| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Journal of Neurophysiology Vol. 79 No. 5 May 1998, pp. 2265-2276
Copyright ©1998 by the American Physiological Society
1 Robert S. Dow Neurological Sciences Institute, Portland 97209; 2 Laboratory of Motor Control, Department of Kinesiology, Catholic University of Leuven, B3001 Heverlee, Belgium; and 3 Department of Physiology and Pharmacology, Oregon Health Sciences University, Portland, Oregon 97201
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Verschueren, Sabine M. P., Paul J. Cordo, and Stephan P. Swinnen. Representation of wrist joint kinematics by the ensemble of muscle spindles from synergistic muscles. J. Neurophysiol. 79: 2265-2276, 1998. Proprioceptive information about movement is transmitted to the central nervous system by a variety of receptor types, which are widely distributed among the muscles, joints, and skin. Muscle spindles are known to be an important and reliable source of information for the perception of movement kinematics. Previous studies that focused on the characteristics of single muscle spindle firing patterns have left the impression that each receptor fires in relation to a number of kinematic variables, leaving the following question unanswered: what role is played by the ensemble of muscle spindles within the same muscle or within synergistic muscles? The study described in this paper addressed whether the perception of joint position and velocity is based on the net input of muscle spindles residing in all synergistic muscles crossing a joint. Normal human adults performed a motor coordination task that required perception of joint velocity and dynamic position at the wrist. The task was to open the left hand briskly as the right wrist was passively rotated in the flexion direction through a prescribed target angle. In randomly occurring trials, the tendons to three muscles [extensor carpi radialis (ECR), extensor carpi ulnaris (ECU), and extensor digitorum (ED)] were vibrated either individually or in different combinations during the performance of the motor task. Tendon vibration is known to distort muscle spindle firing patterns, and consequently, kinesthesia. By comparing performance errors with and without tendon vibration, the relative influences of muscle spindles residing in ECR, ECU, and ED were quantified. Vibration of the individual ECR, ECU, or ED tendons produced systematic undershoot errors in performance, consistent with the misperception of wrist velocity and dynamic position. Performance errors were larger when combinations of, rather than individual, muscle tendons were vibrated. The error resulting from simultaneous vibration of ECR and ECU was roughly equal to the sum of the errors produced by vibration of the individual tendons. These effects of vibrating synergistic tendons at the wrist suggest that kinesthesia is derived from the integrated input of muscle spindles from all synergistic muscles.
The central nervous system (CNS) uses proprioceptive information to coordinate a wide variety of motor activities (e.g., Hasan 1992 A total of 14 human subjects (aged 24-47 years) with no known neuromuscular deficits participated in this study, most in more than one experiment. Each subject provided written consent for participation according to local human subjects procedures. The subjects sat at a table with an hydraulic cylinder that passively rotated the wrist, and they were instructed to open the left hand briskly when the right wrist was rotated through a prescribed target angle (see also Cordo et al. 1994 Experimental apparatus
The experimental apparatus consisted of a manipulandum table that controlled the angular position of the wrist (Fig. 1). Each subject sat at the table and inserted the right hand into a U-shaped cuff that held the wrist in the neutral position, which was defined as 0°. The upper arm was abducted ~60° from the vertical, and the forearm was oriented parallel to the table surface. The forearm was constrained so that movements were restricted primarily to the wrist joint. The wrist was passively rotated by an hydraulic actuator, and a potentiometer transduced the wrist angle. Hand opening was detected by breaking the electrical continuity through two metal contacts circling the thumb and the index finger of the left hand. The left hand was chosen for practical reasons; however, the results should not have been influenced by which hand performed the hand-opening task as previously shown in a direct comparison of ipsilateral versus contralateral hand opening (Cordo et al. 1994
Procedures
The general methodology has been reported previously and will be described briefly here (Cordo et al. 1995b
Data analysis
In the practice sessions, the average wrist angle at hand opening was calculated for each wrist velocity for each subject. Two standard measures were used to define task performance: constant error and variable error (e.g., Poulton 1974 Vibration of the extensor tendons at the wrist disrupted proprioceptive information, and consequently, subjects produced systematic errors in the motor task. Two single trials from the same subject at the same wrist velocity (15°/s) are illustrated in Fig. 3, one without vibration (A) and one with vibration of the ECR tendon of the moving hand (B). In the trial without vibration, the subject opened the contralateral hand when the wrist passed through the target angle. In contrast, when the ECR tendon was vibrated, the subject opened the hand ~400 ms early, resulting in a 7.5° undershoot error. Undershoot errors are consistent with the perception that the joint is more flexed and rotating faster than it actually is (e.g., Cordo et al. 1995b
Practice and accuracy without tendon vibration
The subjects' performances with the motor task improved across the three practice sessions (i.e., without tendon vibration). The height of each bar in Fig. 4 represents the average error for nine subjects. In each group of three bars, the left bar represents the error from the first practice session, the middle bar the error from the second practice session, and the right bar the error from the third practice session.
Effect of ECR, ECU, and ED vibration
Tendon vibration produced systematic errors in subjects' performance of the task. Vibration of the individual ECR, ECU, or ED tendons during performance of the motor task resulted in undershoot errors, although the size of error depended on the muscle vibrated. Vibration of each of the three tendons significantly affected the constant error compared with the control trials (contrasts: ECR: F[1,12] = 15.52, P < 0.005; ECU: F[1,12] = 35.89, P < 0.0001; ED: F[1,12] = 55.62, P < 0.0001). In Fig. 6, negative values on the ordinate represent mean constant error in the undershoot direction. As the velocity of wrist rotation increased, the undershooting caused by vibration decreased (e.g., Cordo et al. 1995b
Vibration cross talk and differences in the two vibrators
Vibration of the ED tendon produced significantly larger undershoot errors compared with ECR or ECU vibration (see Figs. 6 and 7). One possible cause of this disparity is that vibration of one tendon was transmitted to tendons of synergistic muscles on either side through bone and soft tissues. To rule out this possibility, afferent responses were recorded and failed to show significant evidence of cross talk (see Fig. 2). We acknowledge that afferent responses were obtained in this control with the wrist stationary rather than during wrist rotation, as during the behavioral task; however, there is no obvious reason why vibration should spread more easily during movement compared with static conditions.
Vibration and skin stretch at the wrist and hand
In the motor task performed in this study, vibration of the ED tendon at the wrist produced errors in the perception of wrist angle and velocity. To understand more fully why vibration of the tendons to finger extensors should influence wrist kinesthesia, two control experiments were carried out.
The CNS uses the information provided by proprioceptors to coordinate a wide variety of motor activities (Hasan 1992 Use of proprioceptive input in the coordination of the wrist-hand task
In previous studies (Cordo 1990 Integration of proprioceptive information
The hypothesis underlying this study was that kinesthesia originates from the integrated input of the receptor ensemble among synergistic muscles at a joint. This hypothesis is based on the idea that kinematic information represented by the receptor ensemble is not present at all, or is present with lower precision, in the firing patterns of individual receptors. A number of previous studies have hypothesized that both motor output and somatosensory input are based on ensembles of neurons.
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
; McCloskey and Prochazka 1995; Rothwell et al. 1982), and an important source of this information is muscle spindle primary and secondary afferents (Burgess et al. 1982; Gandevia and Burke 1992; Matthews 1982, 1988). Input from proprioceptive afferents informs the CNS about which joints are rotating, and the movement direction and kinematics at each joint. The identification of which joints are rotating presumably arises from the anatomic separation of afferent pathways in the periphery and the somatotopic organization of somatosensory cortex (Erickson 1968; Kaas et al. 1979). The identification of movement direction has been shown to arise from the integrated input from muscle receptors in orthogonally and antagonistic muscles at each joint (Gilhodes et al. 1986; Roll and Gilhodes 1995). The peripheral coding of direction has been hypothesized to be based on a population vector (Roll and Gilhodes 1995), similar to that described for directionally sensitive neurons in motor cortex and other areas of the brain (e.g., Kalaska et al. 1983). In contrast, relatively little is known about how kinematic information is extracted from the population of proprioceptive afferents (Burgess et al. 1982; Matthews 1988).
,b; Houk et al. 1981; Hulliger et al. 1982; Wei et al. 1986; see, however, Bergenheim et al. 1995; Prochazka et al. 1989). The characteristics of single muscle spindle firing patterns (e.g., Houk et al. 1981) leave the impression that each receptor fires in relation to a number of kinematic variables, including muscle length and velocity, which raises the question of what role is played by the hundreds or thousands of similar receptors within the same muscle or across synergistic muscles (Banks and Stacey 1988; Chin et al. 1962; Voss 1971).
), there are a number of reasons to suggest that kinesthesia is derived from the ensemble of afferents responding to joint rotation. Each muscle spindle afferent in a given muscle has a relatively individualistic response to the same muscle stretch (e.g., Burgess et al. 1982). Similarly, joint rotations often stretch groups of synergistic muscles with somewhat different mechanical actions, which could lead to different average response characteristics from the receptor population of each muscle. Finally, the response of each individual receptor includes a number of nonlinear characteristics, including rate sensitivity, hysteresis, adaptation, and thixotropy (e.g., Burgess et al. 1982; Gregory et al. 1988; Proske et al. 1993). The ensemble response could provide a means for the CNS to derive an accurate sense of kinesthesia from a population of idiosyncratic, nonlinear receptors (Clark 1992; Cordo et al. 1994; Hall and McCloskey 1983).
) or cutaneous receptors (Clark et al. 1985; Collins and Prochazka 1996; Ferrell and Milne 1989), to correctly interpret sensory input from multiarticular muscles. Consistent with this idea, vibration of tendons to multiarticular muscles has been reported to produce quite different effects, depending on whether or not one of the joints crossed by the muscle was being rotated while the vibration was being applied (Burgess et al. 1982).
; Cordo et al. 1994, 1995a,b). In this task, the subject indicates with a ballistic finger movement when another passively rotating joint, such as the elbow, passes through a prescribed target angle. Because of neuromuscular delays and the absence of visual feedback, the accurate performance of this task requires the use of proprioceptive information related to the dynamic position and velocity of the passively rotated joint.
; Burke et al. 1976). The purpose of the experiment described in this paper was to determine whether the CNS uses proprioceptive input from all synergistic muscles at a joint, both monoarticular and multiarticular, to construct a perception of joint kinematics. We hypothesized that the CNS uses all proprioceptive information from synergistic monoarticular and multiarticular muscles to determine accurately the kinematics of movement, and that cutaneous input is used to eliminate ambiguity from sensory information provided by receptors in multiarticular muscles.
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
). In preselected trials, vibration was applied transversely to the tendons of several extensor muscles crossing the wrist joint. The kinematics of wrist rotation, time of hand opening, and in some subjects, electromyographic (EMG) activity were recorded. The importance of proprioceptive input from each muscle was inferred from the influence of tendon vibration on the subject's accuracy with the task.
).
View larger version (19K):
[in a new window]
FIG. 1.
Experimental setup. A: subject sits at the manipulandum table with the right hand in a wrist rotator and an electrical continuity detector on the left thumb and index finger. B: 3 vibrators are applied to the wrist extensor tendons, 2 small vibrators on the extensor carpi radialis (ECR) and extensor carpi ulnaris (ECU) and a large, servo-controlled vibrator on the extensor digitorum (ED).
). The motor task was to open the left thumb and index finger briskly when the right wrist was passively rotated in the flexion direction through a prescribed target angle. The starting angle was 0°, the target angle was 25°, and the final angle was 35°. The velocity of wrist rotation (15, 20, 25, 30°/s) was constant in each trial and randomized across trials, so subjects were unaware of the velocity of the upcoming trial. Without vision of the movement nor the ability to predict the wrist velocity, subjects were left with proprioception as the only useful source of information to control this motor task.
).
. With all three vibrators, the vibration amplitude was set to 0.7-1 mm and the frequency to 60 Hz. The static pressure of the probes on the tendons was adjusted to evoke the most pronounced sensation of displacement and movement with the wrist at the start position. In most subjects, vibration of the ECR tendon alone, with the wrist held in the start position, produced an illusion of pronation/flexion of the wrist, vibration of ECU produced an illusion of supination/flexion, and vibration of ED tendon alone produced an illusion of pure wrist flexion. In vibration trials during the experiment, the onset of vibration occurred 2 s before the onset of wrist rotation.
) were used to record from single muscle spindle afferents in three subjects (Fig. 2). The purpose of this experiment was to ensure that the effects of vibrating the tendon to a particular muscle were largely confined to the sensory receptors in that muscle, i.e., that vibration of one tendon did not activate muscle spindle receptors in neighboring synergistic muscles through cross talk. The activity of six muscle spindle afferents (2 from ED and 4 from ECR) was recorded during 60-Hz vibration of ED, ECR, or ECU.
View larger version (36K):
[in a new window]
FIG. 2.
Control experiment to measure vibration "cross talk." A-D: nerve recordings from an ED muscle spindle afferent in the radial nerve: with no vibration (A), ED vibration (B), ECR vibration (C), and ECU vibration (D). E: displacement of the tendon vibrator (60 Hz).
View larger version (13K):
[in a new window]
FIG. 8.
Comparison of effects with 2 tendon vibrators. The height of each bar represents the average amount of undershoot caused by tendon vibration of the ECR or ED. Filled bars indicate vibration was produced with the servo-controlled vibrator and unfilled bars indicate vibration was produced with the small vibrator. Data from 2 different subjects are shown in A and B.
). Constant error was defined as the difference between the target angle and the wrist angle at hand opening for each test condition in each subject. Variable error was defined as the standard deviation of the constant error for each test condition in each subject. The grand means (SDs) of the constant and variable errors were determined for all nine subjects. The effect of practice was defined by the change in error across the three practice sessions. The slope of the linear regression of the relationship between wrist velocity and constant error was calculated for each session. The value of this slope reflects the use of velocity information in the control of this task; lower values of slope indicating more effective use of velocity information (Cordo et al. 1994). Repeated measures analysis of variance (ANOVA) was used in the statistical analysis of practice session data. The ANOVA on the constant error data were performed on the absolute value of constant error (i.e., "absolute error") (Poulton 1974), because negative and positive constant errors would otherwise cancel each other out, thereby hiding the learning effect.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
; Goodwin et al. 1972; Sittig et al. 1985, 1987). The EMG traces of the ECR shown in Fig. 3 show that the subject was generally relaxed as instructed, although a small stretch reflex is evident in later phases of wrist rotation.
View larger version (22K):
[in a new window]
FIG. 3.
Effect of ECR vibration in a single trial. A: control trial shows hand opening when the wrist passes through the target angle. B: vibration trial shows an undershoot error. In top records, stippled line represents target angle at 25°. Hand opening is represented by a discrete change in voltage from 0 to 5 V. Stippled line representing vibration in B shows the onset and offset timing. The electromyographic (EMG) record shows a weak stretch reflex produced by passive wrist rotation at 15°/s.
View larger version (21K):
[in a new window]
FIG. 4.
Reduction of error with practice. A: constant error (as defined in METHODS). B: variable error (as defined in METHODS). Each bar represents the grand mean of the error in 9 subjects. In each group of 3 bars, the left bar represents the 1st practice session, the middle bar the 2nd practice session, and the right bar the 3rd practice session. Each group of 3 bars represents the error in trials at a single wrist velocity (i.e., 15, 20, 25, or 30°/s). The regression lines in A are plotted to the 4 grand mean values for a given practice session.
The triggering angle represents the wrist position, at a particular velocity, at which the triggering process must be initiated for the hand to open when the wrist reaches the target angle. The triggering angle is close to the target angle in slow wrist rotations, and it is farther away in fast wrist rotations. The neuromuscular delay is ~200 ms (Table 1), based on the simple proprioceptive reaction time measured by asking subjects to open the left hand as soon as they detected the onset of a 30°/s rotation of the right wrist (see also Cordo et al. 1994
(1)
). Reaction time was defined as the delay between the onset of right wrist rotation and the break in electrical contact between the left thumb and index finger. If wrist velocity were not used for prediction, the subject would undershoot the target angle at slow velocities and overshoot at higher velocities (Cordo et al. 1994). Thus, if the subject uses velocity information to predict accurately the triggering angle and dynamic position information to detect when the triggering angle has been reached, the slope of the relationship between velocity and constant error should be zero. When the slope was steeper, the subject used less velocity information to predict how far the wrist would rotate during the neuromuscular delay.
View this table:
TABLE 1.
Proprioceptive reaction time at the wrist
View this table:
TABLE 2.
Slope of linear regression of constant error versus velocity over three practice sessions
). In the elbow-hand task, the variable error decreased, but the constant error did not. Figure 5 compares subjects' performances in the elbow-hand (Cordo et al. 1994) and wrist-hand tasks over comparable velocity ranges. In Fig. 5A, which shows data from the first practice session, both positive and negative constant errors (data points) were considerably larger in the wrist-hand task (
) compared with the elbow-hand task (
). As also shown in Fig. 4, the slope of the relationship between constant error and velocity is relatively high in the wrist-hand task, indicating that velocity information was not effectively used. During the first practice session (Fig. 5A), the variable error (error bars) was also larger in the wrist-hand task. In Fig. 5B, which shows data from the third practice session, the constant error was roughly the same for the two tasks, and the variable error (and slope) in the wrist-hand task decreased to a level only marginally greater than that in the elbow-hand task. Thus, before practice, precision was higher in the elbow, whereas after practice it was comparable in the two joints.
View larger version (18K):
[in a new window]
FIG. 5.
Performance error in the wrist-hand and elbow-hand task. A: constant error is contrasted for the wrist-hand task ( ; 9 subjects) and elbow-hand task (; 9 subjects) in the 1st practice session. B: constant error is contrasted for the 3rd practice session. Error bars represent the SD of the subject means.
). On the average, vibrating the ECR () or ECU (
) tendons produced undershooting errors within a range of 2-3°, whereas vibrating ED (
) produced undershooting errors within a range of 3-6°. The effects of ECR and ECU vibration were not significantly different from each other (P > 0.05), although both differed significantly from the effect of ED vibration (ECR: F[1,11] = 5.42, P < 0.05; ECU: F[1,11] = 5.35, P < 0.05).
View larger version (20K):
[in a new window]
FIG. 6.
Effect of vibrating tendons to individual wrist extensors. Lines represent linear regression fits for trials with vibration of ECR ( ), ECU (), and ED (). Error bars represent the SD of the subject means (n = 14). Negative values on the ordinate represent undershoot errors.
View larger version (38K):
[in a new window]
FIG. 7.
Effects of vibrating the tendons to individual and combinations of muscles. Each bar represents the mean undershoot error for 14 subjects. The effect of vibrating combinations of muscles is indicated with the ampersand symbol (e.g., ECR&ECU). The mathematical sum of individual effects is indicated with the plus symbol (e.g., ECR+ECU). Each bar graph represents the undershoot errors at a different wrist velocity.
; Sittig et al. 1985), so that the 8° maximum shown in Fig. 7 for vibration of all three wrist tendons might represent a saturation effect. As with vibration of individual muscles (Fig. 6), the effect of vibrating a combination of tendons decreased with increasing velocity (main effect for velocity in 5 vibration conditions × 4 velocities ANOVA: F[3,33] = 31.65, P < 0.0001).
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
; McCloskey and Prochazka 1995; Rothwell et al. 1982), and muscle spindles are likely to be an important source of this information (Burgess et al. 1982; Gandevia and Burke 1992; Matthews 1982). The importance of muscle spindle input in kinesthesia has been inferred from the proprioceptive illusions evoked by tendon vibration (Goodwin et al. 1972; Sittig et al. 1985) and from the close relationship between muscle spindle firing patterns and movement kinematics (Crowe and Matthews 1964a; Houk et al. 1981; Matthews 1963).
; Cordo et al. 1994, 1995a), it was shown that the CNS uses proprioceptive information related to both the dynamic position and velocity of elbow rotation to trigger a sequential movement of the hand within an elbow-hand movement task. The study reported in this paper involved the identical motor task with the exception that a passive wrist rotation, rather than elbow rotation, preceded a coordinated movement of the hand. Although the main purpose of the study described in this paper was to address the representation of movement kinematics by the ensemble of proprioceptors from synergistic wrist extensors, it seemed relevant to compare the performance characteristics in the wrist-hand task with those in the elbow-hand task, because this is the first published report of the wrist-hand task, and because proprioception is more precise at proximal compared with distal joints in the upper and lower extremities (e.g., De Dominico and McCloskey 1987; Hall and McCloskey 1983; McCloskey 1993).
). The only learning observed in the elbow-hand task was a small decrease in the variable error at the fastest velocity of elbow rotation, and only during the 1st 30 trials of the 1st practice session. In the wrist-hand task, however, the accuracy increased significantly over three practice sessions as shown by a decrease in constant error (Fig. 4). Practice was also associated with a significant decrease in the slope of the relationship between the hand opening angle and velocity of wrist rotation (Table 2), which suggests that practice allowed subjects to use velocity information more effectively. At the end of the three practice sessions, there was little difference in the performance of the wrist-hand and elbow-hand tasks (Fig. 5). The constant error was nearly identical for both joints, and the variable error was only 10-20% larger in the wrist-hand task (Fig. 5). The larger variable error in the third practice session with the wrist-hand task might have been caused by the cast and tendon vibrators on the forearm, and with additional practice it might have decreased further.
). In contrast, larger errors were observed in centrally controlled throwing movements when throwing was performed with the nondominant hand (Hore et al. 1996). Previous studies have demonstrated that proprioceptive acuity is better in proximal joints compared with distal joints of the extremities. This trend for proprioceptive acuity has been demonstrated for the detection of movement (Goldscheider 1889; Hall and McCloskey 1983; Refshauge et al. 1995) and for the perception of static position (Clark 1992; De Dominico and McCloskey 1987), and it has been hypothesized to result from peripheral factors such as the distributions of muscle spindles (Scott and Loeb 1994) or muscle length (Hall and McCloskey 1983; McCloskey 1993). However, the results of the study reported in this paper suggest that this proximal-to-distal gradient in proprioceptive acuity might not be exclusively related to peripheral factors, because this gradient disappeared in the wrist after practice. This gradient might also be related to central factors. Practice-related changes in proprioceptive acuity are possible, because the cortical representations of the digits have been shown to increase dynamically as a result of practice (Nudo et al. 1996), whereas these representations decrease in size when sensory input from that part of a limb diminishes (Merzenich et al. 1990). The subjects participating in the study reported in this paper were subjected to a total of 420 trials over 4 practice and experimental sessions, which took place over a total period of 11/2-2 h. Such extensive experience with the motor task could have led to improved sensory acuity; however, a more comprehensive learning study of proprioceptive acuity at different joints would be desirable.
, 1995; Georgopoulos et al. 1986; Schwartz 1992, 1993), premotor cortex (Caminiti et al. 1991; Fu et al. 1993, 1995), parietal cortex (Kalaska et al. 1983), and cerebellum (Fortier et al. 1989). According to Schwartz (1993), movement velocity as well as direction is represented in the vectorial population code of motor cortex cells.
,b) showed that an ensemble of cutaneous afferents in the raccoon is needed to represent unambiguously the intensity and location of tactile stimuli. Crago et al. (1982) and Hulliger et al. (1995) showed in the cat that populations of Golgi tendon organs represent more reliably the total active force produced by a muscle than individual tendon organs. Bergenheim and co-workers (1995) found that a population of muscle spindle afferents in the cat discriminated better among different speeds and amplitudes of muscle stretch than did individual afferents. Interestingly, this difference between single units and ensembles disappeared after cutting the ventral root, which led to the hypothesis that the gamma system plays an important role in ensemble coding by muscle spindles. Fusimotor activity appears to decorrelate the firing patterns of individual muscle spindles, which enhances the information transmission by ensembles of muscle spindle afferents (Bergenheim et al. 1995; Inbar et al. 1979; Johansson et al. 1995; Milgram and Inbar 1976).
concluded that the CNS must continuously integrate the input from all muscle spindles involved in a movement to decode movement direction. Similarly, vibration of antagonistic muscles can change or completely eliminate the vibration-induced illusion of joint rotation (Gilhodes et al. 1986). The observation that stimulation of a single muscle spindle afferent did not evoke a perception (Macefield et al. 1990) demonstrates that the brain requires input from more than one muscle spindle afferent to perceive movement.
).
; Simon et al. 1982). Activation of ED muscle spindles in the absence of ECR and ECU activity could signal pure finger flexion, and activation of ECR and ECU muscle spindles in the absence of ED activity could signal simultaneous wrist flexion and finger extension at a velocity that exactly counterbalances the stretch evoked by wrist rotation. However, muscle afferent input alone seems insufficient to differentiate all possible combinations of wrist and finger motion. Activation of ECR and ECU muscle spindles in combination with those from ED does not always signal a pure wrist movement, because part of ED muscle spindle activity can be caused by additional finger movement. The illusion of wrist flexion produced by ED vibration in the stationary wrist would seem to require information from some other source, because in normal movement conditions, wrist flexion would never occur without the activation of ECR and ECU muscle spindles.
; Collins and Prochazka 1996; Ferrell and Milne 1989). One control experiment showed that skin stretch on the dorsum of the hand or the dorsum of wrist is a good predictor of finger versus wrist flexion. Thus, during normal movement, cutaneous information from these two sites could clarify exactly where movement is occurring. In the second control experiment, vibration of the ED tendon at these two locations activated different groups of cutaneous and intrinsic afferents under the vibrator probe while evoking completely different illusions. With vibration applied to the index, middle, and ring finger tendons 2-cm distal to the wrist, the illusion was finger flexion at the metacarpophalangeal joint; with vibration applied to the dorsum of the wrist, the illusion was wrist flexion. Vibration 2-cm distal to the wrist might have spread to the interossei muscles, but the activation of afferents from these muscles would be expected to evoke primarily illusions of lateral finger movement, with perhaps some finger flexion at the interphalangeal joint. Vibration at this site might also have spread to extensor indicis and extensor digiti minimi, but these are long-tendon muscles located in the forearm, and afferents from these muscles would not be able to disambiguate wrist flexion from finger flexion.
), the CNS is informed of the precise location of the stimulated point. Microneurographic studies on palmar skin afferents (Burke et al. 1988) and skin afferents on the dorsum of the hand (Edin 1992; Edin and Abbs 1991) have shown that these afferents can provide the information necessary to differentiate wrist from finger motion. Alternatively, cutaneous input could serve to potentiate proprioceptive input in a manner similar to that previously described in experiments in which proprioceptive acuity from muscle receptors was reduced by cutaneous anesthesia (Goodwin et al. 1972; Marsden et al. 1979).
proposed that the velocity of perceived movement can be represented by the length of the vector sum of discharge from all muscle spindles in all muscles acting at a joint; however, these authors did not take into account that some of the muscle spindles activated by tendon vibration were located in multiarticular muscles. In the case of multiarticular muscles, the CNS must be able to distinguish activity evoked by movement of the different joints crossed by the muscle to decode movement velocity (or direction) at any one particular joint. If movement velocity is represented by the length of the vector sum, it must be weighted with respect to the contribution of each particular muscle according to its mechanical action, especially in the case of multiarticular muscles. Our result showing the summation of vibration effects (Fig. 7) demonstrates that, on a perceptual level, there is a weighting of the inputs from synergistic muscles, possibly supported by input from cutaneous afferents. These results suggest that models of sensory coding should take into account the specific actions of muscles as well as all receptor types responding to the movement.
; Scott and Loeb 1994). From this perspective, the location and the actual number of the active sensory afferents has no meaning to the CNS, as in a pure frequency code. However, in a pure frequency code, different stimuli can produce the same average response (Ray and Doetsch 1990a,b), which makes the average response ambiguous, as suggested by Burgess et al. (1982) for the representation of static position. A more plausible hypothesis is that the input of each individual afferent is significant (Burke and Gandevia 1995), and that the CNS uses a recruitment code in addition to a frequency code. This hypothesis has been proposed by Ray and Doetsch (1990a,b) for cutaneous afferents where each "across fiber response" has a specific shape according to the units that are active to represent stimulus location, and the average discharge frequency across the receptor population represents stimulus intensity (Erickson 1974). Given the requirement for proprioceptive afferents to encode dynamic position and velocity in motor tasks such as the wrist-hand and elbow-hand tasks, a combination of recruitment and frequency codes might be required by the CNS to encode multiple kinematic variables simultaneously and independently.
| |
ACKNOWLEDGEMENTS |
|---|
This work was supported by National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant R01 AR-31017. S.M.P. Verschueren is a Research Associate of the Fund for Scientific Research, Flanders, Belgium.
| |
FOOTNOTES |
|---|
Address for reprint requests: P. J. Cordo, Robert S. Dow Neurological Sciences Institute, 1120 NW 20th Ave., Portland, OR 97209.
Received 18 August 1997; accepted in final form 21 January 1998.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Dimitriou and B. B. Edin Discharges in human muscle spindle afferents during a key-pressing task J. Physiol., November 15, 2008; 586(22): 5455 - 5470. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Sturnieks, J. R. Wright, and R. C. Fitzpatrick Detection of simultaneous movement at two human arm joints J. Physiol., December 15, 2007; 585(3): 833 - 842. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. P Westlake, Y. Wu, and E. G Culham Sensory-Specific Balance Training in Older Adults: Effect on Position, Movement, and Velocity Sense at the Ankle Physical Therapy, May 1, 2007; 87(5): 560 - 568. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. F. Collins, K. M. Refshauge, G. Todd, and S. C. Gandevia Cutaneous Receptors Contribute to Kinesthesia at the Index Finger, Elbow, and Knee J Neurophysiol, September 1, 2005; 94(3): 1699 - 1706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Cordo, C. Flores-Vieira, S. M. P. Verschueren, J. T. Inglis, and V. Gurfinkel Position Sensitivity of Human Muscle Spindles: Single Afferent and Population Representations J Neurophysiol, March 1, 2002; 87(3): 1186 - 1195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. E Jones, J. Wessberg, and A. B Vallbo Directional tuning of human forearm muscle afferents during voluntary wrist movements J. Physiol., October 15, 2001; 536(2): 635 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D F Collins, K M Refshauge, and S C Gandevia Sensory integration in the perception of movements at the human metacarpophalangeal joint J. Physiol., December 1, 2000; 529(2): 505 - 515. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
