, USA, 253–259. doi:10.1109/VS-GAMES.2017.8056610 [76] Rebecca Rouse. 2019. Someone Else’s Story: An Ethical Approach to Interactive Narrative Design for Cultural Heritage. In Interactive Storytelling, Rogelio E. Cardona-Rivera, Anne Sullivan, and R. Michael Young (Eds.). Springer International Publishing, Cham, 47–60. [77] Pranab Sahoo, Ayush Kumar Singh, Sriparna Saha, Vinija Jain, Samrat Mondal, and Aman Chadha. 2024. A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications. arXiv:2402.07927 [cs.AI] https://arxiv.org/abs/2402.07927 [78] R. C. Schank and C. K. Riesbeck. 1981. Inside Computer Understanding: Five Programs Plus Miniatures. Lawrence Erlbaum Associates, Hillsdale, New Jersey. [79] United Nations Educational, Scientific and Cultural Organization. 2003. Text of the Convention for the Safeguarding of the Intangible Cultural Heritage. https://ich.unesco.org/en/convention [80] Joyce Siette, Christopher Campbell, Patrick Adam, and Celia Harris. 2024. Exploring the usability of the virtual reality module LEAF CAFÉ: a qualitative think-aloud study. BMC Geriatrics 24 (02 2024). doi:10.1186/s12877-024-04767-y [81] Sketchfab. 2020. Ball python. https://sketchfab.com/3d-models/ball-python74166573c34c40bc977c7a8db5b92ff4 [82] Sketchfab. 2024. Ashanti Moon Mask. https://sketchfab.com/3d-models/ashantimoon-mask-a9ea6430a5c44be998b94040712da159 [83] srcnalt. 2024. An unofficial OpenAI Unity Package. https://github.com/srcnalt/ OpenAI-Unity [84] Ingibergur Sindri Stefnisson and David Thue. 2018. Mimisbrunnur: AI-assisted authoring for interactive storytelling. In Proceedings of the Fourteenth AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE’18). AAAI Press, Edmonton, Alberta, Canada, Article 34, 7 pages. [85] Unity Asset Store. 2024. Tropical Forest Pack. https://assetstore.unity.com/ packages/3d/environments/tropical-forest-pack-49391 [86] Kim Hua Tan, Shirley Ngoi, Nasruddin Yunos, Jamsari Alias, and Norazila Mat. 2024. Tales in Tech: Understanding Educational Impact and Challenges of Digital Storytelling. Conhecimento & Diversidade 16 (07 2024), 617–642. doi:10.18316/rcd.v16i42.11779 [87] Karim Tharani. 2019. When tradition meets technology: Curating digital collections to enhance learning of traditional knowledge. In Proceedings of EdMedia 2019, Johan Van Braak, Mark Brown, Lorenzo Cantoni, Manuel Castro, Rhonda Christensen, Gayle V. Davidson-Shivers, Koen DePryck, Martin Ebner, Mikhail Fominykh, Catherine Fulford, Stylianos Hatzipanagos, Gerald Knezek, Karel Kreijns, Gary Marks, Erkko Sointu, Elsebeth Korsgaard Sorensen, Jarmo Viteli, Joke Voogt, Peter Weber, Edgar Weippl, Olaf ZawackiRichter, and Theo Bastiaens (Eds.). Association for the Advancement of Computing in Education (AACE), Amsterdam, Netherlands, 1641–1644. https: //www.learntechlib.org/p/210186 [88] THeK3nger. 2024. Unity script for using ElevenLabs TTS service. https: //gist.github.com/THeK3nger/882a31f52bb002dac155ad95529c3680 [89] B. Toelken. 1996. Dynamics Of Folklore. Utah State University Press, Logan, UT, USA. https://books.google.de/books?id=-fKAAAAAMAAJ [90] Lingwei Tong, Robert W. Lindeman, Heide Lukosch, Rory Clifford, and Holger Regenbrecht. 2024. Applying Cinematic Virtual Reality with Adaptability to Indigenous Storytelling. J. Comput. Cult. Herit. 17, 2, Article 28 (March 2024), 25 pages. doi:10.1145/3647996 [91] Georgios Trichopoulos, Georgios Alexandridis, and George Caridakis. 2023. A Survey on Computational and Emergent Digital Storytelling. Heritage 6, 2 (2023), 1227–1263. doi:10.3390/heritage6020068 [92] Kudakwashe Tuwe. 2016. The African oral tradition paradigm of storytelling as a methodological framework: Employment experiences for African communities in New Zealand. In Proceedings of the 38th AFSAAP Conference: 21st Century Tensions and Transformation in Africa. African Studies Association of Australasia and the Pacific (AFSAAP), Australia, 1 pages. https://afsaap.org.au/assets/ Kudakwashe_Tuwe_AFSAAP2015.pdf [93] Marina Umaschi. 1996. SAGE storytellers: learning about identity, language and technology. In Proceedings of the 1996 International Conference on Learning Sciences (ICLS ’96). International Society of the Learning Sciences, Evanston, Illinois, 526–531. [94] UNESCO. 2024. Oral traditions and expressions including language as a vehicle of the intangible cultural heritage. https://ich.unesco.org/en/oral-traditions-

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. and-expressions-00053 Retrieved July 10, 2024. [95] UNESCO. 2024. What is Intangible Cultural Heritage? https://ich.unesco.org/ en/what-is-intangible-heritage-00003 Retrieved July 10, 2024. [96] Putri Utami and Hendi Sama. 2022. An Exploratory Study of Digital Traditional Folklore and Its Acceptance Among Generation Z in Indonesia. Jurnal INFOKUM 10, 5 (December 2022), 460–467. https://infor.seaninstitute.org/index.php/ infokum/article/view/989 ISSN: 2302-9706. [97] Michael Vallance and Phillip Towndrow. 2022. Perspective: Narrative Storyliving in Virtual Reality Design. Frontiers in Virtual Reality 3 (03 2022), 779148. doi:10. 3389/frvir.2022.779148 [98] Lieke Van Duin. 2007. Anansi as a classical hero. Journal of Caribbean Literatures 5, 1 (2007), 33–42. https://www.jstor.org/stable/40986316 [99] Jan Vansina. 1985. Oral Tradition as History. James Currey, Oxford, United Kingdom. [100] Anthony Viera and Joanne Garrett. 2005. Understanding Interobserver Agreement: The Kappa Statistic. Family medicine 37 (06 2005), 360–3. [101] Jules White, Quchen Fu, Sam Hays, Michael Sandborn, Carlos Olea, Henry Gilbert, Ashraf Elnashar, Jesse Spencer-Smith, and Douglas C. Schmidt. 2023. A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT. arXiv:2302.11382 [cs.SE] https://arxiv.org/abs/2302.11382 [102] Ashlee Cunsolo Willox, Sherilee L Harper, Victoria L Edge, ‘My Word’: Storytelling, Digital Media Lab, and Rigolet Inuit Community Government. 2013. Storytelling in a digital age: digital storytelling as an emerging narrative method for preserving and promoting indigenous oral wisdom. Qualitative Research 13, 2 (2013), 127–147. doi:10.1177/1468794112446105 [103] William Winn, Mark Windschitl, Ruth Fruland, and Yenling Lee. 2002. When does immersion in a virtual environment help students construct understanding. In Proceedings of the International Conference of the Learning Sciences, ICLS, Vol. 206. International Society of the Learning Sciences (ISLS), Online / no fixed address, 497–503. [104] Zhiqing Wu, Duotun Wang, Shumeng Zhang, Yuru Huang, Zeyu Wang, and Mingming Fan. 2024. Toward Making Virtual Reality (VR) More Inclusive for Older Adults: Investigating Aging Effect on Target Selection and Manipulation Tasks in VR. In Proceedings of the CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI ’24). Association for Computing Machinery, New York, NY, USA, Article 24, 17 pages. doi:10.1145/3613904.3642558 [105] Lyumanshan Ye, Jiandong Jiang, Danni Chang, and Pengfei Liu. 2024. Storypark: Leveraging Large Language Models to Enhance Children Story Learning Through Child-AI collaboration Storytelling. arXiv:2405.06495 [cs.HC] https://arxiv.org/abs/2405.06495 [106] Zhiyuan Yu, Cheng-Hung Lo, Mutian Niu, and Hai-Ning Liang. 2023. Comparing Cinematic Conventions through Emotional Responses in Cinematic VR and Traditional Mediums. In SIGGRAPH Asia 2023 Technical Communications (Sydney, NSW, Australia) (SA ’23). Association for Computing Machinery, New York, NY, USA, Article 10, 4 pages. doi:10.1145/3610543.3626175 [107] Zaihasriah Zahidi, Yan Peng Lim, and Peter Charles Woods. 2013. User Experience for Digitization and Preservation of Cultural Heritage . In 2013 International Conference on Informatics and Creative Multimedia (ICICM). IEEE Computer Society, Los Alamitos, CA, USA, 13–16. doi:10.1109/ICICM.2013.11 [108] Runcong Zhao, Wenjia Zhang, Jiazheng Li, Lixing Zhu, Yanran Li, Yulan He, and Lin Gui. 2023. NarrativePlay: Interactive Narrative Understanding. arXiv:2310.01459 [cs.CL] https://arxiv.org/abs/2310.01459 A Participant Demographics B Researchers Positionality Statement As the main researcher in this study, I acknowledge myself as an educated Asian researcher, my approach to studying and revitalizing folklore is informed by a background that blends academic diligence with a passion for cultural heritage. My work is driven by a commitment to understanding and preserving diverse narratives through technology. This position allows me to reinterpret folklore with a fresh perspective, aiming to make these traditional stories relevant and reachable to contemporary audiences. My values emphasize respect for cultural authenticity and the power of storytelling in bridging cultural divides. C Pre-Assessment Questionnaire 327 Table 4: Participant Demographics and Pre-Assessment Scores. Likert scale: 1 (Not at all Familiar) to 5 (Very Familiar). 𝑛 = 12 per condition. Metric NVNP NVP VNP VP M SD M SD M SD M SD Age 24.5 3.2 25.8 4.1 26.3 3.7 27.2 4.0 Gender Distribution (%) Female 52 Male 48 Storytelling Fam. VR Familiarity Oral Traditions Fam. African Folklore Fam. Interest in Cultures Comfort with Personalization 3.3 3.1 2.4 1.3 3.6 3.8 - 47 53 - 53 47 - 48 52 - 1.0 1.2 1.0 0.6 1.2 1.0 3.5 2.6 2.2 1.2 4.2 3.8 1.0 0.7 1.0 0.4 0.6 0.9 3.7 3.0 2.9 1.8 4.0 3.3 1.3 0.9 1.3 0.9 0.6 1.1 3.8 2.6 3.0 1.6 3.8 3.8 0.8 0.8 1.3 0.9 0.9 0.8 Table 5: Pre-Assessment Questionnaire: Items and Response Options. Question Response Options Cultural background (ethnicity, upbringing, media exposure) Open-ended Education level High School, Bachelor’s, Master’s, Doctoral Digital storytelling engagement Never, Rarely, Sometimes, Of(audiobooks, podcasts, interac- ten, Very Often tive stories) Familiarity with virtual reality Not at all, Slightly, Moderately, Very, Extremely Familiarity with oral traditions Not at all, Slightly, Moderately, Very, Extremely Familiarity with African folk- Not at all, Slightly, Moderately, lore Very, Extremely Interest in learning about other cultures Not at all, Slightly, Moderately, Very, Extremely Comfort with adaptive tech (personalized content) Very Uncomfortable, Uncomfortable, Neutral, Comfortable, Very Comfortable Preferred methods Books, Museums, Interactive, Films, Online, Travel, Conversations D cultural learning User Engagement Scale Long Form

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Table 6: Summary of Factors and Items from UES-LF Questionnaire. Likert scale: 1 (Strongly Disagree) to 5 (Strongly Agree). Table 7: Summary of Cultural Interest Questionnaire Items and Likert Scale. Questions Factor Item Question Focused Attention (FA) FA.1 FA.2 FA.3 FA.4 FA.5 FA.6 FA.7 I felt immersed. I was so involved I lost track of time. I ignored things around me. I lost awareness of my surroundings. Time passed quickly. I was fully absorbed. I let myself go. PU.1 PU.2 PU.3 PU.4 PU.5 PU.6 PU.7 PU.8 I felt frustrated. I found it confusing. I felt annoyed. I felt discouraged. Using this was tiring. The experience was demanding. I felt in control. I couldn’t do some tasks. Aesthetic Appeal (AE) AE.1 AE.2 AE.3 AE.4 AE.5 The experience was visually appealing. It was aesthetically pleasing. I liked the graphics. The visuals were engaging. The screen layout looked good. Rewarding (RW) RW.1 The experience was valuable. RW.2 I consider it successful. RW.3 It didn’t work as expected. RW.4 My experience was rewarding. RW.5 I would recommend it. RW.6 I would use it again. RW.7 It sparked my curiosity. RW.8 I was drawn into it. RW.9 I felt engaged. RW.10 The experience was fun. Perceived Usability (PU) Likert Scale (1-5) How interested are you in learning Not interested at all more about Ghanaian culture after read- to Very interested ing the Anansi story? How relevant do you think the moral lessons from the Anansi story are to modern life? Not relevant at all to Very relevant How likely are you to seek out more African folklore after reading the Anansi story? Not interested at all to Very interested Anansi stories feature other interest- Not interested at all ing characters, such as Tiger, Turtle, to Very interested and Monkey. How interested are you in learning about these characters and their stories? After the user study, would you be in- Not interested at all terested in using the same system to to Very interested learn about oral traditions from other cultures, such as Grimm’s Fairy Tales? G E Cultural Interest Questionnaire F User-generated Narrations 328 Thematic Tables

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Lau et al. Beginning - The jungle whispered with life as Anansi, the clever spider, called out. 'Follow the path to your left, where you’ll find Onini, the python, coiled around her tree stump as always. We don’t have much time!' Story 1 Story 2 _____ walked down the narrow path and soon came _____ walked the path, and soon the great Onini across Onini, her massive coils shimmering in the appeared before them, her scales shimmering in the sunlight. "Greetings, _____," Onini said in a silky voice. "I dappled sunlight. "Ah, _____," Onini hissed softly, her am Onini, the mighty snake, whose coils could outmatch voice full of pride. "I am Onini, the serpent of magnificent any creature in length. If you wish to see my full length, unmatched by any in these lush jungles. Truly, to grandeur, lay a bamboo cane beside me." see my full extent, you’d need something as long as your courage. Lay down a bamboo cane on the sand beside _____ complied but noticed Onini was still wrapped me, and maybe I will stretch out to compare." around her stump. "_____, have you brought something to measure my grand length against?" Onini asked slyly. "A _____ hesitated, trying to figure out what exactly to do, bamboo cane on the sand would suffice." before eventually laying the bamboo cane on the sand. Onini watched closely _____ placed the bamboo, but Onini chuckled. "Ah, I see "Ah, _____, you’re not quite sure what to do, are you? No your bamboo cane, _____. But I find this tree stump quite worries, simply place the bamboo along the sand next to comfortable. Convince me why I should stretch out me, and we can truly see if it matches my grand scale!" alongside your cane." she chuckled, her laughter resonating through the trees. _____ hesitated and said, "I'm pretty sure it’s more _____ paused but decided to press forward. "Hey, uh, can comfortable over there. Come on." you see the bamboo?" view, her shimmering emerald scales catching the light. "Ah, _____," Onini greeted, her voice smooth and commanding. "I am Onini, the magnificent snake, whose length could astound even the Sky God. If you wish to see my full stretch, place a bamboo cane beside me." _____ carefully placed the bamboo on the sand, but Onini didn’t move. She chuckled softly. "_____, your bamboo is tempting, but why should I leave this comfortable stump? You’ll need to give me a better reason." _____ hesitated for a moment before offering, "Come closer, and you can return to your stump after." T , _____, but lying next to the bamboo needs a more Onini’s eyes glinted with curiosity. " hat’s an interesting idea compelling reason. What else can you offer?" _____ fumbled slightly. "Uh, not now, right?" she said, uncertain. Onini tilted her head slightly, amused by _____’s response. "Ah, comfort is tempting indeed, but hardly Onini tilted her head, amused. "_____, seeing the bamboo convincing," Onini replied. "Offer me a thrill or a won’t convince me. Why should I stretch along it? Give challenge instead, something worthy of my grandeur!" me a compelling reason to show my magnificent _____ stammered. "It's easier to wrap around the length." Onini let out another soft laugh, amused by the exchange. "_____, I’m waiting for your reason. Once I’m convinced, you’ll witness my full grandeur. Why should it be now?" bamboo." _____, perhaps grasping at straws, finally replied, "You Onini chuckled again. "Ah, _____, if you believe otherwise, Story 3 _____ walked down the path, and soon Onini came into are so beautiful and strong." dare me to compare my length to your bamboo cane.” Onini’s eyes glimmered with pleasure. "Ah, flattery! It does have its appeal, _____.” Gathering her thoughts, _____ replied, "There’s a mouse over there. You can prey on it and then return." A, !" she exclaimed. "Very well!” Onini’s eyes gleamed with excitement. " h a feast waiting for me Ending - With that, Onini slowly uncoiled from her stump, her immense body stretching out along the ground, her shimmering scales catching the light as she revealed her full, awe-inspiring length. 'Now, behold the longest snake in all of Ghana!' she declared proudly, her voice echoing through the jungle. Figure 9: Figure illustrates three user-generated stories in Anansi the Spider VR. Blue denotes user inputs via speech-to-text, actively shaping the tale, while red illustrates AI-generated responses via text-to-speech, demonstrating the dynamic interplay between participant choices and Gen-AI storytelling. 329

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates Themes Quotes Sub-themes A - Immersion: Participants felt physically present in the virtual environment, with sensory and emotional stimuli deepening their immersive experience. Environmental Realis At some moments, I was like, grab me, I’m going to fall. I was feeling like I was really in a space, but no one was around me, and I was just going through the story. (VP8 I felt like I was part of a different world, a virtual world. (VNP2 The biggest compliment would be that I just lost track of time and also completely felt the surroundings. Like, I was completely immersed in the jungle. (VNP5 Sometimes I forgot that I was in the room, and I hit the table because I thought the controller could go through. (VNP1) A1 - Sensory Experience: The sensory inputs, such as sight and sound, play a crucial role in enhancing the immersiveness of the VR experience. This heightened immersion often leads to participants experiencing physical sensations, such as tickling or the fear of hitting objects, as their senses respond to the virtual environment. Physical Sensatio When I was on top of the mountain, I actually felt a strange, tickling feeling on my feet. (VNP3 When I had to walk through and there were trees, I was bending my head because I felt like they were going to hit me. (VP1) It was really realistic, there were times when it was really scary, so I focused on every part. It was nice to explore and try different strategies. (VNP6 It felt real. You weren't using your legs, though, but moving. I saw these animals, and I felt a bit scared. (VP5 Sometimes the visuals felt too real and scary, and I felt like I was really in danger or was going to fall. (VP8 It felt weirdly real to the point where I got a bit uncomfortable when I got to a height—I am scared of heights. (VP10) A2 - Emotional Reactions: Emotional responses like realism, fear, and anxiety are intensified by the VR environment, making the experience more engaging and memorable. This emotional depth enhances the overall immersion. The gods were very detailed and they looked very nice, so it really imprinted in my memory how they looked. (VP9 Compared to other mediums, I think this aspect blocks other stimuli in the background so you really focus on the story. If I was reading, I will have a lot of distractions around me or if I was watching something on a laptop, but this forces you to fully focus on the environment and it also has the element of view interacting with the environment, so it makes it feel more tangible and real. (VNP2) A3 - Cognitive Engagement: The VR experience enhances focus by immersing participants in detailed visuals and narrative, reducing distractions and making the experience feel more tangible and real. Figure 10: Thematic analysis results: Immersion. B - Personalization: The experience tailored story elements to individual users, such as including their names and offering choices, which increased their connection to and immersion in the narrative. Having to walk through different environments and maneuvering and taking actions made me feel like I was the center of the story. (VNP2 I thought it was really interesting because I really felt that I had to participate in the story, and this contributed to my interest and attention. (VNP6 The aspect that stood out to me the most was that it was personalized to my name. It felt like I was playing a character in a way that I was involved in the story. (NVP1 I feel like the fact that the story calling out my name, makes me feel like I am needed and I would like to keep going in the story. (VP7 The first time [I heard my name] I was like, oh they know my name, and they’re pronouncing it well, so I was like wow. (VP4 The experience made me feel almost like a real character. It felt like I was chatting with my friends or someone who really knows me. (VP6 It felt like it was talking to me, which was pretty nice, felt more direct or personalized than general life stories that can be shared across the board. (VP10) B1 - Story Involvement: Personalisation made participants feel more central to the narrative, enhancing their sense of involvement by directly incorporating their names and choices into the story, making them feel like active participants rather than passive observers. It feels like I am the main hero helping Anansi. (NVP2 I am open to embracing any medium that brings the user, or the hero of the story, closer to the reality of the narrative. (VNP2 It felt like a personalized touch, like I am one of the main characters and I was playing a role and I felt like I’m doing something great. (VP5 Hearing my name fascinated me. When you see a movie, and say a character’s name, it was just like that, saying my name. (VP8) B2 - Identification with Protagonist: Through The character was speaking to me, not just going on telling a story, and gave me tasks, and I think the personalization was great without changing the story too much. (VP3 The method I used to catch the snake because it was kind of smart, but I convinced it by saying I will bring it to a safer place and trick it. I felt good, my method was intelligence and also resourceful so I got the moral message too. (VP8 I felt wise tricking the snake and it was rewarding to hear you did a good job. (VP5) B3 - Character Interaction: Personalisation enhanced interactions with story characters, making these exchanges feel more meaningful and tailored, thereby enriching the overall narrative experience. Figure 11: Thematic analysis results: Personalization. 330 personalisation, participants identified themselves as the main character or hero of the story, increasing their connection and emotional investment in the narrative.

AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates C - Engagement: Participants were consistently involved in the experience. Lau et al. Physical Interaction The story was organized in a way that constantly asked you to do something, like move to a place or grab something, which kept you engaged and on your toes. (VNP5 It was nice to move our hands and literally be there, instead of just looking at a screen. (VNP1 The movement in the virtual world around me, turning in the real world and seeing the virtual world also increase my engagement. (VP3 The interactivity was what made it so engaging and immersive. (VP3) C1 - Interactive Elements: Physical movement and task-oriented activities within the VR environment enhance participant engagement by making them active participants in the story. The moral of the story could have impacted me more if I was more involved. (NVNP2 It was like feeling like the main character, the story would be more engraved in my mind. (NVNP2 The waterfall and nature sounds, that is when VR and AR is really helpful to take you out of the room. (VNP2 VR will take my attention or motivation quickly, listening to the story in the classroom settings but in VR is different because there is more context inside the story. (VP2 Compared to oral storytelling or watching a movie, the effort it takes to get engaged was less in VR, and my interest grew much easier. (VP8) C2 - Immersive Experience: The immersive nature of VR, characterized by a strong sense of presence and the feeling of being the main character, makes the experience more contextual and engaging. This was the most beautiful VR experience I have ever had. (VNP3 The graphics were amazing, it really got me into the game. (VP4 I think this was my best experience in VR. It felt super real, graphic scene, unlike lots of games have too many shapes, polygons, but this felt real. (VP6 I learned a story, I learned about a culture today. Because of the illustrations, I lived it. I experienced it but I also lived it, so I call this learning, learning is to live it. (VNP3) C3 - Aesthetic Appeal: The visual quality of the VR environment captures participants' attention, contributing to an engaging experience. I think this story will stay with me for a lifetime. Because stories that I heard when I was younger, I forget the details, but here, I was experiencing it. I felt, in the moment, and you can never forget what you feel physically, and the memory captures moments better than just listening. (VP5 After this, I am going to Google more about the Anansi story because I want to know more. (VP4) C4 - Continuous Learning: The continuous flow of the VR experience sustains participants' interest, leading to ongoing engagement both during and after the activity. Figure 12: Thematic analysis results: Engagement. D - Reflection: Participants deeply reflected on the personal and cultural implications of the story, applying its lessons to their own lives. The story reminds me of recent situations where I have been working on a bunch of jobseeking applications. Instead of personalizing each one, I decided to personalize only the positions I am truly aiming for and mass-send the rest. (VP7 As a university student, this reminded me a lot about being a good student, being intelligent. And being in a foreign culture, you sort of think of the many times where you had to rely on your intelligence rather than just brute strength. I think the story passes on this good moral message. I think the moral of the story is still very, very relevant. (VNP2) D1 - Reflections on Personal Experience: Participants related the story to their own life experiences, such as job-seeking and navigating foreign cultures, emphasizing the relevance of the story's lessons in their personal lives. Maybe after getting mugged, it’s important to be clever and have good timing to get out of a dangerous situation and stay alive. (VP3 You have to think before acting, and you have to make smart moves. The snake is very strong, but you can be stronger with your intelligence. (VP5 There are also people who are strong and can do everything in daily life, but if you use the correct path, you can do everything you want. For example, knowing how to use technology can make you win. (VP2 Something I can take from this experience would be I could be better at communicating because, you know, on the first try, I could not convince the snake to do what I wanted her to do, but I complimented and said some nice words, and it convinced her. (VP5 Instead of just power, being smarter might make the world much more peaceful. (VP6 It made me think about how you have to be patient. (VNP6 Smart work is better than hard work. (NVNP1) D2 - Reflections on Daily Life Situations: Participants drew parallels between the story and everyday challenges, such as communication, patience, and strategic thinking, recognizing the story's applicability to common life scenarios. Coming from a Catholic background, it reminded me of David and Goliath. It’s in the Bible, but my grandfather used to tell it to me like a fairy tale, and it reminded me of that. (VP4 My family used to tell me stories, and they come from different cities, so they would tell me stories and show me pictures. (VNP6) D3 - Reflection on Cultural and Religious Connections: Participants connected the story to their cultural and religious backgrounds, finding similarities with familiar tales and traditions, which deepened their engagement with the narrative. hen someone tells you a story, there is a personal touch to it. But this is more like escapism. One feels less warm and connected, but this does feel more novel and free. (NVP4 I think nowadays, a lot of people try to trick you into doing something, like scammers online, so just watch out for it and be clever, so you don't get tricked. (VP9 Be humble, don’t let others trick you because you’re not humble. (VP9) D4 - Moral Lessons and Awareness: Participants reflected on the moral lessons presented in the story, considering how these teachings applied to their behavior and awareness in situations like online scams and humility. W Figure 13: Thematic analysis results: Reflection. 331

Gen-AI Personalization and VR Immersion in Oral Storytelling AHs ’25, March 16–20, 2025, Masdar City, Abu Dhabi, United Arab Emirates E - Culture: The storytelling increased appreciation I think African stories are very far away from me. But this game made it so close to me. So I E2 - Appreciation and Interest: The immersive for African folklore, enriching participants' was more interested, even though I am not, usually. (VP6 storytelling made the culture feel more accessible understanding of cultural heritage and promoting I appreciated all cultures and find it fascinating to see how different cultures differ in so and relevant, even for those who initially felt distant cultural exchange. many activities but share common values of life, but it was nice to learn the characters that from it. were represented like the sky god and the spiders. (VNP5 After this experience I am eager to learn more about this story and culture. (VNP6) Although familiar with Anansi and spider stories, said that in his region of Kenya, they are not E3 - Cultural Authenticity and Representation: very popular. Animal morals contain animals specific to the area that the culture comes Reflections on the authenticity of cultural portrayals, from. It is usually more about animals from the wild, and not spiders where I am from. (VNP2 emphasizing the importance of accurate and Growing up, every time my family members would say a word which we don’t know, they resonant cultural elements. always had a story behind it, so they have to tell us a story to show us what it actually means and why it’s still relevant today. (VP8) : About African culture I learned that animals represent some emotions and authority figures, and want to share a moral through animal figures. It is also like this in my culture, T urkish culture, because animals were there before humans. T hey use animals that we face 4 E P - Relation to between A ersonal Cultural xp immersive e p erience k frican fol p cultural traditions, dee authority figures, like snakes, huge spiders, you may be scared of them if you see them in their heritage. ? B kg ac round: The ted connections lore and in our daily lives, and they need to represent you know, some deeper meaning, like strong real life, right p rom p p ' pp artici ants ening their a own reciation of So that’s why, I think that’s how they are trying to explain something bigger in 3 the story. (VNP I have this group of people from India and we meet every week and we talked about how in today’s generation, it is less storytelling that what kids are attracted to, it is more like video games, so we talked about video games to share the routes that the heritage and they are coming from, maybe some of the old warriors. (VP8) :W After the story, I thought to myself ow, two incidents about Africa I need to learn about. 5 E x - Cultural E g chan p e and p S g: harin p (NVP2 The storytelling ins I will also tell my niece and nephew about this story. (NVP2) newfound cultural insights with others, ired wider dissemination of artici ants to share their A p romoting frican narratives. Figure 14: Thematic analysis results: Culture. E - Technological Intervention: This was more interesting than previous VR experiences because I felt like I was E1 - Changes in interest or The changes of medium such learning something. Previously, I was doing tiny actions and couldn’t see a clear Attitudes towards African Culture: as VR and AI, transform the purpose, but while doing this experience, I felt like I was learning about a new culture The use of VR shifted participants' way participants engage with and I found that really interesting. (VNP6) perceptions, increasing their and learn about culture, interest in and appreciation for enhancing both interest and African culture through a more understanding through engaging and impactful immersive and interactive experience. experiences. Oh big time. Because, of course, the generation before us, they are already grown up E2 - Educational Impact and and already know these things, but the generation right now and the generation which Learning: VR and AI facilitated is coming would be more interested towards technology, AI, and VR, because they are deeper cultural learning by already playing games, have mobile phones, etc. It is a way to make them understand providing an immersive experience their culture. (VP8 that made complex cultural It’s sometimes difficult for people to travel to Peru to experience how the culture is, and aspects more accessible and it’s different than reading about it. I think VR could be a really good alternative. (VNP6 easier to understand. You can educate about different cultures, not just folklore stories, but things like culture aspects that you can only learn by going to Africa, a particular city in Africa, to get the experience of how the people live their life there, so we can appreciate it more. (VNP5) Figure 15: Thematic analysis results: Technological Intervention. 332