Emotional information processing in mood disorders: a review of

behavioral and neuroimaging findings

Jukka M. Leppänen

Purpose of review Introduction

A relatively long history of research has shown that mood Mood disorders are associated with abnormalities in the
disorders are associated with abnormalities in the way emotional stimuli are perceived, responded to, and
processing of emotional stimuli. Only the most recent stored in memory. Compared with healthy controls, indi-
studies, however, have begun to elucidate the specificity viduals diagnosed with mood disorder exhibit, for exam-
and neural basis of these abnormalities. This article reviews ple, decreased accuracy of recognizing facial expressions
and discusses the results of these studies. [1–3], biased interpretation of facial expressions [4–6],
Recent findings diminished emotional reactions in response to happy
Individuals diagnosed with major depressive disorder facial expressions [7], and an enhanced memory for pre-
exhibit an attentional bias toward negative emotional cues viously presented negative facial expressions [8,9]. These
(e.g. sad faces), an attentional bias away from positive impairments and biases in the processing of emotional
emotional cues (e.g. happy faces), and an enhanced and social stimuli may underlie problems in mood and
memory for negative emotional material. Compared with interpersonal behavior, and contribute to the onset and
healthy controls, individuals with major depressive disorder maintenance of mood disorders [2]. For these reasons,
show increased neural activity in response to sad faces and understanding the specific nature and mechanisms of the
diminished neural activity in response to happy faces in emotion processing abnormalities may prove to be help-
emotion-related brain circuits (e.g. amygdala and ventral ful in the diagnosis, treatment, and prevention of mood
striatum). Some of these abnormalities in the processing of disorders. The purpose of this review is to examine the
emotional information persist after symptom remission and findings of the most recent studies on the relation bet-
they have also been found in healthy individuals who are at ween mood disorders and the processing of different
heightened risk for the development of mood disorders. types of emotional stimuli (i.e. studies published since
Summary 2004). The majority of these studies have examined emo-
The reviewed data show that major depressive disorder tion processing in major depressive disorder (MDD) and,
involves specific abnormalities in the cognitive and neural hence, this review largely focuses on this specific dis-
processing of emotional information and that these order. Behavioral and electrophysiological studies will be
abnormalities may potentially contribute to the vulnerability reviewed first, followed by a review of neuroimaging
for negative emotion and onset of depressive episodes. studies and studies with high-risk individuals.

Keywords
bipolar disorder, emotion recognition, major depressive Behavioral and electrophysiological studies
disorder Recent behavioral and electrophysiological studies have
attempted to go beyond the abnormalities in emotion
Curr Opin Psychiatry 19:34–39. ß 2006 Lippincott Williams & Wilkins. processing in MDD to examine what specific cognitive
operations are biased in this disorder, and also to examine
University of Tampere, Tampere, Finland the content and the diagnostic specificity of these biases
Correspondence to Jukka M. Leppänen, Human Information Processing (i.e. whether they are specific to a particular type of
Laboratory, Department of Psychology, FIN-33014 University of Tampere, emotional stimuli and whether they are found in other
Tampere, Finland
Tel: +358 3 3551 6537; fax: +358 3 3551 7710; e-mail: [email protected] psychiatric disorders). It has long been thought that
cognitive operations involved in stimulus selection
Current Opinion in Psychiatry 2006, 19:34–39
(attention) and stimulus identification are not as much
Abbreviations biased in MDD as are operations involved in the elab-
ACC anterior cingulate cortex oration and storage of already identified stimuli. Two
ERP event-related brain potential
fMRI functional magnetic resonance imaging recent studies suggest, however, that this is not necess-
MDD major depressive disorder arily the case [10,11]. These studies used a so-called
face dot-probe task to assess the allocation of attention
ß 2006 Lippincott Williams & Wilkins between an emotional and a neutral face positioned
0951-7367
in different regions in the visual space (Fig. 1). The
studies consistently showed that individuals diagnosed
with MDD without any comorbid psychiatric disorder
34
 Emotional information processing Leppänen 35

Figure 1. A dot-probe paradigm used to study attentional individuals also have difficulty in performing a task that
biases toward emotional stimuli requires an ability to selectively focus on task-relevant
dimension of a stimulus while ignoring other, task-
irrelevant stimulus dimensions (e.g. classify faces
based on gender while ignoring variations in emotional
expressions) [14].

MDD has also been associated with an attentional bias

away from happy facial expressions [15] and reduced
perceptual sensitivity to happy facial expressions
[16,17]. There is also evidence that healthy individuals
who score high on measures of anhedonia exhibit heigh-
tened recognition thresholds for happy facial expressions
[18]. These findings fit with the hypothesis that respon-
siveness to positive emotional cues is blunted in depres-
sion. It is noteworthy, however, that not all studies have
reported a bias away from happy faces in MDD [10]. It
is also unresolved whether the bias away from happy
faces is due to MDD or whether it reflects comorbid
anxiety [11,15,19].

Besides the biases in attention and stimulus selection,

biases in memory processes characterize MDD. Com-
Each trial in this task starts with a presentation of a fixation cross for pared with healthy controls and patients with generalized
500 ms. Next, two faces, one neutral and one emotional, are displayed social phobia, patients with MDD tend to recall a higher
on the screen for 1000 ms. Immediately after the offset of the face
pictures, a small dot probe is presented in the location where one of the
proportion of previously presented negative words and a
faces had been. Participants are asked to indicate the position of the lower proportion of previously presented positive words
probe (left/right) by pressing a response key. Shorter probe detection [13]. These biases have long been thought to result
times for probes presented in the same location as the emotional face
than probes presented in the location of the neutral face indicate a bias
from a tendency of depressed individuals to allocate
toward the emotional stimulus. Note that the timing parameters of the more processing resources to negative than to positive
task can vary across studies. In addition, photographs of real faces emotional cues. It has been difficult, however, to directly
instead of schematic faces are typically used in pictorial dot-probe
experiments.
test this hypothesis by using behavioral methods because
these methods do not permit one to directly assess the
operations that occur in between the presentation of
a stimulus and the registration of a behavioral response.
preferentially attended to sad facial expressions over To avoid this limitation, Deveney and Deldin [20]
simultaneously presented neutral facial expressions. measured event-related brain potentials (ERPs) in indi-
Importantly, this bias was only found in patients with viduals with and without MDD while they hold pictures
MDD. It was not found in healthy controls, patients with of happy, neutral, and sad emotional stimuli in memory
general anxiety disorder, or patients with social phobia. for a short (5 s) interval. A slow-wave component of the
Furthermore, the bias was specific to cues that signaled ERP evoked in this type of a task is known to reflect
loss or sadness. There was no evidence for preferential neural activity related to the elaboration and retention of
allocation of attention to cues that signaled other negative stimulus material in working memory. The researchers
emotions (e.g. anger), which rules out the hypothesis that found that, in healthy individuals but not in individuals
the bias toward sad faces in MDD reflects a more general with MDD, the amplitude of the slow-wave component
bias toward negative emotional stimuli. One potential was reduced to sad faces compared with happy and
explanation for these results is that individuals with neutral faces. This may indicate that whereas healthy
MDD have difficulty in ignoring (loss-related) emotional individuals actively avoid processing sad faces, individ-
cues and in disengaging attention from them. Broadly uals with MDD process them similarly as happy and
consistent with this hypothesis, there is evidence show- neutral faces. In other words, individuals with MDD may
ing that patients with MDD interpret emotionally neutral fail to show the normal inhibition of the elaboration of
faces as sad [12]. Additionally, students who score high negative interpersonal stimuli. Another ERP study from
on measures of depressive symptomatology or have a the same group, which employed words instead of facial
history of depressive episode fail to inhibit attending and expressions as stimuli, showed that individuals with
processing negative emotional distractor stimuli while MDD also exhibit decreased elaboration of positive
performing a word classification task [13]. Dysphoric relative to neutral and negative stimuli [21].
36 Mood disorders

Figure 2. A simplified diagram showing the psychological processes involved in the processing of (visual) emotional signals and the
neural structures that subserve these processes

Modified from models presented in [22–25]. ACC,

anterior cingulate cortex; DLPFC, dorsolateral
prefrontal cortex; DMPFC, dorsomedial prefrontal Emotion regulation
cortex; FG, fusiform gyrus; OFC, orbitofrontal ACC, DMPFC, DLPFC
cortex; STS, superior temporal sulcus.

Analysis of stimulus Emotion recognition and

Stimulus
features emotional response
V1/V2, FG, STS Amygdala, insula, OFC,
ventral striatum

Neuroimaging studies MDD differ from healthy controls in neural responses to

Modern brain imaging technologies are increasingly different emotional stimuli. Two recent studies used
utilized to study whether mood disorders are associated event-related fMRI to measure neural responses to happy
with structural and functional abnormalities in emotion- and sad facial expressions in individuals with and without
related neural systems. The psychological processes MDD. In one of these studies [31], healthy controls but
involved in the recognition of emotional stimuli and not patients with MDD showed a linear increase in
the neural substrates of these processes in healthy indi- activity in bilateral fusiform gyri and ventral striatum
viduals are now relatively well defined (Fig. 2). Distinct (right putamen) in response to increasingly intense happy
brain structures have been implicated in the perceptual facial expressions. A reverse pattern of results was
processing of emotional stimuli such as facial expressions observed in response to sad facial expressions. That is,
(i.e. fusiform gyrus and superior temporal sulcus in the patients with MDD but not healthy controls showed a
occipitotemporal cortex), in emotion recognition and linear increase in activity in right fusiform gyrus, ventral
generation of emotional reactions in response to the striatum (left putamen), and left parahippocampal gyrus
stimulus (amygdala, anterior insula, orbitofrontal cortex, extending to left amygdala in response to increasingly
and ventral striatum), and in the regulation of emotional intense sad faces. Consistent with this finding, another
reactions evoked in response to the stimulus [anterior recent fMRI study showed exaggerated activity in the
cingulated cortex (ACC), prefrontal cortex] [22–25]. ventral striatum (putamen/globus pallidus) and the amyg-
dala in response to sad faces in individuals with MDD
Recent structural imaging studies using magnetic reson- [32]. Interestingly, the increased responses to sad faces
ance imaging (MRI) have replicated earlier findings attenuated after 8 weeks of antidepressant treatment
showing enlarged amygdala volumes in individuals with with a selective serotonin reuptake inhibitor (fluoxetine
recent-onset MDD [26] and individuals with bipolar hydrochloride). Taken together, these fMRI findings
disorder [27]. Contrary to these findings in adults, show that activity in neural circuits underlying the pro-
decreased amygdala volumes have been reported in cessing of facial emotion is altered in MDD. More
children and adolescents with MDD [28], and in ado- specifically, individuals with MDD show increased
lescents with bipolar disorder [29,30]. Abnormal age- activity in response to sad faces and decreased activity
related increases in the amygdala volume have also been in response to happy faces in these neural systems.
found in bipolar adolescents [29], which may help to
explain the enlarged amygdala volumes in adult age. It is noteworthy that the neural processing of facial
Although these findings point to a neuroanatomical basis expressions may differ in MDD and bipolar disorder. A
of dysfunctional emotion processing in mood disorders, recent study showed that patients with bipolar disorder I
the functional correlates of the amygdala volume mania show reduced sensitivity to facial sadness in a
abnormalities and their origins (i.e. whether they predate behavioral rating task and also reduced levels of activity
or follow illness onset) have not been elucidated. in the amygdala and subgenual cingulate in response to
sad facial expressions [33]. This study did not report
Functional magnetic resonance imaging (fMRI) provides abnormalities in responses to happy faces in manic
a means to more directly assess whether patients with patients. Another study with bipolar disorder patients
 Emotional information processing Leppänen 37

in euthymic or depressed state, however, showed in- functional polymorphism in the promoter region of the
creased activity in subcortical affect-related brain regions serotonin transporter gene (5-HTT), a gene that
in response to happy (and fearful) facial expressions [34]. regulates 5-HTT expression and affects brain serotonin
function, has been associated with mood disorders. There
The increased activity in the amygdala and other is evidence that carriers of a short (S) variant or allele of
emotion-related regions in response to negative the 5-HTT promoter polymorphism have an increased
emotional stimuli in MDD may partly reflect reduced risk for the development of depression in relation to
functional connectivity and feedback between certain stressful life events [40–42]. More interestingly in the
cortical and limbic brain structures. Cortico-limbic con- present context, neuroimaging studies have shown that,
nections are thought to play a central role in the modu- compared with individuals with long (L) allele of the
lation and inhibition of maladaptive emotional reactions. 5-HTT, S allele carriers show a robust increase in amyg-
To test this hypothesis, Anand and colleagues [35] dala response to threat-related facial expressions and
examined activity in ACC and limbic brain regions (i.e. other aversive emotional stimuli [43,44,45] (see also
the amygdala, pallidostriatum, and medial thalamus) as [46]). Additionally, S allele carriers show reduced gray
well as the connectivity between these brain regions in matter volumes in the perigenual ACC and the amygdala
patients with MDD and healthy controls. The connec- as well as reduced functional correlation of activity
tivity was measured by using a recently developed fMRI- between these regions during processing of threatening
based method in which connectivity between distal brain facial expressions [47]. This suggests that 5-HTT pro-
regions is inferred from correlations of low-frequency moter polymorphism alters those ACC–amygdala con-
blood oxygen related fluctuations. Compared with nections that play a role in inhibition of amygdala-based
healthy controls, patients with MDD showed increased emotional reactions.
activation of the amygdala, pallidostriatum, insula,
anterior cingulate cortex, and anteromedial prefrontal Together, the studies with high-risk populations raise a
cortex in response to negative pictures. In addition, it possibility that subtle individual differences in cognitive
was found that depressed patients showed decreased and neural processing of emotional stimuli may, in inter-
connectivity between ACC and limbic regions during action with other pathogenic factors (e.g. environmental
emotional stimulation. This finding supports the view stressors), play a role in the development of mood dis-
that increased activity in the limbic emotion-related orders. This raises a question of whether the styles
brain structures and maladaptive emotional reactions in of processing emotional information can be changed
MDD are partly attributable to reduced cortico-limbic through interventions and whether such changes, in
connectivity and disruption of normal emotion regula- turn, lead to therapeutic effects. Initial experimental evi-
tion processes. dence shows that repeated practice can, indeed, result
in changes in the deployment of attention to emotional
Studies with individuals at high risk for stimuli and in the interpretation of ambiguous emotional
mood disorders cues, and these experimentally induced changes have a
It is of great interest for research on mood disorders to clear effect on emotional reactions elicited in subsequent
determine whether the biases in the cognitive and neural stressful situations [48]. Further research is needed,
processing of emotional stimuli represent stable trait however, to find the most effective methods of altering
characteristics that contribute to vulnerability for the emotional information processing, and also to study how
development of mood disorders. Researchers have begun long the achieved changes endure and how they generalize
to address this question by studying whether a height- to different types of stimuli and events [48]. Ultimately,
ened risk for the development of mood disorders is the knowledge gained from these studies may shed light
associated with abnormalities in emotion processing. on the mechanisms that underlie the positive effects of
interventions such as cognitive psychotherapy [49–51],
Recent studies with high-risk individuals have shown and also help to sharpen and increase the effectiveness of
that individuals who have recovered from a major these therapeutic approaches.
depressive episode, and who are vulnerable to experience
another depressive episode, show biases in cognitive Conclusion
processing of emotional signals [12,14,36]. Additionally, Recent research on mood disorders has given us an
there is evidence that offspring of parents with bipolar increasingly detailed picture of the emotion processing
disorder [37] and offspring of parents with MDD [38] abnormalities in these disorders and of the cognitive and
exhibit altered emotional behavior, although the evi- neural mechanisms that underlie these abnormalities. To
dence to date is preliminary and not all studies have summarize, the data indicate that MDD is associated
found such effects [39]. Recent studies have also begun with a specific attentional and memory bias toward
to link a genetic susceptibility for mood disorders with negative (sad) cues and away from positive cues. These
altered emotional information processing. Specifically, biases may be associated with abnormalities in the
38 Mood disorders

11 Gotlib IH, Kasch KL, Traill S, et al. Coherence and specificity of information-
excitability of the brain emotion-related circuits as well as  processing biases in depression and social phobia. J. Abnorm Psychol 2004;
disruption of the cortico-limbic connections that are 113:386–398.
Well designed study that addressed the important question of whether the
important in the regulation of emotional responses. The information processing biases found in unipolar depression are specific toward
recent data also provide some support for the hypoth- loss and sadness-related emotional cues, whether they are found across different
tasks used to assess attention and memory, and whether they are diagnostic
esis that biases in the processing of emotional infor- specific (i.e. whether they are found in other affective disorders).
mation represent stable trait characteristics that affect 12 Leppänen JM, Milders M, Bell JS, et al. Depression biases the recognition of
vulnerability for mood disorders. A challenge for future  neutral faces. Psychiatry Res 2004; 128:123–133.
This study used a follow-up design to demonstrate that biases in the processing of
studies will be to employ multilevel analysis of emotion neutral faces in MDD persist after symptom remission.
processing in order to better integrate the findings from 13 Joormann J. Attentional biases in dysphoria: the role of inhibitory processes.
neuroscientific and behavioral studies. For example, it is  Cognit Emot 2004; 18:125–147.
This study used a clever experimental design to demonstrate that the attentional
not known how the abnormalities at the neural level bias in depression may specifically involve an inability to actively inhibit the
translate to recognition and interpretation of emotional processing of nongoal-relevant negative emotional cues.
and social information at the behavioral level, and how 14 Gilboa-Schechtman E, Ben-Artzi E, Jeczemien P, et al. Depression impairs the
 ability to ignore the emotional aspects of facial expression: evidence from the
these abnormalities, in turn, are related to the problems Garner task. Cognit Emot 2004; 18:209–231.
in mood (e.g. anhedonia, irritability) and interpersonal The authors introduce a new perceptual paradigm to study attention to emotional
stimuli and demonstrate that depressed individuals have difficulty ignoring the
behavior. Further research with high-risk individuals as emotional aspects of visual stimuli even when they are specifically instructed to do
well as research examining the most effective methods so.
for controlling and changing emotional information pro- 15 Suslow T, Dannlowski U, Lalee-Mentzel J, et al. Spatial processing of facial
emotion in patients with unipolar depression: a longitudinal study. J Affect
cessing is also required to determine the potential of this Disord 2004; 83:59–63.
approach for the identification, prevention, and treatment 16 Surguladze SA, Young A, Senior C, et al. Recognition accuracy and response
of mood disorders. bias to happy and sad facial expressions in patients with major depression.
Neuropsychology 2004; 18:212–218.
17 Hayward G, Goodwin GM, Cowen PJ, Harmer CJ. Low-dose tryptophan
Acknowledgements  depletion in recovered depressed patients induces changes in cognitive
The writing of this paper was made possible, in part, by grants from the processing without depressive symptoms. Biol Psychiatry 2005; 57:517–
524.
Academy of Finland and Finnish Cultural Foundation. This innovative study linked neurochemistry with cognitive neuropsychology by
demonstrating that experimentally induced reduction in brain serotonin affects
cognitive processing and emotion recognition in recovered depressed patients but
References and recommended reading not in healthy controls.
Papers of particular interest, published within the annual period of review, have
18 Coupland NJ, Sustrik RA, Ting P, et al. Positive and negative affect differen-
been highlighted as:
tially influence identification of facial emotions. Depress Anxiety 2004;
 of special interest
19:31–34.
 of outstanding interest
Additional references related to this topic can also be found in the Current 19 Gilboa-Schechtman E, Presburger G, Marom S, Hermesh H. The effects of
World Literature section in this issue (pp. 107–108). social anxiety and depression on the evaluation of facial crowds. Behav Res
Ther 2005; 43:467–474.
1 Feinberg TE, Rifkin A, Schaffer C, Walker E. Facial discrimination and 20 Deveney CM, Deldin PJ. Memory of faces: a slow wave ERP study of major
emotional recognition in schizophrenia and affective disorders. Arch Gen  depression. Emotion 2004; 4:295–304.
Psychiatry 1986; 43:276–279. The authors used recordings of event-related brain potentials (ERPs) to provide
2 Persad SM, Polivy J. Differences between depressed and nondepressed direct evidence that individuals with and without MDD differentially maintain and
individuals in the recognition of and response to facial emotional cues. elaborate emotional facial expressions in working memory.
J Abnorm Psychol 1993; 102:358–368.
21 Shestyuk AY, Deldin PJ, Brand JE, Deveney CM. Reduced sustained brain
3 Rubinow DR, Post RM. Impaired recognition of affect in facial expression in  activity during processing of positive emotional stimuli in major depression.
depressed patients. Biol Psychiatry 1992; 31:947–953. Biol Psychiatry 2005; 57:1089–1096.
4 Gur RC, Erwin RJ, Gur RE, et al. Facial emotion discrimination. II: Behavioral The authors examined the electrophysiological correlates of sustained processing
findings in depression. Psychiatry Res 1992; 42:241–251. of positive, negative, and neutral self-relevant words in individuals with and without
MDD, and demonstrated that MDD may be specifically associated with diminished
5 Hale WW. Judgment of facial expressions and depression persistence. neural responses to positive emotional cues.
Psychiatry Res 1998; 80:265–274.
22 Adolphs R. Neural systems for recognizing emotion. Curr Opin Neurobiol
6 Hale WW, Jansen JHC, Bouhuys AL, van den Hoofdakker RH. The judgment 2002; 12:169–177.
of facial expressions by depressed patients, their partners and controls.
J Affect Disord 1998; 47:63–70. 23 Haxby JV, Hoffman EA, Gobbini MI. Human neural systems for face recogni-
tion and social communication. Biol Psychiatry 2002; 51:59–67.
7 Sloan DM, Bradley MM, Dimoulas E, Lang PJ. Looking at facial expressions:
dysphoria and facial EMG. Biol Psychiatry 2002; 60:79–90. 24 Phillips ML, Drevets WC, Rauch SL, Lane R. Neurobiology of emotion
perception. I: The neural basis of normal emotion perception. Biol Psychiatry
8 Gilboa-Schechtman E, Erhard-Weiss D, Jeczemien P. Interpersonal deficits 2003; 54:504–514.
meet cognitive biases: memory for facial expressions in depressed and
anxious men and women. Psychiatry Res 2002; 113:279–294. 25 Phillips ML, Drevets WC, Rauch SL, Lane R. Neurobiology of emotion
perception. II: Implications for major psychiatric disorders. Biol Psychiatry
9 Ridout N, Astell AJ, Reid IC, et al. Memory bias for emotional facial expres- 2003; 54:515–528.
sions in major depression. Cognit Emot 2003; 17:101–123.
26 Lange C, Irle E. Enlarged amygdala volume and reduced hippocampal volume
10 Gotlib IH, Krasnoperova E, Neubauer D, Joormann J. Attentional biases for  in young women with major depression. Psychol Med 2004; 34:1059–1064.
 negative interpersonal stimuli in clinical depression. J. Abnorm Psychol 2004; This study replicates and confirms previous findings showing structural abnorm-
113:127–135. alities in the amygdala and hippocampus in patients with recent-onset major
This paper reports a carefully designed and conducted study that employed a depression.
widely used ‘dot-probe’ task to demonstrate that patients with MDD exhibit a
selective attentional bias toward sad facial expressions. A similar attention bias 27 Strakowski SM, DelBello MP, Adler CM. The functional neuroanatomy of
was not found in healthy individuals or in patients diagnosed with generalized  bipolar disorder: a review of neuroimaging findings. Mol Psychiatry 2005;
anxiety disorder, nor was it found for other emotional facial expressions than sad 10:105–116.
faces (i.e. angry or happy faces). These findings suggest that the bias toward sad A good review of the structural and functional brain imaging studies of bipolar
facial expressions is both diagnostic and content specific. disorder.
 Emotional information processing Leppänen 39

28 Rosso IM, Cintron CM, Steingard RJ, et al. Amygdala and hippocampus 39 Pine DS, Lissek S, Klein RG, et al. Face-memory and emotion: associations
 volumes in pediatric major depression. Biol Psychiatry 2005; 57:21–26.  with major depression in children and adolescents. J Child Psychol Psychiatry
This is one of the first structural imaging studies examining amygdala and 2004; 45:1199–1208.
hippocampus volumes in pediatric MDD. This study examined how MDD and a familiar risk for MDD are associated with the
ability to recall those facial expressions that specifically engage the amygdala
29 Chen BK, Sassi R, Axelson D, et al. Cross-sectional study of abnormal
circuitry (i.e. fearful faces).
 amygdala development in adolescents and young adults with bipolar disorder.
Biol Psychiatry 2005; 56:399–405. 40 Caspi A, Sugden K, Moffitt TE, et al. Influence of life stress on depression:
This study showed different age-related changes in the amygdala volume in moderation by a polymorphism in the 5-HTT gene. Science 2002; 301:386–
adolescents and young adults with and without bipolar disorder, pointing to 389.
abnormal development of the amygdala in bipolar disorder.
41 Kendler KS, Kuhn JW, Vittum J, et al. The interaction of stressful life events and
30 DelBello MP, Zimmerman ME, Mills NP, et al. Magnetic resonance imaging a serotonin transporter polymorphism in the prediction of episodes of major
 analysis of amygdala and other subcortical brain regions in adolescents with depression: a replication. Arch Gen Psychiatry 2005; 62:529–535.
bipolar disorder. Bipolar Disord 2004; 6:43–52.
42 Eley TC, Sugden K, Corsico A, et al. Gene–environment interaction analysis
This study shows decreased amygdala volumes in adolescents with bipolar
of serotonin system markers with adolescent depression. Mol Psychiatry
disorder.
2004; 9:908–915.
31 Surguladze S, Brammer MJ, Keedwell P, et al. A differential pattern of neural
43 Hariri AR, Mattay VS, Tessitore A, et al. Serotonin transporter genetic variation
 response toward sad versus happy facial expressions in major depressive
and the response of the human amygdala. Science 2002; 297:400–403.
disorder. Biol Psychiatry 2005; 57:201–209.
By using fMRI, the authors of this study demonstrated that, compared to healthy 44 Hariri AR, Drabant EM, Munoz KE, et al. A susceptibility gene for affective
controls, patients with MDD showed increased responses to sad faces and  disorders and the response of the human amygdala. Arch Gen Psychiatry
decreased responses to happy faces in neural regions implicated in facial 2005; 62:146–152.
expression processing. These findings corroborate the finding of behavioral This study provides an important replication of the earlier study by the same
studies and point to a neural basis of the emotion recognition biases in MDD. authors in which they found that polymorphism in the serotonin transporter gene is
associated with amygdala reactivity.
32 Fu CHY, Williams SCR, Cleare AJ, et al. Attenuation of the neural responses
 to sad faces in major depression by antidepressant treatment. Arch Gen 45 Heinz A, Braus DF, Smolka MN, et al. Amygdala–prefrontal coupling depends
Psychiatry 2004; 61:877–889.  on a genetic variation of the serotonin transporter. Nat Neurosci 2005; 8:20–
This is among the first studies showing that MDD is associated with exagger- 21.
ated neural responses to sad faces in regions implicated in facial expression This study provides an independent replication of the findings of genetically driven
processing, and that these neural responses are attenuated by antidepressant variation in the amygdala responses to negative emotional stimuli, and further
treatment. demonstrates that the polymorphism in the serotonin transporter gene is also
associated with the coupling of activity between the amygdala and the ventro-
33 Lennox BR, Jacob R, Calder AJ, et al. Behavioural and neurocognitive
medial prefrontal cortex.
 responses to sad facial affect are attenuated in patients with mania. Psychol
Med 2004; 34:795–802. 46 Battaglia M, Ogliari A, Zanoni A, et al. Influences of the serotonin transporter
The study provided converging behavioral and neuroimaging evidence that mania  promoter gene and shyness on children’s cerebral responses to facial
is associated with reduced sensitivity to sad facial expressions. expressions. Arch Gen Psychiatry 2005; 62:85–94.
The results of this study show that polymorphism in the serotonin transporter
34 Lawrence NS, Williams AM, Surguladze S, et al. Subcortical and ventral
promoter gene is associated with brain electrophysiologic responses to threaten-
 prefrontal cortical neural responses to facial expressions distinguish patients
ing facial expressions in children.
with bipolar disorder and major disorder. Biol Psychiatry 2004; 55:578–
587. 47 Pezawas L, Meyer-Lindenber A, Drabant EM, et al. 5-HTTLPR polymorphism
This is the first neuroimaging study that compared neural responses to facial  impact human cingulate–amygdala interactions: a genetic susceptibility
expressions in patients with bipolar disorder and MDD. mechanism for depression. Nat Neurosci 2005; 8:828–834.
This study provides an interesting extension of the earlier studies showing that a
35 Anand A, Li Y, Wang Y, et al. Activity and connectivity of brain mood regulating
polymorphism in the 5-HTT gene is associated with the excitability of the brain
 circuit in depression: a functional magnetic resonance study. Biol Psychiatry
circuits underlying negative emotion. The authors used morphometrical and
2005; 57:1079–1088.
functional imaging analyses to show that the S allele of the 5-HTT polymorphism
This study used a novel fMRI-based technique to show that cortico-limbic con-
is associated with structural abnormalities and reduced activation in the cingu-
nectivity is reduced in MDD, supporting the hypothesis that the cortical regulation
late–amygdala feedback circuitry (i.e. a circuitry that plays an important role in the
of limbic activity is disrupted in depression.
regulation and inhibition of negative emotional reactions).
36 Bhagwagar Z, Cowen PJ, Goodwin GM, Harmer CJ. Normalization of en-
48 Mathews A. On the malleability of emotional encoding. Behav Res Ther 2004;
hanced fear recognition by acute SSRI treatment in subjects with a previous
 42:1019–1036.
history of depression. Am J Psychiatry 2004; 161:166–168.
The author reviews recent innovative experiments from his own and others’
37 Gotlib IH, Traill SK, Montoya RL, et al. Attention and memory biases in the laboratories showing that emotional information processing can be altered by
 offspring of parents with bipolar disorder: indications from a pilot study. J Child training and that these alterations influence vulnerability for negative emotion.
Psychol Psychiatry 2005; 46:84–93.
49 Fava GA, Ruini C, Rafanelli C, et al. Six-year outcome of cognitive behavior
This is the first well designed study of information processing biases in children
therapy for prevention of recurrent depression. Am J Psychiatry 2004;
with an elevated risk for mood disorders (i.e. offspring of parents with bipolar
161:1872–1876.
disorder). The study provides preliminary evidence that negative biases in attention
and memory may represent vulnerability markers for mood disorders. 50 Paykel ES, Scott J, Cornwall PL, et al. Duration of relapse prevention after
cognitive therapy in residual depression: follow-up of controlled trial. Psychol
38 Grillon C, Warner V, Hille J, et al. Families at high and low risk for depression: a
Med 2005; 35:59–68.
 three-generation startle study. Biol Psychiatry 2005; 57:953–960.
By measuring anxiety-related behavior (startle reflex) in children and grandchildren 51 Hollon SD, DeRubeis RJ, Shelton RC, et al. Prevention of relapse following
of parents with MDD, the authors of this study found evidence that increased cognitive therapy vs medications in moderate to severe depression. Arch Gen
startle magnitude may represent a vulnerability marker for MDD. Psychiatry 2005; 62:417–422.