Multiagent Narrative Experience Management as Story Graph Pruning

Abstract

In many intelligent interactive narratives, the player controls an avatar while an experience manager controls non-player characters (NPCs). The space of all stories can be viewed as a story graph, where nodes are states and edges are actions taken by the player, by NPCs, or by both jointly. In this paper, we cast experience management as a story graph pruning problem. We start with the full graph and prune intelligently until each NPC has at most one action in every state. Considering the entire graph allows us to foresee the long-term consequences of every pruning decision on the space of possible stories. By never pruning player actions, we ensure the experience manager can accommodate any choice. When used to control the story of an adventure game, players found our technique generally produced higher agency and more believable NPC behavior than a control.

SceneCraft: Automating Interactive Narrative Scene Generation in Digital Games with Large Language Models

“…By representing interactive narratives as story graphs (Riedl and Young 2006), with nodes representing world states and edges representing causal transitions, experience managers can actively track the narrative's progress in the graph toward the authorial goals. Narrative control is sometimes accomplished by pruning (Ware et al 2022) and other similar operations on the graph. We use dialogue graphs in our system to track branched interaction.…”

Planning Stories Neurally

“…Planning algorithms have been used to create many interactive narratives [8]- [11]. There has also been considerable research on augmenting planning algorithms with models of story structure [12]- [14]. Planningbased systems differ from reactive systems by performing an extensive backtracking search over the space of all possible stories.…”