- Email: [email protected]
The development of a novel automated taste stimulus delivery system for fMRI studies on the human cortical segregation of taste
Journal of Neuroscience Methods 172 (2008) 48–53
Contents lists available at ScienceDirect
Journal of Neuroscience Methods journal homepage: www.elsevier.com/locate/jneumeth
The development of a novel automated taste stimulus delivery system for fMRI studies on the human cortical segregation of taste Yukiko N. Kami a,1 , Tazuko K. Goto a,∗,1 , Kenji Tokumori a , Takashi Yoshiura b , Koji Kobayashi c , Yasuhiko Nakamura c , Hiroshi Honda b , Yuzo Ninomiya d , Kazunori Yoshiura a a
Department of Oral and Maxillofacial Radiology, Faculty of Dental Science, Kyushu University, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Department of Medical Technology, Kyushu University Hospital, Fukuoka, Japan d Section of Oral Neuroscience, Graduate School of Dental Science, Kyushu University, Fukuoka, Japan b c
a r t i c l e
i n f o
Article history: Received 2 February 2008 Received in revised form 1 April 2008 Accepted 6 April 2008 Keywords: fMRI Taste The primary taste cortex Delivery system Humans
a b s t r a c t fMRI indicated that the primary taste cortex is activated not only by taste but also by non-taste information from oral stimuli. Head movements caused by swallowing are very critical problem in fMRI and inherent difficulties to modulate taste stimuli in the mouth exist to elucidate functional segregation of human brain. We developed a novel automated taste stimulus delivery system for fMRI studies to segregate the pure taste area in the primary taste cortex in humans. As a novel intra-oral device, an elliptic cylinder was attached to an individual mouthpiece and then subject placed the tongue tip in it. Using a computercontrolled extra-oral device, the solutions ran through the intra-oral device in constant conditions. Three adult volunteers participated in the experimental session, alternately consisting of 30 pairs of taste stimuli (0.5 mol/l sucrose solution) and control (water) blocks. The typical findings of the three subjects revealed activation only in the primary taste cortex (P < 0.001), and none in the secondary taste cortex. This is the first system that delivers the taste stimuli automatically to a standardized area on the subject’s tongue under constant conditions, thus allowing us to successfully segregate the pure taste area in the primary taste cortex in humans. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The primary taste cortex represents the identity and intensity of taste, and the secondary taste cortex reflects the emotional aspect of taste. Functional magnetic resonance imaging (fMRI) studies have indicated that the taste stimuli activate the insula and frontal operculum (primary taste cortex), the orbitofrontal cortex (secondary taste cortex) (Frank et al., 2003; Guest et al., 2007; Kobayashi et al., 2004; Kringelbach et al., 2003; Nitschke et al., 2006; O’Doherty et al., 2001, 2002; Ogawa et al., 2005; Schoenfeld et al., 2004; Small et al., 1999, 2003, 2004; Smits et al., 2007). The primary taste cortex, however, represents not only taste but also the non-taste information of oral stimuli, such as odorants (Cerf-Ducastel and ¨ Murphy, 2001a; Osterbauer et al., 2006; Small et al., 2004), texture (De Araujo and Rolls, 2004), temperature (Guest et al., 2007), and lingual somatosensory stimuli (Cerf-Ducastel et al., 2001b). In the non-human primate, the experiments that studied the responses
∗ Corresponding author. Tel.: +81 92 642 6407; fax: +81 92 642 6410. E-mail address: [email protected] (T.K. Goto). 1 Both the authors contributed equally to this work. 0165-0270/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jneumeth.2008.04.009
of neurons in the primary taste cortex of the cynomolgus monkey showed that gustatory neurons comprise only 2.1–10.9% of the neurons in the insula and operculum. The remaining neurons responded during movement of the mouth or tongue (20.7%), or tactile stimulation of the tongue (9.6%). A few neurons responded in anticipation of stimulus delivery (1.7%), but the majority of neurons tested were unresponsive to the tested stimuli (56.6%) (Pritchard, 1991). Such functional segregation of the taste cortex is more complex in humans, but it is still difficult to make a more precise brain mapping of the taste. The fMRI is an appropriate tool for this functional segregation of the human brain because it is non-invasive, while also providing a high spatial resolution; however, there are some limitations in fMRI experiments for the taste. The subject lies supine within the MRI scanner and is not allowed to move his/her head because head movement caused by swallowing is a very critical problem in fMRI experiment; therefore, it is difficult to give a subject a taste stimulus and then be able to rinse it out. In most previous studies, small volumes of solutions (0.5–2.0 ml) have been delivered to the subject’s mouth through some tubes and the subject was cued by visual or tone signals to swallow the taste solution (De Araujo et al., 2003a,b; De Araujo and Rolls, 2004; Guest et al., 2007;
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
49
Haase et al., 2007; Kringelbach et al., 2003; Marciani et al., 2006; O’Doherty et al., 2002; Schoenfeld et al., 2004; Small et al., 2003, 2004). Kobayashi et al. (2004) developed a perfusion system, which was set on the tongue of the subject. The system consisted of inlet tubes that provide solutions surrounded by a thick outlet tube with holes for continuous suction. Using this system, the subjects do not have to swallow the solution or pay attention to the cues and do not sense odors during a session. Furthermore, this system improved the efficiency of washing out the perfused solutions. However, the possibility remains that some of the liquid could escape from the space between the outlet tube and the subject’s tongue, and this may lead to the subject’s irregular swallowing. Furthermore, this system still creates a tactile stimulus and the difficulty of holding the tube between the lips. To overcome those drawbacks, Ogawa et al. (2005) used a gustatory apparatus (Kobayakawa et al., 1996, 1999) from which the subject received a series of fluids on the dorsal surface of the tongue tip placed in a small hole (7 mm × 2.8 mm) in a Teflon tube through which water and taste solution were delivered. Using this apparatus, the swallowing and tactile stimulus is reduced and the subjects do not have to pay attention to the cues. However, the subjects have to hold the tubes between the lips, the stimulated area is limited to a small area on the dorsal surface of the tongue, and the flow rate of the stimuli is very fast for fMRI. We assume that the further improvements in the system would make it possible to segregate the functional area in the taste cortex. The aim of this study was to develop a novel automated taste stimulus delivery system for fMRI studies to segregate the taste area in the primary taste cortex in humans. 2. Materials and methods 2.1. Taste stimuli delivery system We developed a novel automated delivery system to administer the liquid stimuli, which consisted of an intra-oral and an extra-oral device. 2.1.1. Intra-oral device A rigid removable mouthpiece was fabricated for the mandibular dental arch of each subject by using an ethylene-vinyl acetate resin (round sheet; diameter, 120 mm; thickness, 2.0 mm; ERKOFLEX/ERKODENT® , Pfalzgrafenweiler, Germany) (Fig. 1). An elliptic cylinder (major axis: 15 mm, minor axis: 8 mm, height: 15 mm) made of the same material was attached on the incisor region of the mouthpiece. Its lingual side was open so that subjects could place their tongue tip in the elliptic cylinder at the resting position. Two tubes that provide solutions were then connected to one side of the elliptic cylinder and another tube for the outflow of the solution was connected to another side. The latter tube’s end was put into a waste container. This intra-oral device was placed on the subject while comfortably lying in the supine position in the MRI scanner. The taste solutions and deionized water run through the elliptic cylinder of the intra-oral device in which the subjects fit their tongue tip. Since the flow of the liquid generated a slightly negative pressure, the tongue of the subject was sucked slightly into the elliptic cylinder, and therefore, no liquid leaked into the mouth. 2.1.2. Extra-oral device The extra-oral device was placed outside the MRI scanner room (Fig. 2). The two plastic tubes of the intra-oral device for delivering solutions were connected to the extra-oral device. Compressed air from a compact air pump was controlled by regulators, and pressured taste solutions and deionized water was held in each stainless steel pressure container. The liquid flowed through tubes under pressure. The flow rate of the solutions was monitored
Fig. 1. The intra-oral device of the taste stimulus delivery system. It consisted of a rigid removable mouthpiece on a mandibular dental arch and an elliptic cylinder (15 mm × 8 mm × 15 mm). The elliptic cylinder was attached on the incisor region of the mouthpiece with its lingual side open. The subjects placed their tongue tip in the elliptic cylinder at the rest position. The taste solutions and deionized water from the clear plastic tubes connected to the extra-oral device came into one side of the elliptic cylinder, and after passing through in the elliptic cylinder, flowed out from another side to a waste container. Because of the slight negative pressure generated by the flow of the liquid, no solutions leaked into the subject’s mouth.
by flow meters and was set at 0.8 ml/s. The timing and period to deliver the taste solutions and deionized water in the tubes was switched by solenoid valves, which were controlled by an originally written program on a personal computer. 2.2. Subjects Three healthy female volunteers (age = 31, 34, and 44 years old) participated in this study. They had no history of either any neurological or psychiatric illness. The Human Experimentation Committee of Kyushu University approved all experimental procedures. Before the experiment, the participants were informed in detail about the nature of the experiment, and all gave their written informed consent to participate in the study. 2.3. Experimental design To determine the reproducibility of the novel system, three subjects participated in the experiment twice on two different days, 6 months apart. The same protocol was performed on each day. For the stimulus, 0.5 mol/l sucrose, a prototypical taste solution, was employed. Deionized water was used for the control and rinsing. Both solutions were kept at room temperature. We used a block design shown schematically in Fig. 3. A session consisted of 30 pairs of ON (taste solution) and OFF (deionized water) blocks alternately. The duration of the ON and OFF blocks was 8 and 16 s, respectively. A block-designed paradigm was used
50
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
Fig. 2. The extra-oral device of the taste stimulus delivery system. Compressed air from a compact air pump was gathered in a stainless steel pressure container, and then being controlled by regulators, the taste solution and deionized water were each stored in a stainless steel pressure container. The liquids flowed out through each tube toward the MRI scanner under pressure. The timing and duration to deliver the solutions was switched by solenoid valves, which were controlled by an originally written program on a personal computer. The flow rates of the solutions were monitored by flow meters.
to obtain the image of brain responses to a block of taste stimulation in contrast to those during the control period with water only.
0), detectable but very weak (1), weak (2), moderate (3), strong (4), to very strong (5). They were also asked whether or not they felt any odors or tactile sensations.
2.4. fMRI data acquisition The fMRI scan was performed using a 1.5 T scanner (Symphony; Siemens AG, Erlangen, Germany). A T2*-weighted gradient echo-planar imaging (EPI) sequence was employed (TR = 4000 s, TE = 50 ms, the field of view = 230 mm, matrix size = 64 × 64 pixels, voxel size = 3.6 mm × 3.6 mm × 3.0 mm, slice thickness = 3.0 mm). Each volume datum consisted of 32 slices to cover the whole brain, and a 180-volume data set was acquired in each scanning run (session). For anatomical reference, T1-weighted magnetization-prepared rapid gradient echo (MPRAGE) images (TR = 1900 ms, TE = 3.93 ms, Flip Angle = 15◦ , FOV = 230 mm, matrix size = 256 × 256 pixels, slice thickness = 1.0 mm) were obtained for each subject with the identical location parameters of EPI. After the session, the subjects were asked about the quality of the taste solutions and the perceived intensity of the sweetness on an ordinal scale of six levels, from not detectable (level
Fig. 3. The fMRI paradigm. Block design consisted of 30 pairs of ON (0.5 mol/l sucrose solutions as taste solution) and OFF (deionized water as control and rinse) durations was employed. The durations of the ON and OFF blocks were 8 and 16 s, respectively.
2.5. Data analysis The data were analyzed using statistical parametric mapping (SPM5; Wellcome Department of Imaging Neuroscience, University of London, UK). The first three volumes of each session of functional images were discarded due to unsteady magnetization, and the remaining 177 volumes per session were used for the analysis. Slight head motion within each session was corrected with the realignment program, and all subsequent scans were realigned to the first scan. In this procedure, the head motion of each session was calculated in six degrees of freedom (translation and rotation) in time-series for the entire session. The fMRI data were then spatially normalized into the standard space defined by a template image, that is to say, the MNI coordinate system (Montreal Neurological Institute), and smoothed with an 8 mm full width at half maximum isotropic Gaussian kernel and global scaling. A General Linear Model was used to identify the voxels with stimulus-related signal changes. Significant correlations between the observed response and the modeled response were estimated, thus yielding t-value maps. The level of an uncorrected P-value of <0.001 was considered to be significant. At the group level to test, conjunction analyses (Friston et al., 1999a,b) were employed to investigate the typical characteristics of the three subjects for each day. The activations on the primary taste cortex (the insula and the frontal operculum) and those of the secondary taste cortex (the orbitofrontal cortex) were investigated and anatomically
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
51
Fig. 5. A result of head movement through a session. The head movement of this subject was less than 0.6 mm in translation and less than 0.7◦ in rotation.
4. Discussion Fig. 4. A diagram of the anatomical location of the insula and frontal operculum. (䊉) Insula, () the frontal operculum. The anatomical drawing was obtained from the Atlas of the HUMAN BRAIN, Elsevier Academic Press.
identified according to a brain atlas (Mai et al., 2004) (Fig. 4). The MNI coordinates of the voxel that showed a maximum increased activity due to the stimulus versus the control condition in each interested cluster were further detected.
3. Results All subjects (Sub A, Sub B, Sub C) were able to detect and rate the taste (sweetness) through the intra-oral device. Sub A felt a moderate degree of the sweetness (level 3) on both days, Sub B felt a strong degree (level 4) on both days, and Sub C felt a strong degree (level 4) on the first day and moderate (level 3) on the second day. None of the subjects noted any evoked odor or tactile sensations. Throughout all sessions, the intra-oral device was fixed to the dental arch of each subject; therefore, the subjects had to do nothing except for tasting. A result of the head movement through a session was shown with the output of the SPM realign function that estimated the translation and rotation of each image (Fig. 5). The head motions of all subjects were less than 1 mm in translation and less than 1.5◦ in rotation in all sessions. On both days of testing, three subjects mainly revealed rCBF (regional cerebral blood flow) in the insula and posterior part of the frontal operculum. The activated areas by sweet taste had peaks at [40, −6, 10] on the first day, and [40, 0, 10] on the second day, in the MNI coordinates. These activation cortical areas are consistent with earlier studies (Ogawa et al., 2005), and therefore, will be the primary taste cortex. No activation was observed in the orbitofrontal cortex (Table 1). Fig. 6 shows the major activated areas on the hemispheres.
This is the first system that totally automated for fMRI and delivered a taste stimulus. It delivered the clear and constant taste stimuli to the tongue, and eliminated the non-taste stimuli. This allowed us to successfully segregate the pure taste area in the primary taste cortices in humans. The results obtained on the two different days were the same, thus demonstrating this system to be reproducible and useful for examining the activations in the primary taste cortices in humans. The significant point in this study was the development of a novel intra-oral device. First, it administered the taste solutions into a standardized area on the subject’s tongue. The area of the tongue where the taste solution was received was made as large as possible by using an elliptic cylinder. This made it possible to acquire the stronger sweet stimulus on the tongue tip area, although the intra-oral device that was developed by Kobayakawa et al. (1996) provides the taste solution by a small hole in the tube on the dorsal surface of the tongue tip. This is more effective for fMRI experiment in measuring higher level of function for taste, such as memory, identification or recognition with adding such tasks. In addition, our device enabled sufficient rinsing of the tongue. Second, the intra-oral device was fixed on the mandibular dental arch even in the wet oral environment. This provided the benefit that the sub-
Table 1 Region of activations Day
First Second
Primary taste cortex
Secondary taste cortex
Insula
Frontal operculum
Orbitofrontal cortex
o o
o o
– – (P < 0.001 uncorrected)
First, the first day of the experiment. Second, the second day of the experiment. The typical characteristics of the three subjects were analyzed by conjunction analyses, a level of P < 0.001 was considered to indicate a significant difference. o, the activation occurrence; –, no activation.
52
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
Fig. 6. Brain activation: typical characteristics of three subjects. (A) Activated area in three dimensions: (a) sagittal, (b) coronal, (c) axial sections. (B) The magnified image of the activated cluster in coronal section: (a) the activation on the insula, (b) on posterior part of the frontal operculum. The map is shown with an uncorrected P-value of <0.001 and overlaid on the SPM5 MNI template.
jects were able to concentrate on ‘the taste’. The subjects were thus freed from the stress of holding the tubes between their lips, and the taste solution did not flow out from this intra-oral device. Third, odor stimulus could not occur; therefore, brain activation by the non-taste information of oral stimuli, such as odorants in the pri¨ mary taste cortex (Cerf-Ducastel and Murphy, 2001a; Osterbauer et al., 2006; Small et al., 2004) could be eliminated in our results. Fourth, the advantage of this intra-oral device was that it removed the tactile stimulus of the tongue. The primary taste cortex represents not only taste, but also somatosensory stimuli in the human (Cerf-Ducastel et al., 2001b) and tactile stimulation of the mouth ´ 1999). In addition, region in the monkey (Scott and Plata-Salaman, this design prevented the tactile stimuli on the oral mucosa or surface skin of the face. Fifth, the subjects did not swallow the taste solution. This prevented the head movement during swallowing. The head movement was less than 1.0 mm in any direction or 1.5◦ by rotation through all experimental sessions in our study. Head movement in fMRI causes movement artifacts. The movement itself can be out of the alignment of the images. In addition, even after realignment during the analysis by SPM5, there is residual movement related variance present in the fMRI time-series, causing a loss of sensitivity and, potentially, also specificity (Andersson et al., 2001). It is suggested that it is best to avoid the head movement itself. If the subjects swallow the taste solutions and create head movement, we must know the timing of the swallowing in order to eliminate the data obtained during swallowing. In our study, however, we could simplify the experimental design and not use visual or tone signals. The subjects were lying down and relaxed, and used an eye-mask so that they could concentrate on tasting. Because of this simple process, they did not have to be trained before the session and also did not feel fatigue either during or after the session. The second development of our system was that it provided the taste stimulus under a constant condition. The extra-oral device was the system that delivered the solutions from the containers to the intra-oral device. It was set up in the MRI operation room (outside the MRI scanner room) and then it was controlled automatically by an originally written program on a personal computer. In fMRI studies, the correlation between the MRI signal time course and the stimulus paradigm is very important (Van de Moortele et al., 1997). The manual delivery of taste solutions to the subjects is
usually done by an experimenter who stays beside the MRI scanner. In comparison to delivering stimuli manually, an automatic computer-controlled system ensures accurate timing between the application of the stimulus and the acquisition period by the MRI scanner. The flow-rate of the taste stimuli was kept constant at 0.8 ml/s. We considered three points in setting the flow rate of the liquid. Those were (1) the subject did not sense the tactile stimulus by the flowing liquid into the elliptic cylinder, (2) the subject clearly sensed every change of the control to taste stimuli, (3) the “wash out” of the taste stimuli was sufficient. This extra-oral delivery system allowed us precise control over stimulus delivery which thus ensured the reproducibility of the paradigm. The automatic extra-oral devices have been used in previous studies (Frank et al., 2003; Haase et al., 2007; Marciani et al., 2006; Small et al., 2004; Wagner et al., 2008). However, the intra-oral devices in their studies were different from ours, and they usually used tubes to spray or push out the taste solution onto the tongue. Thereafter, the subjects sensed the taste along with the tactile stimuli and odor, and then swallowed the liquid during the experimental session according to a visual or tone signal. The combination of our novel intra-oral device and extra-oral device allowed us to establish a new totally automated delivery system, as well as to remove the non-taste stimuli and also reduce the level of subject strain. In conclusion, this is the first system that delivers taste stimuli to a standardized area on the subject’s tongue, and the subjects were free from swallowing, tactile stimulus, odor, paying attention to the visual or tone signals, or having to hold the tubes by themselves. This system delivered the solutions under constant conditions and allowed us to successfully segregate the pure taste area in the primary taste cortices in humans. This device will, therefore, make it possible to perform more precise functional mapping with segregation of the taste cortex in the future studies. Acknowledgements We are grateful to Dr. Osamu Takizawa, Chief scientist, Siemens Asahi Meditech Co Ltd., Tokyo, Japan for his support in the MR examinations. We also thank Prof. Brian Quinn for critically reading the English manuscript. This study was supported by a Grant-in-Aid
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
for Scientific Research from the Ministry of Education, Japan (No. 19390479) (T.K. G.) (No. 18109013 and 18077004) (Y. N.) and also by the Kyushu University Foundation (T.K. G.). References Andersson JL, Hutton C, Ashburner J, Turner R, Friston K. Modeling geometric deformations in EPI time series. Neuroimage 2001;13:903–19. Cerf-Ducastel B, Murphy C. fMRI activation in response to odorants orally delivered in aqueous solutions. Chem Senses 2001a;26:625–37. Cerf-Ducastel B, Van de Moortele PF, MacLeod P, Le Bihan D, Faurion A. Interaction of gustatory and lingual somatosensory perceptions at the cortical level in the human: a functional magnetic resonance imaging study. Chem Senses 2001b;26:371–83. De Araujo IE, Kringelbach ML, Rolls ET, Hobden P. Representation of umami taste in the human brain. J Neurophysiol 2003a;90:313–9. De Araujo IE, Kringelbach ML, Rolls ET, McGlone F. Human cortical responses to water in the mouth, and the effects of thirst. J Neurophysiol 2003b;90:1865–76. De Araujo IE, Rolls ET. Representation in the human brain of food texture and oral fat. J Neurosci 2004;24:3086–93. Frank GK, Kaye WH, Carter CS, Brooks S, May C, Fissell K, et al. The evaluation of brain activity in response to taste stimuli-a pilot study and method for central taste activation as assessed by event-related fMRI. J Neurosci Meth 2003;131: 99–105. Friston KJ, Holmes AP, Worsley KJ. How many subjects constitute a study? Neuroimage 1999a;10:1–5. ¨ Friston KJ, Holmes AP, Price CJ, Buchel C, Worsley KJ. Multisubject fMRI studies and conjunction analyses. Neuroimage 1999b;10:385–96. Guest S, Grabenhorst F, Essick G, Chen Y, Young M, McGlone F, et al. Human cortical representation of oral temperature. Physiol Behav 2007;92:975–84. Haase L, Cerf-Ducastel B, Buracas G, Murphy C. On-line psychophysical data acquisition and event-related fMRI protocol optimized for the investigation of brain activation in response to gustatory stimuli. J Neurosci Meth 2007;159: 98–107. Kobayakawa T, Endo H, Saito S, Ayabe-Kanamura S, Kikuchi Y, Yamaguchi Y, et al. Trial measurements of gustatory-evoked magnetic fields. Electroenceph Clin Neurophysiol Suppl 1996;47:133–41. Kobayakawa T, Ogawa H, Kaneda H, Ayabe-Kanamura S, Endo H, Saito S. Spatiotemporal analysis of cortical activity evoked by gustatory stimulation in humans. Chem Senses 1999;24:201–9. Kobayashi M, Takeda M, Hattori N, Fukunaga M, Sasabe T, Inoue N, et al. Functional imaging of gustatory perception and imagery: “top-down” processing of gustatory signals. Neuroimage 2004;23:1271–82.
53
Kringelbach ML, O’Doherty J, Rolls ET, Andrews C. Activation of the human orbitofrontal cortex to a liquid food stimulus is correlated with its subjective pleasantness. Cereb Cortex 2003;13:1064–71. Mai JK, Assheuer J, Paxinos G. Atlas of the human brain. Second ed. Amsterdam: Elsevier; 2004. Marciani L, Pfeiffer JC, Hort J, Head K, Bush D, Taylor AJ, et al. Improved methods for fMRI studies of combined taste and aroma stimuli. J Neurosci Meth 2006;158:186–94. Nitschke JB, Dixon GE, Sarinopoulos I, Short SJ, Cohen JD, Smith EE, et al. Altering expectancy dampens neural response to aversive taste in primary taste cortex. Nat Neurosci 2006;9:435–42. O’Doherty J, Rolls ET, Francis S, Bowtell R, McGlone F. Representation of pleasant and aversive taste in the human brain. J Neurophysiol 2001;85:1315–21. O’Doherty JP, Deichmann R, Critchley HD, Dolan RJ. Neural responses during anticipation of a primary taste reward. Neuron 2002;33:815–26. Ogawa H, Wakita M, Hasegawa K, Kobayakawa T, Sakai N, Hirai T, et al. Functional MRI detection of activation in the primary gustatory cortices in humans. Chem Senses 2005;30:583–92. ¨ Osterbauer RA, Wilson JL, Calvert GA, Jezzard P. Physical and physiological consequences of passive intra-oral shimming. Neuroimage 2006;29:245–53. Pritchard TC. The primate gustatory system. In: Getchell TV, et al., editors. Smell and taste in health and disease. New York: Raven Press; 1991. p. 109–25. ¨ Schoenfeld MA, Neuer G, Tempelmann C, Schußler K, Noesselt T, Hopf JM, et al. Functional magnetic resonance tomography correlates of taste perception in the human primary taste cortex. Neuroscience 2004;127:347–53. ´ Scott TR, Plata-Salaman CR. Taste in the monkey cortex. Physiol Behav 1999;67:489–511. Small DM, Zald DH, Jones-Gotman M, Zatorre RJ, Pardo JV, Frey S, et al. Human cortical gustatory areas: a review of functional neuroimaging data. Neuroreport 1999;10:7–14. Small DM, Gregory MD, Mak YE, Gitelman D, Mesulam MM, Parrish T. Dissociation of neural representation of intensity and affective valuation in human gustation. Neuron 2003;39:701–11. Small DM, Voss J, Mak YE, Simmons KB, Parrish T, Gitelman D. Experience-dependent neural integration of taste and smell in the human brain. J Neurophysiol 2004;92:1892–903. Smits M, Peeters RR, van Hecke P, Sunaert S. A 3 T event-related functional magnetic resonance imaging (fMRI) study of primary and secondary gustatory cortex localization using natural tastants. Neuroradiology 2007;49:61–71. Van de Moortele PF, Cerf B, Lobel E, Paradis AL, Faurion A, Le Bihan D. Latencies in fMRI time-series: effect of slice acquisition order and perception. NMR Biomed 1997;10:230–6. Wagner A, Aizenstein H, Mazurkewicz L, Fudge J, Frank GK, Putnam K, et al. Altered insula response to taste stimuli in individuals recovered from restricting-type anorexia nervosa. Neuropsychopharmacology 2008;33:513–23.
Contents lists available at ScienceDirect
Journal of Neuroscience Methods journal homepage: www.elsevier.com/locate/jneumeth
The development of a novel automated taste stimulus delivery system for fMRI studies on the human cortical segregation of taste Yukiko N. Kami a,1 , Tazuko K. Goto a,∗,1 , Kenji Tokumori a , Takashi Yoshiura b , Koji Kobayashi c , Yasuhiko Nakamura c , Hiroshi Honda b , Yuzo Ninomiya d , Kazunori Yoshiura a a
Department of Oral and Maxillofacial Radiology, Faculty of Dental Science, Kyushu University, 3-1-1, Maidashi, Higashi-ku, Fukuoka 812-8582, Japan Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Department of Medical Technology, Kyushu University Hospital, Fukuoka, Japan d Section of Oral Neuroscience, Graduate School of Dental Science, Kyushu University, Fukuoka, Japan b c
a r t i c l e
i n f o
Article history: Received 2 February 2008 Received in revised form 1 April 2008 Accepted 6 April 2008 Keywords: fMRI Taste The primary taste cortex Delivery system Humans
a b s t r a c t fMRI indicated that the primary taste cortex is activated not only by taste but also by non-taste information from oral stimuli. Head movements caused by swallowing are very critical problem in fMRI and inherent difficulties to modulate taste stimuli in the mouth exist to elucidate functional segregation of human brain. We developed a novel automated taste stimulus delivery system for fMRI studies to segregate the pure taste area in the primary taste cortex in humans. As a novel intra-oral device, an elliptic cylinder was attached to an individual mouthpiece and then subject placed the tongue tip in it. Using a computercontrolled extra-oral device, the solutions ran through the intra-oral device in constant conditions. Three adult volunteers participated in the experimental session, alternately consisting of 30 pairs of taste stimuli (0.5 mol/l sucrose solution) and control (water) blocks. The typical findings of the three subjects revealed activation only in the primary taste cortex (P < 0.001), and none in the secondary taste cortex. This is the first system that delivers the taste stimuli automatically to a standardized area on the subject’s tongue under constant conditions, thus allowing us to successfully segregate the pure taste area in the primary taste cortex in humans. © 2008 Elsevier B.V. All rights reserved.
1. Introduction The primary taste cortex represents the identity and intensity of taste, and the secondary taste cortex reflects the emotional aspect of taste. Functional magnetic resonance imaging (fMRI) studies have indicated that the taste stimuli activate the insula and frontal operculum (primary taste cortex), the orbitofrontal cortex (secondary taste cortex) (Frank et al., 2003; Guest et al., 2007; Kobayashi et al., 2004; Kringelbach et al., 2003; Nitschke et al., 2006; O’Doherty et al., 2001, 2002; Ogawa et al., 2005; Schoenfeld et al., 2004; Small et al., 1999, 2003, 2004; Smits et al., 2007). The primary taste cortex, however, represents not only taste but also the non-taste information of oral stimuli, such as odorants (Cerf-Ducastel and ¨ Murphy, 2001a; Osterbauer et al., 2006; Small et al., 2004), texture (De Araujo and Rolls, 2004), temperature (Guest et al., 2007), and lingual somatosensory stimuli (Cerf-Ducastel et al., 2001b). In the non-human primate, the experiments that studied the responses
∗ Corresponding author. Tel.: +81 92 642 6407; fax: +81 92 642 6410. E-mail address: [email protected] (T.K. Goto). 1 Both the authors contributed equally to this work. 0165-0270/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jneumeth.2008.04.009
of neurons in the primary taste cortex of the cynomolgus monkey showed that gustatory neurons comprise only 2.1–10.9% of the neurons in the insula and operculum. The remaining neurons responded during movement of the mouth or tongue (20.7%), or tactile stimulation of the tongue (9.6%). A few neurons responded in anticipation of stimulus delivery (1.7%), but the majority of neurons tested were unresponsive to the tested stimuli (56.6%) (Pritchard, 1991). Such functional segregation of the taste cortex is more complex in humans, but it is still difficult to make a more precise brain mapping of the taste. The fMRI is an appropriate tool for this functional segregation of the human brain because it is non-invasive, while also providing a high spatial resolution; however, there are some limitations in fMRI experiments for the taste. The subject lies supine within the MRI scanner and is not allowed to move his/her head because head movement caused by swallowing is a very critical problem in fMRI experiment; therefore, it is difficult to give a subject a taste stimulus and then be able to rinse it out. In most previous studies, small volumes of solutions (0.5–2.0 ml) have been delivered to the subject’s mouth through some tubes and the subject was cued by visual or tone signals to swallow the taste solution (De Araujo et al., 2003a,b; De Araujo and Rolls, 2004; Guest et al., 2007;
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
49
Haase et al., 2007; Kringelbach et al., 2003; Marciani et al., 2006; O’Doherty et al., 2002; Schoenfeld et al., 2004; Small et al., 2003, 2004). Kobayashi et al. (2004) developed a perfusion system, which was set on the tongue of the subject. The system consisted of inlet tubes that provide solutions surrounded by a thick outlet tube with holes for continuous suction. Using this system, the subjects do not have to swallow the solution or pay attention to the cues and do not sense odors during a session. Furthermore, this system improved the efficiency of washing out the perfused solutions. However, the possibility remains that some of the liquid could escape from the space between the outlet tube and the subject’s tongue, and this may lead to the subject’s irregular swallowing. Furthermore, this system still creates a tactile stimulus and the difficulty of holding the tube between the lips. To overcome those drawbacks, Ogawa et al. (2005) used a gustatory apparatus (Kobayakawa et al., 1996, 1999) from which the subject received a series of fluids on the dorsal surface of the tongue tip placed in a small hole (7 mm × 2.8 mm) in a Teflon tube through which water and taste solution were delivered. Using this apparatus, the swallowing and tactile stimulus is reduced and the subjects do not have to pay attention to the cues. However, the subjects have to hold the tubes between the lips, the stimulated area is limited to a small area on the dorsal surface of the tongue, and the flow rate of the stimuli is very fast for fMRI. We assume that the further improvements in the system would make it possible to segregate the functional area in the taste cortex. The aim of this study was to develop a novel automated taste stimulus delivery system for fMRI studies to segregate the taste area in the primary taste cortex in humans. 2. Materials and methods 2.1. Taste stimuli delivery system We developed a novel automated delivery system to administer the liquid stimuli, which consisted of an intra-oral and an extra-oral device. 2.1.1. Intra-oral device A rigid removable mouthpiece was fabricated for the mandibular dental arch of each subject by using an ethylene-vinyl acetate resin (round sheet; diameter, 120 mm; thickness, 2.0 mm; ERKOFLEX/ERKODENT® , Pfalzgrafenweiler, Germany) (Fig. 1). An elliptic cylinder (major axis: 15 mm, minor axis: 8 mm, height: 15 mm) made of the same material was attached on the incisor region of the mouthpiece. Its lingual side was open so that subjects could place their tongue tip in the elliptic cylinder at the resting position. Two tubes that provide solutions were then connected to one side of the elliptic cylinder and another tube for the outflow of the solution was connected to another side. The latter tube’s end was put into a waste container. This intra-oral device was placed on the subject while comfortably lying in the supine position in the MRI scanner. The taste solutions and deionized water run through the elliptic cylinder of the intra-oral device in which the subjects fit their tongue tip. Since the flow of the liquid generated a slightly negative pressure, the tongue of the subject was sucked slightly into the elliptic cylinder, and therefore, no liquid leaked into the mouth. 2.1.2. Extra-oral device The extra-oral device was placed outside the MRI scanner room (Fig. 2). The two plastic tubes of the intra-oral device for delivering solutions were connected to the extra-oral device. Compressed air from a compact air pump was controlled by regulators, and pressured taste solutions and deionized water was held in each stainless steel pressure container. The liquid flowed through tubes under pressure. The flow rate of the solutions was monitored
Fig. 1. The intra-oral device of the taste stimulus delivery system. It consisted of a rigid removable mouthpiece on a mandibular dental arch and an elliptic cylinder (15 mm × 8 mm × 15 mm). The elliptic cylinder was attached on the incisor region of the mouthpiece with its lingual side open. The subjects placed their tongue tip in the elliptic cylinder at the rest position. The taste solutions and deionized water from the clear plastic tubes connected to the extra-oral device came into one side of the elliptic cylinder, and after passing through in the elliptic cylinder, flowed out from another side to a waste container. Because of the slight negative pressure generated by the flow of the liquid, no solutions leaked into the subject’s mouth.
by flow meters and was set at 0.8 ml/s. The timing and period to deliver the taste solutions and deionized water in the tubes was switched by solenoid valves, which were controlled by an originally written program on a personal computer. 2.2. Subjects Three healthy female volunteers (age = 31, 34, and 44 years old) participated in this study. They had no history of either any neurological or psychiatric illness. The Human Experimentation Committee of Kyushu University approved all experimental procedures. Before the experiment, the participants were informed in detail about the nature of the experiment, and all gave their written informed consent to participate in the study. 2.3. Experimental design To determine the reproducibility of the novel system, three subjects participated in the experiment twice on two different days, 6 months apart. The same protocol was performed on each day. For the stimulus, 0.5 mol/l sucrose, a prototypical taste solution, was employed. Deionized water was used for the control and rinsing. Both solutions were kept at room temperature. We used a block design shown schematically in Fig. 3. A session consisted of 30 pairs of ON (taste solution) and OFF (deionized water) blocks alternately. The duration of the ON and OFF blocks was 8 and 16 s, respectively. A block-designed paradigm was used
50
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
Fig. 2. The extra-oral device of the taste stimulus delivery system. Compressed air from a compact air pump was gathered in a stainless steel pressure container, and then being controlled by regulators, the taste solution and deionized water were each stored in a stainless steel pressure container. The liquids flowed out through each tube toward the MRI scanner under pressure. The timing and duration to deliver the solutions was switched by solenoid valves, which were controlled by an originally written program on a personal computer. The flow rates of the solutions were monitored by flow meters.
to obtain the image of brain responses to a block of taste stimulation in contrast to those during the control period with water only.
0), detectable but very weak (1), weak (2), moderate (3), strong (4), to very strong (5). They were also asked whether or not they felt any odors or tactile sensations.
2.4. fMRI data acquisition The fMRI scan was performed using a 1.5 T scanner (Symphony; Siemens AG, Erlangen, Germany). A T2*-weighted gradient echo-planar imaging (EPI) sequence was employed (TR = 4000 s, TE = 50 ms, the field of view = 230 mm, matrix size = 64 × 64 pixels, voxel size = 3.6 mm × 3.6 mm × 3.0 mm, slice thickness = 3.0 mm). Each volume datum consisted of 32 slices to cover the whole brain, and a 180-volume data set was acquired in each scanning run (session). For anatomical reference, T1-weighted magnetization-prepared rapid gradient echo (MPRAGE) images (TR = 1900 ms, TE = 3.93 ms, Flip Angle = 15◦ , FOV = 230 mm, matrix size = 256 × 256 pixels, slice thickness = 1.0 mm) were obtained for each subject with the identical location parameters of EPI. After the session, the subjects were asked about the quality of the taste solutions and the perceived intensity of the sweetness on an ordinal scale of six levels, from not detectable (level
Fig. 3. The fMRI paradigm. Block design consisted of 30 pairs of ON (0.5 mol/l sucrose solutions as taste solution) and OFF (deionized water as control and rinse) durations was employed. The durations of the ON and OFF blocks were 8 and 16 s, respectively.
2.5. Data analysis The data were analyzed using statistical parametric mapping (SPM5; Wellcome Department of Imaging Neuroscience, University of London, UK). The first three volumes of each session of functional images were discarded due to unsteady magnetization, and the remaining 177 volumes per session were used for the analysis. Slight head motion within each session was corrected with the realignment program, and all subsequent scans were realigned to the first scan. In this procedure, the head motion of each session was calculated in six degrees of freedom (translation and rotation) in time-series for the entire session. The fMRI data were then spatially normalized into the standard space defined by a template image, that is to say, the MNI coordinate system (Montreal Neurological Institute), and smoothed with an 8 mm full width at half maximum isotropic Gaussian kernel and global scaling. A General Linear Model was used to identify the voxels with stimulus-related signal changes. Significant correlations between the observed response and the modeled response were estimated, thus yielding t-value maps. The level of an uncorrected P-value of <0.001 was considered to be significant. At the group level to test, conjunction analyses (Friston et al., 1999a,b) were employed to investigate the typical characteristics of the three subjects for each day. The activations on the primary taste cortex (the insula and the frontal operculum) and those of the secondary taste cortex (the orbitofrontal cortex) were investigated and anatomically
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
51
Fig. 5. A result of head movement through a session. The head movement of this subject was less than 0.6 mm in translation and less than 0.7◦ in rotation.
4. Discussion Fig. 4. A diagram of the anatomical location of the insula and frontal operculum. (䊉) Insula, () the frontal operculum. The anatomical drawing was obtained from the Atlas of the HUMAN BRAIN, Elsevier Academic Press.
identified according to a brain atlas (Mai et al., 2004) (Fig. 4). The MNI coordinates of the voxel that showed a maximum increased activity due to the stimulus versus the control condition in each interested cluster were further detected.
3. Results All subjects (Sub A, Sub B, Sub C) were able to detect and rate the taste (sweetness) through the intra-oral device. Sub A felt a moderate degree of the sweetness (level 3) on both days, Sub B felt a strong degree (level 4) on both days, and Sub C felt a strong degree (level 4) on the first day and moderate (level 3) on the second day. None of the subjects noted any evoked odor or tactile sensations. Throughout all sessions, the intra-oral device was fixed to the dental arch of each subject; therefore, the subjects had to do nothing except for tasting. A result of the head movement through a session was shown with the output of the SPM realign function that estimated the translation and rotation of each image (Fig. 5). The head motions of all subjects were less than 1 mm in translation and less than 1.5◦ in rotation in all sessions. On both days of testing, three subjects mainly revealed rCBF (regional cerebral blood flow) in the insula and posterior part of the frontal operculum. The activated areas by sweet taste had peaks at [40, −6, 10] on the first day, and [40, 0, 10] on the second day, in the MNI coordinates. These activation cortical areas are consistent with earlier studies (Ogawa et al., 2005), and therefore, will be the primary taste cortex. No activation was observed in the orbitofrontal cortex (Table 1). Fig. 6 shows the major activated areas on the hemispheres.
This is the first system that totally automated for fMRI and delivered a taste stimulus. It delivered the clear and constant taste stimuli to the tongue, and eliminated the non-taste stimuli. This allowed us to successfully segregate the pure taste area in the primary taste cortices in humans. The results obtained on the two different days were the same, thus demonstrating this system to be reproducible and useful for examining the activations in the primary taste cortices in humans. The significant point in this study was the development of a novel intra-oral device. First, it administered the taste solutions into a standardized area on the subject’s tongue. The area of the tongue where the taste solution was received was made as large as possible by using an elliptic cylinder. This made it possible to acquire the stronger sweet stimulus on the tongue tip area, although the intra-oral device that was developed by Kobayakawa et al. (1996) provides the taste solution by a small hole in the tube on the dorsal surface of the tongue tip. This is more effective for fMRI experiment in measuring higher level of function for taste, such as memory, identification or recognition with adding such tasks. In addition, our device enabled sufficient rinsing of the tongue. Second, the intra-oral device was fixed on the mandibular dental arch even in the wet oral environment. This provided the benefit that the sub-
Table 1 Region of activations Day
First Second
Primary taste cortex
Secondary taste cortex
Insula
Frontal operculum
Orbitofrontal cortex
o o
o o
– – (P < 0.001 uncorrected)
First, the first day of the experiment. Second, the second day of the experiment. The typical characteristics of the three subjects were analyzed by conjunction analyses, a level of P < 0.001 was considered to indicate a significant difference. o, the activation occurrence; –, no activation.
52
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
Fig. 6. Brain activation: typical characteristics of three subjects. (A) Activated area in three dimensions: (a) sagittal, (b) coronal, (c) axial sections. (B) The magnified image of the activated cluster in coronal section: (a) the activation on the insula, (b) on posterior part of the frontal operculum. The map is shown with an uncorrected P-value of <0.001 and overlaid on the SPM5 MNI template.
jects were able to concentrate on ‘the taste’. The subjects were thus freed from the stress of holding the tubes between their lips, and the taste solution did not flow out from this intra-oral device. Third, odor stimulus could not occur; therefore, brain activation by the non-taste information of oral stimuli, such as odorants in the pri¨ mary taste cortex (Cerf-Ducastel and Murphy, 2001a; Osterbauer et al., 2006; Small et al., 2004) could be eliminated in our results. Fourth, the advantage of this intra-oral device was that it removed the tactile stimulus of the tongue. The primary taste cortex represents not only taste, but also somatosensory stimuli in the human (Cerf-Ducastel et al., 2001b) and tactile stimulation of the mouth ´ 1999). In addition, region in the monkey (Scott and Plata-Salaman, this design prevented the tactile stimuli on the oral mucosa or surface skin of the face. Fifth, the subjects did not swallow the taste solution. This prevented the head movement during swallowing. The head movement was less than 1.0 mm in any direction or 1.5◦ by rotation through all experimental sessions in our study. Head movement in fMRI causes movement artifacts. The movement itself can be out of the alignment of the images. In addition, even after realignment during the analysis by SPM5, there is residual movement related variance present in the fMRI time-series, causing a loss of sensitivity and, potentially, also specificity (Andersson et al., 2001). It is suggested that it is best to avoid the head movement itself. If the subjects swallow the taste solutions and create head movement, we must know the timing of the swallowing in order to eliminate the data obtained during swallowing. In our study, however, we could simplify the experimental design and not use visual or tone signals. The subjects were lying down and relaxed, and used an eye-mask so that they could concentrate on tasting. Because of this simple process, they did not have to be trained before the session and also did not feel fatigue either during or after the session. The second development of our system was that it provided the taste stimulus under a constant condition. The extra-oral device was the system that delivered the solutions from the containers to the intra-oral device. It was set up in the MRI operation room (outside the MRI scanner room) and then it was controlled automatically by an originally written program on a personal computer. In fMRI studies, the correlation between the MRI signal time course and the stimulus paradigm is very important (Van de Moortele et al., 1997). The manual delivery of taste solutions to the subjects is
usually done by an experimenter who stays beside the MRI scanner. In comparison to delivering stimuli manually, an automatic computer-controlled system ensures accurate timing between the application of the stimulus and the acquisition period by the MRI scanner. The flow-rate of the taste stimuli was kept constant at 0.8 ml/s. We considered three points in setting the flow rate of the liquid. Those were (1) the subject did not sense the tactile stimulus by the flowing liquid into the elliptic cylinder, (2) the subject clearly sensed every change of the control to taste stimuli, (3) the “wash out” of the taste stimuli was sufficient. This extra-oral delivery system allowed us precise control over stimulus delivery which thus ensured the reproducibility of the paradigm. The automatic extra-oral devices have been used in previous studies (Frank et al., 2003; Haase et al., 2007; Marciani et al., 2006; Small et al., 2004; Wagner et al., 2008). However, the intra-oral devices in their studies were different from ours, and they usually used tubes to spray or push out the taste solution onto the tongue. Thereafter, the subjects sensed the taste along with the tactile stimuli and odor, and then swallowed the liquid during the experimental session according to a visual or tone signal. The combination of our novel intra-oral device and extra-oral device allowed us to establish a new totally automated delivery system, as well as to remove the non-taste stimuli and also reduce the level of subject strain. In conclusion, this is the first system that delivers taste stimuli to a standardized area on the subject’s tongue, and the subjects were free from swallowing, tactile stimulus, odor, paying attention to the visual or tone signals, or having to hold the tubes by themselves. This system delivered the solutions under constant conditions and allowed us to successfully segregate the pure taste area in the primary taste cortices in humans. This device will, therefore, make it possible to perform more precise functional mapping with segregation of the taste cortex in the future studies. Acknowledgements We are grateful to Dr. Osamu Takizawa, Chief scientist, Siemens Asahi Meditech Co Ltd., Tokyo, Japan for his support in the MR examinations. We also thank Prof. Brian Quinn for critically reading the English manuscript. This study was supported by a Grant-in-Aid
Y.N. Kami et al. / Journal of Neuroscience Methods 172 (2008) 48–53
for Scientific Research from the Ministry of Education, Japan (No. 19390479) (T.K. G.) (No. 18109013 and 18077004) (Y. N.) and also by the Kyushu University Foundation (T.K. G.). References Andersson JL, Hutton C, Ashburner J, Turner R, Friston K. Modeling geometric deformations in EPI time series. Neuroimage 2001;13:903–19. Cerf-Ducastel B, Murphy C. fMRI activation in response to odorants orally delivered in aqueous solutions. Chem Senses 2001a;26:625–37. Cerf-Ducastel B, Van de Moortele PF, MacLeod P, Le Bihan D, Faurion A. Interaction of gustatory and lingual somatosensory perceptions at the cortical level in the human: a functional magnetic resonance imaging study. Chem Senses 2001b;26:371–83. De Araujo IE, Kringelbach ML, Rolls ET, Hobden P. Representation of umami taste in the human brain. J Neurophysiol 2003a;90:313–9. De Araujo IE, Kringelbach ML, Rolls ET, McGlone F. Human cortical responses to water in the mouth, and the effects of thirst. J Neurophysiol 2003b;90:1865–76. De Araujo IE, Rolls ET. Representation in the human brain of food texture and oral fat. J Neurosci 2004;24:3086–93. Frank GK, Kaye WH, Carter CS, Brooks S, May C, Fissell K, et al. The evaluation of brain activity in response to taste stimuli-a pilot study and method for central taste activation as assessed by event-related fMRI. J Neurosci Meth 2003;131: 99–105. Friston KJ, Holmes AP, Worsley KJ. How many subjects constitute a study? Neuroimage 1999a;10:1–5. ¨ Friston KJ, Holmes AP, Price CJ, Buchel C, Worsley KJ. Multisubject fMRI studies and conjunction analyses. Neuroimage 1999b;10:385–96. Guest S, Grabenhorst F, Essick G, Chen Y, Young M, McGlone F, et al. Human cortical representation of oral temperature. Physiol Behav 2007;92:975–84. Haase L, Cerf-Ducastel B, Buracas G, Murphy C. On-line psychophysical data acquisition and event-related fMRI protocol optimized for the investigation of brain activation in response to gustatory stimuli. J Neurosci Meth 2007;159: 98–107. Kobayakawa T, Endo H, Saito S, Ayabe-Kanamura S, Kikuchi Y, Yamaguchi Y, et al. Trial measurements of gustatory-evoked magnetic fields. Electroenceph Clin Neurophysiol Suppl 1996;47:133–41. Kobayakawa T, Ogawa H, Kaneda H, Ayabe-Kanamura S, Endo H, Saito S. Spatiotemporal analysis of cortical activity evoked by gustatory stimulation in humans. Chem Senses 1999;24:201–9. Kobayashi M, Takeda M, Hattori N, Fukunaga M, Sasabe T, Inoue N, et al. Functional imaging of gustatory perception and imagery: “top-down” processing of gustatory signals. Neuroimage 2004;23:1271–82.
53
Kringelbach ML, O’Doherty J, Rolls ET, Andrews C. Activation of the human orbitofrontal cortex to a liquid food stimulus is correlated with its subjective pleasantness. Cereb Cortex 2003;13:1064–71. Mai JK, Assheuer J, Paxinos G. Atlas of the human brain. Second ed. Amsterdam: Elsevier; 2004. Marciani L, Pfeiffer JC, Hort J, Head K, Bush D, Taylor AJ, et al. Improved methods for fMRI studies of combined taste and aroma stimuli. J Neurosci Meth 2006;158:186–94. Nitschke JB, Dixon GE, Sarinopoulos I, Short SJ, Cohen JD, Smith EE, et al. Altering expectancy dampens neural response to aversive taste in primary taste cortex. Nat Neurosci 2006;9:435–42. O’Doherty J, Rolls ET, Francis S, Bowtell R, McGlone F. Representation of pleasant and aversive taste in the human brain. J Neurophysiol 2001;85:1315–21. O’Doherty JP, Deichmann R, Critchley HD, Dolan RJ. Neural responses during anticipation of a primary taste reward. Neuron 2002;33:815–26. Ogawa H, Wakita M, Hasegawa K, Kobayakawa T, Sakai N, Hirai T, et al. Functional MRI detection of activation in the primary gustatory cortices in humans. Chem Senses 2005;30:583–92. ¨ Osterbauer RA, Wilson JL, Calvert GA, Jezzard P. Physical and physiological consequences of passive intra-oral shimming. Neuroimage 2006;29:245–53. Pritchard TC. The primate gustatory system. In: Getchell TV, et al., editors. Smell and taste in health and disease. New York: Raven Press; 1991. p. 109–25. ¨ Schoenfeld MA, Neuer G, Tempelmann C, Schußler K, Noesselt T, Hopf JM, et al. Functional magnetic resonance tomography correlates of taste perception in the human primary taste cortex. Neuroscience 2004;127:347–53. ´ Scott TR, Plata-Salaman CR. Taste in the monkey cortex. Physiol Behav 1999;67:489–511. Small DM, Zald DH, Jones-Gotman M, Zatorre RJ, Pardo JV, Frey S, et al. Human cortical gustatory areas: a review of functional neuroimaging data. Neuroreport 1999;10:7–14. Small DM, Gregory MD, Mak YE, Gitelman D, Mesulam MM, Parrish T. Dissociation of neural representation of intensity and affective valuation in human gustation. Neuron 2003;39:701–11. Small DM, Voss J, Mak YE, Simmons KB, Parrish T, Gitelman D. Experience-dependent neural integration of taste and smell in the human brain. J Neurophysiol 2004;92:1892–903. Smits M, Peeters RR, van Hecke P, Sunaert S. A 3 T event-related functional magnetic resonance imaging (fMRI) study of primary and secondary gustatory cortex localization using natural tastants. Neuroradiology 2007;49:61–71. Van de Moortele PF, Cerf B, Lobel E, Paradis AL, Faurion A, Le Bihan D. Latencies in fMRI time-series: effect of slice acquisition order and perception. NMR Biomed 1997;10:230–6. Wagner A, Aizenstein H, Mazurkewicz L, Fudge J, Frank GK, Putnam K, et al. Altered insula response to taste stimuli in individuals recovered from restricting-type anorexia nervosa. Neuropsychopharmacology 2008;33:513–23.