The balance of a standing human is precarious. Postural reflexes are continually evoked to maintain the stability. Three main sensory systems contribute inputs to these reflexes:
  • visual
  • vestibular
  • somatosensory



    Normally, all of these are active, but in certain circumstances one or more may break down. For instance, visual afferents are of no use in darkness, and may be counter-productive when the visual field moves markedly, as in motion sickness. In such circumstances, the central nervous system relies upon the vestibular and somatosensory systems to maintain stability. Together, these make up the proprioceptive system, which, as its name suggests, is responsible for collecting information about the body itself.

    Standing bio-mechanics

    To understand and assess balance, it is important to become familiar with some basic bio-mechanical terminology.

    Firstly, like any other object, the body has a centre of gravity (CG) or center of mass (COM), which acts vertically downwards (towards the centre of the earth). In order for an object to remain stable this line from the CG to the ground must fall within the base of support. For the body, this base of support is composed of the two feet and the area between them.

    In order to understand how the nervous system maintains the CG over the base of support, another important concept must be introduced. This is the ground reaction vector (GRV), which is the force (reaction) of the ground on the body, and is a function not only of the position of the CG, but also the muscle actions at the joints of the body. Were the body to be perfectly still, this vector would lie exactly over the CG line. However, in normal standing there is always some postural sway, and the GRV is the mechanism which the nervous system uses to control this sway.

    To see how this happens, look at fig. 1. At any given instant of time, the GRV crosses the ground at the so-called centre of pressure (CP). This is often confused with the CG, but is quite different. It is simply that point at which the pressure over the soles of the feet would be if it were concentrated in one spot. It can be measuredvery easily by a force plate, which we will meet in this laboratory. The important concept to grasp is that when this CP lies to one side of the CG, it causes the CG (and body with it) to accelerate in the opposite direction. It is rather like a Venetian gondolier who uses a pole to push the boat away from the place at which its planted. Thus, by careful alignment of the CP, the sensorimotor system can correct for perturbations in the position of the CG. Normally, the sway of the body CG is kept very small indeed (typically < 5mm), but note that in doing so the CP may move over a much more greater distance. It is rather like a sheep-dog controlling a herd of sheep - the dog moves rapidly to and fro around the herd, while the sheep themselves maintain roughly the same direction.

    Motion of the CP is a result of muscle actions in the body, and certain strategies are used preferentially by the sensorimotor system. For example, in the sagittal (antero-posterior) plane, the ankle joint (triceps and anterior tibial) musculature normally acts first to control sway - the so-called ankle strategy. Should this prove inadequate, more proximal joints (knee, hip etc.) will be activated to restore stability. Finally, if all else fails, the CG moves outside the base of support, and a corrective step is taken in a desperate attempt to regain stability.

    Frontal plane (medio-lateral) sway, on the other hand, is primarily corrected for by a hip strategy, involving hip abductors and adductors. The body moves from side to side as a parallelogram. Should this fail, the ankle joint (peroneal and posterior tibial) musculature may be activated, but note that the very short moment arm of these muscles makes them ill-suited to the task. If necessary, flailing arm movements, or a sideways corrective step can be made.

    Force plate posturography

    Centre of Pressure Measurement

    A force plate is an instrument for measuring the ground reaction vector. It consists of a metal base plate supported at its corners by four force transducers. Simple equations give the co-ordinates of the centre of pressure:

    xCP = (My + Dz.Fx)/Fz

    yCP = (Mx + Dz.Fy)/Fz

    where xCP & yCP are co-ordinates of the CP in the sagittal and frontal planes, Mx &My are the moments (turning forces) around these axes, Dz = constant (=0.0533m for the plate used in this laboratory), and Fx & Fy are the shear forces in the two planes.

    By measuring the position of the CP at regular intervals, e.g. every 1/50th of a second, a plot of its locus, often called a spaghetti diagram (for obvious reasons!) is obtained. A typical example is shown in fig. 2. Quantification of this diagram is normally done by finding the mean and standard deviation of the centre of pressure. The larger the standard deviation, the larger the sway. However, note that this measure is not a true record of sway (which is motion of the body, or CG) but rather a measure of the activity of the motor system in moving the CP. The two are related, but it is possible to have an extremely large CP motion whilst having very little sway. In fact, this happens in the elderly, who maintain a large effort in maintaining the CG within a very small region. This is done, presumably, because their control over the CG is poor outside of this region due to nervous system degeneration.

    Despite this fundamental limitation in the use of CP recordings for sway estimation, their ease of measurement by force plates (compared to the more problematic measurement of CG) has contributed to their popularity, such that they are now often referred to as stabillograms.

    Centre of Gravity Measurement

    It is also possible to measure CG motion with a force plate, but the method is more complex. It involves a double integration of the horizontal (shear) force components, and as a result gives only the relative motion of the CG, not its absolute position:

    By Newton's second law of motion,

    F = m.a

    where F = force acting on body, m = mass of body, and a = acceleration of the CG of the body.

    Thus, a = F /m

    So, if we know the mass of the body and the shear (horizontal) forces on the body, we can calculate the accelerations of the CG in the horizontal plane.

    Further, since acceleration is the second derivative of position, position is simply the double integral of acceleration:

    xCG = Double integral (F/m ) + k

    where xCG = locus of CG line in the sagittal plane, andk is the integration constant. Unfortunately, the latter is unknown, so the absolute position of the CG is unknown, but for sway estimation this is not important - the relative motion is all that is required. In practice the CG tracing can be adjusted to coincide with the CP tracing at rest.

    CoM by low-pass Filtering

    In practice, another method for obtaining an estimate of the CoM tends to be more popular. This makes use of an observation (Benda et al, 1994) that low-pass filtering the CoP signal at 0.5 Hz results in a very close approximation of the CoM. Typically a 4th order Butterworth filter is used.

    Benda BJ, Riley PO & Krebs DE (1994) Biomechanical relationship between the center of gravity and center of pressure during standing. IEEE Trans. on Rehabil. Engg. 2: 3-10.

    Like to add something? Email me!