Facilitating Individuals' Sensemaking about Sedentary Behavior via Contextualized Data

2025-09-23

AI Summary

Prolonged sedentary behavior significantly increases chronic disease risk, yet users often struggle to identify its patterns and contextual drivers from raw sensor data. To address this, we propose MotionShift—a novel system that, for the first time, deeply integrates multi-dimensional contextual information (e.g., location, weather, calendar events) with step-count data to enable situated feedback and visual analytics for sedentary behavior. Leveraging mobile sensing, context-aware computing, and user self-reporting, MotionShift supports reflective sensemaking and mitigates interpretation bias. A user study demonstrates that MotionShift significantly enhances users’ contextual awareness of sedentary episodes, improves behavioral classification accuracy, and effectively fosters data-informed motivation and behavior change. This work advances context-aware health information systems by introducing a principled interaction design and behavior intervention framework grounded in situated cognition and participatory sensing.

Technology Category

• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services

Abstract

The sedentary lifestyle increases individuals' risks of developing chronic diseases. To support individuals to be more physically active, we propose a mobile system, MotionShift, that presents users with step count data alongside contextual information (e.g., location, weather, calendar events, etc.) and self-reported records. By implementing and deploying this system, we aim to understand how contextual information impacts individuals' sense-making on sensor-captured data and how individuals leverage contextualized data to identify and reduce sedentary activities. The findings will advance the design of context-aware personal informatics systems, empowering users to derive actionable insights from sensor data while minimizing interpretation biases, ultimately promoting opportunities to be more physically active.

Problem

• Understanding how contextual information impacts sense-making of sensor data
• Identifying and reducing sedentary activities using contextualized data
• Advancing context-aware personal informatics systems design

Innovation

• Mobile system presenting step data with context
• Contextual information impacts sedentary behavior sensemaking
• Context-aware design minimizes interpretation biases