Chem. Senses 28: 245-251,
2003
© Oxford University Press 2003
Magnetoencephalographic Study of Cortical Activity Evoked by Electrogustatory Stimuli
Department of Behavioral Physiology, Graduate School of Human Sciences, Osaka University, 1-2 Yamadaoka, Suita, Osaka 565-0871 1 Life Electronics Laboratory, National Instistute of Advanced Industrial Science and Technology (AIST), 1-8-31, Midorigaoka, Ikeda, Osaka 563-8577, Japan
Correspondence to: Takashi Yamamoto, Department of Behavioral Physiology, Graduate School of Human Sciences, Osaka University, 1-2 Yamadaoka, Suita, Osaka 565-0871, Japan. e-mail:yamamoto{at}hus.osaka-u.ac.jp
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Electrogustometry is a convenient method to examine taste acuity in clinical situations. Some basic properties of neural activity in the cerebral cortex in response to electrogustatory stimulation were revealed by measuring magnetoencephalography (MEG) signals with a whole-cortex-type system in response to varying intensities of anodal DC currents focally applied to the tongue surface in human subjects. Independent component analysis was used to eliminate stimulus artifacts in MEG signals. Electrogustatory stimulation with intensities of induced electric taste evoked responses bilaterally, mainly in the opercular–insular cortex with a mean onset latency of
350 ms,
while subthreshold electrogustatory stimulation induced modest responses in
the cortex. Stronger stimulation induced a tingling sensation and elicited
large transient responses in both the opercular–insular and somatic
sensory cortices. This is the first description of the basic properties of
human MEG responses to electrogustatory stimulation.
Key words: electric taste, human, magnetoencephalography, taste
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Magnetoencephalography (MEG) is a non-invasive functional brain imaging technique that has good temporal resolution compared to functional magnetic resonance imaging (fMRI) or positron emission tomography (PET). It can give a good estimation of the localization of the activities with more precision than electroencephalography (EEG), and has been used as a tool to analyze cortical taste responses (Kobayakawa et al., 1996
,
1999;
Murayama et al.,
1996).
To obtain a good summation of gustatory evoked brain responses that can be
attained by a sophisticated taste-delivery system, one of the requirements in
an MEG study is to apply repetitive taste stimulation with a well-synchronized
onset timing (Kobayakawa et al.,
1999). By applying an anodal DC current to the tongue, which is
known to elicit a unique taste called ‘electric taste’
(Bujas, 1971;
Frank and Smith, 1991), this
requirement can be easily fulfilled due to the rectangular pulses that result
from this electrogustatory stimulation. Further, while usually mouth rinsing
must follow stimulation with taste solutions, in this situation it is not
necessary, which leads to shorter experimental times. Electrogustatory
stimulation also has the advantage of stimulating a confined area of the
tongue. Because of these advantages, various electrogustometric analyses have
been developed and extensively used by clinicians to evaluate taste disorders
using electrogustometers (Tomita and
Ikeda, 2002).
However, with the exception of the classical EEG studies (Plattig,
1969,
1971;
Plattig and Kobal, 1977;
Schaupp, 1971) and the recent
fMRI study by Barry et al. (Barry
et al., 2001), the response characteristics of the human
brain to electrogustatory stimuli have not been well analyzed. The present
study aimed to use MEG to reveal some basic response characteristics of
cortical neurons to electrogustatory stimulation, and also to address the
question of whether projections of gustatory information to the human cortex
are ipsilateral or contralateral to the side of the tongue stimulated.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Subjects
Twelve healthy right-handed subjects (six females and six males, aged 21–40 years) participated in this study. All were fully informed of the nature of the experiments, and agreed to be subjects. The experiment was conducted in accordance with the revised version of the Helsinki Declaration and was approved by the Osaka University Ethical Committee.
Stimuli
A commercially available electrogustometer (TR-05, Rion Co., Tokyo, Japan)
was used to deliver anodal DC currents with varying strengths and durations.
Two circular (7 mm in diameter) anode electrodes plates
(Ag–AgCl2) were placed on the edge of the anterior area of
the tongue and a similar cathode electrode was taped to the tip of the chin.
The electrodes were connected by a twisted-pair of copper wires to the
electrogustometer. The anodes were securely placed on the left and right
lateral edges, 2 cm from the midline
(Tomita and Ikeda, 2002), with
a special chamber made of silicon rubber. Currents were applied to the anodes
of either side. After a pilot study, we decided on a stimulus duration of 200
ms and an interstimulus interval of 20 s. Current strength varied among
subjects due to differences in sensitivity to the anodal currents applied.
Therefore, prior to the MEG recording session, subjects were asked to verbally
express what they had perceived after the delivery of various current
strengths ranging from 5 µA (minimum) to 210 µA (maximum) by decibel
steps (Tomita and Ikeda,
2002). A consistently detectable range of stimulus strength was
chosen for each subject.
Recording
Each subject was comfortably seated on a non-magnetic chair in a magnetically shielded room. The onset of electrical stimulation provided a trigger signal for MEG averaging. The subjects were instructed not to change their head position, to keep their eyes open and fixate on a point in front of them.
Each subject participated in three sessions per day with intermissions: (i) a weak current eliciting an unidentified sensation; (ii) a moderate current eliciting taste sensation; and (iii) a strong current eliciting a somatosensory tingling sensation (or irritation). Each subject received 60 trials of electric stimulation per session, and data were averaged on-line. Trials containing eye blink artifacts were rejected from the averaging process.
Brain magnetic fields were recorded with a whole-cortex, 122-channel SQUID system (Neuromag-122TM, Neuromag Ltd, Helsinki, Finland). The MEG sensor positions with respect to each subject's head were determined by measuring the magnetic fields generated by three marker coils located on the scalp, whose locations in relation to three landmark points at nasion and two preauricular points were determined before the experiments using a three-dimensional digitizer (Polhemus, Inc., Colchester, VT). Stimulus related epochs of 1100 ms, including a 100 ms pre-stimulus baseline, were recorded with a pass-band of 0.03–100 Hz and a sampling rate of 400 Hz.
Data processing
Stimulus-related artifacts in the MEG signals were removed by independent
component analysis (ICA) (Iwaki et
al., 2003). We adopted an ICA technique based on an
information maximization (infomax) approach
(Makeig et al.,
1997), which minimizes mutual information among the output
independent components by maximizing the joint entropy of the output of a
simple neural network that is an ensemble of ‘sphered’ (zero-mean)
input vectors, linearly transformed and sigmoidally compressed
(Makeig et al.,
1997). Infomax ICA was applied to the 60 daily trials of each
subject measured by 122 magnetometer channels producing 122 temporally
independent components, and the trials for each component were averaged
separately. Independent components obviously representing the stimulus-related
artifacts were excluded, and the remaining components were projected onto the
original MEG signal space to reconstruct the artifact-free MEG signals. After
averaging the data, source estimation was carried out using single equivalent
current dipole (ECD) analysis (Imada et
al., 2001). In the single ECD analysis, all the MEG sensors
were divided into 34 overlapping local sensor groups having 14–20
sensors each. The single equivalent current dipole was estimated for each
local sensor group with a time interval of 2.5 ms, and only dipoles that
continuously (>10 ms) and simultaneously satisfied the criteria that (i)
the GOF values were >80% and (ii) the confidence volume was <2000
mm3 were selected for further analysis. Estimated ECDs were
superimposed on MR images.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Figure 1 shows intensity–perception profiles for seven subjects when different strengths of anodal DC currents were applied to the right side of the tongue surface. Although the strength and range of current varied among subjects, they generally felt no sensation or unidentified sensations (or weak tingling) (<25 µA), taste sensations (25–50 µA) and finally tingling sensations (somatosensory but not a taste sensation) (>50 µA) as the strength of electric current increased. Unidentified sensations occurred before (n = 2) or after (n = 2) taste sensations, but were not reported by three subjects.
|
Figure 2 shows averaged MEG wave forms obtained from one of 122 channels in a subject undergoing electrogustatory stimulation with a strength of 32 µA and duration of 200 ms. As shown in Figure 2A, the original MEG signal contains a large artificial deflection corresponding to the current delivery period. Figure 2B shows six typical components decomposed by ICA. Components a and b are stimulus-related noise components which should be disregarded. Figure 2C shows the MEG waveform after reduction of these noise components.
|
Figure 3 shows three
superimposed traces from 24 channels overlying the left temporal region in a
subject. Each trace represents MEG averaged over 60 trials at one channel. At
4 µA stimulation, this subject felt no particular sensation and showed
small, transient responses (Figure
3A). At 32 µA, the subject perceived electric taste and showed
prolonged responses during and after the course of analysis
(Figure 3B). At 80 µA, the
subject felt a tingling sensation without any taste sensation (230 ms)
and showed large, but brief (400 ms) responses with a shorter onset
latency than that for the taste response.
|
When the first ECD after onset of stimulation was estimated in the cerebral
cortex using a single-dipole model and was superimposed on the subject's MRI,
the source was detected in the anterior insular and adjoining frontal
operculum known as the ‘primary taste area’ (PTA)
(Kinomura et al.,
1994; Kobayakawa et
al., 1996; Murayama
et al., 1996; Cerf
et al., 1998; Small
et al., 1999). Figure
4 shows locations of ECDs in the PTA in the left hemisphere when
the right side of the tongue was stimulated (32 µA). Group analysis
suggested that the areas commonly activated across subjects by electric taste
stimuli included a considerable anteroposterior part of the
insular–operculum (from the frontal to parietal operculum), and the
superior part of the insular rather than the interior part.
|
MEG responses to electrical stimulation of the tongue were also observed in cortical areas other than the PTA, such as the pre-and post-central gyri, superior temporal gyrus, cuneus, angular gyrus, parahippocampal gyrus and supramarginal gyrus. The cortical areas activated differed depending on current intensity. Figure 5 shows the cumulative numbers of estimated ECDs in the bilateral opercular–insular region, frontal, parietal, temporal and occipital lobes counted every 100 ms after onset of stimulation in seven subjects. At a current intensity inducing electric taste, the most prominent responses were detected in the opercular–insular cortex: ECDs were observed not only in the first 500 ms but also in the second 500 ms with more frequent occurrences (Figure 5A), which corresponds to the prolonged MEG taste responses (see Figure 3B). Estimated ECDs to strong currents inducing irritation were observed in both the opercular–insular region (Figure 5A), the interior part of the post-central gyrus that is found in the parietal lobe (Figure 5B), and the superior temporal gyrus that is found in the temporal lobe (Figure 5D). The ECDs were found most frequently within 200–300 ms after onset of stimulation, and the number of ECDs decreased quickly corresponding to the large, transient MEG responses to strong stimulation (see Figure 3C). ECDs were also detected in the frontal (Figure 5C) and occipital (Figure 5E) lobes.
|
To examine ipsilateral or contralateral dominance of the projections from the tongue, electrogustatory stimulation was applied separately to the left or right sides of the anterior area of the tongue, and the latency of the MEG response was measured. Figure 6 shows intensity–perception profiles for six subjects when different strengths of anodal DC currents were applied to either the left or right sides of the tongue. Similar to Figure 1, although the strength and range of current varied among subjects, they generally felt no sensation, weak tingling, a taste sensation and finally a tingling sensation as the strength of the electric current increased. The intensity–perception profiles are very similar for right and left stimulations in each subject.
|
Figure 7 shows the latencies
of the first detection of ECDs in the left or right opercular–insular
region (or PTA) to electrogustatory stimulation of the left or right areas of
the tongue. The mean latencies (350 ms) of the ipsilateral responses,
i.e. responses in the right PTA to right side stimulation or in the left PTA
to left side stimulation, tended to be shorter than those of contralateral
responses. However, differences among the four latencies were not
statistically significant [one-way ANOVA, F(1,12) = 1.34, P
= 0.27].
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Electrogustometry is widely used by clinicians to examine taste acuity (Krarup, 1958
; Fons and
Osterhammel, 1966; Nakazato et
al., 2002; Tamita and Ikeda, 2002). Anodal DC currents are
delivered to focal regions of the dorsal area of the tongue to elicit a unique
and complex taste called ‘electric taste’ that consists mainly of
sour and metallic tastes (Bujas,
1971). In the present study, a commercially available TR-05
electrogustometer, originally developed by Tomita
(Tomita and Ikeda, 2002), was
used. This equipment provides a quantitative evaluation of taste in decibels
ranging from –8 dB (3 µA) to 34 dB (400 µA). Thus,
electrogustometry has the following advantages
(Tomita and Ikeda, 2002): (i)
the range of measurement can be kept constant; (ii) quantitative control of
the intensity of the stimulation is possible; and (iii) only a short period of
time is required for testing.
A disadvantage of using electrogustometry is the elicitation of unfamiliar
unique tastes, which are often difficult for subjects to verbally characterize
(Bujas, 1971). Another drawback
is the induction of large stimulus artifacts, especially when recording neural
activities of the brain with EEG (Plattig,
1969) or MEG (present study). Artifacts interfere with recordings
of evoked brain activity corresponding to electric stimuli. In the present
study, however, independent component analysis successfully eliminated
artifact components from the stimulus-elicited MEG signals.
Threshold current intensities for eliciting taste sensation varied
considerably among subjects. This might have been due to the difficulty of
identifying the electric taste even after familiarization to it through
pre-experimental practices. Consequently, in the present study, we had to
deliver different current intensities to different subjects to elicit this
electric taste. Generally, subjects felt three types of sensations depending
on the strength of the current delivered: a weak tingling during weak
stimulation (<25 µA), followed by electric taste during mild stimulation
(25–50 µA) and tingling or irritation without any taste sensation
during strong stimulation (>50 µA). These results suggest that strong
electric currents stimulate trigeminal afferents
(Frank and Smith, 1991;
Murphy et al., 1995),
and that somatosensory information may interfere with that via taste
afferents.
In agreement with the results of an fMRI study on electrogustatory-evoked
cortical activity (Barry et al.,
2001), activation was seen in many areas of the brain, but most
prominently in the opercular–insular region known as the PTA
(Kinomura et al.,
1994; Kobayakawa et
al., 1996; Murayama
et al., 1996; Cerf
et al., 1998; Small
et al., 1999): numbers of estimated ECDs were most
frequently counted in this region. More precisely, the ECDs were observed
within 200–1000 ms after onset of electrogustatory stimulation, with the
number of ECDs more frequently observed in the second half of the 1 s analysis
time than in the first half, indicating a long-lasting activation of neurons
in this region to electrogustatory stimulation. Although estimated ECDs to
strong currents were also observed in this region, the differences are that
the ECDs were found most frequently within 200–300 ms after onset of
stimulation, the number of ECDs decreased quickly and were not observed in the
second half of the 1 s analysis time (see
Figure 3). These results
indicate that the strong tingling comes first and disappears quickly, whereas
electric taste comes slowly and lasts longer. The present MEG study suggests
that electric taste stimulation (25–50 µA) is not a complex stimulus,
but has an exclusively gustatory component since the elicited responses were
seen mostly in the PTA. On the other hand, stronger stimulation (>50 µA)
eliciting a tingling sensation may be a complex stimulus, which activates both
the post-central gyrus [somatosensory area
(Pardo et al., 1997)]
and the opercular–insular cortex. However, elicited activities in both
regions were transient, but not as long-lasting as exhibited in
electrogustatory MEG responses. There is a possibility that concurrent strong
somatosensory inputs interfere with taste-elicited responses in the
cortex.
Electric stimulation can be applied focally and specifically on the left or
right sides using a small electrode; a 7 mm diameter electrode in the present
study. This method may address the issue of whether taste information projects
from the tongue to the cortex ipsilaterally or contralaterally. Judging from
the latency of the initial response in the opercular–insular region, the
results here suggest that taste information is conveyed bilaterally. This is
consistent with the report of Genow et al.
(Genow et al., 1998)
showing that latency and amplitude of taste-evoked human EEG recordings were
identical in both hemispheres with unilateral stimulation. Barry et
al. (Barry et al.,
2001) found in an fMRI study that electrogustatory stimulation
induced responses exclusively in the right hemisphere in right-handed
subjects. The results of the present study together with Barry et
al.'s data, suggest that taste information may extend to the cortex
almost bilaterally symmetrically, but the information may predominantly be
processed in the right side.
Recording of MEG responses to taste stimulation using chemical solutions can be challenging because a stimulus delivery system that allows repetitive stimulation with a quick rise time of onset to induce synchronized neural activity for a good summation of evoked responses is required. From this aspect, electrogustatory stimulation has an advantage since electrical stimulation can trigger synchronized neural activity in the brain. Further, it does not require rinsing after each stimulation, which can contribute to subjects' fatigue and increase the number of stimulus applications for obtaining better responses. MEG responses to electrogustatory stimulation are eminently suitable for objective evaluations of taste acuity in clinical situations, and the present results should provide fundamental and useful data on which to base analyses of brain activities induced by electrogustatory stimulation.
|
Acknowledgments |
|---|
This work was supported by a Grant-in-Aid for scientific research (No. 14370593 to T.Y.) from the Japan Society for the Promotion of Science, The Salt Science Research Foundation (No. 0244) and The Mishimakaiun Research Foundation.
|
Notes |
|---|
February 25, 2003
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Barry, M.A., Gatenby, J.C., Zeiger, J.D. and Gore, J.C. (2001) Hemispheric dominance of cortical activity evoked by focal electrogustatory stimuli. Chem. Senses,26 , 471–782.
Bujas, Z. (1971) Electrical taste. In Beidler, L.M. (ed.), Handbook of Sensory Physiology. Vol.IV . Taste. Springer, New York, pp.180 –199.
Cerf, B., Lebihan, D., Van de Moortete, P.F., Macleod, P. and Faurion, A. (1998) Functional lateralization of human gustatory cortex related to handedness disclosed by fMRI study. Ann. N. Y. Acad. Sci.,855 ,575 –578.[CrossRef][Web of Science][Medline]
Fons, M. and Osterhammel, P.A. (1968) Electro-gustometry. Arch. Otolaryngol,83 , 538–542.
Frank, M.E. and Smith, D.V. (1991) Electrogustometry: a simple way to test taste. In Getchell, T.V., Doty, R.L., Bartoshuk, L.M. and Snow, J.B. (eds), Smell and Taste in Health and Disease. Raven Press, New York, pp.503 –514.
Genow, A., Hummel, T., Kroger, H., Bagla, R. and Bigelow, D.C. (1998) Gustatory event-related potentials in healthy controls and patients with hypoguesia or aguesia. Chem. Senses, 26,606 –607.
Imada, T., Mashiko, T., and Sekine, H. (2001) Estimation errors of single dipole model applied to twin dipole activeity: computer simulation study. In Nenonen, J., Ilmoninemi, R.J. and Katila, T. (eds), Biomag2000, Proceedings of the 12th International Conference on Biomagnetism. Helsinki University of Technology, Espoo, 2001, pp.733 –737.
Iwaki, S., Yamamoto, C., Takehara, S., Tonoike, M. and Yamamoto, T. (2003) Rejection of electro-gustatory stimulus-related MEG artifacts using independent component analysis.Abstract submitted to the 9th International Conference on Functional Mapping of the Human Brain , New York City, in press.
Kinomura, S., Kawashima, R., Yamada, K., Ono, S., Itoh, M., Yoshioka, S., Yamaguchi, T., Matsui, H., Miyazawa, H., Itoh, H., Goto, R., Fujiwara, T., Satoh, K. and Fukuda, H. (1994) Functional anatomy of taste perception in the human brain studied with positron emission tomography. Brain Res.,659 ,263 –266.[CrossRef][Web of Science][Medline]
Kobayakawa, T., Endo, H., Ayabe-Kanemura, S., Kumagai, T., Yamaguchi, Y., Kikuchi, Y., Takeda, T., Saito, S. and Ogawa, H. (1996) The primary gustatory area in human cerebral cortex studied by magnetoencephalography. Neurosci. Lett.,212 ,155 –158.[CrossRef][Web of Science][Medline]
Kobayakawa, T., Ogawa, H., Kaneda, H., Ayabe-Kanemura, S.,
Endo, H. and Saito, S. (1999) Spatio-temporal
analysis of cortical activity evoked by gustatory stimulation in humans.Chem. Senses
, 24,201
–209.
Krarup, B. (1958) Electro-gustometry: a method for clinical taste examinations. Acta Otolaryngol.,49 , 294–305.[Medline]
Makeig, S., Jung, T-P., Bell, A.J., Ghahremani, D. and
Sejnowski, T.J. (1997) Blind separation of
event-related brain responses into independent components. Proc.
Natl Acad. Sci. USA, 94,10979
–10984.
Murayama, N., Nakasato, N., Hatanaka, K., Fujita, S., Igasaki, T., Kanno, A. and Yoshimoto, T. (1996) Gustatory evoked magnetic fields in humans. Neurosci. Lett., 210,121 –123.[CrossRef][Web of Science][Medline]
Murphy, C., Quininez, C. and Nordin, S.
(1995) Reliability and validity of electrogustometry and its
application to young and elderly persons. Chem. Senses,20
, 499–503.
Nakazato, M., Endo, S., Yoshimura, I. and Tomita, H. (2002) Influence of aging on electrogustometry thresholds. Acta Otolaryngol., Suppl. 546,16 –26.
Pardo, J.V., Wood, T.D., Costello, P.A. Pardo, P.J. and Lee, J.T. (1997) PET study of the localization and laterality of lingual somatosensory processing in humans.Neurosci. Lett. , 234,23 –26.[CrossRef][Web of Science][Medline]
Plattig, K.-H. (1969) Uber den elektrischen geschmack. Z. Biol., 116,161 –211.[Medline]
Plattig, K.-H. (1971) Taste sensations and evoked brain potentials after electric stimulation of tongue in man. In Ohloff, G. and Thomas, A.F. (eds), Gustation and Olfaction. Academic Press, London, pp. 73–86.
Plattig, K.-H. and Kobal, G. (1977) Olfactory and gustatory responses in human electroencephalogram (EEG). In Katsuki, Y., Sato, M. Takagi, S.F. and Omura, Y. (eds),Food Intake and Chemical Senses. University of Tokyo Press, Tokyo, pp. 51–70.
Schaupp, H. (1971) Probleme der objektiven Gustometrie. Z. Laryngol. Otol.,50 , 208–213.
Small, D.M., Zald, D.H., Jones-Gotman, M., Zatorre, R.J., Pardo, J.V., Frey, S. and Petrides, M. (1999) Human cortical gustatory areas: a review of functional neuroimaging data. Neuroreport, 10,7 –14.[Web of Science][Medline]
Tomita, H. and Ikeda, M. (2002) Clinical use of electrogustometry: strengths and limitations.Acta Otolaryngol. , Suppl. 546,27 –38.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Onoda, T. Kobayakawa, M. Ikeda, S. Saito, and A. Kida Laterality of Human Primary Gustatory Cortex Studied by MEG Chem Senses, October 1, 2005; 30(8): 657 - 666. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. P. Halpern Temporal Characteristics of Human Taste Judgements as Calibrations for Gustatory Event-related Potentials and Gustatory Magnetoencephalographs Chem Senses, January 1, 2005; 30(suppl_1): i228 - i229. [Full Text] [PDF]
|
|
|
|
|
|
A. Faurion, T. Kobayakawa, and B. Cerf-Ducastel Cerebral Imaging in Taste Chem Senses, January 1, 2005; 30(suppl_1): i230 - i231. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||