Blunted Activation in Orbitofrontal Cortex
During Mania: A Functional Magnetic Resonance
Lori L. Altshuler, Susan Y. Bookheimer, Jennifer Townsend, Manuel A. Proenza, Naomi Eisenberger,
Fred Sabb, Jim Mintz, and Mark S. Cohen
Background: Patients with bipolar disorder have been reported to have abnormal cortical function during mania. In this study, we
sought to investigate neural activity in the frontal lobe during mania, using functional magnetic resonance imaging (fMRI).
Specifically, we sought to evaluate activation in the lateral orbitofrontal cortex, a brain region that is normally activated during
activities that require response inhibition.
Methods: Eleven manic subjects and 13 control subjects underwent fMRI while performing the Go-NoGo task, a neuropsychological
paradigm known to activate the orbitofrontal cortex in normal subjects. Patterns of whole-brain activation during fMRI scanning were
determined with statistical parametric mapping. Contrasts were made for each subject for the NoGo minus Go conditions. Contrasts
were used in a second-level analysis with subject as a random factor.
Results: Functional MRI data revealed robust activation of the right orbitofrontal cortex (Brodmann’s area [BA] 47) in control subjects
but not in manic subjects. Random-effects analyses demonstrated significantly less magnitude in signal intensity in the right lateral
orbitofrontal cortex (BA 47), right hippocampus, and left cingulate (BA 24) in manic compared with control subjects.
Conclusions: Mania is associated with a significant attenuation of task-related activation of right lateral orbitofrontal function. This
lack of activation of a brain region that is usually involved in suppression of responses might account for some of the disinhibition seen
in mania. In addition, hippocampal and cingulate activation seem to be decreased. The relationship between this reduced function
and the symptoms of mania remain to be further explored.
Key Words: Functional magnetic resonance imaging, mania,
risk-taking behavior), impaired attention (distractibility), and
increased motor activity (e.g., increased movement, increased
talkativeness). Because many of these symptoms suggest an
impairment in normal brain inhibitory mechanisms, impairment
in orbitofrontal function might contribute to this symptom pre-
sentation (Damasio 2000; Horn et al 2003). Human studies have
shown that patients with orbitofrontal lesions can present with
disinhibition, distractibility, hyperactivity, euphoria, and impul-
sivity (Fuster 1989; Starkstein et al 1988, 1990). Animal studies
demonstrate a role for the orbitofrontal cortex in inhibition of
movement, and lesions in this brain region might result in
increased motor activity (Kawashima et al 1996a).
Resting-state imaging studies have implicated disturbances in
functioning in the lateral and medial prefrontal cortex and in limbic
regions in patients with mania (al Mousawi et al 1996; Baxter et al
1985, 1989; Blumberg et al 2000; Drevets et al 1992; Goodwin e t a l
1997; Gyulai et al 1997; Kishimoto et al 1987; Migliorelli et al
1993; O’Connell et al 1995). Very few activation studies have
n addition to an alteration in mood, the clinical state of mania
comprises a cluster of symptoms involving increased impul-
sivity (e.g., overspending, hypersexual behavior, increased
been performed with neurocognitive paradigms to probe the
function of specific brain regions during mania. This is probably
owing to the difficulty of having a manic patient remain still
during scanning. The few activation imaging studies, however,
suggest an attenuation of orbitofrontal functioning during mania
(Blumberg et al 1999, 2003; Elliott et al 2004; Rubinsztein et al
2001). The aim of the present study was to further evaluate
orbitofrontal functioning in manic subjects compared with con-
trol subjects, with the use of a standard neuropsychological task
that specifically requires behavioral inhibition.
Methods and Materials
Our study protocol was approved by the institutional review
boards at the University of California, Los Angeles (UCLA) and at
the Department of Veterans Affairs (VA) Greater Los Angeles
Healthcare System, and each subject gave written consent before
their inclusion in the study. We recruited subjects with bipolar I
disorder through the UCLA Mood Disorders Clinic and the
Bipolar Disorders Clinic of the VA Greater Los Angeles Health-
care System in West Los Angeles, as well as the inpatient units of
both hospitals; subjects enrolled in other research projects of the
UCLA Mood Disorders Research Program were also invited to
participate. We recruited control subjects by advertisements
placed in local newspapers and campus flyers. All subjects were
interviewed with the Structured Clinical Interview for DSM-IV
(Spitzer et al 1996). Control subjects were excluded if they had a
current or past psychiatric diagnosis (including history of sub-
stance abuse) or were taking any medications for medical
reasons. Subjects with bipolar illness were included if they met
criteria for bipolar I disorder and had current mania or hypoma-
nia. They were excluded if they had any other active Axis I
comorbidity. Additional exclusion criteria for both control and
bipolar subjects included left-handedness, hypertension, neuro-
logic illness, metal implants, and a history of skull fracture or
From the Department of Psychiatry (LLA, JM), Department of Veterans Af-
fairs Greater Los Angeles Healthcare System, West Los Angeles Health-
care Center, Los Angeles; Departments of Psychiatry and Biobehavioral
California, Los Angeles; and the Ahmanson-Lovelace Brain Mapping
Center (FS, MSC), UCLA School of Medicine, Los Angeles, California.
Address reprint requests to Lori L. Altshuler, M.D., UCLA Neuropsychiatric
Institute and Hospital, 300 UCLA Medical Plaza, Suite 1544, Box 957057,
Los Angeles, CA 90095-7057; E-mail: firstname.lastname@example.org.
Received March 30, 2005; revised August 9, 2005; accepted September 14,
BIOL PSYCHIATRY 2005;58:763–769
© 2005 Society of Biological Psychiatry
head trauma with loss of consciousness for more than 5 min. On
the day of the scan, we rated mood symptoms in the bipolar
subjects, using the Young Mania Rating Scale (YMRS; Young et al
1978) and the 21-item Hamilton Depression Rating Scale (Ham-
ilton 1960) to assess for current severity of mania and depression
In total, 11 subjects (7 [64%] women) with bipolar I disorder,
currently manic or hypomanic, and 13 control subjects (8 [62%]
women) were included. The mean (?SD) age for the 11 manic
subjects was 36 ? 7.6 years, and the mean age for the 13 control
subjects was 31 ? 6.7 years (t ? 1.94, p ? .07). Both groups were
primarily Caucasian [manic subjects were 63% Caucasian, 18%
Hispanic, 9% African American, 9% Asian; control subjects were
69% Caucasian, 13% Hispanic, 8% African American; ?2(3) ? 1.3,
p ? .73].
At the time of the scan, 4 manic subjects were not taking any
medication, and 7 manic subjects were taking a range of medi-
cations, including lithium (n ? 1) anticonvulsants (divalproex
sodium, lamotrigine; n ? 6), and antipsychotics (olanzapine; n ?
2) to treat their mania. The mean YMRS score for the 11 manic
subjects at the time of the scan was 16.9 ? 3.9, and the mean
21-item Hamilton Depression Rating Scale score was 5.36 ? 4.41.
Mean duration of the current manic episode at the time of the
scan was 6.3 ? 4.29 weeks.
Magnetic resonance imaging scans were obtained on a 3-T
instrument (General Electric, Waukesha, Wisconsin) with echo
planar imaging (EPI) capability (Advanced NMR Systems, Wil-
mington, Massachusetts). Functional MRI (fMRI) scanning was
conducted with a gradient echo, echo planar acquisition se-
quence. First, an automated shim procedure was applied to
maximize magnetic field homogeneity. Second, a sagittal scout
(T2weighted) was obtained to identify locations for both struc-
tural and functional images. Third, coplanar EPI high-resolution
structural images were obtained, consisting of 26 slices (time to
repetition/time to echo [TR/TE] ? 4000 msec/54 msec, 4 mm
thick, 1-mm gap, matrix 1282, field of view [FOV] ? 20 cm)
coplanar to the functional imaging scans. Finally, functional
images were obtained with an asymmetric spin echo sequence
(Hoppel et al 1993). This sequence was used to reduce suscep-
tibility artifacts and covered 16 slices from the midtemporal lobe
region and upward (TR/TE/180° pulse offset ? 2500 msec/70
msec/25 msec, 4 mm thick, 1-mm gap, matrix 642, FOV ? 20 cm).
The Go-NoGo paradigm was used to assess orbitofrontal
activation. A central feature of the task is the requirement to
inhibit a prepotent motor response. This task also requires the
recruitment of anterior cingulate cortex and prefrontal cortex
function related to attention response conflict (Cabeza and
Nyberg 2000). Prior imaging studies in normal subjects have
reported selective activation in the orbitofrontal cortex (Brod-
mann’s areas [BA] 10, 11, 47) and cingulate (BA 24, BA 31)
during the response inhibition component (i.e., the NoGo
minus the Go task) of the paradigm (Elliott et al 2004; Horn et al
2003; Kawashima et al 1996b).
During both the control and experimental tasks, subjects
monitored a sequence of letters presented visually one at a time,
evaluated their identity, and responded to a target by pressing or
not pressing a button. Before beginning the task, subjects were
instructed to use their right index finger to press the key of a
button box. The task began with a 30-sec rest block followed by
eight alternating 30.5-sec blocks of Go and NoGo conditions,
ending with a 30-sec rest block. During the rest block, subjects
passively viewed the word “Rest” at the center of a white screen.
During the experiment, each condition was preceded by an
instruction that lasted 2.5 sec. The Go (control) condition was
preceded by the instruction “Press for all Letters.” In the control
condition, subjects were presented with a series of random letters,
to which they would press the button. The NoGo (experimental)
condition was preceded by the instruction “Press for all except X.”
During the NoGo condition, subjects were shown random letters
50% of the time and the letter “X” 50% of the time. Subjects were
instructed to press the button for each letter as it appeared on the
screen but to refrain from pressing the button for the letter “X.”
The order of the appearance of the letter “X” in the experimental
block was random. Thus, the task required the subject to
sometimes respond and sometimes refrain from responding to a
trigger letter (in this case the letter “X”). Within each condition
(Go and NoGo), stimulus presentation was .5 sec, with an
interstimulus interval of 1.5 sec, so that the subjects would see a
letter appear on the screen every 2 sec.
Performance Data. Response times and accuracy of perfor-
mance of the task were recorded for patients and control subjects
for both the Go and NoGo conditions. Differences between
groups on each task were assessed with a mixed-effects analysis
of variance model (unconstrained covariance matrix), with diag-
nosis as a grouping variable and task as a repeated measure.
Preprocessing and Statistical Parametric Mapping.
functional image volumes were examined closely for time points
containing severe motion or spike artifacts. Single corrupted
volumes (i.e., those containing spike artifacts and volume series
or runs containing significant head motion of 2 voxels or greater)
were removed from further analysis. Head motion correction and
spatial normalization were performed with automated image
registration (AIR) tools (Woods et al 1998). First, the images from
the high-resolution echo planar anatomic scans were aligned
automatically to a site-specific atlas (Woods et al 1999). The
coplanar functional scans were concatenated and corrected for
linear head motion with a six-parameter algorithm in AIR. After
this, the data were smoothed with a 6-mm full width at half
maximum Gaussian kernel. The high-resolution file was then
resampled to match the functional files. Next, all transformation
parameters from realignment and spatial normalization were
applied to the functional files, which were now realigned to
correct for head motion in atlas space. Within-subject masking
was then applied, retaining only those voxels for which there
was signal in all images/scans.
The group preprocessing consisted of cropping images not
shared across all subjects (i.e., eliminating planes that did not
have brain images across all subjects). Next we cropped the
functional files and then processed the group data statistically
with statistical parametric mapping (SPM). Contrasts were run in
The NoGo compared with Go condition (NoGo minus Go)
was used in the current analysis to evaluate degree of brain
activation specific for response inhibition, as opposed to general
attention or processes associated with letter identification and
motor output. Contrasts were first made for the NoGo minus Go
comparison within each group (patients and control subjects
separately). The output from this analysis was then entered into
a second-level analysis with subject as a random factor (random-
effects analysis). Random-effects comparisons were constrained
764 BIOL PSYCHIATRY 2005;58:763–769
L. Altshuler et al
with a mask, such that only voxels demonstrating significant
activity in the within-group analyses were entered into the
between-groups comparisons. This approach minimizes false-
positive errors due to random differences in pixel values be-
tween groups and reduces the need to correct for multiple
In addition, to determine whether the duration and the
severity of the manic episode affected fMRI activity, we per-
formed correlational analyses that identified brain voxels whose
magnitude of activation in the NoGo versus Go comparisons
were significantly related to duration (measured by weeks of the
episode) and severity (measured by the YMRS scores).
Performance data (response time and accuracy) are given in
Table 1. For response times, there was the expected significant
main effect of task, with response time being significantly faster
in both groups during performance of the Go task compared
with the NoGo task [F(1,18) ? 16.31, p ? .0009]. There was,
however, no significant main effect of diagnosis (patient vs.
control groups) on response times during performance of either
the Go or NoGo tasks [F(1,18) ? .07, p ? .8], and there was no
significant diagnosis ? task interaction [F(1,18) ? 2.17, p ? .16].
Similarly, in regard to accuracy, there was the expected main
effect of task, with the Go task being performed significantly
more accurately than the NoGo task in both groups [F(1,18) ?
6.42, p ? .02]. But again, there was no main effect of diagnosis on
the accuracy of performance for either the Go or NoGo tasks
[F(1,18) ? .8, p ? .38] (although accuracy was more variable in
the manic subjects), and there was no significant task ? diagno-
sis interaction [F(1,18) ? 1.24, p ? .28].
SPM Analyses of fMRI Results During NoGo Minus Go
Table 2 indicates the location, spatial extent, and magnitude
of activation separately for manic and control subjects, based on
the within-group SPM analysis uncorrected for multiple compar-
isons. When significance levels were set at p ? .001 corrected for
multiple comparisons, no activation in any brain region was seen
in subjects with mania. Even when significance thresholds were
lowered to p ? .001 uncorrected for multiple comparisons,
manic subjects exhibited no significant regions of activation with
an extent threshold of 10 pixels or more. Thus, lack of activation
in the manic group could not be explained readily by the use of
an overly stringent statistical threshold. Evaluation by SPM of the
four manic subjects who were not taking medications continued
to demonstrate this blunted activation in lateral orbitofrontal
cortex (BA 47) (z ? 4.16 for manic vs. z ? 7.36 for control
Control subjects showed significant activation in the right
lateral orbitofrontal cortical region (BA 47), right lateral prefrontal
cortex (BA 45), and right hippocampus that seemed not to be
activated in the manic group. Both groups also demonstrated
significant activation in the left rostral cingulate (BA 24), although
manic subjects’ activation was attenuated (k ? 10) compared with
control subjects. Figure 1 demonstrates SPM orbitofrontal activa-
tion and hippocampal activation in the control versus manic
Table 3 and Figure 2 illustrate results from the random-effects
analyses of the control versus manic subjects masked for the
control region of activation in the NoGo minus Go task. The
differences in activation between control subjects and manic
Table 1. Mean Behavioral Scores for Reaction Time and Accuracy
ManicControl Manic Control
Reaction Times (msec)
Accurate Responses (%)
.39 ? .07
93 ? 19
.43 ? .11
99 ? 2
.50 ? .06
91 ? 10
.48 ? .07
92 ? 7
Data are presented as mean ? SD.
aBehavioral data unavailable on two manic and two control subjects.
Table 2. Regions of Significant Activation Within Groups from SPM on the NoGo Compared with Go Tasks
(NoGo Minus Go)a
Maximally Activated Voxel Coordinates
Bipolar, Manic (n ? 11)Control Subjects (n ? 13)
kxyzz Scorekxyzz Score
L BA 47
R BA 47
L BA 10
R BA 10
L BA 45
R BA 45
L BA 24
R BA 24
107 36 28
(7) (36) (36)(10) (4.27)
12 38284 4.60
?14 34 5.01
SPM, statistical parametric mapping; L, left; R, right; BA, Brodmann’s area.
aHeight threshold T ? 3.73, extent threshold ?10 voxels, p ? .0001, uncorrected. Values in parentheses show
subthreshold activation (k ? 10).
L. Altshuler et al
BIOL PSYCHIATRY 2005;58:763–769 765
subjects remained significant with random-effects analyses in
right BA 47 and right hippocampus, as well as left cingulate (BA
24) (p ? .01, uncorrected).
Relationship of fMRI Activity and Episode Duration
Correlational analyses showed no significant relationship
between mania severity, as measured by the YMRS, and fMRI
activity on the NoGo versus Go comparison. There was, how-
ever, a significant negative correlation between the duration of
the manic episode and fMRI activity, such that patients with the
longest duration of the current episode showed the least activity
in the right frontal lobe, whereas those with shorter episode
duration had relatively greater fMRI increases in this region (k ?
20 voxels; x, y, z ? 38, 38, 8; p ? .005, uncorrected).
Performance of the NoGo task requires behavioral inhibition,
in that prepotent responses to distractors must be suppressed.
The neural substrates involved in this inhibition response have
previously been reported to include activation of ventral prefron-
tal regions and specifically the right lateral orbitofrontal cortex in
normal subjects (Cabeza and Nyberg 2000; Casey et al 1997;
Elliott et al 2004; Garavan et al 2002; Horn et al 2003; Kawashima
et al 1996a). Animal studies have shown single unit neuronal
firing in specific prefrontal cortical regions in relation to the
NoGo response during the Go-NoGo task (Kubota and Komatsu
1985; Sakagami and Niki 1994; Watanabe 1986). These animal
findings are similar to the human findings of the current study, in
which a strong activation of the right prefrontal cortex was seen
in relation to the NoGo minus Go conditions (e.g., response
The increased blood oxygenation level–dependent response
seen in our control group while performing the NoGo task was
robust and was similar to that previously reported. Our results
suggest that there are specific cortical fields in the right prefrontal
cortex that are activated in the generation of the NoGo response
in normal subjects but not in manic subjects. Subjects with mania
demonstrated significantly less magnitude in signal intensity in
the right orbital region compared with the control subjects. Other
functional imaging studies of subjects with mania in which
cognitive probes of frontal lobe function were used have simi-
larly reported an attenuation in orbital activity. In one study in
word-generation activation paradigm was used, decreases in
orbitofrontal activity bilaterally during rest and a decrease in right
rostral and orbital prefrontal cortex activity during activation
were found in 5 manic subjects (Blumberg et al 1999). In another
PET activation study assessing the role of the frontal lobe in a
decision-making task, Rubinsztein et al (2001) found that task-
related activation was decreased in the right frontopolar (BA 10)
and right lateral orbital (BA 47) regions in 6 manic subjects
compared with 10 control subjects. In a recent fMRI study in
which the Stroop paradigm was used to measure activation in the
prefrontal cortex, Blumberg et al (2003) found a relative decrease
in right prefrontal cortical activation in 11 manic subjects com-
pared with 15 euthymic subjects, suggesting that the finding is
state related. Most recently, Elliott et al (2004) used the Go-NoGo
paradigm to assess 8 manic and 11 control subjects. Manic
subjects demonstrated an attenuated orbitofrontal response when a
semantic task, similar to our design, was given. Given the small
number of activation imaging studies performed to date with
manic subjects, the degree of overlap in findings is striking. Our
study adds to this literature.
The functional significance of decreased orbitofrontal activa-
tion in mania—and the meaning of the negative correlation
between weeks manic and this activation—is unclear. Neuroim-
aging studies have demonstrated a role for medial and lateral
regions of the orbitofrontal cortex in mood regulation (Baker et al
15O positron emission tomography (PET) during a
Figure 1. Statistical parametric mapping results for control and manic sub-
jects on NoGo minus Go task. Top: Activation in control subjects (n ? 13)
during NoGo minus Go condition (p ? .0001, uncorrected, k ? 10 voxels).
condition (p ? .0001, uncorrected, k ? 10 voxels).
versus manic subjects on NoGo minus Go task. Statistical parametric map-
of control subjects with that of manic patients on NoGo minus Go task.
Results were masked with activation of control subjects on the same
Table 3. Regions of Significantly Greater SPM Activation in Controls
Versus Manic Subjects on Random-Effects Analysis: The NoGo Minus
R BA 47 (OFC)
L BA 24 (cingulate)
OFC, orbitofrontal cortex.
aMasked for control regions of activation (height threshold T ? 2.51,
extent threshold k ? 10 voxels, p ? .01, uncorrected).
766 BIOL PSYCHIATRY 2005;58:763–769
L. Altshuler et al
1997; Northoff et al 2000) and in associative emotional memory
functions (Bookheimer 2002; Cabeza and Nyberg 2000; Dapretto
and Bookheimer 1999; Price 2003). The orbitofrontal cortex also
plays an important role in the regulation of aggressive behavior,
and activation of BA 10 and BA 47 might enable persons to
inhibit aggressive behavioral responses (Dougherty et al 1999,
2004; Pietrini et al 2000). Lesions in orbital prefrontal cortex often
have resulted in behavioral disinhibition: dramatic behavioral
changes that resemble mania, including impulsivity, poor plan-
ning, poor judgment, irritability, and high risk-taking, reckless
behaviors (Clark and Davison 1987; Fuster 2001; Joseph 1986;
Paradiso et al 1999; Starkstein et al 1987; Stuss 1991).
How attenuation of orbitofrontal activation might pathophysi-
ologically be associated with manic symptomatology requires
further exploration. It is possible that a defect in frontal lobe
functioning might have effects in neural circuits that regulate
mood. In primates, reciprocal connections exist between the
lateral edge of the orbitofrontal cortex and the medial prefrontal
network (Amaral and Price 1985; Carmichael and Price 1995,
1996). Interestingly, the amygdala contributes as one component
in this distributed neural network (Price 2003). Medial and
ventrolateral orbitoprefrontal areas exchange sensory informa-
tion through extensive reciprocal connections with the amyg-
dala, anterior temporal, and anterior cingulate brain regions
(Mega et al 1997; Ongur and Price 2000). Several groups have
reported that bipolar disorder might be associated with alter-
ations in structure or function in the anterior limbic network
(Altshuler et al 1998; Blumberg et al 2003; Ketter et al 2001;
Strakowski 2002; Strakowski et al 1999). Recent work by our
group found a significantly greater activation of the amygdala in
manic versus control subjects while performing a task that
normally activates the amygdala (Altshuler et al 2005). Hariri et al
(2000) have shown that the right prefrontal cortex modulates
(inhibits) the intensity of the amygdala response bilaterally to
stimuli that usually activate this brain structure. In light of these
findings, our current finding of reduced activation in the right
lateral orbital prefrontal cortex is intriguing when considering the
etiology of our recent findings of amygdala hyperreactivity.
Although it is possible that amygdala hyperreactivity is part of a
primarily pathologic process in mania, it is alternatively possible
that a primary deficit (hypoactivity) in a brain region, such as the
orbitofrontal cortex, that might normally exert an inhibitory/
modulatory effect on the amygdala could result in a disruption of
a primarily inhibitory prefrontal–amygdala circuit. One clinical
result (symptom) of this functional neuroanatomic dysregulation
could be an increase in impulsivity or unstable mood.
The cause of reduced activation of the orbitofrontal cortex in
mania also requires further study. An attenuated functional
response of frontal lobe activation in mania could occur owing to
structural dysfunction. Deficits in white matter volume (Kieseppa
et al 2003) and white matter tracts (Adler et al 2004) have both
been recently reported in the prefrontal region of subjects with
bipolar disorder. Disruption in the integrity of white matter tracts
connecting specific areas in the frontal lobe to other brain
regions could result in an apparent functional frontal lobe
deficiency, even if neurons in the orbitofrontal lobe are intact.
Further evaluation of the underlying reasons for orbitofrontal
hypoactivity is needed.
Manic subjects in our study also demonstrated an attenuated
response in left cingulate and right hippocampus compared with
normal control subjects. The attenuation in cingulate response
of distractibility seen in manic subjects. A role for the anterior
cingulate in modulating attention and a role for the hippocampus in
memory have been well described. Interactions between atten-
tional and emotional brain networks have been described and
are believed to be neuroanatomically moderated through the
anterior cingulate (Mayberg et al 1999; Mega et al 1997; Stra-
kowski et al 2004; Yamasaki et al 2002). Several recent studies in
euthymic bipolar subjects have demonstrated persistent atten-
tional difficulties in bipolar subjects even during euthymia (Clark
and Goodwin 2004; Clark et al 2002). Additional euthymic
subjects seem to have limbic system circuits that are overly active
compared with control subjects when performing non-emo-
tional, attentional tasks (Strakowski et al 2004). These studies
suggest that both attentional difficulties and limbic hyperreactiv-
ity might be trait related rather than state related, which might
represent a disturbed neural circuitry in subjects with bipolar
disorder. The cingulate changes might play a role in these clinical
findings. How these systems contribute to vulnerability to mood
episodes is currently not known. Furthermore, the interactive
associations, if any, between alterations in activity in these brain
regions and the attenuation seen in the orbitofrontal region
remain to be further studied.
Several limitations exist in the present study. First, the number
of patients scanned in our study was small; however, in previ-
ously published studies that, like ours, involve the use of
activation paradigms in subjects who are manic in the scanner,
the number of subjects reported has been 5 (Blumberg et al
1999), 6 (Rubinsztein et al 2001), 11 (Blumberg et al 2003), and
8 (Elliott et al 2004). The logistical problem of having a manic
patient remain at rest for a period of time no doubt accounts for
the relatively small number of imaging studies, as well as the
small group sizes of subjects in the manic phase of bipolar
disorder in each of these studies. In this regard, our study is no
exception, but it nonetheless represents one of the larger studies
involving manic subjects (n ? 11). Despite the small number of
studies and the small number of subjects in each study, a pattern
indicating pathological function in orbitofrontal cortex during
mania has consistently been reported. In our study, no significant
correlations were found, however, between severity of mania
and any SPM regional activation. The range of YMRS scores was
narrow because the more severely manic subjects had data that
could not be included in the current study owing to motion
artifact. Thus, if there were a relationship, it might not have been
revealed because of the restricted range of YMRS scores.
A second limitation of the present study is that most of the
manic patients studied were taking antimanic medications at the
time of scanning, and the impact of these medications on cortical
blood flow has not been well established. Divalproex sodium
(the most commonly used medication by the manic group) and
lithium have been shown to either decrease cerebral blood flow
(Gaillard et al 1996; Leiderman et al 1991) or to have no effect
(Oliver and Dormehl 1998; Theodore 2000). It is possible that
medication contributed to our finding of decreased frontal
responsivity; however, the signal intensity changes in the four
subjects taking no medication also showed blunted response. A
third limitation of the present study is that the manic subjects
were slightly (not significantly) older, and this could have
contributed to a reduced response. A fourth possibility for
reduced activation could be that manic subjects were attending
less to the task; however, response times of patients and control
subjects during the task suggest that all subjects were attending
to the task and make this explanation of our findings unlikely.
Mania is a mood state associated with overall disinhibition.
This manifests itself as an increased reactivity to social and
L. Altshuler et al
BIOL PSYCHIATRY 2005;58:763–769 767
emotional stimuli, increased impulsivity, and increased motor
activity. The present study adds to converging data suggesting
orbitofrontal dysfunction during mania. A blunting of orbitofron-
tal function might be a physiologic marker of some of the
symptoms seen during mania. Functional neuroimaging studies
involving activation paradigms that probe orbitofrontal function
in bipolar depression and euthymia might help distinguish state
versus trait functional neuroanatomic abnormalities in this brain
region. A lack of activation in an orbitofrontal region might affect
other brain regions (e.g., amygdala, basal ganglia, hippocampus,
cingulate) in ways that result in the composite of emotional,
motoric, and attentional symptoms seen in mania. Studies involv-
ing paradigms that specifically activate other brain regions are
needed to help evaluate the neural circuitry in mania. Whether
the blunted activation in orbitofrontal cortex is a primary source
of neurobiologic, pathophysiologic vulnerability to developing
mania or represents a pathologic change that has occurred as a
result of a primary pathologic process elsewhere in the brain
(e.g., the cingulate, amygdala) remains to be further elucidated.
Funding for this study was provided by The Stanley Medical
Research Institute, the National Alliance for Research on Schizo-
phrenia and Depression, and the National Institute of Mental
Health (grant K24 MH01848). For generous support, we also
thank the Brain Mapping Medical Research Organization, Brain
Mapping Support Foundation, Pierson-Lovelace Foundation, The
Ahmanson Foundation, Tamkin Foundation, Jennifer Jones-
Simon Foundation, Capital Group Companies Charitable Foun-
dation, Robson Family, Northstar Fund, the National Institute of
Drug Abuse (grant DA13054), and the National Center for
Research Resources (grants RR12169, RR13642, and RR08655).
We thank Joaquin Fuster, M.D., Ph.D., for consultation
during the development of this manuscript, and Teri English for
editorial assistance during manuscript preparation.
Adler CM, Holland SK, Schmithorst V, Wilke M, Weiss KL, Pan H, et al (2004):
Abnormal frontal white matter tracts in bipolar disorder: A diffusion
tensor imaging study. Bipolar Disord 6:197–203.
al Mousawi AH, Evans N, Ebmeier KP, Roeda D, Chaloner F, Ashcroft GW
(1996): Limbic dysfunction in schizophrenia and mania. A study using
18F-labelled fluorodeoxyglucose and positron emission tomography.
(2005): Increased amygdala activation during mania: A functional mag-
netic resonance imaging study. Am J Psychiatry 162:1211–1213.
Altshuler LL, Bartzokis G, Grieder T, Curran J, Mintz J (1998): Amygdala
enlargement in bipolar disorder and hippocampal reduction in schizo-
phrenia: An MRI study demonstrating neuroanatomic specificity. Arch
Amaral DG, Price JL (1985): Amygdalo-cortical projections in the monkey
Baker SC, Frith CD, Dolan RJ (1997): The interaction between mood and
cognitive function studied with PET. Psychol Med 27:565–578.
(1985): Cerebral metabolic rates for glucose in mood disorders. Studies
with positron emission tomography and fluorodeoxyglucose F 18. Arch
Baxter LR Jr, Schwartz JM, Phelps ME, Mazziotta JC, Guze BH, Selin CE, et al
(1989): Reduction of prefrontal cortex glucose metabolism common to
three types of depression. Arch Gen Psychiatry 46:243–250.
Blumberg HP, Leung HC, Skudlarski P, Lacadie CM, Fredericks CA, Harris BC,
et al (2003): A functional magnetic resonance imaging study of bipolar
disorder: State- and trait-related dysfunction in ventral prefrontal corti-
Blumberg HP, Stern E, Martinez D, Ricketts S, de Asis J, White T, et al (2000):
Increased anterior cingulate and caudate activity in bipolar mania. Biol
Blumberg HP, Stern E, Ricketts S, Martinez D, de Asis J, White T, et al (1999):
Rostral and orbital prefrontal cortex dysfunction in the manic state of
Bookheimer S (2002): Functional MRI of language: New approaches to un-
derstanding the cortical organization of semantic processing. Annu Rev
PET and fMRI studies. J Cogn Neurosci 12:1–47.
prefrontal cortex in macaque monkeys. J Comp Neurol 363:615–641.
Casey BJ, Castellanos FX, Giedd JN, Marsh WL, Hamburger SD, Schubert AB,
et al (1997): Implication of right frontostriatal circuitry in response inhi-
Clark AF, Davison K (1987): Mania following head injury. A report of two
cases and a review of the literature. Br J Psychiatry 150:841–844.
Clark L, Goodwin GM (2004): State- and trait-related deficits in sustained
Damasio AR (2000): A neural basis for sociopathy. Arch Gen Psychiatry 57:
Dapretto M, Bookheimer SY (1999): Form and content: Dissociating syntax
and semantics in sentence comprehension. Neuron 24:427–432.
Ventromedial prefrontal cortex and amygdala dysfunction during an
anger induction positron emission tomography study in patients with
major depressive disorder with anger attacks. Arch Gen Psychiatry 61:
Anger in healthy men: A PET study using script-driven imagery. Biol
Drevets WC, Videen TO, Price JL, Preskorn SH, Carmichael ST, Raichle ME
(1992): A functional anatomical study of unipolar depression. J Neurosci
Abnormal ventral frontal response during performance of an affective
go/no go task in patients with mania. Biol Psychiatry 55:1163–1170.
Fuster JM (1989): The Prefrontal Cortex. New York: Raven Press.
Fuster JM (2001): The prefrontal cortex—an update: Time is of the essence.
Gaillard WD, Zeffiro T, Fazilat S, DeCarli C, Theodore WH (1996): Effect of
valproate on cerebral metabolism and blood flow: An 18F-2-deoxyglu-
cose and 15O water positron emission tomography study. Epilepsia 37:
Garavan H, Ross TJ, Murphy K, Roche RA, Stein EA (2002): Dissociable exec-
utive functions in the dynamic control of behavior: Inhibition, error
detection, and correction. Neuroimage 17:1820–1829.
Goodwin GM, Cavanagh JT, Glabus MF, Kehoe RF, O’Carroll RE, Ebmeier KP
(1997): Uptake of 99mTc-exametazime shown by single photon emis-
sion computed tomography before and after lithium withdrawal in bi-
polar patients: Associations with mania. Br J Psychiatry 170:426–430.
Gyulai L, Alavi A, Broich K, Reilley J, Ball WB, Whybrow PC (1997): I-123
iofetamine single-photon computed emission tomography in rapid cy-
cling bipolar disorder: A clinical study. Biol Psychiatry 41:152–161.
Hamilton M (1960): A rating scale for depression. J Neurol Neurosurg Psychi-
Hariri AR, Bookheimer SY, Mazziotta JC (2000): Modulating emotional re-
Hoppel BE, Weisskoff RM, Thulborn KR, Moore JB, Kwong KK, Rosen BR
Horn NR, Dolan M, Elliott R, Deakin JF, Woodruff PW (2003): Response inhi-
bition and impulsivity: An fMRI study. Neuropsychologia 41:1959–1966.
Joseph R (1986): Confabulation and delusional denial: Frontal lobe and
Kawashima R, Itoh H, Ono S, Satoh K, Furumoto S, Gotoh R, et al (1996a):
Changes in regional cerebral blood flow during self-paced arm and
finger movements. A PET study. Brain Res 716:141–148.
768 BIOL PSYCHIATRY 2005;58:763–769
L. Altshuler et al
Kawashima R, Satoh K, Itoh H, Ono S, Furumoto S, Gotoh R, et al (1996b):
Functional anatomy of GO/NO-GO discrimination and response selec-
tion—a PET study in man. Brain Res 728:79–89.
Ketter TA, Kimbrell TA, George MS, Dunn RT, Speer AM, Benson BE, et al
(2001): Effects of mood and subtype on cerebral glucose metabolism in
treatment-resistant bipolar disorder. Biol Psychiatry 49:97–109.
(2003): Reduced left hemispheric white matter volume in twins with
bipolar I disorder. Biol Psychiatry 54:896–905.
Kishimoto H, Takazu O, Ohno S, Yamaguchi T, Fujita H, Kuwahara H, et al
(1987): 11C-glucose metabolism in manic and depressed patients. Psy-
Kubota K , Komatsu H (1985): Neuron activities of monkey prefrontal cortex
during the learning of visual discrimination tasks with GO/NO-GO per-
formances. Neurosci Res 3:106–129.
Leiderman DB, Balish M, Bromfield EB, Theodore WH (1991): Effect of val-
proate on human cerebral glucose metabolism. Epilepsia 32:417–422.
ing PET findings in depression and normal sadness. Am J Psychiatry
Mega MS, Cummings JL, Salloway S, Malloy P (1997): The limbic system: An
anatomic, phylogenetic, and clinical perspective. J Neuropsychiatry Clin
Northoff G, Richter A, Gessner M, Schlagenhauf F, Fell J, Baumgart F, et al
(2000): Functional dissociation between medial and lateral prefrontal
cortical spatiotemporal activation in negative and positive emotions: A
combined fMRI/MEG study. Cereb Cortex 10:93–107.
O’Connell RA, Van Heertum RL, Luck D, Yudd AP, Cueva JE, Billick SB, et al
(1995): Single-photon emission computed tomography of the brain in
acute mania and schizophrenia. J Neuroimaging 5:101–104.
Oliver DW, Dormehl IC (1998): Cerebral blood flow effects of sodium val-
proate in drug combinations in the baboon model. Arzneimittelfors-
imaginal aggressive behavior assessed by positron emission tomogra-
phy in healthy subjects. Am J Psychiatry 157:1772–1781.
Rubinsztein JS, Fletcher PC, Rogers RD, Ho LW, Aigbirhio FI, Paykel ES, et al
(2001): Decision-making in mania: A PET study. Brain 124:2550–2563.
Sakagami M, Niki H (1994): Spatial selectivity of go/no-go neurons in mon-
key prefrontal cortex. Exp Brain Res 100:165–169.
Spitzer RL, Williams JB, Gibbon M, First MB (1996): Structured Clinical Inter-
view for DSM-IV. New York: Biometrics Research Department, NYC Psy-
Starkstein SE, Boston JD, Robinson RG (1988): Mechanisms of mania after
brain injury. 12 case reports and review of the literature. J Nerv Ment Dis
(1990): Mania after brain injury: Neuroradiological and metabolic find-
ings. Ann Neurol 27:652–659.
Starkstein SE, Pearlson GD, Boston J, Robinson RG (1987): Mania after brain
injury. A controlled study of causative factors. Arch Neurol 44:1069–
Strakowski SM, Adler CM, Holland SK, Mills N, DelBello MP (2004): A prelim-
lar disorder. Neuropsychopharmacology 29:1734–1740.
et al (1999): Brain magnetic resonance imaging of structural abnormali-
ties in bipolar disorder. Arch Gen Psychiatry 56:254–260.
Stuss DT (1991): Disturbances of self-awareness after frontal systems dam-
injury: Clinical and theoretical issues. New York: Oxford University Press,
Theodore WH (2000): PET: Cerebral blood flow and glucose metabolism—
pathophysiology and drug effects. Adv Neurol 83:121–130.
Watanabe M (1986): Prefrontal unit activity during delayed conditional Go/
No-Go discrimination in the monkey. II. Relation to Go and No-Go re-
sponses. Brain Res 382:15–27.
Woods RP, Dapretto M, Sicotte NL, Toga AW, Mazziotta JC (1999): Creation
ear intersubject registration, and analysis of functional imaging data.
Woods RP, Grafton ST, Holmes CJ, Cherry SR, Mazziotta JC (1998): Auto-
Yamasaki H, LaBar KS, McCarthy G (2002): Dissociable prefrontal brain
Young RC, Biggs JT, Ziegler VE, Meyer DA (1978): A rating scale for mania:
Reliability, validity and sensitivity. Br J Psychiatry 133:429–435.
L. Altshuler et al
BIOL PSYCHIATRY 2005;58:763–769 769