TL;DR
This paper surveys the rapidly expanding field of computational and emergent digital storytelling, categorizing existing methods, frameworks, and tools. It highlights the increasing role of AI and mixed reality technologies in creating dynamic, personalized narratives, while also raising crucial ethical questions about human versus machine creativity.
What Problem Does This Paper Address?
The authors address the challenge of navigating the vast and interdisciplinary landscape of digital storytelling (DS). While DS has been a fundamental human activity for millennia, its computational and emergent forms are rapidly evolving, making it difficult to track current trends, identify key methodologies, and understand future research directions. The paper aims to provide a structured overview of this field, focusing on works that leverage computational and emergent methods for narrative creation, and to offer a classification system for these diverse approaches.
Methods
The paper employs a systematic survey methodology. The authors collected and filtered hundreds of works based on their interactivity and potential for computational content creation, with a specific focus on works published after 2015. They then categorized these works into three main groups: DS Authoring Tools, DS Systems and Applications, and DS Methods, Frameworks, and Case Studies. A secondary classification was applied based on the scope and purpose of each work (e.g., education, cultural heritage, games, healthcare). The survey also discusses various media and interaction types, including tangible IDN, gesture recognition, embodied systems, and the mixing of AR, VR, and 360° video.
Datasets
Not discussed in the paper.
Key Insights and Ideas
- The "narrative paradox": There's an inherent tension between user freedom in shaping a story and the author's intended narrative design and coherence.
- Emergent vs. Top-Down Approaches: Narrative creation can be emergent (user-driven, bottom-up) or top-down (system-managed with predefined rules).
- AI's Growing Role: AI is increasingly being used to generate and understand stories, assist human storytellers, and model narrative elements, raising questions about the future of human creativity.
- Multimodality is Key: Digital storytelling thrives on integrating images, music, narratives, and voice to create richer, more immersive experiences.
- Beyond Screens: While screen-based interactions dominate, there's a growing interest in tangible and embodied storytelling methods that engage users physically.
- Ethical Considerations: The rise of AI in storytelling brings ethical dilemmas, including the potential for bias, stereotypes, and the displacement of human creative roles.
- Interdisciplinary Nature: The field draws from computer science, linguistics, psychology, and the arts, highlighting its broad applicability.
Key Results
The survey identified and categorized 92 DS works. The authors found that:
- DS Authoring Tools are crucial for speeding up narrative creation and enriching content with multimedia.
- DS Systems and Applications are increasingly leveraging VR, AR, and robotic mechanisms to enhance user experience and immersion.
- DS Methods, Frameworks, and Case Studies are exploring theoretical underpinnings, AI integration, and new narrative structures.
- The research is highly active, with a significant portion of works focusing on general-purpose computational and emergent storytelling, indicating a broad interest in understanding and advancing the field.
- The paper highlights the increasing use of AI technologies like GPT-3 for narrative generation and the growing importance of mixed reality (AR/VR) in creating immersive experiences.
Limitations and Future Work
- Limitations: The authors acknowledge that the analysis is based on a selection of works and that the field is constantly evolving. They also note that the availability of implementation details can vary, making direct comparison challenging. The ethical implications of AI in storytelling are a significant concern that requires further exploration.
- Future Work: The paper suggests several future directions:
- AI-driven narrative generation: Further research into training AI models with valid scientific data to create personalized narratives for specific contexts (e.g., museum exhibits).
- University education: The emergence of programs for interactive narrative design in both theoretical and technological departments.
- School education: Leveraging storytelling technologies to make interactive narrative creation accessible even to elementary students.
- Social robots: Increased use of social robots for companionship and communication, with advancements in speech and conversational algorithms.
- Ethical frameworks: Developing robust ethical guidelines for AI-generated narratives to prevent bias and ensure responsible use.
- Bridging human and artificial creativity: Exploring how AI can augment, rather than replace, human creativity in storytelling.
Summary updated