Comparison of different methods to identify and quantify balance control

Research output: Contribution to journalArticleAcademicpeer-review

142 Citations (Scopus)

Abstract

The goal of this paper is to clarify the methodological aspects of studies of human balance during quiet standing and perturbed standing. Centre of mass (CoM), centre of pressure (CoP) and electromyogram (EMG) or similar measures are commonly recorded to quantify human balance control. In this paper we show that to identify the rigid body dynamics and the physiological mechanism that controls the body separately, one has to externally perturb the body with known perturbations and to use the indirect (IA) or joint input–output approach (JA) for identification. However, in many balance control studies the direct approach (DA) have been used, which is well suited to study open-loop systems but will give erroneous results when applied to a closed-loop system, as in human balance control. The cross-correlation function and linear regression are examples of the erroneous application of the DA approach in human balance control studies. The consequences of this erroneous DA are given. In addition a new application of the JA is presented that identifies physiological mechanisms that control balance, including passive and active feedback pathways. This new method is compared with existing identification schemes that use the IA and an existing JA that estimates the active pathway. Also it is shown how descriptive measures such as the power spectral densities (PSD) or the stabilogram diffusion plot (SDP) of the CoP and/or CoM depends on the PSD of internal perturbations and sensor noise, which are not measured. Although descriptive measures can be used to describe the state of the balance control system for a particular situation, it does not separate the dynamics of unknown processes that perturb balance from the dynamics of the active and passive feedback mechanisms that controls balance. Only the IA and the preferred JA can give estimates of the passive and active passive feedback mechanisms that control balance.
Original languageUndefined
Pages (from-to)175-203
Number of pages28
JournalJournal of neuroscience methods
Volume145
Issue number1-2
DOIs
Publication statusPublished - 2005

Keywords

  • Identification
  • Linear regression
  • METIS-227858
  • Balance control
  • IR-77413
  • Stabilogram diffusion analysis
  • Cross-correlation function
  • Closed loop systems
  • Power spectral density

Cite this