WatchAnxiety: A Transfer Learning Approach for State Anxiety Prediction from Smartwatch Data

2025-09-17

AI Summary

This study addresses the challenge of real-time state anxiety prediction for individuals with social anxiety using wearable physiological signals (e.g., heart rate) to enable personalized, just-in-time interventions. We propose a dynamic prediction framework integrating transfer learning and meta-learning: a base model is first pre-trained on large-scale external heart rate data, then fine-tuned using ecological momentary assessment (EMA) labels and individual trait anxiety scores; probabilistic modeling is incorporated to improve uncertainty quantification. Our approach is the first to apply transfer learning to cross-subject, real-time state anxiety prediction and introduces trait-state joint modeling to enhance personalization. Evaluated on data from 91 participants, the method achieves a balanced accuracy of 60.4%; on the independent TILES-18 dataset, it attains 59.1% accuracy—outperforming prior state-of-the-art by ≥7%—demonstrating strong generalizability and clinical applicability.

Technology Category

• Intelligent Robots: State Estimation
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Large-scale security measurements

Abstract

Social anxiety is a common mental health condition linked to significant challenges in academic, social, and occupational functioning. A core feature is elevated momentary (state) anxiety in social situations, yet little prior work has measured or predicted fluctuations in this anxiety throughout the day. Capturing these intra-day dynamics is critical for designing real-time, personalized interventions such as Just-In-Time Adaptive Interventions (JITAIs). To address this gap, we conducted a study with socially anxious college students (N=91; 72 after exclusions) using our custom smartwatch-based system over an average of 9.03 days (SD = 2.95). Participants received seven ecological momentary assessments (EMAs) per day to report state anxiety. We developed a base model on over 10,000 days of external heart rate data, transferred its representations to our dataset, and fine-tuned it to generate probabilistic predictions. These were combined with trait-level measures in a meta-learner. Our pipeline achieved 60.4% balanced accuracy in state anxiety detection in our dataset. To evaluate generalizability, we applied the training approach to a separate hold-out set from the TILES-18 dataset-the same dataset used for pretraining. On 10,095 once-daily EMAs, our method achieved 59.1% balanced accuracy, outperforming prior work by at least 7%.

Problem

• Predicting momentary anxiety from smartwatch data
• Measuring intra-day anxiety fluctuations in social situations
• Developing real-time personalized mental health interventions

Innovation

• Transfer learning from external heart rate data
• Fine-tuned probabilistic predictions with meta-learner
• Smartwatch-based system for anxiety detection