Amygdalofrontal Functional Disconnectivity and Aggression in Schizophrenia
Matthew J. Hoptman1,2,3, Debra D’Angelo2,
A. M. Clare Kelly5,6, Francisco X. Castellanos5,6,7, Daniel
C. Javitt3,4, and Michael P. Milham5
2Division of Clinical Research, Nathan Kline Institute,
School of Medicine, New York, NY;4Program in Cognitive
Neuroscience and Schizophrenia, Nathan Kline Institute,
Orangeburg, NY;5The Phyllis Green and Randolph C? owen
Institute for Pediatric Neuroscience, New York University Child
Study Center, New York, NY;6Department of Child and
Adolescent Psychiatry, New York University School of Medicine,
New York, NY;7Office of the Director, Nathan Kline Institute,
A significant proportion of patients with schizophrenia
demonstrate abnormalities in dorsal prefrontal regions in-
cluding the dorsolateral prefrontal and dorsal anterior cin-
gulate cortices. However, it is less clear to what extent
abnormalities are exhibited in ventral prefrontal and limbic
regions, despite their involvement in social cognitive dys-
function and aggression, which represent problem domains
for patients with schizophrenia. Previously, we found that
reduced white matter integrity in right inferior frontal
regions was associated with higher levels of aggression.
Here, we used resting-state functional magnetic resonance
imaging to examine amygdala/ventral prefrontal cortex
(vPFC) functional connectivity (FC) and its relation to ag-
gression in schizophrenia. Twenty-one healthy controls and
25 patients with schizophrenia or schizoaffective disorder
participated. Aggression was measured using the Buss
Perry Aggression Questionnaire. Regions of interest
were placed in the amygdala based on previously published
work. A voxelwise FC analysis was performed in which the
mean time series across voxels for this bilateral amygdala
seed was entered as a predictor in a multiple regression
model with motion parameters and global, cerebrospinal
fluid, and white matter signals as covariates. Patients
showed significant reductions in FC between amygdala
and vPFC regions. Moreover, in patients, the strength of
this connection showed a significant inverse relationship
with aggression, such that lower FC was associated with
higher levels of self-rated aggression. Similar results
were obtained for 2 other measures—Life History of Ag-
gression and total arrests. These results suggest that amyg-
dala/vPFC FC is compromised in schizophrenia and that
this compromise is associated with aggression.
Keywords: amygdala/functional connectivity/aggression/
connectivity.1This hypothesis has been tested using diffu-
ter integrity is disrupted in a number of different regions in
individuals with schizophrenia,2–4although the pattern of
visual evoked potential amplitudes,6hallucinations,7,8neg-
ative symptoms,9aggression and impulsivity,10,11and poor
cognitive performance.8,12Moreover, they are present in
schizophrenia populations. Taken together, these studies
provide a strong structural basis for the idea that deficits
in connectivity are implicated in the pathophysiology of
The confirmation of deficiencies in white matter integ-
rity in schizophrenia suggests that dynamic functional
interactions between brain regions would also be disrup-
ted in the disorder. These relationships can be examined
using resting-state functional connectivity (FC), which
refers to the temporal correlation of brain activity across
disparate regions. In 1995, Biswal et al15discovered that
spontaneous low-frequency oscillations in the resting-
state blood oxygen level–dependent signal were synchro-
nized across motor cortices. Since that time, a number of
different cortical and subcortical functionally connected
networks have been identified under resting-state condi-
tions, as well as during task performance, during sleep,
and under light sedation.16–20These ‘‘intrinsic connectiv-
ity networks’’ (ICNs21) correspond to functionally
relevant systems in the brain.15,16,19,22
Several studies of FC have been conducted in schizo-
phrenia. Widespread functional disconnectivity and
1To whom correspondence should be addressed; Division of
Clinical Research, Nathan Kline Institute, 140 Old Orangeburg
Road, Building 35, Orangeburg, NY 10962; tel: 845-398-6569, fax:
845-398-6566, e-mail: Hoptman@nki.rfmh.org.
Schizophrenia Bulletin vol. 36 no. 5 pp. 1020–1028, 2010
Advance Access publication on March 30, 2009
? The Author 2009. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved.
For permissions, please email: email@example.com.
decreased regional homogeneity (a measure of the
strength of correlations between voxels that are members
of the same ICN) have been found under resting condi-
tions in schizophrenia.23,24The FC of the default-mode
network (DMN) and its anticorrelated ‘‘task-positive’’
network also seem to be abnormal in schizophrenia,
and these abnormalities have been correlated with
more severe symptomatology.25,26Moreover, reduced
FC is seen between the dorsolateral prefrontal cortex
(DLPFC) and the parietal lobe, posterior cingulate, thal-
amus, and striatum of first-episode schizophrenia
patients.27In addition, reduced frontotemporal FC dur-
ing an n-back working memory task was found in med-
ication-naive patients with schizophrenia.28Finally,
patients with schizophrenia have reduced task-based
FC between DLPFC and hippocampal cortices.29
has been paid to the role of amygdalar/ventral prefrontal
cortex (vPFC) circuitry specifically,30,31despite its role in
volvement of these circuits is supported by reduced
asymmetryofwhite matter integrity in theuncinate fascic-
dial temporal regions, in patients with schizophrenia.
Amygdala/vPFC circuitry is thought to play an impor-
tant role in the suppression of impulsive/aggressive
behaviors.31,34For example, damage to the vPFC is as-
Patients with schizophrenia show increased levels of ag-
gression, which is particularly augmented by comorbid
substance abuse, although the association is present
even in nonusers of illicit substances.36Aggression in
schizophrenia has been associated with abnormal total
OFC volumes37,38and reduced white matter integrity
in right inferior frontal regions.10Similar results for orbi-
tofrontal volumes have been observed in suicidality.39
Accordingly, we sought to examine resting-state FC in
amygdala/vPFC circuitry in patients with schizophrenia
or schizoaffective disorder compared with healthy volun-
teers. In order to do so, we selected the amygdala regions
of interest (ROIs) defined by Stein et al40in their analysis
of connectivity in a frontolimbic network comprising the
cingulate, insula, DLPFC, and parahippocampal gyrus,
as well as the amygdala and OFC, which was based on
known anatomical connectivity patterns in macaques.
Based on the findings of elevated aggression in schizo-
phrenia and our prior work demonstrating that aggres-
sion was associated with reduced integrity of ventral
prefrontal white matter in patients with schizophrenia,10
we also examined the relationship between amygdala/
vPFC FC and aggressive attitudes in these same patients.
We predicted that (1) patients would show reduced FC
between amygdala and vPFC; (2) patients would show
increased levels of aggressive attitudes; and (3) among
patients, measures of aggressive attitudes would be asso-
ciated with FC deficits. To examine the specificity of the
dala, we also examined correlations between the Buss
Perry Aggression Questionnaire (BPAQ)41and FC to
other seeds in the Stein et al40network (ie, bilateral
OFC, subgenual and supragenual anterior cingulate,
and bilateral DLPFC). Because we had no a priori hy-
pothesis regarding the laterality of effects, we used bilat-
eral seeds in amygdala, OFC, and DLPFC.
met Diagnostic and Statistical Manual of Mental Disor-
ders (Fourth Edition, Text Revision) (DSM-IV-TR) cri-
teria for schizophrenia (n = 21) or schizoaffective
disorder (n = 4) after a Structured Clinical Interview
for DSM-IV-TR Axis I Disorders, Patient’s Version.42
The patient group was comprised of 18 inpatients and
7 outpatients. With regard to schizophrenia subtypes,
one patient has schizophrenia—disorganized, 7 have
schizophrenia—paranoid, 3 have schizophrenia—residual,
and 10 have schizophrenia—undifferentiated.
Participants with a lifetime history of substance depen-
dence or who met criteria for a substance abuse diagnosis
within the 6 months prior to assessment were excluded
from the study, as were participants with a history of
electroconvulsive treatment, head injury with loss of con-
or human immunodeficiency virus seropositivity. Urine
toxicology screens for substances of abuse were negative
for all subjects. Medication dosages (chlorpromazine
equivalents) were computed according to the American
Psychiatric Association guidelines.43The conversion fac-
tors for the 2 patients who were taking risperidone Consta
were obtained from the Schizophrenia Patients Outcomes
Research Team treatment recommendations (http://
are given in table 1. All participants signed informed con-
sent as approved by the local institutional review boards.
Psychopathology was assessed using the Positive and
Negative Syndrome Scale.45Aggressive attitudes were
measured using the BPAQ,41which is a 29-item self-
report scale.The total score served asthe dependent mea-
sure in this study. Data were missing for one patient on
Aggressive histories were measured using the total
score on the Life History of Aggression (LHA46), which
is an 11-item self-report interview. Data were missing for
3 subjects on this measure.
In a prior study on men with schizophrenia that
included patients with comorbid substance abuse/
Amygdala Functional Connectivity in Schizophrenia
dependence,10we found that scores on the Buss Durkee
Hostility Inventory (BDHI)47(the predecessor of the
BPAQ) were significantly correlated with arrests for vi-
olent charges as gleaned from New York State criminal
history records, which are available for inpatients at the
study site, r = .65, n = 14, P = .02. This suggests that the
BDHI could be used as a valid measure of aggression in
patients. The BPAQ, used in the current study, is a more
refined set of items derived from the BDHI. The range of
arrests for violent offenses was restricted in the current
sample(range = 0–2),possiblybecausenoneofthepatients
in the current study had significant substance abuse/
dependence histories. However, BPAQ total scores were
correlated with total arrests (range = 0–5) in the current
sample (r = .78, n = 17, P = .0002, arrest data unavailable
for 9 patients). These results support the validity of the
BPAQ as a measure for aggression in our population.
Magnetic Resonance Imaging
Scanning took place on the 1.5T Siemens Vision Scanner
netization prepared rapidly acquired gradient echo
[TR] = 11.6 ms, echo time [TE] = 4.9 ms, inversion
time [TI] = 1122 ms, matrix = 256 3 256, field of view
(FOV) = 256 mm, number of excitations (NEX) = 1, slice
thickness = 1 mm, 190 slices, no gap) and a resting-state
functional magnetic resonance imaging (fMRI) scan
(TR = 2000 ms, TE = 50 ms, matrix = 64 3 64, field of
view (FOV) = 224 mm, number of excitations (NEX) =
180, 5-mm slice thickness, 22 slices, no gap). For the rest-
ing-state scan, participants were instructed to close their
eyes and remain awake.
Resting-state data were processed as described elsewhere
in detail.19,48,49Briefly, the first 10 volumes were dis-
carded to eliminate T1 relaxation effects. Thereafter,
images were motion corrected, time shifted, and despiked
using AFNI.50Next, time series were smoothed using
a 6-mm full width-half maximum (FWHM) Gaussian
kernel, temporally filtered, and normalized to Montreal
Neurological Institute (MNI) space (1 3 1 3 1 mm3reso-
lution) using FSL (www.fmrib.ox.ac.uk/fsl).
The MPRAGE also was normalized to MNI space and
was segmented using FSL’s FAST software. The gray
matter, cerebrospinal fluid (CSF), and total brain signal
time series were then extracted using masks derived from
the MPRAGE segmentation. These time series as well as
the time series for the 6 motion parameters were used as
covariates in a general linear model (GLM).
To obtain time series for each seed in each participant, we
(1) transformed the subject’s time series into MNI space
using a 12 degrees of freedom linear affine registration
implemented in FMRIB’s Linear Image Registration
Tool (FLIRT) (voxel size = 1 3 1 3 1 mm3) and (2) calcu-
lated the spatial mean time series (across all voxels) for 2
Table 1. Demographics of Study Participants
Variables Controls (N = 21) Patients (N = 25)
Age (y)40.4 6 10.8 36.7 6 10.5
Education (y)15.5 6 3.012.3 6 2.1 .00001
105.5 6 12.194.3 6 13.5.003
50.8 6 12.5 61.1 6 17.4 .015
10.2 6 5.314.2 6 10.31.60 .061
— 78.7 616.0——
— 18.9 6 6.2——
— 20.9 6 6.3——
CPZ equivalents (mg)— 1157.8 6 627.2——
Note: FSIQ = Full-Scale IQ from Wechsler Abbreviated Scale of Intelligence,44BPAQ = Buss Perry Aggression Questionnaire total
score, LHA = Life History of Aggression total score, PANSS = Positive and Negative Syndrome Scale, CPZ = chlorpromazine.
bMissing for 2 subjects.
cMissing for 1 subject.
dMissing for 3 subjects.
eBy Fisher exact test.
M. J. Hoptman et al.
4 mm) based on the coordinates reported by Stein et al40
(amygdala coordinates: x = 626, y = 0, z = ?20). Addi-
tional identically sized spheres were placed in other parts
of the effective connectivity network identified by Stein
25) and OFC (646, 31, ?9).
For each ROI, individual participant analyses were
carried out with the GLM implemented in FSL’s
FEAT toolbox using the seed-based regression approach
employed in our prior work.19,48The time series for the
ROI as well as for the nuisance covariates (time series
regressors for global signal intensity, white matter,
CSF, and 6 motion parameters) were entered as predic-
tors. We produced individual subject-level maps of all
positively correlated voxels for each ROI seed, correcting
for multiple comparisons at the cluster level using Gauss-
ian randomfieldtheory (minimum z > 2.3;cluster signif-
icance: P < .05, corrected).
Group-level analyses were conducted using FMRIB’s
Local Analysis of Mixed Effects (FLAME), which pro-
duced thresholded z-score maps of activity associated
with each ROI. These maps revealed networks for
patients and controls, as well as difference maps (ie,
group difference maps). Talairach coordinates were de-
from the Talairach Daemon (http://www.talairach.org/
daemon.html). Surface maps of images in Talairach space
were generated using SUMA, a part of AFNI.50
Correlations between aggression and resting-state FC
were performed using a 2-tailed a level of .05.
Talairach coordinates and cluster information are given
in table 2. Across groups, FC analyses for the amygdala
seed revealed significant correlations within a very large
cluster that included bilateral lentiform nuclei and para-
area [BA] 47), and right superior temporal gyrus (STG;
Group analyses revealed marked reductions in FC
between the amygdala seed and portions of vPFC (see
figure 1). More specifically, compared with controls,
patients exhibited reduced amygdala-based FC bilater-
ally with rostral anterior cingulate cortex (BAs 24 and
32) and medial prefrontal gyrus (BA 10). They also
showed reduced amygdala-based FC with left inferior
frontal gyrus (BA 47) and middle frontal gyrus, as well
Table 2. Amgydala/Ventral Prefrontal Cortex Resting-State Activity in Controls and Patients With Schizophrenia or Schizoaffective
Patients < controls
5.22 3 10?10
Note: BA = Brodmann area, PHG = parahippocampal gyrus, Nucl = nucleus, STG = superior temporal gyrus, IFG = inferior frontal
gyrus, ACC = anterior cingulate cortex, MedFG = medial frontal gyrus.
bNumber of voxels.
Amygdala Functional Connectivity in Schizophrenia
as left lentiform nucleus. Overall, the FC regions appear
to be more diffuse in patients (figure 1, second and fourth
rows). A scatterplot of group differences in FC is
presented in figure 2.
For the OFC-seeded network, we observed significant
correlations within the inferior frontal gyrus bilaterally,
in the left supramarginal and superior frontal gyri, and in
right STG and precentral gyrus. FC was lower in patients
than controls in the insula, claustrum, STG, and anterior
cingulate cortex (BA 24) bilaterally; in the left thalamus
parietal lobule and postcentral gyrus (see Supplementary
Figure 1, Supplementary Table 1). There were no sig-
nificant group differences in FC for the subgenual ante-
rior cingulate cortex (ACC), supragenual ACC, or
DLPFC networks (see Supplementary Figures 2–4,
Supplementary Tables 2–4).
Mean BPAQ scores are given in table 1. As expected,
patients had higher scores than controls on the BPAQ
total score, t = 2.25, P = .015, one-tailed. Similar results
were obtained when the patient group was limited to only
those with a diagnosis of schizophrenia, t = 1.99,
P = .026, one-tailed.
Mean LHA scores are given in table 1. As expected,
patients trended toward higher scores than controls on
the LHA total score. Similar results were obtained
when the patient group was limited to only those with
adiagnosisofschizophrenia. TheLHAand BPAQscores
were highly correlated in both patients (r = .66, P = .001)
and controls (r = .63, P = .002).
Relations Between FC and Behavior
Across groups, amygdala/vPFC FC in the region in
which patients and controls differed correlated strongly
and negatively with BPAQ scores, r = ?.55, n = 45,
P = 8.6 3 10?5. This correlation was highly significant
in patients, r = ?.66, n = 24, P = .0004 (see figure 3; cor-
relations were similar for each hemisphere, rleft= ?.44,
P = .033; rright= ?.56, P = .005). The correlation be-
tween BPAQ and amygdala/vPFC FC in the difference
row, right lateral view.
M. J. Hoptman et al.
region was similar, r = ?.68, n = 20, P = .0009, when the
4 patients with schizoaffective disorder were excluded
regions are shown in Supplementary Tables 5 and 6 and
Supplementary Figures 5 and 6. Similar results were
obtained for number of the LHA total score (rleft= ?.24,
n = 22, P = .29; rright= ?.41, P = .057; rbilateral= ?.46,
P = .029) and total arrests (rleft= ?.34, n = 17, P = .18;
rright= ?.64, P = .005; rbilateral= ?.65, P = .005). Thus,
as predicted, higher levels of aggression in patients were
correlated with lower amygdala/vPFC FC.
The correlation with BPAQ was not significant in con-
trols in the region of group differences for the left or bi-
lateral amygdala seeds (rleft= ?.12, n = 21, P = .60;
rbilateral = ?.14, P = .54) but approached significance
for the right-sided seed, r = ?.37, P = .10. Correlations
with LHA were not significant for right or bilateral seeds
(r’s > ?.28, n = 21, P’s > .21) but approached signifi-
cance for the left-sided seed (rleft= ?.37, P = .10).
Total scores on the BPAQ were not significantly cor-
related with FC in OFC control-patient difference
regions or with FC associated with supragenual cingu-
late, subgenual cingulate, or DLPFC regions (there
were no group differences for those latter seeds). None
of the correlations were significant for either group
(P’s > .15).
Partial correlation analyses showed that the relation-
ship between total BPAQ and amygdala/vPFC FC re-
gression coefficients in patients remained significant
even after accounting for age. Furthermore, neuroleptic
medication dosage did not account for this relationship
for the patient group.
To further specify the location of the BPAQ/FC rela-
tionship for the amygdala-seeded network, we entered
the demeaned BPAQ score in a voxelwise analysis on
the amygdala/vPFC data for patients, with the search
space limited only to the region in which patients had
lower FC than controls. Scores on the BPAQ were signif-
icantly correlated with the regression coefficient for
this contrast in an 865-voxel cluster with cluster maxima
at (?8, 20, 1) that was significant at P = 2.6 3 10?5, cor-
rected, and a 373-voxel cluster at (19, 57, ?2), P = .016,
We found that FC between the amgydala and vPFC is
disrupted in medicated patients with schizophrenia and
schizoaffective disorder. Moreover, in patients, amyg-
dala/vPFC FC showed a robust inverse relationship
with self-reported aggression, especially with the right-
seeded network, such that higher levels of aggressive atti-
tudes were associated with a reduced positive, trending
dition, in patients, the results were consistent across 2
other measures of aggression, albeit less strongly in the
case of the LHA. On balance, then, the results support
the idea that amygdalofrontal functional disconnectivity
is associated with aggression and antisocial behavior.
These correlational results were specific to amygdala/
vPFC FC because they were not found for resting-state
FC derived from seeds in OFC, supra- or subgenual cin-
gulate, or bilateral DLPFC.
These results are consistent with increasing evidence
that ventral prefrontal regions involved in social cogni-
tion are dysfunctional in schizophrenia.31,51Recent
work, including that from our laboratory, has shown
Fig. 3. Correlation Between Buss Perry Aggression Questionnaire
Total Score and Amygdala/Ventral Prefrontal Cortex (vPFC)
Functional Connectivity (FC) Regression Parameter Estimates
(Group Difference Region) in Patients With Schizophrenia or
Fig. 2. Scatterplot of Group Differences for Amygdala/Ventral
Prefrontal Cortex (vPFC) Functional Connectivity (FC)
Regression Parameter Estimates.
Amygdala Functional Connectivity in Schizophrenia
that aggression is associated with abnormalities in this
region. For example, in men with schizophrenia, we
found that abnormal right ventral prefrontal white mat-
ter integrity was associated with increased impulsivity
and aggression.10These effects were specific to ventral
prefrontal regions because they were not observed in
more dorsal frontal regions. Volumetric abnormalities
in OFC also have been associated with aggression.37,38
not simply a reflection of a general reduction in FC in
schizophrenia. Although FC was reduced in networks as-
sociated with OFC seeds, such was not the case for net-
Our results also contrast with findings that reduced FC
between amygdala and anterior cingulate has been asso-
ciated with depression,52suggesting that different pat-
terns of amygdalar disconnectivity result in different
symptomatologies. Similarly, abnormalities in FC be-
tween the amygdala and DLPFC have been found in re-
sponse to facial expressions of anger presented during
scanning in patients with schizophrenia.53The develop-
mental pattern of these connections is not known, but
they may reflect cortical immaturity in schizophrenia,
as has been proposed.54
The current study has a number of limitations. The re-
gion of reduced connectivity lies in a region prone to
magnetic susceptibility artifacts. However, such artifact
might diminish the likelihood of detecting group differ-
ences. Second, the amygdala is a small and heterogenous
structure that does not comprise a single functional unit.
However, the ROI chosen for the current article was
based on that used by Stein et al40in their anatomically
based effective connectivity study of the amygdala.
Moreover, at the current resolution afforded by fMRI,
it is difficult to study functional subunits. This remains
is the unconstrained nature of the resting state. We
should point out in this regard that Fransson55examined
so-called task-unrelated thoughts (concerning inner
speech, imagery, planning for the future, episodic mem-
ory, and task-unrelated attention) and found that these
were correlated with neither activation during a working
memory task nor resting-state activation in the DMN.
Moreover, the unconstrained nature of the task would
presumably add noise to the results, weakening statistical
power. This stands in contrast to the strong effects com-
monly observed in such studies.48,56Most significantly in
this regard, our group has shown short- and long-term
test-retest reliability in the moderate to high range for
FC during rest.57Finally, patient participants all had
chronic schizophrenia or schizoaffective disorders and
were all on antipsychotic medication. However, medica-
tion dosages did not account for our group differences in
voxel-based analyses or for the relationship between de-
creased FC and increased aggression. Moreover, in a
subanalysis in which patients with schizoaffective disor-
der were omitted, the correlation between total BPAQ
mates in the group difference region was almost identical
to that in the larger analysis.
The current findings show that FC between amygdala
and frontal regions are disrupted in schizophrenia and
add to a growing literature suggesting abnormalities in
both those regions. The significant correlation between
reduced amygdala/vPFC FC in patients and aggression
is consistent with our prior DTI findings in schizophre-
nia.10These results point toward the potential merits of
employing multimodal assessments of structural and
functional frontolimbic connectivity in patients with
schizophrenia in future studies.
Supplementary tables 1–6 and figures 1–6 are available at
National Institutes of Health (RO1 MH64783 to M.J.H.,
R37 MH49334 to D.C.J.); National Alliance for Re-
search on Schizophrenia and Depression (to F.X.C.);
gifts from Linda and Richard Schaps and Jill and Bob
Smith and Taubman Foundation (to F.X.C.).
We thank Raj Sangoi (RT) (R) (MR) for assistance in
assistance with image processing and analysis.
1. Friston KJ, Frith CD. Schizophrenia: a disconnection syn-
drome? Clin Neurosci. 1995;3:89–97.
2. Lim KO, Hedehus M, Moseley M, de Crespigny A, Sullivan
EV, Pfefferbaum A. Compromised white matter tract integ-
rity in schizophrenia inferred from diffusion tensor imaging.
Arch Gen Psychiatry. 1999;56:367–374.
3. Buchsbaum MS, Tang CY, Peled S, et al. MRI white matter
diffusion anisotropy and PET metabolic rate in schizophre-
nia. Neuroreport. 1998;9:425–430.
4. Ardekani BA, Nierenberg J, Hoptman MJ, Javitt DC, Lim
KO. MRI study of white matter diffusion anisotropy in
schizophrenia. Neuroreport. 2003;14:2025–2029.
5. Kubicki M, McCarley R, Westin CF, et al. A review of diffu-
sion tensor imaging studies in schizophrenia. J Psychiatr Res.
6. Butler PD, Zemon V, Schechter I, et al. Early-stage visual
processing and cortical amplification deficits in schizophre-
nia. Arch Gen Psychiatry. 2005;62:495–504.
M. J. Hoptman et al.
7. Hubl D, Koenig T, Strik W, et al. Pathways that make voices:
white matter changes in auditory hallucinations. Arch Gen
8. Szeszko PR, Robinson DG, Ashtari M, et al. Clinical and
neuropsychological correlates of white matter abnormalities
in recent onset schizophrenia. Neuropsychopharmacology.
9. Wolkin A, Choi SJ, Szilagyi S, Sanfilipo M, Rotrosen JP, Lim
KO. Inferior frontal white matter anisotropy and negative
symptoms of schizophrenia: a diffusion tensor imaging study.
Am J Psychiatry. 2003;160:572–574.
10. Hoptman MJ, Volavka J, Johnson G, Weiss E, Bilder RM,
Lim KO. Frontal white matter, aggression and impulsivity
in men with schizophrenia. Biol Psychiatry. 2002;52:9–14.
11. Hoptman MJ, Ardekani BA, Butler PD, Nierenberg J, Javitt
DC, Lim KO. DTI and impulsivity in schizophrenia: a first
voxelwise correlational study. Neuroreport. 2004;15:2467–2470.
12. Lim KO, Ardekani BA, Nierenberg J, Butler PD, Javitt DC,
Hoptman MJ. Voxelwise correlational analyses of white mat-
ter integrity in multiple cognitive domains in schizophrenia.
Am J Psychiatry. 2006;163:2008–2010.
13. Szeszko PR, Ardekani BA, Ashtari M, et al. White matter ab-
normalities in first-episode schizophrenia or schizoaffective
disorder: a diffusion tensor imaging study. Am J Psychiatry.
14. Kumra S, Ashtari M, Cervellione KL, et al. White matter ab-
normalities in early-onset schizophrenia: a voxel-based diffu-
sion tensor imaging study. J Am Acad Child Adolesc
15. Biswal B, Yetkin FZ, Haughton VM, Hyde JS. Functional
connectivity in the motor cortex of resting human brain using
echo-planar MRI. Magn Reson Med. 1995;34:537–541.
16. Damoiseaux JS, Rombouts SA, Barkhof F, et al. Consistent
resting-state networks across healthy subjects. Proc Natl
Acad Sci U S A. 2006;103:13848–13853.
17. Fransson P, Skiold B, Horsch S, Nordell A, Blennow M,
Lagercrantz H. Resting-state networks in the infant brain.
Proc Natl Acad Sci U S A. 2007;104:15531–15536.
18. Greicius MD, Menon V. Default-mode activity during a pas-
sive sensory task: uncoupled from deactivation but impacting
activation. J Cogn Neurosci. 2004;16:1484–1492.
19. Margulies DS, Kelly AM, Uddin LQ, Biswal BB, Castellanos
FX, Milham MP. Mapping the functional connectivity of an-
terior cingulate cortex. NeuroImage. 2007;37:579–588.
20. Greicius MD, Kiviniemi V, Tervonen O, et al. Persistent
default-mode network connectivity during light sedation.
Hum Brain Mapp. 2008;29:839–847.
21. Seeley WW, Menon V, Schatzberg AF, et al. Dissociable in-
trinsic connectivity networks for salience processing and exec-
utive control. J Neurosci. 2007;27:2349–2356.
22. Fox MD, Snyder AZ, Zacks JM, Raichle ME. Coherent
spontaneous activity accounts for trial-to-trial variability in
human evoked brain responses. Nat Neurosci. 2006;9:23–25.
23. Liang M, Zhou Y, Jiang T, et al. Widespread functional dis-
connectivity in schizophrenia with resting-state functional
magnetic resonance imaging. Neuroreport. 2006;17:209–213.
24. Liu H, Liu Z, Liang M, et al. Decreased regional homogene-
ity in schizophrenia: a resting state functional magnetic reso-
nance imaging study. Neuroreport. 2006;17:19–22.
25. Bluhm RL, Miller J, Lanius RA, et al. Spontaneous low-
frequency fluctuations in the BOLD signal in schizophrenic
patients: anomalies in the default network. Schizophr Bull.
26. Garrity AG, Pearlson GD, McKiernan K, Lloyd D, Kiehl KA,
Calhoun VD. Aberrant ‘‘default mode’’ functional connectiv-
ity in schizophrenia. Am J Psychiatry. 2007;164:450–457.
27. Zhou Y, Liang M, Jiang T, et al. Functional dysconnectivity of
the dorsolateral prefrontal cortex in first-episode schizophrenia
using resting-state fMRI. Neurosci Lett. 2007;417:297–302.
28. Meyer-Lindenberg A, Poline JB, Kohn PD, et al. Evidence for
ory in schizophrenia. Am J Psychiatry. 2001;158:1809–1817.
29. Meyer-Lindenberg AS, Olsen RK, Kohn PD, et al. Regionally
specific disturbance of dorsolateral prefrontal-hippocampal
functional connectivity in schizophrenia. Arch Gen Psychia-
30. Waltz JA, Gold JM. Probabilistic reversal learning impair-
ments in schizophrenia: further evidence of orbitofrontal dys-
function. Schizophr Res. 2007;93:296–303.
31. Hoptman MJ, Nolan KA. The role of prefrontal abnormali-
ties in schizophrenia. In: Javitt DC, Kantrowitz JT, eds.
Schizophrenia. 3rd edn. New York, NY: Springer; 2008.
32. Penn DL, Corrigan PW, Bentall RP, Racenstein JM, New-
man L. Social cognition in schizophrenia. Psychol Bull.
33. Kubicki M, Westin C-F, Maier SE, et al. Uncinate fasciculus
findings in schizophrenia: a magnetic resonance diffusion ten-
sor imaging study. Am J Psychiatry. 2002;159:813–820.
34. Volavka J. Neurobiology of Violence. 2nd edn. Washington,
DC: APA Publishing; 2002.
35. Grafman J, Schwab K, Warden D, Pridgen A, Brown HR,
Salazar AM. Frontal lobe injuries, violence, and aggression:
a report of the Vietnam Head Injury Study. Neurology.
36. Steadman HJ, Mulvey EP, Monahan J, et al. Violence by peo-
ple discharged from acute psychiatric inpatient facilities and
by others in the same neighborhoods. Arch Gen Psychiatry.
37. Hoptman MJ, Volavka J, Weiss EM, et al. Quantitative MRI
measures of orbitofrontal cortex in patients with chronic
schizophrenia or schizoaffective disorder. Psychiatry Res.
chiatric patients. Psychiatry Res. 2006;147:213–220.
39. Rusch N, Spoletini I, Wilke M, et al. Inferior frontal white
matter volume and suicidality in schizophrenia. Psychiatry
40. Stein JL, Wiedholz LM, Bassett DS, et al. A validated net-
workof effective amygdala
41. Buss AH, Perry M. The Aggression Questionnaire. J Pers Soc
42. First MB, Spitzer RL, Gibbon M, Williams JBW. Structured
Clinical Interview for DSM-IV-TR Axis I Disorders-Patient
Edition (SCID-I/P, 11/2002 Revision). New York, NY: Bio-
metrics Research Department, New York State Psychiatric
43. American Psychiatric Association.Practice guideline for the
treatment of patients with schizophrenia. Am J Psychiatry.
44. Psychological Corporation.Wechsler Abbreviated Scale of In-
telligence (WASI) Manual. San Antonio, TX: Psychological
45. KaySR, OplerLA, FiszbeinA.PositiveandNegative Syndrome
Amygdala Functional Connectivity in Schizophrenia
46. Coccaro EF, Berman ME, Kavoussi RJ. Assessment of life Download full-text
history of aggression: development and psychometric charac-
teristics. Psychiatry Res. 1997;73:147–157.
47. Buss A, Durkee A. An inventory for assessing different kinds
of hostility. J Consult Psychol. 1957;21:343–349.
interactions: a new locus of dysfunction in adult attention-deficit/
hyperactivity disorder. Biol Psychiatry. 2008;63:332–337.
49. Kelly AMC, Uddin LQ, Biswal BB, Castellanos FX, Mil-
ham MP. Competition between functional brain networks
mediates behavioral variability. NeuroImage. 2008;39:527–
50. Cox RW. AFNI: software for analysis and visualization of
functional magnetic resonance neuroimages. Comput Biomed
51. Brunet-Gouet E, Decety J. Social brain dysfunctions in
schizophrenia: a review of neuroimaging studies. Psychiatry
52. Pezawas L, Meyer-Lindenberg A, Drabant EM, et al. 5-
HTTLPR polymorphism impacts human cingulate-amygdala
interactions: a genetic susceptibility mechanism for depres-
sion. Nat Neurosci. 2005;8:828–834.
53. Radulescu AR, Mujica-Parodi LR. A systems approach to
prefrontal-limbic dysregulation in schizophrenia. Neuropsy-
54. Lewis DA, Levitt P. Schizophrenia as a disorder of neurode-
velopment. Annu Rev Neurosci. 2002;25:409–432.
55. Fransson P. How default is the default mode of brain func-
tion? Further evidence from intrinsic BOLD signal fluctua-
tions. Neuropsychologia. 2006.
56. Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van E,
Raichle ME. The human brain is intrinsically organized
into dynamic, anticorrelated functional networks. Proc Natl
Acad Sci U S A. 2005;102:9673–9678.
57. Shehzad Z, Kelly AMC, Reiss PT, et al. The resting brain:
unconstrained yet reliable. Cereb Cortex. In press.
M. J. Hoptman et al.