Below are some projects that I am interested in pursuing (pending funding availability and student interest). This list is not complete, but it's a good start. Each of the project areas below have links to papers that touch on different aspects of the problems I'm describing; consider these to be good conversation starters.

If any of these sound exciting to you, consider joining the lab!

Expanding the Expressive Range of Automated Storytelling

Automated storytelling systems have grown in complexity over time, yet there is still a lot they cannot express in narrative-theoretic terms. Concretely, while we have systems that can generate stories with believable characters (those that act with intent) and can support generating stories where dramatic conflict exists, we still cannot generate a variety of situations common to a variety of plot effects (e.g. we're only very recently beginning to reason about modeling stories where characters can be mistaken about the narrative state of the world).

Working in this area will involve identifying some design problem context for which automated storytelling systems would be great for, and figuring out what state-of-the-art narrative generation technologies can be brought to bear to help solve the problem. More than likely, we will be looking at what automated storytelling systems can currently do and critically evaluating them to identify what they (rhetorically, aesthetically, or communicatively) cannot. Then it's up to us to figure out how we can get them to generate stories they couldn't before, in order to solve the problem they couldn't solve before. In other words, we're looking to expand their expressive range in a way that affords solving a concrete design problem. Importantly, we will be looking to flip the dominant storytelling paradigm of simulation-then-presentation in favor of presentation-then-simulation approaches. We will also be looking at new storytelling affordances made available through emerging technologies (such as virtual reality).

Mixed-Initiative Game Design

Games are a kind of complex system which feature a large number of interacting components (players, rules, and so forth). It should come as no surprise that designing games (both digital and analog) is a non-trivial task. What computational tools can be brought to bear to designing games? Learning game mechanics? Exploring the logical ramifications of rule sets? What if, in addition to developing well-formed games, we're interested in games that achieve some communicative purpose? How can we create "the digital camera of games"?

Working in this area will involve studying the process of game design as practiced by local game designers, in an attempt to identify concrete points of failure during the design process (e.g. bottlenecks in development, cost blowup in testing) and developing intelligent tools that designers would actually be willing to use to alleviate some of the authorial burden involved in the craft.

Modeling the Determinants of Player Choice

Sid Meier once famously quipped "a game is a series of meaningful choices." Today, choice structures are arguably one of the most defining characteristics of games since they represent a form of activity that effects some change in the state of the game universe (whether it be an actual change in the logic of the game or a change the player thinks actually happened).

Despite their importance, not many folks have looked at understanding what influences players when they make choices. Working in this area will involve building off prior work that I have carried out alongside colleagues to understand what we label the "determinants of player choice" in the context of interactive narratives and games.

Automated Dungeon Mastering

Often touted as a "grand challenge" of interactive storytelling, this involves creating an AI agent capable of running a tabletop role-playing game (i.e. serving as a "dungeon master"). Simple right?

On the surface of it, the AI agent can automate a lot of the necessary minutiae that happens during a tabletop session. For example, it can calculate perception checks, attack rolls, movement for characters, and so on. But an AI dungeon master (AI-DM) must also determine several things that are less obvious: it needs to perform dynamic difficulty adjustment, it needs to improvise new story situations based on the actions of characters, and it needs to treat gameplay as a form of conversation. Working in this area will involve combining technologies around natural language processing (e.g. Amazon's Lex) with technologies around story generation (e.g. automated planning), and understanding how to navigate tradeoffs between authorial control and player agency.

Modeling the Narrative Mind as it Relates to Sense-making

The narratologist David Herman argues in his book Storytelling and the Sciences of Mind that there are two ways with which we reason about narrative structure.

The first is the more intuitive engagement with narrative: narrative as a target of interpretation, where the narrative itself is reified and exists as media to consume (for example, when you read a book or watch a film). The second is the less intuitive, but equally prevalent, engagement with narrative: narrative as a framework for sense-making, where the narrative is constructed by "narrativising" actions that observed in some task environment (and in daily life). While a significant amount of computational work exists around the former, the latter has been less explored. Working in this area will involve looking at prior work that has attempted to narrativize task environments as well as thinking about how we might improve upon it in an environment that is wholly unique to the University of Utah: the Entertainment Arts and Engineering eSports program.