| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Physiology, Monash University, Melbourne, Victoria; and 2Prince of Wales Medical Research Institute, University of New South Wales, Sydney, Australia
Submitted 11 October 2005 ; accepted in final form 19 December 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
15%. This was accompanied by a significant increase in matching errors for the unsupported matching task from 2.7 ± 0.5 to 0.8 ± 0.7° but not for counterweighted (1.4 ± 0.2 to –0.2°± 1.1°) or horizontal matching (–1.3 ± 0.7° to –1.8 ± 0.7°). This, it is postulated, is because the reduced voluntary torque after exercise was accompanied by a greater effort required to support the arms, leading to larger matching errors. However, effort is only able to provide positional information for unsupported matching where gravity plays a role. In gravity-neutral tasks like counterweighted or horizontal matching, a change in the effort-force relationship after exercise leaves matching accuracy unaffected. effort; fatigue; kinesthesia; muscle spindle
A number of recent observations have made it necessary to reassess these views. It is a common experience after a period of intense physical activity to feel unsteady on the legs and clumsy in the execution of skilled movements. Such observations led us to carry out experiments on the effects of exercise on the senses of limb position and movement. Exercise, sufficiently intense to be accompanied by a significant fall in muscle force, led subjects to make errors in a forearm position matching task (1, 20). The direction of the errors suggested that, in matching position, subjects were making use of signals of both peripheral and central origin. The central signal, associated with the sense of effort, became disturbed after the exercise as a result of fatigue. These observations suggested that, in the absence of vision, we determine the position of our limbs in space, by reliance, in part, on the amount of effort required to maintain limb position against gravity. This proposition would help account for the proprioceptive disturbances known to occur in high-gravity (11) or low-gravity (27) environments and for differences in motor learning strategies dependent on whether subjects are instructed to adopt a spatial or effort goal (10).
In the present study, we have extended our observations to include conditions where the force of gravity cannot contribute to position sense. The specific hypothesis tested was that muscle fatigue from exercise would have no effect on position sense in an unloaded matching task in gravity-neutral circumstances. Three experiments have been carried out. First, position sense was measured with the hands in front, unsupported, as in our laboratory's previous studies (1, 20). Then we repeated position sense measurements in the same posture after both arms had been counterweighted so that now movement of the forearms from one position to another was essentially effortless. Finally, we measured position sense in a matching task in which the forearm was moved in the horizontal plane. Here, too, the arm could be moved with a minimum of effort. Horizontal matching was chosen as an additional task because it was considered that movement in the vertical plane without effort was rather unnatural. Each of the three sets of measurements was repeated before and after a period of eccentric exercise that led to a significant fall in force. It was postulated that, if the sense of effort played a role in position sense in any of these tasks, position errors after the exercise should increase in proportion to the fall in muscle force. Specifically, it was predicted that, after the exercise, position errors for the unsupported matching task would increase in proportion to the fall in force (20); for the counterweighted and horizontal matching tasks, it would not change because gravity played no part in these tasks.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Each subject attended several sessions. A series of control trials was carried out to familiarize subjects with the equipment and procedures. In the unsupported position matching task, blindfolded subjects were required to achieve an accuracy of 
3° of error and a standard deviation of 3.5°. Four subjects were excluded from the experiments because they were unable to achieve the required matching accuracy. Another exclusion criterion was the size of the force drop at 24 h after the exercise. When it was 15% maximal voluntary contraction (MVC), the subject was not used. This excluded a further two subjects. These two subjects were nevertheless still used for preexercise control measurements. It meant that eight subjects provided control data and that six subjects were used to study the effects of exercise. In these experiments we used two test angles, 30 and 60° from the horizontal position, which was assigned 0°. These angles had been selected in a previous study and found to provide suitable measures of elbow positional accuracy (20).
Position Sense
The measurement procedure used was the same as that described in a study by Walsh et al. (20) in 2004. The subject had both forearms strapped to lightweight paddles, with the paddle hinges arranged to be coincident with the elbow joint (Fig. 1). The height of the apparatus was adjusted so that, when the forearms were strapped to the paddles, the upper arms were at 
45° to the horizontal. Potentiometers attached to the paddle hinges provided a voltage signal proportional to the elbow joint angle. Position signals were acquired at 100 Hz using MacLab 8/s running Chart software (AD Instruments, Castle Hill, NSW, Australia). Resolution of elbow angles was 0.2° with the vertical apparatus, and 0.5° with the horizontal apparatus.
|
In the first experiment, the subject's reference arm was moved by the experimenter from the horizontal position (0°) to the test angle. Subjects were told to hold their arm in that position, which was 30 or 60° from the horizontal. The two test positions differed in that the 30° test position, being closer to horizontal, required 1.6 times more flexion torque to support the weight of the arm, which represented 3.4% MVC compared with 2.1% MVC for 60°. Subjects were asked to match the position of their reference arm by placement of the matching arm. The angles adopted by the two arms were recorded. After each match, the subject returned their forearms to the horizontal position, ready for the next match. A total of five matches was carried out at 30° and a further five matches at 60° for each subject. The order of presentation of angles was randomized. Matching errors were calculated as the difference in angle between reference and matching arms. Errors by the matching arm in the direction of extension were assigned a positive value; in the direction of flexion they were assigned a negative value.
Matching With Counterweighted Arms
In a second experiment, matching trials were carried out with both arms counterweighted (Fig. 1). Counterweighting was achieved by attaching a rigid steel shaft directed backward at each paddle hinge. A 1-kg weight could be slid up or down the shaft, and, when it precisely counterbalanced the weight of the forearm at the test angle, it was locked in position. The position of the counterbalance was adjusted for each subject, individually, at the start of a set of measurements. It meant that the subject's forearm could be placed at a set angle, and it would remain there without effort. The aim of this experiment was to try to minimize any cues from gravitational torques about the elbow joint while giving subjects access to other, peripherally derived positional information about the location of the elbow and hand.
Horizontal Matching
The third matching task involved matching the positions of the forearms in the horizontal plane, where there would be no gravitational cues. The subject sat at a table, with forearm and upper arm supported by a cradle that was hinged at a point coincident with the elbow joint (Fig. 1). The arms were moved across a surface subdivided into angular positions in degrees, the locations of the arms being indicated by a laser light below each hand. The horizontal movement was almost frictionless and required no effort to maintain a given elbow angle. When the forearm was extended at 90° to the trunk, in the horizontal plane, this was assigned 0°, which correspond to an included elbow angle of 130°. The test elbow angles used for horizontal position matching corresponded to the angles used in vertical matching. Position errors were calculated as for the vertical matching task by taking the difference between reference angle and matching angle. A positive error meant that the matching arm had adopted a more extended position than the reference arm.
In preliminary experiments using the counterweighted and horizontal matching equipment, subjects' biceps and triceps of the reference arm had surface electromyograph (EMG) electrodes attached over the belly of the muscle. Recordings showed that, once the reference arm had been placed in its test position, elbow muscles remained electrically silent.
Eccentric Exercise
Before the exercise, subjects carried out several MVCs. MVCs were measured for elbow flexors of both arms. For this the forearm paddles were locked in the vertical position, that is, giving an elbow angle of 90°, which approximates to the optimum angle for torque in elbow flexors (23). The paddle was held in that position by a horizontal metal bar that was in series with a strain gauge. Subjects were required to flex their elbow maximally for three contractions, each 3 s long and separated by a 1-min rest period. The mean peak value of force was used as the MVC measurement.
The eccentric exercise consisted of lowering a weight with the arm. The weight was adjusted to be 30% MVC for elbow flexors. It had been shown previously that this was a sufficiently intense exercise to produce a significant and prolonged force deficit (20), and it was well tolerated by subjects. The subject was required to lower the weight in a controlled way, making sure that the lengthening contraction continued until the arm was fully extended. This was important because previous experience (8, 16) had shown that damage to elbow flexors occurred particularly at the longer muscle lengths. Subjects were asked to take 2–3 s to carry out each contraction. They carried out five sets of 10 eccentric contractions with 20-s rests between each set. This comprised one exercise bout. Typically, subjects completed four to five sets (4.3 ± 0.3) depending on their fitness. As soon as subjects began to have difficulty in supporting the weight, the exercise was stopped. Six of the eight subjects showed a fall in MVC of 15–50% (33.8 ± 9.0%) 24 h after the exercise.
Subjects were asked to attend the laboratory on two occasions. During the first visit, the control measurements were carried out for the three experimental conditions: the unsupported arms, counterweighted arms, and matching in the horizontal plane. Then subjects were exercised. Subjects' position sense was not remeasured immediately after the exercise because at this point they began to tire and lose concentration. Position sense was therefore measured at 24 h postexercise.
Statistical Analysis
Statistical analysis was carried out on all matching values, including data from both the 30 and 60° trials. For the unsupported and the counterweighted matching tasks errors were calculated as angle (reference arm) – angle (indicator arm), where 0° = horizontal forearm and 90° = vertical forearm. For the horizontal matching task, errors were measured in the same way, but where 0° = forearm at 90° to the trunk, in the horizontal plane. Data were analyzed using the software Igor Pro version 5 (Wavemetrics, Lake Oswego, OR). Statistical analysis used Data Desk 6.2 (Data Description, Ithaca, NY). A two-way interactive ANOVA with repeated measures was used to test the differences in position errors between the pairs, unsupported vs. counterweighted reference arm and unsupported vs. horizontal reference arm, both before and after exercise. Comparisons of standard deviations were carried out using an F distribution to test the null hypothesis that the ratio of variances equaled unity. A pooled t-test was used to test for significant differences between the pooled pre- and postexercise errors for the three matching conditions. Significance was accorded a P value <0.05. Results in the text are given as means ± SE.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Unsupported arms. An example of position errors in a vertical matching task is shown in Fig. 2. The subject matched the reference elbow angle by adopting a more extended position with their matching arm (positive error). Errors for this subject were larger with a 30° reference angle compared with 60°. For the sake of clarity, data in the plots of Figs. 3, 4, and 6 are shown for 30° only, although statistical analysis was carried out on all values. Mean position errors for the unsupported arms were 2.7 ±0.5°.
|
|
|
|
Horizontal arms. When errors for the horizontal matching task were compared with values for unsupported matching, there were two trends. Errors were again less negative compared with the unsupported vertical matching task, and their range was somewhat wider (Fig. 3). Mean position errors were 1.3 ± 0.7°. Position errors in the horizontal posture were not significantly different from errors in the unsupported or counterweighted condition. However, in a comparison of standard deviations, mean values for horizontal matching (6.8°) were significantly larger than for unsupported matching (4.6°; P < 0.05; F test), but not significantly different from errors for counterweighted matching.
Position Matching After Exercise
For two subjects, the falls in MVC measured after the exercise had largely reversed during the 24 h so that the persisting deficit in force was 15% or more in only six subjects. Postexercise data were collected for only these subjects because it was considered that the effects of falls in force of <15% were likely to be small and therefore unlikely to emerge against a background of variability.
The prolonged fall in force indicated that the muscles had been damaged by the exercise. In support of that view, when elbow flexors were palpated, contracted, or stretched, all subjects reported modest levels of pain, delayed-onset muscle soreness (DOMS) known to result from this kind of exercise (9, 14, 22). A third indicator of damage was that subjects, standing with their arms hanging at their sides, tended on the exercised side to have their elbow more flexed. Such a flexed posture is indicative of a rise in passive tension in elbow flexors (9, 16, 22).
Measurements, made after exercise on the subject illustrated in Fig. 2, are shown in Fig 5. In these matching trials, the exercised matching arm was used to indicate the position of the unexercised reference. We chose to do the experiment this way around because previous experience had shown that it brought out exercise-related errors more clearly (20). The exercised indicator arm adopted a systematically more flexed position in matching the position of the reference. The trend was clearer for 30°, although it was present for 60° as well. These findings were consistent with our laboratory's earlier observations (20).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Because up to this point all of our studies had involved forearm position matching in the vertical plane, we had assumed that effort only played a role in position sense under circumstances where the gravity vector was acting (20). The experiments reported here have explored the possibility that a sense of effort may also contribute to position sense when the force of gravity is not able to provide positional cues.
We employed a three-pronged approach. We compared position sense at the elbow joint for the unsupported forearm, using the normal, hands-in-front position, with position sense when the arms were made essentially weightless, by means of counterweights. Second, because effortless movements in the vertical plane are rather unnatural, another test was sought, one in which gravity did not play a role, yet involving movements not unlike those encountered in everyday activities. We therefore measured position sense in the horizontal plane. Finally, to ensure that an influence from effort signals, if present, would be detected, we disturbed the effort-force relationship by damaging the elbow flexors with eccentric exercise. Eccentric exercise was chosen because it is the only form of exercise that reliably produces a sustained fall in muscle force. Matching accuracy was compared before and 24 h after the exercise. At 24 h, there was a significant fall in muscle force due to muscle damage (15). Such a fall would mean that a greater effort was required to maintain the unsupported position of the arm. No extra effort would, of course, be needed for the counterweighted or horizontal matching tasks. It was therefore predicted that, in these tasks, reducing muscle force with exercise would have no effect on position matching accuracy.
Because all of the postexercise measurements were made at 24 h, it meant that the muscle had long recovered from any fall in force due to metabolic fatigue. However, by 24 h, subjects had become sore; DOMS had developed in the exercised elbow flexors. It raised the possibility that, for the unsupported matching, errors could, in part, be the result of influences attributable to the soreness. It is known that subjects systematically underestimate reference isometric forces when they are matched by a sore matching arm (24). However, the experiments reported here involved only low forces in the exercised muscle, making it unlikely that DOMS contributed to matching errors. Certainly, subjects did not report any discomfort during matching procedures.
Unsupported Position Matching
The observations made on position sense with the arms-in-front, unsupported, essentially confirmed our laboratory's previous observations, as did the effects of the exercise (1, 20). Before the exercise, subjects tended to match the reference elbow angle by adopting a more extended position with their matching arm (Figs. 2 and 3). In another recent study using a similar protocol, this was attributed to the way the experiment was carried out. In that study, as well as in the experiments reported here, at the start of the matching trial, both arms lay flat on the table in front of the subject. The subject then had the reference arm moved to the test angle and held it there. The subject was asked to move the matching arm to the same angle. The movement of the reference arm from full extension to the test angle was likely to condition its elbow flexors, leading the matching arm to adopt a more extended position (25).
It was found that position errors at a test angle of 30° tended to be larger than at 60°. This was a similar trend to that reported before (20). Our laboratory has speculated previously that, as the arm was placed more nearly horizontally, a larger vector of the force of gravity would be acting on it (Ref. 20, Fig. 6). However, the situation is more complicated than that. There is an accompanying change in elbow flexors' length-tension relation as the arm is extended, and this is not paralleled by the change in torque as the lever arm changes. After making a number of assumptions we estimated that the flexor torque required to support the arm at 30° would be 1.6 times higher than at 60°.
After exercise, for the unsupported matches, errors lay in a systematically more flexed direction; that is, they were more negative (Figs. 5 and 6). Indeed, the errors were now distributed approximately uniformly about the reference angle (Fig. 6). Pooling the data confirmed these trends and showed that the exercise had led to a significant change in the distribution of the errors (Fig. 7).
Our interpretation is as follows. We believe that position sense at the more proximal joints such as the elbow arises, in part, from peripheral signals coming from muscle spindles, with a supplementary contribution from cutaneous receptors (2). In addition, positional cues are provided by the sense of effort. The effort sense is believed to be generated centrally, within the brain. So, whenever we carry out a voluntary contraction, it is postulated that a copy of the motor command reaching the motor cortex is sent to sensory areas to generate the effort sensation. For a review of this subject see McCloskey et al. (13). When subjects held their reference arm at the test angle, this would be signaled in part by the effort sensation accompanying the force generated in elbow flexors to support the weight of the arm against gravity. After the muscles of the matching arm had been damaged by the eccentric exercise, as evidenced by the persisting force deficit and the soreness, the matching arm adopted a more flexed position to match the reference angle because maintenance of a given angle against the force of gravity was associated with a larger effort signal (20). Here it is assumed that subjects attend the sense of effort generated on the reference side and place their matching arm at an angle where the efforts on the two sides match. Because the indicator elbow flexors were weaker, the arm adopted a position closer to the vertical where the gravity vector was less (20).
One possible explanation for the postexercise matching errors was that the eccentric exercise had damaged the intrafusal fibers of muscle spindles. However, it is known that muscle spindle responsiveness remains unaltered after a series of eccentric contractions (6). So, matching errors cannot be attributed to disturbed receptor properties. Another spindle-based explanation is that the increased effort required to support the arm leads to a greater level of fusimotor coactivation (18) and therefore higher spindle discharge rates. However, the evidence does not support fusimotor-evoked spindle activity as generating conscious sensations (13).
Position Sense With Counterweighted Arms
The counterweighted matching, although a somewhat unnatural task, had the advantage that it allowed a comparison with unsupported position matching under near identical conditions, with only the effort to support the arm removed. When subjects matched their counterweighted arms, the bias in position errors in the direction of extension was smaller than with unsupported matching, although still present. Presumably, overcoming inertial forces to move the counterweighted arm from the extended position produced a small amount of conditioning, although less than for the unsupported matching. As a consequence, position matching errors were more uniformly distributed about the target angle (Fig. 3).
A more important feature of the counterweighted matching was that subjects became less sure about their arm position. This showed up as a significant increase in the standard deviation of the matching errors (Fig. 4). A similar observation was made by Gooey et al. (5), who found that absolute errors in position matching were no bigger for the counterweighted condition compared with the unsupported condition. However, they too found an increase in the standard deviation. One interpretation of these findings is that a positional cue is provided by the effort accompanying the contraction supporting the arm against the force of gravity. Removing the need for such a contraction removes one source of positional information. That led to a more erratic matching performance by subjects. In a related experiment, Winter et al. (25) compared position sense at the elbow with the arm supported by the experimenter and without support. Again, position sense improved when the arm was unsupported.
After muscles of the matching arm had been damaged and weakened by the exercise, the important result was that with counterweighted matching there was no significant increase in the size of the matching errors or their distribution. The finding supports the view that, with counterweighted matching, an increase in effort from muscle damage had no effect on position matching accuracy. There was a small trend for the counterweighted matching arm to adopt a more flexed position than before (Figs. 6 and 7), although this was not significant. Perhaps this was due to the extra effort required in overcoming inertial forces.
Horizontal Position Matching
We resorted to a horizontal matching task because we considered counterweighted matching in the vertical plane as rather unnatural. In everyday activities, we routinely move our arms in the horizontal plane so we considered this a more realistic task. The pattern of errors and their variability for horizontal position matching were similar to those seen when the arms were counterweighted. There was a more uniform distribution of the errors about the target compared with unsupported position matching (Fig. 3), and the standard deviations of the errors were significantly larger (Fig. 4).
After exercise, there was no detectable change in the sizes of the position errors or their variability, although a small, nonsignificant change in position bias in the direction of flexion was apparent (Figs. 6 and 7). The lack of any effect from the fall in torque after exercise supports our hypothesis that the sense of effort only plays a role when muscle activity is required to maintain the position of the arm.
Horizontal position matching, as we have done it here, has been carried out before (3, 7, 21). The general view is that position sense with this posture arises primarily in muscle spindles. Indeed the experiments on muscle vibration support this (3, 7). It is also consistent with our views of the source of proprioceptive signals when the arms are supported (20, 25), or counterweighted, keeping in mind a possible contribution from cutaneous receptors (2).
In conclusion, we have shown that, when subjects carry out a forearm position matching task, they become more erratic in their performance if the arms are counterweighted or if the task is carried out in the horizontal plane. We propose that these effects are due to withdrawal of a positional cue normally available to subjects when matching the unsupported arms in the vertical plane. The perceived effort in supporting the arms against the force of gravity provides that cue. The closer the forearm is to the horizontal plane, the greater the gravitational vector and therefore the larger the required muscle torque (20). This leads to a larger perceived effort. Our current working hypothesis is that, during development, we learn to routinely associate effort sensations with movements as signaled by proprioceptive feedback from the moving muscles. Eventually we begin to use the sense of effort as a proprioceptive signal in its own right, much as had been proposed by Von Helmholtz (19).
Our findings on the effects of exercise are consistent with the effort hypothesis. The fall in force after exercise led to a significant change in the distribution of the errors in the unsupported matching task but not in the tasks where gravity played no role. An important question for the future is how well subjects perform in the three tasks under conditions where their arm muscles are bearing a load. Muscle spindles would be expected to become coactivated (18), and so the reliability of their positional signal would have to be reassessed. This would be expected to show up more prominently in circumstances like counterweighted matching and horizontal matching, where effort does not appear to provide positional information.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. J. Givoni, T. Pham, T. J. Allen, and U. Proske The effect of quadriceps muscle fatigue on position matching at the knee J. Physiol., October 1, 2007; 584(1): 111 - 119. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Allen, G. E. Ansems, and U. Proske Effects of muscle conditioning on position sense at the human forearm during loading or fatigue of elbow flexors and the role of the sense of effort J. Physiol., April 15, 2007; 580(2): 423 - 434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. E. Ansems, T. J. Allen, and U. Proske Position sense at the human forearm in the horizontal plane during loading and vibration of elbow muscles J. Physiol., October 15, 2006; 576(2): 445 - 455. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. G. Semmler and T. S. Miles Exercise, effort, and limb position sense J Appl Physiol, April 1, 2006; 100(4): 1099 - 1100. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, left) and another 5 trials at 60° (
, right). The position of the reference arm is given by the dotted line at zero error. When the matching arm adopted a more extended position relative to the reference arm, this was assigned a positive value, if more flexed it was negative. For unsupported matching, at both test angles the matching arm adopted a more extended position relative to the target.
, Before the exercise;