Behavior Latticing: Inferring User Motivations from Unstructured Interactions

2026-04-08Human-Computer Interaction

Human-Computer Interaction
AI summary

The authors propose a new way for AI to better understand people by looking beyond just what they do to why they do it. They introduce a method called behavior latticing that connects different actions over time to uncover the user’s deeper motivations and needs. Their approach helps AI systems provide more meaningful support, not just repeating tasks but addressing real user goals. They tested this and found their system gives more insightful user understanding and improves how well AI meets user needs.

personal AIbehavior latticinguser motivationsuser needsbehavior synthesisinteraction dataAI user modelinguser insightshuman-computer interaction
Authors
Dora Zhao, Michelle S. Lam, Diyi Yang, Michael S. Bernstein
Abstract
A long-standing vision of computing is the personal AI system: one that understands us well enough to address our underlying needs. Today's AI focuses on what users do, ignoring why they might be doing such things in the first place. As a result, AI systems default to optimizing or repeating existing behaviors (e.g., user has ChatGPT complete their homework) even when they run counter to users' needs (e.g., gaining subject expertise). Instead we require systems that can make connections across observations, synthesizing them into insights about the motivations underlying these behaviors (e.g., user's ongoing commitments make it difficult to prioritize learning despite expressed desire to do so). We introduce an architecture for building user understanding through behavior latticing, connecting seemingly disparate behaviors, synthesizing them into insights, and repeating this process over long spans of interaction data. Doing so affords new capabilities, including being able to infer users' needs rather than just their tasks and connecting subtle patterns to produce conclusions that users themselves may not have previously realized. In an evaluation, we validate that behavior latticing produces accurate insights about the user with significantly greater interpretive depth compared to state-of-the-art approaches. To demonstrate the new interactive capabilities that behavior lattices afford, we instantiate a personal AI agent steered by user insights, finding that our agent is significantly better at addressing users' needs while still providing immediate utility.