A Network Approach to Analysing Heart-Rate Variability
Abstract
Heart rate variability (HRV) reflects autonomic nervous system functioning and is commonly examined using statistical methods that compare individual HRV indices across demographic groups. Such approaches, however, may overlook group differences in how HRV components relate to one another. This study applies a network approach to delineate gender differences in the conditional relationships among short-term HRV indices and demographic characteristics in a Singaporean adult sample (N = 2147). Conventional regression analyses identified gender differences in average heart rate and low-frequency/high-frequency (LF/HF) ratio, whereas differences in root mean square of successive differences (RMSSD) and standard deviation of normal-to-normal intervals (SDNN) were minimal and largely confined to specific age groups. In contrast, mixed graphical models revealed more pronounced gender differences in the conditional organisation of HRV. Among males, age was negatively associated with SDNN, SDNN was strongly associated with RMSSD, and weaker inverse relationships emerged between age and average heart rate and LF/HF ratio. The female network was substantially sparser, with an association observed primarily between ethnicity and LF/HF ratio, and no association between SDNN and RMSSD. These findings demonstrate that network analysis can reveal structural differences across gender that may be overlooked in traditional analyses. More broadly, they highlight the utility of psychological network analysis as an exploratory, hypothesis-generating tool for characterising multivariate conditional structures, even when the network theory underlying the approach is not assumed to hold.
Related
- Dominant Parasympathetic Modulation of Heart Rate and Heart Rate Variability in a Wild-Caught Seabird
- Comparison of Four Yogic Breathing Techniques Regarding Their Impact on Heart-Rate Variability and Breathing Frequency
- Self-reported sleep quality, physical activity and anxiety associated with heart rate variability in youth tiers derived via unsupervised machine learning
- Heart Rate Variability in Cardiometabolic Disease: From Molecular Mechanisms to Personalized Exercise Prescription
- Association of Total Atherosclerosis Burden of Baroreceptor-Resident Arteries with Heart Rate Variability and Its Circadian Rhythm in Patients with Ischemic Stroke
BibTeX
@misc{sheng2026network,
title = {{A Network Approach to Analysing Heart-Rate Variability}},
author = {Ho Han Sheng and Emily Ortega},
howpublished = {PsyArXiv},
year = {2026},
month = oct,
doi = {10.31234/osf.io/uxb4s_v1},
url = {https://osf.io/preprints/psyarxiv/uxb4s_v1/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - A Network Approach to Analysing Heart-Rate Variability
AU - Ho Han Sheng
AU - Emily Ortega
PY - 2026
DA - 2026/10/03
PB - PsyArXiv
DO - 10.31234/osf.io/uxb4s_v1
UR - https://osf.io/preprints/psyarxiv/uxb4s_v1/
AB - Heart rate variability (HRV) reflects autonomic nervous system functioning and is commonly examined using statistical methods that compare individual HRV indices across demographic groups. Such approaches, however, may overlook group differences in how HRV components relate to one another. This study applies a network approach to delineate gender differences in the conditional relationships among short-term HRV indices and demographic characteristics in a Singaporean adult sample (N = 2147). Conventional regression analyses identified gender differences in average heart rate and low-frequency/high-frequency (LF/HF) ratio, whereas differences in root mean square of successive differences (RMSSD) and standard deviation of normal-to-normal intervals (SDNN) were minimal and largely confined to specific age groups. In contrast, mixed graphical models revealed more pronounced gender differences in the conditional organisation of HRV. Among males, age was negatively associated with SDNN, SDNN was strongly associated with RMSSD, and weaker inverse relationships emerged between age and average heart rate and LF/HF ratio. The female network was substantially sparser, with an association observed primarily between ethnicity and LF/HF ratio, and no association between SDNN and RMSSD. These findings demonstrate that network analysis can reveal structural differences across gender that may be overlooked in traditional analyses. More broadly, they highlight the utility of psychological network analysis as an exploratory, hypothesis-generating tool for characterising multivariate conditional structures, even when the network theory underlying the approach is not assumed to hold.
ER -