Identification of the Unstable Human Postural Control System.

Sungjae Hwang,Peter Agada,Tim Kiemel,John J Jeka

doi:10.3389/fnsys.2016.00022

Sungjae Hwang, Peter Agada + Show 2 more

Open Access

https://doi.org/10.3389/fnsys.2016.00022

Copy DOI

Abstract

Maintaining upright bipedal posture requires a control system that continually adapts to changing environmental conditions, such as different support surfaces. Behavioral changes associated with different support surfaces, such as the predominance of an ankle or hip strategy, is considered to reflect a change in the control strategy. However, tracing such behavioral changes to a specific component in a closed loop control system is challenging. Here we used the joint input–output (JIO) method of closed-loop system identification to identify the musculoskeletal and neural feedback components of the human postural control loop. The goal was to establish changes in the control loop corresponding to behavioral changes observed on different support surfaces. Subjects were simultaneously perturbed by two independent mechanical and two independent sensory perturbations while standing on a normal or short support surface. The results show a dramatic phase reversal between visual input and body kinematics due to the change in surface condition from trunk leads legs to legs lead trunk with increasing frequency of the visual perturbation. Through decomposition of the control loop, we found that behavioral change is not necessarily due to a change in control strategy, but in the case of different support surfaces, is linked to changes in properties of the plant. The JIO method is an important tool to identify the contribution of specific components within a closed loop control system to overall postural behavior and may be useful to devise better treatment of balance disorders.

Highlights

Human upright standing is intrinsically unstable, requiring sensory feedback to remain upright
In order to decode the mechanistic basis underlying the changes in the coordinative relationship between the trunk/leg segments on different support surfaces, we identified all four closed-loop frequency response functions (FRFs) [Hvu(f ), Hvy(f ), Hdu(f ), and Hdy(f )] and four open-loop FRFs [P(f), F(f), S(f), and M(f)] by using a MIMO system identification method
By identifying FRFs for each sensory perturbation relative to each muscle, we found that visual upweighting is greater for ankle EMG than for hip EMG, suggesting that sensory reweighting differs across actuators when the subject stands on short support surface

Summary

Introduction

Human upright standing is intrinsically unstable, requiring sensory feedback to remain upright. Individuals with bilateral vestibular loss cannot stay upright in Condition 6 of the Sensory Organization Test, which removes visual and proprioceptive input through sway-referencing (Peterka and Black, 1990) To solve this problem, we recently implemented an approach called the joint input–output (JIO) method (e.g., Fitzpatrick et al, 1996; Katayama, 2005; van der Kooij et al, 2005; Kiemel et al, 2008, 2011). The JIO method “opens the loop” for human upright standing by calculating multiple linear input–output relationships (gain and phase), between various components of the control loop and using mathematical techniques to isolate a particular process (e.g., feedback, see Materials and Methods Section below). Considering the many subsystems/processes involved in human upright stance control, such findings may have implications for populations with poor balance control, fostering more precise diagnosis, and treatment

Objectives

Methods

Results

Discussion

Conclusion