Distributed digit somatotopy in primary somatosensory cortex
Introduction
High-resolution studies of cortical response to cutaneous stimulation in the human have come from preoperative exploration in brain surgery patients. Such studies are restricted to superficial cortex (Gelnar et al., 1998) and may further be limited by deviant cortical organization (Maegaki et al., 1995, Maldjian et al., 1996, Weiller et al., 1993). While the advent of functional magnetic resonance imaging (fMRI; Ogawa et al., 1992) has enabled comparably high-resolution mapping of somatosensory responses (Maldjian et al., 1999), few investigators have examined the precise spatial organization of this activity in humans, even for a body surface as small and accessible as the digits. Researchers have been limited by magnetic field strength (Blankenburg et al., 2003, Gelnar et al., 1998, Kurth et al., 1998), or by restricting their stimulation to two digits (Francis et al., 2000, Kurth et al., 1998, McGlone et al., 2002) or just the tips of the digits (Gelnar et al., 1998, Maldjian et al., 1999, McGlone et al., 2002). Investigators then typically delineate a somatosensory organization from distinct activation hotspots in response to cutaneous stimulation (Francis et al., 2000, Gelnar et al., 1998, Kurth et al., 1998, Maldjian et al., 1999, McGlone et al., 2002). Even when multiple areas are stimulated on the same finger (Blankenburg et al., 2003), the sites are stimulated in separate experiments, thus making it difficult to compare neighboring somatosensory responses.
Here, we present results of high-resolution somatotopic mapping, achieved by using a high-field (4.0 T) magnet and by stimulating the entire surface of three digits: the thumb, index, and ring fingers. While higher-resolution scanning limits the signal-to-noise ratio in individual voxels (Bandettini, 2001), this effect is countervailed by an analysis based on the temporal relation of blood-oxygenation level dependent (BOLD) activity to the stimulation time course. To study this relation, we used the sliding-window stimulation technique that Servos et al. (1998) developed to map out the somatotopic representations of the human arm. This method is based on the phase-analysis technique that was originally employed to yield precise retinotopic maps of visual cortex (Engel et al., 1994, Sereno et al., 1995). While our basic pattern of stimulation is a square wave of stimulation interspersed with non-stimulation periods, the skin surface experiences the stimulation in a moving window that cycles over the surface repeatedly during the experiment. In the case of digit stimulation, the general direction of stimulation is always the same, either from the base of the digit to the fingertip (the proximal-to-distal condition) or from the tip to the base (distal-to-proximal). One such cycle is depicted in Fig. 1, with six channels of stimulation available over the course of the experiment, but only three stimulation locations active at any one time. The two directions of stimulation act as control experiments for one another, with only the temporal pattern of stimulation across the digit acting to distinguish the conditions. Within a somatotopic map of a given digit, the hemodynamic activity should approximate the on/off stimulation pattern, with characteristic delays in each voxel depending on the voxel's receptive area along the digit. Furthermore, such a somatotopic map would necessarily have an early response to tip relative to base stimulation in a distal-to-proximal experiment, but relatively late phase of tip response in proximal-to-distal stimulation.
We individually tested each of the three digits (thumb, index and ring finger) of the right hand of right-handed subjects. Scanning focused on the contralateral sensorimotor region, where the somatotopic representation of the hand is known to occupy a particularly large surface area (Penfield and Boldrey, 1937). Analysis focused on Brodmann's areas 1 and 3b as well as area 3a (all in primary somatosensory cortex, SI) and, further rostrally, area 4 (corresponding to primary motor cortex, MI). Each of these areas has been suggested to contain a parallel somatotopic sensory representation of the body surface among primates (Kaas et al., 1979, Nelson et al., 1980, Strick and Preston, 1978a, Strick and Preston, 1978b, Wong et al., 1978). Areas 3a and 4, however, are involved in motor production (Huffman and Krubitzer, 2001) and are typically responsive to kinesthetic afferents rather than cutaneous inputs (Kaas et al., 1979, Mountcastle, 1998). We therefore treated them as controls for areas 3b and 1, which are responsive to more superficial stimuli (Mountcastle, 1998). Area 2, though also in SI, was not included in the analysis both because it is thought to be secondary to area 3 processing (Künzle, 1978, Vogt and Pandya, 1977) and because the somatotopy here was expected to be obscured by overlapped, multifingered representation of the digits (Darian-Smith et al., 1984, Iwamura et al., 1980, Iwamura et al., 1985).
We also investigated whether the digit representations in these areas reflected several details evident from electrophysiological results, including nonlinearities in the multiplicity, scaling, and arrangement of the maps. More precisely, we located maps of the digits that showed phase reversal coinciding with reversal in stimulation direction, and compared: (1) the frequency of these representations in SI and its controls (areas 4/3a and a simulation of the data described in Materials and methods) as well as within SI areas; (2) the differential representation of the three digits (and the surface within each digit) in terms of map volume and voxel populations within each map; and (3) the topography of the digit responses.
Section snippets
Materials and methods
We obtained high-precision maps of the neural response to hand stimulation by using 4.0 T fMR imaging and a custom-built pneumatic device with a grid of closely spaced stimulation sites. We tested each of three digits (the thumb, index and ring fingers) twice, in opposite directions of stimulation. In data analysis, we paid particular attention to the delineation of Brodmann's areas 4, 3a, 3b, and 1. Functional analysis was applied to these regions, but was not restricted to the knob region
Somatotopic phase maps
We located continuous bands of functional voxels showing a reversed ordering of phase delays associated with a reversal of stimulation direction between experiments on the same digit and subject. Although the search algorithms were designed in principle to look for phasic relations in three dimensions, in practice, all such phase bands were located within-slice in the original pseudocoronal scan planes. As a result of the individually applied ROI masks, these within-subject representations were
Discussion
Three patterns of results deserve further discussion. First, the presence of stimulation-related phase reversing bands in areas 3b and 1 but not in areas 3a or 4 suggests that the input organization of the former areas is relatively discrete and somatotopic. Second, our results in SI confirm those obtained by more invasive recordings as to the nonlinear representation of the body surface in terms of phase band arrangement, scaling, and frequency compared to the digit surfaces stimulated.
Acknowledgment
The authors thank Joe Gati and Andrea Santi for their technical assistance, and James Overduin and Denise Henriques for their comments on the manuscript. Part of this work was supported by grants from the Natural Sciences and Engineering Research Council of Canada to S.A.O. and P.S. and from the Canadian Institutes of Health Research, the Canada Research Chairs program, and the Ontario Premier's Research Excellence Award program to P.S.
References (66)
- et al.
Brodmann's areas 17 and 18 brought into stereotaxic space—Where and how variable?
NeuroImage
(2000) - et al.
Responses of monkey precentral neurons to passive movements and phasic muscle stretch: relevance to man
Electroencephalogr. Clin. Neurophysiol.
(1990) - et al.
fMRI of the responses to vibratory stimulation of digit tips
NeuroImage
(2000) - et al.
Fingertip representation in the human somatosensory cortex: an fMRI study
NeuroImage
(1998) - et al.
Areas 3a, 3b, and 1 of human primary somatosensory cortex. 1. Microstructural organization and interindividual variability
NeuroImage
(1999) - et al.
Areas 3a, 3b, and 1 of human primary somatosensory cortex
NeuroImage
(2000) Deconvolution of impulse response in event-related BOLD fMRI
NeuroImage
(1999)- et al.
Overlapping representation of fingers in the somatosensory cortex (area 2) of the conscious monkey
Brain Res.
(1980) Cortico-cortical efferents of primary motor and somatosensory regions of the cerebral cortex in Macaca fascicularis
Neuroscience
(1978)- et al.
The sensory somatotopic map of the human hand demonstrated at 4 Tesla
NeuroImage
(1999)
Functional neuroimaging studies of human somatosensory cortex
Behav. Brain Res.
Segregation of multiple somatosensory maps within the human postcentral gyrus using fMRI (Abstract)
NeuroImage
I feel my hand moving: a new role of the primary motor cortex in somatic perception of limb movement
Neuron
Distribution of human pacinian corpuscles in the hand. A cadaver study
J. Hand Surg., Br.
Multiple representation in the primary motor cortex
Brian Res.
Sorting of somatosensory afferent information in primate motor cortex
Brain Res.
Noninvasive identification of human central sulcus: a comparison of gyral morphology, functional MRI, dipole localization, and direct cortical mapping
NeuroImage
Brain representation of active and passive movements
NeuroImage
Human cortical potentials evoked by stimulation of the median nerve: I. Cytoarchitectonic areas generating short-latency activity
J. Neurophysiol.
Selection of the optimal pulse sequence for functional MRI
Evidence for a rostral-to-caudal somatotopic organization in human primary somatosensory cortex with mirror-reversal in areas 3b and 1
Cereb. Cortex
Four channels mediate the mechanical aspects of touch
J. Acoust. Soc. Am.
The tangible features of textured surfaces: their representation in the monkey's somatosensory cortex
FMRI of human visual cortex
Nature
Why does the somatosensory homunculus have hands next to face and feet next to genitals? A hypothesis
Neural Comput.
Statistical Analysis of Circular Data
Area 3a: topographic organization and cortical connections in marmoset monkeys
Cereb. Cortex
Receptive field integration and submodality convergence in the hand area of the post-central gyrus of the alert monkey
J. Physiol.
Functional subdivisions representing different finger regions in area 3 of the first somatosensory cortex of the conscious monkey
Exp. Brain Res.
Comparison of the hand and finger representation in areas 3, 1, and 2 of the monkey somatosensory cortex
Tactile sensibility in the human hand: Relative and absolute densities of four types of mechanoreceptive units in glabrous skin
J. Physiol.
Projection pattern of functional components of thalamic ventrobasal complex on monkey somatosensory cortex
J. Neurophysiol.
The somatosensory system of primates
Comp. Primate Biol.
Cited by (80)
Brain Functional Imaging Anatomy
2022, Neuroimaging Clinics of North AmericaThe laminar pattern of resting state in human cerebral cortex
2021, Magnetic Resonance ImagingFunctional Brain Anatomy
2021, Neuroimaging Clinics of North AmericaCitation Excerpt :The sensory and motor systems are both organized topographically at the precentral and postcentral gyrus, respectively. This organization means that each portion of the body has a specific location on the cortex,12,13 which is represented in a topographically organized map known as the “homunculus.” Because the amount of cortex represented in the homunculus is proportional to the degree of precision or discrimination necessary for optimal function of its representative body part, the resultant maps are distorted to emphasize the face, lips, fingers, and hands.
Brain (re)organisation following amputation: Implications for phantom limb pain
2020, NeuroImageCitation Excerpt :Other studies have focused on detailed investigations of one body part (most commonly the hand). Here the classical studies were restricted by limited resolution, making the dissociation of individual digits relatively noisy (Overduin and Servos, 2004). Recent advances in fMRI, and in particular techniques that provide increased signal to noise ratio and spatial resolution (multiband sequences and 7T MRI), offer new opportunities to overcome some of these technical issues.