- Email: [email protected]
Cortical potentials following voluntary and passive finger movements
Electroencephalography and Clinical Neurophysiology, 1980, 5 0 : 2 0 1 - - 2 1 3
201
© Elsevier/North-Holland Scientific Publishers, Ltd.
CORTICAL POTENTIALS FOLLOWING VOLUNTARY AND PASSIVE FINGER MOVEMENTS H. SHIBASAKI 1, G. BARRETT 2, ELISE HALLIDAY and A.M. HALLIDAY 2
Institute of Neurology, The National Hospital, Queen Square, London WC1 (England) (Accepted for publication: May 7, 1980)
In a previous study (Shibasaki et al. 1980) the authors identified 8 components in the movement-related cortical potential, 4 premotion and 4 post-motion. The present study was undertaken to compare the cortical potentials associated with a voluntary brisk finger movement with those evoked by a similar passive movement, in order to clarify the time relationship and functional significance of these components, especially those following the movement. A negative c o m p o n e n t occurring over the frontal region in very close association with the finger movement (N + 50) seemed to be of particular interest, because this is the most consistent and well-defined of the post-motion components, and also because this c o m p o n e n t has been controversial in regard to its time relationship to the movement onset (Gerbrandt et al. 1973; Wilke and Lansing 1973; Shibasaki and Kato 1975; Gerbrandt 1978). It appears probable that some previous studies, by failing to distinguish this frontal c o m p o n e n t from the precentral N--10 peak, may have identified either c o m p o n e n t indiscriminately as the N2 of Vaughan et al. (1968). The present study also aimed to investigate the scalp topography of the cerebral responses I On leave from the Department of Neurology, Neurological Institute, Kyushu University, Fukuoka, Japan, sponsored by the Japan Society for the Promotion of Science and the Royal Society. 2 Member of the External Staff of the Medical Research Council.
evoked by passive movement, which does not appear to have been studied in any detail previously.
Methods Seven healthy right-handed volunteers (1 male and 6 female) aged between 19 and 32 years (mean 22 years) acted as subjects. Full details of the m e t h o d of recording the cortical potentials associated with voluntary finger movement have been given in an earlier paper (Shibasaki et al. 1980) and only significant variations from the earlier procedure will be described here. The movement employed for both voluntary and passive conditions was a brisk extension of the middle finger at the metacarpophalangeal joint. Voluntary movements were repeated at variable intervals exceeding 3 sec at a self-paced rate. The subject was requested to make the movements as briskly as possible. The extent of the movement was approximately 10--20 ° . To elicit passive movement, one end of a soft hemp string was bound round the distal portion of the middle phalanx of the middle finger and the other end of the string was pulled up briskly by the examiner. The string was approximately 10 cm long and kept lightly stretched between passive movements in order to avoid a shock effect which might have been given to the finger if the string was kept slack before being pulled. The move-
202
ments were repeated at intervals exceeding 3 sec at an irregular rate paced by the examiner. The string was pulled up so that the subject's middle finger was extended at the metacarpophalangeal joint by 10--20 °. Return movements to the resting position were made slowly in both voluntary and passive conditions. For both movements, 50 trials formed one session. Four sessions of voluntary movement and 2 sessions of passive movement were given for each hand, and both hands were tested separately. In one subject, only one hand was tested for voluntary movement. A trigger pulse was obtained using a photometer for both voluntary and passive movements. A narrow beam of light was projected onto a photometer, and the middle finger being tested was placed so that raising the finger tip interrupted the beam. When the finger tip was displaced either voluntarily or passively, a signal was produced in the photometer which activated a Digitimer model D4030. For both conditions, the EMG was recorded with a time constant of 0.03 sec by a pair of silver-silver chloride cup electrodes placed approximately 3 cm apart over the muscle which was contracting during the voluntary movement. The EEG was recorded, simultaneously with the EMG from the contracting muscles, with the same montage (Fig. 1) and using the same filter settings as in the previous study (Shibasaki et al. 1980). Eye movements were also monitored as before. The 14-channel EEG, EOG and rectified EMG were input to a PDP12 computer, and were averaged with an opisthochronic averaging program, using the photometer pulse as a trigger. With an ordinate period of 7.5 msec this program provided a window time of 1920 msec covering a period from 1500 msec before to 420 msec after the trigger pulse. Fifty trials were averaged in each session, and later the trials for all sessions of each movement were added together (giving a total of 200 for voluntary movement and 100 for passive movement). The resultant wave forms were plotted on an X-Y plotter
ft, S I I I B A S A K I ET AL
2
Fig. 1. E l e c t r o d e p l a c e m e n t for the EEG and EOG used in the present study. The same m o n t a g e was used in the previous study o n the m o v e m e n t - r e l a t e d cortical potential (Shibasaki et al. 1980).
and stored digitally on magnetic tape. Measurements were made on the individual record by a cursoring program. The peak latency of each component was measured with respect to the p h o t o m e t e r trigger pulse. The amplitude of each c o m p o n e n t was measured from the preceding peak of opposite polarity. In voluntary movement, components were named according to the terminology used in our previous study (Shibasaki et al. 1980), which was based on the polarity and the time interval in msec between the peak of each component and the peak of the averaged, rectified EMG. In passive movement, each recognizable c o m p o n e n t was named
ACTIVE AND PASSIVE MOTOR POTENTIALS
according to its polarity and time interval in msec between the peak of the c o m p o n e n t and t'he p h o t o m e t e r trigger. For illustrative purposes, a grand average of the data from 4 subjects was computed, for each kind of movement. Voluntary and passive movements were compared in terms of the general wave form of potentials, the latency and amplitude of each c o m p o n e n t , and the scalp topography. Results
(I) Cortical potentials associated with voluntary middle finger extension The wave form of the averaged cortical potential associated with voluntary middle finger extension was similar to that reported in the previous study (Shibasaki et al. 1980), although the p h o t o m e t e r was used to obtain a trigger pulse in the present study whereas the EMG onset was used in the previous one. With regard to pre-motion components, an early symmetric, diffuse slow negativity (Bereitschaftspotential, BP) and a later contralateral precentral negative slope (NS'(--500 to --90) were clearly seen (Figs. 2 and 3). The premotion positivity (P--50) was less clearly recognized in the present study. In the grand average record, it is suggested only b y a small positive inflection over the midline and ipsilateral central region following right middle finger extension (Fig. 2) and a slightly earlier ipsilateral positive inflection following the negative peak in the parietal region (Figs. 2 and 3). The m o t o r potential (N--10) was also less clearly seen in most subjects in the present study, and in the grand average records it appeared only as a culminating peak of the NS'(--500 to --90) at the contralateral precentral area (Figs. 2 and 3). The other 4 postm o t i o n components a r e described in more detail in order to compare them with the potentials associated with passive movement. The peak of the averaged, rectified EMG occurred at a mean interval of 49 msec (n = 13, S.D. = 18) before the p h o t o m e t e r trigger, with EMG onset at 143 msec ( n = 1 3 , S ~ . = 26) before the trigger.
203
(1) N + 50. This c o m p o n e n t (Fig. 4) was identified in all records (right and left middle finger extension) of all subjects. The peak occurred at a mean interval of 15 msec (n = 13, S.D. = 30) before the p h o t o m e t e r trigger, b u t 34 msec (n = 13, S.D. = 27) after the EMG peak. Its onset occurred at a mean interval of 83 msec (n = 12, S.D. = 26) before the p h o t o m e t e r trigger, b u t 58 msec (n = 12, S.D. = 30) after the EMG onset. The mean amplitude measured at the midline frontal electrode (Fz') was 3.8 pV ( n = 12, S.D. = 1.2). It was localized over the frontal and frontopolar region, slightly larger on the contralateral side (Figs. 2 and 3). (2) P ÷ 90. This c o m p o n e n t (Fig. 4) was recognized in 8 o u t of 13 records (5 o u t of 7 subjects). The peak occurred at a mean interval of 5 msec (n = 8, S.D. = 28) before the p h o t o m e t e r trigger, b u t 40 msec (n = 8, S.D. = 27) after the EMG peak. Its onset occurred at a mean interval of 82 msec (n = 8, S J). = 38) before the p h o t o m e t e r trigger, b u t 50 msec (n=8, S.D. = 25) after the EMG onset. Therefore, the time relationship of this component to either the p h o t o m e t e r trigger or the EMG was similar to that of N + 50. The mean amplitude measured at the contralateral parietal electrode (P3' and P4') was 4.2 pV (n = 8, S.D. = 1.4). It was localized to the contralateral parietal region, extending with diminishing amplitude across the ipsilateral parietaJ region, and at much smaller amplitude to the precentral region (Figs. 2 and 3). (3) N + 1 6 0 . Over the parietal region, P +9-0 was followed b y a negative wave, which was recognizable in all b u t one of the records for all subjects (Fig. 4). The peak occurred at a mean interval of 97 msec (n = 12, S.D. = 27) after the p h o t o m e t e r trigger and 144 msec (n = 12, S.D. = 29) after the EMG peak. The mean amplitude measured at the contralateral parietal electrode (P3' and P4') was 4.6 pV (n = 12, S~D. = 1.6). It was localized to the parietal region with a contralateral predominance, and extended at much smaller amplitude to the precentral region (Figs. 2 and 3).
204
tt. SHIBASAKI ET AL,
4 subjects Rt M F Ext.
Voluntary
Passive
r'"'"'T'"'"'T"""T'"'"'T"'"T""T"
i....,....I....,....i....,....i....i....i....,....i....,....I...
FI' FZ p
•
F~'
LHM
/
C1'
~]
iIto~v
C~' RHM P"'
/
Pt' Pz'
I
Pg'
/
p~'
/
EOG
EMG I......... I......... I......... [ ......... I ......... [........L.
I.........I.........I.........I.........I.........I.........I...
1500
900
0
300msec
Fig. 2. Grand average records of cortical potentials associated with passive and voluntary extension of the right middle finger from 4 subjects. 100 movements each for passive extension and 200 movements each for voluntary extension. The photometer trigger coincides with time zero.
ACTIVE A N D PASSIVE M O T O R
POTENTIALS
205
(4) P + 300. A diffuse, large positivity was seen f o l l o w i n g t h e N + 160 in all subjects and in 11 o u t o f 13 r e c o r d s (Fig. 4). T h e peak o c c u r r e d at a m e a n interval o f 271 msec
(n = 11, S.D. = 80) a f t e r t h e p h o t o m e t e r trigget a n d 316 msec (n = 11, S 2 . = 76) a f t e r t h e E M G peak. T h e a m p l i t u d e varied f r o m subject t o subject, t h e m e a n a m p l i t u d e m e a s u r e d at
4 subjects Lt M F E x t . Passive
Voluntary
r'""'"I'""'"T'"'"'T'"'"'T""'"'I'"":"T" Fpz' FI' Fz' F2'
LHM C1'
iI
Cz
I0~¥
C2' RHM P3' PI" Pz' P2' P4'
EOG
EMG I..
I......... I......... 1......... I......... I.........'1.........1...
i.J......... J......... J......... i......... t......... L..
1500
900
0
300reset
Fig. 3. Grand average records of cortical potentials associated with passive and voluntary extension of the left middle finger from 4 subjects. 100 movements each for passive and 200 movements each for voluntary extension. The photometer trigger coincides with time zero.
20(-;
H. SHIBASAKI E T A ~ 4 subjects Rt M F Ext. Passive
Voluntary
r""T""'T'"'""r"""T"'"'"i'"'""T N+56
N 70
1 Fz'
IO pV
Cl' ~
N
190
-,, " \ P-270
-.,.. P+300
T P65
EMG
"'
I....,....I....,....I....,....I....,...J....,....I....,....I... o
I....,....I....,....1....,....I....,....I....,....I....,....I...
- 1500
- 900
0 +300msec
Fig. 4. Terminology used to label each recognizable post-motion peak. Selected channels from the grand average records of passive and voluntary right middle finger extension (Fig. 2).
the vertex (Cz) being 10.7 pV (n = 11, S.D. = 5.1). It was maximal at the vertex (Cz) or the contralateral precentral electrode (C1' or C2'), and bilaterally widespread (Figs. 2 and 3). (II) Cortical potentials associated with passive middle finger ex tension Although in the grand average records, there appears to be a low amplitude slow
negative/positive swing in the 1.5 sec period before the onset of movement, especially in right finger movement, there was no significant consistent potential change before the movement onset in the individual records, except for a few records which were contaminated with eye movement artifacts. A sharp negative wave was seen over the frontal and contralateral precentral region
207
ACTIVE AND PASSIVE MOTOR POTENTIALS TABLE I C o m p o n e n t s o f the cortical potential evoked by voluntary and passive m o v e m e n t s (mean value). Voluntary
N+50
Interval f r o m p h o t o m eter trigger (msec) * Interval f r o m the EMG peak (msec) * A m p l i t u d e (pV) ** Maximal location ***
P+90
N+160
--5
+97
--15 +34 3.8 F l' or F2'
+40 4.2 P 3 ' or P4'
P+---3"00
+271
+144 4.6 P3' or P4'
+316 10.7 Cz
Pa~ive
N15
N70
P65
P130
N1-T~
N190
P270
Latency after p h o t o m eter trigger (msec) A m p l i t u d e (pV) ** Maximal l o c a t i o n ***
16 7.6 C1' or C2'
73 ? F l ' or F2 p
65 7.9 P3 s or P4 r
132 13.1
140 7.7 P3 t or P4 t
193 9.4 C1 ~ or C2'
271 11.2 Cz
C1 t or
C2 t
* Mean interval b e t w e e n the peak of each c o m p o n e n t and the p h o t o m e t e r trigger or the EMG peak. ' - - ' indicates occurrence of the peak before the p h o t o m e t e r trigger or EMG peak, and ' + ' after. ** Measured f r o m the preceding peak of opposite polarity. *** Contralateral to the m o v e m e n t whenever asymmetric.
coinciding with the onset of passive movement, and followed b y several positive and negative peaks. This initial negativity was among the most consistent and prominent features of the response. Closer observation showed that this negativity was longer in its duration over the frontal region than over the precentral region and that the frontal wave appeared to be a composite of 2 negative peaks occurring in very close time relationship (Figs. 2, 3 and 4). At least 7 distinct peaks were identified on the basis of different peak latencies and scalp topography (Table I). (1) N15. Immediately after the onset of passive movement, a sharp negative wave occurred over the contralateral precentral region (Fig. 4). This c o m p o n e n t was recognized in all b u t one record for all subjects. The peak occurred at a mean latency of 16 msec (n = 13, S.D. = 20) after the p h o t o m e t e r trigger and the onset preceded the trigger by a mean interval of 36 msec (n = 13, S.D. = 18). The mean amplitude measured at the contralateral precentral electrode (C1' and C2') was 7.6 pV (n = 13, S.D. = 2.3). It was maximal at the contralateral precentral electrode halfway
between the vertex and the hand m o t o r area (C1' or C2'), extending more anteriorly than posteriorly (Figs. 2 and 3). Over the precentral region, it was lateralized to the contralateral hemisphere with only a small extension to the ipsilateral hemisphere. Over the frontal region, however, it was seen bilaterally, although slightly larger on the contralateral side. Posteriorly, it extended at much smaller amplitude to the parietal region again with a slight contralateral predominance. (2) P65. A positive peak occurred over the parietal region just after the movement onset (Fig. 4). This c o m p o n e n t was identified in 11 o u t of 14 records (6 o u t of 7 subjects). The mean peak latency after the p h o t o m e t e r trigger was 65 msec (n = 11, S.D. = 22). The mean amplitude measured at the contralateral parietal' electrode (P3' and P4') was 7.9 pV (n = 11, S.D. = 3.5). It was localized to the contralateral parietal region, extending at diminishing amplitude across to the ipsilateral side (Figs. 2 and 3). Forward extension was n o t detectable due to the succeeding positivity with a different peak latency in the precentral region.
208
(3) N70. This is the second of a composite of 2 negative peaks occurring over the frontal region immediately after the onset of passive movement (Fig. 4). This peak could be clearly distinguished from the initial negative peak (N15) in 8 out of 14 records (5 out of 7 subjects). Its peak occurred at a mean latency of 73 msec (n = 8, S.D. -- 18) after the photometer trigger, which is within 8 msec of the mean peak latency of the parietal P65. The onset latency and amplitude could not be determined because this peak occurred so close to N15, forming a composite double negative peak over the frontal region. The N70 component appeared to be localized to the frontal region with slight contralateral predominance, extending at smaller amplitude to the frontopolar and also to the precentral region (Figs. 2 and 3). In the precentral region, however, it is more clearly seen at the midline and adjoining ipsilateral electrode. (4) P130. This prominent peak (Fig. 4) was interpreted as being distinct from P65 because of its different timing and scalp topography. It was identified in 8 out of 14 records (4 out of 7 subjects). The peak occurred at a mean latency of 132 msec (n = 8, S.D. = 22) after the photometer trigger. Its onset followed the peak of N15 or N70. The mean amplitude measured at the contralateral precentral electrodes (C~' or C2') was 13.1 pV (n = 8, S.D. = 6.8). It was distributed over the precentral and frontal region with contralateral predominance (Figs. 2 and 3) but the lateralisation was more marked centrally than frontally. (5) N140. Over the parietal region, P65 was followed by a negative wave (Fig. 4). This component was identified in 11 out of 14 records (6 out of 7 subjects). The mean peak latency occurred 140 msec (n = 11, S.D. = 30) after the photometer trigger which is within 8 msec of the mean latency of the more anteriorly distributed P130. The mean amplitude measured at the contralateral parietal electrode (Pa' and P4') was 7.7 pV (n = 11, S.D. = 3.2}. It was localized over the contralateral parietal region, extending with diminishing
I! SHIBASAKI ~;'l' :\ amplitude across to the ipsilateral parietal region (Figs. 2 and 3). Forward extension could not be detected due ~o the succeeding negativity of a different peak latency (N190~ recorded in the precentral region. (6) ~,<190. This prominent negative peak (Fig. ,4) appeared to be distinct from N140 on the grounds of its different peak latency and scalp topography. It was identified in 10 out of 14 records (5 out of 7 subjects). The peak occurred at a mean latency of 193 msec (n : 10, S.D. = 29) after the photometer trigger. The onset followed the peak of P130. The mean amplitude measured at the vertex electrode (Cz) where it was largest was 9.4 gV (n = 10, S.D.= 6.7). Like P130, N190 was recorded over the precentral and frontal region bilaterally with a more or less symmetrical distribution about the midline (Figs. 2 and 3). (7) P270. This is a large, diffusely distributed positive wave with a mean peak latency of 271 msec (n = 11, S.D. = 40) after the photometer trigger (Fig. 4). This component was seen in 11 out of 14 records (6 out of 7 subjects). The mean amplitude measured at the vertex electrode (Cz) was 11.2 pV (n = 11, S.D. = 4.9). It was maximal at the vertex and widely distributed in a bilaterally symmetric fashion (Figs. 2 and 3).
(III) Comparison between voluntary and passive movements With regard to averaged potentials before the onset of movement, there was a marked difference in the wave form between voluntary and passive movements (Figs. 2 and 3). In voluntary middle finger extension, a symmetric slow negativity (BP) started 1--1.5 sec before the movement onset, and increased its gradient, especially over the contralateral precentral region, approximately 500 msec before the movement onset (NS' (--500 to --90)). There was no significant potential change before the movement in passive middle finger extension. With regard to averaged potentials after the onset of movement, certain c o m m o n features
ACTIVE AND PASSIVE MOTOR POTENTIALS were noted between voluntary and passive movements, although a greater number of sharper peaks were seen in passive movement than in voluntary movement (Figs. 2 and 3). The initial frontal negative component in voluntary movement (N + 50) was essentially similar to N70 in passive movement in its wave form and scalp topography, although, with respect to the photometer trigger, the peak of N + 50 occurred 88 msec earlier (Table I). A similar relationship was found between P + 90 in voluntary movement and P65 in passive movement with a corresponding discrepancy of 70 msec and also between N + 160 and N140 with a discrepancy of 43 msec. The initial sharp negative peak at the contralateral precentral area consistently seen in passive movement (N15) was never recognized in voluntary movement. A prominent positive-negative complex seen over the precentral and frontal region in passive movement (P130, N190) was not seen in voluntary movement. The later diffuse positive components, P + 300 in voluntary and P270 in passive movement , were similar in scalp topography and latency although the latter peak appeared much sharper.
Discussion With regard to cortical potentials preceding voluntary middle finger extension, th'e present study produced similar results to our previous study (Shibasaki et al. 1980) except for the fact that 2 small components, P--50 and N--10, were less clearly demonstrated in the present experiment. This can be attributed to the different methods of deriving the trigger pulse used in these 2 studies; the EMG onset being used in the previous study as against the interruption of a light beam by the movement in the present one. At any rate, the absence of premotion components in passive movement confirms that these potentials are related to voluntary movement as originally proposed by Kornhuber and Deecke (1965). With regard to the prominent frontal nega-
209 rive component in close association with the movement onset (N + 50), its peak preceded the photometer trigger by 15 msec in the present study. But, since its peak was preceded by the EMG peak by 34 msec and its onset was preceded by the EMG onset by 58 msec, this potential can be considered to occur after movement onset. The present finding of a similar negative potential in the frontal region following passive movement (N70) supports this conclusion, although there are definite differences in regard to precise latency between the two, possibly related to inevitable uncontrolled differences in the active and passive movement conditions. The backward extension of N70 to the midline and ipsilateral precentral electrodes was a feature also noted for N + 50 in our earlier study of active finger extension, using EMG triggering (Shibasaki et al. 1980). This component seems to correspond to the small negative potential found by Bates (1951) on the human contralateral central scalp beginning 20--35 msec, and peaking at 55--75 msec, after the EMG onset. Similar potentials were recorded by Kornhuber and Deecke (1965), Deecke et al. (1976) and Papakostopoulos et al. (1975). Arezzo and Vaughan {1975) and Arezzo et al. (1977), by making epidural and intracortical recordings in the monkey trained to perform self-paced wrist movements, recorded N2 beginning at 90 msec before the EMG onset at the contralateral precentral area, P2 at 150 msec after the EMG onset at the contralateral central area, and P3 at 265 msec after the EMG onset in bilateral area 5. Later, they identified 2 components in their N2 (Arezzo and Vaughan 1980). N2a occurred as the only antecedent phasic component limited to a small region immediately anterior to the central sulcus. N2b appeared as a composite with the preceding N2a in the precentral region, but in isolation more anteriorly, beginning at 15-20 msec after the EMG onset. The N + 50 found in our previous study as well as the present study seems to conform with this N2b in its wave form, time relationship to the movement onset and localization. Arezzo and
2i0 Vaughan (1980) recorded a positive component in the post-central cortex (P2a) with the same timing as N2b, and postulated that the sulcal cortex acts as a dipole source projecting a negative field potential anteriorly and superiorly as N2b and a positive one posteriorly and inferiorly as Pea- In this regard, it is of particular interest that, in the present recordings from the human scalp, the parietal positivity ( P + 9 0 ) occurred with approximately the same timing as the frontal negativity (N + 50). Only a few studies have been reported on human cortical potentials evoked by passive joint movement. Kornhuber and Deecke (1965) recorded a complex consisting of several peaks following manually produced passive fist-clenching or passive wrist movement produced by a string in a similar way to our present experiment. Their early positive component, occurring 90--100 msec after the onset of the experimenter's EMG, was contralaterally predominant, whereas later components were bilaterally symmetric. Rodin et al. (1969) studied the response evoked by passive movement of the index finger using a DC solenoid. They recorded a double-peaked negative potential in the contralateral premotor to anterior parietal lead between 50 and 90 msec after the movement onset, followed by a positive c o m p o n e n t at 180 msec. Papakostopoulos et al. (1974) studied cortical potentials in response to mechanical displacement of the index finger by both cortical and scalp recording. They found a positive wave at 34 msec in the contralateral precentral cortex and at 42 msec in the contralateral postcentral cortex, the latter being larger, followed by a negative wave at 68 msec in the precentral and at 97 msec in the postcentral cortex. However, t h e y did n o t record any corresponding activity from the prefrontal cortex. The early responses recorded in the present study seem to be in conformity with those recorded by Rodin et al. (1969). In particular, the double-peaked negative wave followed by a positive peak recorded by them in the contralateral premotor to anterior parietal
1{. SHIBASAKI ET AL lead appears to correspond to the N15--N70 complex and P130 recorded by us in the f r o n tal leads. The present study on passive movement evoked potentials made use of simultaneous multichannel EEG recording and averaging to delineate the detailed topography of each component. Among the 7 peaks identified in this way, 2 pairs of components of opposit~ polarity, one N70 and P65 and the other P130 and N140, had constituent peaks with approximately the same latency. In both of these pairs, the former peak was recorded from the frontal and precentral electrodes whereas the latter peak was recorded from the parietal electrodes. With regard to their coronal distribution, all these peaks, although bilaterally recorded, were larger over the contralateral hemisphere. These findings seem to be of special interest in view of the current theory of kinesthetic input into the sensorimotor cortex. In extracellular unit recording of responses to muscle nerve stimulation in the baboon (Phillips et al. 1971; Heath et al. 1976), the m o n k e y (Ros6n and Asanuma 1972) and the cat (Zarzecki et al. 1978), group I muscle afferents from the forelimb were shown to project first to area 3a, which is located in the depth of the central sulcus, at a latency of 5--10 msec. Similar results were also found in the m o n k e y following passive movement of the wrist (Yumiya et al. 1974) and stretch of a limb muscle (Lucier et al. 1975). Although a short-latency input to area 4 has been shown to exist in the m o n k e y , the minimum latency being 6--8 msec from the hand muscle (Lemon 1979), it is now believed t h a t neurones in area 4 receive the muscle afferents mainly via area 3a (Lucier et al. 1975; Zarzecki et al. 1978). Assuming that the kinesthetic input in the human also projects to area 3a, it appears reasonable to postulate that N70 and P65 are a negative and positive potential field projected antero-superiorly and posterodnferiorly, respectively, from a dipole source within the central sulcus. Similarly, P130 and N140 could also be the result of the succeeding dipole source of op-
ACTIVE AND PASSIVEMOTOR POTENTIALS posite polarity within the central sulcus. In view of the marked resemblance of the N + 50[P + 90 complex of voluntary movement to the N70/P65 complex of passive movement, it seems reasonable to conclude that the N + 50/P + 90 reflects the kinesthetic feedback from brisk voluntary movement. It might also be related to the long servo-loop from periphery to cortex proposed by Phillips (1969) and Marsden et al. (1972). It has also to be remembered that in voluntary movement, the sensory cortex is thought to receive, in addition to the sensory afferents from peripheral receptors, 'corollary discharges' or non-sensory inputs from the motor cortex via cortico~ortical connections (Evarts 1972; Matthews 1977). However, the resemblance of N70/P65 recorded in passive movement to the N + 50/P + 90 complex suggests a direct sensory input from the periphery. The functional significance of the sharp negative response, recorded over the contralateral precentral area 16 msec after the photometer trigger in passive movement (N15), remains to be resolved. The actual latency may be slightly longer because the photometer pulse used as a reference point will only be initiated a little after the onset of finger displacement. This time lag has not been measured in this study. But, from its similar distribution and latency to the early component of the somatosensory evoked potential to electrical stimulation of peripheral nerves, the N15 component might represent a response of the somatosensory cortex to sensory stimulation. It m a y , for example, represent the response to tactile stimulation of the finger by the pull of the string which produced the passive movement. It should be noted that with the method of producing passive movement used in the present study, skin afferents will inevitably be stimulated along with joint and muscle proprioceptors. Summary In order to clarify the time relationship and functional significance of post-motion corn-
211 ponents of the movement-related cortical potential, averaged cortical potentials associated with voluntary and passive movements were compared mainly with respect to their scalp topography. Fourteen channels of scalp EEG, together with EOG and EMG, were simultaneously recorded in 7 healthy adult subjects while the subject was either repeating a selfpaced brisk extension of a middle finger or while the experimenter was extending the middle finger by pulling up a string attached to the finger. Potentials associated with the movement were averaged opisthochronically in relation to a trigger actuated by the finger interrupting a beam of light. Seven peaks were identified in the passive movement-evoked potential. A sharp negative peak occurred over the contralateral precentral region 16 msec after the photometer trigger (N15). Another negative component (N70) formed a composite of double-peaked negativity with N15 and was seen over the frontal region with a contralateral predominance. A positive peak (P65) was recorded over the contralateral parietal region with a similar latency to N70. This N70/P65 complex has some marked similarities in terms of wave form and spatial relationship with the N + 50/ P + 90 complex recorded with voluntary movement of the same finger. It is postulated that these components may be the projected potential fields from a dipole source within the central sulcus and may represent a kinesthetic feedback from the muscle afferents.
R~sum~ Potentiels corticaux consdcutifs d des mouvements volontaires ou passifs du doigt
Afin d'~lucider les relations temporelles et la signification fonctionnelle des composants 'post-moteurs' du potentiel cortical lib au mouvement, on a compar~ les potentiels corticaux moyenn~s associ~s ~ des mouvements volontaires ~ ceux dfis ~ des mouvements passifs, en particulier du point de vue de leur
212
topographie sur le scalp. On a enregistr~ simultan~ment 14 canaux d'EEG, I d'EOG et 1 d'EMG chez 7 sujets adultes sains, tandis que le sujet, tantSt pratiquait a volont~ une extension rapide du majeur, tantSt subissait cette extension de la part de l'exp~rimentateur qui tirait sur une ficelle attach~e au doigt. Les potentiels associ~s au mouvement 6taient moyenn~s grace ~ un top de synchronisation donn~ par le doigt en interrompant un rayon lumineux. Sept pics ont ~t~ identifies dans le potentiel li~ au mouvement. Un pic raide, n~gatif, apparait dans la r6gion pr~centrale contralat~rale 16 msec apr~s le d~clenchement photo~lectrique (N15). Une autre composante n~gative (N70) formait un complexe n~gatif double pic avec N15 et se recueillait dans la r6gion frontale, avec une predominance contralat~rale. Un pic positif (P65) ~tait enregistr~ dans la r~gion pari~tale contralat~rale avec une latence similaire ~ N70. Ce complexe N70/P65 avait de notables similitudes, du point de vue de sa configuration et de ses relations spatiales avec le complexe N + 50/P + 90 enregistr~ lors du mouvement volontaire du m~me doigt. On propose que ces composants pourraient ~tre dfis ~ un champ de potentiels dont la source serait un dip61e situ~ dans le sillon central et pourraient ainsi traduire une activit~ kinesth6sique en retour issue des aff~rents musculaires.
References Arezzo, J. and Vaughan, Jr., H.G. Cortical potentials associated with voluntary movements in the monkey. Brain Res., 1975, 88: 99--104. Arezzo, J. and Vaughan, Jr., H.G. Intracortical sources and surface topography of the motor potential and somatosensory evoked potential in the monkey. In: H.H. Kornhuber and L. Deecke (Eds.), Motivation, Motor and Sensory Processes of the Brain: Electrical Potentials, Behaviour and Clinical Use, Progr. Brain Res., Vol. 54. Elsevier, Amsterdam, 1980: 77--83. Arezzo, J., Vaughan, Jr., H.G. and Koss, B. Relationship of neuronal activity to gross movementrelated potentials in monkey pre- and postcentral
H. SHIBASAKI ET AI~. cortex. Brain Res., 1977, 132: 362--369. Bates, J.A.V. Electrical activity of the cortex accom panying movement. J. Physiol. (Lond.), 1951. 113: 240--257. Deecke, L., GrSzinger, B. and Kornhuber, H.H. Voluntary finger movement in man: cerebral potentials and theory. Biol. Cybernet., 1976, 2 3 : 9 9 ..... 119. Evarts, E.V. Contrasts between activity of preeentrai and postcentral neurons of cerebral cortex during movement in the monkey. Brain Res., 1972, 40: 25--31. Gerbrandt, L.K. Methodological criteria for the validation of movement-related potentials. In: D.A. Otto (Ed.), Multidisciplinary Perspectives in EventRelated Brain Potential Research. U.S. Environmental Protection Agency, Washington, D.C., ]978: 97--104. Gerbrandt, L.K., Golf, W.R. and Smith, D.B. Distribution of the human average movement potential. Electroenceph. clin. Neurophysiol., 1973, 34: 461--474. Heath, C.J., Hore, J. and Phillips, C.G. Inputs from low threshold muscle and cutaneous afferents of hand and forearm to areas 3a and 3b of baboon's cerebral cortex. J. Physiol. (Lond.), 1976, 257: 199--227. Kornhuber, H.H. und Deecke, L. Hirnpotentialllnderungen bei Willkiirbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pfliigers Arch. ges. Physiol., 1965, 284: 1--17. Lemon, R.N. Short-latency peripheral inputs to the motor cortex in conscious monkeys. Brain Res., 1979, 161: 150--155. Lucier, G.E., Riiegg, D.C. and Wiesendanger, M. Responses of neurons in motor cortex and in area 3A to controlled stretches of forelimb muscles in cebus monkeys. J. Physiol. (Lond.), 1975, 251: 833--853. Marsden, C.D., Merton, P.A. and Morton, H.B. Servo action in human voluntary movement. Nature (Lond.), 1972, 238: 1 4 0 - 1 4 3 . Matthews, P.B.C. Muscle afferents and kinaesthesia. Brit. reed. Bull., 1977, 33: 137--142. Oldfield, R.C. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia, 1971, 9: 97--113. Papakostopoulos, D., Cooper, R. and Crow, H.J. Cortical potentials evoked by finger displacement in man. Nature (Lond.), 1974, 252: 582--584. Papakostopoulos, D., Cooper, R. and Crow, H.J. Inhibition of cortical evoked potentials and sensation by self-initiated movement in man. Nature (Lond.), 1975, 258: 321--324. Phillips, C.G. Motor apparatus of the baboon's hand (The Ferrier Lecture, 1968). Proc. roy. Soc. B, 1969, 173: 141--174.
ACTIVE AND PASSIVE MOTOR POTENTIALS Phillips, C.G., Powell, T.P.S. and Wiesendanger, M. Projection from low-threshold muscle afferents of hand and forearm to area 3a of baboon's cortex. J. Physiol. (Lond.), 1971, 217: 419--446. Rodin, E., Wasson, S. and Porzak, J. Objective evaluation of joint sense and touch in the human. Neurology (Minneap.), 1969, 19: 247--257. Ros~n, I. and Asanuma, H. Peripheral afferent inputs to the forelimb area of the monkey m o t o r cortex: input-output relations. Exp. Brain Res., 1972, 14: 257--273. Shibasaki, H. and Kato, M. Movement-associated cortical potentials with unilateral and bilateral simultaneous hand movement. J. Neurol., 1975, 208: 191--199. Shibasaki, H., Barrett, G., Halliday, E. and Halliday, A.M. Components of the movement-related cortical potential and their scalp topography. Elec-
213 troenceph, clin. Neurophysiol., 1980, 49: 213-226. Vaughan, Jr., H.G., Costa, L.D. and Ritter, W. Topography of the human m o t o r potential. Electroenceph. clin. Neurophysiol., 1968, 25: 1--10. Wilke, J.T. and Lansing, R.W. Variations in the m o t o r potential with force exerted during voluntary arm movements in man. Electroenceph. clin. Neurophysiol., 1973, 35: 259--265. Yumiya, H., Kubota, K. and Asanuma, H. Activities of neurons in area 3a of the cerebral cortex during voluntary movements in the monkey. Brain Res., 1974, 78: 169--177. Zarzecki, P., Shinoda, Y. and Asanuma, H. Projection from area 3a to the m o t o r cortex by neurons activated from group I muscle afferents. Exp. Brain Res., 1978, 33: 269--282.
201
© Elsevier/North-Holland Scientific Publishers, Ltd.
CORTICAL POTENTIALS FOLLOWING VOLUNTARY AND PASSIVE FINGER MOVEMENTS H. SHIBASAKI 1, G. BARRETT 2, ELISE HALLIDAY and A.M. HALLIDAY 2
Institute of Neurology, The National Hospital, Queen Square, London WC1 (England) (Accepted for publication: May 7, 1980)
In a previous study (Shibasaki et al. 1980) the authors identified 8 components in the movement-related cortical potential, 4 premotion and 4 post-motion. The present study was undertaken to compare the cortical potentials associated with a voluntary brisk finger movement with those evoked by a similar passive movement, in order to clarify the time relationship and functional significance of these components, especially those following the movement. A negative c o m p o n e n t occurring over the frontal region in very close association with the finger movement (N + 50) seemed to be of particular interest, because this is the most consistent and well-defined of the post-motion components, and also because this c o m p o n e n t has been controversial in regard to its time relationship to the movement onset (Gerbrandt et al. 1973; Wilke and Lansing 1973; Shibasaki and Kato 1975; Gerbrandt 1978). It appears probable that some previous studies, by failing to distinguish this frontal c o m p o n e n t from the precentral N--10 peak, may have identified either c o m p o n e n t indiscriminately as the N2 of Vaughan et al. (1968). The present study also aimed to investigate the scalp topography of the cerebral responses I On leave from the Department of Neurology, Neurological Institute, Kyushu University, Fukuoka, Japan, sponsored by the Japan Society for the Promotion of Science and the Royal Society. 2 Member of the External Staff of the Medical Research Council.
evoked by passive movement, which does not appear to have been studied in any detail previously.
Methods Seven healthy right-handed volunteers (1 male and 6 female) aged between 19 and 32 years (mean 22 years) acted as subjects. Full details of the m e t h o d of recording the cortical potentials associated with voluntary finger movement have been given in an earlier paper (Shibasaki et al. 1980) and only significant variations from the earlier procedure will be described here. The movement employed for both voluntary and passive conditions was a brisk extension of the middle finger at the metacarpophalangeal joint. Voluntary movements were repeated at variable intervals exceeding 3 sec at a self-paced rate. The subject was requested to make the movements as briskly as possible. The extent of the movement was approximately 10--20 ° . To elicit passive movement, one end of a soft hemp string was bound round the distal portion of the middle phalanx of the middle finger and the other end of the string was pulled up briskly by the examiner. The string was approximately 10 cm long and kept lightly stretched between passive movements in order to avoid a shock effect which might have been given to the finger if the string was kept slack before being pulled. The move-
202
ments were repeated at intervals exceeding 3 sec at an irregular rate paced by the examiner. The string was pulled up so that the subject's middle finger was extended at the metacarpophalangeal joint by 10--20 °. Return movements to the resting position were made slowly in both voluntary and passive conditions. For both movements, 50 trials formed one session. Four sessions of voluntary movement and 2 sessions of passive movement were given for each hand, and both hands were tested separately. In one subject, only one hand was tested for voluntary movement. A trigger pulse was obtained using a photometer for both voluntary and passive movements. A narrow beam of light was projected onto a photometer, and the middle finger being tested was placed so that raising the finger tip interrupted the beam. When the finger tip was displaced either voluntarily or passively, a signal was produced in the photometer which activated a Digitimer model D4030. For both conditions, the EMG was recorded with a time constant of 0.03 sec by a pair of silver-silver chloride cup electrodes placed approximately 3 cm apart over the muscle which was contracting during the voluntary movement. The EEG was recorded, simultaneously with the EMG from the contracting muscles, with the same montage (Fig. 1) and using the same filter settings as in the previous study (Shibasaki et al. 1980). Eye movements were also monitored as before. The 14-channel EEG, EOG and rectified EMG were input to a PDP12 computer, and were averaged with an opisthochronic averaging program, using the photometer pulse as a trigger. With an ordinate period of 7.5 msec this program provided a window time of 1920 msec covering a period from 1500 msec before to 420 msec after the trigger pulse. Fifty trials were averaged in each session, and later the trials for all sessions of each movement were added together (giving a total of 200 for voluntary movement and 100 for passive movement). The resultant wave forms were plotted on an X-Y plotter
ft, S I I I B A S A K I ET AL
2
Fig. 1. E l e c t r o d e p l a c e m e n t for the EEG and EOG used in the present study. The same m o n t a g e was used in the previous study o n the m o v e m e n t - r e l a t e d cortical potential (Shibasaki et al. 1980).
and stored digitally on magnetic tape. Measurements were made on the individual record by a cursoring program. The peak latency of each component was measured with respect to the p h o t o m e t e r trigger pulse. The amplitude of each c o m p o n e n t was measured from the preceding peak of opposite polarity. In voluntary movement, components were named according to the terminology used in our previous study (Shibasaki et al. 1980), which was based on the polarity and the time interval in msec between the peak of each component and the peak of the averaged, rectified EMG. In passive movement, each recognizable c o m p o n e n t was named
ACTIVE AND PASSIVE MOTOR POTENTIALS
according to its polarity and time interval in msec between the peak of the c o m p o n e n t and t'he p h o t o m e t e r trigger. For illustrative purposes, a grand average of the data from 4 subjects was computed, for each kind of movement. Voluntary and passive movements were compared in terms of the general wave form of potentials, the latency and amplitude of each c o m p o n e n t , and the scalp topography. Results
(I) Cortical potentials associated with voluntary middle finger extension The wave form of the averaged cortical potential associated with voluntary middle finger extension was similar to that reported in the previous study (Shibasaki et al. 1980), although the p h o t o m e t e r was used to obtain a trigger pulse in the present study whereas the EMG onset was used in the previous one. With regard to pre-motion components, an early symmetric, diffuse slow negativity (Bereitschaftspotential, BP) and a later contralateral precentral negative slope (NS'(--500 to --90) were clearly seen (Figs. 2 and 3). The premotion positivity (P--50) was less clearly recognized in the present study. In the grand average record, it is suggested only b y a small positive inflection over the midline and ipsilateral central region following right middle finger extension (Fig. 2) and a slightly earlier ipsilateral positive inflection following the negative peak in the parietal region (Figs. 2 and 3). The m o t o r potential (N--10) was also less clearly seen in most subjects in the present study, and in the grand average records it appeared only as a culminating peak of the NS'(--500 to --90) at the contralateral precentral area (Figs. 2 and 3). The other 4 postm o t i o n components a r e described in more detail in order to compare them with the potentials associated with passive movement. The peak of the averaged, rectified EMG occurred at a mean interval of 49 msec (n = 13, S.D. = 18) before the p h o t o m e t e r trigger, with EMG onset at 143 msec ( n = 1 3 , S ~ . = 26) before the trigger.
203
(1) N + 50. This c o m p o n e n t (Fig. 4) was identified in all records (right and left middle finger extension) of all subjects. The peak occurred at a mean interval of 15 msec (n = 13, S.D. = 30) before the p h o t o m e t e r trigger, b u t 34 msec (n = 13, S.D. = 27) after the EMG peak. Its onset occurred at a mean interval of 83 msec (n = 12, S.D. = 26) before the p h o t o m e t e r trigger, b u t 58 msec (n = 12, S.D. = 30) after the EMG onset. The mean amplitude measured at the midline frontal electrode (Fz') was 3.8 pV ( n = 12, S.D. = 1.2). It was localized over the frontal and frontopolar region, slightly larger on the contralateral side (Figs. 2 and 3). (2) P ÷ 90. This c o m p o n e n t (Fig. 4) was recognized in 8 o u t of 13 records (5 o u t of 7 subjects). The peak occurred at a mean interval of 5 msec (n = 8, S.D. = 28) before the p h o t o m e t e r trigger, b u t 40 msec (n = 8, S.D. = 27) after the EMG peak. Its onset occurred at a mean interval of 82 msec (n = 8, S J). = 38) before the p h o t o m e t e r trigger, b u t 50 msec (n=8, S.D. = 25) after the EMG onset. Therefore, the time relationship of this component to either the p h o t o m e t e r trigger or the EMG was similar to that of N + 50. The mean amplitude measured at the contralateral parietal electrode (P3' and P4') was 4.2 pV (n = 8, S.D. = 1.4). It was localized to the contralateral parietal region, extending with diminishing amplitude across the ipsilateral parietaJ region, and at much smaller amplitude to the precentral region (Figs. 2 and 3). (3) N + 1 6 0 . Over the parietal region, P +9-0 was followed b y a negative wave, which was recognizable in all b u t one of the records for all subjects (Fig. 4). The peak occurred at a mean interval of 97 msec (n = 12, S.D. = 27) after the p h o t o m e t e r trigger and 144 msec (n = 12, S.D. = 29) after the EMG peak. The mean amplitude measured at the contralateral parietal electrode (P3' and P4') was 4.6 pV (n = 12, S~D. = 1.6). It was localized to the parietal region with a contralateral predominance, and extended at much smaller amplitude to the precentral region (Figs. 2 and 3).
204
tt. SHIBASAKI ET AL,
4 subjects Rt M F Ext.
Voluntary
Passive
r'"'"'T'"'"'T"""T'"'"'T"'"T""T"
i....,....I....,....i....,....i....i....i....,....i....,....I...
FI' FZ p
•
F~'
LHM
/
C1'
~]
iIto~v
C~' RHM P"'
/
Pt' Pz'
I
Pg'
/
p~'
/
EOG
EMG I......... I......... I......... [ ......... I ......... [........L.
I.........I.........I.........I.........I.........I.........I...
1500
900
0
300msec
Fig. 2. Grand average records of cortical potentials associated with passive and voluntary extension of the right middle finger from 4 subjects. 100 movements each for passive extension and 200 movements each for voluntary extension. The photometer trigger coincides with time zero.
ACTIVE A N D PASSIVE M O T O R
POTENTIALS
205
(4) P + 300. A diffuse, large positivity was seen f o l l o w i n g t h e N + 160 in all subjects and in 11 o u t o f 13 r e c o r d s (Fig. 4). T h e peak o c c u r r e d at a m e a n interval o f 271 msec
(n = 11, S.D. = 80) a f t e r t h e p h o t o m e t e r trigget a n d 316 msec (n = 11, S 2 . = 76) a f t e r t h e E M G peak. T h e a m p l i t u d e varied f r o m subject t o subject, t h e m e a n a m p l i t u d e m e a s u r e d at
4 subjects Lt M F E x t . Passive
Voluntary
r'""'"I'""'"T'"'"'T'"'"'T""'"'I'"":"T" Fpz' FI' Fz' F2'
LHM C1'
iI
Cz
I0~¥
C2' RHM P3' PI" Pz' P2' P4'
EOG
EMG I..
I......... I......... 1......... I......... I.........'1.........1...
i.J......... J......... J......... i......... t......... L..
1500
900
0
300reset
Fig. 3. Grand average records of cortical potentials associated with passive and voluntary extension of the left middle finger from 4 subjects. 100 movements each for passive and 200 movements each for voluntary extension. The photometer trigger coincides with time zero.
20(-;
H. SHIBASAKI E T A ~ 4 subjects Rt M F Ext. Passive
Voluntary
r""T""'T'"'""r"""T"'"'"i'"'""T N+56
N 70
1 Fz'
IO pV
Cl' ~
N
190
-,, " \ P-270
-.,.. P+300
T P65
EMG
"'
I....,....I....,....I....,....I....,...J....,....I....,....I... o
I....,....I....,....1....,....I....,....I....,....I....,....I...
- 1500
- 900
0 +300msec
Fig. 4. Terminology used to label each recognizable post-motion peak. Selected channels from the grand average records of passive and voluntary right middle finger extension (Fig. 2).
the vertex (Cz) being 10.7 pV (n = 11, S.D. = 5.1). It was maximal at the vertex (Cz) or the contralateral precentral electrode (C1' or C2'), and bilaterally widespread (Figs. 2 and 3). (II) Cortical potentials associated with passive middle finger ex tension Although in the grand average records, there appears to be a low amplitude slow
negative/positive swing in the 1.5 sec period before the onset of movement, especially in right finger movement, there was no significant consistent potential change before the movement onset in the individual records, except for a few records which were contaminated with eye movement artifacts. A sharp negative wave was seen over the frontal and contralateral precentral region
207
ACTIVE AND PASSIVE MOTOR POTENTIALS TABLE I C o m p o n e n t s o f the cortical potential evoked by voluntary and passive m o v e m e n t s (mean value). Voluntary
N+50
Interval f r o m p h o t o m eter trigger (msec) * Interval f r o m the EMG peak (msec) * A m p l i t u d e (pV) ** Maximal location ***
P+90
N+160
--5
+97
--15 +34 3.8 F l' or F2'
+40 4.2 P 3 ' or P4'
P+---3"00
+271
+144 4.6 P3' or P4'
+316 10.7 Cz
Pa~ive
N15
N70
P65
P130
N1-T~
N190
P270
Latency after p h o t o m eter trigger (msec) A m p l i t u d e (pV) ** Maximal l o c a t i o n ***
16 7.6 C1' or C2'
73 ? F l ' or F2 p
65 7.9 P3 s or P4 r
132 13.1
140 7.7 P3 t or P4 t
193 9.4 C1 ~ or C2'
271 11.2 Cz
C1 t or
C2 t
* Mean interval b e t w e e n the peak of each c o m p o n e n t and the p h o t o m e t e r trigger or the EMG peak. ' - - ' indicates occurrence of the peak before the p h o t o m e t e r trigger or EMG peak, and ' + ' after. ** Measured f r o m the preceding peak of opposite polarity. *** Contralateral to the m o v e m e n t whenever asymmetric.
coinciding with the onset of passive movement, and followed b y several positive and negative peaks. This initial negativity was among the most consistent and prominent features of the response. Closer observation showed that this negativity was longer in its duration over the frontal region than over the precentral region and that the frontal wave appeared to be a composite of 2 negative peaks occurring in very close time relationship (Figs. 2, 3 and 4). At least 7 distinct peaks were identified on the basis of different peak latencies and scalp topography (Table I). (1) N15. Immediately after the onset of passive movement, a sharp negative wave occurred over the contralateral precentral region (Fig. 4). This c o m p o n e n t was recognized in all b u t one record for all subjects. The peak occurred at a mean latency of 16 msec (n = 13, S.D. = 20) after the p h o t o m e t e r trigger and the onset preceded the trigger by a mean interval of 36 msec (n = 13, S.D. = 18). The mean amplitude measured at the contralateral precentral electrode (C1' and C2') was 7.6 pV (n = 13, S.D. = 2.3). It was maximal at the contralateral precentral electrode halfway
between the vertex and the hand m o t o r area (C1' or C2'), extending more anteriorly than posteriorly (Figs. 2 and 3). Over the precentral region, it was lateralized to the contralateral hemisphere with only a small extension to the ipsilateral hemisphere. Over the frontal region, however, it was seen bilaterally, although slightly larger on the contralateral side. Posteriorly, it extended at much smaller amplitude to the parietal region again with a slight contralateral predominance. (2) P65. A positive peak occurred over the parietal region just after the movement onset (Fig. 4). This c o m p o n e n t was identified in 11 o u t of 14 records (6 o u t of 7 subjects). The mean peak latency after the p h o t o m e t e r trigger was 65 msec (n = 11, S.D. = 22). The mean amplitude measured at the contralateral parietal' electrode (P3' and P4') was 7.9 pV (n = 11, S.D. = 3.5). It was localized to the contralateral parietal region, extending at diminishing amplitude across to the ipsilateral side (Figs. 2 and 3). Forward extension was n o t detectable due to the succeeding positivity with a different peak latency in the precentral region.
208
(3) N70. This is the second of a composite of 2 negative peaks occurring over the frontal region immediately after the onset of passive movement (Fig. 4). This peak could be clearly distinguished from the initial negative peak (N15) in 8 out of 14 records (5 out of 7 subjects). Its peak occurred at a mean latency of 73 msec (n = 8, S.D. -- 18) after the photometer trigger, which is within 8 msec of the mean peak latency of the parietal P65. The onset latency and amplitude could not be determined because this peak occurred so close to N15, forming a composite double negative peak over the frontal region. The N70 component appeared to be localized to the frontal region with slight contralateral predominance, extending at smaller amplitude to the frontopolar and also to the precentral region (Figs. 2 and 3). In the precentral region, however, it is more clearly seen at the midline and adjoining ipsilateral electrode. (4) P130. This prominent peak (Fig. 4) was interpreted as being distinct from P65 because of its different timing and scalp topography. It was identified in 8 out of 14 records (4 out of 7 subjects). The peak occurred at a mean latency of 132 msec (n = 8, S.D. = 22) after the photometer trigger. Its onset followed the peak of N15 or N70. The mean amplitude measured at the contralateral precentral electrodes (C~' or C2') was 13.1 pV (n = 8, S.D. = 6.8). It was distributed over the precentral and frontal region with contralateral predominance (Figs. 2 and 3) but the lateralisation was more marked centrally than frontally. (5) N140. Over the parietal region, P65 was followed by a negative wave (Fig. 4). This component was identified in 11 out of 14 records (6 out of 7 subjects). The mean peak latency occurred 140 msec (n = 11, S.D. = 30) after the photometer trigger which is within 8 msec of the mean latency of the more anteriorly distributed P130. The mean amplitude measured at the contralateral parietal electrode (Pa' and P4') was 7.7 pV (n = 11, S.D. = 3.2}. It was localized over the contralateral parietal region, extending with diminishing
I! SHIBASAKI ~;'l' :\ amplitude across to the ipsilateral parietal region (Figs. 2 and 3). Forward extension could not be detected due ~o the succeeding negativity of a different peak latency (N190~ recorded in the precentral region. (6) ~,<190. This prominent negative peak (Fig. ,4) appeared to be distinct from N140 on the grounds of its different peak latency and scalp topography. It was identified in 10 out of 14 records (5 out of 7 subjects). The peak occurred at a mean latency of 193 msec (n : 10, S.D. = 29) after the photometer trigger. The onset followed the peak of P130. The mean amplitude measured at the vertex electrode (Cz) where it was largest was 9.4 gV (n = 10, S.D.= 6.7). Like P130, N190 was recorded over the precentral and frontal region bilaterally with a more or less symmetrical distribution about the midline (Figs. 2 and 3). (7) P270. This is a large, diffusely distributed positive wave with a mean peak latency of 271 msec (n = 11, S.D. = 40) after the photometer trigger (Fig. 4). This component was seen in 11 out of 14 records (6 out of 7 subjects). The mean amplitude measured at the vertex electrode (Cz) was 11.2 pV (n = 11, S.D. = 4.9). It was maximal at the vertex and widely distributed in a bilaterally symmetric fashion (Figs. 2 and 3).
(III) Comparison between voluntary and passive movements With regard to averaged potentials before the onset of movement, there was a marked difference in the wave form between voluntary and passive movements (Figs. 2 and 3). In voluntary middle finger extension, a symmetric slow negativity (BP) started 1--1.5 sec before the movement onset, and increased its gradient, especially over the contralateral precentral region, approximately 500 msec before the movement onset (NS' (--500 to --90)). There was no significant potential change before the movement in passive middle finger extension. With regard to averaged potentials after the onset of movement, certain c o m m o n features
ACTIVE AND PASSIVE MOTOR POTENTIALS were noted between voluntary and passive movements, although a greater number of sharper peaks were seen in passive movement than in voluntary movement (Figs. 2 and 3). The initial frontal negative component in voluntary movement (N + 50) was essentially similar to N70 in passive movement in its wave form and scalp topography, although, with respect to the photometer trigger, the peak of N + 50 occurred 88 msec earlier (Table I). A similar relationship was found between P + 90 in voluntary movement and P65 in passive movement with a corresponding discrepancy of 70 msec and also between N + 160 and N140 with a discrepancy of 43 msec. The initial sharp negative peak at the contralateral precentral area consistently seen in passive movement (N15) was never recognized in voluntary movement. A prominent positive-negative complex seen over the precentral and frontal region in passive movement (P130, N190) was not seen in voluntary movement. The later diffuse positive components, P + 300 in voluntary and P270 in passive movement , were similar in scalp topography and latency although the latter peak appeared much sharper.
Discussion With regard to cortical potentials preceding voluntary middle finger extension, th'e present study produced similar results to our previous study (Shibasaki et al. 1980) except for the fact that 2 small components, P--50 and N--10, were less clearly demonstrated in the present experiment. This can be attributed to the different methods of deriving the trigger pulse used in these 2 studies; the EMG onset being used in the previous study as against the interruption of a light beam by the movement in the present one. At any rate, the absence of premotion components in passive movement confirms that these potentials are related to voluntary movement as originally proposed by Kornhuber and Deecke (1965). With regard to the prominent frontal nega-
209 rive component in close association with the movement onset (N + 50), its peak preceded the photometer trigger by 15 msec in the present study. But, since its peak was preceded by the EMG peak by 34 msec and its onset was preceded by the EMG onset by 58 msec, this potential can be considered to occur after movement onset. The present finding of a similar negative potential in the frontal region following passive movement (N70) supports this conclusion, although there are definite differences in regard to precise latency between the two, possibly related to inevitable uncontrolled differences in the active and passive movement conditions. The backward extension of N70 to the midline and ipsilateral precentral electrodes was a feature also noted for N + 50 in our earlier study of active finger extension, using EMG triggering (Shibasaki et al. 1980). This component seems to correspond to the small negative potential found by Bates (1951) on the human contralateral central scalp beginning 20--35 msec, and peaking at 55--75 msec, after the EMG onset. Similar potentials were recorded by Kornhuber and Deecke (1965), Deecke et al. (1976) and Papakostopoulos et al. (1975). Arezzo and Vaughan {1975) and Arezzo et al. (1977), by making epidural and intracortical recordings in the monkey trained to perform self-paced wrist movements, recorded N2 beginning at 90 msec before the EMG onset at the contralateral precentral area, P2 at 150 msec after the EMG onset at the contralateral central area, and P3 at 265 msec after the EMG onset in bilateral area 5. Later, they identified 2 components in their N2 (Arezzo and Vaughan 1980). N2a occurred as the only antecedent phasic component limited to a small region immediately anterior to the central sulcus. N2b appeared as a composite with the preceding N2a in the precentral region, but in isolation more anteriorly, beginning at 15-20 msec after the EMG onset. The N + 50 found in our previous study as well as the present study seems to conform with this N2b in its wave form, time relationship to the movement onset and localization. Arezzo and
2i0 Vaughan (1980) recorded a positive component in the post-central cortex (P2a) with the same timing as N2b, and postulated that the sulcal cortex acts as a dipole source projecting a negative field potential anteriorly and superiorly as N2b and a positive one posteriorly and inferiorly as Pea- In this regard, it is of particular interest that, in the present recordings from the human scalp, the parietal positivity ( P + 9 0 ) occurred with approximately the same timing as the frontal negativity (N + 50). Only a few studies have been reported on human cortical potentials evoked by passive joint movement. Kornhuber and Deecke (1965) recorded a complex consisting of several peaks following manually produced passive fist-clenching or passive wrist movement produced by a string in a similar way to our present experiment. Their early positive component, occurring 90--100 msec after the onset of the experimenter's EMG, was contralaterally predominant, whereas later components were bilaterally symmetric. Rodin et al. (1969) studied the response evoked by passive movement of the index finger using a DC solenoid. They recorded a double-peaked negative potential in the contralateral premotor to anterior parietal lead between 50 and 90 msec after the movement onset, followed by a positive c o m p o n e n t at 180 msec. Papakostopoulos et al. (1974) studied cortical potentials in response to mechanical displacement of the index finger by both cortical and scalp recording. They found a positive wave at 34 msec in the contralateral precentral cortex and at 42 msec in the contralateral postcentral cortex, the latter being larger, followed by a negative wave at 68 msec in the precentral and at 97 msec in the postcentral cortex. However, t h e y did n o t record any corresponding activity from the prefrontal cortex. The early responses recorded in the present study seem to be in conformity with those recorded by Rodin et al. (1969). In particular, the double-peaked negative wave followed by a positive peak recorded by them in the contralateral premotor to anterior parietal
1{. SHIBASAKI ET AL lead appears to correspond to the N15--N70 complex and P130 recorded by us in the f r o n tal leads. The present study on passive movement evoked potentials made use of simultaneous multichannel EEG recording and averaging to delineate the detailed topography of each component. Among the 7 peaks identified in this way, 2 pairs of components of opposit~ polarity, one N70 and P65 and the other P130 and N140, had constituent peaks with approximately the same latency. In both of these pairs, the former peak was recorded from the frontal and precentral electrodes whereas the latter peak was recorded from the parietal electrodes. With regard to their coronal distribution, all these peaks, although bilaterally recorded, were larger over the contralateral hemisphere. These findings seem to be of special interest in view of the current theory of kinesthetic input into the sensorimotor cortex. In extracellular unit recording of responses to muscle nerve stimulation in the baboon (Phillips et al. 1971; Heath et al. 1976), the m o n k e y (Ros6n and Asanuma 1972) and the cat (Zarzecki et al. 1978), group I muscle afferents from the forelimb were shown to project first to area 3a, which is located in the depth of the central sulcus, at a latency of 5--10 msec. Similar results were also found in the m o n k e y following passive movement of the wrist (Yumiya et al. 1974) and stretch of a limb muscle (Lucier et al. 1975). Although a short-latency input to area 4 has been shown to exist in the m o n k e y , the minimum latency being 6--8 msec from the hand muscle (Lemon 1979), it is now believed t h a t neurones in area 4 receive the muscle afferents mainly via area 3a (Lucier et al. 1975; Zarzecki et al. 1978). Assuming that the kinesthetic input in the human also projects to area 3a, it appears reasonable to postulate that N70 and P65 are a negative and positive potential field projected antero-superiorly and posterodnferiorly, respectively, from a dipole source within the central sulcus. Similarly, P130 and N140 could also be the result of the succeeding dipole source of op-
ACTIVE AND PASSIVEMOTOR POTENTIALS posite polarity within the central sulcus. In view of the marked resemblance of the N + 50[P + 90 complex of voluntary movement to the N70/P65 complex of passive movement, it seems reasonable to conclude that the N + 50/P + 90 reflects the kinesthetic feedback from brisk voluntary movement. It might also be related to the long servo-loop from periphery to cortex proposed by Phillips (1969) and Marsden et al. (1972). It has also to be remembered that in voluntary movement, the sensory cortex is thought to receive, in addition to the sensory afferents from peripheral receptors, 'corollary discharges' or non-sensory inputs from the motor cortex via cortico~ortical connections (Evarts 1972; Matthews 1977). However, the resemblance of N70/P65 recorded in passive movement to the N + 50/P + 90 complex suggests a direct sensory input from the periphery. The functional significance of the sharp negative response, recorded over the contralateral precentral area 16 msec after the photometer trigger in passive movement (N15), remains to be resolved. The actual latency may be slightly longer because the photometer pulse used as a reference point will only be initiated a little after the onset of finger displacement. This time lag has not been measured in this study. But, from its similar distribution and latency to the early component of the somatosensory evoked potential to electrical stimulation of peripheral nerves, the N15 component might represent a response of the somatosensory cortex to sensory stimulation. It m a y , for example, represent the response to tactile stimulation of the finger by the pull of the string which produced the passive movement. It should be noted that with the method of producing passive movement used in the present study, skin afferents will inevitably be stimulated along with joint and muscle proprioceptors. Summary In order to clarify the time relationship and functional significance of post-motion corn-
211 ponents of the movement-related cortical potential, averaged cortical potentials associated with voluntary and passive movements were compared mainly with respect to their scalp topography. Fourteen channels of scalp EEG, together with EOG and EMG, were simultaneously recorded in 7 healthy adult subjects while the subject was either repeating a selfpaced brisk extension of a middle finger or while the experimenter was extending the middle finger by pulling up a string attached to the finger. Potentials associated with the movement were averaged opisthochronically in relation to a trigger actuated by the finger interrupting a beam of light. Seven peaks were identified in the passive movement-evoked potential. A sharp negative peak occurred over the contralateral precentral region 16 msec after the photometer trigger (N15). Another negative component (N70) formed a composite of double-peaked negativity with N15 and was seen over the frontal region with a contralateral predominance. A positive peak (P65) was recorded over the contralateral parietal region with a similar latency to N70. This N70/P65 complex has some marked similarities in terms of wave form and spatial relationship with the N + 50/ P + 90 complex recorded with voluntary movement of the same finger. It is postulated that these components may be the projected potential fields from a dipole source within the central sulcus and may represent a kinesthetic feedback from the muscle afferents.
R~sum~ Potentiels corticaux consdcutifs d des mouvements volontaires ou passifs du doigt
Afin d'~lucider les relations temporelles et la signification fonctionnelle des composants 'post-moteurs' du potentiel cortical lib au mouvement, on a compar~ les potentiels corticaux moyenn~s associ~s ~ des mouvements volontaires ~ ceux dfis ~ des mouvements passifs, en particulier du point de vue de leur
212
topographie sur le scalp. On a enregistr~ simultan~ment 14 canaux d'EEG, I d'EOG et 1 d'EMG chez 7 sujets adultes sains, tandis que le sujet, tantSt pratiquait a volont~ une extension rapide du majeur, tantSt subissait cette extension de la part de l'exp~rimentateur qui tirait sur une ficelle attach~e au doigt. Les potentiels associ~s au mouvement 6taient moyenn~s grace ~ un top de synchronisation donn~ par le doigt en interrompant un rayon lumineux. Sept pics ont ~t~ identifies dans le potentiel li~ au mouvement. Un pic raide, n~gatif, apparait dans la r6gion pr~centrale contralat~rale 16 msec apr~s le d~clenchement photo~lectrique (N15). Une autre composante n~gative (N70) formait un complexe n~gatif double pic avec N15 et se recueillait dans la r6gion frontale, avec une predominance contralat~rale. Un pic positif (P65) ~tait enregistr~ dans la r~gion pari~tale contralat~rale avec une latence similaire ~ N70. Ce complexe N70/P65 avait de notables similitudes, du point de vue de sa configuration et de ses relations spatiales avec le complexe N + 50/P + 90 enregistr~ lors du mouvement volontaire du m~me doigt. On propose que ces composants pourraient ~tre dfis ~ un champ de potentiels dont la source serait un dip61e situ~ dans le sillon central et pourraient ainsi traduire une activit~ kinesth6sique en retour issue des aff~rents musculaires.
References Arezzo, J. and Vaughan, Jr., H.G. Cortical potentials associated with voluntary movements in the monkey. Brain Res., 1975, 88: 99--104. Arezzo, J. and Vaughan, Jr., H.G. Intracortical sources and surface topography of the motor potential and somatosensory evoked potential in the monkey. In: H.H. Kornhuber and L. Deecke (Eds.), Motivation, Motor and Sensory Processes of the Brain: Electrical Potentials, Behaviour and Clinical Use, Progr. Brain Res., Vol. 54. Elsevier, Amsterdam, 1980: 77--83. Arezzo, J., Vaughan, Jr., H.G. and Koss, B. Relationship of neuronal activity to gross movementrelated potentials in monkey pre- and postcentral
H. SHIBASAKI ET AI~. cortex. Brain Res., 1977, 132: 362--369. Bates, J.A.V. Electrical activity of the cortex accom panying movement. J. Physiol. (Lond.), 1951. 113: 240--257. Deecke, L., GrSzinger, B. and Kornhuber, H.H. Voluntary finger movement in man: cerebral potentials and theory. Biol. Cybernet., 1976, 2 3 : 9 9 ..... 119. Evarts, E.V. Contrasts between activity of preeentrai and postcentral neurons of cerebral cortex during movement in the monkey. Brain Res., 1972, 40: 25--31. Gerbrandt, L.K. Methodological criteria for the validation of movement-related potentials. In: D.A. Otto (Ed.), Multidisciplinary Perspectives in EventRelated Brain Potential Research. U.S. Environmental Protection Agency, Washington, D.C., ]978: 97--104. Gerbrandt, L.K., Golf, W.R. and Smith, D.B. Distribution of the human average movement potential. Electroenceph. clin. Neurophysiol., 1973, 34: 461--474. Heath, C.J., Hore, J. and Phillips, C.G. Inputs from low threshold muscle and cutaneous afferents of hand and forearm to areas 3a and 3b of baboon's cerebral cortex. J. Physiol. (Lond.), 1976, 257: 199--227. Kornhuber, H.H. und Deecke, L. Hirnpotentialllnderungen bei Willkiirbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pfliigers Arch. ges. Physiol., 1965, 284: 1--17. Lemon, R.N. Short-latency peripheral inputs to the motor cortex in conscious monkeys. Brain Res., 1979, 161: 150--155. Lucier, G.E., Riiegg, D.C. and Wiesendanger, M. Responses of neurons in motor cortex and in area 3A to controlled stretches of forelimb muscles in cebus monkeys. J. Physiol. (Lond.), 1975, 251: 833--853. Marsden, C.D., Merton, P.A. and Morton, H.B. Servo action in human voluntary movement. Nature (Lond.), 1972, 238: 1 4 0 - 1 4 3 . Matthews, P.B.C. Muscle afferents and kinaesthesia. Brit. reed. Bull., 1977, 33: 137--142. Oldfield, R.C. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia, 1971, 9: 97--113. Papakostopoulos, D., Cooper, R. and Crow, H.J. Cortical potentials evoked by finger displacement in man. Nature (Lond.), 1974, 252: 582--584. Papakostopoulos, D., Cooper, R. and Crow, H.J. Inhibition of cortical evoked potentials and sensation by self-initiated movement in man. Nature (Lond.), 1975, 258: 321--324. Phillips, C.G. Motor apparatus of the baboon's hand (The Ferrier Lecture, 1968). Proc. roy. Soc. B, 1969, 173: 141--174.
ACTIVE AND PASSIVE MOTOR POTENTIALS Phillips, C.G., Powell, T.P.S. and Wiesendanger, M. Projection from low-threshold muscle afferents of hand and forearm to area 3a of baboon's cortex. J. Physiol. (Lond.), 1971, 217: 419--446. Rodin, E., Wasson, S. and Porzak, J. Objective evaluation of joint sense and touch in the human. Neurology (Minneap.), 1969, 19: 247--257. Ros~n, I. and Asanuma, H. Peripheral afferent inputs to the forelimb area of the monkey m o t o r cortex: input-output relations. Exp. Brain Res., 1972, 14: 257--273. Shibasaki, H. and Kato, M. Movement-associated cortical potentials with unilateral and bilateral simultaneous hand movement. J. Neurol., 1975, 208: 191--199. Shibasaki, H., Barrett, G., Halliday, E. and Halliday, A.M. Components of the movement-related cortical potential and their scalp topography. Elec-
213 troenceph, clin. Neurophysiol., 1980, 49: 213-226. Vaughan, Jr., H.G., Costa, L.D. and Ritter, W. Topography of the human m o t o r potential. Electroenceph. clin. Neurophysiol., 1968, 25: 1--10. Wilke, J.T. and Lansing, R.W. Variations in the m o t o r potential with force exerted during voluntary arm movements in man. Electroenceph. clin. Neurophysiol., 1973, 35: 259--265. Yumiya, H., Kubota, K. and Asanuma, H. Activities of neurons in area 3a of the cerebral cortex during voluntary movements in the monkey. Brain Res., 1974, 78: 169--177. Zarzecki, P., Shinoda, Y. and Asanuma, H. Projection from area 3a to the m o t o r cortex by neurons activated from group I muscle afferents. Exp. Brain Res., 1978, 33: 269--282.