- Email: [email protected]
Cerebral potentials preceding voluntary toe, knee and hip movements and their vectors in human precentral gyrus
BrainResearch, 376(1986)175-179 Elsevier
175
BRE21587
Cerebral potentials preceding voluntary toe, knee and hip movements and their vectors in human precentral gyrus JORGEN BOSCHERT and LODER DEECKE*
Abteilung Neurologie, Universitiit Ulm, D-7900 Ulm (F.R.G.) and Neurologische Universitiitsklinik Wien, A-1097 Vienna (Austria) (Accepted February 18th, 1986)
Key words: Bereitschaftspotential - - readiness potential - - voluntary movement - - lower extremity - precentral gyms - - motor homunculus
In this study, the brain potentials related to voluntary, self-paced plantarflexions of the left toes, flexion of the right knee and isometric extensions of the left hip were examined in 3 groups of 10 subjects each. In the first half of the foreperiod, the Bereitschaftspotential (BP) or readiness potential for all movements was symmetrically distributed over both hemispheres. For toe and knee movements, an ipsilateral preponderance of the BP developed in the later foreperiod, which was statisticallysignificant for toe movements. For hip movements, BP topography was symmetrical during the entire foreperiod including its second half. Since finger, hand or shoulder movements of previous experiments show a strong contralateral preponderance of the BP, the results are discussed as further support of the hypothesis that the lateralisation of the BP is due to the orientation of the precentral electrical field vector generated by an active source in the MI motor cortex. Thus, part of the somatotopic representation of the human precentral gyms can be mapped by this non-invasive means: upper limb movements are located on the convexity; toe, foot and knee movements are generated on the mesial cortex between the hemispheres; and hip movements seem to be located at the mantle edge.
The Bereitschaftspotential (BP) or readiness p o tential is a slow negativity preceding voluntary finger m o v e m e n t by 1 s o r m o r e 14,15. W h i l e it starts symmetrically over b o t h hemispheres even prior to unilateral m o v e m e n t s , in the later f o r e p e r i o d the BP prior to u p p e r limb m o v e m e n t s exhibits a contralateral p r e p o n d e r a n c e 6. T h e m o t o r potential (MP) 9A° starting about 60 ms prior to the onset of m o v e m e n t further contributes to the total contralateral p r e p o n derance of negativity. A positive potential, the premotion positivity (PMP, latency 90 ms) is usually seen when previously active areas start to relax (cf. refs. 11, 12). These include the fronto-central midline with the s u p p l e m e n t a r y m o t o r area ( S M A ) as well as parietal and ipsilateral p r e c e n t r a l areas. The BP associated with foot m o v e m e n t s surprisingly showed a paradoxical lateralisation, i.e. an ipsilateral p r e p o n d e r a n c e 4,5. Such ipsilateral p r e p o n d e r ance of the BP was also found prior to toe movements 2. Controversial reports in the literature in-
elude some authors, who did n o t find any side difference of the BP prior to foot dorsiflexions 18 and others, who even r e p o r t e d on a contralateral p r e p o n d e r ance of the BP prior to foot dorsiflexion 19. In reviewing the literature, Brunia et al. n pointed to the fact that authors reporting on an ipsilateral BP p r e p o n d e r ance for lower limb m o v e m e n t used plantarflexion, whereas authors who either found a symmetric BP or a contralateral p r e p o n d e r a n c e e m p l o y e d dorsiflexion. Thus, they supposed that the direction of movement might influence BP lateralisation. H o w e v e r , when comparing plantar flexions and dorsiflexions of the toes within subjects in a r a n d o m i z e d fashion, B0schert et al. 3 found an ipsilateral BP p r e p o n d e r ance in both cases without any difference between conditions. Thus, lateralisation of the BP does not d e p e n d on the direction of the m o v e m e n t but rather on the locus of m o v e m e n t generation in the p r i m a r y m o t o r cortex (MI). Activation of mesial parts of M I (e.g. toe or
* Present address: Neurologische Universitfitsklinik, Lazarettgasse 14, A-1090 Wien, Austria. Correspondence: L. Deecke, University of Vienna, Neurological Clinic, Lazarettgasse 14, A-1097 Vienna, Austria.
176 tralateral leg and the trunk. We therefore decided to employ brisk isometric contractions of the gluteal muscles. After some training, using the E M G as a monitor as in biofeedback, subjects were capable of confining these to only one side (left). Each session (with the third group of subjects) consisted of 128 artifact-free trials. Subjects fixed their eyes on a designated point straight ahead of them and avoided blinking or any other movement. Trials contaminated by artifacts were excluded by the experimenters on observation of both the subject and his E E G . Movements were performed at irregular intervals of 4-12 s. Beckman Ag/AgC1 non-polarizing electrodes, 16 m m in diameter, were affixed to the scalp using EC 2 electrode cream after gently rubbing the skin with sand paper or similar tools. Recording positions included left precentral, mid-precentral and right precentral leads (C 1, C z and C 2 of the 10/20 system), referenced against linked earlobes. Impedances were usually below 1 kff2. In addition, bipolar contralateral vs ipsilateral recordings were derived by computer (left vs right precentral for right knee flexions, and right vs left precentral for left toe or hip movements). The E O G was recorded between electrodes affixed to the medial upper and the lateral lower orbital rim. The
foot representation) produces a dipole pointing towards the ipsilateral hemisphere. Consequently, the BP shows an ipsilateral preponderance. Activity on the convexity of MI as with finger movements obviously elicits dipoles directed towards the contralateral hemisphere, thus giving rise to a contralateral BP preponderance. If the active focus is at the mantle edge, the dipole would point straight upward in the sagittal plane, causing a symmetric BP throughout the foreperiod. It was the purpose of the present study to test for this hypothesis by examining the premovement potentials prior to unilateral toe, knee and hip movements. The experiments were performed on 3 groups of subjects. Each group consisted of 5 female and 5 male healthy paid students, aged 18-30 years. Subjects reclined in a chair and performed unilateral movements of the left toes, the right knee or the left hip muscles in a voluntary, self-paced manner. Toe movements were left toe plantarflexions, and knee movements were flexions of the right knee. In both groups, 200 artifact-free movements were collected per subject. For hip movements, pilot experiments revealed that isotonic movements around the hip joint provoked compensating movements of the con-
Toes ips.
Knee
Hip
J
~,.-J Cz
J
v
__._.__..J
COl3. Diff.
'
"
--
~
J
i
<._ EMG
EOG
-5#V
~ sec
-~v
-SjuV -1.0
-0.5
m -10
-0.5
-I.O -05
Fig. 1. Grand averages of movement-related potentials for plantarflexions of the left toes, flexions of the right knee, and isometric extensions of the left hip (contractions of the gluteal muscles). Bipolar contralateral vs ipsilateral recordings (Diff.) are derived by computer.
177 E M G was recorded bipolarly from the flexor digitorum plantae communis, biceps femoris or gluteus maximus muscles, respectively. The rectified E M G served for trigger. Data were recorded on a Siemens Mingograf E E G machine using an overall time constant of 3 s (EMG 0.01 s) and a low pass filter of 70 Hz (EMG 700 Hz). No smoothing of the data was carried out. Artifact-free trials (n = 200) of toe or knee movements were digitized and averaged over an epoch of 3 s, 2.25 of which were pre-trigger. A number of good gluteal contractions (128) were averaged over 3 s, two of which were pre-trigger. The digital grid consisted of 256 sampling points for toe or knee movements and of 1024 points for hip movements. For data analysis, a baseline was computed for each averaged trace using the first 500 ms of the epoch. BP amplitude was measured (in relation to the base line) at trigger level (BP0). Lateral differences were measured and statistically analyzed. The grand averages for toe, knee and hip movements are shown in Fig. 1. Throughout the foreperiod, the BP maximum is at the vertex for all 3 types of movement. Means, standard errors, differences, and t-values of BP amplitudes at the onset of E M G activity (trigger, BP0) are given in Table I. As can be seen from the bipolar graphs ('Diff.' in Fig. 1), toe movements are preceded by BPs that are larger ipsilaterally than contralaterally. This is in confirmation of earlier findings 2. For knee movements as well, an ipsilateral preponderance is obvious from the bipolar graphs. However, when comparing the ipsilateral preponderance for knee flexions with the one for toe movements, the former has less than half the amplitude of the latter and lacks statistical significance. In order to establish the best possible kind of hip movement, many pilot experiments were necessary: polygraphic E M G recordings revealed that it seems to be
an a priori impossibility to execute isotonic movements of one hip in isolation, since accompanying E M G activity of other ipsilateral muscles and compensating movements of the contralateral leg or even the trunk are inevitable. The only way out of this dilemma was to employ isometric contractions of the gluteal musculature and to train its unilateral performance by E M G monitoring as in biofeedback. As seen in Fig. 1, the BP for hip movements is absolutely symmetrical throughout the foreperiod, the bipolar derivation between lateral electrodes showing an isoelectric course. In the early foreperiod of movement preparation, a rather widespread activity in both cerebral hemispheres occurs, as indicated by a symmetric BP. As suggested by BP recordings in parkinsonism 7,8 and by regional cerebral blood flow studies 17, the early BP is probably generated by the SMA, which plays a key role in motor preparation, perhaps by contributing the initiating ' G o ' c o m m a n d (cf. ref. 13). The SMA's involvement appears to be bilateral. On approaching the onset of movement, negativity becomes more focussed on the cortical areas directly implicated in movement generation. The 'final common pathway' for movements of the extremities is the MI and the pyramidal system. The MI cortex is known from cortical stimulation experiments 16 and from clinical pathology (e.g. stroke) to be strictly unilaterally organized, i.e. it only controls movements of the contralateral half of the body. It is the MI cortex, thus, that imposes lateralisation on BP topography, which is a contralateral preponderance for upper limb movements. For movements of the lower extremities, BP lateralisation is 'paradoxical': for foot movements 4 and for toe movements 2, it was demonstrated that the BP has an ipsilateral preponderance. This is obviously due to the orientation of an electrical dipole on the
TABLE I Bereitschaftspotential at the onset o f E M G activity (BPo) over the precentral area
Means, standard errors, matched-pair differences (contralateral minus ipsilteral) and their t-values. Sign of ,X, positive or negative BP amplitudes (pV) vs linked earlobes; Sign of d, larger ipsilateral (+) negative BP amplitudes. S.E.M., standard error of the mean: S.E.D., standard error of the difference. Toes
Knee S.E.M. d
midline contralateral ipsilateral
-17.11 1.89 - 9.20 1.11 -10.81 1.06
S.E.D. t
+1.61 0.70
X
Hip S.E.M. d
-13.46 2.08 2.30 - 7.40 1.23 - 8.19 1.37
S.E.D. t
+0.79 0.37
X
S.E.M. d
-12.05 1.90 2.14 - 8.86 1.48 - 9.04 1.51
S.E.D. t
+0.18 0.59
0.31
178 tX
Toes
Fig. 2. Schematic illustration of the orientation of electrical field vectors produced by different active sources in the MI motor cortex. The tip of the arrows indicates the negative pole. Close to the mantle edge, the field vector stands upright in the sagittal plane. Activity from this area does not impose any lateralizing bias on BP topography. By finding muscle groups, whose contractions are preceoeo oy a symmetric t~r, mese would be controlled by motor cortex cells at or near the mantle edge, as is the case with hip movements. mesial cortical surface. From animal experiments a and other data we know that during the time of the BP the cortical surface becomes negative as referred to the depth. Thus, the integral vector produced by the activity of a certain somatotopic MI area is perpendicular to the surface with its negative pole towards the surface and its positive pole towards the depth. Fig. 2 illustrates the orientations of these precentral dipoles by arrows. Their tips (indicating negativity) point in different directions according to the curvature of the gyrus. Sources on the convexity (finger, hand or shoulder movement) produce vectors oriented in a contralateral upward direction, thus giving rise to a contralateral preponderance of the BP. Dipoles from sources on the medial surface of the gyrus (toe, foot or knee movements) point towards the ipsilateral hemisphere and, consequently, cause an ipsilteral BP preponderance. The present study is an attempt to test for this hy-
pothesis on the basis of the following assumption: an active focus close to the mantle edge should produce a vector that stands upright in the sagittal plane. Such a vector would no longer impose any lateralising bias on BP topography over the scalp. Consequently, if voluntary unilateral movements of a given group of muscles would produce a symmetric BP topography, such muscles would be controlled by an MI area at or near the edge. In other words, we would be able to map part of the somatotopic MI representation in intact man by functional, non-invasive means only based upon voluntary movement physiology and not on artificial electrical stimulation of the cortex as in ref. 16. Indeed, the present experiments show that the ipsilateral BP preponderance decreases from toe via knee to hip movements. For knee movements, the ipsilateral BP preponderance has only half the amplitude as the one for toe movements. Also, statistical significance is reached only for toe movements but not for knee movements. Hip movements (isometric contractions of the left gluteal musculature), the most proximal movements of the lower limb available, produced a BP topography that was symmetrical; the bipolar derivation showed an isoelectric course, Thus, hip movements seem to be generated at the mantle edge. Unfortunately, hip movements are the most proximal lower limb movements available and contractions of the trunk can by no means be performed unilaterally. Shoulder movements on the other hand already show a clearcut contralateral BP preponderance 9. Thus, in our series of lower limb movements, we cannot ascend beyond movements of the hip. However, experiments are underway to map the functional topography of the postcentral gyrus by means of somatosensory evoked potentials. Supported by the Deutsche Forschungsgemeinschaft (De 302/3-1), the Secretariat on Science, Research and Development, and the Programmes of Distinction of British Columbia, Canada.
179 1 Arezzo, J. and Vaughan, H.G. Jr., Cortical potentials associated with voluntary finger movements in the monkey, Brain Research, 88 (1975) 99-104. 2 Boschert, J., Hink, R.F. and Deecke, L., Finger movement versus toe movement-related potent~ls:-further evidence for supplementary motor area (SMA) participation prior to voluntary action, Exp. Brain Res., 55 (1983) 73-80. 3 Boschert, J., Brickett, P., Weinberg, H. and Deecke, L., Movement-related potentials preceding toe plantarflexion and dorsiflexion, Human Neurobiol., 2 (1983) 87-90. 4 Brunia, C.H.M. and Vingerhoets, A.J.J.M., Opposite hemisphere differences in movement-related potentials preceding foot and finger flexions, Biol. Psychol., 13 (1981) 261-269. 5 Brunia, C.H.M. and Van den Bosch, W.E.J., The influence of response side on the readiness potential prior to foot and finger movements. In R. Karrer, J. Cohen and P. Tueting (Eds.), Brain and Information: Event-Related Potentials, Ann. N.Y. Acad. Sci., 425, 1984, 434-437. 6 Deecke, L., Functional significance of cerebral potentials preceding voluntary movement. In D.A. Otto (Ed.), Multidisciplinary Perspectives in Event-Related Brain Potential Research, Washington, U.S. Government Printing Office, 1978, pp. 87-91. 7 Deecke, L., Cerebral potentials related to voluntary actions, parkinsonian and normal subjects. In P.J. Delwaide and A. Agnoli (Eds.), Clinical Neurophysiology in Parkinsonism. Restorative Neurology, Elsevier Science Publishers Amsterdam, 1985, Vol. 2, pp. 91-105. 8 Deecke, L. and Kornhuber, H.H., An electrical sign of participation of the mesial 'supplementary' motor cortex in human voluntary finger movements, Brain Research, 159 (1978) 473-476. 9 Deecke, L., Scheid, P. and Kornhuber, H.H., Distribution of readiness potential, pre-motion positivity, and motor potential of the human cerebral cortex preceding voluntary finger movements, Exp. Brain Res., 7 (1969) 158-168. 10 Deecke, L., GrSzinger, B. and Kornhuber, H.H., Voluntary finger movement in man: cerebral potentials and theory, Biol. Cybernet., 23 (1976) 99-119.
11 Deecke, L., Bashore, T., Brunia, C., Griinewald-Zuberbier, E., Griinewald, G. and Kristeva, R., Movement associated potentials and motor control. In R. Karrer, J. Cohen and P. Tueting (Eds.), Brain and Information: Event-Related Potentials, Ann. N.Y. Acad. Sci., 425, 1984, pp. 398-428. 12 Deecke, L., Heise, B., Kornhuber, H.H., Lang, M. and Lang, W., Brain potentials associated with voluntary manual tracking: Bereitschaftspotential, conditioned pre-motion positivity (cPMP), directed attention potential (DAP), and relaxation potential (RXP); anticipatory activity of the limbic and frontal cortex. In R. Karrer, J. Cohen and P. Tueting (Eds.), Brain and Information: Event-Related Potentials, Ann. N.Y. Acad. Sci., 425, 1984, pp. 450-464. 13 Deecke, L., Kornhuber, H.H. Lang, W., Lang, M. and Schreiber, H., Timing function of the frontal cortex in sequential motor and learning tasks, Human Neurobiol., 4 (1985) 143-154. 14 Kornhuber, H.H. and Deecke, L., Hirnpotentialfinderungen beim Menschen vor und nach Willktirbewegungen, dargestellt mit Magnetbandspeicherung und Riickw/irtsanalyse, Pfliigers Arch., 281 (1964) 52. 15 Kornhuber, H.H. and Deecke, L., Hirnpotentiaifinderungen bei Willkiirbewegungen und passiven Bewegungen des Meuschen: Bereitschaftspotential und reafferente Potentiale, PfliigersArch., 284 (1965) 1-17. 16 Penfield, W. and Rasmussen, T., The Cerebral Cortex of Man, Macmillan, New York, 1950. 17 Roland, P.E., Larsen, B., Lassen, N.A. and Skinhoj, E., Supplementary motor area and other cortical areas in organization of voluntary movements in man, J. Neurophysiol., 43 (1980) 118-136. 18 Shibasaki, H., Barrett, G., Halliday, E. and Halliday, A.M., Cortical potentials associated with voluntary foot movement in man, Electroencephalogr. Clin. Neurophysiol., 52 (1981) 507-516. 19 Vaughan, H.G., Costa, L.D. and Ritter, W., Topography of the human motor potential, Electroencephalogr. Clin. Neurophysiol., 25 (1968) 1-10.
175
BRE21587
Cerebral potentials preceding voluntary toe, knee and hip movements and their vectors in human precentral gyrus JORGEN BOSCHERT and LODER DEECKE*
Abteilung Neurologie, Universitiit Ulm, D-7900 Ulm (F.R.G.) and Neurologische Universitiitsklinik Wien, A-1097 Vienna (Austria) (Accepted February 18th, 1986)
Key words: Bereitschaftspotential - - readiness potential - - voluntary movement - - lower extremity - precentral gyms - - motor homunculus
In this study, the brain potentials related to voluntary, self-paced plantarflexions of the left toes, flexion of the right knee and isometric extensions of the left hip were examined in 3 groups of 10 subjects each. In the first half of the foreperiod, the Bereitschaftspotential (BP) or readiness potential for all movements was symmetrically distributed over both hemispheres. For toe and knee movements, an ipsilateral preponderance of the BP developed in the later foreperiod, which was statisticallysignificant for toe movements. For hip movements, BP topography was symmetrical during the entire foreperiod including its second half. Since finger, hand or shoulder movements of previous experiments show a strong contralateral preponderance of the BP, the results are discussed as further support of the hypothesis that the lateralisation of the BP is due to the orientation of the precentral electrical field vector generated by an active source in the MI motor cortex. Thus, part of the somatotopic representation of the human precentral gyms can be mapped by this non-invasive means: upper limb movements are located on the convexity; toe, foot and knee movements are generated on the mesial cortex between the hemispheres; and hip movements seem to be located at the mantle edge.
The Bereitschaftspotential (BP) or readiness p o tential is a slow negativity preceding voluntary finger m o v e m e n t by 1 s o r m o r e 14,15. W h i l e it starts symmetrically over b o t h hemispheres even prior to unilateral m o v e m e n t s , in the later f o r e p e r i o d the BP prior to u p p e r limb m o v e m e n t s exhibits a contralateral p r e p o n d e r a n c e 6. T h e m o t o r potential (MP) 9A° starting about 60 ms prior to the onset of m o v e m e n t further contributes to the total contralateral p r e p o n derance of negativity. A positive potential, the premotion positivity (PMP, latency 90 ms) is usually seen when previously active areas start to relax (cf. refs. 11, 12). These include the fronto-central midline with the s u p p l e m e n t a r y m o t o r area ( S M A ) as well as parietal and ipsilateral p r e c e n t r a l areas. The BP associated with foot m o v e m e n t s surprisingly showed a paradoxical lateralisation, i.e. an ipsilateral p r e p o n d e r a n c e 4,5. Such ipsilateral p r e p o n d e r ance of the BP was also found prior to toe movements 2. Controversial reports in the literature in-
elude some authors, who did n o t find any side difference of the BP prior to foot dorsiflexions 18 and others, who even r e p o r t e d on a contralateral p r e p o n d e r ance of the BP prior to foot dorsiflexion 19. In reviewing the literature, Brunia et al. n pointed to the fact that authors reporting on an ipsilateral BP p r e p o n d e r ance for lower limb m o v e m e n t used plantarflexion, whereas authors who either found a symmetric BP or a contralateral p r e p o n d e r a n c e e m p l o y e d dorsiflexion. Thus, they supposed that the direction of movement might influence BP lateralisation. H o w e v e r , when comparing plantar flexions and dorsiflexions of the toes within subjects in a r a n d o m i z e d fashion, B0schert et al. 3 found an ipsilateral BP p r e p o n d e r ance in both cases without any difference between conditions. Thus, lateralisation of the BP does not d e p e n d on the direction of the m o v e m e n t but rather on the locus of m o v e m e n t generation in the p r i m a r y m o t o r cortex (MI). Activation of mesial parts of M I (e.g. toe or
* Present address: Neurologische Universitfitsklinik, Lazarettgasse 14, A-1090 Wien, Austria. Correspondence: L. Deecke, University of Vienna, Neurological Clinic, Lazarettgasse 14, A-1097 Vienna, Austria.
176 tralateral leg and the trunk. We therefore decided to employ brisk isometric contractions of the gluteal muscles. After some training, using the E M G as a monitor as in biofeedback, subjects were capable of confining these to only one side (left). Each session (with the third group of subjects) consisted of 128 artifact-free trials. Subjects fixed their eyes on a designated point straight ahead of them and avoided blinking or any other movement. Trials contaminated by artifacts were excluded by the experimenters on observation of both the subject and his E E G . Movements were performed at irregular intervals of 4-12 s. Beckman Ag/AgC1 non-polarizing electrodes, 16 m m in diameter, were affixed to the scalp using EC 2 electrode cream after gently rubbing the skin with sand paper or similar tools. Recording positions included left precentral, mid-precentral and right precentral leads (C 1, C z and C 2 of the 10/20 system), referenced against linked earlobes. Impedances were usually below 1 kff2. In addition, bipolar contralateral vs ipsilateral recordings were derived by computer (left vs right precentral for right knee flexions, and right vs left precentral for left toe or hip movements). The E O G was recorded between electrodes affixed to the medial upper and the lateral lower orbital rim. The
foot representation) produces a dipole pointing towards the ipsilateral hemisphere. Consequently, the BP shows an ipsilateral preponderance. Activity on the convexity of MI as with finger movements obviously elicits dipoles directed towards the contralateral hemisphere, thus giving rise to a contralateral BP preponderance. If the active focus is at the mantle edge, the dipole would point straight upward in the sagittal plane, causing a symmetric BP throughout the foreperiod. It was the purpose of the present study to test for this hypothesis by examining the premovement potentials prior to unilateral toe, knee and hip movements. The experiments were performed on 3 groups of subjects. Each group consisted of 5 female and 5 male healthy paid students, aged 18-30 years. Subjects reclined in a chair and performed unilateral movements of the left toes, the right knee or the left hip muscles in a voluntary, self-paced manner. Toe movements were left toe plantarflexions, and knee movements were flexions of the right knee. In both groups, 200 artifact-free movements were collected per subject. For hip movements, pilot experiments revealed that isotonic movements around the hip joint provoked compensating movements of the con-
Toes ips.
Knee
Hip
J
~,.-J Cz
J
v
__._.__..J
COl3. Diff.
'
"
--
~
J
i
<._ EMG
EOG
-5#V
~ sec
-~v
-SjuV -1.0
-0.5
m -10
-0.5
-I.O -05
Fig. 1. Grand averages of movement-related potentials for plantarflexions of the left toes, flexions of the right knee, and isometric extensions of the left hip (contractions of the gluteal muscles). Bipolar contralateral vs ipsilateral recordings (Diff.) are derived by computer.
177 E M G was recorded bipolarly from the flexor digitorum plantae communis, biceps femoris or gluteus maximus muscles, respectively. The rectified E M G served for trigger. Data were recorded on a Siemens Mingograf E E G machine using an overall time constant of 3 s (EMG 0.01 s) and a low pass filter of 70 Hz (EMG 700 Hz). No smoothing of the data was carried out. Artifact-free trials (n = 200) of toe or knee movements were digitized and averaged over an epoch of 3 s, 2.25 of which were pre-trigger. A number of good gluteal contractions (128) were averaged over 3 s, two of which were pre-trigger. The digital grid consisted of 256 sampling points for toe or knee movements and of 1024 points for hip movements. For data analysis, a baseline was computed for each averaged trace using the first 500 ms of the epoch. BP amplitude was measured (in relation to the base line) at trigger level (BP0). Lateral differences were measured and statistically analyzed. The grand averages for toe, knee and hip movements are shown in Fig. 1. Throughout the foreperiod, the BP maximum is at the vertex for all 3 types of movement. Means, standard errors, differences, and t-values of BP amplitudes at the onset of E M G activity (trigger, BP0) are given in Table I. As can be seen from the bipolar graphs ('Diff.' in Fig. 1), toe movements are preceded by BPs that are larger ipsilaterally than contralaterally. This is in confirmation of earlier findings 2. For knee movements as well, an ipsilateral preponderance is obvious from the bipolar graphs. However, when comparing the ipsilateral preponderance for knee flexions with the one for toe movements, the former has less than half the amplitude of the latter and lacks statistical significance. In order to establish the best possible kind of hip movement, many pilot experiments were necessary: polygraphic E M G recordings revealed that it seems to be
an a priori impossibility to execute isotonic movements of one hip in isolation, since accompanying E M G activity of other ipsilateral muscles and compensating movements of the contralateral leg or even the trunk are inevitable. The only way out of this dilemma was to employ isometric contractions of the gluteal musculature and to train its unilateral performance by E M G monitoring as in biofeedback. As seen in Fig. 1, the BP for hip movements is absolutely symmetrical throughout the foreperiod, the bipolar derivation between lateral electrodes showing an isoelectric course. In the early foreperiod of movement preparation, a rather widespread activity in both cerebral hemispheres occurs, as indicated by a symmetric BP. As suggested by BP recordings in parkinsonism 7,8 and by regional cerebral blood flow studies 17, the early BP is probably generated by the SMA, which plays a key role in motor preparation, perhaps by contributing the initiating ' G o ' c o m m a n d (cf. ref. 13). The SMA's involvement appears to be bilateral. On approaching the onset of movement, negativity becomes more focussed on the cortical areas directly implicated in movement generation. The 'final common pathway' for movements of the extremities is the MI and the pyramidal system. The MI cortex is known from cortical stimulation experiments 16 and from clinical pathology (e.g. stroke) to be strictly unilaterally organized, i.e. it only controls movements of the contralateral half of the body. It is the MI cortex, thus, that imposes lateralisation on BP topography, which is a contralateral preponderance for upper limb movements. For movements of the lower extremities, BP lateralisation is 'paradoxical': for foot movements 4 and for toe movements 2, it was demonstrated that the BP has an ipsilateral preponderance. This is obviously due to the orientation of an electrical dipole on the
TABLE I Bereitschaftspotential at the onset o f E M G activity (BPo) over the precentral area
Means, standard errors, matched-pair differences (contralateral minus ipsilteral) and their t-values. Sign of ,X, positive or negative BP amplitudes (pV) vs linked earlobes; Sign of d, larger ipsilateral (+) negative BP amplitudes. S.E.M., standard error of the mean: S.E.D., standard error of the difference. Toes
Knee S.E.M. d
midline contralateral ipsilateral
-17.11 1.89 - 9.20 1.11 -10.81 1.06
S.E.D. t
+1.61 0.70
X
Hip S.E.M. d
-13.46 2.08 2.30 - 7.40 1.23 - 8.19 1.37
S.E.D. t
+0.79 0.37
X
S.E.M. d
-12.05 1.90 2.14 - 8.86 1.48 - 9.04 1.51
S.E.D. t
+0.18 0.59
0.31
178 tX
Toes
Fig. 2. Schematic illustration of the orientation of electrical field vectors produced by different active sources in the MI motor cortex. The tip of the arrows indicates the negative pole. Close to the mantle edge, the field vector stands upright in the sagittal plane. Activity from this area does not impose any lateralizing bias on BP topography. By finding muscle groups, whose contractions are preceoeo oy a symmetric t~r, mese would be controlled by motor cortex cells at or near the mantle edge, as is the case with hip movements. mesial cortical surface. From animal experiments a and other data we know that during the time of the BP the cortical surface becomes negative as referred to the depth. Thus, the integral vector produced by the activity of a certain somatotopic MI area is perpendicular to the surface with its negative pole towards the surface and its positive pole towards the depth. Fig. 2 illustrates the orientations of these precentral dipoles by arrows. Their tips (indicating negativity) point in different directions according to the curvature of the gyrus. Sources on the convexity (finger, hand or shoulder movement) produce vectors oriented in a contralateral upward direction, thus giving rise to a contralateral preponderance of the BP. Dipoles from sources on the medial surface of the gyrus (toe, foot or knee movements) point towards the ipsilateral hemisphere and, consequently, cause an ipsilteral BP preponderance. The present study is an attempt to test for this hy-
pothesis on the basis of the following assumption: an active focus close to the mantle edge should produce a vector that stands upright in the sagittal plane. Such a vector would no longer impose any lateralising bias on BP topography over the scalp. Consequently, if voluntary unilateral movements of a given group of muscles would produce a symmetric BP topography, such muscles would be controlled by an MI area at or near the edge. In other words, we would be able to map part of the somatotopic MI representation in intact man by functional, non-invasive means only based upon voluntary movement physiology and not on artificial electrical stimulation of the cortex as in ref. 16. Indeed, the present experiments show that the ipsilateral BP preponderance decreases from toe via knee to hip movements. For knee movements, the ipsilateral BP preponderance has only half the amplitude as the one for toe movements. Also, statistical significance is reached only for toe movements but not for knee movements. Hip movements (isometric contractions of the left gluteal musculature), the most proximal movements of the lower limb available, produced a BP topography that was symmetrical; the bipolar derivation showed an isoelectric course, Thus, hip movements seem to be generated at the mantle edge. Unfortunately, hip movements are the most proximal lower limb movements available and contractions of the trunk can by no means be performed unilaterally. Shoulder movements on the other hand already show a clearcut contralateral BP preponderance 9. Thus, in our series of lower limb movements, we cannot ascend beyond movements of the hip. However, experiments are underway to map the functional topography of the postcentral gyrus by means of somatosensory evoked potentials. Supported by the Deutsche Forschungsgemeinschaft (De 302/3-1), the Secretariat on Science, Research and Development, and the Programmes of Distinction of British Columbia, Canada.
179 1 Arezzo, J. and Vaughan, H.G. Jr., Cortical potentials associated with voluntary finger movements in the monkey, Brain Research, 88 (1975) 99-104. 2 Boschert, J., Hink, R.F. and Deecke, L., Finger movement versus toe movement-related potent~ls:-further evidence for supplementary motor area (SMA) participation prior to voluntary action, Exp. Brain Res., 55 (1983) 73-80. 3 Boschert, J., Brickett, P., Weinberg, H. and Deecke, L., Movement-related potentials preceding toe plantarflexion and dorsiflexion, Human Neurobiol., 2 (1983) 87-90. 4 Brunia, C.H.M. and Vingerhoets, A.J.J.M., Opposite hemisphere differences in movement-related potentials preceding foot and finger flexions, Biol. Psychol., 13 (1981) 261-269. 5 Brunia, C.H.M. and Van den Bosch, W.E.J., The influence of response side on the readiness potential prior to foot and finger movements. In R. Karrer, J. Cohen and P. Tueting (Eds.), Brain and Information: Event-Related Potentials, Ann. N.Y. Acad. Sci., 425, 1984, 434-437. 6 Deecke, L., Functional significance of cerebral potentials preceding voluntary movement. In D.A. Otto (Ed.), Multidisciplinary Perspectives in Event-Related Brain Potential Research, Washington, U.S. Government Printing Office, 1978, pp. 87-91. 7 Deecke, L., Cerebral potentials related to voluntary actions, parkinsonian and normal subjects. In P.J. Delwaide and A. Agnoli (Eds.), Clinical Neurophysiology in Parkinsonism. Restorative Neurology, Elsevier Science Publishers Amsterdam, 1985, Vol. 2, pp. 91-105. 8 Deecke, L. and Kornhuber, H.H., An electrical sign of participation of the mesial 'supplementary' motor cortex in human voluntary finger movements, Brain Research, 159 (1978) 473-476. 9 Deecke, L., Scheid, P. and Kornhuber, H.H., Distribution of readiness potential, pre-motion positivity, and motor potential of the human cerebral cortex preceding voluntary finger movements, Exp. Brain Res., 7 (1969) 158-168. 10 Deecke, L., GrSzinger, B. and Kornhuber, H.H., Voluntary finger movement in man: cerebral potentials and theory, Biol. Cybernet., 23 (1976) 99-119.
11 Deecke, L., Bashore, T., Brunia, C., Griinewald-Zuberbier, E., Griinewald, G. and Kristeva, R., Movement associated potentials and motor control. In R. Karrer, J. Cohen and P. Tueting (Eds.), Brain and Information: Event-Related Potentials, Ann. N.Y. Acad. Sci., 425, 1984, pp. 398-428. 12 Deecke, L., Heise, B., Kornhuber, H.H., Lang, M. and Lang, W., Brain potentials associated with voluntary manual tracking: Bereitschaftspotential, conditioned pre-motion positivity (cPMP), directed attention potential (DAP), and relaxation potential (RXP); anticipatory activity of the limbic and frontal cortex. In R. Karrer, J. Cohen and P. Tueting (Eds.), Brain and Information: Event-Related Potentials, Ann. N.Y. Acad. Sci., 425, 1984, pp. 450-464. 13 Deecke, L., Kornhuber, H.H. Lang, W., Lang, M. and Schreiber, H., Timing function of the frontal cortex in sequential motor and learning tasks, Human Neurobiol., 4 (1985) 143-154. 14 Kornhuber, H.H. and Deecke, L., Hirnpotentialfinderungen beim Menschen vor und nach Willktirbewegungen, dargestellt mit Magnetbandspeicherung und Riickw/irtsanalyse, Pfliigers Arch., 281 (1964) 52. 15 Kornhuber, H.H. and Deecke, L., Hirnpotentiaifinderungen bei Willkiirbewegungen und passiven Bewegungen des Meuschen: Bereitschaftspotential und reafferente Potentiale, PfliigersArch., 284 (1965) 1-17. 16 Penfield, W. and Rasmussen, T., The Cerebral Cortex of Man, Macmillan, New York, 1950. 17 Roland, P.E., Larsen, B., Lassen, N.A. and Skinhoj, E., Supplementary motor area and other cortical areas in organization of voluntary movements in man, J. Neurophysiol., 43 (1980) 118-136. 18 Shibasaki, H., Barrett, G., Halliday, E. and Halliday, A.M., Cortical potentials associated with voluntary foot movement in man, Electroencephalogr. Clin. Neurophysiol., 52 (1981) 507-516. 19 Vaughan, H.G., Costa, L.D. and Ritter, W., Topography of the human motor potential, Electroencephalogr. Clin. Neurophysiol., 25 (1968) 1-10.