in humans, we measured regional cerebral blood flow (rCBF) with PET while subjects tasted an aversive saline solution. Attempting to taste water served as the ...
Brain (1998), 121, 1143–1154
Aversive gustatory stimulation activates limbic circuits in humans David H. Zald,1,2 Joel T. Lee,1 Kevin W. Fluegel1 and Jose´ V. Pardo1,2 1Cognitive
Neuroimaging Unit, Psychiatry Service, Minneapolis Veterans Affairs Medical Center, 2Division of Neuroscience Research, Department of Psychiatry, University of Minnesota, Minneapolis, Minnesota, USA
Correspondence to: Jose´ V. Pardo, Cognitive Neuroimaging Unit (11P), VAMC, One Veterans Drive, Minneapolis, MN 55417, USA
Summary Animal studies implicate the amygdala and its connections in the recognition of aversive stimuli. A recent PET study demonstrated that the human amygdala and left orbitofrontal cortex show substantial increases in regional cerebral blood flow (rCBF) during exposure to aversive odourants. To examine if aversive gustatory stimuli similarly activate these regions, nine healthy women tasted an aversive saline solution, pure water and chocolate while rCBF was measured with PET. The aversive saline condition, when contrasted with the water condition,
increased activity in the right amygdala, left anterior orbitofrontal cortex, medial thalamus, pregenual and dorsal anterior cingulate, and the right hippocampus. The right amygdala, left orbitofrontal cortex and pregenual cingulate remained significantly activated when saline was compared with chocolate. The present results indicate that the amygdala and orbitofrontal cortex respond to aversive stimuli in both the olfactory and gustatory modalities, and highlight the role of the pregenual cingulate in negative emotional processing.
Keywords: amygdala; medial thalamus; orbitofrontal; PET; taste Abbreviations: BA 5 Brodmann area; rCBF 5 regional cerebral blood flow
Introduction Electrophysiological and lesion studies in a number of mammalian species indicate that the amygdala plays a critical role in evaluating the affective significance of stimuli in many sensory modalities (LeDoux, 1987; Davis, 1992). Recently, we demonstrated that humans show large bilateral increases in amygdala activity during exposure to highly aversive odourants (Zald and Pardo, 1997). This response exceeded that seen with pleasant or neutral odourants. Although the amygdala receives particularly direct projections from the olfactory system, it seems unlikely from the animal literature that the response of the human amygdala to aversive stimuli is confined to the olfactory domain. Like the olfactory system, the gustatory system possesses relatively direct projections to the amygdala. Substantial gustatory information reaches the lateral, basal and central nuclei of the amygdala directly from the insular primary gustatory region (Turner, 1980). Additional gustatory afferents may also reach the amygdala from the caudolateral orbitofrontal cortex and gustatory responsive nuclei in the brainstem and thalamus (Norgren, 1976; Beckstead et al., 1980; Turner and Herkenham, 1981; Yasui et al., 1984; Amaral et al., 1992). Consistent with these multiple projections, several discrete nuclei within the amygdala © Oxford University Press 1998
possess cells with gustatory responses (Azuma et al., 1984; Scott et al., 1993; Yasoshima et al., 1995). Electrophysiological and lesion data indicate that the amygdala is probably not necessary for basic sensory perception and discrimination of taste (Aggleton, 1992; Scott et al., 1993) However, lesions of the amygdala in animals critically impair their ability to directly associate gustatory stimuli with stimuli from other sensory modalities (Kennie and Nagel, 1973; Arthur, 1975; Mikulka et al., 1977; Gaffan and Murray, 1990). Moreover, bilateral amygdalectomies in non-human primates produce marked alterations in feeding behaviour including the ingestion of substances that normal animals find unpalatable (Klu¨ver and Bucy, 1937). Based upon this behavioural evidence and our previous findings with olfactory stimuli, we hypothesized that exposure to aversive gustatory stimuli would significantly increase activity within the human amygdala. We similarly hypothesized that aversive gustatory stimuli would increase activity within the orbitofrontal cortex. The orbitofrontal cortex receives significant projections from both the primary gustatory cortex and the amygdala (Amaral et al., 1992; Baylis et al., 1995; Carmichael and Price, 1995; Zald and Kim, 1996a), placing it in a position to code or process
1144
D. H. Zald et al.
information about the motivational value of gustatory stimuli. Lesions of the orbitofrontal cortex in non-human primates produce impairments that resemble those observed following amygdala lesions, including deficits in associating gustatory reinforcers with neutral stimuli and Klu¨ver–Bucy-like abnormalities in eating behaviour (Butter et al., 1969; Ursin et al., 1969; Gaffan and Murray, 1990; Baylis and Gaffan, 1991). Furthermore, orbitofrontal cortex activity (particularly in the left hemisphere) has been observed in a number of PET studies involving negative emotional inductions (Pardo et al., 1993; Rauch et al, 1994, 1995; Fischer et al., 1996), including exposure to aversive odourants (Zald and Pardo, 1997). To examine whether an unconditioned aversive gustatory stimulus would activate the amygdala and orbitofrontal cortex in humans, we measured regional cerebral blood flow (rCBF) with PET while subjects tasted an aversive saline solution. Attempting to taste water served as the primary control condition, and as an additional gustatory control condition, subjects were scanned while tasting chocolate. This allowed an examination of whether regions activated by saline are similarly activated by other gustatory stimuli, and whether saline, perceived as highly unpleasant, differentially activate limbic regions relative to a highly pleasant gustatory stimulus.
Methods Subjects Ten right-handed female subjects (mean age 37 years, range 25–62 years) were studied with PET while tasting either saline solution, pure water or chocolate. Only women were studied in order to maximize the likelihood of intense emotional experiences (Shields, 1991), and for consistency with our previous study of aversive olfaction which utilized only women (Zald and Pardo, 1997). All subjects gave informed consent approved by the VAMC Human Subjects Committee and Radioactive Drug Research Committee. One subject was excluded due to motion during the saline condition (see below) leaving a total of nine subjects with valid data for all conditions.
Experimental design The stimulus in the aversive saline condition consisted of a 5% solution of iodized NaCl dissolved in deionized distilled water at room temperature. The stimulus for the pure water condition consisted of deionized distilled water at room temperature. The chocolate stimulus consisted of 3 g of Hershey’s Symphony Chocolate (Hershey’s, Hershey, Pa., USA). In the saline and pure water conditions, 3 ml of the fluid was injected ~5 s before the start of scan acquisition into the mouth through a small plastic cannula held between the teeth. An additional 2 ml of fluid was slowly injected into the subject’s mouth during the course of the next 45 s. Prior to fluid injection, subjects received the following
instructions: ‘You are about to receive a liquid in your mouth. Close your eyes, and see if you can taste anything. When you feel the liquid in your mouth, swish it around a couple of times and then allow your tongue to rest. If there is too much fluid in your mouth, go ahead and swallow.’ In the chocolate condition, the chocolate was placed on the tip of the subject’s tongue 5 s before the start of scan acquisition, and subjects were instructed, ‘Close your eyes, and see if you can taste anything. When you feel something on your tongue, close your mouth, swish it around a couple of times, and then allow your tongue to rest.’ After each condition, subjects rated the stimulus for pleasantness–unpleasantness on an 11-point visual analogue scale with anchors at 0 (extremely unpleasant), 5 (neutral) and 10 (extremely pleasant), and intensity, also on an 11-point visual analogue scale with anchors at 0 (undetectable) and 10 (extremely intense). Subjects were informed that they would receive an unpleasant taste during one scan condition, but were blind to the scan number for that condition, the identity of the stimulus and to its degree of unpleasantness. Subjects also did not know what other gustatory stimuli they would receive for the other scans. However, when subjects completed the saline condition before the water and/or chocolate condition, they were told that they would not receive another unpleasant taste. Five of the subjects with valid scans received the chocolate condition before the saline condition, while four of the subjects received the saline first. Because of the odd number of subjects providing valid scans, it was not possible to fully counterbalance the conditions. However, post hoc analysis provided no evidence of an order effect.
PET imaging and analysis The rCBF was estimated from normalized (1000 counts) tissue radioactivity using a Siemens ECAT 953B camera (Knoxville, Tenn., USA) with septae retracted; a ‘slow-bolus’ injection of H215O (an initial dose of 814 MBq or 22 mCi infused at a constant rate over 30 s; Silbersweig et al., 1993) was followed by a 90 s scan acquisition beginning upon radiotracer arrival into the brain (10 min inter-scan intervals). Images were initially reconstructed with a 3D reconstruction algorithm using a 0.5 cycles-per-pixel Hanning filter (Kinahan and Rogers, 1989) and attenuation correction using a 2D transmission scan. Measured coincidences were corrected for random events and electronic dead time, but not for decay or scatter. Following reconstruction, images were visually inspected for motion artefacts using ANALYZE (BRU, Mayo Foundation, Rochester, Minn., USA). Movement between scans was empirically quantified through examination of shift and rotation parameters from automated co-registration files (Minoshima et al., 1992). One subject was excluded because she moved her head substantially during the course of the saline condition. This problem did not occur with the other subjects, but initial examination of motion artefacts and registration data indicated that some of the subjects moved between scans. Such movement may cause artefacts in ventral
Limbic responses to saline regions of the brain as a result of misalignment of emission scans with transmission scans (Huang et al., 1979). To eliminate this potential source of error, we implemented a realignment procedure similar to that described by Andersson et al. (1995); before attenuation correction, each emission scan was coregistered to the subject’s first emission scan, using Automated Image Registration (Woods et al., 1992). Normalization for global activity, co-registration within each study session, placement of the intercommissural line from image fiducials, and linear warping of each subject’s scans to a reference stereotactic atlas (Talairach and Tournoux, 1988) were subsequently accomplished with software provided by Minoshima and co-workers (Minsohima et al., 1992, 1993, 1994). Images were blurred with a 3-pixel 3D Gaussian filter producing a final image resolution of ~10 mm full-width at half-maximum and a mapping resolution of ,2 mm (Fox et al., 1986). Pixel-wise subtractions were performed to determine activations occurring in the saline condition relative to the water condition. Statistical analysis employed the global variance of all intracerebral pixels (Worsley et al., 1992). A threshold of P , 0.0005 (equivalent to a Z-score of 3.3) was selected for the analysis of the contrast between saline and pure water conditions. This threshold is slightly more conservative than the P , 0.001 cutoff frequently used in pixel-wise analyses of PET images. The more stringent P , 0.0005 threshold was based on a bootstrapping study of 60 subjects scanned twice while resting with their eyes closed. This bootstrapping analysis indicated that, on average, the imaging techniques used in this study produce approximately one false positive focus (emerging due to chance) with a sample size of nine subjects. Because the goal of the saline versus chocolate comparison was to elucidate the nature of activity in these areas rather than to perform another exploratory analysis, a threshold of P , 0.001 was adopted for areas already identified in the other comparison. However, regions which failed to reach significance in the initial saline versus water comparison are only reported if they meet the P , 0.0005 criterion in the saline versus chocolate comparison.
Results All nine subjects with valid scans rated the saline solution as highly aversive (mean 5 0.9; range 0–2.5 on the 11-point scale) and highly intense (mean 5 8.9; range 6–10). In contrast, only one subject reported detecting any taste during the pure water condition and this was described as barely detectable. The chocolate was rated as highly pleasant (mean 5 9.0; range 8–10) and moderately intense (mean 5 6.8; range 5–8.5). These ratings indicate that the chocolate and saline conditions were well matched for hedonic strength (i.e. absolute deviation from a neutral rating of 5.0), but the saline was slightly more perceptually intense than the chocolate. Subjects frequently described experiencing muscle tension during the saline condition. None of the subjects
1145
Table 1 Locations of increased rCBF when tasting 5% saline versus pure water Area
x
y
z
Z-score
Right medial thalamus Right amygdala Left pregenual cingulate (BA 32) Right cingulate cortex (BA 24) Left cingulate cortex (BA 24) Left pregenual cingulate (BA 24) Left orbitofrontal cortex (BA 11) Right hippocampus
1 26 –1 3 –6 –3 –24 26
–22 1 39 10 –22 30 41 –15
9 –16 0 40 36 0 –7 –20
4.2 3.8 3.7 3.7 3.5 3.5 3.4 3.4
Stereotactic coordinates (mm) identify the location of the rCBF maxima according to the atlas of Talairach and Tourneau (1988). x 5 medial–lateral position relative to the midline (right hemisphere positive); y 5 anterior–posterior position relative to the anterior commissure (anterior positive); z 5 inferior–superior position relative to the intercommissural plane (superior positive).
Table 2 Locations of rCBF maxima when tasting saline versus chocolate Area
x
y
z
Z-score
Left pregenual cingulate (BA 24) Right motor cortex (BA 4) Right insula Right motor cortex (BA 4) Left orbitofrontal cortex (BA 11) Right amygdala
–1 57 33 46 –21 26
32 –15 14 –15 39 –8
0 38 7 34 –7 –18
4.2 3.9 3.7 3.5 3.5 3.1
reported feeling disgust or fear, but they all indicated a dislike for the saline. In comparison, subjects frequently stated that the chocolate relaxed them, and several subjects asked for additional chocolate upon completion of the scan session. Table 1 shows brain regions activated in the contrast between tasting saline and tasting pure water. The largest rCBF increase in the saline condition was localized to the medial thalamus in the region of the dorsomedial and midline (intermediodorsal) nuclei. The peak of this diencephalic focus occurred in the right hemisphere, but the focus extended into the left hemisphere and may represent a bilateral response (see Fig. 1). Tasting saline also induced significant rCBF increases in the pregenual and dorsal anterior cingulate, left orbitofrontal cortex, right amygdala and right anterior hippocampus. Figure 2 shows the location of rCBF increases in the right amygdala and hippocampus. The focus in the left orbitofrontal cortex involved an anterior medial region located along the superior apex of the ‘H-shaped’ orbital sulcus (Mai et al., 1997). Also a modest non-significant activation was localized to the right orbitofrontal cortex in Brodmann area (BA) 11 (at coordinates x 5 –17, y 5 37, z 5 –14; Z-score 5 3.0), indicating that the orbitofrontal cortex response to aversive saline was not completely lateralized. Many of these regions exhibited statistically significant activity in the contrast between the saline and chocolate
1146
D. H. Zald et al.
Fig. 1 Medial thalamic and cingulate activation arising in the contrast between tasting aversive saline and tasting pure water. The image displays activity above a significance threshold of P , 0.001 (hotter colours denote greater activation) superimposed on a surface rendering of the medial wall of the brain. The image only displays the maximum activity occurring within 4 pixels (9 mm) of the medial wall. The left hemisphere of the brain is displayed at the top of the figure.
conditions (see Table 2). Foci emerging in this comparison included the pregenual cingulate, left orbitofrontal cortex and right amygdala. Figure 3 shows the location of the left orbitofrontal cortex focus in the saline versus chocolate comparison. Subjects also demonstrated hippocampal
(x 5 –28, y 5 –19, z 5 –18; Z-score 5 2.8) and thalamic activity (x 5 –1, y 5 –22, z 5 7; Z-score 5 2.5), but neither of these reached statistical significance at the P , 0.001 cutoff. Unexpectedly, the saline condition produced significantly greater activity than the chocolate condition in
Limbic responses to saline
Fig. 2 Right amygdala and hippocampal activation arising in the contrast between tasting aversive saline and tasting pure water. The image displays rCBF activation above a threshold of P , 0.001 (hotter colours denote greater activation) superimposed on the corresponding transverse slice (z 5 –16) of a standard T1weighted MRI. The hippocampal maxima localized ~4 mm below this slice. The right side of the MRI corresponds to the left side of the brain.
the right anterior insula in the vicinity of presumed primary gustatory cortex, even though both conditions involved gustatory stimulation. The primary motor cortex (BA 4) also showed increased activity in this contrast, suggesting that there was more tongue movement during the saline condition. The contrast between the chocolate and pure water conditions revealed no significant increases in activity in the amygdala, orbitofrontal cortex or pregenual cingulate. It may be noted that some subjects showed rCBF increases in the right amygdala in the chocolate condition, but these increases were inconsistent, and several of the subjects showed rCBF decreases. Interestingly, the two subjects showing the greatest amygdala responses (11.5% and 13.9% rCBF increases) in the chocolate versus water contrast also showed the greatest rCBF increases (22.1% and 22.6%) in the contrast between the saline and water conditions. The most robust increase in the chocolate versus water contrast was localized to the posterior cingulate [BA 31; two peaks, at x 5 –12, y 5 – 40, z 5 36 (Z-score 5 4.6) and x 5 –3, y 5 –42, z 5 36 (Z-score 5 4.5)]. A dorsal anterior cingulate focus also emerged in BA 24 (x 5 –6, y 5 –24, z 5 38; Z-score 5 4.3) at similar coordinates to those appearing in the contrast
1147
Fig. 3 Surface rendering of the ventral aspect of the brain demonstrating the location of the left orbitofrontal cortex focus in the saline versus chocolate comparison. Only rCBF increases with significance greater than P , 0.001 are shown with hotter colours denoting greater activation. The right side of the image corresponds to the left side of the brain.
between the saline and water conditions. Therefore, the dorsal anterior cingulate responds similarly to both pleasant and aversive gustatory stimuli, which hence cancel each other in the contrast between the saline and chocolate conditions. Finally, the chocolate condition produced activation in the right dorsomedial thalamic region which failed to reach statistical significance (x 5 6, y 5 –15, z 5 9; Z-score 5 2.8). The restricted range of hedonic (pleasantness– unpleasantness) ratings of saline limits the utility of performing parametric or correlational analyses between rCBF and psychoperceptual ratings. Five subjects rated the stimulus as extremely unpleasant (i.e. hedonic ratings of 0.0) and four subjects rated it as moderately unpleasant (ratings of 2.0–2.5). To determine whether the perceived degree of aversiveness influenced rCBF, we split subjects into those with moderately unpleasant and extremely unpleasant experiences, based on their subjective ratings to saline. The percentage rCBF increase for the two groups was calculated by averaging all pixels within spherical (4.5 mm radius) regions of interest. These regions of interest were centred on the peak coordinates defined by the saline minus water comparison, except in the case of the right amygdala, where the region of interest was centred on the peak coordinates from the saline minus chocolate comparison to ensure that the region of interest remained limited to the amygdala
1148
D. H. Zald et al.
Table 3 Percentage rCBF increase within regions of interest for the saline versus pure water in subjects experiencing the saline as extremely aversive versus moderately aversive Region
Group with extremely unpleasant experience (n 5 5)
Group with moderately unpleasant experience (n 5 4)
Right hippocampus Pregenual cingulate Right amygdala Left orbitofrontal cortex Medial thalamus
14.5 9.8 11.1 11.4 13.4
4.4 1.5 10.4 10.7 14.1
proper. Table 3 lists the rCBF increases for activity within pregenual cingulate, medial thalamic, orbitofrontal, amygdala and hippocampal regions of interest for the two subject groups. The hippocampus showed the most dramatic difference between the two groups, increasing 10% more in the extremely unpleasant group than in the moderately unpleasant group. The pregenual cingulate only showed substantial rCBF increases in the subjects who perceived the stimulus as extremely aversive. In contrast, the right amygdala, medial thalamus and left orbitofrontal cortex all showed robust responses regardless of whether subjects rated saline as moderately or extremely unpleasant. To determine whether additional regions were specific to the sub-sample experiencing the saline as extremely aversive, we performed a post hoc pixel-wise comparison between the saline and water conditions for the five subjects rating the saline as extremely aversive. This increased the detection sensitivity which might have been decreased in the full group comparison. Several right hemisphere foci emerged in this comparison which had not reached statistical significance in the full group analysis. The largest of these additional responses arose in the right frontal pole (x 5 21, y 5 50, z 5 2; Z-score 5 3.9). A right pregenual cingulate focus (BA 32; x 5 10, y 5 28, z 5 –9; Z-score 5 3.8) appeared in addition to the previously reported left cingulate focus. This confirms that the right hemisphere pregenual cingulate activity observed in Fig. 2 most probably reflects bilateral activity rather than blurring from a left hemisphere focus. The analysis also revealed bilateral medial thalamic activation (x 5 6, y 5 –19, z 5 9; Z-score 5 3.7 and x 5 –1, y 5 –22, z 5 9; Z-score 5 3.8). Finally, a right anterior insular region was activated in this condition (x 5 –35, y 5 10, z 5 –2; Z-score 5 3.6), with four of the five subjects showing a .12% increase in right insular rCBF. These analyses indicate that some brain regions only respond robustly when saline is experienced as highly aversive, and these rCBF responses may be obscured when subjects with more moderate hedonic responses are pooled with subjects with more dramatic subjective responses to saline. To examine the functional interactions between limbic regions activated by saline, we examined the correlation
between rCBF within regions of interest selected from the saline and water contrast. Non-subtracted rCBF for regions of interest within the saline and water conditions were separately submitted to correlational analysis. Interpretation of such an analysis must be considered exploratory due to the small sample size and the multiple structures activated in the contrast between saline and water conditions. Nevertheless, such analyses provide valuable information in instances where brain regions are functionally coupled (Zald et al., 1998). Regions of interest were defined as described above; however, instead of examining the change in rCBF between conditions, rCBF was calculated separately within the saline and water conditions. These analyses revealed a tight coupling of activities in the left pregenual cingulate and the left anterior orbitofrontal cortex during the saline condition (r 5 0.81, P , 0.01). In contrast, the activity in these regions were not significantly correlated in the water condition (r 5 0.27, P . 0.10). No other regions showed significant functional coupling in the saline or water condition, although activity in the right amygdala showed a tendency to be linked with that in the right hippocampus (r 5 0.55, P , 0.10) and the left orbitofrontal cortex (r 5 0.51, P , 0.10) in the saline condition.
Discussion The current study demonstrates that exposure to an aversive gustatory stimulus activates a network of limbic structures involving the amygdala, pregenual cingulate and orbitofrontal cortex. Subjects showed increased activity in these regions when tasting aversive saline, relative to attempting to taste pure water and to tasting chocolate. Because the saline and pure water conditions were matched for temperature and volume of fluid, the observed activations cannot be attributed to non-gustatory sensory aspects of intraoral stimulation (e.g. somatosensory or thermal coding). These activations probably do not reflect basic gustatory processing, since these areas remained significantly activated when saline was contrasted with another gustatory stimulus. Furthermore, the ability of saline, relative to chocolate, to activate these regions does not appear to reflect differences in the hedonic strength of the two stimuli, since subjects rated both stimuli as strong. Rather, the rCBF increases in the amygdala, pregenual cingulate and anterior orbitofrontal cortex appear to reflect specifically the recognition, experience and/or response to the aversive quality of the saline. The ability of an aversive gustatory stimulus to activate the amygdala robustly converges with our previous finding that exposure to aversive odourants activates the amygdala (Zald and Pardo, 1997). It is also consistent with a recent PET study of unpleasant visual stimulation, which demonstrated an increase in amygdala rCBF (Lane et al., 1997b). Taken together, these findings support the hypothesis that the human amygdala plays a multimodal role in the recognition, evaluation and/or response to aversive stimuli. In both the present gustatory study and our previous olfactory study (Zald
Limbic responses to saline and Pardo, 1997), subjects rated the stimuli as moderately to extremely aversive and highly arousing. We have not observed substantial increases in amygdala activity when subjects rate odours or tastes as only mildly unpleasant. The strength of the subjective experience of aversiveness may underlie the ability of these gustatory and olfactory paradigms to activate the amygdala strongly. This interpretation converges with animal studies which indicate that amygdala lesions critically impair the acquisition of conditioned emotional responses to neutral stimuli that are paired with highly aversive or highly arousing stimuli, while leaving conditioning to mildly unpleasant, non-arousing stimuli unimpaired (Cahill and McGaugh, 1990). Unexpectedly, the amygdala activation occurred only in the right hemisphere. This contrasts with our previous finding of bilateral amygdala activation during aversive olfaction (Zald and Pardo, 1997). The reason for this lateralized response is unclear, although it may reflect a right dominance for some aspects of gustatory processing. Patients with right anterior temporal lobectomies have been observed to demonstrate elevated taste recognition thresholds for citric acid relative to patients with left anterior temporal lobectomies (Small et al., 1997a; but see Henkin et al., 1977). This pattern of right dominance for anterior temporal lobe processing of citric acid has also been confirmed with PET (Small et al., 1997a). In the present study, a similar pattern of laterality was evident in the insula, which only showed increased activity in the right hemisphere. The peak focus in the amygdala region for the saline minus water comparison lies at the extreme anterior end of the amygdala as defined by the atlases of Talairach and Tournoux (1988) and Mai et al. (1997). Given the spatial resolution of current PET techniques and inter-subject anatomical variability, this focus may include cortex slightly anterior to the amygdala. Indeed, we have observed activity in the vicinity of the pyriform cortex (anterior to the amygdala) in some taste experiments (D.H.Z. and J.V.P., unpublished observations). The right anterior temporal lobe focus reported by Small et al. (1997a) in subjects tasting citric acid also falls slightly anterior and medial to the Talairach coordinates for the amygdala. However, activity in the anterior amygdala/pyriform area largely cancels when tasting saline is compared with tasting chocolate, leaving a discrete focus centred more posteriorly within the amygdala. This more posterior focus falls solidly within the Talairach and Tournoux (1988) and Mai et al. (1997) coordinates for the amygdala, and remains clearly distinct from the hippocampal focus which lies 11 mm posterior. Recordings of single units in non-human primates suggest that gustatory reward may also activate the amygdala (Azuma et al., 1984; Ono and Nishijo, 1992). However, we did not observe consistent evidence of amygdala activation in the contrast between the chocolate and water conditions. Furthermore, the aversive saline condition activated the amygdala at similar levels when compared with either water or chocolate. Although some subjects showed rCBF increases
1149
in the amygdala region of interest when tasting chocolate relative to tasting water, these increases were inconsistent across subjects. The possibility that water to some extent obscured the ability to observe amygdala responses to chocolate requires consideration. A few amygdala cells show responses to water in electrophysiological studies of nonhuman primates (Nishijo et al., 1988). Indeed, we have observed moderate rCBF increases in the right amygdala region in subjects attempting to taste water relative to a resting condition (D.H.Z. and J.V.P., unpublished observation). This may reflect the role of water as a reinforcer in its own right. Nevertheless, saline activates this area substantially more than water does, whereas chocolate does not. A small percentage of amygdala cells also respond to intra-oral thermal or tactile information (Azuma et al., 1984; Nishijo et al., 1988; Scott et al., 1993). Such thermal and tactile factors were well matched in the saline versus pure water comparison, but not adequately controlled in comparisons involving chocolate and either water or saline. Despite this potential problem, the greater amygdala activation during aversive as compared with pleasant gustatory stimulation converges with a growing literature demonstrating greater involvement of the human amygdala in negative emotional processing than positive emotional processing (Gloor, 1990; Adolphs et al., 1995; Breiter et al., 1996; Irwin et al., 1996; Ketter et al., 1996; Morris et al., 1996; Young et al., 1996; Lane et al., 1997b; Scott et al., 1997). In addition to activating the right amygdala, saline activated several closely interconnected limbic and paralimbic regions. The amygdala, pregenual cingulate, orbitofrontal cortex and medial thalamus (dorsomedial and midline nuclei) all connect with one another through direct and mostly reciprocal connections (Pandya et al., 1981; Vogt and Pandya, 1987; Amaral et al., 1992; Neafsey et al., 1993; Ray and Price, 1993; Van Hoesen et al., 1993; Groenewegen and Berendse, 1994; Baylis et al., 1995; Carmichael and Price, 1995; 1996; Zald and Kim, 1996a; Bachevalier et al., 1997). The areas activated by saline thus conform closely to a widespread distributed limbic/paralimbic network. Anatomical and functional studies highlight the close functional interaction between the orbitofrontal cortex and amygdala (Zald and Kim, 1996a, b; Zald et al., 1998). The ability of aversive saline to activate the human orbitofrontal cortex converges with electrophysiological data from nonhuman primates demonstrating the presence of orbitofrontal cortex cells with specific responses to aversive saline (Thorpe et al., 1983; Rolls et al., 1990). Single cell recordings in monkeys suggest that the human orbitofrontal cortex should also possess cells responsive to rewarding gustatory stimuli such as chocolate (Thorpe et al., 1983; Rolls et al., 1990). However, we failed to observe significant increases in the orbitofrontal cortex during exposure to chocolate compared with water. In fact, half of the subjects showed greater rCBF within the anterior orbitofrontal region of interest when tasting water than when tasting chocolate. Furthermore,
1150
D. H. Zald et al.
comparison of saline with chocolate instead of water did not reduce the magnitude of the left orbitofrontal cortex focus. The orbitofrontal cortex (particularly the left orbitofrontal cortex) emerges as one of the most frequently activated regions during aversive sensory and psychological experiences (Pardo et al., 1993; Rauch et al., 1994; 1995; Fischer et al., 1996; Zald and Pardo, 1997). However, the specific foci activated within the orbitofrontal cortex tend to vary depending on the sensory modality and experimental paradigm, perhaps reflecting the large differences in sensory and limbic afferents that distinguish orbitofrontal cortex subregions (Carmichael and Price, 1995, 1996). It may be noted that the left orbitofrontal cortex focus in the current study appears too anterior to represent the caudolateral orbitofrontal cortex area identified as secondary gustatory cortex in monkeys (Rolls, 1989; Rolls et al., 1990). This anterior-medial region is more likely to represent a heteromodal association region which receives gustatory information secondary to the caudolateral orbitofrontal cortex (Rolls and Baylis, 1994; Carmichael and Price, 1995; 1996). The medial thalamus emerged as the region with the largest magnitude activation in the saline versus water comparison. Like lesions of the orbitofrontal cortex and amygdala, dorsomedial thalamic lesions disrupt the ability to form direct associations between visual stimuli and gustatory reinforcers (Gaffan and Murray, 1990). Thalamic lesions that include the dorsomedial nucleus frequently disrupt the acquisition of avoidance learning and other conditioned emotional responses (Buchanan and Powell, 1993; Gabriel, 1993). Human data also indicate that medial thalamic areas are activated during exposure to visual stimuli perceived as disgusting (Lane et al., 1997a; Paradiso et al., 1997). However, the dorsomedial thalamic activity appears less clearly specific to the aversive condition. Chocolate (when compared with water) produced moderate, albeit non-significant, increases in right medial thalamic rCBF, and the comparison between saline and chocolate failed to produce a statistically significant activation in this region. Furthermore, this area showed robust rCBF increases (.11%) regardless of whether subjects found the saline moderately or extremely aversive. This indicates that the medial thalamus probably plays a relatively balanced role in the processing of gustatory stimuli, which is not restricted by the intensity or direction of the hedonic experience. This conclusion converges with evidence that visual stimuli that induce positive emotions often produce rCBF increases in the thalamus similar to those with visual stimuli that induce negative emotions (Lane et al., 1997a, b; Reiman et al., 1997). It also appears consistent with a recent case report in which a dorsomedial thalamic lesion in a 68-year-old woman produced changes in the hedonic perception of tastes, particularly causing previously pleasant stimuli to be perceived as neutral or unpleasant, without altering basic olfactory and gustatory identification (Rousseaux et al., 1996). Exposure to aversive saline induced activity in multiple portions of the cingulate cortex. However, only the activity in the inferior, pregenual portions of the cingulate (BA 24
and adjacent BA 32) remained significant when saline was compared with chocolate, indicating a valence-specific response (unpleasant distinct from pleasant). Furthermore, this region appeared heavily influenced by the strength of the subjective experience with large rCBF increases (.6%) in both the left and right pregenual cingulate occurring only in subjects who found the stimulus extremely aversive. The large dependence of pregenual cingulate activity upon the hedonic significance of the stimulus argues against a role in basic taste processing. This inferior pregenual cingulate region has previously been implicated in affective behaviour (Devinsky et al., 1995). Animal studies implicate this region in aspects of avoidance learning and emotional conditioning (Buchanan and Powell, 1993; Gabriel, 1993). Recent neuroimaging studies of anxiety, fear, dysphoria and depression also implicate the inferior-medial frontal lobe in unpleasant emotional states (George et al., 1995; Rauch et al., 1995, 1996; Drevets et al., 1997; Mayberg et al., 1997; Shin et al., 1997). The current findings thus converge with both animal and human data, in highlighting the importance of the inferior-medial frontal cortex in negative emotional experiences. In contrast, the more dorsal anterior and midcingulate regions did not exhibit a similar level of specificity to aversive gustatory stimulation. For instance, the dorsal anterior cingulate responded equally well to both chocolate and saline in comparison to water. The hippocampal activation was unexpected. Although the hippocampus is directly or indirectly connected with several other areas activated by saline, including the amygdala (Amaral, 1986; Saunders et al., 1988), midline thalamus (Groenewegen and Berendse, 1994), pregenual cingulate and anterior medial orbitofrontal cortex (Vogt and Pandya, 1987; Van Hoesen et al., 1993; Carmichael and Price, 1995; Zald and Kim, 1989), it has not been commonly observed in previous neuroimaging studies of emotion. Interestingly, in a recent PET study, there was coactivation of the left amygdala and left hippocampus in subjects viewing unpleasant pictures (Lane et al., 1997b). In both the study of Lane et al. (1997b) and the present study, the hippocampal activation occupied an anterior portion of the hippocampus, and occurred ipsilateral to the amygdala activation. The present data further suggest that hippocampal activation primarily occurred among subjects who found the saline extremely aversive. This could reflect a gating process through which highly arousing stimuli preferentially gain access to hippocampal mnemonic processes. Data from both humans and animals highlight the importance of the amygdala for the emotional enhancement of memory (Markowitsch et al., 1994; Cahill et al., 1995, 1996; McGaugh et al., 1996). The current data raise the possibility that the amygdala accomplishes this enhancement by modulating hippocampal processing when stimuli are perceived as extremely aversive or arousing. The right anterior insula showed activations in both the contrast of saline versus chocolate, and the contrast of saline versus water, when the latter contrast was limited to subjects who perceived the saline as extremely aversive. This insular
Limbic responses to saline focus is consistent with the location of primary gustatory cortex as identified in non-human primates (Yaxley et al., 1990; Norgren, 1990). The insular focus in the saline versus water contrast arose despite the ability of water to activate this region relative to resting conditions (D.H.Z. and J.V.P., unpublished observations). Whether this response reflects taste processing itself, or a neural correlate specific to highly aversive gustation, remains unresolved, since the insular activation failed to reach significance when all subjects were included in the analysis. That this insular focus again emerged in the saline minus chocolate comparison (even though both conditions involved gustatory stimulation) supports the hypothesis that insular activity reflects the aversive nature of saline. However, a recent PET study demonstrated greater insular activity during exposure to basic tastants than during combined gustatory/olfactory stimulation (Small et al., 1997b). Since chocolate possesses both gustatory and olfactory properties, it remains possible that the observed insular activity in the saline versus chocolate comparison reflects basic taste processing, rather than being a specific correlate of aversive gustation. A potential limitation of the current study involves the use of a single continuous stimulus presentation per 90 s scan period. Cells in the gustatory system frequently show rapid habituation during extended exposure to stimuli. This may explain the inconsistencies in anterior insular activity and the lack of significant rCBF increases in the frontal operculum. While the continuous presentation of gustatory stimuli may reduce activation in primary gustatory cortex, the long duration of stimulation may increase the intensity of the aversive experience. Subjects frequently commented that the length of the scan made the saline condition more aversive because they were unable to stop or ignore the unpleasant sensation. This protracted experience may thus aid the ability to activate areas involved in the hedonic processing of tastants relative to regions processing their primary sensory properties. In conclusion, tasting aversive saline activates a network of interconnected limbic and paralimbic structures. The present data indicate that several of these areas participate in the evaluative or motivational aspects of the aversive experience rather than reflecting basic sensory coding of gustatory stimuli. Disruption of such evaluative or motivational processes by lesions in these regions could lead to an inability to recognize and learn which foods are unpalatable (as occurs in the Klu¨ver–Bucy syndrome). Moreover, the capacity of these regions to respond during exposure to aversive stimuli may underlie their role in avoidance learning and emotional conditioning.
Acknowledgements We wish to thank our volunteers and the staff of the PET Imaging Service. This work was supported, in part, by the Department of Veterans Affairs, NARSAD and the Minnesota Obesity Center (P30 DK50456–02). D.H.Z. was supported by a NRSA grant (1 F32 MH11641–01A1).
1151
References Adolphs R, Tranel D, Damasio H, Damasio AR. Fear and the human amygdala. J Neurosci 1995; 15: 5879–91. Aggleton JP. The functional effects of amygdala lesions in humans: a comparison with findings from monkeys. In: Aggleton JP, editor. The amygdala: neurobiological aspects of emotion, memory, and mental dysfunction. New York: Wiley-Liss; 1992. p. 485–503. Amaral DG. Amygdalohippocampal and amygdalocortical projections in the primate brain. In: Schwarcz R, Ben-Ari Y, editors. Excitatory amino acids and epilepsy. Adv Exp Med Biol, Vol 203. New York: Plenum Press; 1986. p. 3–17. Amaral DG, Price JL, Pitka¨nen A, Carmichael ST. Anatomical organization of the primate amygdaloid complex. In: Aggleton JP, editor. The amygdala: neurobiological aspects of emotion, memory, and mental dysfunction. New York: Wiley-Liss; 1992. p. 1–66. Andersson JL, Vagnhammar BE, Schneider H. Accurate attenuation correction despite movement during PET imaging. J Nucl Med 1995; 36: 670–8. Arthur JB. Taste aversion learning is impaired by interpolated amygdaloid stimulation but not by post-training amygdaloid stimulation. Behav Biol 1975; 13: 369–76. Azuma S, Yamamoto T, Kawamura Y. Studies on gustatory responses of amygdaloid neurons in rats. Exp Brain Res 1984; 56: 12–22. Bachevalier J, Meunier M, Lu MX, Ungerleider LG. Thalamic and temporal cortex input to medial prefrontal cortex in rhesus monkeys. Exp Brain Res 1997; 115: 430–44. Baylis LL, Gaffan D. Amygdalectomy and ventromedial prefrontal ablation produce similar deficits in food choice and in simple object discrimination learning for an unseen reward. Exp Brain Res 1991; 86: 617–22. Baylis LL, Rolls ET, Baylis GC. Afferent connections of the caudolateral orbitofrontal cortex taste area of the primate. Neuroscience 1995; 64: 801–12. Beckstead RM, Morse JR, Norgren R. The nucleus of the solitary tract in the monkey. Projections to the thalamus and brain stem nuclei. J Comp Neurol 1980; 190: 259–82. Breiter HC, Etcoff NL, Whalen PJ, Kennedy WA, Rauch SL, Buckner RL, et al. Response and habituation of the human amygdala during visual processing of facial expression. Neuron 1996; 17: 875–87. Buchanan SL, Powell DA. Cingulothalamic and prefrontal control of autonomic function. In: Vogt BA, Gabriel M, editors. Neurobiology of cingulate cortex and limbic thalamus: a comprehensive handbook. Boston (MA): Birkha¨user; 1993. p. 381–414. Butter CM, McDonald JA, Snyder DR. Orality, preference behavior, and reinforcement value of nonfood object in monkeys with orbital frontal lesions. Science 1969; 164: 1306–7. Cahill L, McGaugh JL. Amygdaloid complex lesions differentially affect retention of tasks using appetitive and aversive reinforcement. Behav Neurosci 1990; 104: 532–43. Cahill L, Babinsky R, Markowitsch HJ, McGaugh JL. The amygdala and emotional memory [letter]. Nature 1995; 377: 295–6.
1152
D. H. Zald et al.
Cahill L, Haier RJ, Fallon J, Alkire MT, Tang C, Keator D, et al. Amygdala activity at encoding correlated with long-term, free recall of emotional information. Proc Natl Acad Sci USA 1996; 93: 8016–21.
et al. Anterior paralimbic mediation of procaine-induced emotional and psychosensory experiences [see comments]. Arch Gen Psychiatry 1996; 53: 59–69. Comment in: Arch Gen Psychiatry 1997; 54: 763–5.
Carmichael ST, Price JL. Limbic connections of the orbital and medial prefrontal cortex in macaque monkeys. J Comp Neurol 1995; 363: 615–41.
Kinahan PE, Rogers JG. Analytic 3D image reconstruction using all detected events. IEEE Trans Nucl Sci 1989; 36: 964–86.
Carmichael ST, Price JL. Connectional networks within the orbital and medial prefrontal cortex of macaque monkeys. J Comp Neurol 1996; 371: 179–207. Davis M. The role of the amygdala in fear and anxiety. [Review]. Annu Rev Neurosci 1992; 15: 353–75. Devinsky O, Morrell MJ, Vogt BA. Contributions of anterior cingulate cortex to behaviour. [Review]. Brain 1995; 118: 279–306. Drevets WC, Price JL, Simpson JR Jr, Todd RD, Reich T, Vannier M, et al. Subgenual prefrontal cortex abnormalities in mood disorders [see comments]. Nature 1997; 386: 824–7. Comment in: Nature 1997; 386: 769–70.
Klu¨ver H, Bucy PC. ‘Psychic blindness’ and other symptoms following bilateral temporal lobectomy in rhesus monkeys. Am J Physiol 1937; 119: 352–3. Lane RD, Reiman EM, Ahern GL, Schwartz GE, Davidson RJ. Neuroanatomical correlates of happiness, sadness, and disgust. Am J Psychiatry 1997a; 154: 926–33. Lane RD, Reiman EM, Bradley MM, Lang PJ, Ahern GL, Davidson RJ, et al. Neuroanatomical correlates of pleasant and unpleasant emotion. Neuropsychologia 1997b; 35: 1437–44. LeDoux JE. Emotion. In: Mountcastle VB, Plum F, editors. Handbook of physiology, Sect. 1, Vol. 5, Pt 1. Bethesda (MD): American Physiological Society; 1987. p. 419–59.
Fischer H, Wik G, Fredrikson M. Functional neuroanatomy of robbery re-experience: affective memories studied with PET. Neuroreport 1996; 7: 2081–6.
Mai JK, Assheuer J, Paxinos G. Atlas of the human brain. San Diego (CA): Academic Press; 1997.
Fox PT, Mintun MA, Raichle ME, Miezin FM, Allman JM, Van Essen DC. Mapping human visual cortex with positron emission tomography. Nature 1986; 323: 806–9.
Markowitsch HJ, Calabrese P, Wurker M, Durwen HF, Kessler J, Babinsky R, et al. The amygdala’s contribution to memory—a study on two patients with Urbach–Wiethe disease. Neuroreport 1994; 5: 1349–52.
Gabriel M. Discriminative avoidance learning. In: Vogt BA, Gabriel M, editors. Neurobiology of cingulate cortex and limbic thalamus: a comprehensive handbook. Boston (MA): Birkha¨user; 1993. p. 478–523.
Mayberg HS, Brannan SK, Mahurin RK, Jerabek PA, Brickman JS, Tekell JL, et al. Cingulate function in depression: a potential predictor of treatment response. Neuroreport 1997; 8: 1057–61.
Gaffan D, Murray EA. Amygdalar interaction with the mediodorsal nucleus of the thalamus and the ventromedial prefrontal cortex in stimulus-reward associative learning in the monkey. J Neurosci 1990; 10: 3479–93. George MS, Ketter TA, Parekh PI, Horwitz B, Herscovitch P, Post RM. Brain activity during transient sadness and happiness in healthy women. Am J Psychiatry 1995; 152: 341–51. Gloor P. Experiential phenomena of temporal lobe epilepsy: facts and hypotheses. [Review]. Brain 1990; 113: 1673–94. Groenewegen HJ, Berendse HW. The specificity of the ‘nonspecific’ midline and intralaminar thalamic nuclei. [Review]. Trends Neurosci 1994; 17: 52–8. Henkin R, Comiter H, Deio P, O’Doherty D. Defects in taste and smell recognition following temporal lobectomy. Trans Am Neurol Assoc 1977; 102: 146–50. Huang SC, Hoffman EJ, Phelps ME, Kuhl DE. Quantitation in positron emission computed tomography: 2. Effects of inaccurate attenuation correction. J Comput Assist Tomogr 1979; 3: 804–14. Irwin W, Davidson RJ, Lowe MJ, Mock BJ, Sorenson JA, Turski PA. Human amygdala activation detected with echo-planar functional magnetic resonance imaging. Neuroreport 1996; 7: 1765–9. Kennie ED, Nagel JA. Failure to form a learned taste aversion in rats with amygdaloid lesions. Bull Psychonomic Soc 1973; 2: 155–6. Ketter TA, Andreason PJ, George MS, Lee C, Gil DS, Parekh PI,
McGaugh JL, Cahill L, Roozendaal B. Involvement of the amygdala in memory storage: interaction with other brain systems. [Review]. Proc Natl Acad Sci USA 1996; 93: 13508–14. Mikulka PJ, Freeman FG, Lidstrom P. The effect of training technique and amygdala lesions on the acquisition and retention of a taste aversion. Behav Biol 1977; 19: 509–17. Minoshima S, Berger KL, Lee KS, Mintun MA. An automated method for rotational correction and centering of three-dimensional functional brain images. J Nucl Med 1992; 33: 1579–85. Minoshima S, Koeppe RA, Mintun MA, Berger KL, Taylor SF, Frey KA, et al. Automated detection of the intercommissural line for stereotactic localization of functional brain images. J Nucl Med 1993; 34: 322–9. Minoshima S, Koeppe RA, Frey KA, Kuhl DE. Anatomic standardization: linear scaling and nonlinear warping of functional brain images. J Nucl Med 1994; 35: 1528–37. Morris JS, Frith CD, Perrett DI, Rowland D, Young AW, Calder AJ, et al. A differential neural response in the human amygdala to fearful and happy facial expressions. Nature 1996; 383: 812–5. Neafsey EJ, Terreberry RR, Hurley KM, Ruit KG, Frysztak RJ. Anterior cingulate cortex in rodents: connections, visceral control functions, and implications for emotion. In: Vogt BA, Gabriel M, editors. Neurobiology of cingulate cortex and limbic thalamus: a comprehensive handbook. Boston (MA): Birkha¨user; 1993. p. 206–23.
Limbic responses to saline
1153
Nishijo H, Ono T, Nishino H. Topographic distribution of modalityspecific amygdalar neurons in alert monkey. J Neurosci 1988; 8: 3556–69.
efferents of the amygdala and the hippocampal formation in the rhesus monkey: II. Reciprocal and non-reciprocal connections. J Comp Neurol 1988; 271: 185–207.
Norgren R. Taste pathways to hypothalamus and amygdala. J Comp Neurol 1976; 166: 17–30.
Scott SK, Young AW, Calder AJ, Hellawell DJ, Aggleton JP, Johnson M. Impaired auditory recognition of fear and anger following bilateral amygdala lesions. Nature 1997; 385: 254–7.
Norgren R. Gustatory system. In: Paxinos G, editor. The human nervous system. San Diego: Academic Press; 1990. p.845–61. Ono T, Nishijo H. Neurophysiological basis of the Klu¨ver–Bucy syndrome: responses of monkey amygdaloid neurons to biologically significant objects. In: Aggleton JP, editor. The amygdala: neurobiological aspects of emotion, memory, and mental dysfunction. New York: Wiley-Liss; 1992. p. 167–90. Pandya DN, Van Hoesen GW, Mesulam MM. Efferent connections of the cingulate gyrus in the rhesus monkey. Exp Brain Res 1981; 42: 319–30. Paradiso S, Robinson RG, Andreasen NC, Downhill JE, Davidson RJ, Kirchner PT, et al. Emotional activation of limbic circuitry in elderly normal subjects in a PET study. Am J Psychiatry 1997; 154: 384–9. Pardo JV, Pardo PJ, Raichle ME. Neural correlates of self-induced dysphoria [see comments]. Am J Psychiatr 1993; 150: 713–9. Comment in: Am J Psychiatry 1994; 151: 784–5. Rauch SL, Jenike MA, Alpert NM, Baer L, Breiter HC, Savage CR, et al. Regional cerebral blood flow measured during symptom provocation in obsessive-compulsive disorder using oxygen 15labeled carbon dioxide and positron emission tomography [see comments]. Arch Gen Psychiatry 1994; 51: 62–70. Comment in: Arch Gen Psychiatry 1995; 52: 159–60. Rauch SL, Savage CR, Alpert NM, Miguel EC, Baer L, Breiter HC, et al. A positron emission tomographic study of simple phobic symptom provocation. Arch Gen Psychiatry 1995; 52: 20–8. Rauch SL, van der Kolk BA, Fisler RE, Alpert NM, Orr SP, Savage CR, et al. A symptom provocation study of posttraumatic stress disorder using positron emission tomography and script-driven imagery. Arch Gen Psychiatry 1996; 53: 380–7.
Scott TR, Karadi Z, Oomura Y, Nishino H, Plata-Salaman CR, Lenard L, et al. Gustatory neural coding in the amygdala of the alert macaque monkey. J Neurophysiol 1993; 69: 1810–20. Shields SA. Gender in the psychology of emotion: a selective research review. In: Strongman KT, editor. International review of studies on emotion. New York: Wiley; 1991. p. 227–45. Shin LM, Kosslyn SM, McNally RJ, Alpert NM, Thompson WL, Rauch SL, et al. Visual imagery and perception in posttraumatic stress disorder: a positron emission tomographic investigation. Arch Gen Psychiatry 1997; 54: 233–41. Silbersweig DA, Stern E, Frith CD, Cahill C, Schnorr L, Grootoonk S, et al. Detection of thirty-second cognitive activations in single subjects with positron emission tomography: A new lowdose H215O regional cerebral blood flow three-dimensional imaging technique. J Cereb Blood Flow Metab 1993; 13: 617–29. Small DM, Jones-Gotman M, Zatorre RJ, Petrides M, Evans AC. A role for the right anterior temporal lobe in taste quality recognition. J Neurosci 1997a; 17: 5136–42. Small DM, Jones-Gotman M, Zatorre RJ, Petrides M, Evans AC. Flavor processing: more than the sum of its parts. Neuroreport 1997b; 8: 3913–7. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain. Stuttgart: Thieme; 1988. Thorpe SJ, Rolls ET, Maddison S. The orbitofrontal cortex: neuronal activity in the behaving monkey. Exp Brain Res 1983; 49: 93–115. Turner B, Herkenham M. An autoradiographic study of thalamoamygdaloid connections in the rat [abstract]. Anat Rec 1981; 199: 260A.
Ray JP, Price JL. The organization of projections from the mediodorsal nucleus of the thalamus to orbital and medial prefrontal cortex in macaque monkeys. J Comp Neurol 1993; 337: 1–31.
Turner BH, Mishkin M, Knapp M. Organization of the amygdalopetal projections from modality-specific cortical association areas in the monkey. J Comp Neurol 1980; 191: 515–43.
Reiman EM, Lane RD, Ahern GL, Schwartz GE, Davidson RJ, Friston KJ, et al. Neuroanatomical correlates of internally generated human emotion. Am J Psychiatry 1997; 154: 918–25.
Ursin H, Rosvold HE, Vest B. Food preference in brain lesioned monkeys. Physiol Behav 1969; 4: 609–12.
Rolls ET. Information processing in the taste system of primates. [Review]. J Exp Biol 1989; 146: 141–64.
Van Hoesen GW, Morecraft RJ, Vogt BA. Connections of the monkey cingulate cortex. In: Vogt BA, Gabriel M, editors. Neurobiology of cingulate cortex and limbic thalamus: a comprehensive handbook. Boston (MA): Birkha¨user; 1993. p. 249–84.
Rolls ET, Baylis LL. Gustatory, olfactory, and visual convergence within the primate orbitofrontal cortex. J Neurosci 1994; 14: 5437–52. Rolls ET, Yaxley S, Sienkiewicz ZJ. Gustatory responses of single neurons in the caudolateral orbitofrontal cortex of the macaque monkey. J Neurophysiol 1990; 64: 1055–66. Rousseaux M, Muller P, Gahide I, Mottin Y, Romon M. Disorders of smell, taste, and food intake in a patient with a dorsomedial thalamic infarct. Stroke 1996; 27: 2328–30. Saunders RC, Rosene DL, Van Hoesen GW. Comparison of the
Vogt BA, Pandya DN. Cingulate cortex of the rhesus monkey: II. Cortical afferents. J Comp Neurol 1987; 262: 271–89. Woods RP, Cherry SR, Mazziotta JC. Rapid automated algorithm for aligning and reslicing PET images. J Comput Assist Tomogr 1992; 16: 620–33. Worsley KJ, Evans AC, Marrett S, Neelin P. A three-dimensional statistical analysis for CBF activation studies in human brain [see comments]. J Cereb Blood Flow Metab 1992; 12: 900–18. Comment in: J Cereb Blood Flow Metab 1993; 13: 1040–2.
1154
D. H. Zald et al.
Yasoshima Y, ShimuraT, Yamamoto T. Single unit responses of the amygdala after conditioned taste aversion in conscious rats. Neuroreport 1995; 6: 2424–8.
Zald DH, Kim, SW. Anatomy and function of the orbital frontal cortex, II. Function and relevance to obsessive-compulsive disorder. J Neuropsychiatry Clin Neurosci 1996b; 8: 249–61.
Yasui Y, ltoh M, Mizuno N. Projections from the parvocellular part of the posteromedial ventral nucleus of the thalamus to the lateral amygdaloid nucleus in the cat. Brain Res 1984; 292: 151–5.
Zald DH, Kim SW. The orbitofrontal cortex. In: Salloway S, Malloy P, editors. The frontal lobes and neuropsychiatric illness. Washington (DC): American Psychiatric Press 1998. In press.
Yaxley S, Rolls ET, Sienkiewicz ZJ. Gustatory responses of single neurons in the insula of the macaque monkey. J Neurophysiol 1990; 63: 689–700.
Zald DH, Pardo JV. Emotion, olfaction, and the human amygdala: amygdala activation during aversive olfactory stimulation. Proc Natl Acad Sci USA 1997; 94: 4119–24.
Young AW, Hellawell DJ, Van De Wal C, Johnson M. Facial expression processing after amygdalotomy. Neuropsychologia 1996; 34: 31–9.
Zald DH, Donndelinger MJ, Pardo JV. Elucidating dynamic brain interactions with across-subjects correlational analyses of PET data: the functional connectivity of the amygdala and orbitofrontal cortex during olfactory tasks. J Cereb Blood Flow Metab 1998. In press.
Zald DH, Kim, SW. The anatomy and function of the orbital frontal cortex, I. Anatomy, neurocircuitry, and obsessive-compulsive disorder. J Neuropsychiatry Clin Neurosci 1996a; 8: 125–38.
Received September 11, 1997. Revised December 26, 1997. Accepted January 21, 1998
