| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Respiratory
Medicine, We have used
voluntary tongue contraction to test whether we can image activation of
the hypoglossal nuclei within the human brain stem by using functional
magnetic resonance imaging (fMRI). Functional images of the whole brain
were acquired in eight subjects by using T2-weighted echo planar
imaging (blood oxygen level development) every 6.2 s. Sequences of
images were acquired during 12 periods of 31-s "isometric"
rhythmic tongue contraction alternated with 12 periods of 31-s tongue
relaxation. Noise arising from cardiac- and respiratory-related
movement was removed either by filtration (high pass; cutoff 120 s) or
by inclusion in the statistical analysis as confounding effects of no
interest. For the group, tongue contraction was associated with
significant signal increases (P < 0.05 corrected for multiple comparisons) in the sensorimotor cortex,
supplementary motor area, operculum, insula, thalamus, and cerebellum.
For the group and for six of eight individuals, significant signal
increases were also seen within the medulla
(P < 0.001, predefined region of
interest with no correction for multiple comparisons); this signal is
most likely to reflect neuronal activation associated with the
hypoglossal motor nuclei. The data demonstrate that fMRI can be used to
detect, simultaneously, the cerebral and brain stem control of tongue movement.
functional magnetic resonance imaging; hypoglossal nucleus; genioglossus; medulla
IN HUMANS THE CONTROL of tongue movement is essential
for a wide variety of behaviors that include speech, mastication, and swallowing. Additionally, the muscles of the tongue, in concert with
other muscles of the oro- and nasopharynx, play an important role in
maintaining upper airways patency (16). Investigations during
neurosurgery (7, 22) have demonstrated that the tongue is represented,
bilaterally, on the inferior aspect of the "motor homunculus"
close to the lateral fissure. The motor cortex mediates the voluntary
and behavioral control of tongue movement via corticobulbar connections
to the lower motor neurons. The cell bodies of these neurons are
located within the hypoglossal nuclei of the medulla (1). Presumably,
inputs to the hypoglossal motor neurons from the respiratory centers of
the brain stem (1) mediate the respiratory-related activities of the
tongue (16, 18). The anatomic locations of the hypoglossal motor nuclei
are well described; the nuclei are situated, bilaterally, on the dorsal
surface of the medulla at the level of the obex in the midline. In
humans, each nucleus is ~2 mm in diameter and extends rostrocaudally
from the obex over a length of ~20 mm (4, 5, 31). Imaging of brain
function by using positron-emission tomography has confirmed the
location of the representation of the tongue in the motor cortex (13), determined originally by electrophysiological studies (7, 22); however,
we are not aware of any reports that describe the functional anatomy of
the hypoglossal nuclei in the brain stem.
Blood oxygen level dependent (BOLD)-functional magnetic resonance
imaging (fMRI) has now become established as a technique that can be
used to detect focal brain activation in the cortex (2); potentially,
the technique has sufficient sensitivity and spatial resolution to
detect activations within the lower brain stem. However, tissue
movements related to the respiratory and cardiac cycles are prominent
in the brain stem (23), and functional imaging of this region of the
brain is potentially more problematic than cortical imaging. Therefore,
the purpose of the present study was to determine the BOLD-related
signal changes in the brain stem and the cortex that are associated
with the brain activation mediating voluntary tongue contraction. Using this paradigm, we hope to establish appropriate methodologies and
analysis strategies that will allow us to undertake further functional
imaging studies of the whole brain, including the brain stem. The
results of this study have been presented in preliminary form (3).
Subjects.
Eight healthy right-handed subjects were studied (age range 24-47
yr; 4 women); three were authors of the present study. All subjects
gave informed consent and were studied with local ethical approval
(Joint Ethical Committees of The Institute of Neurology and The
National Hospital for Neurology and Neurosurgery). None reported a
history of snoring or was clinically obese.
Imaging.
All brain images were obtained by using a Siemens Vision magnetic
resonance imaging (MRI) scanner, operating at 2 Tesla, with a gradient
booster system and a head volume radio-frequency coil. Each subject lay
supine in the scanner. Head movements were minimized with foam padding.
To reduce the scanner noise, each subject wore foam earplugs; these
incorporated plastic tubes that were connected to a microphone in the
control room. This arrangement allowed verbal instructions to be given
to the subject. After an initial positioning image had been performed,
a T1-weighted "structural" MRI of the subject's brain was
obtained. Subsequently, sequences of functional brain images were
acquired by using T2-weighted echo planar imaging; each sequence
comprised 128 consecutive functional images of the entire brain. These
images were acquired every 6.2 s; each consisted of 64 sequential
transverse planes (spin echo time: 40 ms/plane) with an isotropic voxel
resolution of 3 mm and a matrix size of 64 × 64 pixels. Gaps of
~10 min separated the acquisition of each sequence of functional images.
Physiological monitoring.
In four subjects the electrocardiogram was recorded throughout the
scanning session by using electrodes placed on the posterior surface of
the chest wall. Off-line filtering was performed to remove scanning
artifacts and to obtain information on the timing of the QRS complex
relative to each scanning image. This signal was further filtered to
obtain a low-frequency component that reflected changes in lung volume
related to breathing. Additionally, breathing was monitored by using a
bellows pneumograph. The pneumograph was constructed by using
corrugated rubber tubes placed around the chest and abdomen; these
tubes were connected to a differential pressure transducer (MP45,
Validyne) located outside the scanner room. Respiratory timing
(inspiratory and expiratory times) and an uncalibrated measurement of
tidal volume were then derived from this signal. The partial pressure
of end-tidal CO2
(PETCO2) was determined
from gas continuously sampled via a nasal cannula and analyzed by using
a quadrupole mass spectrometer (MGA 900, Case Medical).
Tongue movement paradigm.
The tongue contraction was designed to be isometric and to produce
minimal displacement of the tongue or jaw; it consisted of a rhythmic
(~1-Hz) self-paced pushing of the tongue against the roof of the
mouth and front upper teeth. For the control condition, the tongue was
relaxed in the same position. We had previously measured electrical
activity from the muscles of the tongue and jaw during this maneuver
and confirmed that the tongue contraction could be performed with
little or no jaw muscle "coactivation." For each functional
imaging sequence, each subject performed 12 periods of 31-s (i.e., 5 whole brain images/period) rhythmic tongue contraction, alternated with
12 periods of 31-s control (tongue relaxed). Auditory cues were given
to begin each period (contraction and control). Four sequences of data
were obtained in the four subjects in whom the physiological monitoring
was performed; two sequences of data were obtained in the remaining
four subjects.
Data analysis.
fMRI data were analyzed by using SPM96 software (Wellcome Dept. of
Cognitive Neurology, Institute of Neurology, London;
http://www.fil.ion.ucl.ac.uk/spm) according to the following procedure.
To account for T1-related relaxation effects, the first five brain
images of each sequence (i.e., the first 31 s) were discarded from the
analysis. Images were then realigned to the first of the remaining
images, normalized into standard stereotaxic space, and spatially
filtered (filter size, full width, half-maximum: 4.5 × 4.5 × 4.5 mm). Statistical analyses were then performed by using the
principles of the general linear model (10) extended to allow the
analysis of fMRI data as a time series (11). The data were first
temporally smoothed by convolving the data with a hemodynamic response
function chosen to represent the relationship of the neuronal
activation to blood flow changes (11). Signal changes relating to the
effect of interest (i.e., tongue activation vs. control) were modeled
in the analysis by using a square wave "boxcar" function smoothed with a five-point Hanning window (19); this function was chosen to
represent the signal changes likely to be induced by using the study's
"on/off" paradigm (11).
ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Respiratory and cardiac data.
Overall, breathing was similar in the two conditions (Table
1). However, there was a small but
significant increase in respiratory rate during tongue contraction;
this was due principally to a shortening of expiratory time.
Inspiratory time, tidal volume, PETCO2, and cardiac interval
were not different between the two conditions (Table 1).
|
Signal increases associated with tongue contraction. Initial analyses were performed by using models 1 and 2 with the data from the four individuals in whom the physiological monitoring was performed. For both the group and the individuals, quantitatively and qualitatively similar results were seen by using both models 1 and 2; consequently, analyses were performed in the entire group (n = 8) by using only model 1; the results for model 1 alone are reported.
For the group (n = 8), tongue contraction, when corrected for multiple comparisons, was associated with significant signal increases bilaterally in the primary sensorimotor cortex, extending down into the upper bank of the operculum and insula (Fig. 1). Additionally, activations were seen in the supplementary motor area (extending down into the cingulate gyrus), the putamen, thalamus, and cerebellum (Fig. 1). The locations of the most significant foci of activation for these regions are reported in Table 2.
|
|
|
|
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
Imaging methodology. The signal changes reported here are consistent with neuronal activations within the cerebral cortex, the subcortical grey matter, and the brain stem. However, this interpretation depends on a number of factors relating to the imaging methodology that are discussed below.
BOLD contrast fMRI detects small signal changes that are related to changes in the magnetization of protons within the blood. These changes are produced by changes in the concentration of deoxyhemoglobin, which is paramagnetic. Normally, neuronal activation is associated with an increase in the strength of the T2-weighted magnetic resonsance signal, which reflects a reduction in deoxyhemoglobin concentration. It has been proposed that deoxyhemoglobin is diluted during neuronal activation because a relative hyperemia occurs in the blood vessels supplying the area of activation (2). Although this method has potentially high sensitivity and spatial resolution, the signal-to-noise ratio is poor, and confounding signal changes can be induced directly and indirectly by brain motion. This motion may be caused by changes in head position between successive images or by tissue movements related to the respiratory and cardiac cycles. Cardiac- and respiratory-related effects are thought to be particularly prominent in subcortical and brain stem structures (23). Furthermore, movement of tissue outside the brain, such as the face or mouth, may change the local magnetic field and induce further artifacts. Such effects can be marked and may increase the variance of the data so that real signal changes become obscured. Alternatively, if the movement is correlated with the activation condition, it may produce an artifactual activation-related signal (14). We have taken a number of approaches to remove such effects from our data during the analysis. As a first step, all the functional brain images for each subject were realigned by using a "rigid body algorithm"; this adjusts the position of each brain image in the series to a chosen reference image by a series of displacements and rotations (8). Additionally, the parameters used for realignment were included within our statistical model as effects of no interest. This has two advantages. First, it will reduce the variance in the data that is related to head motion; this will increase the power of the model to detect significant changes related to the tongue-activation task. Second, it will remove any effects of head motion that correlate with tongue contraction. This is a conservative approach to the analysis, for it also means that any real neuronal activations due to tongue contraction that correlate with the effect of no interest will also be eliminated. The strengths and weaknesses of this approach were highlighted when we reran the analysis, excluding these factors. Generally, the strength and extent of the activations throughout the brain were stronger. In one individual, in whom there was a particularly strong correlation between the realignment parameters and the tongue-contraction state, reanalysis of these data (Fig. 2E) revealed a region of signal increase within the brain stem that was absent in the original analysis (Fig. 2D). Even in those subjects in whom activations were seen with the original analysis, the true extent and location of the activation may be distorted. Although it is likely that the more extensive brain stem signal both in the group and in this subject is due to the neuronal activity associated with the tongue contraction, we cannot exclude the possibility that it is an artifact related to head movement. Effects related to the cardiac and respiratory cycles were accounted for in either of two ways. In model 1, the data were high-pass filtered by using a temporal filter. This filter has two functions; first, it will remove any low-frequency temporal drift in the data (15). Second, because the frequency components of the respiratory and cardiac cycles are high relative to the sampling frequency of the brain images (i.e., TR = 6.2 s), any respiratory and cardiac effects will be undersampled and aliased to low frequencies. The high-pass filter will therefore remove such effects from the data (15); the filter cutoff was chosen to be as high as possible without compromising frequency components related to the experimental conditions. In model 2, by using the approach outlined above, data on the timing of the cardiac cycle and of chest wall position relative to each image were included as effects of no interest within the statistical analysis; this approach allows the confounding effects of these signals to be accounted for with a high temporal resolution despite the low sampling rate of brain images relative to the frequency of the respiratory and cardiac cycles. These two approaches appear equally valid for the data in the present study because the results of the statistical analyses from each model were similar. Although this result may be surprising at first sight, it may in large part be explained by the following. One would anticipate that the improvements due to model 2 would be greatest in regions in which task-related neuronal activation coincided with regions in which signal variance related to cardiac and respiratory effects was high. Using the cardiac and respiratory data as "effects of interest" in a further analysis, we were able to determine the locations of the strongest cardiac and respiratory effects. Cardiac effects were strongest in subcortical areas. For the brain stem, cardiac-related signals appeared to correlate with the position of blood vessels on the surface of the medulla and pons outside the brain tissue. The respiratory-related signal was strongest on tissue and/or cerebrospinal fluid boundaries, in particular on the anterior and posterior cortical surfaces (i.e., on the frontal and parietal lobes). Therefore, because the locations of these signals do not correlate closely with the sites of the tongue-related signal, we would not expect model 2 to improve the data quality substantially. The robustness of these observations, using two different analytical approaches, gives us confidence that the tongue-related signal changes we see, even in the brain stem, are not artifactual. This conclusion is further supported by the data from the physiological monitoring (Table 1) that demonstrated no condition-related changes in heart rate and minimal respiratory changes.Brain stem activation. We chose to inspect the data at two levels of statistical probability. A threshold, corrected for multiple comparisons, was set for the whole brain by using the approach of Friston et al. (9); we can therefore be confident that, although we report significant signal changes in regions about which we have weak or no prior hypotheses, these changes are unlikely to be due to chance. The physiological significance of these activations is discussed further in Activations in primary sensorimotor cortex and Other sites of activation. For the medulla, we set a lower level of significance; because we had an a priori hypothesis, constrained by precise anatomy, no correction for multiple comparisons was needed to test for an activation within this region (9). At this level of significance (P < 0.001), regions of activation within the medulla were seen in the group and in six subjects, although the sites of these activations were not always related to the known anatomic position of the hypoglossal nuclei. In humans, these nuclei are reported to lie on the dorsal surface of the medulla, on either side of the midline, and extend for ~2 cm caudally from a level just caudal to the pontomedullary junction (4, 5, 31). We are not aware of any reports that have systematically studied the variability of the locations of these nuclei. Comparison of the sections of DeArmond et al. (4) and Duvernoy (5) reveals differences in the ventrodorsal position of the nuclei. A probability atlas has been constructed for the regions of the cortex that allow statistical confidence to be assigned to the locations of functional activations (27); such a facility might also be useful to the present study; however, no comparable atlas exists for the nuclei of the brain stem.
Further points suggest that the BOLD signal change may not be anatomically superimposed on the nuclei. BOLD is most likely to detect signal changes in the venules and small veins draining a region of activation. Because the hypoglossal nucleus is supplied by branches of the anterior spinal artery, which also supplies the anterior aspect of the medulla, some of the more distal regions of activation may reflect the venous drainage of this supply. In addition to this, the neurophysiological activity associated with tongue contraction may not be localized to the nuclei; anatomic studies in the rat have shown that the dendrites from the hypoglossal cell bodies extend up to 2 mm beyond the boundaries of the nuclei (28, 32). With these points in mind (including the methodological aspects of the analysis discussed above), the locations of the signal changes reported in the present study appear anatomically consistent with the locations of any signal changes that might result from activation of the hypoglossal nuclei in humans. It is, however, possible that the more ventral sites of activation in the brain stem may reflect activations within the principal olivary nucleus, the voluntary tongue contraction being associated with substantial cerebellar activity. Given the involvement of the cerebellum in such motor tasks, it may not be possible to dissociate these alternative explanations. Although the rostrocaudal extent of the activation in the medulla was less than the known length of the hypoglossal nuclei, the tongue contraction performed in the present study will not have recruited all the muscles innervated via the hypoglossal nucleus (for example, there was no retraction of the tongue). We would not therefore expect to produce activation throughout the entire nuclei. Before the present study, we did not know whether the metabolic and hemodynamic "uncoupling" that determines the activation-related BOLD signal within the cortex would also occur in the medulla. Our data (cf. Fig. 3) suggest that the characteristics of the signal changes in the medulla are largely similar to those observed in the cortex. However, signal changes in the medulla, related to the tongue contraction, were generally smaller and more variable than those in the cortex. Slight differences in the relationship among neuronal activation, metabolic demand, and blood flow, coupled with greater tissue movement in the brain stem, may explain why statistically weaker signal changes were seen in the medulla than in other activated brain regions. It is not clear what factors may lead to more significant signal changes in one individual compared with another. Differences due to individual variability in physiology cannot be excluded. However, gender differences are not thought to be important, and we saw no association of signal sensitivity with gender in the present study. It is only recently that neuroimaging methods have had sufficient sensitivity to detect cortical activations reliably by using single-subject data. Perhaps, then, it is not surprising that in the present study medullary activations were not detected in all individuals. Presently, this may restrict studies addressing brain stem physiology to either a grouped analysis or to a preselected subset of individuals in which brain stem activation has been previously demonstrated by using a paradigm similar to that used in the present study. Others have also attempted to image neuronal activations within the pons and medulla by using BOLD fMRI. Gozal et al. (12) measured signal changes during CO2-stimulated breathing to determine respiratory-related brain activations; the authors reported significant signal decreases in a number of brain regions, including the cortex and medulla, during stimulation. Clearly, such activation-related signal decreases are unexpected given the putative origin of the activation-related BOLD signal; their findings may suggest that CO2 substantially changes the relationship between the BOLD signal and neuronal activation. Additionally, in this early fMRI study, Gozal et al. were unable to utilize more recent developments of signal processing that improve signal detection and exclude signal artifacts. Recently, others have successfully reported activation-related signal foci in the pons and medulla in response to an auditory stimulus (20); for their study, the authors used electrocardiogram-triggered acquisition as a means of reducing cardiac-related artifacts. In the present study, to ensure a strong motor activation of the tongue, but with minimal physical movement, subjects were required to tense the tongue against the roof of the mouth and the back of the upper front teeth. This activation produced a clear condition-related sensory stimulation; the sensory signal from the mouth and teeth would be mediated by the trigeminal (V) nerve. For the anterior two-thirds of the tongue, general sensations (as opposed to taste) are also mediated via the fifth nerve; general sensations from the posterior one-third of the tongue are mediated via the lingual branch of the glossopharyngeal (IX) nerve (29). Sensory information from the fifth nerve will be relayed within the trigeminal sensory nucleus located, bilaterally, within the pons. It is not clear where the twelveth nerve afferents, which relate to general sensation from the tongue, synapse centrally, and this might, in part, be within the nucleus of the solitary tract. It is therefore possible that some parts of the reported medullary signals may be correlates of the sensory-related activation of the tongue. However, it is notable that no activation was detected within the pons that would reflect the sensory-related activation of the trigeminal nucleus. Experimentally, such sensory signals from the tongue might be accounted for, in part, by including a control task that moves the tongue passively; however, it is hard to see how this could be achieved satisfactorily within the scanning environment, when any tissue movements of the head and neck must be minimized.Activations in primary sensorimotor cortex. Both Foerster (7) and Penfield and Boldrey (22) "mapped" the primary motor cortex in humans by electrically stimulating the exposed scalp during neurosurgery. Both of these studies demonstrated that the tongue was represented on the inferior aspect of the primary motor cortex close to the lateral fissure. These early observations are in good agreement with the representation of the tongue in the primary motor cortex imaged by using positron-emission tomography (13). Although initial attempts to determine the representation of the tongue in the motor cortex by using fMRI were unsuccessful (24), a more recent study using fMRI (30) has reported a representation that is consistent with that obtained previously by using positron-emission tomography. Our own unpublished observations, using focal transcranial magnetic stimulation, have shown that stimulation of the tongue is most effective at a site on the scalp that corresponds to an inferolateral representation on the motor strip. The sensory representation of the tongue has been obtained by using fMRI by electrically stimulating the tip of the tongue (25); with this stimulus, a focus of activation was detected in the contralateral postcentral gyrus in a position that is lateral to the representation obtained for the fingertips.
In the present study, we report relatively large bilateral regions of activation that encompass the inferolateral primary sensory and motor cortexes; a number of distinct local foci are present within this region. Because the precise position of the central sulcus is uncertain in our averaged group results (cf. Fig. 1), we have not ascribed our principal focus of activation to the pre- or postcentral sulcus and report this simply as primary sensorimotor cortex. This location corresponds well with the previously reported representations of the tongue (7, 22, 25, 30). Although the tongue activation was performed with slight or no detectable coactivation of the jaw muscles, it is possible that some of the anatomically more superior motor representations reflect such muscle recruitment. Some of the more superior sensory-related signals are likely to reflect the representation of hard palate and upper teeth in addition to that of the tongue.Other sites of activation. Studies in nonhuman primates have shown that there are many "nonprimary" motor areas in the cortex that are associated with the voluntary control of movement (reviewed in Ref. 6). Some of these areas show somatotopic mapping; many have direct connections to the primary motor cortex or spinal cord. Fink et al. (6) used positron-emission tomography to map the location of such areas in humans by using voluntary movements of the hand, shoulder, and leg. It is notable that all the nonprimary cortical areas associated with tongue contraction observed in the present study (i.e., the supplementary motor area, premotor cortex, the upper bank of the operculum, the insula, and upper parietal cortex) were also identified during hand, shoulder, and leg movements (6). Additionally, in the present study, tongue contraction was associated with significant signal increases within the thalamus and cerebellum, activations that are typical of those associated with the voluntary control of movement (21).
Conclusions. We have used fMRI to characterize the motor areas in the brain that are associated with voluntary tongue contraction. Using this paradigm, we have identified signal changes within the primary motor cortex and other motor-related areas of the cortex together with signal changes in subcortical structures that include the medulla. The signal changes within the medulla are most likely to reflect the activation of the hypoglossal motor nuclei, although additional components relating to the activation of olivary nuclei may be present. Thus we believe that it is now possible to use fMRI to image, simultaneously, activation of the primary motor cortex and activation associated with the recruitment of the "lower" motor neuron. The methodologies described here may be suitable for other brain stem-related activation studies; in particular, the methodology may allow the investigation of "nonvoluntary" motor-related activations, such as those that occur with automatic breathing.
| |
ACKNOWLEDGEMENTS |
|---|
D. R. Corfield, G. R. Fink, R. S. J. Frackowiak, O. Josephs, and R. Turner are supported by the Wellcome Trust.
| |
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: D. R. Corfield, Dept. of Respiratory Medicine, National Heart and Lung Institute, Imperial College School of Medicine, Charing Cross Campus, St. Dunstan's Road, London W6 8RP, UK (E-mail: d.corfield{at}ic.ac.uk).
Received 26 September 1997; accepted in final form 14 January 1999.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
1.
Berger, A. J.,
and
M. C. Bellingham.
Mechanisms of respiratory motor output.
In: Regulation of Breathing, edited by J. Dempsey,
and A. Pack. New York: Dekker, 1996, p. 71-149.
2.
Cohen, M. S.
Rapid MRI and functional application.
In: Brain Mapping
The Methods, edited by A. W. Toga,
and J. C. Mazziotta. London: Academic, 1996, p. 223-255.
3.
Corfield, D. R.,
K. Murphy,
O. Josephs,
G. R. Fink,
R. S. J. Frackowiak,
L. Adams,
and
R. Turner.
Brainstem activation during tongue movement imaged with fMRI (Abstract).
Neuroimage
2:
S254,
1997.
4.
DeArmond, S. J.,
M. M. Fusco,
and
M. M. Dewey.
Structure of the Human BrainA Photographic Atlas. New York: Oxford Univ. Press, 1989.
5.
Duvernoy, H. M.
The Human Brainstem and Cerebellum. Vienna: Springer-Verlag, 1995.
6.
Fink, G. R.,
R. S. J. Frackowiak,
U. Pietrzyk,
and
R. E. Passingham.
Multiple non-primary motor areas in the human cortex.
J. Neurophysiol.
77:
2164-2174,
1997
7.
Foerster, O.
Motorische Felder und Bahnen. Berlin: Springer, 1936.
8.
Friston, K. J.,
J. Ashburner,
C. D. Frith,
J.-B. Poline,
J. D. Heather,
and
R. S. J. Frackowiak.
Spatial registration and normalisation of images.
Human Brain Mapping
2:
165-189,
1995.
9.
Friston, K. J.,
A. Holmes,
J.-B. Poline,
C. J. Price,
and
C. D. Frith.
Detecting activations in PET and fMRI: levels of inference and power.
Neuroimage
40:
223-235,
1996.
10.
Friston, K. J.,
A. P. Holmes,
K. Worsley,
J.-B. Poline,
C. Frith,
and
R. S. J. Frackowiak.
Statistical parametric maps in functional imaging: a general linear approach.
Human Brain Mapping
2:
189-210,
1995.
11.
Friston, K. J.,
P. Jezzard,
and
R. Turner.
Analysis of functional MRI time series.
Human Brain Mapping
1:
153-171,
1994.
12.
Gozal, D.,
G. M. Hathout,
K. A. Kirlew,
H. Tang,
M. S. Woo,
J. Zhang,
R. B. Lufkin,
and
R. M. Harper.
Localization of putative neural respiratory regions in the human by functional magnetic resonance imaging.
J. Appl. Physiol.
76:
2076-2083,
1994
13.
Grafton, S. T.,
R. P. Woods,
J. C. Mazziotta,
and
M. E. Phelps.
Somatotopic mapping of the primary motor cortex in humans: activation studies with cerebral blood flow and positron emission tomography.
J. Neurophysiol.
66:
735-743,
1991
14.
Hajnal, J. V.,
R. Myers,
A. Oatridge,
J. E. Schwieso,
I. R. Young,
and
G. M. Bydder.
Artifacts due to stimulus correlated motion in functional imaging of the brain.
Magn. Reson. Med.
31:
289-291,
1994.
15.
Holmes, A. P.,
O. Josephs,
C. Buechel,
and
K. J. Friston.
Statistical modelling of low-frequency confounds in fMRI (Abstract).
Neuroimage
5:
S480,
1997.
16.
Horner, R. L.
Motor control of the pharyngeal musculature and implications for the pathogenesis of obstructive sleep apnea.
Sleep
19:
827-853,
1997.
17.
Josephs, O., A. M. Howseman, K. Friston, and R. Turner.
Physiological noise modelling of EPI fMRI using SPM (Abstract).
Br. Chapt. Int. Soc. Magn. Reson.
Med. A27, 1996.
18.
Kobayashi, I.,
A. Perry,
J. Rhymer,
B. Wuyam,
P. Hughes,
K. Murphy,
J. A. Innes,
J. McIvor,
A. D. Cheesman,
and
A. Guz.
Inspiratory coactivation of the genioglossus enlarges retroglossal space in laryngectomized humans.
J. Appl. Physiol.
80:
1595-1604,
1996
19.
Lynn, P. A.
An Introduction to the Analysis and Processing of Signals. London: Macmillan Education, 1982.
20.
Melcher, J. R.,
B. C. Fullerton,
and
R. M. Weisskoff.
Imaging human auditory function from brainstem to cortex.
Neuroimage
2:
S172,
1997.
21.
Passingham, R. E.
The Frontal Lobes and Voluntary Action. Oxford: Oxford Univ. Press, 1993.
22.
Penfield, W.,
and
E. Boldrey.
Somatic motor and sensory representation in the cerebral cortex as studied by electrical stimulation.
Brain
15:
389-443,
1938.
23.
Poncelet, B. P.,
V. J. Wedeen,
R. M. Weisskoff,
and
M. S. Cohen.
Brain parenchyma motion: measurement with cine echo-planar MR imaging.
Radiology
185:
645-651,
1992
24.
Rao, S. M.,
J. R. Binder,
T. A. Hammeke,
P. A. Bandettini,
J. A. Bobholz,
J. A. Frost,
B. M. Myklebust,
R. D. Jacobson,
and
J. S. Hyde.
Somatotopic mapping of the human primary motor cortex with functional magnetic resonance imaging.
Neurology
45:
919-924,
1995[Abstract].
25.
Sakai, K.,
E. Watanabe,
Y. Onodera,
H. Itagaki,
E. Yamamoto,
H. Koizumi,
and
Y. Miyashita.
Functional mapping of the human somatosensory cortex with echo-planar MRI.
Magn. Reson. Med.
33:
736-743,
1995[Medline].
26.
Talairach, J.,
and
P. Tournoux.
Coplanar Stereotaxic Atlas of the Human Brain. Stuttgart: Thieme, 1988.
27.
Thompson, P. M.,
C. Schwartz,
and
A. W. Toga.
High-resolution random mesh algorithms for creating a probabilistic 3D surface atlas of the human brain.
Neuroimage
3:
19-34,
1996[Medline].
28.
Wan, X. S. T.,
J. Q. Trojanowski,
J. O. Gonatas,
and
C. N. Liu.
Cytoarchitecture of the extra-nuclear and commisural dendrites of hypoglossal nucleus neurones as revealed by conjugates of horseradish peroxidase with cholera toxin.
Exp. Neurol.
78:
167-175,
1982[Medline].
29.
Weddell, G.,
J. A. Harpman,
and
D. G. Lambley.
The innervation of the musculature of the tongue.
J. Anat.
74:
255-266,
1940[Medline].
30.
Wildgruber, D.,
H. Ackermann,
U. Klose,
B. Kardatzki,
and
W. Grodd.
Functional lateralization of speech production at primary motor cortex: a fMRI study.
Neuroreport
7:
2791-2795,
1996[Medline].
31.
Williams, P. L.,
and
R. Warwick
(Editors).
Gray's Anatomy. Edinburgh: Churchill Livingstone, 1980Williams, P. L., and R. Warwick (Editors).
Gray's Anatomy. Edinburgh:
Churchill Livingstone, 1980.
32.
Withington-Wray, D. J.,
S. W. Mifflin,
and
K. M. Spyer.
Intracellular analysis of respiratory-modulated hypoglossal motoneurons in the cat.
Neuroscience
25:
1041-1051,
1988[Medline].
This article has been cited by other articles:
|
|
 |
|
  T. Saarinen, H. Laaksonen, T. Parviainen, and R. Salmelin Motor Cortex Dynamics in Visuomotor Production of Speech and Non-speech Mouth Movements Cereb Cortex, February 1, 2006; 16(2): 212 - 222. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. E. Martin, B. J. MacIntosh, R. C. Smith, A. M. Barr, T. K. Stevens, J. S. Gati, and R. S. Menon Cerebral Areas Processing Swallowing and Tongue Movement Are Overlapping but Distinct: A Functional Magnetic Resonance Imaging Study J Neurophysiol, October 1, 2004; 92(4): 2428 - 2493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Nakahara, N. Nakasato, A. Kanno, S. Murayama, K. Hatanaka, H. Itoh, and T. Yoshimoto Somatosensory-evoked Fields for Gingiva, Lip, and Tongue Journal of Dental Research, April 1, 2004; 83(4): 307 - 311. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. C. McKay, K. C. Evans, R. S. J. Frackowiak, and D. R. Corfield Neural correlates of voluntary breathing in humans J Appl Physiol, September 1, 2003; 95(3): 1170 - 1178. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Shinagawa, T. Ono, Y. Ishiwata, E. Honda, T. Sasaki, M. Taira, A. Iriki, and T. Kuroda Hemispheric Dominance of Tongue Control Depends on the Chewing-side Preference Journal of Dental Research, April 1, 2003; 82(4): 278 - 283. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.J. Lavigne, T. Kato, A. Kolta, and B.J. Sessle NEUROBIOLOGICAL MECHANISMS INVOLVED IN SLEEP BRUXISM Critical Reviews in Oral Biology & Medicine, January 1, 2003; 14(1): 30 - 46. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Komisaruk, K. M. Mosier, W.-C. Liu, C. Criminale, L. Zaborszky, B. Whipple, and A. Kalnin Functional Localization of Brainstem and Cervical Spinal Cord Nuclei in Humans with fMRI AJNR Am. J. Neuroradiol., April 1, 2002; 23(4): 609 - 617. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. S. Berns, S. M. McClure, G. Pagnoni, and P. R. Montague Predictability Modulates Human Brain Response to Reward J. Neurosci., April 15, 2001; 21(8): 2793 - 2798. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Sawczuk and K.M. Mosier Neural Control of Tongue Movement With Respect To Respiration and Swallowing Critical Reviews in Oral Biology & Medicine, January 1, 2001; 12(1): 18 - 37. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
4 mm,
z = 18 mm;
Z score = 9.12); this maximum is
identified by intersection of crosshairs on each slice in
B and by tips of white arrowheads in
A. R, right; A, anterior.
, signal intensity obtained from
statistical model corrected for effects of no interest
(model 1; see text for details); thick
lines, model estimate (increasing signal intensity corresponding to
periods of tongue activation). Right:
same data averaged image by image (±SE) for 12 periods of tongue
contraction. 
, periods of tongue contraction.
Signal intensity has been normalized to mean signal intensity
determined for whole brain. Data are from subject represented in Fig.