Live Blog of Symposium presentations
Information about the registration and accomodation for the symposium
Information about a few scholarship opportunities for the symposium
More Information about the Improvised Comedy Games Workshop to be held at the symposium
The submission page is open!
An updated Call for Ppaers is out.
The Call for Special Session on work falling outside classical AI has been added.
The Call for Papers is out.
Narrative is a pervasive aspect of all human societies. Human beings make sense of the world by constructing stories and listening to the stories of others. In addition, stories as a form of entertainment play a central role in our social and leisure lives. As a result, story and narrative have become a key interest for Artificial Intelligence researchers. The 1970's and early 80's saw a substantial amount of work in story understanding and generation springing up in the AI community. For example, Roger Schank's research group looked at the kinds of knowledge structures and processes used by humans to understand natural language. The work at that time included systems for story understanding (Schank, 1981) and story generation (Wilensky, 1981) using scripts, plans and goals as the underlying knowledge structures (Meehan, 1977). The next generation of work in the 90's included other creative endeavors such as modeling daydreaming (Mueller 1990) and work in storytelling (Turner 1992). In recent years, artificial intelligence research applied to narrative has focused on narrative generation, targeting the development of interactive narrative systems where the user is situated in a virtual world and is an active participant in an ongoing narrative.
The role of narrative as a primary mechanism for organizing human experience has been recognized in many different fields. As a result, work in narrative has become increasingly multidisciplinary with influences from many fields including art, psychology, cultural and literary studies, as well as drama. The diverse perspectives obtained from the different fields of study have produced exciting results in areas including: 1) interfaces that use narrative as a system design principle; 2) systems for understanding that model the processes by which a human understands a narrative; 3) systems for generation that seek to model the knowledge and processes necessary to produce a narrative; 4) systems for interaction that aim to develop experiences including the audience as an active participant in an ongoing narrative; and 5) embodied agent design that use narrative to structure an agent's behavior.
More recently, there has been an increasing level of interest at the intersection of narrative and games. Historically, the relationship between narrative and games has been problematic giving rise to intense debate (Frasca, 1999). Interactivity in games allows for a more subjective experience than is allowed for by non-interactive games. At the same time, the interactive nature of games can cause a reduction in the degree of authorial control, thereby providing more possibilities for the game player and bestowing upon her some of the author's responsibilities to create a narrative experience. Granting the player an active role in a narrative, however, creates the potential for tension between the author's narrative goals and player's freedom to interact in pursuit of their own goals. Artificial Intelligence systems built upon the techniques of player modeling, story generation, and story coordination and management have shown considerable promise. With current state of the art techniques, these systems have the intelligence to predict player behavior, adapt to it, and even generate narrative content in response to player choices. Despite the significant advances relative to the early work in the 70's, there are many challenging open problems facing the research community today.
The 1999 AAAI Fall Symposium on Narrative Intelligence was a successful multidisciplinary gathering of researchers emphasizing both interactive as well as non-interactive forms of narrative intelligence. The symposium focused on various topics including narrative theory, autonomous performance agents, narrative interfaces, narrative generation, narrative understanding, and interactive story. The conference on AI and Interactive Digital Entertainment (AIIDE), that began in 2005, has a focus on intelligence applied to interactive narrative systems, but excludes the non-interactive narrative AI research included in the 1999 symposium. Furthermore, a sizable portion of the work presented at AIIDE since its inception has been from the broader field of game AI in general, with an emphasis on intelligent adversaries in real-time strategy games, path planning, and animation. The emphasis on interactivity at AIIDE is one aspect of the larger body of research investigating narrative and intelligence. Unfortunately, there is no recurring publication or meeting venue that is openly receptive to the complete spectrum of narrative and AI research. A few small-scale conferences in Europe such as the International Conference on Technologies for Interactive Digital Storytelling and Entertainment (TIDSE) and the International Conference on Virtual Storytelling (ICVS) are receptive to AI narrative research, although these conferences are not held for the artificial intelligence community and thus are not ideal places to disseminate narrative AI research. AI researchers are also known to submit papers about narrative to conferences and journals on education, literary studies, linguistics, cognitive science, and computer graphics; however, this research tends to be regarded as fringe or peripherally relevant to these venues.
Since the 1999 symposium, there has been only one venue receptive to the breadth of AI and narrative research that our proposed symposium will target. This venue was the AAAI'07 Symposium on Intelligent Narrative Technologies. Serving as a forum where the full range of narrative and AI research could be discussed, it brought together AI researchers of both interactive and non-interactive narrative technologies, cognitive psychologists, and narrative theorists with the goal of discussing the fundamental issues in representing, presenting, adapting, and reasoning about narrative. The discussions and presentations at the symposium resulted in many interesting questions, several of which remained unanswered at the time of the symposium's end. At the time of our proposed symposium would be held, a year and a half will have passed, presenting an ideal opportunity for the research community to come together once more to discuss our progress in addressing those questions, and to articulate a new set of questions that have arisen since then. Our symposium will bring together many of the researchers that attended the 2007 symposium, as well as new researchers from many of the disciplines we have described. The goal of our symposium will be to discuss innovations and work that has been conducted in the time since the 2007 paper deadline, and to further focus on how computer systems can be designed to reason about, perform, and adapt narrative structures for interactive and non-interactive technologies. In addition, we plan to revisit in detail the two most hotly discussed issues of the 2007 symposium: natural language understanding and generation, and authoring paradigms and tools. The output of the proposed symposium will include paper proceedings, a write-up of the event for AI Magazine, and a possible publication of select papers for wider dissemination, if appropriate.
References: