Zooming into the Complex Dynamics of Electrodermal Activity Recorded during Emotional Stimuli: A Multiscale Approach

Area of Science:

• Physiological complexity analysis
• Autonomic nervous system research
• Nonlinear dynamics in biosignals

Background:

• Physiological signals display complex, multi-time scale behaviors.
• Chaos theory techniques show promise in differentiating healthy and pathological states.
• Electrodermal activity (EDA) reflects autonomic nervous system state and exhibits fractal properties.

Purpose of the Study:

• To introduce the multiscale complexity index of electrodermal activity (MComEDA) for analyzing EDA signals.
• To assess MComEDA's ability to discern varied autonomic responses.
• To evaluate MComEDA's performance on ultra-short physiological time series.

Main Methods:

• Developed MComEDA by extending the ComEDA algorithm with a coarse-graining procedure.
• Applied MComEDA to EDA signals from the CASE and AMIGOS datasets.
• Utilized Wilcoxon signed rank test with Bonferroni's correction for statistical analysis.

Main Results:

• MComEDA values significantly differed between high- and low-arousal stimuli (p < 0.05).
• MComEDA demonstrated superior performance over single-scale methods in discriminating valence levels within high-arousal stimuli (e.g., scary vs. amusing, p = 0.024).
• The multiscale approach enhanced information extraction from short EDA time series.

Conclusions:

• MComEDA provides a robust method for analyzing multi-time scale complexity in EDA signals.
• A multiscale nonlinear analysis of EDA improves the characterization of task-specific autonomic responses.
• This approach is effective even with limited data duration, crucial for real-time applications.