Exp Brain Res (1992) 91 : 539 542
Experimental BrainResearch 9 Springer-Verlag 1992
Research Note
Independent control of the digits: changes in perceived heaviness over a wide range of force S.L. Kilbreath and S.C. Gandevia Department of Clinical Neurophysiology, The Prince Henry Hospital and Prince of Wales Medical Research Institute, University of New South Wales, Sydney 2036, Australia Received June 24, 1992 / Accepted August 5, 1992
Summary. Perceived heaviness of a weight lifted by flexion of the distal joint of one digit increases when an adjacent digit concurrently lifts a weight. The present study confirmed this finding for relatively low weights (representing 3-5% maximal voluntary force) and a method was adapted to show that this effect occurs for much larger weights (20-25% maximal force). Thus, the increase in perceived heaviness is likely to operate over a wide range of muscle force generated by the hand. As perceived heaviness is biased by the magnitude of the central motor commands, these findings may reflect a lack of complete independence of motor commands to the long flexor muscles acting on the digits. Key words: Motor commands - Hand function - Perceived heaviness - Human
Introduction The perceived heaviness of a lifted weight is biased by the magnitude of central motor commands rather than signals related to peripheral muscle force (for reviews, see McCloskey 1981; Gandevia 1987; Cafarelli 1988; Jones 1988). Weight-matching tasks, in which subjects compare the heaviness of lifted weights, have thus been used to assess central motor commands indirectly (e.g. Gandevia and McCloskey 1977a, b; Aniss et al. 1988; Gandevia and Kilbreath 1990; Kilbreath and Gandevia 1991). The major aim of the present study was prompted by the observation that the perceived heaviness of a weight lifted by flexion of one digit increases when an adjacent digit concurrently lifts a separate weight (Kilbreath and Gandevia 1991). Thus, the force achieved by one muscle acting on the hand cannot be perceived independently of that achieved by another related muscle. This increase occurred when weights were lifted by the long flexors of the digits (i.e. flexor pollicis longus Correspondence to: S.C. Gandevia
and digital portions of flexor digitorum profundus). The magnitude of the perceived heaviness increased with the size of the concurrently lifted weight and with fatigue. The illusion was not present when one weight was lifted by flexor digitorum profundus and the other weight was concurrently lifted by a muscle in the lower limb. The reference and concurrent weights lifted represented less than 5% of maximal voluntary forces (MVC) of the relevant muscles. The present study aimed to determine whether this lack of independence in the perception of forces generated by the digits occurred when the reference and concurrent weights represented 20-25% maximal force. To make psychophysical judgements of these higher forces with little likelihood of fatigue required a more rapid and controlled psychophysical method of weight estimation. Perceived heaviness is usually measured with the method of adjustment, involving a "reference" weight (or force) on one side and a "variable" weight on the other (McCloskey etal. 1974; Gandevia and McCloskey 1977a, b; Jones and Hunter 1983, 1990; Aniss et al. 1988; Gandevia and Kilbreath 1990; Kilbreath and Gandevia 1991). Thus, we first aimed to determine whether the estimates are the same when a rapid invariant method of weight estimation is used as when the traditional method of adjustment is used. The method of limits, based upon t h e " up-and-down" rule (Wetherill et al. 1966), was selected because it is robust and has previously been used for weight-matching (e.g. Holway et al. 1937; Wetherill et al. 1966; Bertelsmann et al. 1985). Materials and methods These studies were performed on 15 subjects. The procedures were approved by the local ethics committee. In the first study, each subject matched a 500 g reference weight with the method of adjustment and with the method of limits. Subjects (n = 9) simultaneously lifted the reference and the variable weights with flexor pollicis longus by flexion at the interphalangeal joint of the thumbs (e.g. Gandevia and Kilbreath 1990). The right
540 800
1'3
9 Adjustments
600
Estimate of
heaviness (g)
400
O
200
Initial 1
value
6
12
Progressive weight changes
Fig. 1. Data from a single subject. On the left are shown the mean (+SD) estimates for 500 g reference weight with the adjustment method (filled circle) and with the limits method (open circle). On the right are the subject's raw data from the two extended trials with the method of limits. The peaks and troughs are weights at which the subject perceived the variable weight to be below or above the reference weight and so requested a change in direction. With this technique, estimates are derived from the midpoints between sequential peaks and troughs
thumb lifted the 500 g reference weight and the left thumb lifted a variable weight. These weights were not visible to the subject. Subjects simultaneously lifted the two weights up to three times before making a decision on their relative heaviness. Subjects were given no further instructions. In the "adjustment" method, a subject requested that the variable weight be adjusted in the direction appropriate to make it feel the same as the reference weight. A trial was completed when the subject reported that the variable weight felt the same as the 500 g reference weight. Each subject completed ten trials for each condition. In half of the trials, the variable weight was initially higher than the reference weight and, in the other half, it was lower. Subjects were unaware that the changes in the variable weight were standardized: at the beginning of each trial, the variable weight was changed by 50 g; when the subject requested a change in direction, i.e. the reference weight had been "overshot", the weights were usually changed by 25 g. Occasionally, the weights were changed by a larger amount. Two extended trials, using the method of limits, were undertaken by each subject. This method was based upon the up-and-down rule (Wetherill et al. 1966) and used a two-alternative, forced-choice paradigm. A subject was constrained to request only an increase or decrease in the variable weight. Thus, each time the subject perceived that the apparent heaviness of the reference weight had been overshot, a "change of direction" in loading of the variable weight occurred. A trial involved 12 changes in direction. The variable weight was changed initially by 50 g until the subject requested the first change in direction and thereafter by 25 g. Data from a single subject, indicating the successive peaks and troughs associated with changes in direction, are shown in Fig. 1. The subject's estimates were taken as the midpoints between successive troughs and peaks (and vice-versa) which signal the subject's perception that an over- or under-estimation of the reference weight had occurred (see Fig. 1). The second study also used both methods of weight-matching to compare the effect on perceived heaviness when a weight was
concurrently lifted. Subjects (n=8) matched a 200 g reference weight lifted by flexion of the interphalangeal joint of the thumb. In addition, they matched the reference weight while concurrently lifting a 300 g weight by flexion of the adjacent distal interphalangem joint of the index finger. The reference and concurrent weights were lifted simultaneously by the digits of the right hand together with the variable weight lifted by the left thumb (see Kilbreath and Gandevia 1991). Movement of the index finger was limited to active flexion of the distal interphalangeal joint and any cocontraction produced by the long thumb and finger extensor muscles was eliminated by anaesthesia of the radial nerve (6-8 ml of 2% lignocaine with adrenaline at the mid-humerus level). The block remained complete for the duration of each experiment. Ten control trials and ten trials with concurrent lifting were completed in blocks with the adjustment method. As in the first study, half of the trials commenced with the initial variable weight higher than the reference weight and half lower. Four extended trials, using the method of limits, were completed: in one control trial and one trial with concurrent lifting, the variable weight was initially lighter than the reference weight and in the other two trials it was initially heavier. The order of trials was randomized between subjects. Because the effect of concurrent lifting on perceived heaviness was seen for all studied pairs of digital flexors, in the third study, two digital portions of flexor digitorum profundus were used to lift the reference and concurrent weights. Subjects matched a 1000 g weight with the index finger while concurrently lifting a 1000 g weight with the middle finger (see Kilbreath and Gandevia 1991). The hand posture adopted for the third study eliminated any active extensor torque (Kilbreath and Gandevia 1991). To reduce the likelihood of fatigue, only the method of limits, as above, was used. Furthermore, only seven changes in direction were used in this study and the second one as analysis of the data from the first study showed no further change to the estimate after seven changes. Data were analysed with analysis of variance (ANOVA) and paired t-tests. Statistical significance was set at the 5% level.
Results T h e m e a n e s t i m a t e (_+ S D ) o f the 500 g reference w e i g h t using the a d j u s t m e n t m e t h o d was 4 2 0 + 103 g a n d 422 ( + 107) g w h e n the m e t h o d o f limits was used. A l t h o u g h there was no difference in the e s t i m a t e o f p e r c e i v e d heaviness b e t w e e n the t w o m e t h o d s , e x p e r i m e n t s using the m e t h o d o f limits t o o k less t h a n h a l f the time r e q u i r e d for the a d j u s t m e n t m e t h o d a n d i n v o l v e d a p p r o x i m a t e l y o n e - h a l f the n u m b e r o f m u s c l e c o n t r a c t i o n s . W h e n the trials f r o m the a d j u s t m e n t m e t h o d were c a t e g o r i z e d acc o r d i n g to w h e t h e r the initial v a r i a b l e w e i g h t was h e a v i e r o r lighter t h a n the reference weight, A N O V A also revealed no significant difference b e t w e e n the two m e a n s : 4 1 8 + 101 g w h e n the initial v a r i a b l e weights were light a n d 421 _+ 101 g w h e n the initial v a r i a b l e weights were heavy. Thus, the e s t i m a t e s o f the reference weight were n o t b i a s e d b y o u r l o a d i n g strategy. T h e 500 g reference weight o n the r i g h t was m a t c h e d w i t h a slightly l i g h t e r w e i g h t o n the left (t-test, P < 0 . 0 5 ) . T h e difference in heaviness b e t w e e n the m a t c h e d reference a n d v a r i a b l e weights m a y reflect a difference in a b s o l u t e s t r e n g t h o f the r i g h t a n d left flexor pollicis longus. I n the s e c o n d study, the results f r o m b o t h p s y c h o p h y s i c a l m e t h o d s s h o w e d t h a t w h e n a 300 g weight was lifted c o n c u r r e n t l y b y the i n d e x finger, p e r c e i v e d heaviness o f the 200 g reference w e i g h t lifted b y the t h u m b
541 200
1200
150
1000
Estimate of heaviness
Fig. 2. On the left, results from matching a 200 g ref-
(g)
100
50
800
' Control (200 g)
' Concurrent lift (300 9)
Adjustment method
'
'
600
I
i
Control (1000 g) Limit method
Concurrent lift (1000 g)
Limit method
flexion increased (Fig. 2, P
Experimental BrainResearch 9 Springer-Verlag 1992
Research Note
Independent control of the digits: changes in perceived heaviness over a wide range of force S.L. Kilbreath and S.C. Gandevia Department of Clinical Neurophysiology, The Prince Henry Hospital and Prince of Wales Medical Research Institute, University of New South Wales, Sydney 2036, Australia Received June 24, 1992 / Accepted August 5, 1992
Summary. Perceived heaviness of a weight lifted by flexion of the distal joint of one digit increases when an adjacent digit concurrently lifts a weight. The present study confirmed this finding for relatively low weights (representing 3-5% maximal voluntary force) and a method was adapted to show that this effect occurs for much larger weights (20-25% maximal force). Thus, the increase in perceived heaviness is likely to operate over a wide range of muscle force generated by the hand. As perceived heaviness is biased by the magnitude of the central motor commands, these findings may reflect a lack of complete independence of motor commands to the long flexor muscles acting on the digits. Key words: Motor commands - Hand function - Perceived heaviness - Human
Introduction The perceived heaviness of a lifted weight is biased by the magnitude of central motor commands rather than signals related to peripheral muscle force (for reviews, see McCloskey 1981; Gandevia 1987; Cafarelli 1988; Jones 1988). Weight-matching tasks, in which subjects compare the heaviness of lifted weights, have thus been used to assess central motor commands indirectly (e.g. Gandevia and McCloskey 1977a, b; Aniss et al. 1988; Gandevia and Kilbreath 1990; Kilbreath and Gandevia 1991). The major aim of the present study was prompted by the observation that the perceived heaviness of a weight lifted by flexion of one digit increases when an adjacent digit concurrently lifts a separate weight (Kilbreath and Gandevia 1991). Thus, the force achieved by one muscle acting on the hand cannot be perceived independently of that achieved by another related muscle. This increase occurred when weights were lifted by the long flexors of the digits (i.e. flexor pollicis longus Correspondence to: S.C. Gandevia
and digital portions of flexor digitorum profundus). The magnitude of the perceived heaviness increased with the size of the concurrently lifted weight and with fatigue. The illusion was not present when one weight was lifted by flexor digitorum profundus and the other weight was concurrently lifted by a muscle in the lower limb. The reference and concurrent weights lifted represented less than 5% of maximal voluntary forces (MVC) of the relevant muscles. The present study aimed to determine whether this lack of independence in the perception of forces generated by the digits occurred when the reference and concurrent weights represented 20-25% maximal force. To make psychophysical judgements of these higher forces with little likelihood of fatigue required a more rapid and controlled psychophysical method of weight estimation. Perceived heaviness is usually measured with the method of adjustment, involving a "reference" weight (or force) on one side and a "variable" weight on the other (McCloskey etal. 1974; Gandevia and McCloskey 1977a, b; Jones and Hunter 1983, 1990; Aniss et al. 1988; Gandevia and Kilbreath 1990; Kilbreath and Gandevia 1991). Thus, we first aimed to determine whether the estimates are the same when a rapid invariant method of weight estimation is used as when the traditional method of adjustment is used. The method of limits, based upon t h e " up-and-down" rule (Wetherill et al. 1966), was selected because it is robust and has previously been used for weight-matching (e.g. Holway et al. 1937; Wetherill et al. 1966; Bertelsmann et al. 1985). Materials and methods These studies were performed on 15 subjects. The procedures were approved by the local ethics committee. In the first study, each subject matched a 500 g reference weight with the method of adjustment and with the method of limits. Subjects (n = 9) simultaneously lifted the reference and the variable weights with flexor pollicis longus by flexion at the interphalangeal joint of the thumbs (e.g. Gandevia and Kilbreath 1990). The right
540 800
1'3
9 Adjustments
600
Estimate of
heaviness (g)
400
O
200
Initial 1
value
6
12
Progressive weight changes
Fig. 1. Data from a single subject. On the left are shown the mean (+SD) estimates for 500 g reference weight with the adjustment method (filled circle) and with the limits method (open circle). On the right are the subject's raw data from the two extended trials with the method of limits. The peaks and troughs are weights at which the subject perceived the variable weight to be below or above the reference weight and so requested a change in direction. With this technique, estimates are derived from the midpoints between sequential peaks and troughs
thumb lifted the 500 g reference weight and the left thumb lifted a variable weight. These weights were not visible to the subject. Subjects simultaneously lifted the two weights up to three times before making a decision on their relative heaviness. Subjects were given no further instructions. In the "adjustment" method, a subject requested that the variable weight be adjusted in the direction appropriate to make it feel the same as the reference weight. A trial was completed when the subject reported that the variable weight felt the same as the 500 g reference weight. Each subject completed ten trials for each condition. In half of the trials, the variable weight was initially higher than the reference weight and, in the other half, it was lower. Subjects were unaware that the changes in the variable weight were standardized: at the beginning of each trial, the variable weight was changed by 50 g; when the subject requested a change in direction, i.e. the reference weight had been "overshot", the weights were usually changed by 25 g. Occasionally, the weights were changed by a larger amount. Two extended trials, using the method of limits, were undertaken by each subject. This method was based upon the up-and-down rule (Wetherill et al. 1966) and used a two-alternative, forced-choice paradigm. A subject was constrained to request only an increase or decrease in the variable weight. Thus, each time the subject perceived that the apparent heaviness of the reference weight had been overshot, a "change of direction" in loading of the variable weight occurred. A trial involved 12 changes in direction. The variable weight was changed initially by 50 g until the subject requested the first change in direction and thereafter by 25 g. Data from a single subject, indicating the successive peaks and troughs associated with changes in direction, are shown in Fig. 1. The subject's estimates were taken as the midpoints between successive troughs and peaks (and vice-versa) which signal the subject's perception that an over- or under-estimation of the reference weight had occurred (see Fig. 1). The second study also used both methods of weight-matching to compare the effect on perceived heaviness when a weight was
concurrently lifted. Subjects (n=8) matched a 200 g reference weight lifted by flexion of the interphalangeal joint of the thumb. In addition, they matched the reference weight while concurrently lifting a 300 g weight by flexion of the adjacent distal interphalangem joint of the index finger. The reference and concurrent weights were lifted simultaneously by the digits of the right hand together with the variable weight lifted by the left thumb (see Kilbreath and Gandevia 1991). Movement of the index finger was limited to active flexion of the distal interphalangeal joint and any cocontraction produced by the long thumb and finger extensor muscles was eliminated by anaesthesia of the radial nerve (6-8 ml of 2% lignocaine with adrenaline at the mid-humerus level). The block remained complete for the duration of each experiment. Ten control trials and ten trials with concurrent lifting were completed in blocks with the adjustment method. As in the first study, half of the trials commenced with the initial variable weight higher than the reference weight and half lower. Four extended trials, using the method of limits, were completed: in one control trial and one trial with concurrent lifting, the variable weight was initially lighter than the reference weight and in the other two trials it was initially heavier. The order of trials was randomized between subjects. Because the effect of concurrent lifting on perceived heaviness was seen for all studied pairs of digital flexors, in the third study, two digital portions of flexor digitorum profundus were used to lift the reference and concurrent weights. Subjects matched a 1000 g weight with the index finger while concurrently lifting a 1000 g weight with the middle finger (see Kilbreath and Gandevia 1991). The hand posture adopted for the third study eliminated any active extensor torque (Kilbreath and Gandevia 1991). To reduce the likelihood of fatigue, only the method of limits, as above, was used. Furthermore, only seven changes in direction were used in this study and the second one as analysis of the data from the first study showed no further change to the estimate after seven changes. Data were analysed with analysis of variance (ANOVA) and paired t-tests. Statistical significance was set at the 5% level.
Results T h e m e a n e s t i m a t e (_+ S D ) o f the 500 g reference w e i g h t using the a d j u s t m e n t m e t h o d was 4 2 0 + 103 g a n d 422 ( + 107) g w h e n the m e t h o d o f limits was used. A l t h o u g h there was no difference in the e s t i m a t e o f p e r c e i v e d heaviness b e t w e e n the t w o m e t h o d s , e x p e r i m e n t s using the m e t h o d o f limits t o o k less t h a n h a l f the time r e q u i r e d for the a d j u s t m e n t m e t h o d a n d i n v o l v e d a p p r o x i m a t e l y o n e - h a l f the n u m b e r o f m u s c l e c o n t r a c t i o n s . W h e n the trials f r o m the a d j u s t m e n t m e t h o d were c a t e g o r i z e d acc o r d i n g to w h e t h e r the initial v a r i a b l e w e i g h t was h e a v i e r o r lighter t h a n the reference weight, A N O V A also revealed no significant difference b e t w e e n the two m e a n s : 4 1 8 + 101 g w h e n the initial v a r i a b l e weights were light a n d 421 _+ 101 g w h e n the initial v a r i a b l e weights were heavy. Thus, the e s t i m a t e s o f the reference weight were n o t b i a s e d b y o u r l o a d i n g strategy. T h e 500 g reference weight o n the r i g h t was m a t c h e d w i t h a slightly l i g h t e r w e i g h t o n the left (t-test, P < 0 . 0 5 ) . T h e difference in heaviness b e t w e e n the m a t c h e d reference a n d v a r i a b l e weights m a y reflect a difference in a b s o l u t e s t r e n g t h o f the r i g h t a n d left flexor pollicis longus. I n the s e c o n d study, the results f r o m b o t h p s y c h o p h y s i c a l m e t h o d s s h o w e d t h a t w h e n a 300 g weight was lifted c o n c u r r e n t l y b y the i n d e x finger, p e r c e i v e d heaviness o f the 200 g reference w e i g h t lifted b y the t h u m b
541 200
1200
150
1000
Estimate of heaviness
Fig. 2. On the left, results from matching a 200 g ref-
(g)
100
50
800
' Control (200 g)
' Concurrent lift (300 9)
Adjustment method
'
'
600
I
i
Control (1000 g) Limit method
Concurrent lift (1000 g)
Limit method
flexion increased (Fig. 2, P
