Interactive Storytelling

Artificial Intelligence methods open up new possibilities in interactive storytelling, enabling the creation of believable characters with rich personalities and emotions, interactive story systems that incorporate player interaction into the construction of dynamic plots, and story generation systems that capture large and well-formed collections of potential stories. The goal of these approaches is not to replace human authorship, but rather to move human authorship to a meta-level, and thus to support a richness and depth of player interaction that is not otherwise possible. However, there are significant authoring challenges in creating AI-based interactive stories. This talk will describe current research efforts to support authors in telling stories in this new medium.

Interactive storytelling has been present at the heart of digital entertainment media for over thirty years, however the breadth of its narrative scope has remained stifled. As computational boundaries are eased, so are many of the perceived technical obstacles to generated narrative content. Furthermore, there is a sense that notable commercial successes are thawing the professional bias towards authored content. Powerful tools that permit vast and complex worlds to be built have mined gameplay in the sandbox genre. With much of the content generated procedurally the designers have still maintained a strong authorial voice. Presenting similar solutions within the narrative scope can win further converts provided that they are sensitive to the commercial requirements. In this presentation we will explore what the digital entertainment industry has done in the field of interactive storytelling, explore where successes might be reinforced and imagine what it might achieve in the immediate future.

The past 20 years have seen Live Action Role Playing evolve to become a complex scene of games in the Nordic countries. The Nordic LARP scene has been a creative, dynamic and experimental environment where questions regarding the design of intrigue and interaction have been explored from many perspectives. Over the past years the knowledge within this subcultural context has spread into several academic fields, into the schools of the nordic countries, to the national theatre stages, into other types of games and also into mainstream media. During this keynote Carl Heath will take us on a journey into this subculture, on an exploratory path aiming to explain key elements in the design of the rich and complex social worlds that can be found within LARPs in the Nordic countries.

Despite advances in game technology, most stories constructed by game designers remain inherently linear in nature, and player actions often have limited impact on the central story. In interactive storytelling approaches, an important challenge is the creation of stable yet dynamic environments to allow the emergence of unscripted stories involving both human-controlled characters and autonomous non-player characters (NPCs). In this paper, we present an architectural design for creating open-ended, interactive storytelling systems in which story structure emerges in real time and in response to player actions, thus providing a greater variety of game experiences than more scripted approaches. We present a partial implementation of the approach in a virtual environment populated by multiple NPCs that exhibit stable but interesting autonomous behavior. Finally, we present experimental results that demonstrate the scalability of the approach and variability of NPC behavior that it produces.

In order for the author to create his/her intended effects using interactive narratives, the user has to be able to understand his/her experience as designed by the author. In this paper, we argue that a key desideratum for interactive narrative frameworks is to model the characters’ motivational consistency during the interaction. This work reports an empirical study for evaluating the importance of using well-motivated characters in interactive narratives. The results demonstrate that inconsistency in the characters’ motivations can confuse the user and affect the user’s expectations and interpretations of the events in the story.

In this paper we consider the issues involved in influencing a user in an interactive storytelling context using results from the social psychology’s area of persuasion. We hypothesize that it is possible to use these results in order to influence the user in predictable ways. Several important concepts of persuasion, such as how people make decisions, and how can we influence that process are discussed. We describe a proposal on how to apply these results in an interactive storytelling setting and describe a small study were we have successfully influenced the players of a story in following a specific path by using an expert source manipulation.

With the increasing number of prototypes and market applications of interactive storytelling, the understanding and optimization of how end users respond to computer-mediated interactive narratives is of growing importance. Based on a conceptual model of user experiences in interactive storytelling, a measurement instrument for empirical user-based research was developed. We report findings from an initial test of the self-report scales that was conducted with N=80 players of the adventure game ”Fahrenheit”. Interactivity was manipulated experimentally in order to validate the measures. Results suggest that the scales will be useful for comparing user responses to ‘real’ interactive storytelling systems.

We investigated 101 online hyperstories created at school by students aged 4-18 using the same authoring tool, to discover recurring characteristics of youngers’ hypermedia narratives and their underlying creative process. The findings of our empirical study offer a contribution to the digital storytelling field from a conceptual and a practical perspective. They provide a deeper understanding of digital storytelling activities performed by youngsters in educational settings, highlighting a tension between regularity and creativity that has some correlations with authors’ age and school context. In addition, the discovered recurrent features can be used as classification means for digital stories and can be translated into guidelines to support the authoring process and to improve the functionalities of storytelling tools.

Adaptation of stories – as a translation between media, such as literature and film – is explored for genres of interactive storytelling that make use of highly-interactive and user-adaptive technology. A concrete case study of transforming and abstracting a Hemingway short story is discussed in detail. The conclusion is that even though Interactive Storytelling content has to follow formal models, these cannot be derived from a written narrative alone and need story creators’ input in order to work for interactivity.

Narratives are widespread in multimedia, especially in linear audiovisuals. However, rare are the approaches that deal with video annotation and editing with respect to the narrative content. In this paper, we introduce an annotation schema for the narrative features of media objects, that relies on a formal theory of story and characters, and we describe a software tool, called Cinematic, for the annotation of video objects and the automatic editing of the annotated objects.

In this paper we present an approach for an association-based story environment, in which a priori unrelated experiences represented in images, are stitched together to guide users through interesting city spaces. We describe the associative stories generated by this system, and outline how the notion of ’hypespot’ facilitates linking the real world with the structure of the story. We outline the overall architecture of the system and provide a generated story example.

The paper presents a framework that allows the collection of multiple story fragments from several sources and/or authors, in the context of social networks, where story fragments tell facts about items that are ontologically modeled in the system. The framework provides tools for threading facts together into stories; by doing so, it shapes a new narratological model, that mixes emergent narrative and authored approaches, and that can be defined as ”collective”.

This paper introduces an application for learning story marketing using a story creation system, Anime de Blog. Anime de Blog is an animation-based consumer-generated media (CGM) application. Users can create animated stories of affiliate advertising by simply inputting basic words using Anime de Blog. Through practical work experience of affiliate advertising on the Web by undergraduate students using Anime de Blog, students learned about an internet application, story marketing, and the concepts of affiliates and affiliate advertisements; they could also critically evaluate the application.

This paper looks at the current state of the art of academic sports commentary generation systems. It then summarises work on how drama is created in professional broadcast commentary using dramatic devices. A novel computational architecture for sports commentary generation based on analyses of professional broadcast commentary is then presented. This work presents a first step towards realising a computational model of real-time sports commentary which explicitly incorporates dramatic narrative devices in the form of dramatic structure, themes and texturing.

This paper presents an expressive activity program and a support system for collaborative storytelling based on multiple compositions: Zuzie. Zuzie’s storytelling is designed so that people can experience rich expressions and extend their interpretation of them through creation of compositions. Each composition is a deliberate arrangement of a single set of figures on a background plane. Coupled with the expressive activity program, this research is intended to design and implement a system that supports multiple compositions. This report describes our Zuzie expressive activity and presents discussion of the expressions through the study and artworks.

In the last few years the word ”documentary” has been loosely used to describe multimedia pieces that incorporate video no matter its nature, technique, language or scope, taking advantage of the fuzzy and fragile boundaries of the documentary definition. The present manifesto aims to give a brief insight on the interactive documentary arena and also to sketch some production remarks for future interactive documentary productions.

This case study of ”Virtuelle Mauer/ReConstructing the Wall,” a prize-winning interactive 3D installation on the Berlin Wall, describes an approach using artistic concepts of abstraction and interactive narratology to simplify production and focus on subtle but powerful user interactions with the virtual world. Interactivity is predetermined and invariable, but after two years of exhibitions in Europe, the USA and South Asia, the authors are confident that it provides an engrossing, affective experience for a general public across cultures and including schoolchildren and their teachers, contemporary witnesses and their families, historians and historic preservationists.

This paper is a study of the different characteristics of the presentation media of artwork that can exist in physical and non-material form. The paper explores how these characteristics affect the design of an artwork. Physical in this context refers to physical installations, and non-material is used to define artworks where the designer has little or no control over the presentation media, such as artwork that exists online.

This paper describes the initial stages for the iLand project, a Location Aware Multimedia Story project that captures and exposes the rich oral culture and traditions at the Island of Madeira, using the Madeira main city, Funchal, as a setting to bring a new level of engagement of the audience with the city and its traditional stories. We developed high quality content to be used in an already existing platform to deliver location aware stories. With the story experience carefully designed, an evaluation was carried out in order to expose the opportunities where such systems can be improved. Finally we discuss the results from the evaluation and explain how we will address them in the design of our new system.

Research on Interactive Storytelling has mainly focused on the opposition between a discourse point of view and a character point of view for a story. In this paper, we propose an approach to model Interactive Storytelling using Linear Logic, which is a support for reasoning on both points of view. Then we give an example on an educational game that shows the possibility of validating a scenario of a story, by applying the proof graph of a Linear Logic sequent. Finally we discuss about issues which should be settled and future works to be done on the Linear Logic approach for the IS modeling.

Our continued investigation into the experience of improvisers as they construct narrative in improvisational theatre provides a meaningful decomposition of its atomic unit, the offer. Our study was conducted with improvisers performing improv “games” in their theatre with each performance video recorded. Individual participants were selectively shown individual performances before being interviewed. This process is meant to elicit deeper information into how the performer chooses specific narrative interactions to develop in an improvisation performance. This paper presents our ongoing findings related to narrative development in improvisational theatre and how they were used to create an improvisational micro-agent. These findings have demonstrated that the use of offers to construct a scene involves the offers’ acceptance and augmentation in a scene more than just the strength of an offer.

The emerging artistic practice of interactive narrative in digital media marks a profound departure from traditional narrative. The application of traditional narrative theory for interactive narrative is problematic, since the affordances of digital media challenge many underlying assumptions of theories related to non-digital media. This paper proposes a theoretical framework for interactive storytelling, which addresses these concerns by foregrounding system (the digital artifact) and process (the user interacting with the system) over the product-centered view of legacy media. On this basis, protostory, narrative design, and narrative vectors are proposed as new terms to more adequately describe the structure of narrative in interactive digital storytelling. This move is also relevant for practical design given the influence theoretical concepts have on concrete implementations.

In this paper we describe a data-driven interactive storytelling system similar to previous work by Gordon & Swanson. We addresses some of the problems of their system, by combining information retrieval, machine learning and natural language processing. To evaluate our system, we leverage emerging crowd-sourcing communities to collect orders of magnitude more data and show statistical improvement over their system. The end result is a computer agent capable of contributing to stories that are nearly indistinguishable form entirely human written ones to outside observers.

One can see interactive Storytelling (IS) as a process where participants and system (through its setting, characters and events) cooperate in order to converge towards a good story experience. The notion of good story experience is very subjective and controversial, not only because it’s directly related to participants’ sense of logical and chronological sequence but also to their model of human experience [22]. Recent IS research inspired by improvisational theatre (improv) has taken this problem to a new level where the IS process occurs in extremely dynamic environments. In this paper we approach the theoretical groundings towards the design of a story appraisal mechanism for autonomous agents in an improv environment.

We have implemented an action-based role-playing game, called Wind’s End, that incorporates a simple, practical algorithm to enforce rising dramatic intensity during game play by controlling the choice of goals by nonplayer characters.

While many forms of storytelling are well-suited to the domain of entertainment, interactive storytelling remains unique in its ability to also afford its audiences a sense of having influence over what will happen next. We propose that the key to encouraging such feelings of agency in interactive stories lies in managing the perceived relevance of the decisions that players make while they play. To this end, we present the design of a system which automatically estimates the relevance of in-game decisions for each particular player, based on a dynamically learned model of their preferences for story content. By actively choosing among several potential consequences of a given player decision, the proposed system highlights the relevance of each decision while accommodating for its players’ preferences over potential story content.

In this paper, we describe an approach to use storytelling in academic teaching as a means for background research to hypermedia and virtual reality topics in computer science. Interactivity within this context means selective authoring rather than immersive interaction. In contrast to existing approaches a Hypermedia Novel environment allows an iterative approach to the narrative content, thereby integrating story authoring and story reception at any time. The narrative practice and background research as well as the resulting product can supplement lecture material with comparable success to traditional academic teaching approaches. In addition there is the added value of soft skill training and a gain of expert knowledge in areas of personal background research.

This paper presents first results from an ongoing research initiative to develop an interactive storytelling application to support teaching English as a second language for children in schools. Considering the necessity to accommodate to both students’ and teachers’ needs, we provide a summarized review of relevant systems and discuss the envisioned challenges concerning the specific aim we are addressing. Furthermore, we present different usage scenarios closely related to the concept of role playing and its application within this problem space.

In this paper, we present a preliminary evaluation of a text-based and graphical version of an interactive fiction game that we created to look at how the user experience varies across the different mediums and modalities.

This paper describes a preliminary investigation into the concept of rereadability in interactive stories. Through a close reading of the text-based interactive fiction

Alabaster

, seven possible motivations for rereading an interactive story are identified. This close reading suggests that, while there are many ways in which rereadability in interactive stories is similar to rereadability in traditional (non-interactive) fiction, there are also forms of rereadability that are unique to interactive storytelling.

The author summarizes his experience creating a story-based simulation from raw documentary footage taken in the aftermath of the 2010 earthquake in Haiti. The use of Propp’s typology to create a heroic framework within which to organize the material is explained. The author introduces the concept of Negotiation Clouds to improve a fold-back story structure and create meaningful agency for the user without compromising author control of the narrative.

Interactive Drama in 3D worlds has great potential for communicating serious themes, however it can become challenging to organize the content in such a way that the theme is communicated clearly while maintaining the feeling of free spatial navigation in the 3D world. In order to address this problem, and to propose a way to structure content, we have developed the Interactive Dramatic Experience Model, which attempts to organize narrative events in a 3D world while keeping the freedom of spatial interactivity. In order to exemplify this model, we have chosen to oppose the classic genre of violent interactive shooter experiences by allowing the participants to experience the feeling of being a victim of war. An evaluation of the implementation indicated that participants experienced free spatial interaction, while still being able to acquire an understanding of the theme being mediated.

Improving the visiting experience of exhibitions and public spaces in general has been the subject of several studies over the past years. In the study presented here, we were particularly interested in understanding the potential of combining explicit and implicit interaction modes between virtual characters and visitors. We present in this paper scenarios that exemplify the combination of these interaction modes, and an initial study with user involvement, based on a software platform that we are currently developing. We report on first feedback from users about the system-level interaction, usability, and visiting experience. The tested case study should create a sense of ubiquity of the virtual characters throughout the visit, and take advantage of their communication skills, at the same time giving freedom to visiting groups to interact with each other and to make visits to the exhibition place at the order they desire.

The author proposes a hypothesis that metalepsis by the narratee into the diegetic level of the story and therefore into the intradiegetic narratee position is a major ingredient in what makes some interactive narratives compelling, and that the move into this position is unconscious and a function of the porosity between narrative layers that interaction can engender in narrative. He illustrates his proposition with a storytelling ritual that is played out between a six year old child and himself.

Sports broadcasting typically involves a play-by-play commentator and a color commentator. The color commentator’s function is to entertain the viewer during the game. The prevalent way of doing so is by telling brief stories relevant to the game in progress. In this paper, we propose that storytelling in sports is a challenging and rich problem for Artificial Intelligence (AI) research for the following reasons. First, storytelling is considered to be a very “human” activity as it requires engaging the audience in a story and appealing to their factual as well as emotional sides. Thus, automating it will be an advance in AI. Second, while automated storytelling in general has received ample attention, storytelling in sports has seen much less research. Third, an AI storyteller can be used as an assistant to a human color commentator in a real game, or autonomously in the context of a sports video game. Fourth, the task is non-trivial as we demonstrate by applying off-the-shelf machine learning methods in an attempt to match baseball stories to baseball game states.

We propose a novel approach - employing of level-of-detail with stories. The key idea is to create a tree representation of a story, where the root represents a single annotated image, and the bottom a selection of several hundreds of annotated images. The time available for each percipient of the LOD story identifies the number of perceivable images from this story. In other words, the root offers the story’s main picture and message (one-liner), while the leaf nodes mean the full-length comics strip. The LOD-story input is the time available to a virtual tourist. The output is given by the presentation itself - and its length.

We document the proof-of-the-concept with a feature film directed by Martin Tapak and produced by Peter Rufus. Matej Zeman art-photographed the movie creation and his hundreds of images are combined with movie script for the LODstory. This LOD-story is intended for use in the first Slovak virtual museum on cratft and art of tinkering.

Digital storytelling refers to use digital tools so that people can tell their stories. That is, it is a technological application that allows people to communicate a

message

to other people.

Due to that reason, from authors’ point of view, the selection of the most adequate communication channel should be a crucial step in order to transmit correctly the story.

Taking seriously Andrew Stern’s aspiration that IDS become a premier art form for the 21

st

century, this paper re-examines agency, understood as the ability to freely control the plot, as a key concept in IDS aesthetics. Tracing the origins of this notion in IDS theory, this paper suggests that ”true” agency is a myth, and that even restricted agency is too constrained to serve as a desirable goal for IDS-as-art.

It is a popular idea that violence and addiction (long term immersion in virtual worlds) are negative outcomes of game’s content and graphic. The paper explores weather those elements can serve as the starting point in research of the role of realistic representation of violence. By analysing basic structural elements of games (topography, narration and action), the paper will propose formal analyse as precondition for phenomenology of games. The game Carmageddon will serve as base for analyse of new type of simulacrum that is ”action oriented”.

Emohawk is a narrative-based serious game designed to be a supportive tool for teaching basics of virtual agents development at universities and high-schools. Emohawk is built utilizing a free version of Unreal Engine 2 and it features an interactive scenario with four virtual agents controlled by an appraisal-driven architecture playing out a story approximately 5-10 minutes long. Students are engaged in solving game-based tasks with increasing complexity and simple programming tasks related to various parts of the virtual agents curricula. The Emohawk distribution includes documentation, graphical debugging tools and tutorials. The project is in continuous development and we plan a large evaluation for the 2010/2011 academic year.

ENIGMA is an experimental platform for collaborative authoring of the behaviour of autonomous virtual characters in interactive narrative applications. The main objective of this system is to overcome the bottleneck of knowledge acquisition that exists in generative storytelling systems through a combination of crowd-sourcing and machine learning. While the authoring front-end of the application is used to create short example stories set in a specific story domain, the server side of the application collects many of those stories and derives behaviour models for autonomous virtual characters such as formal planning operator descriptions from them. A mixed initiative mode increases coherence by feeding already learnt character behaviour back into the client.

Traumatic brain injury (TBI) is a leading cause of disability and has a devastating impact on all family members. The goal of our project is to design, develop, and evaluate a pedagogical interactive drama to support uninjured youths aged 12 to 19 years who have a parent or sibling with a TBI. We used IDtension, a highly interactive drama engine, to develop a text-based interactive game, in which young users play the role of Frank, who has to prepare a dinner with Paul, his father, who suffers from a TBI.

The hyperglobe (HG) will be introduced both as a technological and a semiotic-communicative model. We will argue why and how concepts from the domain of digital storytelling can be used for developing authoring software for a HG (i.e. for a spherical display) and outline the HG’s future potential as a narrative platform for multidisciplinary research.

The manuscript includes information about the objectives, methods, participation requirements and references for the Users and Evaluation of Interactive Storytelling at the 3rd International Conference on Interactive Digital Storytelling – Edinburgh 2010.

The last 10 years of research in the domain of Interactive Digital Storytelling have led to several advanced systems [1], often based on Artificial Intelligence, which have been demonstrated on stories usually created by the systems’ inventors themselves. To leverage the potential of these systems and to create more and particularly more compelling interactive narratives, it is essential to involve experienced story creators as authors. These authors often have an intermediate level of creative experience in some storytelling discipline, as well as certain knowledge of information technology. However, they may lack the technical and conceptual knowledge that is necessary to grasp the complexity of the systems produced by research groups. This explains why few of them have dared to write stories for the above mentioned systems. On the other hand, current storyworld creation by engineers lacks the conceptual knowledge of creative principles for narrative and drama.

In recent years, there has been an increasing interest in exploring virtual environments and computer aided interactive stories as tools in developing health promotion and disease prevention interventions. Applications have been developed to address a range of health related conditions, including stress [1], risky behaviors [2] and post-traumatic stress disorder(PTSD) [3].

Compared to conventional intervention techniques, which usually require face-toface interactions with clinicians, computer aided interactive stories have several advantages. They are less expensive to the user. The user can access the materials at any time and from his/her convenient locations. The privacy provided by computer aided interventions can help engage the user more efficiently. People who are suffering from mental health conditions often do not actively seek treatment because of the perceived stigma. Interacting with a computer program can make them feel less embarrassed and more in control.

Moreover, story itself is a powerful tool to teach and change people’s behaviors. The support of interactivity makes interactive stories even more powerful by allowing the user to experience and learn in context. Further, the story, and intervention messages, can be tailored based on user profiles and the user’s patterns of interaction within the intervention.

As a new research domain matures, it becomes increasingly important for researchers to agree on a shared vocabulary. For researchers in Interactive Digital Storytelling, this is a particular challenge, because researchers come from many different domains and bring their own terminology with them. This workshop exposes and explores the differences in meaning and terminology of key terms such as “story” in different academic disciplines related to Interactive Digital Storytelling. In order to minimize confusion and misunderstandings in the academic discussion within the field and with outside disciplines, the workshop explores ways towards a shared vocabulary for Interactive Digital Storytelling.

The movement from a screen based experience of the internet to one in which everything is connected in the actual world is slowly becoming a reality. The advent of smart phones and data free contracts has provided a cultural context in Europe in which society will begin reading and writing stories to objects of their own. Digital Storytelling methodologies are required to understand how narratives between objects can be constructed as individual objects become connected with others. This workshop proposes research methods to carryout case studies and methods to understand the relationship between the objects and the participants. The implications for the knowledge to support this cultural / technical revolution are widespread – in particular of industry, but also leisure and the arts. We hope that storytellers, academics who are interested in social computing, and programmers will be interested in the workshop.

Korsakow is an open source program that can be used to create nonlinear films following a structure whose foundation is heavily keyword-oriented. An intuitive interface design and a compelling story are key in creating interactive digital stories. We believe that Korsakow offers opportunities for the creation of intuitive interfaces. We also believe that the type of treatment of discrete (linear) stories offers opportunities for compelling interactive narratives.

With this in mind, we designed a workshop in which the participants create an interactive story, based on discrete movies that they will shoot on site and on keyword-based systems that they will design in creative ways. More than simply offering a context in which to explore the tool, we wanted to draw on the knowledge capital offered by the participants concerning other tools in this area.

In the IRIS network of excellence [1], one work package (”Authoring Tools and Creation Methods”) has been tackling accessibility issues of highly-interactive and generative Interactive Storytelling (IS) systems for story creators. While one problem is a current lack of well-designed tools, which needs improvement, the other side of the medal is that nevertheless, prospective interactive storyworld creators need basic knowledge in the currently available approaches. With the vision in mind that in the future, complex IS projects will be accomplished by an interdisciplinary team, future authors do not need to program a story engine, but they would indeed need to know about the characteristics of content conceptualization, knowledge modeling and forms of abstraction, which then are used in a collaborative production process with engineers.