Uwe Proske
Department of Physiology, Monash University, Clayton, Victoria, Australia

https://doi.org/10.36866/pn.66.20

The present-day view for muscles acting about the elbow joint is that an important source of positional information comes from the muscle spindles of elbow flexors and extensors. We believe that muscle spindles signal muscle length. During rotation of the joint, as a muscle is stretched, spindle discharge increases in direct proportion to the size of the length change. In other words, signals of muscle length change are interpreted by the brain as movements about the joint. There are two pieces of evidence in support of the view that muscle spindles provide information about limb position. The first were the ground-breaking observations of Goodwin et al. (1972) who described illusions of movement and changed position of the elbow during vibration of elbow muscles. At that time it was already known that muscle spindles were exquisitely sensitive to vibration (Brown et al. 1967). The second observation supporting a role for muscle spindles as kinaesthetic sensors uses the muscle property called thixotropy. This is the change in passive tension of skeletal muscle following conditioning contractions and length changes. Since the intrafusal fibres of muscle spindles, on which the spindle sensory endings lie, also exhibit thixotropy, it means that by conditioning the muscle it is possible to alter spindle responsiveness without changing the length of the muscle. That, in turn, means subjects make conditioning-dependent errors in a forearm matching task (Gregory et al. 1988).

To explain how we use muscle conditioning to explore position sense I have drawn a simple diagram (Fig. 1). Contracting elbow flexors with the arm held flexed leads the intrafusal fibres in elbow flexor spindles to lie ‘taut’, generating strain on sensory endings and producing high levels of resting discharge in the spindles. When the relaxed arm is placed at an intermediate test angle, activity levels in flexors will continue to be high, in extensors they will be low. The subject interprets the high discharge levels as a muscle that is more stretched than it really is. It leads subjects to make matching errors in the direction of arm extension.

If now the contraction is repeated but with the arm held extended, during the subsequent movement to the test angle, flexor muscles and their spindles fall slack, leading to a low resting discharge (Fig. 1). By contrast, extensor spindle activity will be high. It means that after extension conditioning subjects perceive their forearms to be more flexed than they really are. By using the two forms of conditioning, total errors
in position sense of 20º can be achieved, representing about a quarter of the full range of elbow movements available to the subject under the conditions of our experiment. So the effects of conditioning are considerable.

The usefulness of the conditioning method is that is provides evidence for a role for muscle spindles in position sense that is quite different from the effects of muscle vibration. It is unlikely that other potential contributors to position sense, joint receptors and skin stretch receptors,
will show muscle conditioning­dependent effects.

The question that has consumed me in recent times is how does position sense work when limb muscles are contracting? For the forearms, how do we know where our arms are when we support them against gravity, or when they are bearing a load? The important consideration here is that whenever we carry out a voluntary contraction to support our limbs the intrafusal fibres of muscle spindles are contracted as well. This is called co-activation (Vallbo, 1971). It means that as soon as limb muscles become active, the spindle signal fed back to the brain from those muscles increases dramatically. Now we face a new problem. How is the brain able to distinguish between spindle impulses generated as a result of muscle length changes from impulses generated from intrafusal contraction? It is really that question which fascinates me.

Rather than go through the various hypotheses that have been put forward in an attempt to solve this problem, I would like to describe some recent observations from our laboratory relevant to the topic (Ansems et al. 2006).

We chose to carry out the measurements of position sense about the elbow joint in the horizontal plane. The reason for doing this was that we did not want our observations to be complicated by the effects of gravity. The arm was supported by a cradle that rotated about a pivot point coaxial with the elbow joint. Movement of the relaxed arm was almost effortless. Position of one arm, set by the experimenter, was matched by the other, moved by the subject. In addition, by means of a series of pulleys arm muscles could be made to support a load, or to move it.

The first thing we found was that when elbow muscles began to contract to support a load, the conditioning effects disappeared. They don’t disappear immediately at contraction threshold, but become progressively smaller as the contraction grows until, typically with a strength of a quarter of maximum for elbow flexors, they are no longer significant. For a group of subjects, position errors of 4.5º from muscle conditioning in the relaxed muscle reduced to 0.8º at 25% MVC (maximum voluntary contraction, Fig. 2A).

We explain this result by proposing that where muscles and their spindles were slackened by conditioning (conditioning extended for elbow flexors in Fig. 1), the fusimotor activity accompanying a contraction leads to removal of the slack and to sensitisation of spindles. Therefore during a contraction the errors from flexion and extension conditioning converge towards the flexion conditioned state, where spindles are sensitised.

So the distribution of position errors at the forearm during loading supports the ideas that during a voluntary contraction the fusimotor and skeletomotor systems are co-activated. What, perhaps, is a little unexpected is that co-activation is distributed over about a quarter of the muscle’s working range.

None of this helps us, of course, in understanding how position sense works during load bearing. What we considered to be a clue was that when flexor spindles were in their sensitised state (conditioned flexed, Fig. 1), loading flexor muscles did not introduce additional position errors.

During load bearing, the fusimotor activity would be expected to produce a large increase in spindle discharges, yet position errors remained unchanged. Nor was there any evidence of subjects becoming more erratic in their matching ability. Standard errors of the mean remained about the same (red trace, Fig. 2A). So whatever theory is used to explain position sense during load bearing, it must take such a result into account.

There is one more clue to add to the puzzle. When a passive muscle is vibrated, as mentioned earlier, it leads to illusions of limb movement and changed position. We measured the position error from vibration of the passive muscle and compared it with errors during vibration of a contracting muscle. Interestingly the position errors from vibration were gone at 25% MVC (Ansems et al. 2006). We believe that fusimotor-activated spindles are still vibration sensitive (Brown et al. 1967) but they do not seem to generate any illusions. Is it possible that fusimotor­activated spindles no longer have access to consciousness?

Finally, we tried a slightly different approach. Following muscle conditioning, the subject was required to move a load from the conditioning position to the test angle. This time the difference in conditioning effects did not disappear on loading the arm (Fig. 2B). It was as though moving a load introduced an additional signal that produced errors which added to the errors from conditioning, particularly when the conditioning had sensitised spindles (red trace, Fig. 2B).

To conclude, we have made a number of observations on position sense at the forearm during contraction of elbow muscles. While the picture remains fragmentary, we have provided new directions for future experiments. We no longer believe that the sense of effort accompanying support of a load provides positional information in any simple way (Walsh et al. 2004).

Our current working hypothesis is that when we move a loaded arm the brain listens to the feedback during placement of the arm and compares it with feedback levels generated in the past from similar movements. This information is used in deciding where to place the other arm in a position matching task.

References

Ansems GE, Allen T & Proske U (2006). Position sense at the human forearm in the horizontal plane during loading and vibration of elbow muscles. J Physiol 576, 445-455.

Brown MC, Engberg I & Matthews PB (1967). The relative sensitivity to vibration of muscle receptors of the cat. J Physiol 192, 773-800.

Goodwin GM, McCloskey DI & Matthews PB (1972). The contribution of muscle afferents to kinaesthesia shown by vibration induced illusions of movement and by the effects of paralysing joint afferents. Brain 95, 705-748.

Gregory JE, Morgan DL & Proske U (1988). Aftereffects in the responses of cat muscle spindles and errors of limb position sense in man. J Neurophysiol 59, 1220-1230.

Vallbo AB (1971). Muscle spindle response at the onset of isometric voluntary contractions in man. Time difference between fusimotor and skeletomotor effects. J Physiol 218, 405-431.

Walsh LD, Hesse, CW, Morgan DL & Proske U (2004). Human forearm position sense after fatigue of elbow flexor muscles. J Physiol 558, 705-715.

Wood SA, Gregory JE & Proske U (1996). The influence of muscle spindle discharge on the human H reflex and the monosynaptic reflex in the cat. J Physiol 497, 279-290.