Human emotion recognition using heart rate variability analysis with spectral bands based on respiration

Area of Science:

• Psychophysiology
• Biomedical Engineering
• Affective Computing

Background:

• Human emotion recognition is crucial for understanding well-being and social interaction.
• Heart Rate Variability (HRV) analysis offers a non-invasive method to assess autonomic nervous system activity.
• Traditional HRV analysis may not fully capture the nuances related to respiratory influences on emotional states.

Purpose of the Study:

• To assess the efficacy of HRV analysis in recognizing human emotions.
• To investigate the role of spectral band adjustments based on respiratory frequency (RF) in emotion detection.
• To compare calm-neutral (Relax), positive (Joy), and negative (Fear) emotional states using HRV.

Main Methods:

• Standard time and frequency domain HRV analysis was performed.
• The high-frequency (HF) band was specifically centered on the individual's respiratory frequency (RF).
• Key HRV indices including power in low band (PLF), normalized power in HF band (PHFn), LF/HF ratio, and pNN50 were analyzed.

Main Results:

• PLF, PHFn, and LF/HF distinguished between Relax and Joy states.
• Mean heart rate and RF differed significantly between Relax and Fear states.
• Several HRV indices (pNN50, PLF, PHFn, LF/HF) showed significant differences between Joy and Fear.
• Centering the HF band on RF yielded lower p-values in statistical analysis compared to standard HF bands.

Conclusions:

• HRV analysis, especially with RF-centered spectral bands, is a promising tool for emotion recognition.
• Specific HRV parameters effectively differentiate between distinct emotional states.
• Adjusting HRV analysis to individual respiratory patterns enhances its sensitivity for emotion detection.