This fMRI study investigated the human somatosensory system, especially the secondary somatosensory cortex (SII), with respect to its potential somatotopic organization. Eight subjects received electrical stimulation on their right second finger, fifth finger and hallux. Within SII, the typical finding for both fingers was a representation site within the contralateral parietal operculum roughly halfway between the lip of the lateral sulcus and its fundus, whereas the representation site of the hallux was found more medially to this position at the fundus of the lateral sulcus, near the posterior pole of the insula. Somatotopy in SII seems to be less fine-grained than in primary somatosensory cortex (SI), as, in contrast to SI, no separate representations of the two fingers in SII were observed. A similar somatotopic representation pattern between fingers and the hallux was also observed within ipsilateral SII, indicating somatotopy of contra- as well as ipsilateral SII using unilateral stimulation. Further areas exhibiting activation were found in the superior and inferior parietal lobule, in the supplementary and cingulate motor area, and in the insula.
As well as other functional systems, the somatosensory system encompasses multiple cortical areas. Among these are the primary and the secondary somatosensory cortices (Penfield and Rasmussen, 1950; Woolsey, 1958; Mountcastle, 1980; Burton, 1986; Kaas, 1990). The primary somatosensory cortex (SI) of primates, occupying the postcentral gyrus in the anterior parietal lobe, can be subdivided into different cytoarchitectonic areas that are arranged from anterior to posterior. They are termed areas 3a, 3b, 1 and 2 according to the classification of Brodmann and of Vogt and Vogt (Brodmann, 1909; Vogt and Vogt, 1919). In non-human primates, it has been demonstrated that each of the four areas possibly contains a complete map of the contralateral body surface (Merzenich et al., 1978; Kaas et al., 1979; Nelson et al., 1980; Sur et al., 1980; Kaas, 1983; Pons et al., 1985). Recent neuroimaging studies using positron emission tomography (PET) (Burton et al., 1997) and functional magnetic resonance imaging (fMRI) (Lin et al., 1996; Gelnar et al., 1998; Kurth et al., 1998, 2000; Francis et al., 2000; Moore et al., 2000) suggest a similar organization of human primary somatosensory cortex. The subdivisions of SI seem to have different functional roles; for example, neurons in areas 3a and 2 seem to be sensitive especially to stimulation of deep receptors, whereas neurons in areas 3b and 1 exhibit a predominant responsiveness to stimulation of cutaneous receptors (Powell and Mountcastle, 1959; Iwamura et al., 1993).
Despite extensive research in animals, and in contrast to SI, little information is available on the exact spatial extent, structural organization and function of human secondary somatosensory cortex (SII). The existence of this additional ‘second’ somatosensory area was first suggested by Adrian based on studies in cats. He reported on a second representation of the forefoot and hindfoot which was located laterally to the representation of the face in SI (Adrian, 1940, 1941). While Adrian interpreted this finding as a special feature of the cortex of the cat and stated that it could not be found in the dog or the monkey, Woolsey and colleagues were able to show that this second somatosensory area is not so restricted, but is a common feature shared by monkeys as well as several other mammals (Woolsey, 1943, 1944; Woolsey and Wang, 1945; Woolsey and Fairman, 1946). In the brain of the monkey, SII was located within the upper bank of the lateral sulcus in the region of the parietal operculum (Woolsey, 1943). A gross somatotopic arrangement was suggested in the early work of Woolsey with the representations arranged from the lip of the lateral sulcus to its fundus in the order face, arm and leg (Woolsey and Fairman, 1946). In the following decades the parietal operculum of non-human primates was explored more extensively with electrophysiological (Whitsel et al., 1969; Robinson and Burton, 1980a; Krubitzer et al., 1995) as well as anatomical techniques (Friedman et al., 1980; Burton et al., 1995) leading to even more detailed descriptions of SII. Recent studies in non-human primates have suggested that SII within the parietal operculum is composed of two areas containing two mirror-symmetric somatotopic maps of the body surface (Burton et al., 1995; Krubitzer et al., 1995). In man, it was first proposed by Penfield and colleagues that the parietal operculum could be the location of SII, based on studying the effects of cortical stimulation in patients with epilepsy undergoing surgical treatment (Penfield and Rasmussen, 1950; Penfield and Jasper, 1954). Later, Woolsey and colleagues, performing cortical stimulation as well as measuring evoked potentials from the exposed brain, came to a similar conclusion (Woolsey et al., 1979). Non-invasive neuroimaging studies further supported the existence of human SII and its location within the upper bank of the lateral sulcus (Hari et al., 1983, 1984; Seitz and Roland, 1992; Burton et al., 1993, 1997; Lin et al., 1996; Hodge et al., 1998; Maldjian et al., 1999a; Polonara et al., 1999; Francis et al., 2000), and in one recent fMRI study a somatotopic organization of contralateral SII was suggested due to tactile stimulation of the face, hand and foot (Disbrow et al., 2000).
Using electrical stimulation of different body parts, we have previously been able to demonstrate the somatotopic organization of SI in human subjects by fMRI (Kurth et al., 1998; Villringer et al., 1998; Kurth et al., 2000). In these studies, we were able to demonstrate the fine-grained somatotopic organization of the somatosensory hand area by visualizing the distinct cortical representations of single fingers within SI and their sequential medial-to-lateral arrangement from the fifth finger to the thumb. Furthermore, we succeeded in demonstrating multiple representations of a single finger in SI, i.e. its representations in the different cytoarchitectural subdivisions of SI. Using the same type of stimulation in the present study, we addressed the following questions regarding the organization of the human somatosensory system:
Can functional activation of contralateral secondary somatosensory cortex be visualized, and is there evidence for a somatotopic organization of SII?
If so, how fine-grained is the somatotopy of SII, or, more specifically, can the representations of single fingers in SII be distinguished, as it is possible to do in SI?
As there is evidence that neurons in SII possess symmetrical bilateral receptive fields, is there ipsilateral activation of SII and, if so, does ipsilateral activation also show a somatotopic arrangement?
Material and Methods
Eight healthy volunteers (six male, two female, mean age 26.1 years, range 21–31 years) participated in this study. The study was approved by the local ethics committee, and written consent was obtained from each subject prior to investigation.
The subjects received electrical stimulation at three different sites of the right body side: digits II (second finger) and V (fifth finger) of the hand, and digit I of the foot (hallux). Electrical stimulation consisted of monophasic square-wave pulses generated by a clinical neurostimulator (Neuropack 2, Nihon Kohden). The duration of a single pulse was 200 μs. A pulse frequency of 7 Hz was chosen since this frequency turned out to elicit the maximal changes in the primary somatosensory cortex with respect to the amplitude of the signal intensity changes as well as the size of the activated volume (Repenthin et al., 1998). The electrical stimuli were delivered via ring electrodes (Nicolet Biomedical Inc., Madison, WI) to the stimulation sites (cathode proximal, anode distal). Before starting with the fMRI measurements, the pain thresholds were determined individually for all three stimulation sites. Stimulus intensities applied to the different stimulation sites were chosen as high as possible but were still clearly below the pain threshold in order to avoid a possible confounding effect of painful sensations on cortical signal intensity changes. The mean stimulus intensities applied to the different stimulation sites were 8.2 mA for stimulation of the second finger, 6.3 mA for stimulation of the fifth finger and 17.4 mA for stimulation of the hallux. The generally higher stimulus intensities applied to the hallux were necessary probably because of higher skin resistance in this body region.
fMRI Scanning Technique
All fMRI measurements were performed on a 1.5 T Siemens Magnetom Vision scanner with a standard head coil. The head was immobilized by vacuum pads in order to minimize head movements. As it was intended to obtain functional volumes that were orientated parallel to the ‘ac–pc line’ connecting the anterior commissure (ac) with the posterior commissure (pc), the first step consisted of identifying these structures on sagittal T2-weighted images (TR 2200 ms, TE 128 ms, flip angle 180°, eight slices, slice thickness 3 mm) across the subjects' heads. Next, 20 T2-weighted images (TR 5500 ms, TE 128 ms, flip angle 180°, slice thickness 4 mm, interslice gap 1 mm) orientated parallel to the ac–pc line were acquired in order to select the 16 slices for functional imaging that were most likely to include primary as well as secondary somatosensory cortex in the sample volume. In order to measure fMRI signal intensity changes accompanying somatosensory stimulation, a BOLD (Bandettini et al., 1992; Frahm et al., 1992; Kwong et al., 1992; Ogawa et al., 1992) sensitive FID-EPI sequence (TR 3010 ms, TE 60 ms, flip angle 90°, interleaved acquisition of images, field of view 256 × 256 mm, matrix 64 × 64, 16 slices, slice thickness 4 mm, interslice gap 1 mm) was employed.
The stimulation protocol employed in this study consisted of three functional scans in each of which 253 functional volumes were acquired, resulting in an acquisition time of 12 min 39 s per scan. The first three functional volumes were acquired in order to achieve steady-state magnetization and were subsequently excluded from further analysis. During each functional scan, each body site was electrically stimulated four times. The order of stimulating the different body sites in a scan was randomized. Stimulation of a body site consisted of 30 s of stimulation preceded by a 30 s period of rest. The last period of stimulation in each scan was followed by a final 30 s of rest.
After the acquisition of functional data, for each subject a FLASH (fast low-angle shot) sequence (TR 20 ms, TE 5 ms, flip angle 30°) which offered a spatial resolution of 1 × 1 × 1 mm was employed in order to superimpose the functional data on a T1-weighted three-dimensional anatomical data set.
In the first step of data analysis the functional images were motion-corrected (intrarun realignment, sinc interpolation, first-order correction for movement and spin history) using SPM96 (Wellcome Department of Cognitive Neurology, London, UK). Further data analysis was performed using the BrainVoyager® 3.9 software package, courtesy of R. Goebel, Maastricht, The Netherlands (Goebel et al., 1998). It implied the transformation of the individual three-dimensional anatomical dataset into Talairach space using a piecewise affine and continuous transformation for each of 12 predefined subvolumes (Talairach and Tournoux, 1988). The two-dimensional functional volumes were then aligned to this anatomical data set and interpolated into three-dimensional volume time courses with a spatial resolution of 1 × 1 × 1 mm. The volume time courses were then transformed into Talairach space using the parameters specified for the individual anatomical data sets. Further preprocessing of the functional data included the removal of linear drifts, but smoothing in neither the time domain nor the space domain was performed, the latter since spatial smoothing might be critical when differences of cortical locations are examined that may be relatively small. The three-dimensional volume time courses were then z-transformed and appended in time. Multiple regression analysis was performed on the time courses using the General Linear Model. The model included a 6 s time-shifted square-wave function as a predictor for each of the three stimulation conditions: second finger, fifth finger and hallux. Statistical maps were computed for each of the stimulation conditions in each individual and the group. Furthermore, maps for contrasts between the second finger and the fifth finger as well as the second finger and the hallux were computed. On these maps the contribution of each stimulation condition to the variance of the signal time course was displayed colour-coded for each voxel. All maps were rigorously thresholded at correlation coefficients corresponding to P < 0.00001 (uncorrected). For investigation of a somatotopic organization of human SII, the Talairach coordinates of the most significant voxel of the activation within the parietal operculum (activation peak) were determined for each stimulation condition. For visualization of the results, the statistical maps were either shown superimposed onto coronal and transversal sections through the brain or superimposed onto a reconstructed brain surface. The latter was done by segmenting and tesselating the grey–white matter boundary and by inflating the resulting surface mesh (Linden et al., 1999).
In seven out of the eight subjects electrical stimulation of at least two stimulation conditions led to statistically significant intensity changes within the SII region, whereas in one subject an activation within this region was seen during only one of the stimulation conditions. Since the main purpose of the present study was to assess whether there is a somatotopic arrangement within the SII region, we excluded this subject from further analysis so that the final data set that is reported on was gathered from seven subjects.
Electrical stimulation of the fingers and the hallux led to statistically significant increases in fMRI signal intensity in several cortical areas. The results of the individual subjects are reported first, followed by the results that were obtained in the group analysis.
Results of Individual Subjects
A consistent result of the electrical stimulation of the three body sites was a statistically significant increase in signal intensity within the contralateral upper bank of the lateral sulcus. More exactly, these activations were located within the parietal operculum, extending from the lip of the upper bank of the lateral sulcus to its fundus near the posterior pole of the insula. Activation in this region, which presumably corresponded to SII, due to stimulation of the second finger and the hallux was found in all seven subjects, and was found in six subjects due to stimulation of the fifth finger.
There was an obvious difference within the parietal operculum regarding the locations of the activations evoked by the stimulation of the three different body sites. Stimulating the second or the fifth finger, the activation peak was located within the parietal operculum roughly halfway between the lip of the lateral sulcus and its fundus. Stimulating the hallux, in all subjects the activation was located clearly more deeply (medially) within the parietal operculum, at a location close to the fundus of the lateral sulcus and the posterior pole of the insula. In contrast to this separation between the fingers and the hallux, no difference was noted for the cortical representations of the fingers within the upper bank of the lateral sulcus, as indicated by the very similar Talairach coordinates of the activation peaks of both fingers within the single subjects (see Table 1). Examples of our findings concerning the cortical representation sites in contralateral SII are given in Figures 1 and 2.
In addition to this activation of contralateral SII, four subjects also exhibited activation in the ipsilateral parietal operculum during at least one of the stimulation conditions. Activation of ipsilateral SII during stimulation of one finger and the hallux was found in one subject. Here, the somatotopic arrangement was similar to contralateral SII, i.e. the hallux was represented in the depth of the lateral sulcus near the posterior pole of the insula (x = 32, y = –20, z = 6), whereas the second finger was located in a more lateral position within the parietal operculum (x = 41, y = –27, z = 14).
Furthermore, electrical stimulation consistently led to significant signal intensity changes within the contralateral postcentral gyrus corresponding to the activation of primary somatosensory cortex (see Fig. 3). Activation of SI was observed in all seven subjects during the stimulation of the second finger as well as of the hallux, and in six subjects during the stimulation of the fifth finger. The cortical representation of the hallux was found in the mesial wall of the anterior parietal lobe, in some cases extending onto its convexity at the most superior part of the brain. The representations of the fingers were found in the postcentral gyrus just posterior to the location of the motor hand representation of the precentral gyrus, which can be identified easily on axial slices through the brain due to its characteristic omega- or epsilon-like shape (Yousry et al., 1997). Within the somatosensory hand area, the representation of the fifth finger was located medially and superiorly to the representation of the second finger. In individual subjects, distinct activation foci within the postcentral gyrus due to stimulation of the fingers were seen, presumably reflecting the activation of different subdivisions of SI. The typical finding was an activation located within the anterior wall of the postcentral gyrus and a second focus located more posteriorly on its crown or within the adjacent posterior wall of the postcentral gyrus. A separation of foci in this manner was seen in all seven subjects due to stimulation of the second finger and in five subjects due to stimulation of the fifth finger. Additionally, in two subjects, due to stimulation of the second finger, a third focus was seen in the fundus of the central sulcus.
In addition to the activation sites within SI and SII, electrical stimulation resulted in statistically significant signal intensity increases in a number of other cortical areas (see Table 2). These activations were preferentially seen in the hemisphere contralateral to the stimulation, but in several cases they were also found in the ipsilateral hemisphere. Frequently, activation of the so-called posterior parietal cortex, encompassing the superior and inferior parietal lobule, was seen (see Figs 2 and 3). The activations in the inferior parietal lobule were seen in the supramarginal gyrus and/or the angular gyrus. Further cortical sites exhibiting activation were found in the mesial wall of the frontal lobe. In general, these activations were located anteriorly to the primary motor cortex and within the cingulate sulcus, respectively, presumably corresponding to the supplementary and cingulate motor areas (SMA and CMA; see Fig. 3). Additional areas exhibiting activation were found in the insula, mainly in its posterior part.
Results of the Group Analysis
The group analysis (see Fig. 4) showed the typical activation of contralateral SII that was also observed within the single subjects. Statistically significant signal intensity changes were found during the stimulation of all three body sites. The representation sites of the two fingers were apparently not different from each other, and the activation peaks shared similar Talairach coordinates (x = –45, y = –19, z = 16 for stimulation of the second finger, x = –42, y = –22, z = 17 for stimulation of the fifth finger), whereas the hallux representation was located medially to the finger representations (x = –33, y = –21, z = 10). The group analysis also showed significant activation within the ipsilateral SII region during the stimulation of the second finger and the hallux that failed to reach statistical significance within several single subjects. These activations again occurred at cortical sites that were clearly separated from each other and exhibited the typical arrangement that was also seen within the contralateral SII region within the single subjects as well as within the group, with the second finger located more superficially (x = 56, y = –28, z = 15) and the hallux more deeply within the parietal operculum (x = 33, y = –19, z = 11). Furthermore, the group analysis showed the typical activations in contralateral SI that were also observed within the single subjects. The representation of the hallux was again found in the mesial wall of the anterior parietal lobe, whereas the representations of the fingers were found on the convexity of the anterior part of the parietal lobe, the representation of the second finger located laterally to that of the fifth finger.
The present study provides evidence for a somatotopic organization of contralateral as well as ipsilateral SII of man by demonstrating different representation sites for the fingers and the hallux within the parietal operculum due to unilateral innocuous somatosensory stimulation. In contrast to SI, the representations of the two fingers within SII could not be separated.
Activation of SII
The main findings of this study are the demonstration of a somatotopic arrangement within contralateral as well as ipsilateral SII. In contralateral SII, data on individual subjects as well as the group analysis of all seven subjects yielded a typical pattern of somatotopy. The fingers were represented within the parietal operculum approximately halfway between the lip of the lateral sulcus and its fundus, whereas the representation site of the hallux was found more medially within the parietal operculum, close to the fundus of the lateral sulcus and the posterior pole of the insula. The representation site of the hallux as determined in the current study is in agreement with results obtained in non-human primates (Woolsey and Fairman, 1946; Whitsel et al., 1969; Friedman et al., 1980; Robinson and Burton, 1980a; Burton et al., 1995; Krubitzer et al., 1995). In these studies the representation site of the lower limb was located medially to the representation of the hand, approximately at the posterior pole of the insula or the fundus of the lateral sulcus, in some instances extending onto the surface of the insular cortex. Furthermore, the representation sites of the fingers that were located more laterally (superficially) are also in line with these animal studies. Moreover, our results concerning somatotopy within contralateral SII are qualitatively in accordance with the work of Disbrow et al. on human subjects (Disbrow et al., 2000). The type of stimulus and the kind of stimulation used in our study differed from that of the study by Disbrow et al. Whereas we used an electrical stimulus, Disbrow et al. performed a tactile stimulation with a sponge rubbed across the skin. Furthermore, we stimulated relatively small body areas, such as single fingers and the hallux, while in the study by Disbrow et al. comparatively large areas were stimulated, e.g. the whole hand and the whole foot. Stimulating single fingers enabled us to investigate whether somatotopy in SII is as fine-grained as in SI by comparing the representation pattern evoked by the stimulation of two fingers in SII to that in SI. In SII, we found an overlapping representation pattern for the fingers. This finding may be explained as follows: (i) the spatial resolution of our fMRI-approach was insufficient to demonstrate a potential separation, although using the same technique the separation in SI was clearly seen; or (ii) there is a real lack of spatial separation of digits within SII, which is in line with studies in animals demonstrating that neurons in SII responding to somatosensory stimulation of the hand generally possess relatively large receptive fields that encompass many or even all digits of the hands (Robinson and Burton, 1980a; Sinclair and Burton, 1993). However, in an interesting recent study in humans, Gelnar et al. reported some somatotopy for finger representations (first, second and fifth fingers, respectively) not only in SI but also in SII due to vibrotactile stimulation (Gelnar et al., 1998). The authors of that study pointed out that there was no simple somato-topic representation within the SII region. In an attempt to describe a more subtle somatotopy, the spatial distances between the fingers were calculated. Since the distances between the finger representations (greatest between the thumb and the fifth finger) were not statistically significant different from each other, more work on this issue is needed before a positive conclusion can be reached.
Results from recent work in macaques suggest that the SII region of the monkey contains two somatotopic maps of the body surface, termed SII and PV (the ‘parietal ventral’ area) by Krubitzer et al. (Krubitzer et al., 1995), and anterior and posterior SII by Burton et al. (Burton et al., 1995), respectively. The recent study by Disbrow et al. came to a similar conclusion for humans (Disbrow et al., 2000). In the current study we were not able to demonstrate such an additional second representation of the body within the parietal operculum. This might be explained by the spatial organization of these areas, as Krubitzer et al. (Krubitzer et al., 1995) suggested that the representations in these two areas are mirror-images of each other, and that these two areas share a common boundary at the representations of the fingers and the toes, respectively. Due to this common boundary stimulating a single finger or the hallux may give only a single focus of activation.
Activity changes occurring in the contra- as well as the ipsilateral SII region during unilateral somatosensory stimulation have been reported in several neuroimaging studies in humans using magnetoencephalography (MEG) (Hari et al., 1983, 1984, 1993; Simoes and Hari, 1999), PET (Seitz and Roland, 1992; Burton et al., 1993; Ledberg et al., 1995) or fMRI (Maldjian et al., 1999a; Polonara et al., 1999; Disbrow et al., 2000). These bilateral signal changes probably rely on afferent inputs reaching contra- as well as ipsilateral SII, as it has been shown in previous studies in the monkey that SII consists of neurons with receptive fields that are not just confined to the contralateral body side but are found on the ipsilateral side too (Whitsel et al., 1969; Robinson and Burton, 1980a). However, the signal changes in the ipsilateral SII region reported in the studies on humans mentioned above were less pronounced than those in the contralateral hemisphere. This may be because only some of the neurons within SII have receptive fields corresponding to the ipsilateral body side. Whereas Whitsel and colleagues (Whitsel et al., 1969) reported that up to 90% of the neurons of SII have bilateral receptive fields, Robinson and Burton (Robinson and Burton, 1980a) found that only one-third of the neurons have bilateral or ipsilateral receptive fields. As a result, unilateral stimulation may lead to somewhat weaker neuronal activity within ipsilateral SII, and this may be why our statistical significance criterion was met only by some of the subjects and stimulation conditions. Nevertheless, by performing the group analysis we were able to clearly show activations for the second finger as well as the hallux also within the ipsilateral SII region that were separated from each other, exhibiting the same typical somatotopic arrangement observed within the contralateral SII region, with the second finger located more superficially and the hallux more deeply within the parietal operculum.
Thus, the current study confirms results obtained in the recent study by Disbrow et al. (Disbrow et al., 2000) concerning the somatotopic organization of SII inferred from somatotopically arranged cortical representation sites in contralateral SII. Using unilateral stimulation, in this study we found activation of ipsilateral SII due to stimulation of different body parts, as reported earlier (Maldjian et al., 1999a; Polonara et al., 1999; Disbrow et al., 2000), and, furthermore, we were able to demonstrate that these activations are somatotopically arranged. Therefore this study provides evidence for the somatotopic organization of contra- as well as ipsilateral SII using unilateral stimulation.
Activation of SI
In all subjects and for all stimulation sites except one (stimulation of the fifth finger in one subject), significant signal intensity changes occurred in the contralateral postcentral gyrus, reflecting the activation of SI. The representation site of the hallux in SI was found in the mesial wall of the anterior parietal lobe, whereas the fingers were represented on the convexity of the hemisphere, the fifth finger medial to the second finger. The group analysis gave a similar result. These representation sites in SI reflected the known medial-to-lateral somatotopic organization of human SI that has been repeatedly demonstrated, in more or less detail, by means of different methods, such as cortical stimulation (Penfield and Boldrey, 1937; Woolsey et al., 1979; Wood et al., 1988; Allison et al., 1996), measuring evoked potentials from the cortical surface (Woolsey et al., 1979; Wood et al., 1988; Sutherling et al., 1992; Allison et al., 1996), MEG (Baumgartner et al., 1991; Hari et al., 1993), electroencephalography (Baumgartner et al., 1993), PET (Fox et al., 1987) and fMRI (Gelnar et al., 1998; Kurth et al., 1998; Maldjian et al., 1999b; Stippich et al., 1999). The finding of distinct activation peaks across the postcentral gyrus due to stimulation of a single finger presumably reflects the activation of different subdivisions of SI. Most likely, activations seen within the anterior wall of the postcentral gyrus correspond to area 3b, those within the crown and the adjacent posterior wall of the postcentral gyrus to area 1 or 2, and those within the fundus of the central sulcus to area 3a. Our findings are in general agreement with recent work in humans suggesting, in analogy to monkeys, a representation of the body surface possibly within each of the subdivisions of human SI (Lin et al., 1996; Burton et al., 1997; Gelnar et al., 1998; Kurth et al., 1998, 2000; Francis et al., 2000; Moore et al., 2000).
Activation of Additional Cortical Areas
Additional areas exhibiting activation were found contralaterally in individual subjects in posterior parietal cortex, in the SMA and CMA, and in the insula. Besides contralateral activation, in several cases activation within the ipsilateral hemisphere was also seen, because at least some of the neurons in these areas possess bilateral receptive fields (Robinson and Burton, 1980b; Tanji et al., 1988; Schneider et al., 1993; Iwamura et al., 1994).
The so-called posterior parietal cortex encompasses the superior and inferior parietal lobules. In general, these cortical regions are regarded as a site for higher-order processing of sensory information and sensorimotor integration (Sakata et al., 1973; Mountcastle et al., 1975; Hyvarinen, 1982; Caselli, 1993; Andersen, 1997; Colby and Goldberg, 1999). The posterior parietal cortex of humans differs somewhat from that of monkeys. In humans, areas 5 and 7 are located in the superior parietal lobule and are assumed to be the correlates of area 5 of the monkey (Hyvarinen, 1982), which has been shown to be responsive to simple innocuous somatosensory stimulation, though it seems to be engaged especially in the guidance of upper limb movements on the basis of proprioceptive information (Sakata et al., 1973; Nixon et al., 1992). Therefore, the activation in the superior parietal lobule in the present study is in line with data available on area 5 of the monkey and confirms results of recent neuroimaging studies in humans that also reported on activation of the superior parietal lobule during somatosensory stimulation (Forss et al., 1994; Burton et al., 1997; Gelnar et al., 1998; Hodge et al., 1998; Kurth et al., 1998; Polonara et al., 1999; Francis et al., 2000). As the anterior part of the inferior parietal lobule has been proposed as the human homologue of area 7b of the monkey (Hyvarinen, 1982), our finding of activations within the supramarginal gyrus and the angular gyrus in several subjects is consistent with studies in monkeys demonstrating responsiveness of neurons in area 7b to innocuous somatosensory stimulation (Robinson and Burton, 1980c). Activation of this cortical area has also been found in humans in two recent fMRI studies using vibrotactile stimulation (Gelnar et al., 1998; Hodge et al., 1998).
The signal intensity increases that were observed in mesial parts of the frontal lobe anteriorly to the primary motor cortex and within the cingulate sulcus are likely to represent the activation of the SMA and CMA. The meaning of this activation is not clear, as these areas are mainly related to motor function (Roland and Zilles, 1996; Tanji, 1996). However, activation of these areas due to somatosensory stimulation has been also found in electrophysiological studies in monkeys (Romo et al., 1993; Akazawa et al., 2000) as well as humans (Allison et al., 1992, 1996; Dowman and Schell, 1999), and also in recent neuroimaging studies in humans (Lin et al., 1996; Burton et al., 1997; Korvenoja et al., 1999; Polonara et al., 1999).
Finally, our finding of activations within the insula is in good agreement with electrophysiological studies in monkeys (Robinson and Burton, 1980b,c; Schneider et al., 1993), as well as with recent non-invasive imaging studies in humans (Burton et al., 1993; Coghill et al., 1994; Davis et al., 1998; Gelnar et al., 1998; Hodge et al., 1998; Korvenoja et al., 1999; Francis et al., 2000) demonstrating activation of this multifunctional structure due to innocuous somatosensory stimulation.
In summary, using unilateral electrical stimulation of single fingers and the hallux, this fMRI study has demonstrated the somatotopic organization of contralateral and ipsilateral SII. In both hemispheres, the SII representation site of the hallux is located medially to that of the fingers. In contrast to SI, overlapping representation sites were found for the two fingers within the finger–hand area of contralateral SII, presumably indicating a less fine-grained somatotopy within SII.
Address correspondence to Jan Ruben, MD, Department of Neurology, Charité, Humboldt-University, Schumannstrasse 20/21, D-10117 Berlin, Germany. Email: email@example.com.
|Subject||Second finger||Fifth finger||Hallux|
|Subject||Second finger||Fifth finger||Hallux|
|Posterior parietal cortex|
|Superior parietal lobule||Inferior parietal lobule||SMA/CMA||Insula|
|Contra: contralateral hemisphere; Ipsi: ipsilateral hemisphere.|
|Posterior parietal cortex|
|Superior parietal lobule||Inferior parietal lobule||SMA/CMA||Insula|
|Contra: contralateral hemisphere; Ipsi: ipsilateral hemisphere.|