Unsupervised profiling of meditation-induced autonomic responses using electrodermal and heart rate variability features
Background and Motivation: Meditation practices influence the autonomic nervous system (ANS), as reflected in electrodermal activity (EDA) and heart rate variability (HRV), though individual responses vary. This study aimed to explore whether unsupervised clustering can uncover distinct physiologica...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Results in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590123025025502 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Background and Motivation: Meditation practices influence the autonomic nervous system (ANS), as reflected in electrodermal activity (EDA) and heart rate variability (HRV), though individual responses vary. This study aimed to explore whether unsupervised clustering can uncover distinct physiological patterns during guided meditation, aiding personalized assessment and biofeedback development. Materials and Methods: EDA and HRV were recorded from 14 healthy participants (8 men, 6 women) during guided meditation. Signals were preprocessed via filtering, normalization, and segmentation into 3-minute windows. Extracted features included tonic and phasic EDA components, skin conductance responses (SCRs), and HRV metrics (e.g., RMSSD, SDNN). Fuzzy C-means clustering was applied to identify physiological response subgroups. The optimal number of clusters was determined using the Davies–Bouldin index and silhouette scores. Results: Unsupervised FCM clustering was used to explore physiological responses to meditation. Although clustering yielded four participant-level and five measurement-level groups, post hoc interpretation revealed three overarching response profiles—arousal, balance, and relaxation—based on common feature trends. These interpreted profiles synthesize the clustering results into more accessible representations of individual autonomic variability. Discussion: The clusters reflect heterogeneous ANS responses to meditation, suggesting it does not induce a uniform physiological state. Unsupervised learning offers an objective approach to profiling individual responses, supporting personalized meditation and biofeedback. Conclusion: Fuzzy clustering revealed distinct autonomic patterns during meditation, demonstrating potential for personalized neurotechnology and informing tailored mindfulness interventions based on physiological feedback. |
|---|---|
| ISSN: | 2590-1230 |