Experiments in Paired Storytelling
What makes an interactive story feel well-crafted? What makes a player feel like their participation in the story is meaningful? And in interactive narratives played between humans, such as tabletop role-playing games, how do the participants reason about each other to help support both of these criteria?
To help explore these questions, we are creating a dataset from an interactive narrative game played online between two users: the player, who controls the protagonist, and the game master, who controls all of the other characters. In addition to recording the users' gameplay choices, we ask the users to explain why they made those choices, to speculate on the reasons for their partner's choices, and to rate their experience of story structure and agency throughout the game. By studying how people think about their interactions in the game, we hope to design artificially intelligent agents that understand and are understandable by players with the same flexibility as a human partner. This page will be home to our data as it becomes available, as well as opportunities for the public to participate.
Data Sets
Publications
- Mutual Implicit Question Answering for shared authorship: a pilot study on player expectations. In Proceedings of the 10th Intelligent Narrative Technologies workshop at the 13th AAAI international conference on Artificial Intelligence and Interactive Digital Entertainment, pp. 259-265, 2017. .
- Structure, agency, and intent: preliminary data collection. In Proceedings of the Interactive Narrative Technologies workshop at the 20th AAAI international conference on Artificial Intelligence and Interactive Digital Entertainment, 2024. .