My Research

Interactive Storytelling via Player Modelling


Motivation

Is it possible to tell stories through games? Certainly. Like the carefully written script of a movie, one event unfolds after another, and another, gradually leading the player toward a well-planned conclusion. Unfortunately, in most digital games, the player's path through the story is often quite strict; there may be many opportunities to wander around, explore the game's environment, and experience an array of side-plots, but ultimately nothing important happens until the player performs an action that triggers the next step of the predetermined plot. Is this such a bad thing? For one, it breaks the player's sense of immersion in the game; as soon as she realizes that the world full of diversions has little to no effect on what's actually going to happen, she might wonder why the diversions are there at all (think of reading a novel whose author suggests skipping several chapters, since they have nothing to do with the story being told). Secondly, in story-based games with predetermined plots, there is little motivation to play the game again; the excitement and uncertainty felt during the first-time reading of a novel is lost in later readings.

Inspiration

Interactive Storytelling suggests an answer to the problems given above, by asking game-story authors to relax. Instead of specifying exactly what will happen in the game world, their task becomes the specification of everything that can happen in the world, and the game engine then builds a story by connecting the pieces together. This process is well known to the players of Pen & Paper Role-Playing Games, where a group of players' experiences in a fantasy world are shaped by one or more "Game Masters" - players who choose and describe what transpires in the world in response to the players' actions. Game Masters have a working idea of the world in which the game is set; they know what can conceivably happen, and are able to choose sequences of events that lead the players through a fantasy story of their design. Supposing that a set of authors had been convinced to specify the world of things that can happen, how would the game engine choose which things should happen? In answering this question, the techniques used by Game Masters inspire my approach; they watch, and they learn. By observing their players' reactions to the events that occur throughout the story, Game Masters can gain a sense of what things the players might like to have happen, or what situations they might enjoy being in. Given this knowledge, Game Masters are able to tailor the story being told to the specific preferences of the people who are playing, which hopefully improves the experience for all players involved.

My Goal

The goal of my research is to design and implement a system capable of choosing which events should come next in an interactive story, by mimicking the way in which the Game Masters of Pen & Paper Role-Playing Games learn and exploit their players' preferences. I intend to focus my work on the domain of Computer Role-Playing Games, and am currently working on a prototype system using BioWare's Aurora Neverwinter Toolset, the tool used to create Neverwinter Nights, a multi-award-winning computer role-playing game.

Interested?

Visit the PaSSAGE website!

Curriculum Vitae



Education: Ph. D., Computing Science, University of Alberta (current)
M. Sc., Computing Science, University of Alberta, 2007
B. Sc. Hons, Computer Science, University of Regina, 2005
Supervisor: Vadim Bulitko
Research Group: Intelligent Reasoning, Critiquing, and Learning (IRCL)
Research Lab: Artificial Intelligence Research
Students Supervised: Trevon Romanuik (Summer 2010)
Michael Webb (NSERC USRA, Summer 2008, Summer 2009)
Phillip Silver (High School Internship Program, Summer 2007)
Trevor Bekolay (NSERC USRA, Summer 2007)
Eric Wasylishen (NSERC USRA, Summer 2006)
Full Curriculum Vitae: Updated February 2011 (pdf)

Publications


The following is a list of the work that I've been fortunate enough to have published.

David Thue, Vadim Bulitko, Marcia Spetch, Trevon Romanuik. A Computational Model of Perceived Agency in Video Games. The Seventh Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE): pp. 91-96. AAAI Press. Palo Alto, California, USA. October 12, 2011. [pdf]
Mark Riedl, David Thue, Vadim Bulitko. Game AI as Storytelling. In book: Artificial Intelligence for Computer Games. pp. 125-150. Springer USA. New York, New York, USA. March 2011. [pdf] [url]
David Thue, Vadim Bulitko, Marcia Spetch, Trevon Romanuik. Player Agency and the Relevance of Decisions. The Third Joint International Conference on Interactive Digital Storytelling (ICIDS): pp. 210-215. Springer-Verlag. Edinburgh, Scotland. November 1, 2010. [pdf]
David Thue, Vadim Bulitko, Marcia Spetch, Michael Webb. Socially Consistent Role Passing in Player-Specific Stories. The Sixth Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE): pp. 198-203. AAAI Press. Palo Alto, California, USA. October 10, 2010. [pdf]
David Thue, Vadim Bulitko, Marcia Spetch, Michael Webb. Exaggerated Claims for Interactive Stories. The Second Joint International Conference on Interactive Digital Storytelling (ICIDS): pp. 179-184. Springer-Verlag. Guimaraes, Portugal. December 9, 2009. [pdf]
David Thue, Vadim Bulitko, Marcia Spetch. Making Stories Player-Specific: Delayed Authoring in Interactive Storytelling. The First Joint International Conference on Interactive Digital Storytelling (ICIDS): pp. 230-241. Springer-Verlag. Erfurt, Germany. November 26, 2008. [pdf]
David Thue, Vadim Bulitko, Marcia Spetch. PaSSAGE: A Demonstration of Player Modelling in Interactive Storytelling. The Fourth Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE): pp. 227-228. AAAI Press. Palo Alto, California, USA. October 22, 2008. [pdf]
David Thue, Vadim Bulitko, Marcia Spetch. Simulating the Adaptive Behaviour of Storytellers in Computer Video Games. The Tenth International Conference on the Simulation of Adaptive Behavior (SAB), Last Minute Results track. pp. N/A. Poster via unpublished abstract. Osaka, Japan. July 7, 2008. [pdf]
David Thue, Vadim Bulitko, Marcia Spetch. Player Modeling for Interactive Storytelling: A Practical Approach. AI Game Programming Wisdom 4: pp. 633-646. Charles River Media. Boston, Massachusetts, USA. February 2008. [url]
David Thue, Vadim Bulitko, Marcia Spetch, Eric Wasylishen. Learning Player Preferences to Inform Delayed Authoring. Papers from the AAAI Fall Symposium on Intelligent Narrative Technologies. FS-07-05: pp. 158-161. AAAI Press. Arlington, Virginia, USA. November 9, 2007. [pdf]
David Thue. Player-informed Interactive Storytelling (M.Sc. Thesis). Department of Computing Science. 130 pages. University of Alberta. Edmonton, AB. October 3, 2007. [pdf]
David Thue, Vadim Bulitko, Marcia Spetch, and Eric Wasylishen. Interactive Storytelling: A Player Modelling Approach. The Third Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE). pp. 43-48. Stanford, California, USA. June 6, 2007. [pdf]
David Thue and Vadim Bulitko. Modelling Goal-directed Players in Digital Games. The Second Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE). pp. 86-91. Marina del Rey, California, USA. June 23, 2006. [pdf]