3rd Workshop on Intelligent Narrative Technologies

at the Foundations of Digital Games Conference, Monterey, CA, U.S.A., June 18, 2009.


Proceedings now available through the ACM Digital Library

INT3 Presentation Schedule

Session 1: Improv
8:30 - 9:00
Riedl (Paper #16)
9:00 - 9:30
Magerko (Paper #8)
9:30 - 10:00
Magerko (Paper #15)

10:00 - 10:30  BREAK
Session 2: Interactive Storytelling
10:30 - 11:00
McCoy (Paper #10)
11:00 - 11:30
Lee (Paper #13)
11:30 - 11:45
Sullivan (Paper #11)
11:45 - 12:00
Rowe (Paper #14)

12:00 - 1:30 LUNCH
Session 3: Story Authoring
1:30 - 2:00
Tanenbaum (Paper #6)
2:00 - 2:30
Orkin (Paper #3)
2:30 - 2:45
Tearse (Paper #12)

3:00 - 3:30 BREAK
Session 4: Narrative Discourse
3:30 - 4:00
Tomai (Paper #9)
4:00 - 4:30
Niehaus (Paper #7)
4:30 - 4:45
Hills (Paper #1)

5:00 - 5:15 BREAK
Session 4: Applications
5:15 - 5:30
Chen (Paper #17)
5:30 - 5:45
Madden (Paper #5)

INT3 List of Accepted Papers

1: Damian Hills. A Conversational Framework for Emergent Collaborative Storytelling  

Abstract. This paper proposes a framework that incorporates shared narration through the design and evaluation of a creativity support system that promotes meaning and sense-making for collaborative storytelling in a situated context. The approach is to develop a holistic system that enables the processes during a group
conversation, or the situated modalities associated with oral storytelling.
2: Inderjeet Mani. Predicting Reader Response in Narrative  

Abstract. This paper sketches a theory of how readers (or users, or players, or viewers) respond to narratives. Such a theory can be useful for developing evaluation functions to allow for narrative outcomes that have maximum impact at particular times on the reader. The reader’s response is viewed in terms of character evaluations, namely judgments of sympathy or antipathy for the agent involved in that outcome. Character development is computed in terms of transitions in reader evaluations for an agent over the time course of the narrative. To formally model character evaluations, we begin with a representation of the narrative fabula in terms of the events, their participant roles, and their temporal relations. This representation is implemented on a corpus of narratives with existing tools and standards. Reader evaluations are annotated on events in the fabula. Once high reliability in human character evaluations has been proven, a character evaluation tagger will be trained on these evaluations.
3: Jeff Orkin, Tynan Smith, Hilke Reckman and Deb Roy. Semi-Automatic Task Recognition for Interactive Narratives with EAT & RUN  

Abstract. Mining data from online games provides a potential alternative to programming behavior and dialogue for characters in interactive narratives by hand. Human annotation of course-grained tasks can provide explanations that make the data more useful to an AI system, however human labor is expensive. We describe a semi-automatic methodology for recognizing tasks in gameplay traces, including an annotation tool for non-experts, and a runtime algorithm. Our results show that this methodology works well with a large corpus from one game, and suggests the possibility of refactoring the development process for interactive narratives.
4: Isabel Machado Alexandre, David Jardim and Pedro Faria Lopes. Maths4Kids – Telling Stories with Maths 

Abstract. In this paper, we describe a novel approach to teaching early mathematical concepts to young children. This approach aims to merge storytelling and Maths. Evidence show that through dramatic games and role-playing activities young children (aging from 5-7 years old) learn to master new knowledge, to fit in a new school setting and to socially relate with their peers. So taking this evidence into account, it is possible to devise an innovative collaborative learning scenario that teaches early mathematical concepts by telling and creating stories. More than doing it in the traditional formats (oral or theatrical form), we are investigating the possibility to take this scenario to an innovative computerised platform – Microsoft SurfaceTM.
5: Neil Madden and Brian Logan. User evaluation of virtual reporting agents   

Abstract. Many persistent online environments such as Massively-Multiplayer Online
Games (MMOGs), feature weblogs and/or live reportage of participants' activities in the world. While such reports and commentary can enhance the user's enjoyment
and increase their sense of shared experience, the demands of such large scale reporting on the participants can be considerable. To address this problem a number of in-game reporting and commentary systems have been proposed which use
virtual "reporter'' agents within the game to produce real-time and post-game
commentary  tailored to the interests of individual users. However, to date, there has been no evaluation of these systems from a user perspective. In this  paper we present the results of a live evaluation study performed using an  instance of the online role-playing game Neverwinter Nights augmented  with witness-narrator agents to provide in-game and post-game reports. Our results indicate that reporting does increase enjoyment of the game, and that players play for longer when their activities are recorded on a community web page, suggesting that agent-based reporting is a promising approach to community building in online games and social environments.
6: Joshua Tanenbaum, Karen Tanenbaum and Magy Seif El-Nasr. Authoring Tangible Interactive Narratives  Using Conceptual Hyperlinks 

Abstract. Creating content for different forms of interactive narratives requires a different set of skills and techniques than writing non-interactive stories.  In this paper we describe a prototype tangible interactive narrative system called the Reading Glove and outline the authoring process used to create story content for it.  We begin by discussing different approaches to authoring that are currently in use in interactive digital storytelling.  We then discuss in detail the process of writing the story for the Reading Glove and provide an analysis of the fiction created through this process. From this analysis we put forth the notion of “conceptual hyperlinks”: a design technique that we believe has utility for the authors of future interactive narratives. We conclude with a set of general design recommendations for authoring within interactive storytelling systems.
7: James Niehaus and R. Michael Young. A Method for Generating Narrative Discourse to Prompt Inferences  

Abstract. Narratives that prompt inferences can be more interesting in that they provide the reader with the   opportunitity to reason about the narrative world, participating in its construction. These   narratives can also be more concise and direct, as details can be filled in by the reader. On the   other hand, narratives that leave out important information without the opportunity to infer this   information may be incoherent. To generate narratives that prompt inferences a system must 1)   employ a theory of how inferences are prompted and 2) provide a capacity for creating narratives   that satisfy inference goals. This paper presents is a novel algorithm for generating discourse   plans that prompt inferences according to a theory of online inferencing in narrative discourse.   Though other approaches have generated narrative and discourse structures to influence the   reader's perception of the narrative , this is the first approach to present an empirically based   cognitive model of online inference generation. The algorithm employs a partial-order planning   approach to discourse generation, selecting events to tell the reader from an input story plan.
8: Brian Magerko, Casey Fiesler, Allan Baumer and Daniel Fuller. Bottoms Up: Improvisational Micro-Agents 

Abstract. This paper describes our current approach in implementing computational improvisational micro-agents.  This approach is intended to foster bottom-up research to better understand how to build more complex agent behaviors in a theatrical improvisational setting.  Micro-agent designs are based on our current findings in the Digital Improv Project, a multi-year study focused on studying real life theatrical improvisers with an aim towards better understanding the cognition employed in improvisation at the individual and group level.
9: Emmett Tomai and Ken Forbus. Using Narrative Functions as a Heuristic for Relevance in Story Understanding  

Abstract. Story understanding requires a degree of knowledge and expressiveness beyond the current state of natural language understanding.  We present an approach that addresses these needs, using a large-scale knowledge base, simplified English grammar and a combination of compositional frame semantics and abductive reasoning.  This in turn raises a significant challenge disambiguating complex semantic structures, which requires a pragmatics of narrative for constraint and guidance.  We present a theory of narrative functions that serve as a heuristic for relevance in narrative, and provide evidence that this heuristic is effective for disambiguation that leads to consistent understanding.
10: Josh McCoy, Mike Treanor, Ben Samuel, Brandon Tearse, Michael Mateas and Noah Wardrip-Fruin. Comme il Faut 2:  A fully realized model for socially-oriented gameplay   

Abstract. Social games—common patterns of character interactions that modify the social environment of the story world—provide a useful abstraction when authoring a story composed of interactive characters, making it possible to create games with deep possibility spaces that are about social interaction (which would be intractable if hand-authoring all the options). In this paper, we detail the workings of a major new version of our social artificial intelligence system, Comme il Faut, that enables social game play in interactive media experiences. The workings of Comme il Faut 2 are shown, with running examples, from both knowledge representation and process perspectives. Finally, the paper concludes with a plan for evaluating and demonstrating Comme il Faut 2 through an implementation of an interactive media experience that consists of a playable social space.
11: Anne Sullivan, Michael Mateas and Noah Wardrip-Fruin. Rules of Engagement: Moving Beyond Combat-Based Quests  

Abstract. Computer role-playing games (CRPGs) are known for their strong narrative structure. Over time, quests have become one of the main mechanics for leading a player through the story. Quests are given to the player in the form of a set of tasks to complete with few, if any, options. The options given to the player instead often revolve around combat-oriented actions – requiring the player to engage in combat to progress through the storyline, despite player preference or game story that hints otherwise. We address this issue with the GrailGM, a run-time game master which offers quests and actions to the player based on their history and current world state.
12: Brandon Tearse, Michael Mateas and Noah Wardrip-Fruin. Minstrel Remixed: A Rational Reconstruction   

Abstract. In this paper, we describe introduce Minstrel Remixed, a rational reconstruction of MINSTREL by Scott Turner.  In addition to recreating the landmark story generation system for public usage we also introduce a number of modifications that were made during the reconstruction that allow for investigation into the inner workings of the system.  Additionally we introduce Minstrel Remixed as a platform for use in Interactive Narrative applications and provide a number of concrete examples.
13: Seung Lee, Bradford Mott and James Lester. Investigating Director Agents’ Decision Making in Interactive Narrative: A Wizard-of-Oz Study  

Abstract. Interactive narrative planning offers significant potential for creating engaging narrative experiences that are tailored to individual users. Orchestrating all of the events in a storyworld to create optimal user experiences calls for effective narrative decision-making. A key requirement of this endeavor is understanding the role that different knowledge sources play in narrative decision making. To investigate knowledge sources for interactive narrative, a corpus was collected in a Wizard of Oz (WOZ) study conducted with a narrative-centered learning environment. With narrative planning and natural language dialogue functionalities provided by wizards, the data from the WOZ study offers insight into the knowledge sources involved in narrative decision making and suggests how these knowledge sources can be effectively utilized by a narrative planner to create engaging interactive narratives.
14: Jonathan Rowe, Lucy Shores, Bradford Mott and James Lester. A Framework for Narrative Adaptation in Interactive Story-Based Learning Environments  

Abstract. Over the past several years, the intelligent narrative technologies community has devised a broad range of interactive narrative systems. A key functionality provided by interactive narrative systems is narrative adaptation: augmenting story experiences in response to users’ actions, and tailoring story elements to individual users’ preferences and needs. However, there is little consensus about how to conceptualize interactive narrative adaptation across different systems or applications. This paper presents a framework that synthesizes multiple approaches to narrative adaptation, as well as experience management techniques from related fields such as intelligent tutoring systems. The framework has three components: plot adaptation, discourse adaptation, and user tailoring. Drawing on examples from the intelligent narrative technologies literature, the paper discusses connections to story-based learning systems. The framework is illustrated with examples from CRYSTAL ISLAND, a narrative-centered learning environment for middle school microbiology. The framework offers a first step toward establishing a general model of interactive narrative.
15: Brian Magerko. Shared Mental Models in Improvisational Performance   

Abstract. This paper describes the mental structures called shared mental
models, which are heavily related to group problem solving and cognition, and reports how they are related to theatrical improvisation based on our empirical findings.  We have conducted a series of studies on real life improvisers aimed at
uncovering the underlying cognition involved in improvisation, with the end goal of having a clear understanding of how to build improvisational synthetic characters.  We describe cognitive divergence, when improvisers have conflicting mental models of what is occurring on stage, and cognitive convergence, which is
the process of resolving such conflicts within the performance.  These findings are supported by examples from our study and are used to make conclusions about improvisational synthetic character design.
16. Mark Riedl. A Comparison of Interactive Narrative System Approaches Using Human Improvisational Actors

Interactive narrative is an approach to interactive entertainment or learning in which a system attempts to tell a story to an interactive participant. In this paper we report on a study to compare the theoretical strengths and weaknesses of two approaches to developing computational interactive narrative systems. We compare two approaches to interactive narrative: emergent approaches utilizing autonomous virtual character agents, and drama management approaches utilizing semi-autonomous virtual character agents. Our study uses improvisational theatre as an idealized, human analogue to computational interactive narrative. Results suggest that, regardless of approach, idealized interactive narrative systems should be nearly indistinguishable in terms of character believability and narrative coherence. Results suggest that drama management systems may have an advantage when particular features are required to emerge in players' interactive experiences.

17: Sherol Chen, Adam Smith, Arnav Jhala, Michael Mateas. RoleModel: Towards a Formal Model of Dramatic Roles for Story Generation

RoleModel is a novel story generator organized around explicit formal models of character roles. RoleModel expands the expressiveness of stories generated from arbitrarily partial domain specification by using a formal model of roles within an abductive logic programming framework. Authorial goals in the system can be fully or partially specified as constraints in an abductive logic program. In particular, the RoleModel system focuses on representing and satisfying role constraints of the story characters. This paper discusses the basic architecture for the RoleModel
approach, demonstrates example output from the system through three use-cases, discusses the authorial expressiveness enabled by a “stageless” abductive logic approach to story generation, and proposes the current and future directions.