336 PHAN ET AL. While th e may not be a particular brain both emotion without Cognitive De ation of th flect that ntral MPEC 1 nd 2C).Inte tain a ects may be shared across differ nt emotional vious meta-analyse sof cognition revealed that the tasks.Figure 2A shows that the MPFC was activated rostral-ventral and orbital regions of the MPFC are across multiple individual emotions (four of five spe argely insensitive to cognitive tasks (Duncan and cific emoti ons in at 40%oI stu Owen,2000:Cabeza and Nyberg.2000). he MPEC a to st that the MPFC 2.Regions Associated with Individual Emotions may have a general role in emotional processing.as 2.1 Fear and the amygdala.Specifically,fear induc suggested by Lane. Reiman and colleagues,whe tion had a strong association with the amygdala.Sixty ported that emotional films,pictures and recall percen ed t em y9 12 nd the of these th he is res ponsible for detecting. generating.and maintain- Lane et al 1997c:Re an et al1gg7刀This is co tent with the notion that a number of processes are potentially various emotional tasks (e.g. s(Adolphs et al. der et al. 1996 appraisa n吧 ter et a (D 100 (1997)did find the 005. aubiects inter allv-atte nded to their nal state al 1998b)and in evocation of fearful emotional re not when they externally atte nded to nonaffective sponses from direct stimulation (Halgren et al.,1978). cha acter of a picture stimulus.Furthermore,ac The amygdala al Iso appears importan t in the detection tivity to co elate with envronment threa Scott 1997 Isenberg e emo a 0 an recal et a as ir ate th (Kluber Bucy 193 :Weiskra 1g56 et al 1998)One n ossibility therefore is that the King.1992).Strikingly,of the eight studies that exam- MPFC may be involved in the cognitive aspects (e.g. ned cerebral responses to fearful faces,six pointed to attention to emotion,appraisal emo tical involvement of the a amygd a orris et a of emotional processing (Drevets and 199 en the tivation als tion into other modalities such as words (Ise nbera et al vided into affective and cognitive regions,which has 1999)and vocalizations (Phillips 1998a).Morris et al been observed in the ante rio cingulate cortex (ACC) 1996)found that the amy ygdalar response to fearful (Bush e 2000 ACC is nown to b olved in aces ant raction with th form n that serves emo sing with incre th ta.2000 d to th of facial as subie MPEC (Petrides and Pandya.1999:Devinsky et al. were instructed to classify en notional faces by ender 1995).Figures 1A and 1B show that activatic ns re- not by emotion.Such an interpretation is further ported in the prefrontal cortex in response to different n from studies ng masked are ces wh amygo of e the ate cortex (Accad)(BA rostral 24 anterior/ven erience ubjectively (Morris et al..1998b: tral 32,33).While the activations located in the area of Whalen et al..1998a) the MPFC are more ventral and less dorsal,we did not Given that f ar is the most salient of the individual nnd any e or an alternative interpretation for the amyg. e s a mo. gen tha eaks fell into dor sal MPEC (se Fig 10) L 20011 son to studies with tasks that involved Emotion and Whalen et al.(1998b)observed that the amygdala re- Cognition.Thus.MPFC appears equally sensitive to sponds to fearful faces despite the lack of explicit rec-
While there may not be a particular brain region that is absolutely necessary for all emotional functions, the common activation of the MPFC may reflect that certain aspects may be shared across different emotional tasks. Figure 2A shows that the MPFC was activated across multiple individual emotions (four of five specific emotions in at least 40% of studies). Accordingly, X2 analysis revealed no specific association between the MPFC and Individual Emotion as compared to other regions. These findings suggest that the MPFC may have a general role in emotional processing, as suggested by Lane, Reiman, and colleagues, who reported that emotional films, pictures, and recall as wells as positive and negative emotion, happiness, sadness, disgust, and the mixture of these emotions all separately engaged the MPFC (Lane et al., 1997a; Lane et al., 1997c; Reiman et al., 1997). This is consistent with the notion that a number of processes are potentially common to various emotional tasks (e.g., appraisal/evaluation of emotion, emotional regulation, and emotion-driven decision-making). Lane et al. (1997b) did find that the MPFC (BA9) activated when subjects internally-attended to their emotional state, but not when they externally attended to nonaffective characteristics of a picture stimulus. Furthermore, activity in the MPFC has been shown to correlate with emotional awareness to both film and recall-generated emotion, suggesting its role in detecting emotional signals from both exteroceptive and interoceptive cues (Lane et al., 1998). One possibility therefore is that the MPFC may be involved in the cognitive aspects (e.g., attention to emotion, appraisal/identification of emotion) of emotional processing (Drevets and Raichle, 1998). Given the putative importance of cognition in emotion, we questioned whether the MPFC can be subdivided into affective and cognitive regions, which has been observed in the anterior cingulate cortex (ACC) (Bush et al., 2000). The ACC is known to be involved in a form of attention that serves to regulate both cognitive and emotional processing (Whalen et al., 1998a; Bush et al., 2000), and is closely interconnected to the MPFC (Petrides and Pandya, 1999; Devinsky et al., 1995). Figures 1A and 1B show that activations reported in the prefrontal cortex in response to different Individual Emotions and Induction Methods are located within ventral-rostral BA 9 and 10 of MPFC, and extend into the affective division of rostral anterior cingulate cortex (ACCad) (BA rostral 24, anterior/ventral 32, 33). While the activations located in the area of the MPFC are more ventral and less dorsal, we did not find any evidence for a functional affective-cognitive division of the MPFC. Our Cognitive Demand analysis revealed that relatively much fewer Emotion alone peaks fell into dorsal MPFC (see Fig. 1C), in comparison to studies with tasks that involved Emotion and Cognition. Thus, MPFC appears equally sensitive to both emotional tasks with and without Cognitive Demand, as activations from both conditions cluster in ventral MPFC (see Figs. 1C and 2C). Interestingly, previous meta-analyses of cognition revealed that the rostral-ventral and orbital regions of the MPFC are largely insensitive to cognitive tasks (Duncan and Owen, 2000; Cabeza and Nyberg, 2000). 2. Regions Associated with Individual Emotions 2.1 Fear and the amygdala. Specifically, fear induction had a strong association with the amygdala. Sixty percent of studies that examined fear activated the amygdala (X2 12.57, P 0.01) (Fig. 2A). Several lines of evidence support the notion that the amygdala is responsible for detecting, generating, and maintaining fear-related emotions. Particularly, the amygdala has been implicated in the recognition of fearful facial expressions (Adolphs et al., 1995; Calder et al., 1996), feelings of fear after procaine induction (Ketter et al., 1996), fear conditioning (LeDoux, 1993; Bechara et al., 1995; LaBar et al., 1995; Morris et al., 1998b; Whalen et al., 1998b), and in evocation of fearful emotional responses from direct stimulation (Halgren et al., 1978). The amygdala also appears important in the detection of environment threat (Scott et al., 1997; Isenberg et al., 1999; Phillips et al., 1998a), as well as in the coordination of appropriate responses to threat and danger (Kluber and Bucy, 1939; Weiskrantz, 1956; King, 1992). Strikingly, of the eight studies that examined cerebral responses to fearful faces, six pointed to the critical involvement of the amygdala (Morris et al., 1996; Breiter et al., 1996; Phillips et al., 1997; Phillips et al., 1998a; Morris et al., 1998a; Whalen et al., 1998a). Fear-associated amygdalar activations also extended into other modalities such as words (Isenberg et al., 1999) and vocalizations (Phillips 1998a). Morris et al. (1996) found that the amygdalar response to fearful faces showed a significant interaction with the intensity of emotion (increasing with increasing fearfulness) and that the activation was not contingent upon the explicit processing of facial expression, as subjects were instructed to classify emotional faces by gender not by emotion. Such an interpretation is further strengthened by findings from studies using masked fearful faces which found that the amygdalar response occurred even when the fearful expression was not consciously perceived or even when subjects did not experience fear subjectively (Morris et al., 1998b; Whalen et al., 1998a). Given that fear is the most salient of the individual emotions, an alternative interpretation for the amygdala’s involvement is that it has a more general role for vigilance or for processing salience, or attributes that make stimuli meaningful (Davis and Whalen, 2001). Whalen et al. (1998b) observed that the amygdala responds to fearful faces despite the lack of explicit rec- 336 PHAN ET AL.
FUNCTIONAL NEUROANATOMY OF EMOTION 337 ognition of the expression and that fearful faces are Though this review found that many of the SCC acti- more likely to signify a signal for threat than to induce vations arose from studies in which sadness was tran- fear given that subjects often do not report being ntly induced by autobiographical scripts (George e Hence.th amygdalar ctivati ons may be pri 00 n Derg et al. 9:Liott for proces he ocati ding toother inducti fear faces (Morris et al.1996:Breiter et al.1996 ods The inco istent findings in earlier activation Phillips et al.,1997).aversive pictures (Irwin et al. studies on transient sadness in healthy subjects,par 1996:Taylor et al,1998:Simpson et al,2000),as well ticularly in the subgenual ACC(Gemar et al,1996; as sac et al 1999)and happy faces (Breiter e Pardo et 93:George et al.. 1995,may attrib to the in prov on meth A positi on the fuly all of the with additio task uch as re activated to both pleasant and unpl visualizing emotional memories.Liotti et al.(2000)and Thus,the amygdala may not exclusively respond to Mayberg et al.(1999)attempted to address these per- affectively laden stimuli,but may respond to meaning- ceptua or cogn itive confounds by scanning subjects ul stim general.Ou own findings also concu er they ha eved a de nd. ral amyg the activations in subgenua al.2000e avers BA25). and the arly 70% studies f the Happines structure's role in mediating conditioned r nglia (BG)(F 2A)The cthat this a which enhance information processing to nonaversive may be ir ortant in sitive e otions,such as happi stimuli(Everitt,1991).These findings suggest that the gains support from multiple in vivo investiga- amy ala respond mportanc or stimu tions of addictive substances and behaviors (Breiter et alience egar ence (v e co ent 1997:Stein et al..1998 992:Roch et co 996),rew 999,a with id se(SCR)to affective pictur which der aying a vi n the es (Kocpp to salient arousing stimuli. nd n ive been obs regardless of emotional valence (Lang et al.1993). faces (Whalen et al. 1998a Morris et al 2.2 Sadness and the subcallosal cir ulate Sad oleasant pictures induction was significantly associated with subcallosal et al.,1997a;Lane et al,1999:Davidson and cingulate cortex (SCC)activation. About 46% of sad win,1999), ed recall (George ness in duc lized studies reported nd the Dam region ante P000 a Ci requ any 00 Int olimbic don the basal g or hy has be fou d in ventral striatum is well positioned to res ond to the SCC in resting state studies of patients with clin incentive reward motivation and to pregoal attainment ical depression,a mood disorder with relatively more of positive affect arising from progre on toward a sustained sadness (Baxter et al., 1985:Mayberg,1994 re goal (Dav and 1999 cons Drevets et al As expected,activ ity in he sub gula B 25 nc when o1s8tuali 1999. and th a211005 ulated that d in the BG:60%of s this area may lead those susceptible towards a com- nent of the BG(Fig.2A).Contrary to happ pensatory pattern of hypometabolism. disgust has been theoretically conceptuali as a Reiman and colleagues(1997)found anterior cingulate withdrawal emotion (Davidson et al.1990).The re- activity to rec -genera dD t no -ind ce viewed stu facial y mt rpr ps et taoge he d 07 9 its s (1998)hvp
ognition of the expression and that fearful faces are more likely to signify a signal for threat than to induce fear given that subjects often do not report being afraid. Hence, the amygdalar activations may be primarily for processing affective information in service of imparting danger warnings. Amygdala activations occur throughout various evocative stimuli, including fear faces (Morris et al., 1996; Breiter et al., 1996; Phillips et al., 1997), aversive pictures (Irwin et al., 1996; Taylor et al., 1998; Simpson et al., 2000), as well as sad (Blair et al., 1999) and happy faces (Breiter et al., 1996), and positive pictures (Hamann et al., 1999). A positive correlation of blood flow in the amygdala was found with subsequent recall of pleasant pictures (Hamann et al., 1999): in that study, the amygdala activated to both pleasant and unpleasant pictures. Thus, the amygdala may not exclusively respond to affectively laden stimuli, but may respond to meaningful stimuli in general. Our own findings also concur this interpretation since we have observed that amygdala also responds to nonaversive/neutral (Taylor et al., 2000) and positive (Liberzon et al., submitted) pictures, supporting evidence from animal studies of the structure’s role in mediating conditioned responses which enhance information processing to nonaversive stimuli (Everitt, 1991). These findings suggest that the amygdala responds to emotional importance or stimulus salience, regardless of valence (whether the content is pleasant or aversive/unpleasant). This is also consistent with psychophysiologic evidence of skin conductance response (SCR) to affective pictures which demonstrate a response to salient, arousing stimuli, regardless of emotional valence (Lang et al., 1993). 2.2 Sadness and the subcallosal cingulate. Sadness induction was significantly associated with subcallosal cingulate cortex (SCC) activation. About 46% of sadness induction studies reported activation of the SCC, region localized to the ventral/subgenual anterior cingulate (BA 25), over twice as frequently as any other specific emotion (X2 9.24, P 0.05). Interestingly, hypometabolism or hypoperfusion has been found in the SCC in resting state studies of patients with clinical depression, a mood disorder with relatively more sustained sadness (Baxter et al., 1985; Mayberg, 1994; Drevets et al., 1997). As expected, activity in the subgenual cingulate (BA 25) increased when depressed subjects respond to pharmacologic treatment (Brody et al., 1999; Mayberg et al., 2000). George et al. (1995) speculated that dysphoria-induced hyperactivity in this area may lead those susceptible towards a compensatory pattern of hypometabolism. Because Reiman and colleagues (1997) found anterior cingulate activity to recall-generated but not film-induced sadness, they interpreted that the SCC activations may result more from the cognitive process of internally generating emotion, and less from sadness itself. Though this review found that many of the SCC activations arose from studies in which sadness was transiently induced by autobiographical scripts (George et al., 1995; Lane et al., 1997c; Mayberg et al., 1999; Liotti et al., 2000), the X2 analysis did not support the notion that SCC activation was specifically associated with recall induction, as compared to other induction methods. The inconsistent findings in earlier activation studies on transient sadness in healthy subjects, particularly in the subgenual ACC (Gemar et al., 1996; Pardo et al., 1993; George et al., 1995), may be attributed to the differences in provocation method. Particularly, subjects were partly or fully scanned while they were actively generating the targeted emotional state with additional cognitive tasks such as recalling or visualizing emotional memories. Liotti et al. (2000) and Mayberg et al. (1999) attempted to address these perceptual or cognitive confounds by scanning subjects after they had achieved a desired intensity of sadness, and confirmed the activations in subgenual ACC (BA 25). 2.3 Happiness and the basal ganglia. Nearly 70% happiness induction studies reported activation in the basal ganglia (BG) (Fig. 2A). The notion that this area may be important in positive emotions, such as happiness, gains support from multiple in vivo investigations of addictive substances and behaviors (Breiter et al., 1997; Stein et al., 1998; Koob, 1992; Koch et al., 1996), reward processing (Rolls, 1999), and enjoyable (playing a video game) activities (Koepp et al., 1998). Activations in the basal ganglia, including the ventral striatum and putamen, have been observed in response to happy faces (Whalen et al., 1998a; Morris et al., 1996, 1998a; Phillips et al., 1998b), pleasant pictures (Lane et al., 1997a; Lane et al., 1999; Davidson and Irwin, 1999), happiness-induced recall (George et al., 1996b; Damasio et al., 2000), pleasant sexual and successful competitive arousal (Rauch et al., 1999; Redoute et al., 2000). Given its rich innervation of mesolimbic dopaminergic neurons, the basal ganglia/ ventral striatum is well positioned to respond to incentive reward motivation and to pregoal attainment of positive affect arising from progression toward a desired goal (Davidson and Irwin, 1999), consistent with the notion that happiness can be conceptualized as an approach emotion (Davidson et al., 1990). 2.4 Disgust and the basal ganglia. Interesting, we also found that disgust induction frequently activated the BG; 60% of studies evoking disgust reported engagement of the BG (Fig. 2A). Contrary to happiness, disgust has been theoretically conceptualized as a withdrawal emotion (Davidson et al., 1990). The reviewed studies suggest that, particularly, facial expressions of disgust activated the BG (Phillips et al., 1997, 1998a; Sprengelmeyer et al., 1998). Sprengelmeyer and colleagues (1998) hypothesized a specific FUNCTIONAL NEUROANATOMY OF EMOTION 337