Graphing the Everyday: A Neurosymbolic Approach to Eliciting Routines for Just-In-Time Adaptive Interventions

2026-08-10

AI Summary

This work addresses two key challenges in transforming users’ unstructured daily routines into structured schedules suitable for just-in-time adaptive interventions (JITAIs): the neglect of dynamic fluctuations in psychological states and energy levels, and the cognitive-ecological mismatch between linear large language model (LLM) extraction and humans’ nonlinear narrative styles. To bridge this gap, the authors propose a neuro-symbolic system integrating LLMs with a Neo4j knowledge graph, leveraging natural language replay for human-centered evaluation. The framework introduces design heuristics—routine bridging, adaptive negotiation, and scalable transparency—that effectively mitigate entity fragmentation and ecological misalignment. This approach advances schedule tracking beyond rigid logging toward an empathetic, context-aware proactive intervention agent, offering nuanced, context-sensitive support for sustained health behavior change.

Abstract

Just-In-Time Adaptive Interventions (JITAIs) increasingly rely on conversational agents to elicit user routines, yet translating fluid human dialogue into rigid schedule data remains a significant challenge. We conducted a qualitative investigation of a neurosymbolic pipeline, combining Large Language Models (LLMs) with a Neo4j knowledge graph, to map unstructured verbal narratives into actionable interventions. Through human-centric evaluation using natural-language playbacks, we identified a critical "mental-model gap," where the linear extraction of LLMs clashes with hierarchical, non-linear human storytelling, causing severe entity fragmentation. Furthermore, we articulate an "ecological mismatch," demonstrating that algorithmic schedule availability frequently ignores the user's fluctuating psychological receptivity and physical energy levels. To resolve these tensions, we propose actionable design heuristics, including routine piggybacking, adaptive negotiation, and scalable transparency. Ultimately, these guidelines provide a foundational framework for evolving rigid schedule-trackers into empathetic, context-aware proactive agents capable of supporting long-term health behavior change.

Problem

• Just-In-Time Adaptive Interventions
• routine elicitation
• mental-model gap
• ecological mismatch
• conversational agents

Innovation

• neurosymbolic
• just-in-time adaptive interventions
• knowledge graph
• mental-model gap
• ecological mismatch