• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jphysiolThe Journal of Physiology SiteMembershipSubmissionJ Physiol
J Physiol. Apr 1, 2005; 564(Pt 1): 2.
PMCID: PMC1456057

Paradoxical muscle contractions and the neural control of movement and balance

Over the past three decades, research focused on the sensory aspects of standing has shown how the brain integrates sensory information from many different sources to create a functional balance response that adapts to virtually any situation. Little attention has been given to how the leg muscles transform that neural response into the force that keeps the body upright. Now, in this issue of The Journal of Physiology, Loram et al. (2005a,b) show that the muscles are not behaving as assumed. What does this mean for our ideas of movement and balance control?

When standing, the body behaves more or less as an inverted pendulum that sways about the ankles (Smith, 1957). The amount it leans from the vertical determines the torque required to balance it. The centre of mass of the body is kept in front of the ankles so that the body always tends to fall forwards. The calf muscles contract to pull it backwards. Standing then should just require a continuous contraction at precisely the right level to balance the body.

What keeps this muscle activity at just the right level? The classical Sherringtonian view would be that tonic reflexes provide the basic level of drive to keep the body pointed upright and phasic stretch reflexes are there to control the sway of the body. The vestibular organs and musculature of the trunk form the sensory arc of the tonic reflexes. The lengthening and shortening of the calf muscles that occurs with body sway produces a muscle spindle input to stretch reflexes. Swaying forward should stretch the calf muscles and evoke a reflex contraction that restores the position of the body.

Such formulations encountered problems. It became apparent that human stretch reflex gains are not particularly large (Crago et al. 1976). In any case, it appeared that they could not operate with greater effect because the delays of neural transmission would create feedback instability. In contrast, the natural elastic stiffness of muscle, which is instantaneous, would provide an effective means of maintaining posture (Rack & Westbury, 1974). Thus, a view developed that forward sway stretches the fibres of actively contracting muscles. The muscles themselves act like springs. In producing the appropriate contraction and muscle force to hold a load, the CNS actually sets the muscle's spring constant (stiffness) and the unstretched spring length at which force is zero (Houk, 1979). When applied to standing (Winter et al. 2001), this mechanical tone hypothesis argues that the muscle spring holds the inverted pendulum upright much like a guy rope holds up a tent pole. Once in position, no further intervention is needed.

All of the above tacitly assumes that the tendon, which couples the muscle to the support, is stiff enough for the task. There is no point tying the guy rope to a rubber band because it will keep stretching until the tent has fallen. There is a minimum stiffness of the muscle–tendon guy rope that will support the pendulum. What defines this minimum stiffness? The guy rope must acquire more potential energy by being stretched than the pendulum loses by falling. This energy will return the body to the target position. If the guy rope is not stiff enough, the energy it gains by being stretched is not enough to offset the energy that the pendulum loses as it falls. Seeking to minimize energy, the pendulum stretches the spring until it hits the floor.

This is precisely the problem for human standing. In an earlier paper, Loram & Lakie (2002) showed that the stiffness of Achilles' tendons is only about 90% of the body's load stiffness. It does not matter how stiff the muscle is because the tendon, being in series with the muscle, defines the maximum possible stiffness of the guy rope. This makes the mechanical tone model of standing untenable. It may work for other joints and loads, but for the ankles and standing, the body's load stiffness exceeds the capability of the guy rope.

Loram et al. (2005a) provide beautiful data made by watching gastrocnemius and soleus with real-time ultrasound during normal unperturbed standing. With eight markers on the fascicle images, spatial cross correlation and vector analysis resolves fascicular movements to 10 µm: down to the diameter of the smallest muscle fibres! What they see they call ‘paradoxical’ movements, paradoxical because they go the wrong way. When the body sways forward and the anatomical muscle (origin to insertion) lengthens, the muscle fascicles shorten. During backward sway, they lengthen.

This is a consequence of balancing a load that is stiffer than the tendon. Somehow, the brain has to make the total tendon-plus-muscle stiffness greater than the load stiffness. What if the muscle acts as a negative spring, getting shorter the more it is pulled? This is an inherently unstable situation and the muscle would shorten to its limit as the body continued to fall.

Loram et al. (2005b) go on to show how the brain and the muscle solve this load problem. Since ‘static’ equilibrium cannot be achieved by continuous muscle contraction, the system adopts a pattern of cyclic muscle activation, producing repeated ballistic, catch-and-throw movements of the body. The behaviour resembles keeping a balloon in the air by repeated hits. Over time, the balloon maintains a mean position, which might seem an equilibrium point, but at no time does it stay in static equilibrium; it is either being accelerated upward as it is hit or it is in free fall.

Standing is a dynamic activity. It has been believed that normal body sway comes from small perturbation forces, either internal to the body (respiration) or external (breezes); limited sensory acuity to detect body movement; receptor noise; motor output noise; or movement generated by the brain. What Loram et al. (2005b) show is entirely different. A major cause of human body sway arises from this cyclic pattern of catch-and-throw ballistic muscle activation.

Based on the spatial and temporal accuracy of these measurements of muscle ‘activity’, Loram and colleagues have made a major advance. Their ultrasound and analysis technique will provide a valuable tool for many areas of motor control research.

These observations certainly challenge the simplistic view that the mechanical stiffness of contracting muscles provides a spring-like action that is used for postural control. Standing appears to be one activity that is beyond the domain of that model. These papers show how important it is that we consider muscle action in terms of the load that it works against.

The implications for reflex control are not so clear. We know that sensory input from the eyes, the feet and other cutaneous sources all reduce sway. How is sensory integration melded with this intermittent ballistic model that has developed to address the mechanical aspects of balance control? How are perturbations to the body that need active responses dealt with? Perturbation forces will still stretch the leg muscle fascicles albeit relative to this paradoxical movement. Sensory feedback control remains and with it the problems of delays and instability. Thus, the specific conditions of load and tendon stiffness during standing could create this cyclic movement pattern through instability in the entire neuromuscular control, rather than it being driven centrally by an intermittent control. The causalities of this newly identified behaviour need teasing out.


Articles from The Journal of Physiology are provided here courtesy of The Physiological Society
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...