Our research explores how to create stories with computer colleagues in open-ended, improvisational domains. Started in 2008, the Digital Improv Project examined the cognitive functions of improv actors through the process of protocol analysis and behavioral coding. The end goal of this process has been to establish approaches to computational creativity that can then be employed in co-creative experiences with AI characters. This work yielded several prototype experiences where humans and AI actors can play improv games together, such as Party Quirks and Three Line Scene.
The Computational Play project has explored how humans create stories together through pretend play as a way to understand both human creativity as well as potential computational approaches to pretend play intelligent agents.
The GENII / MImE project examines how to use participatory sensemaking as a means of learning story content through interaction. GENII (Graphical Embodied Narrative Intelligence & Improvisation) intelligent agents control virtual characters within a virtual world known as MImE (Movement Improvisation Environment). GENII agents can learn to improvise embodied narratives with people through observing their actions and inferring how they affect the virtual world in turn.
This material is based upon work supported by the National Science Foundation under Grant Numbers IIS #0757567, #320520, and IIS #1036457.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.