Improving the Scene: Extending LOGTELL to Support a Plan-recognition / Plan-generation Paradigm

How to characterize a literary genre is a much debated problem, which can be approached with useful results by combining models drawn from both Literary Theory and Computer Science. Once a genre is specified with some rigour in a constructive way, it becomes possible not only to determine whether a given plot is a legitimate representative of the genre, but also to generate such plots, an ability of obvious relevance to Storytelling theory and practice. A key issue in interactive storytelling is how to generate stories which are, at the same time, interesting and coherent. On the one hand, it is desirable to provide means for the user to intervene in the story. But, on the other hand, it is necessary to guarantee that user intervention will not introduce events that violate the rules of the intended genre. A conceptual modelling method with the purpose of allowing an interactive narrative is presented, based on a plan recognition/plan generation paradigm. The method leads to the formulation of static, dynamic and behavioural schemas, expressed in temporal logic, and to a multi-stage interactive plot generation, supported by a logic-based tool for the interactive generation and dramatization of stories. This paper describes extensions made to the LOGTELL prototype to better support the genre model and incorporate a plan recognition / plan generation paradigm. The resulting system is of special help for the composition of plots through the adaptation of fully or partially generated plots. A case study, involving a simple Swords and Dragons genre (also extended), illustrates the discussion. Based on the model, the user can interact with the tool at various levels, obtaining a variety of stories agreeable to individual tastes, within the imposed coherence requirements. The system alternates stages of goal inference, planning, plan recognition, user intervention and 3D visualization.