These experimental and analytical methods provide guidelines that aim to understand to role of the many components of the central nervous and musculoskeletal systems in human postural control. Postural control models with meaningful physical parameters can be used to investigate the role and interaction of the sensory systems and their changes due to disease and aging. These methods can be used to assess patient balance problems, to reveal the ideology of the impairment, and to aid in the design of interventions for improving postural control.
These methods can also be used to study the interactions between sensory motor pathologies and balance control. For example, for fall prevention in the elderly. This protocol provides a means to investigate the relative contributions of sensory modalities including proprioceptive and visual systems and their interactions as well as muscle passive contributions to postural control.
To prepare a subject for the measurement of electromyography from ankle muscles, use single differential electrodes with an interelectrode distance of one centimeter. Mark the medial gastrocnemius at the most prominent bulge of the muscle, the lateral gastrocnemius at one third of the line between the head of the fibula and the heel, the soleus at two thirds of the line between the medial condyles of the femur and the medial malleolus, and the tibialis anterior at one third of the line between the tip of the fibula and the tip of the medial malleolus. When all of the points have been marked, use a razor to shave each area and clean the skin with alcohol.
When the skin has dried, use double sided tape to attach one electrode to each area taking care that each electrode is fixed to the skin securely. To prepare the subject for a kinematic measurement, first use a strap to attach a reflective marker as high as possible on the subject's shank and have the subject put on the body harness. Use a strap to attach a reflect marker to the subject's waist and have the subject climb on to the standing apparatus.
Adjust the subject's foot position to align the lateral and medial malleoli of each leg to the pedal axis of rotation and use a marker to outline the foot positions. Instruct the subject to keep their feet in the same locations during the experiments and adjust the vertical position of the laser range finders to point to the center of the reflective markers. Then, adjust the horizontal distance between the laser range finder and the reflective markers so that the range finders work in their mid range and do not saturate during standing experiments.
Before beginning the experiment, inform the subject of what to expect for each trial condition. Instruct the subject to stand quietly with hands at the side while looking forward maintaining their balance as they do so when faced with the real world perturbations. For a quiet standing trial, have the subject stand still for two minutes with no perturbations.
For perturbed experiments, if the objective is to investigate the role of somatosensory system or ankle stiffness in standing, apply pedal perturbations for two to three minutes while recording the data. If the objective is to examine the role of vision in postural control, apply visual perturbations by rotating the visual field using the virtual reality headset for two to three minutes while recording the data. If the objective is to examine the interaction of the two systems in postural control, apply the visual and pedal perturbations simultaneously.
For non-parametric identification of the dynamic relation of the body angle to visual perturbations after loading the visually perturbed trial data into a suitable analysis software program, use the commands as indicated to decimate the raw body angle and the visual perturbation signals and remove the means from the decimated signals. Determine the decimated sampling frequency, then select the lowest frequency of interest to determine the window length and choose the degree of overlap for the estimation of power spectra. Define the vector of frequencies at which the frequency response is to be estimated.
Use the TF Estimate function to find the frequency response of the system as indicated and find the gain and phase of the estimated frequency response as demonstrated. Then, use the command as indicated to calculate the coherence function and plot the gain, phase, and coherence as a function of frequency. In this example of a typical standing trial with visual perturbations, a trapezoidal signal applied by the virtual reality headset can be observed where the field of view rotates from zero to plus or minus 0.087 rad in the sagittal plane.
The ankle and body angles were very similar in this analysis since the foot angle is zero and the shank and upper body move together. The ankle torque was also correlated with the shank and body angles. Electromyographs from the ankle muscles demonstrate that the soleus and the lateral gastrocnemius are continuously active, but the medial gastrocnemius periodically generates large bursts of activity with body sway and that the tibialis anterior is silent.
Here, a frequency response of the transfer function relating the visual input to the body angle for the standing trial data is shown. In this experiment, the coherence was high at low frequencies up to around one hertz and dropped significantly at higher frequencies, meaning the frequency response is meaningful up to one hertz. The gain initially increased from 0.1 to 0.2 hertz before decreasing to one hertz, demonstrating the expected low pass behavior due to the body's high inertia.
The phase also started at zero and decreased almost linearly with the frequency indicating that the output was delayed with respect to the input. Take care to align the ankle axis of rotation with that of the actuator. Make sure that the subject does not generate extra movements and ensure that the appropriate mechanical and visual perturbations are used.
First time users may have difficulty in establishing a consistent repeatable experimental paradigm and in using the appropriate identification methods that account for close loop, non-linear, and time varying effects in postural control. These methods have been used to investigate healthy postural control and its adaptation as well as to quantify changes in balance control under a variety of experimental and clinical conditions.
