Departmeni! of Integrative Physiology, National Institute for Physiological Sciences, Myodaiji, Okazaki 444, Japan
Accepted for publication: 20 June 1995
Abstract
Somatosensory evoked magnetic fields (SEFs) following painful electrical stimulation of the finger were investigated in 5 normal subjects. Equivalent current dipoles (ECDs) of deflections shorter than 100 msec in latency were located in the primary sensory cortex (SI) in the hemisphere contralateral to the stimulated finger following either non-painful or painful stimulation. Two main deflections, Nl00m-P100m and N250m-P250m, were independently identified following painful stimulation, although they were not found in SEFs following non-painful wed: stimulation. ECDs of the N100m-P100m were considered to be located in the bilateral second sensory cortices (SII). ECDs of the N250m-P250m were identified in the bilateral cingulate cortices and SII, but the intersubject difference was large. Therefore, we considered that contralateral SI and bilateral SII were initially activated by painful noxious stimulation, and then multiple areas including bilateral SII and cingulate cortices were activated. In EEG recordings (evoked potentials), no potential corresponding to N100m-P100m was found, probably because it was difficult to record activation in SII by EEG recordings. The P250 potential which corresponded to the N250m-P250m was clearly identified, probably because activation of multiple areas generated large long-duration EEG potentials which were maximal around the vertex, unlike MEG recordings.
Keywords: Somatosensory evoked magnetic fields; Magnetoencephalography; Superconducting quantum interference device (SQUID); Painful stimulation; Cinglate cortex; Second sensor~' cortex
I. Introduct ion
Somatosensory evoked potentials (SEPs), following painful electrical stimula~Lion of the skin, showed predomi- nant middle-latency potentials, N150 and P250, which were not identified when stimulation is not so painful (Chen and Chapman, 1980; Bromm and Scharein, 1982; Stowell, 1984; Katayama et al., 1985; Miltner et al., 1989; Ohara, 1989, 1990; Dowman, 1994a,b; Dowman and Darcey, 1994). Therefore, these particular potentials were considered to be generated by specific pain-related brain activities. However, as these potentials were maximal around the vertex and spread following stimulation of any area of the body surface, their generator sources have not yet been determined. As compared with averaged elec- troencephalography (EEG evoked potentials), magnetoen-
cephalography (MEG) has several theoretical advantages in localizing brain dipoles due to less skull effects and detection of only a specific orientation of brain current tangential to the skull (Sutherling et al., 1992). Several investigators including ourselves have reported somatosen- sory evoked magnetic fields (SEFs) following stimulation of the upper limb (Brenner et al., 1978; Kaufman et al., 1981; Had et al., 1984, 1990; Wood et al., 1985; Huttunen et al., 1987, 1992; Sutherling et al., 1988; Rossini et al., 1989; Tiihonen et al., 1989; Baumgartner et al., 1991; Had, 1991; Narici et al., 1991; Suk et al., 1991; Kakigi, 1994; Kakigi et al., 1995a). However, only one study has been reported concerning pain-related SEFs following painful electrical stimulation (Joseph et al., 1991), and they recorded SEFs from small areas using a single coil. There- fore, the objective of this study was to thoroughly analyze the detailed topography of SEFs following painful electri- cal stimulation of the finger using two sets of 37-channel neuromagnetometers to investigate the generator source of each identifiable component.
464 Y. Kitamura et al. / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474
Sensor Layout
L
C3 C4
Fz Pz
Cz
Fig. 1. Positions of magnetometer placement. The mesurement matrix was centered at 5 different positions, around the C3, C4, Cz, Fz and Pz of the international 10-20 system in each subject.
N20m
W e a k P30m P60m
N30m N60m
M o d e r a t e l y ~ ~ Painful
Painful
toofr L 5ms
-~ ~ 4o ~o Fig. 2. SEFs following fight index finger stimulation recorded at the C3 position in "weak," "moderately painful" and "very painful" sessions in subject 1. Representative waveforms recorded at 2 channels are shown to indicate the nomenclature of each component. The N20m-P20m, N30m-P30m, N40m-P40m and N60m-P60m were clearly identified in all sessions.
o=.
.<
(a)
, 2001T l_ ' ~ ~ ~ ' ~ 20ms
U p p e r (b)
k
t.,
r~
O
L o w e r
Fig. 3. SEFs recorded at all 37 channels of the C3 position following moderately painful stimulation of the fight index finger in subject I (a). The analysis window was 0-50 msec. Isocontour map of N30m-P30m (b) at the period indicated by the vertical bars in (a). Outward going flux is marked by solid lines in the contour plot, inward going flux by dotted lines and thick line shows the zero point. Contour step was 10 IT,
Y. Kitamura et aL / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474 465
2. Subjects and methods
Five normal volunteer,~ (1 female and 4 males; mean age 32 years, range 29-41 years; mean height 167 cm, range 151-175 cm) were studied. Informed consent was obtained from all participants prior to the study.
The electrical stimulus was a constant voltage square- wave pulse delivered transcutaneously to the right index finger. We adopted 3 levels of intensity, "weak , " "mod- erately painful" and "w',ry painful," depending on the subjective feeling of each subject; 2 - 4 mA, 5 -7 mA and 10-13 mA were used as " w e a k , " "moderately painful" and "ve ry painful" stinauli, respectively. Stimuli were applied in random order. The interstimulus interval was between 1500 and 5000 msec and the stimulus duration was 1 msec.
The recordings were made in a magnetically shielded room with the subject sitting or lying on a bed, depending on the recording area of the scalp.
SEFs were measured with two 37-channel biomagne- tometers (Magnes, Biomagnetic Technologies Inc., San
Diego, CA). The detection coils of the biomagnetometer were arranged in a uniformly distributed array in concen- tric circles over a spherically concave surface. Thus, all of the sensor coils were equally sensitive to the brain's weak magnetic signals. The device was 144 mm in diameter and its radius was 122 mm. The outer coils were 72.5 ° apart, Each coil was 20 mm in diameter and the distance between the centers of each coil was 22 mm. Each coil was connected to a superconducting quantum interference de- vice (SQUID).
The measurement matrix was centered at 5 different positions: around the C3, C4, Fz, Cz and Pz of the international 10-20 system in each subject (Fig. 1). The whole head was mostly covered by these positions of magnetometer placement. Additional placements, for ex- ample F3 and F4, were also adopted in some subjects depending on the recorded ECDs positions. Responses were filtered with a 0 .1 - I00 Hz bandpass filter and digi- tized at a sampling rate of 1041.7 Hz. The analysis time was 100 msec before and 900 msec after application of stimuli, and DC offset was achieved using pre-stimulus
N 30 m - P 30 m
R - L
Fig. 4. Localization of represenlative ECDs of the N30m-P30m following moderately painful stimulation overlapped on axial and coronal views of MRI in two subjects. They were located in the hand area of SI in the left hemisphere.
466 Y. Kitamura et al . / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474
period as the baseline. Each session comprised an average of 300 trials, and at least two averages were obtained to ensure reproducibility. The order of each session was randomized.
A spherical model was fitted to the digitized head shape of each subject, and the location (x, y, z positions), orientation and amplitude of a best-fitted single equivalent current dipole (ECD) were estimated for each point in time. The origin of the head-based coordinate system was the midpoint between the preauricular points (Geselowitz, 1970; Sarvas, 1987). The x-axis indicated the coronal plane with a positive value towards the anterior direction;
the y-axis indicated the mid-sagittal plane with positive values towards the left preauricular point. The z-axis lay on the transverse plane perpendicular to the x-y line with positive value towards the upper side. Correlations be- tween the theoretical field generated by the model and the observed field were used to estimate the goodness of fit of the model parameters.
The evoked potentials of the EEG were also recorded simultaneously. Gold disk electrodes (1 cm in diameter) were attached to the scalp with electrode jelly based on the international 10-20 system. Eye movements were moni- tored from an electrode placed at the left supraorbital
C3 C4
Weak
Moder Painful
N250m
Very Painful
1 l f f f f l l l
lOOfT l 50ms
6 100 200 6 160 260 Fig. 5. SEFs waveforms recorded in C3 and C4 positions following "weak," "moderately painful" and "very painful" stimulation in subject 2. Superimposition of all 37 channels in each condition. The analysis window was 20 msec before and 280 msec after stimulation. Nl00m-P100m and N250m-P250m were identified only in "moderately painful" and "very painful" sessions, being much larger in the latter.
(a) Upper
<
(b)
'oo"T+L-+m +
Y. Kitamura et al. / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474 467
t.,
@
L o w e r
Fig. 6. SEFs recorded at all 37 channels in the C3 position following "very painful" stimulation of the right index finger in subject 1 (a). The analysis window was 0-150 msec. Isocontour map of N100m-Pl00m (b) at the period indicated by the vertical bars in (a). Contour step was 10 fT.
ridge. I m p e d a n c e was m a i n t a i n e d at less than 5 k .O. A n
.~xploring e l ec t rode w as p l a c e d at Cz and the l inked ear-
lobes (A1 + A 2 ) we re u s e d as re fe rences . T he t ime con -
gtant and the low pas s f i l te r o f the ampl i f i e r were 2 sec and
50 Hz, r espec t ive ly . A re la t ive pos i t iv i ty at gr id 1 r e su l t ed
in a d o w n w a r d def lec t ion . The tr ials in w h i c h E O G var ia -
t ion e x c e e d e d 80 /zV or E O G de f l ec t i on e x c e e d e d + 100
/xV and the tr ials in w h i c h there was an e r ror r e s p o n s e or a
r e s p o n s e o m i s s i o n were exc luded f r o m S E P averag ing .
M a g n e t i c r e s o n a n c e imag ing ( M R I ) was p e r f o r m e d us-
(a) Upper (b) I
, , . , , , ~ . 6 2 3 , t ~ e I % % % 1 1 1 " . ~ b ~ I 2 4 % J I
Fig. 7. SEFs recorded at all 37 channels in the C4 position following "very painful" stimulation of the right index finger in subject 1 (a). The analysis window was 0-150 msec. Isocoatour map of N10Om-Pl00m (b) at the period indicated by the vertical bars in (a). Contour step was 10 fT.
468 Y. Kitamura et al. / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474
ing a GE Signa 1.0 T system. The Tl-weighted coronal and axial images with contiguous 3 mm slice thickness were adopted for overlays with ECD sources detected by MEG. The nasion and bilateral preauricular points were identified on MRI images with the aid of high contrast cod liver oil capsules (3 mm in diameter).
The positions of the equivalent current dipole (ECD) of each deflection were analyzed every 0.96 msec and super- imposed on MRI. Their ECD was located around the hand sensory area of the SI in the left hemisphere which was contralateral to the stimulated finger (Fig. 4). The direction of ECDs was oriented mainly antero-posteriorly.
3. Results 3.2. Middle-latency components with latencies longer than 100 msec
3.1. Short-latency components with latencies shorter than 100 msec
Four main deflections, N20m, P30m, N40m and P60m, and their counterparts, P20m, N30m, P40m and N60m, were clearly identified in all subjects in all "weak," "moderately painful" and "very painful" sessions (Fig. 2). When the probe was centered around the C3 position, these deflections showed clear polarity reversal (Fig. 3).
Two main deflections, N100m-P100m and N250m- P250m, were identified only in the "moderately painful" and "very painful" sessions, and they were much larger in the latter session (Fig. 5).
The N100m-P100m was clearly identified in all 5 sub- jects. The N100m deflection showed a clear polarity rever- sal against the P100m deflection when the probe was centered at C3 and C4 (Figs. 6 and 7), but the isocontour maps of the N100m-P100m (Figs. 6 and 7) were much
N 100 m - P 100 m ( C3 Position)
R L
Fig. 8. Localization of representative ECDs of the Nl00m-Pl00m following very painful stimulation overlapped on axial and coronal views of MRI in two subjects. The probe was centered around the C3. They were located around the superior bank of the Sylvian fissure on the left hemisphere.
Y. Nitamura et al. /Electroencephalography and clinical Neurophysiology 95 (1995) 463-474 469
different to those of the early deflections; for example N30m-P30m (Fig. 3). Their ECDs were located around the superior bank of the Sylvian fissure on each hemisphere (Figs. 8 and 9). The N1,L00m-P100m was identified in bilateral hemispheres, and there was no significant differ- ence in peak latencies in both hemispheres. The dipole moments of N100m-P100m recorded in each hemisphere were almost the same.
The following large deflection, N250m-P250m, was clearly identified in 4 of 5 subjects. However, the positions of their ECDs were rather complex as compared with the early deflections. When tbe probe was centered at the C3 or C4, ECDs were located in bilateral SII similarly to N100m-P100m (Fig. 10). When the probe was centered at the Cz position, clear pc,larity reversal deflections were also identified in 4 subjects (Figs. 11 and 12). Their ECDs were located along the interhemispheric fissure, probably in the cingulate cortex, on the left side in two subjects and
on the right side in one subject (Fig. 12), and in the strange areas of the white matter in one other subject.
When the probe was centered at the Fz and Pz positions which covered the mid-frontal and mid-parietal areas, no clear polarity-reversed deflections were identified. No con- sistent ECD considered to be generated in the mid-frontal or mid-parietal areas were identified.
Averaged EEGs recorded at the Cz showed clear N150 and P250 in all subjects, so the potentials were dramati- cally increased in amplitude in the "very painful" session (Fig. 13). The N100m-P100m appeared to correspond to the onset of the N150, and the N250m-P250m appeared to correspond to the P250.
4. Discussion
There have been many reports concerning pain-related somatosensory evoked potentials following electrical stim-
N 100 m - P 100 m ( C4 Position)
J j J
/ /
J
R "~ L
Fig. 9. Localization of representative ECDs of the Nl00m-Pl00m following very painful stimulation overlapped on axial and coronal views of MRI in two subjects. The probe was centerecl around the C4. They were located around the superior bank of the Sylvian fissure on the right hemisphere.
470 Y. Kitamura et a L / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474
ulation (Chen and Chapman, 1980; Bromm and Scharein, 1982; Stowell, 1984; Katayama et al., 1985; Miltner et al., 1989; Ohara, 1989, 1990; Dowman, 1994a,b; Dowman and Darcey, 1994), but, to our knowledge, only Joseph et al. (1991) reported MEGs using this method. They used a single coil placed around the sensory areas of the hemi- sphere contralateral to the stimulated finger and found ECDs of pain SEFs in the "frontal operculum," which appeared slightly more frontal as compared with the loca- tion of ECDs found in the present study, SII. They did not find ECDs in the cingulate cortex, probably because they did not place SQUID coils in the central areas.
To our knowledge, 3 other systematic studies of MEG following painful stimulation including ours have been reported. Hari et al. (1983) studied the painful electrical stimulation of tooth pulp, and Huttunen et al. (1986) used CO 2 gas applied to the nasal mucosa. They reported that
the current sources were located at or near SII. We used a CO 2 laser beam applied to the skin of the limbs as a painful stimulus and also found ECDs in the bilateral SII (Kakigi et al., 1995b).
In the present study, ECDs of the early deflections shorter than 100 msec in latency were located in SI, probably area 3b, contralateral to the stimulated nerve. These were identified in all sessions including those with "weak" stimuli. These results were compatible with the previous reports studying SEFs following upper limb stim- ulation (Hari et al., 1984, 1990, 1993; Wood et al., 1985; Rossini et al., 1989; Hari, 1991; Narici et al., 1991; Kakigi, 1994; Kakigi et al., 1995a). However, the subse- quent deflections, N 100m-P100m and N250m-P250m, were found only in the "moderately painful" and "very painful" sessions. Therefore, these deflections appeared to be generated by painful (noxious) stimulation.
N 250 m - P 250 m ( C3 Position)
R ...... L 1
Fig. 10. Localization of representative ECDs of the N250m-P250m following very painful stimulation overlapped on axial and coronal views of MRI in two subjects. The probe was centered around the C3. They were located in the left SII like N100m-Pl00m.
Y. ,~fitamura et al. / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474 471
ECDs of the N100m-P100m were located in bilateral SII, and ECDs of the N250m-P250m appeared to be located in multiple areas including the bilateral SII and cingulate cortices. These results indicated that the ascend- ing signals following painful noxious stimulation initially reached bilateral SII, and multiple areas were activated afterwards.
SII is located along the superior bank of the Sylvian fissure and lies lateral and posterior to the face representa- tion in SI, anterior or medial to the primary auditory areas (Burton, 1986). Although the functions of SII have not been clearly identified, SII probably provides a very funda- mental somatosensory capacity because this area has been found in nearly all mammalian species including animals lacking extensive manipulative skills (see Burton, 1986). SII is the first cortical stage of the somatosensory system in which neurons process information from both sides of the body (Juliano et al., 1983). Therefore, it is reasonable to assume that the bilateral SII are activated following painful stimulation. Burton (1986) proposed that SII pro- vides a "cortical loop" for cutaneous input to the motor cortex, acts as a "highe:r order" association center for tactile learning or has both of these capabilities but to differing extents depending upon the behavioral sophistica- tion of the animal.
The role of neurons in SI and SII concerning the perception of pain in humans remains unclear. However, all studies of MEGs concerning painful stimulation re- ported to date including the present study (Hari et al., 1983; Huttunen et al., 1986; Joseph et al., 1991; Kakigi et
al., 1995b) indicated that SII was the site responsible for receiving signals following painful stimulation.
Cerebral activities to painful stimulation following the initial excitation of neurons in SII should be very complex. We did not consider that N250m and P250m were gener- ated by single ECD. We speculated that multiple areas including bilateral SII and cingulate cortices were acti- vated simultaneously and independently. Therefore, a large inter-individual difference in location of the ECD was recognized, because magnetic fields generated by multiple areas should interfere with each other. Positron emission tomography (PET) has demonstrated that painful heat stim- uli cause significant activation of SI in the hemisphere contralateral to the stimulated hand, bilateral SII and the cingulate cortex (Talbot et al., 1991). This report was fundamentally compatible with the present study.
In the EEG recordings, the N150 and P250 were clearly identified as the pain-related potentials. Peaks N100m- P100m of MEG appeared to correspond to the onset of N150. This finding indicated that the responses generated in SII are very difficult to identify by EEG recordings, whereas MEGs are quite useful for their detection. This is probably because of the following 2 reasons: (1) it is difficult to detect electric potentials generated in SII, be- cause they are very small and their dipoles oriented verti- cally and tangential to the surface scalp EEG electrodes which are placed in the temporal areas, and (2) the refer- ence electrodes of the EEGs, usually linked earlobes, have similar electrical activity to the electrodes placed around the SII. The N250m-P250m of MEG appeared to corre-
(a) Right (b)
+
lOOms
Left
o m .o k
<
Fig. 11. SEFs recorded at all 3"I channels of Cz position following "very painful" stimulation of the right index finger in subject 3 (a). The analysis window was 0-150 msec. Isocontour map of N250m-P250m (b) at the period indicated by the vertical bars in (a). Contour step was 10 fT.
472 Y. Kitamura et a l . / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474
N 250 m - P 250 m ( Cz Position)
R L
Fig. 12. Localization of representative ECDs of the N250m-P250m following very painful stimulation overlapped on axial and coronal views of MRI in 3 subjects. The probe was centered around the Cz. They were located along the interhemispheric fissure, probably in the cingulate cortex.
Y. Kitamura et al. / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474 473
Moderately Painful
Veo t/__ Painful
P250 ~ v L
20ms JL00 200 300ms
Fig. 13. Averaged EEG waveforms recorded at the Cz electrode follow- ing "weak," "moderately painful" and "very painful" stimulation in subject 2. The N150 and P250 were clearly identified only in "mod- erately painful" and "very painful" sessions, being much larger in the latter.
spond to the large N150-P250 complex o f the E E G which
were m a x i m u m around the ver tex. Therefore , we consid-
ered that the act ivi t ies ;generated in var ious areas were
summated and clearly recorded by the EEGs , but interfered
with each other in M E G recordings . This f inding was
probably due to the lack o f c lear ident i f icat ion by M E G of
act ivi t ies generated in deep areas.
In conclus ion, when we rece ive painful electr ical st imu-
lation, SI contralateral to the s t imulated nerve is initially
act ivated and then bilateral SII are act ivated. It is not clear
at present whether bilateral SII are act ivated by signals
r ece ived f rom SI or by signals c o m i n g direct ly f rom the
thalamus. Then, mult iple areas inc luding SII and cingulate
cor tex are act ivated s imul taneous ly and independent ly .
A c k n o w l e d g e m e n t s
The authors are ve ry grateful for the technical help o f
Mr. M. Mori , Mr. Y. Ta2~eshima and Mr. O. Nagata.
This study was supported by the Ueha ra Memor ia l
Foundat ion and the Nai to Foundat ion in Japan, The Integ-
rat ive Studies on Phys io log ica l Funct ions (06NP0101) f rom
The Minis t ry o f Educat ion, Sc ience and Cul ture o f Japan.
References
Baumgarmer, C., Sutherling, W.W., Di, S. and Barth, D.S. (1991) Spatiotemporal modeling of cerebral evoked magnetic fields to me- dian nerve stimulation. Elec:troenceph. clin. Neurophysiol., 79: 27-35.
Brenner, D., Lipton, J., Kanfman, L. and Williamson, SJ. (1978) Somati- cally evoked magnetic fields of the human brain. Science, 199: 81-83.
Bromm, B. and Scharein, E. (1982) Principal component analysis of pain-related cerebral potentials to mechanical and electrical stimula- tion in man. Electroenceph. olin. Neurophysiol., 53: 94-103.
Burton, H. (1986) Second somatosensory cortex and related areas. In: E.G. Jones and A. Peters (Eds.), Cerebral Cortex, Vol. 5. Plenum Press, New York, pp. 31-98.
Chen, A.C.N. and Chapman, C.R. (1980) Aspirin analgesia evaluated by event-related potentials in man: possible central action in brain. Exp. Brain Res., 39: 359-364.
Dowman, R. (1994a) SEP topographies elicited by innocuous and nox- ious sural nerve stimulation. I. Identification of stable periods and individual differences. Electroenceph. clin. Neurophysiol., 92: 291- 302.
Dowman, R. (1994b) SEP topographies elicited by innocuous and nox- ious sural nerve stimulation. II. Effects of stimulus intensity on topographic pattern and amplitude. Electroenceph. clin. Neurophysiol., 92: 303-315.
Dowman, R. and Darcey, T.M. (1994) SEP topographies elicited by innocuous and noxious sural nerve stimulation. IlL Dipole source localization analysis. Electroenceph. clin. Neurophysiol., 92: 373-391.
Geselowitz, D.B. (1970) On the magnetic field generated outside an inhomogeneous volume con ductor by internal current sources. IEEE Trans. Magn., MAG-6: 346-347.
Had, R. (1991) On brain's magnetic responses to sensory stimuli. J. Clin. Neurophysiol., 8: 157-169.
Had, R., Kaukoranta, E., Reinikainen, K., Huopaniemie, T. and Mauno, J. (1983) Neuromagnetic localization of cortical activity evoked by painful dental stimulation in man. Neurosci. Lea., 42: 77-82.
Had, R., Reinikainen, K., Kaukoranta, E., H~irn~il~iinen, M., llmoniemi, R., Penttinen, A., Salminen, J. and Teszner, D. (1984) Somatosensory evoked cerebral magnetic fields from SI and SII in man. Electroen- ceph. clin. Neurophysiol., 57: 254-263.
Had, R., H~im~ilginen, H., H~irn~il~iinen, M., Kekoni, J., Sams, M. and Tiihonen, J. (1990) Separate finger representations at the human second somatosensory cortex. Neuroscience, 37: 245-249.
Had, R., Karhu, J., Hgm~il~iinen, M., Knuutila, J., Salonen, O., Sams, M. and Vilkman, V. (1993) Functional organization of the human first and second somatosensory cortices: a neuromagnetic study. Eur. J. Neurosci., 5: 724-734.
Huttunen, J., Kobal, G., Kaukoranta, E. and Had, R. (1986) Cortical responses to painful CO 2 stimulation of nasal mucosa: a magnetoen- cephalographic study in man. Electroenceph. clin. Neurophysiol., 64: 347-349.
Huttunen, J., Had, R. and Leinonen, L. (1987) Cerebral magnetic re- sponses to stimulation of ulnar and median nerves. Electroenceph. clin. Neurophysiol., 66: 391-400.
Huttunen, J., Ahlfors, S. and Had, R. (1992) Interaction of afferent impulses in the human primary sensorimotor cortex. Electroenceph. clin. Neurophysiol., 82: 176-181.
Joseph, J., Howland, E.W., Wakai, R., Backonja, M., Baffa, O., Potenti, F.M. and Cleeland, C.S. (1991) Late pain-related magnetic fields and electric potentials evoked by intracutaneous electric finger stimula- tion. Electroenceph. clin. Neurophysiol., 80: 46-52.
Juliano, S.L., Hand, P.J. and Whitsel, B.L. (1983) Patterns of metabolic activity in cytoarchitectural area SII and surround cortical fields of the monkey. J. Neurophysiol., 50: 961-980.
Kakigi, R. (1994) Somatosensory evoked magnetic fields following me- dian nerve stimulation. Neurosci. Res., 20: 165-174.
Kakigi, R., Koyama, S., Hoshiyama, M., Watanabe, S., Shimojo, M. and Kitamura, Y. (1995a) Gating of somatosensory evoked responses during active finger movements: magnetoencephalographic studies. J. Neurol. Sci., 128: 195-204.
Kakigi, R., Koyama, S., Hoshiyama, M., Kitamura, Y., Shimojo, M. and Watanabe, S. (1995b) Pain-related magnetic fields following painful CO 2 laser stimulation in man. Neurosci. Lea., 192: 45-48.
Katayama, Y., Tubokawa, T., Harano, T. and Tukiyama, T. (1985) Dissociation of subjective pain report and pain-related late positive
474 Y. Kitamura et al. / Electroencephalography and clinical Neurophysiology 95 (1995) 463-474
components of cerebral evoked potentials in subjects with brain lesions. Brain Res. Bull., 14: 423-426.
Kaufman, L., Okada, Y., Brenner, D. and Williamson, S.J. (1981) On the relation between somatic evoked potentials and fields. Int. J. Neu- rosci., 15: 223-239.
Miltner, W., Johnson, R.J., Braun, C. and Larbig, W. (1989) Somatosen- sory event-related potentials to painful and non-painful stimuli: ef- fects of attention. Pain, 38: 303-312.
Narici, L., Modena, I., Opsomer, R.J., Pizzella, V., Romani, G.L., Torrioli, G., Traversa, R. and Rossini, P.M. (1991) Neuromagnetic somatosensory homunculus: a non-invasive approach in humans. Neu- rosci. Lett., 121: 51-54.
Ohara, K. (1989) Recording conditions of pain related late positive component. Jpn. J. EEG EMG, 17: 365-374.
Ohara, K. (1990) Characteristics and evaluation of pain related late positive component. Jpn. J. EEG EMG, 18: 1-9.
Rossini, P.M., Narici, L., Romani, G.L., Traversa, R., Cecchi, L., Cilli, M. and Urbano, A. (1989) Short latency somatosensory evoked responses to median nerve stimulation in healthy humans: electric and magnetic recordings. Int. J. Neurosci., 46: 67-76.
Sarvas, J. (1987) Basic mathematical and electromagnetic concepts of the biomagnetic inverse problem. Phys. Med. Biol., 32:11-22.
Stowell, H. (1984) Event related brain potentials and human pain: a first objective overview. Int. J. Psychophysiol., 1: 137-151.
Suk, J., Ribary, U., Cappell, J., Yamamoto, T. and Llinfis, R. (1991) Anatomical localization revealed by MEG recordings of the human somatosensory system. Electroenceph. clin. Neurophysiol., 78: 185- 196.
Sutheding, W.W., Crandall, P.H., Dareey, T.M., Becker, D.P., Levesque, M.F. and Barth, D.S. (1988) The magnetic and electric fields agree with intracranial localizations of somatosensory cortex. Neurology, 38: 1705-1714.
Sutherling, W.W., Cuffin, B.N., Hari, R., Hamer, R.N., King, D.W., Mattson, R.H., Reite, M.L., Sato, S. and Weinbcrg, H. (1992) Assess- ment: magnetoencephalography (MEG). Report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology, 41: 1-4.
Talbot, J.D., Marrett, S., Evans, A.C., Meyer, E., Bushnell, M.C. and Duncan, G.H. (1991) Multiple representations of pain in human cerebral cortex. Science, 251: 1355-1358.
Tiihonen, J., Hari, R. and H~niilkinen, M. (1989) Early deflections of cerebral magnetic responses to median nerve stimulation. Electroen- ceph. clin. Neurophysiol., 74: 290-296.
Wood, C.C., Cohen, D., Cuffin, B.N., Yarita, M. and Allison, T. (1985) Electrical sources in human somatosensory cortex: identification by combined magnetic and potential recordings. Science, 227: 1051- 1053.
