A Sensitivity Analysis of an Inverted Pendulum Balance Control Model

Area of Science:

• Biomechanics
• Systems Neuroscience
• Human Movement Science

Background:

• Balance control models are essential for understanding human stability in health and disease.
• The Independent Channel (IC) model is frequently used, but the specific roles of its parameters require detailed investigation.

Purpose of the Study:

• To determine the unique contribution and relative importance of each parameter within the IC balance control model.
• To identify which specific parameters are most crucial for accurately describing balance behavior.

Main Methods:

• Balance behavior was quantified using transfer functions (TFs) relating sensory perturbations to body sway across frequencies.
• Local sensitivity analysis using partial derivatives was performed on each model parameter for TF magnitude and phase.
• An inverted pendulum model with a neuromuscular system controlled by various parameters was utilized.

Main Results:

• Intrinsic stiffness and proportional gain significantly shape low-frequency magnitude (0.1-1 Hz).
• Derivative gain influences the peak and slope of magnitude between 0.5-0.9 Hz.
• Sensory weight affects overall magnitude, while time delay impacts phase above 0.6 Hz; force feedback and intrinsic stiffness have lesser effects.

Conclusions:

• Sensory weight, time delay, derivative gain, and proportional gain uniquely affect transfer functions, highlighting their critical roles.
• Identifying crucial parameters provides deeper insight into the mechanisms underlying balance differences across patient groups.
• This analysis refines our understanding of the IC model, aiding in the diagnosis and treatment of balance disorders.