Pii: s0006-3223(02)01396-3

Cerebral Blood Flow during Anticipation of PublicSpeaking in Social Phobia: A PET Study Maria Tillfors, Tomas Furmark, Ina Marteinsdottir, and Mats Fredrikson Background: The aim was to examine the neural corre-
ously been investigated in animal phobics (Drevets et al lates of anxiety elicited by the anticipation of public 1995; Wik et al 1996) and normal healthy volunteers speaking in individuals with social phobia. (Benkelfat et al 1995; Chua et al 1999; Reiman 1997; Methods: Positron emission tomography and 15O-water
Reiman et al 1989). In anticipation of fear-provoking was used to measure regional cerebral blood flow in stimuli, animal phobics exhibited rCBF decreases in mo- subjects with DSM-IV defined social phobia during anxi- dality-specific primary sensory areas, while elevated per- ety anticipation. Heart rate and subjective anxiety were fusion was not observed (Drevets et al 1995; Wik et al also recorded. While being scanned, subjects were speak- 1996). In healthy, normal individuals, the anterior cingu- ing alone either before or after speaking in public. To late, insula, temporal, prefrontal, and the orbitofrontal evaluate anticipatory anxiety we compared individuals cortices together with the thalamus and the cerebellum speaking alone before they were speaking in front of anaudience with those who did the reverse. have been associated with an increased perfusion duringanticipatory anxiety (Benkelfat et al 1995; Chua et al Results: Heart rate and subjective anxiety measures
1999; Reiman 1997; Reiman et al 1989). In contrast with confirmed anticipatory anxiety in social phobics whoperformed their private speech before their public. This the animal phobics, rCBF decreases were not associated was accompanied by enhanced cerebral blood flow in the with anticipatory anxiety in healthy normals. Neural ac- right dorsolateral prefrontal cortex, left inferior temporal tivity assessed with the electroencephalogram (EEG) has cortex, and in the left amygdaloid-hippocampal region. also been reported to change in social phobics during Brain blood flow was lower in the left temporal pole and anticipation of a public speech (Davidson et al 2000a).
bilaterally in the cerebellum in the anticipation group. These authors observed a right-sided activation in the Conclusions: Brain regions with altered perfusion pre-
anterior temporal and lateral prefrontal scalp regions.
sumably reflect changes in neural activity associated with Individuals with social phobia worry excessively (e.g., worry about anticipated public performance. We speculate Clark 1997) and because the hallmark of anticipatory that anticipatory anxiety in social phobics originates in an anxiety is worry (Barlow et al 1996), it would be of affect sensitive fear network encompassing the amygdal- interest to explore the neural correlates of anticipatory oid-hippocampal region, prefrontal, and temporal areas. anxiety in social phobia (social anxiety disorder). Thus, Biol Psychiatry 2002;52:1113–1119 2002 Society of the aim of the present study was to examine the neural correlates of anxiety elicited by the anticipation of a publicspeaking task in individuals with social phobia. Recently, Key Words: Anticipatory anxiety, social phobia, public
we examined brain activity during symptom provocation speaking, induced anxiety, positron emission tomography,regional cerebral blood flow in social phobics and healthy individuals while speaking infront of an audience and alone (Tillfors et al 2001). Toevaluate anticipatory anxiety, we compared the social Introduction
phobics who were speaking alone before they were speak-ing in front of an audience (i.e., the anticipation group) Anticipatory anxiety is characterized by worry about with those who were speaking alone after they had been future events (Barlow et al 1996). Regional cerebral speaking in front of an audience (i.e., the comparison blood flow (rCBF) during anticipatory anxiety has previ- During social anxiety provocation, rCBF increased From the Department of Social Sciences (MT), O more in the amygdaloid complex in those with social Department of Psychology (TF, MF), Uppsala University, Uppsala; Depart- phobia than in healthy nonphobic individuals (Tillfors et al ment of Neuroscience, Psychiatry (IM), Uppsala University, Uppsala; and theUppsala University PET Center (MF), Uppsala University, Uppsala, Sweden.
2001). Because numerous previous studies (see Davidson Address reprint requests to Dr. M. Tillfors, Orebro University, Department of and Irwin 1999 and Davidson et al 2000b for a review) Social Sciences, SE-70182 Orebro, Sweden.
Received July 25, 2001; revised March 6, 2002; accepted March 15, 2002.
also have identified the amygdaloid complex to play a crucial role in the processing of fear and anxiety, we Trait Anxiety Inventory (STAI-T) (Spielberger et al 1983). Four performed a hypothesis driven search for amygdala acti- unpaired t tests performed in Statview 4.5 for Macintosh were vation in addition to a whole-brain search using traditional used to evaluate possible group differences in trait anxiety.
Methods and Materials
Subjective and physiologic measures of fear and anxiety wereanalyzed with four directional unpaired t tests of significance performed in Statview 4.5 for Macintosh (SAS Institute Inc., This study was part of a larger investigation of the functional Cary, NC). A one-tailed alpha level of 0.05 was applied.
neuroanatomy of social phobia before and after treatment withcognitive-behavioral therapy and citalopram. The methodologyhas been described in detail elsewhere (Furmark et al in press; Tillfors et al 2001). Briefly, 18 right-handed previously untreatedpatients (10 men and 8 women) with DSM-IV social phobia with All individuals with social phobia were given topics of a neutral, a mean age (Ϯ SD) of 35.2 (Ϯ 7.3) years (range 23– 46) who did nonemotional content—i.e., they spoke about a travel experience not medicate were recruited. Criteria for exclusion were: previ- or a vacation. On a random basis, 20 sec extract from each ous or current organic brain disorders and somatic diseases, delivered speech was transformed to text and counted. The current comorbid Axis I disorders, menopause, and pregnancy.
counting words were analyzed with an unpaired t test of Screening included psychiatric (structured clinical diagnostic significance performed in Statview 4.5 for Macintosh.
interviews, SCID; First et al 1998) and medical histories as wellas trait measures of anxiety (Furmark et al in press). SCID-interviews were performed by an experienced psychiatrist. Par- ticipants refrained from tobacco, alcohol and caffeine 12 hoursbefore the PET investigation. The Uppsala University Medical SCANNING. Investigations were performed on a GEMS
Faculty Ethical Review Board and the Uppsala University PC2048 –15B scanner (General Electrics, Milwaukee, WI) with a Isotope Committee approved the study. Informed consent was 10-cm axial field of view (Holte et al 1989), producing 15 slices obtained after the procedure had been fully explained.
with a 6.5 mm slice spacing and a 6 mm axial/transaxialresolution. A venous catheter was inserted and the subjects werefixated in the scanner using a commercial headholder with fast hardening foam. The participants were positioned in the scanner Subjects lay in the scanner with their eyes open during all tasks.
such that the planes were parallel to a line between the anterior During scanning, subjects were either speaking in the presence of and posterior commissure where the most basal slice was below a surrounding audience (6 – 8 silently observing individuals) or the orbita. First, a 10-min transmission scan using a rotating alone during 2.5 min. About 20 min before the initial scan, 68germanium pin source was completed. During the end of the subjects were instructed to prepare a 2.5 min speech about a transmission scan, each subject was given a saline injection to travel experience or a vacation. Each task was repeated and attenuate novelty effects. Before each emission scan, 700 –1300 scanned twice with the order counterbalanced between subjects.
Mbq of 15O-water in 3– 4 mL of water (approximately 15 To evaluate anticipatory anxiety related to social fear we com- Mbq/kg body weight) was injected with 12-min intervals. The pared the nine social phobics who were speaking alone before subjects were told to start speaking immediately following they were speaking in front of an audience (i.e., the anticipation injections and data were collected in fifteen 10-sec frames. Each group) with the nine who were speaking alone after they had subject had a total of four emission scans during two conditions been speaking in front of an audience (i.e., the comparison presented in counterbalanced order between subjects. Finally, an group). Men and women were equally distributed in the two additional injection was performed when the scanner-bed was groups. Scores on Spielberger’s State Anxiety Inventory automatically moved back and forth between two positions 10 (STAI-S; 20 – 80) (Spielberger et al 1983) and ratings of fear and cm apart. Data were collected only at these positions and never distress, ranging from 0 –100, were obtained immediately after while the scanner was moving. This last interleaved scan was each scan. During each scan heart rate was recorded with the performed to obtain a scan with full axial coverage of the brain, PSYLAB6 integrated system for psychophysiology (http//:www.
which aids in the stereotactical normalization of the PET-images.
IMAGE RECONSTRUCTION. Data from the first 70 sec after
arrival of the bolus to the brain were summed. Images were reconstructed from the summation after correction for attenua- The trait anxiety questionnaires included the Social Phobia Scale tion and scatter by using the transmission scan (Bergstro¨m et al (SPS) (Mattick and Clarke 1998), the Social Interaction Anxiety 1983). The additional interleaved scan was reconstructed to a Scale (SIAS) (Mattick and Clarke 1998), the Personal Report of 30-slice image set with a 20 cm coverage in axial (superior to Confidence as a Speaker (PRCS) (Paul 1966) and Spielberger’s inferior) direction (Bergstro¨m et al 1982). All images were Figure 1. Mean values and standard deviations for subjective measures of state anxiety and heart rate during speaking alone before (i.e.,the anticipation group) or after (i.e., the comparison group) speaking in public for 18 individuals with social phobia.
reconstructed to a 128 ϫ 128 matrix with a pixel size of 2 mm ANATOMICAL STANDARDIZATION. Normalization of all
individual CBF images into a standard brain shape (Greitz et al The two groups did not differ in trait anxiety as they had 1991) was performed automatically by matching the scan with similar scores on the SPS (t Ͻ 1, df ϭ 16, ns), the SIAS 20-cm axial coverage to an atlas template (Andersson and (t Ͻ 1, df ϭ 16, ns), the PRCS (t Ͻ 1, df ϭ 16, ns) and the Thurfjell 1997). The images from the four emission scans were STAI-T (t Ͻ 1, df ϭ 16, ns).
automatically aligned to the 20-cm scan (Andersson 1995),bringing them into the stereotactic space and correcting for head movements between scans. The stereotactic space was definedbased on the postmortem slicing of a single subject in which The anticipation group as compared with the comparison brain contours, gyri, sulci, central structures, and Brodmann group displayed higher heart rate (t ϭ 2.5, df ϭ 16, p Ͻ areas were defined (Greitz et al 1991). The cerebral brain atlas .01) and state anxiety ratings (STAI-S: t ϭ 1.8, df ϭ 16, software (Thurfjell et al 1995) allows for identification, in terms p Ͻ .05), but did not differ in fear (t Ͻ 1, df ϭ 16, ns) or of anatomy, cytoarchitecture, and Talairach coordinates (Ta- distress (t Ͻ 1, df ϭ 16, ns) (See Figure 1).
BLOOD FLOW RECORDINGS. An ANOVA (Friston et al
1995) with one between group factor (the anticipation group vs.
the comparison group) was used. To account for between subject When counting words it was found that there were no variance and to evaluate anticipatory anxiety we only used the significant difference between the anticipation group and first of the two normalized rCBF images from the speaking alone the comparison group (t ϭ Ϫ.93; df ϭ 16; ns).
condition to be compatible to a random effects model (Peterssonet al 1999). Global flow was estimated using a predefined mask outlining the brain, but excluding all voxels which changed as aconsequence of study conditions using F-map masking (Anders- The state related anxiety reaction associated with antici- son 1997). PET data were normalized for global flow using linear pation was accompanied by enhanced rCBF in the right scaling (Andersson 1997). The contrast generated a t map that dorsolateral prefrontal and left inferior temporal cortices, was subsequently converted to a z score map through a proba- as well as in the left amygdaloid-hippocampal region (x ϭ bility preserving transformation (Friston et al 1991). The signif- Ϫ27, y ϭ Ϫ13, z ϭ Ϫ13; maximum z value ϭ 2.58, p ϭ icance of the z score maps was evaluated at an omnibus level .005 uncorrected for multiple comparisons) (See Figure 2 using the mean square z score (Worsley et al 1995). Local changes were evaluated using the spatial extent of connected Brain blood flow was lowered in the left temporal pole clusters of voxels with a z score above 2.6 (Friston et al 1994).
This test takes into account multiple comparisons and has a and bilaterally in the cerebellum in the anticipation as cluster-localizing power (Friston et al 1996). The directed search compared with the comparison group (See Table 1).
for amygdala activation was performed by comparing the two Because the design is unbalanced in that individuals in groups using a z threshold of 2.58 corresponding to an uncor- the anticipation group always performed their private rected (one tailed) p of .005.
speech first, we performed a linear trend analysis to 43, z ϭ 32). Thus, task repetition did not influence rCBFin structures associated with anticipatory anxiety except inthe prefrontal cortex where rCBF was attenuated by taskrepetition, but elevated as a function of anticipatoryanxiety. Anticipatory related rCBF-increases were ob-served in the dorsolateral prefrontal cortex while repetitionrelated decreases were located in the prefrontal cortexsuperior to the anticipatory cluster. Because the oppositedirection of change and since the centers of gravity ofthese clusters were separated by 28 mm, it is most likelythat anticipatory anxiety rather than task repetition ac-counts for the observed rCBF alterations in the rightdorsolateral prefrontal cortex.
Discussion
The aim of the present study was to explore the functionalneuroanatomy of anticipatory anxiety in social phobics,elicited by the anticipation of speaking in front of anaudience. Heart rate and subjective anxiety measuresconfirmed anticipatory anxiety in social phobics who weredelivering their private speech before their public (i.e., theanticipation group) as compared with those who had thereverse order (i.e., the comparison group). The increased Figure 2. Enhanced normalized relative regional cerebral blood heart rate and subjective anxiety were associated with flow (rCBF) in the left amygdaloid-hippocampal region in social elevated rCBF in the right dorsolateral prefrontal and left phobics speaking alone before (i.e., the anticipation group) inferior temporal cortices as well as in the left amygdal- compared with after (i.e., the comparison group) speaking in oid-hippocampal region. Regional CBF was lower in the left temporal pole and bilaterally in the cerebellum in theanticipation group. This neural pattern probably reflects evaluate order effects using rCBF data from all conditions emotional processes since the speaking tasks were identi- (n ϭ 4) and subjects (n ϭ 18). As a function of repeated cal for both groups but associated with more anxiety in the performances rCBF increased bilaterally in the retrosple- anticipation than in the control group. Furthermore, the nial area (maximum z value ϭ 4.1; x ϭ Ϫ1, y ϭ Ϫ37, z present pattern most likely can be attributed to anticipa- ϭ 18), in the right parietal cortex (maximum z value ϭ tion, even though we cannot rule out the possibility that 3.8; x ϭ 34, y ϭ Ϫ51, z ϭ 35) and in the left motor cortex the enhanced anxiety in the anticipation group could be an (maximum z-value ϭ 2.6; x ϭ Ϫ29, y ϭ Ϫ12, z ϭ Ϫ12).
effect attributed to relief in the control group. Be that as it Attenuated rCBF was observed in the right superior may, it still reflects cognitive expectancy related pro- prefrontal cortex (maximum z value ϭ 4.7; x ϭ 37, y ϭ cesses. It is not likely that the rCBF alterations are due to Table 1. Relative rCBF in Social Phobics as a Function of Anticipation of Public Speaking R Dorsolateral prefrontal cortex (46a) Brain areas, Talairach coordinates, and maximum voxel z value for significant rCBF differences. The coordinates in millimeters correspond to the stereotactic atlas of Talairach & Tournoux (1988). rCBF, regional cerebral blood flow; R, right hemisphere; L, left hemisphere.
aCorresponding Brodmann area.
bUncorrected for multiple comparisons because of an a` priori hypothesis.
anatomical differences between the two groups since all In previous PET-studies on anticipatory anxiety altered individual CBF images were normalized into a standard neural activity has been observed in modality-specific atlas using the computerized brain atlas software (Thur- primary sensory areas in animal phobics (Drevets et al fjell et al 1995). Because the two conditions were not 1995; Wik et al 1996), and in the anterior cingulate, insula, presented in counterbalanced order we evaluated order temporal, prefrontal and orbitofrontal cortices as well as effects by means of a trend analysis sensitive to order the thalamus and the cerebellum in healthy normal indi- effects. Task repetition per se did not influence rCBF in viduals (Benkelfat et al 1995; Chua et al 1999; Reiman areas associated with anticipatory anxiety except in the 1997; Reiman et al 1989). Thus, the findings of alterations prefrontal cortex. Regional CBF in the right dorsolateral in the prefrontal and temporal cortices as well as in the prefrontal cortex increased with anxiety, whereas it de- cerebellum, in the present study, partly overlap with the creased with repetition in the right prefrontal cortex neuroimaging studies on anticipatory anxiety in healthy superior to the above cluster. This suggests a functional individuals, but not in animal phobics. While there is an segregation within the frontal cortex between processes obvious modality specificity in specific phobia it is more related to anticipatory anxiety and task repetition.
difficult to determine a specific stimulated modality in Anxiety related rCBF alterations in the amygdala are social phobia and also in anticipatory anxiety in healthy consistent with previous studies identifying the amygdala volunteers. The divergent results may mirror this differ- as important for negative affect (e.g., Davidson and Irwin ence. Finally, attenuated rCBF in the temporal pole is 1999; Davidson et al 2000b; Davis and Whalen 2001) similar to that observed during symptom provocation in including social anxiety (Furmark et al in press; Tillfors et social phobia (Tillfors et al 2001), specific phobia al 2001) and dispositional negative affect (Fischer et al (Fredrikson et al 1995) and self-generated fear (Damasio 2001; Abercrombie et al 1998). Further, paradigms like et al 2000) as well as shock-induced unexpected panic classical conditioning have also reported amygdala acti- vation during anticipation of an aversive outcome in The present study is consistent with Davidson and individuals with social phobia (Birbaumer et al 1998).
colleagues’ (2000a) observations of right prefrontal EEG Regional CBF alterations in the right prefrontal cortex activation in social phobics during anticipatory anxiety.
accompany memory retrieval (see e.g., Cabeza and Ny- Consequently, it seems that the neural patterns of antici- berg 2000 for a review). The prefrontal cortex has also patory anxiety in social phobics and healthy individualsare relatively similar and might involve activation of been suggested to participate in the conscious experience affective working memory (e.g., Davidson and Irwin of emotion (e.g., Lane et al 1997; Reiman 1997). Because 1999); however, the imaging data on anticipatory anxiety anticipatory anxiety is characterized by worry about future in healthy individuals (Benkelfat et al 1995; Chua et al events but may also activate memories of the past, we 1999; Reiman 1997; Reiman et al 1989) mainly report speculate that the enhanced perfusion in the right dorso- increased perfusion whereas in the present study both lateral prefrontal cortex reflects affective working mem- enhanced and attenuated rCBF were present during antic- ory. According to Davidson et al (2000b), this process is ipatory anxiety related to social fears. These differences in critical when an individual is anticipating future affective directions of change could represent different functional outcomes. Further support for the importance of the roles in individuals with and without pathologic anxiety, prefrontal cortex in anticipatory anxiety could be in- or reflect the differential effect of various anxiety induc- creased right-sided prefrontal activation found at rest in tion techniques. Yet, some students of emotion have patients with generalized anxiety disorder, a syndrome argued that the direction of change is less important than characterized by excessive worry (Wu et al 1991), and the location of the change because irrespective of direction right-sided prefrontal EEG activations observed during alterations may signify regional engagement (e.g., Malizia anticipation of public speaking in social phobics (David- 1999). If the latter is the case, similar areas seem engaged son et al 2000a). In addition, the present results with rCBF by anticipation in normal healthy individuals and social alterations in the right prefrontal cortex and the amygdal- phobics. One distinctive region separating “normal” from oid-hippocampal complex are consistent with lesion and “pathologic” anxiety seems to be the amygdala because it neuroimaging data where emphasis has been placed on these does not appear to be engaged by anticipatory anxiety in structures as key components forming part of the circuitry of normal individuals, but only in social phobics. An expla- negative emotion (e.g., Davidson and Irwin 1999; Davidson nation could be that both anticipatory and situationally et al 2000b). Hence, anticipatory anxiety in social phobics (Tillfors et al 2001) elicited anxiety in social phobics may be mediated by an affect sensitive neural system originates in an affect sensitive fear network centered in characterized by a reciprocal relation between the prefron- the amygdaloid-hippocampal region and involving inter- tal cortex and the amygdaloid-hippocampal area.
action with the prefrontal cortex. This account is in line with Gorman et al (1989; 2000) who proposed that References
anticipatory anxiety arises after kindling of limbic struc- Abercrombie HC, Schaefer SM, Larson CL, Oakes TR, Lindgren tures such as the amygdala and the hippocampus.
KA, Holden JE, et al (1998): Metabolic rate in the right The right lateralized activation in the amygdaloid- amygdala predicts negative affect in depressed patients.
hippocampal region in situationally elicited anxiety re- ported by Tillfors and coworkers (2001) as compared with Andersson JLR (1995): A rapid and accurate method to realign the left lateralized activation during anxiety anticipation PET scans utilizing image edge information. J Nucl Med observed in the present study may concur with studies trying to distinguish between type of anxiety. Panic as Andersson JLR (1997): How to estimate global activity indepen- compared with worry seems associated with an asymmetry dent of changes in local activity. Neuroimage 60:237–244.
in favor of the right hemisphere, while a left lateralized Andersson JLR, Thurfjell L (1997): Implementation and valida- pattern seems associated with worry (Heller et al 1997).
tion of a fully autonomic system for intra- and interindividualregistration of PET brain scans. J Comput Assist Tomogr This notion is also generally consistent with neuroimaging studies reporting that noncognitive processes like implicit Barlow DH, Chorpita BF, Turovsky J (1996): Fear, panic, emotional memory recall activates the right amygdala anxiety, and disorders of emotion. In: Dienstbier RA, Hope (Morris et al 1998; Rauch et al 2000), whereas more DA, editors. Nebraska Symposium on Motivation. Volume cognitively related processes like explicit emotional mem- 43. Perspectives on Anxiety, Panic and Fear. Lincoln: Uni- ory recall activates the left amygdala (Dolan et al 2000; versity of Nebraska Press, pp 251–328.
Morris et al 1998; Phelps et al 2001). Even though this Benkelfat C, Bradwejn J, Meyer E, Ellenbogen BA, Milot S, suggests laterality differences it should be noted that not Gjedde A, et al (1995): Functional neuroanatomy of CCK - all the neuroimaging data reporting asymmetries have induced anxiety in normal healthy volunteers. Am J Psychi-atry 152:1180 –1184.
rigorously analyzed the interaction of condition withhemisphere, and hence, must be regarded with caution.
Bergstro¨m M, Eriksson L, Bohm C, Blomqvist G, Litton J (1983): Correction for scattered radiation in a ring detector Enhanced rCBF in the dorsolateral prefrontal cortex, positron camera by integral transformations of the projec- inferior temporal cortex and the amygdaloid-hippocampal tions. J Comput Assist Tomogr 7:42–50.
region as well as attenuated rCBF in the temporal pole, as Bergstro¨m M, Litton J, Eriksson L, Bohm C, Blomqvist G a function of anxiety anticipation, may reflect worry about (1982): Determination of object contour from projections for anticipated public performance. One explanation could be attenuation correction in cranial positron emission tomogra- that when anticipating public speaking the amygdaloid- phy. J Comput Assist Tomogr 6:365–372.
hippocampal region is activated. Previous research has Birbaumer N, Grodd W, Diedrich O, Klose U, Erb M, Lotze M, suggested that the orbitofrontal/ventromedial prefrontal et al (1998): FMRI reveals amygdala activation to human cortex exerts inhibitory control of the amygdala (e.g., faces in social phobics. Neuroreport 9:1223–1226.
Davidson et al 2000b; Davidson et al 2000c). One possible Cabeza R, Nyberg L (2000): Imaging cognition II: An empirical explanation may be that increased worry and rumination, review of 275 PET and fMRI studies. J Cogn Neurosci 12:1– reflected in dorsolateral prefrontal activation, modulates the orbitofrontal/ventromedial prefrontal cortex resulting Chua P, Krams M, Toni I, Passingham R, Dolan R (1999): A in reduced inhibition of the amygdaloid complex. In social functional anatomy of anticipatory anxiety. Neuroimage9:563–571.
phobics, the amygdaloid complex seems highly sensitive(Schneider et al 1999; Tillfors et al 2001) and an anxiety Clark DM (1997): Panic disorder and social phobia. In: Clark DM, Fairburn CG, editors. Science and Practice of Cognitive reaction may be more easily elicited by decreased orbito- Behavior Therapy. Oxford: Oxford University Press, pp 121– frontal/ventromedial inhibition than in nonanxious indi- viduals. To conclude, we speculate that anticipatory anx- Damasio AR, Grabowski TJ, Bechara A, Damasio H, Ponto iety is mediated by an affect sensitive functional system LLB, Parvizi J, et al (2000): Subcortical and cortical brain encompassing the amygdaloid-hippocampal area and the activity during the feeling of self-generated emotions. Nature prefrontal cortex possibly related both to state and trait Davidson RJ, Irwin W (1999): The functional neuroanatomy of emotion and affective style. Trends Cognit Sci 3:11–20.
Davidson RJ, Jackson DC, Kalin NH (2000b): Emotion, plastic- ity, context, and regulation: Perspectives from affective neu-roscience. Psychol Bull 126:890 –909.
Supported by the Swedish Council for Research in the Humanities and Davidson RJ, Marshall JR, Tomarken AJ, Henriques JB (2000a): Social Sciences, the Uppsala University, and the Bank of Sweden While a phobic waits: Regional brain electrical and auto- Tercentenary Foundation. We are grateful to the staff of the Uppsala nomic activity in social phobics during anticipation of public University PET-center for providing excellent research conditions.
speaking. Biol Psychiatry 47:1–14.
Davidson RJ, Putnam KM, Larson CL (2000c): Dysfunction in Lane RD, Reiman EM, Ahern GL, Schwartz GE, Davidson RJ the neural circuitry of emotion regulation—A possible pre- (1997): Neuroanatomical correlates of happiness, sadness, lude to violence. Science 289:591–594.
and disgust. Am J Psychiatry 154:926 –933.
Davis M, Whalen PJ (2001): The amygdala: Vigilance and Malizia AL (1999): What do brain imaging studies tell us about emotion. Mol Psychiatry 6:13–34.
anxiety disorders? J Psychopharmacol 13:372–378.
Dolan RJ, Lane R, Chua P, Fletcher P (2000): Dissociable Mattick RP, Clarke JC (1998): Development and validation of temporal lobe activations during emotional episodic memory measures of social phobia scrutiny fear and social interaction retrieval. Neuroimage 11:203–209.
anxiety. Behav Res Ther 36:455–470.
Drevets WC, Burton H, Videen TO, Snyder AZ, Simpson JR, ¨ hman A, Dolan RJ (1998): Conscious and uncon- Raichle ME (1995): Blood flow changes in human somato- scious emotional learning in the human amygdala. Nature First MB, Gibbon M, Spitzer RL, Williams JBW (1998): SCID-I Paul G (1966): Insight vs. Desensitization in Psychotherapy. and SCID-II:Interview protocol (in Swedish). Stockholm: Stanford: Stanford University Press.
Petersson KM, Nichols TE, Poline J-B, Holmes AP (1999): Fischer H, Anderson JLR, Furmark T, Fredrikson M (1998): Statistical limitations in functional neuroimaging I: Non- Brain correlates of an unexpected panic attack: A human inferential methods and statistical models. Philos Trans R Soc positron emission tomographic study. Neurosci Lett 251:137– Lond B Biol Sci 354:1239 –1260.
Phelps EA, O’Connor KJ, Gatenby JC, Gore JC, Grillon C, Davis Fischer H, Tillfors M, Furmark T, Fredrikson M (2001): Dispo- M (2001): Activation of the left amygdala to a cognitive sitional pessimism and amygdala activity: A PET study in representation of fear. Nature Neurosci 4:437–441.
healthy volunteers. Neuroreport 12:1635–1638.
Rauch SL, Whalen PJ, Shin LM, McInerney SC, Macklin ML, Fredrikson M, Wik G, Annas P, Ericson K, Stone-Elander S Lasko NB, et al (2000): Exaggerated amygdala response to (1995): Functional neuroanatomy of visually elicited simple masked facial stimuli in posttraumatic stress disorder: A phobic fear: Additional data and theoretical analysis. Psycho- functional MRI study. Biol Psychiatry 47:769 –776.
Reiman EM (1997): The application of positron emission tomog- Friston KJ, Frith CD, Liddle PF, Frackowiak RSJ (1991): raphy to the study of normal and pathologic emotions. J Clin Comparing functional (PET) images: The assessment of Psychiatry 58(suppl 16):4 –12.
significant change. J Cereb Blood Flow Metab 11:690 –699.
Reiman EM, Fusselman MJ, Fox PT, Raichle ME (1989): Friston KJ, Holmes A, Poline JB, Price CJ, Frith CD (1996): Neuroanatomical correlates of anticipatory anxiety. Science Detecting activations in PET and fMRI: Levels of inferenceand power. Neuroimage 40:223–235.
Friston KJ, Holmes AP, Worsley KJ, Poline J-B, Frith CD, Schneider F, Weiss U, Kessler C, Mu¨ller-Ga¨rtner H-W, Posse S, Frackowiak RSJ (1995): Stastical parametric maps in func- Salloum JB, et al (1999): Subcortical correlates of differential tional imaging: A general linear approach. Human Brain classical conditioning of aversive emotional reactions in social phobia. Biol Psychiatry 45:863–871.
Friston KJ, Worsley KJ, Frackowiak RSJ, Mazziotta JC, Evans Spielberger CD, Gorsuch RL, Lushene R, Vagg PR, Jacobs GA AC (1994): Assessing the significance of focal activations (1983): Manual for the State-trait Anxiety Inventory. Palo using their spatial extent. Human Brain Mapp 1:210 –220.
Alto, CA: Consulting Psychologists Press.
Furmark T, Tillfors M, Marteinsdottir I, Fischer H, Pissiota A, Talairach J, Tournoux P (1988): Co-Planar Stereotaxic Atlas of La˚ngstro¨m B, et al (2002): Common brain targets for seroto- the Human Brain. New York: Thieme Medical.
nergic drug therapy and cognitive-behavioral treatment of Thurfjell L, Bohm C, Bengtsson E (1995): CBA - An atlas-based social anxiety. Arch Gen Psychiatry, in press.
software tool used to facilitate the interpretation of neuroim- Gorman JM, Kent JM, Sullivan GM, Coplan JD (2000): Neuro- aging data. Comput Methods Programs Biomed 47:51–71.
anatomical hypothesis of panic disorder, revised. Am JPsychiatry 157:493–505.
Tillfors M, Furmark T, Marteinsdottir I, Fischer H, Pissiota A, Långstro¨m B, et al (2001): Cerebral blood flow in social Gorman JM, Liebowitz MR, Fyer AJ, Stein J (1989): A neuro- phobics during stressful speaking tasks: A PET-study. Am J anatomical hypothesis for panic disorder. Am J Psychiatry Greitz T, Bohm G, Holte S, Eriksson LA (1991): A computerized Wik G, Fredrikson M, Fischer H (1996): Cerebral correlates of brain atlas: Construction, anatomical content, and some ap- anticipated fear: A PET study of specific phobia. Int J Neu- plications. J Comput Assist Tomogr 15:26 –38.
Heller W, Nitschke JB, Etienne MA, Miller GA (1997): Patterns Worsley KJ, Poline JB, Vandal AC, Friston KJ (1995): Tests for of regional brain activity differentiate types of anxiety. J distributed, nonfocal brain activations. Neuroimage 2:183– Abnorm Psychol 106:376 –385.
Holte S, Eriksson L, Dahlbom M (1989): A preliminary evalu- Wu JC, Buchsbaum MS, Hershey TG, Hazlett E, Sicotte N, ation of the Scanditronix PC2048 –15B brain scanner. Euro Johnson C (1991): PET in generalized anxiety disorder. Biol

Source: http://www-ang.kfunigraz.ac.at/~schulter/se04_social_phobia.pdf