Detections of movements imposed on finger, elbow and...

1983;335;519-533 J. Physiol.

L A Hall and D I McCloskey

shoulder joints.Detections of movements imposed on finger, elbow and

This information is current as of January 15, 2008

This is the final published version of this article; it is available

[email protected] reproduced without the permission of Blackwell Publishing: articles are free 12 months after publication. No part of this article may

The Journal of Physiology Online. http://jp.physoc.org/subscriptions/ go to: The Journal of Physiology OnlineTo subscribe to

Physiological Society. It has been published continuously since 1878. is the official journal of TheThe Journal of Physiology Online

at QUEEN'S UNIVERSITY on January 15, 2008 jp.physoc.orgDownloaded from

http://jp.physoc.org/subscriptions/

mailto:[email protected]

http://jp.physoc.org

for 12 months after publication unless article is open access. This version of the article may not be posted on a public website

http://jp.physoc.org/cgi/content/abstract/335/1/519

at:

[email protected] reproduced without the permission of Blackwell Publishing: articles are free 12 months after publication. No part of this article may

The Journal of Physiology Online. http://jp.physoc.org/subscriptions/ go to: The Journal of Physiology OnlineTo subscribe to

Physiological Society. It has been published continuously since 1878. is the official journal of TheThe Journal of Physiology Online


http://jp.physoc.org/cgi/content/abstract/335/1/519

http://jp.physoc.org/subscriptions/

mailto:[email protected]


J. Physiol. (1983), 335, pp. 519-533 519With 5 text-figures

Printed in Great Britain

DETECTIONS OF MOVEMENTS IMPOSED ON FINGER, ELBOW ANDSHOULDER JOINTS

BY LESLEY A. HALL AND D. I. McCLOSKEYFrom the School of Physiology and Pharmacology, University of New South Wales,

Kensington, Sydney 2033, Australia

(Received 2 March 1982)

SUMMARY

1. The angular displacements necessary for 70O correct detection were determinedin normal subjects at the shoulder and elbow joints and at the terminal joint of themiddle finger. Angular velocities of displacement between 0.1250 and 160'/s weretested. Each joint was tested in the mid-range of its normal excursion. The jointswere carefully supported for testing and the muscles acting at the joints were relaxed.

2. When assessed in terms of angular displacements and angular velocities,proprioceptive performance at the shoulder and elbow joints was superior to that atthe finger joint. Optimal performance at the finger joint was attained over the rangeof angular velocities from 100 to 800/s. Optimal performance at both more proximaljoints was optimal over a wider range (20-800/s). Active pointing movements madewithout vision of the moving part were performed at each joint at velocities withinthe range of optimal proprioceptive performance.

3. However, when detection levels and displ cement velocities were expressed interms of linear displacements and velocities at the finger tip for all three joints, thefinger joint gave the best performance and the shoulder the worst. In practical terms,therefore, displacements of a given linear extent are best detected if they move distaljoints and worst if they move proximal joints.

4. For the elbow and finger joints the detection level and velocity data wereexpressed also in terms of proportional changes in the lengths of muscles operatingat these joints, and as proportional changes in the distance between the points ofattachment of the joint capsules. Analysis in terms of proportional changes of musclelength showed remarkably similar performance levels at both joints. This suggeststhat intramuscular receptors are important determinants of proprioceptive perfor-mance. Analysis in terms ofjoint capsular displacement did not unify the data: on thisform of analysis proprioceptive performance at the elbow joint was superior.

INTRODUCTION

A simple and widely used test of proprioceptive sensibility is to impose movementsof constant angular velocity on a joint and to ask the subject to indicate when heis sure that a movement has occurred. The test was introduced by Goldscheider in1889, and has been variously modified. Sometimes the subject is required to make



L. A. HALL AND D. 1. McCLOSKEYhis detection only when he knows the direction, rather than simply the existence,of imposed movement. This is because it is known that movements can be perceivedbefore their directions are known (Goldscheider, 1889; Laidlaw & Hamilton,1937a, b; and this paper), and it has been argued that only when an awareness ofboth movement and direction is required can tests be regarded as specific forproprioceptive mechanisms (McCloskey, 1978).

In clinical practice the magnitude of displacement before detection is judgedagainst performance on the opposite side of the body (when that is normal), or againstthe examiner's clinical experience of normality. In the experimental situation theresults are usually expressed as the detection threshold, the angular displacementoccurring before detection. In both cases observations are usually confined to slowvelocities of imposed angular rotation to avoid difficulties related to the subject'sresponse time being large in comparison with the time for actually making thedetection. Goldscheider (1889), and later Cleghorn & Darcus (1952), overcame thisproblem by imposing rotations of known amplitude at a given velocity and allowingan interval after conclusion of the movement for the subject to indicate detectionof movement and direction. A similar method was used here.

Goldscheider's (1889) experiments demonstrated that detection thresholds forproximal joints are lower than those for more distal joints. This finding is quotedwidely in modern textbooks of physiology. However, at least some of the experimentswhich led Goldscheider to this conclusion were carried out without requiring thesubject to state the direction of imposed movement, and so have been criticized on

the grounds that they did not test specifically proprioceptive mechanisms (McCloskey,1978). Moreover, the subsequent experiments of Laidlaw & Hamilton (1937 a, b), inwhich detection of both movement and direction was required, showed no consistentdifferences between proximal and distal joints.The present study examined the detection thresholds for rotations imposed on the

shoulder joint, the elbow joint and the terminal joint of the middle finger, over a widerrange of angular velocities than has been reported by previous workers. Part of thework has been reported in abstract form (Hall & McCloskey, 1981).

METHODS

Determinations were made of the 700 detection levels for angular rotations imposed on the finger,elbow and shoulder joints of ten healthy, adult male and female subjects, including one of theauthors (D.I.McC.). Seven of these subjects had participated in the experiments described in theaccompanying paper (Gandevia, Hall, McCloskey & Potter, 1983), and the relevant data obtainedin that study for the distal interphalangeal joint of the middle finger are included again here,together with similar data from the three additional subjects. Data from all ten subjects were

obtained in this study for the elbow joint and the shoulder joint.During all tests the subjects were instructed to relax the muscles at the test joint completely.

The subject's view of the test joint and of the apparatus was blocked, and the apparatus madeno detectable sound when operating.

Apparatus and calibrationDisplacements of the test joints were made by a large electromagnetic vibrator (Advanced

Dynamics Industries Vibrator AV-50 with oscillator-amplifier N-300) driven by a variable ramp

generator with position feedback. Position was recorded by displaying on a storage oscilloscope or

on a pen recorder the output of a position transducer (Schaevitz GPM-101) coupled to the displacingrod.

520



PROXIMAL AND DISTAL JOINTS 521

Fig. 1 is a diagram of the apparatus and postures used for the three joints. The terminal jointof the middle finger was tested, as described in the accompanying paper (Gandevia et al. 1983),by immobilizing the hand in a posture in which all joints except the test joint were held straight.With the subject's muscles relaxed movements could be imposed with ease on the test joint. Testingwas performed with the subject seated, the upper arm and forearm resting on a support, the wriststraight and the hand at shoulder level. The test joint was set at 300-400 into flexion for testing.

een: ~~ ~~~~~~A OA X

Finger Elbow Shoulder

Fig. 1. Apparatus and positions used for determining detection levels for movements ofthe terminal joint of the middle finger, the elbow joint and the shoulder joint. The leftpanel shows, above, the board upon which the hand was fixed for testing the finger jointand, below, the hand placed on the board with the distal phalanx of the middle fingerprotruding beyond it, and fixed to a rod moved by an electromechanical vibrator. Thecentre panel shows, above, the carrier upon which the forearm was rested for testing theelbow joint and, below, the arm in the carrier with the axes of the elbow and of the carrier(A) aligned, and the carrier attached to a rod from the electromechanical vibrator. Theright panel shows, above, the carrier upon which the arm was rested for testing theshoulder joint and, below, the arm in the carrier with the axes of the shoulder (X) andcarrier (A) aligned as closely as possible. Testing was carried out with the arm in position I,which was more comfortable than position II.

Coupling of the finger to the vibrator was by means of a stirrup arrangement attaching to a firmrubber 'thimble' which fitted over the fingertip and extended proximally no further than the baseof the fingernail. Angular calibration was made using a light wire pointer on the tip of the fingerand a large protractor aligned directly beneath it.

The elbow joint was tested while the forearm was supported in a metal splint (Fig. 1). This splintwas mounted on a movable board at one end of which there was an axis which was fixed to a tabletop. The carrier moved in an arc about the axis supported by a small wheel (36 cm from the axis)running freely on the smooth table top. The apparatus was moved by the electromagnetic vibratorwhich was coupled to the carrier 21 cm from the axis. The subject's bare, pronated forearm waslaid in the trough-like metal splint on the carrier with the axis of elbow joint rotation aligned withthe carrier's axis of rotation. The shoulder joint and the elbow joint were set in the same plane.The forearm and hand were wrapped into the splint using a gauze bandage. The upper arm wassupported on sandbags. The initial angle of the elbow was set at 90°, and rotations into flexion andextension were imposed in the horizontal plane. Calibrations were made by placing a largeprotractor under the carrier with its centre beneath the axis of rotation.The carrier used for imposing movements on the elbow joint was modified to support the arm

for movement at the shoulder joint. The upper arm and forearm were secured into a long trough-like



L. A. HALL AND D. 1. McCLOSKEY

splint. Initially the elbow was fixed in extension for this (Fig. 1: position II), but most subjectsfound that this posture became uncomfortable as the experiment proceeded. Therefore, all testswere performed with the elbow fixed in 900 of flexion (Fig. 1: position I), the forearm pronated,and the upper arm, forearm and hand wrapped into the splint using a gauze bandage. The carriermoved about a fixed axis, as before, but in this series of experiments it was impossible to place theaxis of rotation of the shoulder joint directly over the axis of the carrier's rotation: therefore thelatter was set 11 cm distal to the true axis of shoulder rotation. This imposed no obvious distortionsof the tissues around the shoulder for the quite small angular rotations used, nor could the subjectsdetect that the axes were slightly misaligned even when permitted to make much larger, activemovements at the shoulder joint. The values for measured detection levels and velocities wereconverted from rotations of the carrier about its axis into displacements at the shoulder joint. Fortesting, the shoulder was set at right angles to the coronal plane of the body, and the arm movedon its carrier in a horizontal plane.

Testing proceduresDetection levels for each subject at each joint were determined over a range of velocities of

angular rotation. Each joint was tested in a separate 2-3 h session. The finger was tested at 1-25°,25°, 5°, 100, 200, 400, 800 and 1600/s; the elbow at 0-1560, 0-313°, 0.6250, 1-250, 250, 50, 200 and80'/s; and the shoulder at 0-125°, 0-250, 050°, 1°, 2°, 40, 160 and 640/s. The 700% detection levelswere determined by a method similar to that used by Goldscheider (I1889), Cleghorn & Darcus (1952)and Gandevia et al. (1983). Successive sets of movements of the same angular extent were imposedon the test joint at a given angular velocity until the smallest amplitude was found at which thesubject could detect 700% or more displacements. Each set of imposed movements comprised arandom mix of ten flexions and ten extensions and the subject was required to specify the directionof movement correctly within 3 s of its completion to be scored as having detected it. The stepsof amplitude for sets of movements used in testing the finger joint were 0.250, 0 50, 1°, then 1°increments to the maximum possible. For the elbow and shoulder joints they were 0-10, 0.20, 0Q4°,0.60, 1, then 0.50 steps to the maximum possible. If a subject failed to score 70% correct detectionsat the maximum excursion possible, but had made some correct detections at the angular velocitybeing tested, the threshold was recorded as the next increment above the maximum. This wasnecessary here only sometimes, and only for the elbow and shoulder joints (as 20 was the maximumexcursion possible for elbow and shoulder, 2.50 was recorded when necessary). This practice is likelyto have led to an underestimation of the 700% detection level on the relatively few occasions itwas employed: it did, however, provide data on occasions when the apparatus could not producesufficiently large excursions to allow usual procedure to be followed.

RESULTS

At close to the criterion level of 70 0 correct, movement of a joint is usually clearlyperceived even when its direction is not. This was reported by all subjects. For slowermovements, some likened the sensation to that sometimes experienced when a liftstarts to move and one is at first uncertain about whether it is going up or down.At faster velocities, subjects were often in no doubt that a movement had occurred,but made no response because they could not nominate its direction. It is clear that,as Laidlaw & Hamilton (1937 a, b) stressed, the detection of the mere occurrence ofmovement has a lower threshold than the detection ofdirection ofmovement. As notedin the Introduction, it is the latter which is likely to depend on specificallyproprioceptive mechanisms. In the present study subjects were instructed to respondonly when they knew the direction of imposed movement, and thus many movementswhich were known to have occurred were not detected by the criteria used here.Indeed, the frequency of 'false positives' registered (movement detected, directionincorrect) was low - less than 5 0. However, the low-threshold non-specific signals

522




of occurrence of movement may have served as alerting signals for the subjects intheir concentration on detecting direction of movement.

Angular rotationThe 7000 detection levels recorded in experiments on the terminal joint of the

middle finger, the elbow joint and the shoulder joint are shown in Fig. 2. As thesubjects showed no systematic difference between detection ofmovements into flexionand into extension, data for movements in opposite directions are grouped together.

7

r5- .C

C

E0E

Bc3-\C0C.,\

0

00

0-1 0-5 1 5 10 50 100 200Angular velocity 0/s

Fig. 2. Data from ten subjects are shown. The graph illustrates the means (+S.E.M.) ofthe angular displacements necessary for 70 0 detection (including correct nomination ofdirection) of randomly mixed flexions and extensions imposed at various angular velocities(log scale). Data for the finger joint are given as circles, for the elbow joint as trianglesand for the shoulder joint as squares. At all angular velocities tested, larger angulardisplacements were required for 700% detection for the finger joint than for the moreproximal joints.

Fig. 2 shows that proprioceptive acuity at the distal interphalangeal joint of themiddle finger improved with increasing angular velocity, to a broad optimal rangebetween about 200 and 800/s. Performance deteriorated at 1600/s (7000 detectionlevel at 160'/s was greater than at 800/s: paired t test, P < 0 05; n.s. on grouped data).

Proprioceptive performance at the elbow and shoulder joints was superior to thatat the finger joint at all comparable angular velocities (700 detection levelscompared, grouped data, t test, P < 0 001). Shoulder and elbow joints gave similardetection levels, with those for the shoulder joint consistently a little lower than those




for the elbow joint. Optimal performance at both joints was attained at lower angularvelocities than with the finger joint, and was maintained through the broad rangeof angular velocities from 20 to 80'/s, without significant deterioration at the highervelocities tested.

In testing the elbow and shoulder joints the pronated hand and forearm were boundinto a trough-like splint on a carrier. The palm of the hand and the sensitive padsof the digits rested on the splint as the tests were carried out. It seemed possible thatcutaneous receptors in the hand might have been activated by the movement, andhave contributed to the detections, particularly at the faster velocities. Subjectsdenied any awareness of such signals when making their detection. The possibilitywas ruled out in one subject when the elbow joint was tested before and afteranaesthetizing the whole hand, without altering proprioceptive performance. Thehand was anaesthetized by ischaemic/pressure block achieved by inflation of a bloodpressure cuff to 300 mmHg at the wrist (see Goodwin, McCloskey & Matthews, 1972)after obtaining the informed consent of the subject.

Proportions of range of normal movementsThe data shown in Fig. 2 were plotted there in the form in which they were

collected - as detection levels in degrees, and velocities of movement in degrees persecond. In this and the following sections of the Results, the same data are convertedinto different terms for analysis.The normal ranges of angular excursion at the three joints tested are not the same.

The distal interphalangeal joint has a range of 90°, the elbow joint 145°, and theshoulder joint 1700 in the horizontal plane as tested (Kapandji, 1970; confirmed hereas approximately correct) Using these figures it was possible to convert the data ofdegrees and degrees per second into terms of proportions of the range of normalmovement. When expressed in these terms proprioceptive acuity was revealed againas better at proximal than at dismal joints.

Linear displacementsA given angular rotation at the terminal joint of the middle finger moves the tip

of that finger through a certain distance. Rotation of the shoulder joint through asimilar angle carries the same fingertip, at the end of an extended arm and hand,through a much greater distance. Rotation of the elbow joint through a similar angleinvolves a movement at the fingertip ofan intermediate amount. It is clearly possible,therefore, to express the data on angular rotations in terms of linear displacementsand velocities of the remote fingertip. This was done here. Conversion factors, whichwere determined from measurements made on the subjects (who were all adults ofmedium build) were as follows: 10 of rotation of the terminal joint of the middle fingermoves the fingertip 0 4 mm; 1° of rotation at the elbow joint moves the fingertip6-1 mm; and 10 of rotation at the shoulder joint moves the fingertip 13-13 mm. Whenthese factors were applied to the base data, the results shown in Fig. 3 were obtained.

It should be noted that, when proprioceptive acuity is analysed in terms of lineardisplacements and velocities at the end of the moving limb, as in Fig. 3, the previouslydemonstrated order of proprioceptive performance is reversed. The distal joint isshown to be superior, and the proximal joints inferior.

524



PROXIMAL AND DISTAL JOINTS

Changes in the lengths of muscle fasciclesPotential contributors to proprioceptive performance are intramuscular stretch

receptors. In order to analyse the behaviour of this type of receptor to proprioceptiveperformance, measurements were made of the lengths of fascicles in the mucles whichflex and extend the elbow joint and the terminal joint of the middle finger.Measurements at the shoulder joint were not made because of uncertainty about

24-

22-

20-

18-

E

-5

G)a0

_0

16-

14-

12-

10-

6-

4-

o-0.5 1 5 10 50 100

Velocity (mm/s)Fig. 3. The data collected from ten subjects in terms of angular displacements andvelocities, and illustrated in those terms in Fig. 2, are here shown converted to terms ofthe linear displacements and velocities (log scale) occurring at the fingertip as each of thetest joints is rotated. Data for the finger joint are given as circles, for the elbow joint astriangles and for the shoulder joint as squares. Assessed in these terms proprioceptiveperformance is superior at more distal joints. (Note that this reverses the order ofproprioceptive performances given in the presentation in Fig. 2.)

which fascicles and which muscles were appropriate. The measurements were madein the dissected, unembalmed cadaver of an adult male of average build. Individualmuscles operating at the two joints were displayed and closely inspected. Measurementswere made of the lengths of clear, individual muscle fascicles within the muscle, notof the whole muscle bellies. Where the lengths of individual fascicles varied in anyone muscle, the length mid-way between the shortest and longest fascicle measuredwas recorded. The fascicles were measured with the joint held at the position in whichit was tested in the present experiments. Rotation of the joint through a measuredangular excursion (usually 90°) gave a movement together or apart of the points of

525



L. A. HALL AND D.1. MeCLOSKEY

origin and insertion of the muscle: assuming no extension of the tendon in life byrotations imposed on the relaxed joints, the extent of this movement gave a figurefor the change in fascicle length per degree of rotation. In turn, this could be expressedas percentage change in fascicle length per degree, and factors obtained to convertthe data into these terms. The measurements are shown in Table 1.

TABLE 1. Length changes of muscle fascicle during joint rotation

Change in fascicleFasicle length at length per degree

test position of joint rotationMuscle (mm) (mm)

Extensor digitorum 110 0039Flexor digitorum profundus 85 005Biceps (long head) 154 0-62Biceps (short head) 169 0-62Brachialis 140 0 50Brachioradialis 168 0 47Triceps (long head) 240 0 44Triceps (lateral head) 145 0 30Triceps (medial head) 105 0-27

Fascicle lengths are given for the finger joint in 30° flexion, the elbow joint 90° flexion (i.e. testpositions).

Fig. 4 shows the 700 detection levels and the velocities of joint movement plottedin terms of proportional changes in the lengths of muscle fascicles. In this form ofpresentation both the proximal and the distal joints give comparable proprioceptiveperformance. What is to be noted particularly in Fig. 4 is that the data relating toboth joints correspond closely over the fairly broad range of optimal performances.

Joint capsular displacementParticipation of receptors in the joint capsules and ligaments must also be

considered as one of the mechanisms underlying the proprioceptive performancesanalysed here. Measurements were made of the joint capsules of the elbow joint andthe terminal joint of the middle finger in the same cadaver described above.Corresponding measurements for the shoulder were not made because of thecomplexity of the arrangement of the shoulder joint capsule and its attachments. Thedistances between the points of capsular attachment on flexor and extensor surfacesof each joint were measured with the joint held in the position in which it was testedin the present experiments. Rotation of the joint through a measured angularexcursion (600 or 900) and further measurement gave a figure for the changes in thesedistances per degree of angular rotation.With the conversion factors thus obtained, the 7000 detection levels and the

velocities of joint movement were plotted in terms of proportional changes in jointcapsular length. On this form of analysis proprioceptive performance of the elbowjoint was markedly superior to that of the finger joint.

526




Active movementThe velocities at which active movements were performed were measured at each

of the test joints in four of the subjects. In each test the subject sat with the jointto be moved in the same posture and apparatus in which the tests for detection of

A

0-8- < BrachialisV A Biceps (long head)* v Biceps (short head)

0 7 - * Brachioradialis* Triceps (long head)

-s * Triceps (medial head)C' *Triceps (lateral head)a) 0-6- > oFlexor digitorum profundus

XI Extensor digitorumC

0.5Q.5C

.C: 0-5-3

a)

0-0

>A

02U

01

0E * O

0 1- t 0hs

0 r002 02 2 20 40

Velocity (% change in test length/s)Fig. 4. Data for the elbow and finger joints, from the ten subjects illustrated in Fig. 2,are here shown converted to terms of percentage changes in the lengths of fascicles inindividual muscles. Conversion of data was done by measuring the median length ofindividual muscle fascicles in each muscle, and the alteration in those lengths per degreeof angular rotation. Measurements were made with the joints positioned as they were fortesting. Data are given as 70%0 detection levels in percentage change of test length offascicles, and as the velocity of such changes (log scale). Muscles operating at the elbowjoint are shown as filled symbols and muscles operating at the distal joint of the middlefinger are shown as open symbols. Assessed in these terms proprioceptive performance issimilar at the elbow and the finger joint, particularly over the range of optimalperformance.

imposed movements had previously been conducted. The apparatus was disconnectedfrom the electromagnetic vibrator, however, and moved freely in response to smallforces (a load of 50 g applied at the fingertip was sufficient to move each relaxed jointat a velocity faster than any recorded here). Position of the test joint was measured,for finger joint movement with the usual position transducer, and for elbow and

527



528 L. A. HALL AND D. I. McCLOSKEY

shoulder joints with a sensitive potentiometer mounted co-axially with the test joint.The subjects sat facing an array of four objects about a metre in front of them, andwere asked to point an imaginary long wire pointer attached to the middle finger asaccurately as possible at the different targets. The subjects could see the targets,but not the arm, hand or finger. They were permitted to undertake the task in their

40-

20-

r

40-

E0 20-0

6z

40

F inger

Elbow

Shou Ider

40 80 120 160 200 240 280

Angular velocity (0/s)Fig. 5. Distributions of velocities chosen by four subjects in making accurate pointingmovements by actively moving the distal joint ofthe middle finger (upper panel), the elbowjoint (middle panel) and the shoulder joint (lower panel). Each subject made twenty-fivesuch movements with each joint. See text for details.

own time and could choose movements between any targets they chose. The targetswere between 150 and 400 apart. The movements were recorded on a pen recorder.The average velocity of each movement was calculated between a point 40 of rotationfrom commencement of the movement and 40 of rotation from its conclusion. In theconditions of this experiment it was found that angular velocity was fairly constantin this range.

Fig. 5 shows the velocities chosen in these experiments at each of the test joints:the velocities of one hundred movements, twenty-five from each subject, are shownfor each joint. For movements at the distal interphalangeal joint the velocities of

Lus

U, -I_,




active movements varied from 300 to 2850/s. However, the majority (70 00) were madeat velocities between 50° and 1100/s. For the elbow joint slower velocities werechosen: between 50 and 550/s, with 850 of velocities in the range 10-350/s. Evenslower velocities were preferred for shoulder movements: between 50 and 350/s, with620 of velocities in the range 100-200/s.

DISCUSSION

This paper reports the amplitudes of joint rotation required for 7000 correctdetection by normal subjects. Data for three joints in the upper limb are reported,for a broad range of velocities of rotation.The choice of 700 correct as the detection level of interest in this study was a

purely arbitrary one. In subjects where we measured both 500 and 700 levels, theshapes of the relations between amplitude and velocity were unchanged, with slightlysmaller amplitudes required at each velocity when the 500 level was used. In theconditions of this study, the 500% level would not have the same significance as ithas in simple detection tasks where a subject has only to register the presence orabsence of a stimulus in a specified period. In those conditions, of course, consistentdetection at above the 500 level indicates a better than random performance. Here,however, simple detection was not enough, as the proprioceptively significantproperty, direction, was also required. It has already been noted that most subjectsdetected the presence of imposed movements on many occasions when they could notnominate direction, and so remained silent. Also there were very few 'false positives',that is, nominations of the incorrect direction for perceived movements. In suchconditions it is clear that detections at a rate above this low level of false positivesare better than random performance. Therefore, the choice of the 700 detection levelhere represents choice of a criterion level which requires performance clearly abovethreshold level.The results of the present study confirm Goldscheider's (1889) finding that,

expressed in terms of angular rotation, proprioceptive acuity is higher at moreproximal joints. A similar conclusion follows if the data are expressed in terms ofproportions of the range of angular excursion, instead of in degrees and degrees persecond. Consultation of previous work in this field (Goldscheider, 1889; Pillsbury,1901; Winter, 1912; Laidlaw & Hamilton, 1937b; Cleghorn & Darcus, 1952; Provins,1958; Kokmen, Bossmeyer & Williams, 1977) shows that this widely acceptedconclusion hitherto rested entirely on Goldscheider's findings, conducted over a muchnarrower range of angular velocities, and had not been duplicated by others.Furthermore, the present study also confirmed Goldscheider's finding that self-paced,active movements at a joint are usually carried out at velocities within the range ofoptimal proprioceptive performance.The superior proprioceptive performance, measured in terms of angular rotation,

of proximal joints need not imply a greater concern of the central nervous systemfor proximal rather than distal joints. Distal joints are more severely affected bycortical lesions (Head, 1920; Holmes, 1927), and are more strongly perceived inphantom limbs (Henderson & Smyth, 1948), and so appear to have an importancerelated to the area of cortical representation of the body parts (Penfield & Boldrey,

529




1937). What the results of the present work call into question, however, are thecriteria by which proprioceptive performance has conventionally been assessed -criteria involving analyses in terms of angular rotation.A joint is actively moved so as to alter the location of a more distal body part.

If proprioceptive acuity is important in the accomplishment of simple tasks such asthis, then for practical purposes, proprioceptive acuity should be assessed in terms oflinear displacments of that more distal body part. When this is done, as it was herein Fig. 3 by appropriately scaling the data which had been collected in angular terms,distal joints gave clearly superior performance. What Fig. 3 means, for example, isthat if a small linear displacement occurs at the fingertip, at any given linear velocity,it will be most readily detected if it comes about by rotation of only the finger joint,and least readily detected if the rotation is confined to the shoulder joint. Clearly,in these practical terms, acuity is superior at distal joints.The data may also be viewed with underlying mechanisms in mind. Analysis in

terms of movement of the points of attachment of joint capsules at the elbow andterminal joint of the finger does not unify the data: for a given proportional changein the distance between joint capsular attachments, at any velocity of such change,the elbow joint gives superior proprioceptive performance. The points of attachmentof the joint capsule, and their movement with rotation of the joint, were taken hereas measures of joint capsular movement. It is clear from inspection of the capsulesin a fresh cadaver, however, that such movement of attachments is only a poorindicator of capsular stretch. Over most of the range of excursion in each joint thecapsular fibres are not under stretch and the capsular surface is slack, or even folded.Only when the joint nears one limit of its excursion is the capsule tightened. Thisfits with the well-established property of joint receptors, which usually fire maximallyat one, or sometimes both extremes of the joint's range (e.g. Clark & Burgess, 1975;Ferrell, 1980). In so far as movement of the capsule's attachment can be regardedas an indicator of the mechanical disturbance presented to the joint receptors,however, and assuming in addition that the populations of such receptors havecomparable influences centrally, the findings reported here suggest that the jointreceptors cannot be solely responsible for performance - otherwise, performance atboth joints would have been comparable when expressed in terms of capsularmovement. An alternative statement of this conclusion would be that, if jointreceptors do account for performance at the two joints, then either the receptorpopulations at the two joints differ in number, sensitivity or central influence, or thesimilar proportional movements of the capsular attachments impose systematicallydifferent degrees of stretch on capsules. Both these alternatives represent testablehypotheses.Perhaps the most striking feature to emerge from this study is the way that

proprioceptive performance at the finger and elbow joints is unified by analysis interms of the lengths of fascicles of the muscles which operate those joints (Fig. 4).This strongly suggests a common mechanism based upon this variable. Such amechanism implies a receptor population sensitive to very small changes of fasciclelength. For example, the optimal detection levels shown in Fig. 4 lie approximatelybetween 01025 and 0-05 00 of the length of the fascicles, for velocities between about1 and 200 of that length per second. For a fascicle of 10 cm these figures represent

530




25-50 1m, at 1-20 mm/s. The results suggest involvement of the muscle spindles and,indeed, support the existing evidence of a role for these receptors in consciousproprioception (Goodwin et al. 1972; Goodwin, 1976; Matthews, 1977; McCloskey,1978, 1981). The muscle spindles detect length changes with great sensitivity: forexample, Matthews & Stein (1969) reported that 3000 modulation of the dischargeof a single spindle primary could be produced by a movement of 100 /um (probablyabout 0-2 0% of muscle length). Goodwin, Hulliger & Matthews (1975) demonstratedgood spindle responses to 1 Hz stretches of only 10 jtm amplitude (see also Matthews,1972; Hasan & Houk, 1975). Furthermore, these data refer to unitary spindlebehaviour, and it is reasonable to assume that thresholds for delectability ofsummedafferent discharge will be far less. The examples given refer to spindle sensitivityin the cat and it has been argued (Vallbo, 1974) that human spindles may haveconsiderably less position sensitivity than cat hind limb extensors: that conclusion,however, is based upon conversion to proportional muscle lengths ofthe displacementsactivating human spindles, and the conversion factors used may have been inap-propriate. Elsewhere, Vallbo, Hagbarth, Torebjork & Wallin (1979) concluded thathuman spindles exhibit a high sensitivity for small movements. Newsom Davis (1975)argued that biopsied human intercostal spindles in his study had, ifanything, a higherstatic ('position') sensitivity than those of the cat, once proportional changes oflength had been taken into account. More recently, McKeon & Burke (1981) reportedmodulation of single human spindles in synchrony with the arterial pulse, andMcKeon, Burke & Gandevia (1981) reported human spindles to be sensitive to twitchcontractions of single motor units - findings which have been seen with feline spindlesand which indicate that human spindles have comparably high sensitivity.The unification of data on proprioceptive performance made possible by analysis

in terms of proportional changes in the lengths of muscle fascicles need not indicatethat muscle groups at all joints will give similar performance. The elbow joint andthe finger joint tested here both move the same distal body part, and might well begiven equal significance within the central nervous system. It need not follow thatthe proprioceptive performance of joints moving another body part, say the toe, arecomparably 'tuned'. It is likely also that the performance documented here in termsof muscle receptors is one which is dependent upon the summed input from receptorsin all of the muscles operating at a joint. When the spindles in an agonist group ofmuscles discharge in response to rotation of a joint, spindles in the antagonist willdischarge less or fall silent, and both the increased discharge from one group anddecreased discharge from the other are likely to be important in detecting movement.That both sets of receptors are used was demonstrated in the accompanying paper(Gandevia et al. 1983), where the removal of inputs from one ofthe agonist/antagonistmuscle groups was shown to result in a reduction in proprioceptive acuity. Ho ever,when estimation of the extent of a suprathreshold displacement, rather than simplyits detection, is the task in hand, it may well be that receptors in agonist andantagonist muscle groups have very different significance.

This work was supported by the National Health and Medical Research Council of Australia.We thank Drs D. Burke, S. Gandevia, P.B.C. Matthews, E. Potter and A. Prochazka for theircomments on some aspects of the work. Diane Madden provided expert technical assistance.

531



5L. A. HALL AND D. 1. McCLOSKEY

REFERENCES

CLARK, F. J. & BURGESS, P. R. (1975). Slowly adapting receptors in the cat knee joint: can theysignal joint angle? J. Neurophysiol. 38, 1448-1463.

('LEGHORN, T. E. & DARCUS, H. D. (1952). The sensibility to passive movement of the human elbowjoint. Q. J exp. Psychol. 4, 66-77.

FERRELL, W. R. (1980). The adequacy of stretch receptors in the cat knee joint for signalling jointangle throughout the range of movement. J. Physiol. 299, 85-99.

GANDEVIA, S. C., HAI,, L. A., MCCLOSKEY, D. I. & POTTER, E. K. (1983). Proprioceptivesensation at the terminal joint of the middle finger. J. Physiol. 355, 507-517.

GOLDSCHEIDER, A. (1889). Untersuchungen uber den Muskelsinn. Arch. Anat. Physiol. 3, 369-502.GOODWIN, G. M. (1976). The sense of limb position and movement. Exercise Sport Sci. Rev. 4, 87-124.GOODWIN, G. M., HULLIGER, M. & MATTHEWS, P. B. C. (1975). The effects of fusimotor stimulation

during small amplitude stretching on the frequency-response of the primary ending of themammalian muscle spindle. J. Physiol. 253, 175-206.

GOODWIN, G. M., MCCLOSKEY, D. I. & MATTHEWS, P. B. C. (1972). The contribution of muscleafferents to kinaesthesia shown by vibration induced illusions of movement and by the effectsof paralysing joint afferents. Brain 95, 705-748.

GRAY, H. (1980). Gray's Anatomy, 36th edn, ed. WILLIAMS, P. L. & WARWICK, R., pp. 460-473.Edinburgh: Churchill Livingstone.

HALL, L. A. & MCCLOSKEY, D. I. (1981). Detection of displacements imposed at the terminal jointof the middle finger, the elbow joint and the shoulder joint. Proc. Aust. Physiol. Pharmac. Soc'12, 196P.

HASAN, Z. & HOUK, J. C. (1975). Transition in sensitivity of spindle receptors that occurs whenmuscle is stretched more than a fraction of a millimeter. J. Neurophysiol. 38, 673-689.

HEAD, H. (1920). Studies in Neurology, vol. 2. London: Oxford University Press.HENDERSON, W. R. &SMYTH,G. E. (1948). Phantomlimbs.J. Neurol. Neurosurg. Psychiat. 11,88-112.HOLMES, G. (1927). Disorders of sensation produced by cortical lesions. Brain 50, 413-427.KAPANDJI, I. A. (1970). The Physiology of the Joints, 2nd edn., vol. 1. Edinburgh: Churchill

Livingstone.KOKMEN, E., BOSSMEYER, R. W. JR & WILLIAMS, W. J. (1977). Quantitation of motion perception

in the digits: a psychophysical study in normal human subjects. Ann. Neurol. 2, 279-284.LAIDLAW, R. W. & HAMILTON, M. A. (1937a). The quantitative measurement of apperception of

passive movement. Bull. neurol. Inst. N. Y. 6, 145-153.LAIDLAW, R. WV. & HAMILTON, M. A. (1937b). A study of thresholds in apperception of passivemovement among normal control subjects. Bull. neurol. Inst. N. Y. 6, 268-273.

MCCLOSKEY, D. I. (1978). Kinesthetic sensibility. Physiol. Rev. 58, 763-820.MCCLOSKEY, D. I. (198 1). Corollary discharges. In American Handbook ofPhysiology (Motor Control),

pp. 14 15-1448. Bethesda, Maryland: American Physiological Society.McKEON, B. & BURKE, D. (1981). Component of muscle spindle discharge related to the arterial

pulse. J. Neurophysiol. 46, 788-796.McKEON, B., BURKE, D. & GANDEVIA, S. C. (1981). Mechanical coupling between muscle spindlesand motor units in man. Proc. Aust. Physiol. Pharmac. Soc. 12, 195P.

MATTHEWS, P. B. C. (1972). Mammalian Muscle Receptors and their Central Actions. London:Edward Arnold.

MATTHEWS, P. B. C. (1977). Muscle afferents and kinaesthesia. Br. Med. Bull. 33, 137-142.MATTHEWS, P. B. C. & STEIN, R. B. (1969). The sensitivity of muscle spindle afferents to small

sinusoidal changes of length. J. Physiol. 200, 723-743.NEWSOM DAVIS, J. (1975). The response to stretch of human intercostal muscle spindles studied

in vitro. J. Physiol. 249, 561-579.PENFIELD, W. & BOLDREY, E. (1937). Somatic motor and sensory representation in the cerebral

cortex of man as studied by electrical stimulation. Brain 60, 389-443.PILLSBURY, W. B. (1901). Does the sensation of movement originate in the muscle? Am. J. Psychol.

12, 346-353.PROVINS, K. A. (1958). The effect of peripheral nerve block on the appreciation and execution of

finger movements. J. Physiol. 143, 55-67.

532




VALLBO, A. B. (1974). Afferent discharge from human muslce spindles in noncontracting muscles.Steady state impulse frequency as a function of joint angle. Acta physiol. scand. 90, 303-318.

VALLBO, A. B., HAGBATH, K.-E., TOREBJ6RK, H. E. & WALLIN, B. G. (1979). Somatosensory,proprioceptive and sympathetic activity in human peripheral nerves. Physiol. Revi 59,919-957.

WINTER, J. E. (1912). The sensation of movement. Psychol. Rev. 19, 374-385.



1983;335;519-533 J. Physiol.

L A Hall and D I McCloskey shoulder joints.

Detections of movements imposed on finger, elbow and

This information is current as of January 15, 2008

& ServicesUpdated Information

http://jp.physoc.orgbe found at: including high-resolution figures, can

Permissions & Licensing

shtmlhttp://jp.physoc.org/misc/Permissions.entirety can be found online at:

itsarticle in parts (figures, tables) or in Information about reproducing this

Reprints

lhttp://jp.physoc.org/misc/reprints.shtmcan be found online: Information about ordering reprints



http://jp.physoc.org/misc/Permissions.shtml

http://jp.physoc.org/misc/reprints.shtml


Detections of movements imposed on finger, elbow and...

Documents

Transcript of Detections of movements imposed on finger, elbow and...