🤖 AI Summary
Social anxiety disorder (SAD) lacks well-characterized, stage-specific autonomic biomarkers, particularly in low- and middle-income countries (LMICs).
Method: We employed a single-lead wearable ECG device to record heart rate (HR) and heart rate variability (HRV) across four standardized psychological phases—baseline, anticipation, speech delivery, and reflection—within a validated behavioral anxiety paradigm. Time-frequency HRV metrics (RMSSD, LF/HF ratio) and real-time HR were analyzed.
Contribution/Results: As the first controlled experimental study of SAD physiology in LMICs, we identified a distinctive autonomic pattern: significant HRV suppression and HR elevation during anticipation and speech, partially reversing during reflection. This phase-dependent physiological signature achieved 82% accuracy in SAD classification. We publicly release a multimodal, phase-annotated physiological dataset, establishing foundational empirical and theoretical support for non-invasive, wearable-based digital phenotyping of anxiety disorders.
📝 Abstract
This paper investigates physiological markers of Social Anxiety Disorder (SAD) by examining the relationship between Electrocardiogram (ECG) measurements and speech, a known anxiety-inducing activity. Specifically, we analyze changes in heart rate variability (HRV) and heart rate (HR) during four distinct phases: baseline, anticipation, speech activity, and reflection. Our study, involving 51 participants (31 with SAD and 20 without), found that HRV decreased and HR increased during the anticipation and speech activity phases compared to baseline. In contrast, during the reflection phase, HRV increased and HR decreased. Additionally, participants with SAD exhibited lower HRV, higher HR, and reported greater self-perceived anxiety compared to those without SAD. These findings have implications for developing wearable technology to monitor SAD. We also provide our dataset, which captures anxiety across multiple stages, to support further research in this area.