Brain Research Bulletin 114 (2015) 2030

Contents lists available at ScienceDirect

Brain Research Bulletin

journal homepage: www.elsevier.com/locate/brainresbull

Research report

Response inhibition failure to visual stimuli paired with a

single-type stressor in PTSD patients: An fMRI pilot study
Marcella Brunetti a,b, , Gianna Sepede a,b,f , Antonio Ferretti a,b , Gianluca Mingoia e ,
Gian Luca Romani a,b , Claudio Babiloni c,d
a

Institute of Advanced Biomedical Technologies, University of Chieti, Italy

Department of Neuroscience, Imaging & Clinical Science, University of Chieti, Italy
c
Department of Physiology and Pharmacology, University of Rome La Sapienza, Rome, Italy
d
IRCCS San Raffaele Pisana, Rome, Italy
e
IZKF Medicine Faculty of RWTH University, Aachen, Germany
f
Department of Basic Medical Sciences, Neurosciences and Sense Organs, University A. Moro, Bari, Italy
b

a r t i c l e

i n f o

Article history:
Received 11 June 2014
Received in revised form 28 January 2015
Accepted 5 March 2015
Available online 16 March 2015
Keywords:
Amygdala
Conditioning
Emotion
Functional magnetic resonance imaging
(fMRI)
Post-traumatic stress disorder (PTSD)

a b s t r a c t
Patients with post-traumatic stress disorder (PTSD) tend to misinterpret innocuous stimuli as potential
threats, possibly due to a conditioning provoked by traumatic episodes. Previous neuroimaging evidence
has shown an abnormal activation of the amygdala and prefrontal cortex in PTSD patients during fear
conditioning and extinction. Nevertheless, the effects of a single-type adverse stressor on that circuit
remain poorly explored. We tested the hypothesis that a single-type adverse episode is able to affect the
prefrontal cortex and amygdala response to conditioned stimuli. To test this hypothesis, fMRI recordings
were performed in PTSD patients and trauma-exposed controls during the observation of neutral and
negative paired or non-paired pictures with an adverse stimulus by means of a single association.
Results showed that left amygdala activation during negative reinforced stimuli was correlated with
the score of PTSD clinical scale across all subjects. Furthermore, in the traumatized non-PTSD group, the
activation of the dorso-medial prefrontal cortex and bilateral amygdala was lower during the observation
of the reinforced (CS+ ) versus non-reinforced pictures (CS ) in response to emotionally negative stimuli.
This was not the case in the PTSD patients. These results suggest that in PTSD patients, a single-episode
conditioning unveils the failure of an inhibitory mechanism moderating the activity of the prefrontal
cortex and amygdala in response to adverse and neutral stimuli.
2015 Elsevier Inc. All rights reserved.

1. Introduction
Patients with post-traumatic stress disorder (PTSD) experience
intensive fear due to the continuous reliving of the past trauma,
exhibit exaggerated responses to emotionally negative stimuli and
tend to misinterpret innocuous stimuli as potential threats (van der
Kolk, 1994).
Functional topography of brain activation of PTSD patients
observing emotional stimuli has been explored by neuroimaging
studies using functional magnetic resonance imaging (fMRI) and
positron emission tomography (PET). Activation of the amygdala

Corresponding author at: Institute for Advanced Biomedical Technologies &

Department of Neuroscience, Imaging & Clinical Science, University G. dAnnunzio
of Chieti Via dei Vestini, 33 66100 Chieti, CH, Italy. Tel.: +39 0871 3556935;
fax: +39 0871 3556930.
E-mail address: mbrunetti@itab.unich.it (M. Brunetti).
http://dx.doi.org/10.1016/j.brainresbull.2015.03.001
0361-9230/ 2015 Elsevier Inc. All rights reserved.

received special attention since it sub-serves the processing of

negative emotional stimuli and triggers homeostatic neurovegetative responses to stress (Pissiota et al., 2002; Lanius et al., 2006;
Williams et al., 2006). An increased amygdala response to emotional visual stimuli in PTSD patients compared to control subjects
has been shown (Rauch et al., 2000; Williams et al., 2006). This was
true not only during emotional stimuli observation but also during
neutral pictures observation (Brunetti et al., 2010). These results
suggest a hyperactivation of the amygdala in response to neutral
stimuli in PTSD patients.
Moreover, in previous neuroimaging studies the activation of
the amygdala during the processing of emotional stimuli was typically accompanied by the activation of prefrontal areas. It has been
hypothesized that a controlled co-activation of the amygdala and
dmPFC might sub-serve an adaptive mechanism by which vigilance
towards potentially threatening stimuli is modulated. An abnormal
regulation of this mechanism might lead to a misinterpretation
of external stimuli and uncontrolled emotional and vegetative

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

reactions (Robinson et al., 2012). Previous results demonstrated

that the left amygdala and dorsal anterior cingulate cortex (dACC)
showed a very strong activation in response to salient stimuli
in PTSD patients as a possible mechanism of hypervigilance or
maladaptive attention and learning (Bryant et al., 2005). Dorsal
ACC hyperactivation in response to contextual information during
fear extinction in PTSD patients was also observed (RougemontBcking et al., 2011). In addition, a resting state study observed a
reduced anti-correlation between the amygdala and dorsal ACC in
these patients (Sripada et al., 2012). Overall, the role of the coordinated prefrontal cortex-amygdala activity in healthy individuals,
in PTSD patients and during conditioning has been extensively
investigated. Firstly, prefrontal cortex and amygdala involvement
in a brain circuit designated to the representation of learned
classication of stimuli through threat appraisal and emotion regulation processes has been described (Britton et al., 2011; Brunetti
et al., 2014). Furthermore, medial prefrontal cortex represents the
biological relevance of a given stimulus and may convey this information to the dorso-lateral prefrontal cortex (Mechias et al., 2010).
Amygdala may play a pivotal role in the learning processes related
to emotional reactions, fear, and fear extinction (Phelps et al., 2004;
Milad et al., 2009; Britton et al., 2011). Secondly, positive functional
connectivity was found in PTSD patients between the amygdala
and anterior cingulate cortex in response to stimulations related to
emotionally traumatic stimuli (Osuch et al., 2008) and during the
observation of fearful faces under anxious conditions in healthy
subjects (Robinson et al., 2012). Thirdly, fMRI studies investigating conditioning in PTSD patients indicated that, compared to the
control group, the PTSD group showed a greater activation of the
amygdala during extinction learning and greater activation of the
anterior cingulate cortex related to the extinction recall (extinguished conditioned stimuli presented in the absence of shock;
Milad et al., 2009). Furthermore, previous studies have investigated
fear extinction learning in healthy humans using neuroimaging
techniques (Phelps et al., 2004) and in PTSD patients by means of
physiological recordings (SCR measurement) (Milad et al., 2008).
Conventional (Pavlovian) conditioning represents a putative
mechanism triggering PTSD (Bremner et al., 2005). In conventional conditioning, a neutral stimulus (NS) producing no particular
response at a rst presentation is associated with a second stimulus
(unconditioned stimulus US), becoming a conditioned stimulus
(CS). This now elicits a learned response (conditioned response, CR).
This learning typically happens after some repetitions of the association between US and CS, and CR gradually decreases after repeated
presentation of the CS in the absence of the US (i.e. extinction learning). In their 2-day fear conditioning fMRI study in healthy subjects,
Phelps et al. (2004) used a partial reinforcement paradigm; during the acquisition phase, CS+ was paired with US whereas CS
was never paired with US. Immediately after this phase, the day 1
extinction phase followed, consisting of CS+ and CS presentation.
Approximately 24 h after the rst session, day 2 extinction phase
was administered. Interestingly, a signicantly greater amygdala
response to the CS relative to the CS+ during day 1 extinction phase
was observed (Phelps et al., 2004).
Keeping in mind the above data and considerations, testing the
hypothesis that activation of the prefrontal cortex and amygdala
is affected by a single exposure to traumatic (aversive) stimuli can
yield insights. Neuroimaging studies, indeed, indicate that PTSD
patients can develop functional and structural consequences just
after a single traumatic episode (Chen et al., 2012; Liu et al., 2012).
This is an exploratory study to preliminarily test this hypothesis.
We recorded fMRI in PTSD patients and in trauma-exposed controls during a new variant of the Pavlovian conditioning procedure.
In the present conditioning session, pictures with neutral and emotional stimuli (CSs) were or were not associated to aversive stimuli
in the framework of a single episode (i.e. USs; acoustic aversive

21

stimulus as a single exposure to each visual stimulus). In other

words, each CS was one-off associated to the US. After the conditioning session, block-design fMRI recordings were performed
during the observation of the pictures previously associated with
the aversive stimulus (CS+ ) (i.e. single-episode conditioning that
mimicked the mechanism of the single traumatic event type) as
well as during the observation of the pictures not previously associated with this stimulus (CS ) (i.e. no conditioning as a control
condition). An effect of the single-episode conditioning on the BOLD
activation of the prefrontal cortex and amygdala during the observation of the negative pictures is expected.

2. Materials and methods

2.1. Subjects
Participants were 24 bank clerks, victim of one or more armed
bank assaults about 10 months before the experiments (range
of 218 months). Inclusion criteria included right-handedness
assessed by Edinburgh Inventory (Oldeld, 1971), normal or
corrected to normal vision, and age from 20 to 50 years. Exclusion criteria comprised seizure disorder, progressive neurological
and/or systemic disorders, metallic implants, signicant unstable concurrent medical illness, hormone replacement therapy,
electroconvulsive or light therapy, administration of concomitant
medication that could alter mood or cerebral metabolism (e.g. benzodiazepines, antidepressants, mood stabilizers, stimulants, and
steroids) within 30 days prior to screening, history of any substance/alcohol abuse or dependence within the past 6 months
(nicotine dependence was allowed), pregnancy, and not understanding the experimental procedures.
Four out of the 24 participants were excluded for the following reasons: one participant was excluded due to co-morbidity
(panic attack disorder with severe claustrophobia that made the
patient unable to remain still in the MRI scanner), another patient
asked to interrupt the experiment; two subjects were excluded
due to concurrent medical illness. Ten participants who completed
the experiment and could be included in the nal analysis were
assigned to PTSD group, whereas ten age, sex and educationmatched healthy subjects were assigned to control group. Nicotine
use was comparable in the two groups.
All participants underwent an extensive clinical examination
carried out by an expert psychiatrist (GS) and two clinical psychologists (MB, GLM). A broader range of traumatic event types, including
car accidents and criminal attacks, was assessed using the event
checklist of the Clinician Administered PTSD Scale (CAPS) (Blake
et al., 1990). In the majority of subjects (N = 9 PTSD patients, N = 9
traumatized controls) the only type of traumatic event was bank
robbery; only two participants reporting other kind of traumas
(a serious car accident for one PTSD subject, witnessing an earthquake for one traumatized control). Traumas were multiple (more
than one bank robbery in 7 PTSD patients and 5 traumatized controls). The clinical features and demographic information of the two
groups and main statistical comparisons are reported in Table 1.
Standardized clinical instruments were used for the assessment
of DSM-IV diagnoses by trained researchers: CAPS (Blake et al.,
1990; CAPS Italian version: Pieraccini et al., 1999) for the diagnosis and quantication of PTSD and related dissociative features
(MB) and Mini-International Neuropsychiatric Interviews (MINI)
(Sheehan et al., 1994; MINI 5.0 Italian version: Conti et al., 1999)
for diagnoses of DSM-IV axis one disorders (GS). At the time of the
present study, participants met DSM-IV diagnostic criteria for the
following current co-morbid diagnoses: dysthymia (N = 1 PTSD subject), agoraphobia without history of panic attack disorder (N = 1
PTSD subject), and social phobia (N = 1 control subject). None of the

22

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

Table 1
Demographic and clinical characteristics of the two subject groups.
Variable

PTSD group (n = 10)

Control group (n = 10)

Mean age in years (SD)

Mean school educ. in
years (SD)
N females (%)
Trauma load [mean
number of traumatic
events (sd)]
N subjects victims of
multiple traumatic
events (%)

39.7 (6.34)
16 (2.5)

36.8 (12.1)
14 (2.1)

6 (60)
3.6 (2.9)a

6 (60)
2.5 (1.8)a

7 (70)b

5 (50)b

N subjects victims of
traumas other than
bank robberies (%)
Car accident (%)
Earthquake (%)
CAPS PTSD symptom
score, mean (SD)

conrmed by subjects after rst presentation. In spite of the

relatively low intensity of the volume, the aversive stimulus was
able to produce an automatic reaction as demonstrated by the
changes in the reaction time of task performance when it was
presented. Furthermore, in a previous study by our group autonomic responses to a similar sudden acoustic sound in traumatized
subjects were observed (Di Giacinto et al., 2014).
2.3. Experimental design

1 (10)c

1 (10)c

1 (10)
0 (0)

0 (0)
1 (10)

30.8* (11.2)

6.2* (6.9)

Note: sd = standard deviation.

*
Between-group signicant difference. One-way ANOVA result: [F(1,18) = 25.05,
p < 0.0001].
a
No between group effect. One-way ANOVA result: [F(1,18) = 1.05, p = 0.32].
b
No between group differences. Fishers exact two-tailed p = 0.65.
c
No between group differences. Fishers exact two-tailed p = 1.0.

participants had been previously treated with psychopharmacologic interventions.

All subjects received detailed explanations about the study
design and gave written informed consent according to the Declaration of Helsinki (World Medical Association Declaration of Helsinki,
1997). The protocol was approved by the local Ethics Committee
(School of Medicine Ethics Committee, University of Chieti, Italy).
2.2. Stimuli
120 colour pictures were chosen from the International Affective Picture System on the basis of their normative ratings (IAPS,
Lang et al., 1999; Catani et al., 2009). Of these, 60 pictures
depicted unpleasant scenes (e.g. mutilations, assaults, dead bodies, etc.) (IAPS rating: valence < 3, 5.3 < arousal < 7) and 60 pictures
depicted neutral contents (e.g. mushrooms, mugs, neutral faces,
etc.) (4.5 < valence < 5.5 and arousal < 4).
The pictures were presented using in-house software based
on Matlab (http://www.mathworks.com/). Stimuli were projected
over a transparent screen inside the scanner tunnel and were
viewed through a mirror system mounted on the top of the magnetic resonance imaging (MRI) head coil.
A 70 dB buzz sound was used as acoustic aversive stimulus.
The intensity of the acoustic stimulus was established considering
the experimental conditions: subjects lay inside the scanner bore
during conditioning phase. This circumstance emphasized the
perception of the subjects. A louder buzz, (100 dB) as described
in other literature and normally used as unconditioned stimulus,
was tested in a preliminary phase, but it caused excessive motor
reaction and in some cases the request of experiment interruption.
We reduced the intensity of the stimulus in order to make it tolerable. All subjects reported the acoustic startle as disturbing and
alarming. This stimulus was delivered via a non-magnetic and MRIcompatible sound system (Commander XG MRI Audio System) with
a frequency range from 100 Hz to 25 kHz. The electric signals generated by the computer audio board were amplied and sent through
the shielded room penetration panel to an electropneumatic transducer. The aversive sound was presented for 150 ms and its volume
was adapted subject by subject in order to be clearly audible and
able to cause an aversive reaction (scare or surprise) as was

The present experiment was carried out as a part of a larger

study including another scanning session (previously published
data, see Brunetti et al., 2010). The present experiment consisted of
two different experimental sessions described below which represents the second and third run of the larger session.
2.3.1. Conditioning session
120 pictures never seen before were presented according to a
boxcar design, arranged in 24 blocks consisting of 5 pictures each
(Stimulus Onset Asynchrony (SOA) = 3 s, stimulus duration = 2 s, 1 s
interval showing a cross-xation). The ve pictures of individual
blocks were either all unpleasant or all neutral and each block was
followed by a control state (cross-xation) with the same duration.
The order of the unpleasant and neutral blocks was randomized to
avoid order effects, tonic arousal, and fatigue across subjects. Every
block or control state lasted 15 s.
Each picture was presented only once to avoid habituation and
memory effects and also to simulate a single exposure to trauma
condition. Furthermore, 25% of the pictures (both unpleasant and
neutral) were conditioned by association of images with an acoustic aversive stimulus, US (described in Section 2.2); the association
between images and US occurred randomly, in order to avoid prediction of the US; the paired stimulus was not necessarily present
in every block. There are several kinds of conventional conditioning paradigms: forward conditioning, simultaneous conditioning,
second-order and higher-order conditioning, backward conditioning, and temporal conditioning (Chance, 2013). We decided to
use backward conditioning. The purpose of this study is not to
investigate the neural underpinning of unconditioned stimuli or
conditioning process, rather we used the conditioning process to
introduce a bias and disrupt an automatic response which is at the
basis of PTSD. Backward conditioning, although controversial, has
several times been demonstrated to be effective (Hall, 1984; Spetch
et al., 1981). Ayres et al. (1987) demonstrated that one-trial backward conditioning was able to cause lick suppression and defensive
behaviours in rats. In accordance with the backward conditioning
paradigm, a CS immediately followed a US (Chang et al., 2004); consequently, in the present study the aversive stimulus (US) started
150 ms before the associated image (CS) without overlap.
In order to avoid a context effect, the subjects stayed in the
scanner during this session, but no MR images were acquired.
2.3.2. Study session
The 120 pictures presented during the conditioning session
(unpleasant and neutral, paired with US and not paired with US)
were also presented during fMRI acquisition, using the same block
design. Each block was homogeneous for valence and pairing (i.e.
blocks contained ve unpleasant paired pictures OR ve unpleasant
non-paired pictures, OR ve neutral paired pictures OR ve neutral
non-paired pictures). No acoustic stimulus was delivered in this
phase. Fig. 1 illustrates the experimental design.
During both experimental sessions, the subjects task was to
observe each picture and to quickly press one of two buttons on a
compatible MRI keypad using their right hand: a left button when
the pictures contained a vegetable item such as plants, owers,
etc. (target pictures) and a right button if the pictures contained

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

23

Fig. 1. Conditioning session (no scanner acquisition). Examples of negative (unpleasant) and neutral blocks respectively, consisting of 5 pictures each (Stimulus Onset
Asynchrony (SOA) = 3 s, stimulus duration = 2 s) alternated with a 1 s control state (cross-xation). A total of 24 blocks was presented. The subjects task was to observe each
picture and to press as quickly as possible the left button in the case of target detection and the right button in the case of no target detection. Targets were vegetable
elements (e.g. plants, owers) in the pictures. Furthermore, 25% of the pictures (both unpleasant and neutral) were conditioned by association of images with an aversive
acoustic stimulus. The aversive stimulus was presented 150 ms before the associated image. Study session (scanner acquisition). All 120 pictures presented during conditioning
(unpleasant and neutral, paired with US and not paired with US) were presented in the same block design as in the conditioning session. Each stimulation block was presented
in a randomized order alternated with cross-xation. The duration of every block and every control state was 15 s each.

no vegetable item. In each experimental session, the presence of

targets was balanced between unpleasant and neutral pictures, corresponding to 25% of the total number of pictures. The vegetable
items were always neutral in valence (i.e. there was no ring tree
or bloody ower, etc.) irrespective of the pictures main emotional
content.
This procedure was designed to reproduce typical conditions
triggering dysfunctional behaviour in patients with PTSD. These
patients suffer from sudden emotional crises while they are
engaged in daily activities requiring processing of neutral and emotional stimuli available in the environment. In order to control the
attention level, all participants were informed that they had to
perform a memory task after the end of the experiment.

2.4. fMRI recordings

The image acquisition parameters were identical to those in
our previous study (Brunetti et al., 2010). BOLD contrast functional
imaging was performed with a SIEMENS MAGNETOM VISION scanner at 1.5 T by means of T2*-weighted echo planar imaging (EPI),
free-induction decay (FID) sequences with the following parameters: TR 3 s, TE 60 ms, matrix size 64 64, FOV 256 mm, in-plane
voxel size 4 mm 4 mm, ip angle 90 , slice thickness 4 mm and no
gap. A standard head coil was used and the subjects head was xed
with foam pads to reduce involuntary movements. Functional fMRI
volumes consisted of 28 bicommissural transaxial slices including
the cortical regions of interest; 240 volumes (including 3 dummy
volumes) were acquired starting with a control period. A high resolution structural volume was acquired at the end of the session
via a 3D MPRAGE sequence with the following features: sagittal,
TR = 9.7 ms and TE = 4 ms, matrix 256 256, FoV 256 mm, in-plane
voxel size 1 mm 1 mm, ip angle 12 , slice thickness 1 mm,
no gap.

2.5. fMRI data analysis

Pre-processing and statistical analysis of the fMRI data were performed using Brain Voyager QX 2.3 software (Brain Innovation, The
Netherlands). Due to T1 saturation effects, the rst three scans of
each run were discarded from the analysis. Pre-processing of functional scans included slice scan time correction, motion correction
and removal of linear and nonlinear trends from voxel time series.
A three-dimensional motion correction was performed by means of
a rigid body transformation to match each functional volume to the
reference volume (the fourth volume) estimating three translation
and three rotation parameters.
These parameters were stored in log-les and inspected to check
that estimated head movement was not larger than approximately
half a voxel (2 mm) for the functional run and that no taskcorrelated movement had occurred (Friston et al., 1996; Hajnal
et al., 1994). Spatial normalization was performed for structural
and functional data sets. The spatial normalization of the structural volumes was performed in two steps. In the rst step the 3D
MPRAGE data set of each subject was aligned with the stereotactic
axes. For this step the location of the anterior commissure (AC), the
posterior commissure (PC) and two rotation parameters for midsagittal alignment were specied manually in the 3D data set. In
the second step the extreme points of the cerebrum were specied.
These points together with the AC and PC coordinates were then
used to scale the 3D data sets into the dimensions of the standard
brain of the Talairach and Tournoux atlas (Talairach and Tournoux,
1988) using a piecewise afne and continuous transformation. All
investigated areas were included in this space.
To transform the functional data into Talairach space, the preprocessed functional time series were rst re-sampled at a voxel
size of 3 mm 3 mm 3 mm and co-registered with the corresponding structural data set. The co-registration transformation
in BrainVoyager QX was determined by concatenating an initial

24

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

alignment matrix obtained using the Siemens position parameters

of the functional and structural images with a ne-tuning alignment matrix obtained by means of an intensity-driven alignment
algorithm. The alignment between functional and anatomical scans
was nally checked by means of a thorough visual inspection. Then
the rigid-body AC-PC transformation matrix and the piecewise
afne Talairach grid scaling transformation performed for the 3D
anatomical data set were applied to co-registered functional data.
This procedure resulted in a normalized four-dimensional data representation (volume time course) for each functional run. In order
to avoid quality loss due to successive data sampling, spatial normalization was actually performed using a single transformation
matrix obtained by combining the different spatial transformations
described.
A random effect-group analysis was performed using the general linear model (GLM) (Friston et al., 1995) with correction for
temporal autocorrelation (Bullmore et al., 1996; Woolrich et al.,
2001). Four predictors of interest (2 levels of the Pairing factor 2
levels of the Valence factor) were considered, whereas baseline corresponded to the cross-xation blocks. To account for
haemodynamic delay, the boxcar waveform representing the rest
and task conditions was convolved with an empirically founded
haemodynamic response function (Boynton et al., 1996). In this
analysis, the time series from each subject were z-normalized
prior to the statistical computation. Group SPMs were obtained
and thresholded at p < 0.05 corrected for multiple comparisons by
means of the False Discovery Rate (FDR) (Genovese et al., 2002).
Firstly, a contrast map showing a signicant between-group difference in activation (p < 0.05 corrected) was obtained from the
whole-brain group analysis, pooling the four experimental conditions (e.g. 2 levels of the Pairing factor 2 levels of the Valence
factor). Secondly, specic contrasts testing for experimental conditions differences across groups were performed.
In addition to the whole-brain voxel-wise analysis, a regional
comparison of activation across conditions was performed. Regions
of interest (ROIs) were dened using the rst contrast (betweengroup difference, pooling the four experimental conditions) in
order to avoid a bias towards a particular condition. The mean fMRI
time course in a given ROI for each subject was analyzed, and the
individual BOLD responses to the four conditions were characterized by the tted parameters of the GLM (normalized beta values).
For each ROI, the BOLD response was used as input (e.g. dependent
variable) to an ANOVA design including Group, Pairing, and Valence
as factors.
2.6. Statistical analysis
2.6.1. Behavioural data
All statistical analyses were performed by Statistica 6.1 software
(Statsoft Italia srl, 2003). An ANOVA mixed design using Group,
Pairing and Valence as factors (p < 0.05) tested behavioural accuracy and reaction times recorded in the fMRI scanner as dependent
variables. Homogeneity of variance was assessed by means of the
Brown-Forsythe test (p < 0.05). The effects of main interest were
the following:
(1) Group: PTSD patients and trauma-exposed controls [(TC)
between factor];
(2) Valence: neutral and negative valence of the pictures (within
factor);
(3) Pairing: paired with US and not paired with US pictures (within
factor).
Consequently, within factors were dened as following:
NegCS+ = negative picture paired with US; NegCS = negative

picture not paired with US; NeuCS+ = neutral picture paired with
US; NeuCS = neutral picture not paired with US.
Signicant effects were assessed by means of the LSD post hoc
test (p < 0.05). Spearman rank order coefcients to examine the
relationship among behavioural data (reaction time; accuracy),
CAPS score, and the BOLD beta values of brain areas showing statistically signicant Group effects in the mentioned ANOVAs (p < 0.05)
were computed.
2.6.2. Regional analysis of fMRI data
All statistical analyses were performed by Statistica 6.1 software (Statsoft Italia srl, 2003). The comparison of BOLD activation
in regions of interest was undertaken by means of the analysis of
variance (ANOVA) for repeated measures. The dependent variable
of the ANOVA analysis was the BOLD activation (normalized beta
value) in each ROI, while the ANOVA factors were the following:
(1) Group: PTSD patients and trauma-exposed controls [(TC)
between factor];
(2) Valence: neutral and negative valence of the pictures (within
factor);
(3) Pairing: paired with US and not paired with US pictures (within
factor).
Again, within factors were dened as following:
NegCS+ = negative picture paired with US; NegCS = negative
picture not paired with US; NeuCS+ = neutral picture paired with
US; NeuCS = neutral picture not paired with US.
An LSD post hoc test to evaluate the hypothesis of a statistically signicant effect involving the Group and Pairing factors
(p < 0.05) was performed. As a secondary analysis, Spearman rank
order coefcients examined the relationship of the BOLD activation
between the prefrontal cortex and amygdala (p < 0.05). Finally, differential BOLD activation CS+ minus CS was used as a dependent
variable within an ANOVA design with Group and Valence as factors
(p < 0.05).
3. Results
3.1. Behavioural data
Table 2 reports the main behavioural results (i.e. accuracy and
reaction time, RT, as dependent variables) of the ANOVA mixed
designs using Group (PTSD, TC), Valence (Neg, Neu) and Pairing
(CS+ , CS ) as factors. Concerning the accuracy, the Group factor
showed no statistically signicant main effect. In contrast, a main
effect of the Pairing factor pointed to lower accuracy for the CS
(86.0%) than for the CS+ (94.2%). Furthermore, a main effect of
the Valence factor showed lower accuracy for the negative (83.7%)
than for the neutral pictures (96.5%). There was also an interaction
between the Pairing and Valence factors, showing that the accuracy
was reduced by the Pairing factor (i.e. CS ) together with Valence
factor (i.e. negative valence).
Concerning the RT, the Group factor showed no statistically
signicant main effect, whereas main effects of the Pairing and
the Valence factor pointed to a slower reaction time for the CS
(794 ms) than for the CS+ (764 ms), and for the negative (840 ms)
than for the neutral pictures (718 ms). There was also an interaction between the Pairing and Valence factors, showing a slower
reaction time for the negative CS (876 ms) than for the negative
CS+ (803 ms).
In summary, there was no statistically signicant difference
between the PTSD patients and the traumatized control group in
terms of accuracy or RT, thus the subsequent between-group differences in fMRI (BOLD) activation cannot be merely explained

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

Table 2
Behavioural results and statistical analysis.
Percentage correct responses (accuracy)
NegCS

Variables

ALL
TC
PTSD

20
10
10

NeuCS

NegCS+

NeuCS+

Mean

SD

Mean

SD

Mean

SD

Mean

SD

77.1
76.9
77.3

6.0
5.7
6.5

94.9
93.9
96.0

3.8
3.3
4.2

90.3
90.0
90.7

9.0
8.3
10.0

98.0
96.7
99.3

3.7
4.4
2.1

2 group 2 valence ANOVA results

Factor

Effect
df

Error
df

Pairing
Valence
Pairing Valence

1
1
1

18
18
18

89.380
97.732
42.976

<0.0001
<0.0001
<0.0001

Mean reaction time (RT)

NegCS

Variables

ALL
TC
PTSD

20
10
10

NeuCS

NegCS+

NeuCS+

Mean

SD

Mean

SD

Mean

SD

Mean

SD

876.2
873.9
878.5

141.6
175.3
107.7

711.2
705.3
717.2

97.6
116.3
80.8

803.2
805.9
800.5

96.3
115.3
79.0

725.5
729.6
721.4

125.5
147.4
107.2

2 group 2 valence ANOVA results

Factor

Effect
df

Error
df

Pairing
Valence
Pairing Valence

1
1
1

18
18
18

12.859
68.623
9.969

0.002
<0.0001
0.005

Note: SD = standard deviation; df = degrees of freedom.

NegCS+ = negative picture paired with US; NegCS = negative picture unpaired with
US; NeuCS+ = neutral picture paired with US; NeuCS = neutral picture unpaired with
US; TC = traumatized controls.

in terms of differences in behavioural performance. Furthermore,

both groups showed the general negative effect of the emotional
valence and conditioning on behavioural performance (p < 0.05),
conrming the effective impact of these experimental manipulations in the present experiment.
3.2. fMRI results
3.2.1. Whole-brain voxel-wise analysis
Fig. 2 shows the contrast map obtained from the whole-brain
group analysis considering those voxels showing a signicant
between-group difference in activation (p < 0.05 corrected), pooling the four experimental conditions (e.g. 2 levels of the Pairing
factor 2 levels of the Valence factor). In this contrast a statistically signicant difference (p < 0.05) was observed in the following
cortical areas: bilateral dorso-lateral prefrontal cortex (dlPFC) BA
9, right dorso-lateral prefrontal cortex BA 46, right dorsomedial
prefrontal cortex (dmPFC) BA 8, medial posterior superior parietal
gyrus (precuneus-PCUN, BA 7), right lingual gyrus (LING-BA 18) and
bilateral amygdala (AMY).
Furthermore, the specic contrast NegCS+ versus NegCS , separately performed on the two groups, showed in the TC a signicant
lower activation of the bilateral AMY during NegCS+ than during
NegCS presentation (p < 0.05, Fig. 3a). No statistical results were
observed in the PTSD group. Moreover, the specic contrast Neu
versus Neg revealed a signicant larger activation of the right DLPFC
9 during neutral than during negative pictures observation in the
TC group (p < 0.05, Fig. 3b), while in PTSD the Neg versus Neu contrast revealed a signicant larger activation of the right dmPFC BA
8 (p < 0.05, Fig. 3c).

25

3.2.2. ROI analysis

The regional activation analysis performed by means of the
ANOVA design including Group, Pairing, and Valence as factors
in ROIs dened as described in the methods section yielded the
following results. All statistical values are displayed in Table 3.
Beside the expected ANOVA Group main effect, there was a
main effect of the Valence factor. On the one hand, the main
effect of the Group factor showed statistical differences for the
following regions: bilateral DLPFC BA 9, right DLPFC BA 46, right
LING BA 18, right dmPFC BA 8, PCUN BA7 and bilateral AMY.
Specically, compared to the TC, the PTSD group showed a lower
BOLD activation in bilateral DLPFC BA9, right DLPFC BA46, and
right LING BA 18. The PTSD group was characterized by a higher
BOLD activation in right dmPFC BA8, right PCUN BA7 and bilateral
AMY.
On the other hand, the main effect of the Valence factor was
found in right DLPFC BA 46, regardless of Group and Pairing factors.
A higher BOLD activation in this cortical area during the observation
of the negative pictures compared to the neutral ones has been
observed.
In addition to the above main effects, there was an ANOVA interaction between the Group and the Valence factor in right DLPFC BA
9 and in right dmPFC BA 8. Specically, the negative pictures had
a different impact in the two groups. The TC group showed that
BOLD activation in right DLPFC BA 9 was lower during the observation of the negative than the neutral pictures while the PTSD
group showed that BOLD activation in right dmPFC BA 8 was higher
during the observation of the negative compared to the neutral
pictures.
Finally, Fig. 4 shows the ANOVA interaction among the Group,
Pairing and Valence factors in bilateral AMY and in right dmPFC
BA 8. Post hoc tests indicated that the TC group showed that BOLD
activation in right (p < 0.05) and left (p < 0.01) AMY was lower during the observation of the negative paired (NegCS+ ) compared to
the negative non-paired pictures (NegCS ). This effect was absent
in the PTSD group. Activation in right dmPFC BA 8 was characterized by a similar BOLD pattern, however post hoc comparisons
(NegCS+ versus NegCS ) did not reach statistical signicance in the
trauma-exposed controls.
To take into account the possible confounding effect of other
psychiatric conditions on fMRI results, we repeated the ANOVA
after removing the three subjects with a comorbid DSM-IV diagnosis (a disthymic PTSD patient, an agoraphobic PTSD patient and
a traumatized control affected by social phobia). All the above mentioned main effects and interactions remained signicant, with
the exception of the Group Pairing Valence interaction in the
right dorso-medial prefrontal cortex BA 8 which did not reach
statistical signicance anymore. Moreover, in order to control for
trauma load, we entered the level of trauma exposure as a covariate
in the ANOVA of the above-described ROIs and all the mentioned
effects remained signicant.

3.3. Control analyses

The main data analysis showed that in the TC group (but not in
the PTSD group), BOLD activation in left and right AMY was lower
during the observation of the NegCS+ compared to the NegCS . An
interesting issue is whether this effect could be observed by a direct
comparison between the two groups. To test this control hypothesis, we used differential BOLD activation CS+ > CS as a dependent
variable for a control ANOVA design with Group and Valence as factors. Results showed an interaction between the Group and Valence
factors in left AMY. Post hoc tests indicated a higher BOLD activation CS+ > CS of left AMY in the PTSD than in the TC during
the observation of the negative pictures (p < 0.05; Fig. 5), lending

26

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

Fig. 2. Contrast maps obtained from the whole-brain group analysis considering those voxels showing a signicant between-group difference in activation (TC > PTSD, p < 0.05
corrected), pooling the four experimental conditions (e.g. 2 levels of the Pairing factor 2 levels of the Valence factor). Areas with larger activation for the traumatized control
group (TC) are depicted in red. Activation was observed in: bilateral dorso-lateral prefrontal cortex (DLPFC) BA 9 and right dorso-lateral prefrontal cortex (DLPFC) BA 46
[y = 34]; right dorsomedial prefrontal cortex (dmPFC) BA 8 and medial posterior superior parietal gyrus/precuneus (PCUN) BA 7 [z = 50]; right lingual gyrus (LING BA 18)
[z = 7]; bilateral amygdala (AMY) [y = 8]. R = right hemisphere.

further support to the evidence that, compared to TC, the PTSD

patients are characterized by exaggerated AMY responses during
the observation of negative pictures associated to a single aversive
experience (e.g. aversive stimulus).
Another interesting issue is whether the above effects of the
experimental conditions on the prefrontal cortex and amygdala
were statistically interrelated, suggesting a correlated activity
during the processing of the negative pictures associated to conditioning stimuli (e.g. aversive stimulus). To test this control
hypothesis, Spearman coefcient was computed on BOLD activation related to the negative pictures in the PTSD and TC considered
as a whole group (Table 4). Results showed a signicant negative correlation between left DLPFC BA 9 and left AMY during the
observation of the NegCS . Furthermore, there was a signicant
negative correlation between right DLPFC BA 46 and right AMY
during the observation of the NegCS+ . These results indicate an

intra-hemispherical correlated activity between DLPFC and AMY

in the processing of negative pictures.
The main data analysis conrmed the key role of human AMY in
the processing of negative pictures and aversive stimuli (NegCS+ ).
Therefore, another interesting issue is whether the effects of the
present experimental conditions on AMY were related to clinical
features of PTSD. To test this control hypothesis, we investigated
the statistical correlation between BOLD activation in right or left
AMY and CAPS score. To this aim, Spearman coefcient was computed between the BOLD activation of right or left AMY resulting
from the contrast NegCS+ versus baseline and the total CAPS score
in all PTSD and TC considered as a whole group. Results showed
a signicant positive correlation between the BOLD activation of
left AMY during negative conditioned pictures and the total CAPS
score (Rs = 0.46, p < 0.05), thus conrming the relevance of the AMY
function in PTSD.

Fig. 3. Specic contrast maps: (a) NegCS+ > NegCS contrast, separately performed on the two groups, showed in the traumatized control group a signicant larger activation
of the bilateral amygdala [y = 8] during NegCS than during NegCS+ presentation (p < 0.05). No signicant statistical differences were observed in the PTSD group. (b)
Neg > Neu pictures contrast revealed a signicant larger activation during negative than during neutral pictures observation in the controls right dorso-lateral prefrontal
cortex BA 9 [y = 34] (p < 0.05). (c) Neg > Neu pictures contrast revealed a signicant larger activation of the right dorsomedial prefrontal cortex [z = 50] during negative than
during neutral pictures observation in the PTSD group (p < 0.05).

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

Table 3
FMRI results. 2 group 2 Pairing factor 2 valence random effect general linear
model (RFX-GLM). Brain regions showing signicant effects. The statistical threshold was set at p 0.05, corrected for multiple comparisons, using False Discovery
Rate (FDR). The baseline was the cross xation condition. Signicant effects were
controlled for trauma load (entering the number of traumatic events as a covariate) and for psychiatric comorbidity (removing the subjects with other psychiatric
diagnoses).
Main effects
Group effect
ROI name

Brodmann
area

Right dorso-lateral
prefrontal cortex
Left dorso-lateral
prefrontal cortex
Right dorso-lateral
prefrontal cortex
Right lyngual gyrus
Right dorso-medial
prefrontal cortex
Medial posterior superior
parietal gyrus/precuneus
Right amygdala
Left amygdala

df

Effect

32.10

1.18

<0.001

PTSD < TC

19.96

1.18

<0.001

PTSD < TC

46

7.88

1.18

<0.05

PTSD < TC

18
8

9.50
5.69

1.18
1.18

<0.01
<0.05

PTSD < TC
PTSD > TC

5.46

1.18

<0.05

PTSD > TC

5.24
5.95

1.18
1.18

<0.05
<0.05

PTSD > TC
PTSD > TC

Valence effect
ROI name

Brodmann
area

df

Effect

Right dorso-lateral
prefrontal cortex

46

7.10

1,18

<0.05

Negative > neutral

Interaction effects
Group Valence effect
ROI name

Brodmann
area

df

Post hoc signicant

comparisons

Right dorso-lateral
prefrontal cortex
Right dorso-medial
prefrontal cortex

4.63

1,18

<0.05

7.97

1,18

<0.05

In TC:
negative < neutral
In PTSD:
negative > neutral

Group Pairing Valence effect

ROI name

Brodmann
area

df

Right dorso-medial
prefrontal cortex
Right amygdala

4.50

1.18

<0.05a

5.22

1.18

<0.05

6.22

1.18

<0.05

Left amygdala

Post hoc signicant

comparisons

In TC:
NegCS+ < NegCS
In TC:
NegCS+ < NegCS

df = degree of freedom; PTSD = PTSD patients group; TC = traumatized controls

group; NegCS = negative not paired stimuli; NegCS+ = negative paired stimuli.
a
Not signicant after removing the 3 subjects with psychiatric comorbidity:
F(115) = 3.47, p = 0.08.

Table 4
Spearman rank order coefcients between BOLD beta values of different activated
areas in the whole sample (PTSD and control subjects) [p < .05].
R DLPFC/Ba46

L DLPFC/Ba9

PCUN

R amygdala

NegCS : 0.55
NegCS : 0.59
NeuCS+ : 0.72
NeuCS : 0.56
R amygdala NegCS+ : 0.53
L Amygdala

NegCS+ : 0.46

NegCS : 0.46

dmPFC

NegCS+ : 0.62
NegCS : 0.54
NegCS+ : 0.53
NegCS : 0.47

R LING

Note: NegCS+ = negative picture paired with US; NegCS = negative picture unpaired
with US; NeuCS+ = neutral picture paired with US; NeuCS = neutral picture unpaired
with US.

27

4. Discussion
Are the prefrontal cortex and amygdala affected by a single
exposure to traumatic (aversive) stimuli in PTSD subjects? Here this
issue was addressed by an exploratory study using a new variant of
the Pavlovian conditioning procedure combined with the recording
of fMRI in PTSD and in trauma-exposed participants. In a blockdesign procedure, some pictures with negative or neutral contents
were associated, by means of a single exposure, to an aversive
acoustic stimulus as a model of the traumatic episode, while other
pictures with negative or neutral contents were presented with
no conditioning as a control condition. Main results lend support
to the working hypothesis. In the traumatized non-PTSD group,
the activation in the right dlPFC was lower during the observation of the negative compared to the neutral pictures. Furthermore,
in the same control group, the activation in the bilateral amygdala was lower during the observation of the NegCS+ compared
to the NegCS . In PTSD group we observed the following results:
amygdala responses were not reduced by the observation of the
NegCS+ compared to the NegCS . Since no statistically signicant
difference between the PTSD patients and the traumatized controls
groups in terms of accuracy or RT were observed, the described
between-group differences in dlPFC and amygdala activation cannot be merely explained in terms of differences in behavioural
performance. Rather, the present results might be explained by a
neurophysiological mechanism operating in the traumatized nonPTSD controls for the moderation of the amygdala responses during
the processing of visual pictures with negative and aversive content. Specically, higher amygdala activity towards NegCS than
NegCS+ has been observed in healthy humans (Merz et al., 2013) as
well as in rodents (Herry et al., 2008) during early extinction phase.
In our traumatized control group, this response could indicate a
learned extinction by means of a down-regulation of the amygdala
in response to the conditioned emotional stimulus. This mechanism
may fail in the PTSD patients. Even if the small sample size does not
allow a denitive conclusion, one can speculate that the difference
in amygdala responses between PTSD and traumatized participants is the basis of the clear-cut division in individual responses
to trauma. Furthermore, when contrasting the two groups global
maps, in our PTSD group a lower activation of the bilateral dlPFC
and a higher activation of the right dmPFC and bilateral amygdala
were observed. In addition, in the left hemisphere the activation
of the dlPFC and amygdala pointed to a negative (inverse) correlation during the observation of the negative stimuli associated to
the aversive stimuli. Finally, in the same condition there was a positive correlation between the global activation of the left amygdala
and the total CAPS score, thus conrming the relevance and specicity of the amygdala function in PTSD symptoms. Although these
nding as a whole suggest a clear segregation of functional mechanisms between PTSD and non-PTSD traumatized individuals, they
are probably not sufcient to denitively understand the peculiarity of the syndrome development. Indeed, lack of results in the
PTSD group could also suggest that the conditioning response in
this group is not strong enough to be observed in fMRI. Nevertheless, to our knowledge, the results of the present exploratory
study provide the rst experimental evidence that a single stressful episode (a visual negative stimulus only once associated to a
single aversive stimulus) induces differences in the activation of
the prefrontal cortex and amygdala in PTSD patients compared to
traumatized non-PTSD participants.
This evidence complements the ndings of the following three
main research lines. In the rst research line, it has been reported
that the prefrontal cortex and amygdala are part of brain circuits
of extreme importance for the representation of the homeostatic
value of the visual stimuli (Arana et al., 2003). In this theoretical
framework, dorso-lateral and medial prefrontal areas are involved

28

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

Fig. 4. Results of BOLD activation in the corresponding ROIs as dened in the main text. A signicant Group Pairing Valence interaction was observed within both the
right and left amygdala [right, F(118) = 5.22, p < 0.05; left, F(118) = 6.22, p < 0.05] and right dorsomedial prefrontal cortex BA 8 [F(118) = 4.5, p < 0.05]. Bar plots show LSD
post hoc test results: in the TC BOLD activation in the right (p < 0.05) and left (p < 0.01) amygdala (top) was lower during the observation of the negative paired with US
(NegCS+ ) compared to the negative unpaired with US pictures (NegCS ). This effect was not found in the PTSD group. In right dmPFC (bottom) a similar BOLD pattern was
observed but post hoc comparisons did not reach statistical signicance (p = 0.20) [*p < 0.05; **p < 0.01].

in the elaboration of learned classication of stimuli through

threat appraisal and emotion regulation processes (for a review
see Britton et al., 2011). Furthermore, the medial prefrontal cortex represents the biological relevance of a given stimulus and may

convey this information to the dlPFC (Mechias et al., 2010). Finally,

the amygdala may play a pivotal role in the learning processes
related to emotional reactions, fear, and fear extinction (Phelps
et al., 2004; Milad, 2009; Britton et al., 2011). It sends signals to the

Fig. 5. Signicant Group Valence interaction within the left amygdala measured on differential value (CS+ > CS ). LSD post hoc test results: the traumatized control group
TC left amygdala showed a signicant BOLD response decrease with respect to PTSD left amygdala activation (F(118) = 6.23 p < 05) in response to negative pictures. No
between-group differences were observed during neutral stimulation [*p < 0.05]. R = right hemisphere.

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

hypothalamus for the activation of the sympathetic nervous system, to the reticular nucleus for increasing reexes, to the nuclei of
the trigeminal nerve and facial nerve for facial expressions of fear,
and to the ventral tegmental area, locus coeruleus, and laterodorsal
tegmental nucleus for the activation of dopamine, norepinephrine
and epinephrine release, respectively (Talarovicova et al., 2007).
Consistently, our traumatized non-PTSD group showed an amygdala response that resembled a learned mechanism of fear
extinction by means of amygdala down-regulation.
In the second research line, it has been reported that brain
circuits encompassing the prefrontal cortex and amygdala are disrupted in PTSD patients. In a PET study, a positive functional
connectivity was found in PTSD patients between amygdala and
anterior cingulate cortex in response to stimulations related to
emotionally traumatic events (Osuch et al., 2008). Furthermore, a
positive functional connectivity between the amygdala and dmPFC
was observed during the observation of fearful faces under anxious conditions in healthy subjects (Robinson et al., 2012) and
during resting state in PTSD patients (Sripada et al., 2012). The
functional relationship between amygdala, anterior cingulate and
dmPFC might be either adaptive (e.g. increasing vigilance towards
threatening stimuli) or maladaptive if generalized (Grillon and
Charney, 2011; Robinson et al., 2011). Even if the present study
was based on BOLD response rather than connectivity analysis,
our results in the PTSD group seem to be in agreement with this
research line, since the amygdala and dmPFC showed a similar
behaviour in this group. More investigations on the functional connectivity among the dlPFC, dmPFC, and amygdala could support
this line.
In the third research line, the neural basis of conditioning was
investigated in PTSD patients. An fMRI study used a 2-day fear
conditioning and extinction protocol in PTSD patients and in traumatized non-PTSD controls (Milad et al., 2009). Results indicated
that compared to the control group, the PTSD group showed a
greater activation of the amygdala during day 1 of the experiment
(extinction learning) and of the anterior cingulate cortex related
to the extinction recall on day 2 (extinguished conditioned stimuli
presented in the absence of shock). It was concluded that impaired
fear extinction in PTSD could represent the outcome of dysfunctional activation in brain structures that mediate fear extinction
(Milad et al., 2009). Our results partially agree with this study. In
fact, Milad and colleagues in their traumatized non-PTSD controls
also observed a bilateral vmPFC activation during extinction recall.
This regions activation was not observed in our sample, probably
since our paradigm did not include a delayed extinction learning
(i.e. extinction recall, see also Milad et al., 2007). Nevertheless,
our results in the traumatized non-PTSD group are coherent with
extinction learning mechanism as described in previous studies
(Phelps et al., 2004).
Interestingly, our data also showed a decreased right lingual
gyrus activity in PTSD patients compared to traumatized controls
in all conditions. Previous studies, comparing PTSD patients and
controls, observed decreased functional connectivity as well as
reduced amplitude of low-frequency uctuation in the right lingual
gyrus (Yin et al., 2011; Qin et al., 2012). This visual association cortex, strongly connected to the amygdala, receives projections from
the latter that affect attention to visual information (Taylor and
Fragopanagos, 2005). Our results indicate that in PTSD patients this
cortico-limbic communication is altered, indicating the involvement of sensorial cortices in the emotional processing.
As a novelty, the results of the present study extend the previous ndings by demonstrating that a single stressing episode affects
brain structures, including amygdala, that mediate fear extinction
in PTSD patients (Milad et al., 2009). Furthermore, they parallel
recent structural neuroimaging ndings showing that a single prolonged exposure to trauma was associated with a smaller grey

29

matter volume of left dorsal anterior cingulate cortex in PTSD

patients than in control subjects (Chen et al., 2012). In addition,
a single prolonged exposure to trauma was associated to cortical
thinning in frontal besides in parietal and parahippocampal regions
(Liu et al., 2012).
In conclusion, the results of the present exploratory study suggest that single-episode conditioning affects the activation of the
amygdala and prefrontal cortex during negative and neutral visual
stimuli processing. It can be speculated that a single traumatic
event type might reduce a physiological inhibitory control of the
dlPFC on dmPFC and amygdala in the processing of visual stimuli
associated to those seen in the single aversive episode. A limit of
the present study is the lack of psychophysiological measures such
as skin conductance and heart rate, that could support the conclusions from imaging data and at the same time could add important
vegetative information. Another limit could be represented by the
small sample size, thus probably resulting in a low statistical power.
Nevertheless, our ndings motivate a conclusive validation fMRI
study simultaneous to psychophysiological recordings in a larger
sample of PTSD and control subjects aimed at quantifying the functional connectivity among the dlPFC, dmPFC, and amygdala during
the model of the aversive event and subsequent exposure to visual
stimuli associated to such an event.
Conict of interest
The authors declare that there are no conicts of interest.
Acknowledgments
The authors would like to thank local Bank Labour Union for
collaboration in this research. The authors would also like to thank
Dott. Gianni Perrucci for technical support.
References
Arana, F.S., Parkinson, J.A., Hinton, E., Holland, A.J., Owen, A.M., Roberts, A.C., 2003.
Dissociable contributions of the human amygdala and orbitofrontal cortex to
incentive motivation and goal selection. J. Neurosci. 23 (29), 96329638.
Ayres, J.J.B., Haddad, C., Albert, M., 1987. One-trial excitatory backward conditioning
as assessed by conditioned suppression of licking in rats: concurrent observations of lick suppression and defensive behaviors. Anim. Learn. Behav. 15 (2),
212217.
Blake, D., Weathers, F., Nagy, L., Kaloupek, D., Klauminzer, G., Charney, D., Keane, T.,
1990. A clinician rating scale for assessing current and lifetime PTSD: the CAPS-I.
Behav. Ther. 18, 187L 188.
Boynton, G.M., Engel, S.A., Glover, G.H., Heeger, D.J., 1996. Linear systems analysis of functional magnetic resonance imaging in human V1. J. Neurosci. 16,
42074241.
Bremner, J.D., Vermetten, E., Schmahl, C., Vaccarino, V., Vythilingam, M., Afzal, N.,
Grillon, C., Charney, D.S., 2005. Positron emission tomographic imaging of neural
correlates of a fear acquisition and extinction paradigm in women with childhood sexual-abuse-related post-traumatic stress disorder. Psychol. Med. 35 (6),
791806.
Britton, J.C., Lissek, S., Grillon, C., Norcross, M.A., Pine, D.S., 2011. Development of
anxiety: the role of threat appraisal and fear learning. Depress. Anxiety 28 (1),
517.
Brunetti, M., Perrucci, M.G., Di Naccio, M.R., Ferretti, A., Del Gratta, C., Casadio, C.,
Romani, G.L., 2014. Framing deductive reasoning with emotional content: an
fMRI study. Brain Cogn. 87 (June), 153160.
Brunetti, M., Sepede, G., Mingoia, G., Catani, C., Ferretti, A., Merla, A., Del Gratta, C.,
Romani, G.L., Babiloni, C., 2010. Elevated response of human amygdala to neutral
stimuli in mild post traumatic stress disorder: neural correlates of generalized
emotional response. Neuroscience 168 (3), 670679.
Bryant, R.A., Felmingham, K.L., Kemp, A.H., Barton, M., Peduto, A.S., Rennie, C.,
Gordon, E., Williams, L.M., 2005. Neural networks of information processing in
posttraumatic stress disorder: a functional magnetic resonance imaging study.
Biol. Psychiatry 58 (2), 111118.
Bullmore, E., Brammer, M., Williams, S., Rabe-Hesketh, S., Janot, N., David, A., Mellers,
J., Howard, R., Sham, P., 1996. Statistical methods of estimation and inference
for functional MR image analysis. Magn. Reson. Med. 35 (2), 261277.
Catani, C., Adenauer, H., Keil, J., Aichinger, H., Neuner, F., 2009. Pattern of cortical
activation during processing of aversive stimuli in traumatized survivors of war
and torture. Eur. Arch. Psychiatry Clin. Neurosci. 259 (6), 340351.
Chance, P., 2013. Learning and Behavior. Cengage Learning.

30

M. Brunetti et al. / Brain Research Bulletin 114 (2015) 2030

Chang, R.C., Stout, S., Miller, R.R., 2004. Comparing excitatory backward and forward
conditioning. Q. J. Exp. Psychol. B: Comp. Physiol. Psychol. 57 (1), 123.
Chen, Y., Fu, K., Feng, C., Tang, L., Zhang, J., Huan, Y., Cui, J., Mu, Y., Qi, S., Xiong, L., Ma,
C., Wang, H., Tan, Q., Yin, H., 2012. Different regional gray matter loss in recent
onset PTSD and non PTSD after a single prolonged trauma exposure. PLoS ONE
7 (11), e48298.
Conti, L., Rossi, A., Donda, P., 1999. In: Conti, L. (Ed.), Repertorio delle scale di Valutazione in Psichiatria SEE Firenze, vol. 174., pp. 153.
Di Giacinto, A., Brunetti, M., Sepede, G., Ferretti, A., Merla, A., 2014. Thermal signature
of fear conditioning in mild post traumatic stress disorder. Neuroscience 266,
216223.
Friston, K.J., Holmes, A.P., Worsley, K.J., Poline, J.P., Frith, C.D., Frackowiak, R.S.J., 1995.
Statistical parametric maps in functional imaging: a general linear approach.
Hum. Brain Mapp. 2, 173181.
Friston, K.J., Williams, S., Howard, R., Frackowiak, R.S.J., Turner, R., 1996. Movement
related effects in fMRI time series. Magn. Reson. Med. 35, 346355.
Genovese, C.R., Lazar, N.A., Nichols, T., 2002. Thresholding of statistical maps in functional neuroimaging using the false discovery rate. NeuroImage 15 (4), 870878.
Grillon, C., Charney, D.R., 2011. In the face of fear: anxiety sensitizes defensive
responses to fearful faces. Psychophysiology 48, 17451752.
Hajnal, J.V., Myers, R., Oatridge, A., Schwieso, J.E., Young, I.R., Bydder, G.M., 1994.
Artifacts due to stimulus correlated motion in functional imaging of the brain.
Magn. Reson. Med. 31, 283291.
Hall, J.F., 1984. Backward conditioning in Pavlovian type studies: reevaluation and
present status. Pavlov. J. Biol. Sci. 19, 163169.
Herry, C., Ciocchi, S., Senn, V., Demmou, L., Mueller, C., Luethi, A., 2008. Switching
on and off fear by distinct neuronal circuits. Nature 454, 600-U28.
Lang, P.J., Bradley, M.M., Cuthbert, B., 1999. International Affective Picture System
(IAPS): Technical Manual and Affective Ratings. The Center for Research in Psychophysiology, University of Florida, Gainesville, FL.
Lanius, R.A., Bluhm, R., Lanius, U., Pain, C., 2006. A review of neuroimaging studies
in PTSD: heterogeneity of response to symptom provocation. J. Psychiatr. Res.
40 (8), 709729.
Liu, Y., Li, Y.J., Luo, E.P., Lu, H.B., Yin, H., 2012. Cortical thinning in patients with recent
onset post-traumatic stress disorder after a single prolonged trauma exposure.
PLoS ONE 7 (6), e39025.
Mechias, M.L., Etkin, A., Kalisch, R., 2010. A meta-analysis of instructed fear studies:
implications for conscious appraisal of threat. Neuroimage 49 (2), 17601768.
Merz, C.J., Hermann, A., Stark, R., Wolf, O.T., 2013. Cortisol modies extinction learning of recently acquired fear in men. Soc. Cogn. Affect. Neurosci., nst137.
Milad, M.R., Wright, C.I., Orr, S.P., Pitman, R.K., Quirk, G.J., Rauch, S.L., 2007. Recall
of fear extinction in humans activates the ventromedial prefrontal cortex and
hippocampus in concert. Biol. Psychiatry 62, 446454.
Milad, M.R., Orr, S.P., Lasko, N.B., Chang, Y., Rauch, S.L., Pitman, R.K., 2008. Presence
and acquired origin of reduced recall for fear extinction in PTSD: results of a
twin study. J. Psychiatr. Res. 42 (7), 515520.
Milad, M.R., Pitman, R.K., Ellis, C.B., Gold, A.L., Shin, L.M., Lasko, N.B., Zeidan, M.A.,
Handwerger, K., Orr, S.P., Rauch, S.L., 2009. Neurobiological basis of failure to
recall extinction memory in posttraumatic stress disorder. Biol. Psychiatry 66
(12), 10751082, http://dx.doi.org/10.1016/j.biopsych.2009.06.026.
Oldeld, R.C., 1971. The assessment and analysis of handedness: the Edinburgh
Inventory. Neuropsychologia 9, 97113.
Osuch, E.A., Willis, M.W., Bluhm, R., CSTS Neuroimaging Study Group, Ursano, R.J.,
Drevets, W.C., 2008. Neurophysiological responses to traumatic reminders in the
acute aftermath of serious motor vehicle collisions using [15O]H2 O positron
emission tomography. Biol. Psychiatry 64, 327335.

Phelps, E.A., Delgado, M.R., Nearing, K.I., LeDoux, J.E., 2004. Extinction learning in
humans: role of the amygdala and vmPFC. Neuron 43, 897905.
Pieraccini, F., Bossini, L., Lombardelli, A., DellErba, A., Pappagallo, E., Mantovani,
A., Castrogiovanni, P., 1999. Conti L Repertorio delle scale di Valutazione in
Psichiatria SEE Firenze, vol. 1272., pp. 1253.
Pissiota, A., Frans, O., Fernandez, M., von Knorring, L., Fischer, H., Fredrikson,
M., 2002. Neurofunctional correlates of posttraumatic stress disorder: a PET
symptom provocation study. Eur. Arch. Psychiatry Clin. Neurosci. 252 (2),
6875.
Qin, L.D., Wang, Z., Sun, Y.W., Wan, J.Q., Su, S.S., Zhou, Y., Xu, J.R., 2012. A preliminary
study of alterations in default network connectivity in post-traumatic stress
disorder patients following recent trauma. Brain Res. 1484, 5056.
Rauch, S.L., Whalen, P.J., Shin, L.M., McInerney, S.C., Macklin, M.L., Lasko, N.B.,
Orr, S.P., Pitman, R.K., 2000. Exaggerated amygdala responses to masked facial
stimuli in posttraumatic stress disorder: a functional MRI study. Biol. Psychiatry
47, 769776.
Robinson, O., Letkiewicz, A., Overstreet, C., Ernst, M., Grillon, C., 2011. The effect of
induced anxiety on cognition: threat of shock enhances aversive processing in
healthy individuals. Cogn. Affect. Behav. Neurosci. 11 (2), 217227.
Robinson, O.J., Charney, D.R., Overstreet, C., Vytal, K., Grillon, C., 2012. The adaptive
threat bias in anxiety: amygdala-dorsomedial prefrontal cortex coupling and
aversive amplication. Neuroimage 60 (1), 523529.
Rougemont-Bcking, A., Linnman, C., Zefro, T.A., Zeidan, M.A., Lebron-Milad, K.,
Rodriguez-Romaguera, J., Rauch, S.L., Pitman, R.K., Milad, M.R., 2011. Altered
processing of contextual information during fear extinction in PTSD: an
fMRI study. CNS Neurosci Ther. 17 (4), 227236, http://dx.doi.org/10.1111/
j.1755-5949.2010.00152.x.
Sheehan, D.V., Lecrubier, Y., Janavs, J., Knapp, E., Weiller, E., Bonora, L.I., Amorim, P.,
Lpine, J.-P., Sheehan, M.F., Baker, R.R., Sheehan, K.H., 1994. Mini International
Neuropsychiatric Interview (MINI). University of South Florida Institute for
Research in Psychiatry/INSERM-Hpital de la Salptrire, Tampa, Florida/Paris,
France.
Spetch, M.L., Wilkie, D.M., Pinel, J.P., 1981. Backward conditioning: a reevaluation of
the empirical evidence. Psychol. Bull. 89 (1), 163175.
Sripada, R.K., King, A.P., Garnkel, S.N., Wang, X., Sripada, C.S., Welsh, R.C., Liberzon,
I., 2012. Altered resting-state amygdala functional connectivity in men with
posttraumatic stress disorder. J. Psychiatry Neurosci. 37. (4).
Talairach, J., Tournoux, P., 1988. Co-planar Stereotaxic Atlas of the Human Brain.
Thieme, New York.
Talarovicova, A., Krskova, L., Kiss, A., 2007. Some assessments of the amygdala role
in suprahypothalamic neuroendocrine regulation: a minireview. Endocr. Regul.
41 (4), 155162.
Taylor, J.G., Fragopanagos, N.F., 2005. The interaction of attention and emotion.
Neural Netw. 18 (4), 353369.
van der Kolk, B.A., 1994. The body keeps the score: memory and the evolving psychobiology of posttraumatic stress. Harv. Rev. Psychiatry 1 (5), 253265.
Williams, L.M., Kemp, A.H., Felmingham, K., Barton, M., Olivieri, G., Peduto, A.,
Gordon, E., Bryant, R.A., 2006. Trauma modulates amygdala and medial
prefrontal responses to consciously attended fear. Neuroimage 29 (2),
347357.
Woolrich, M.W., Ripley, B.D., Brady, M., Smith, S.M., 2001. Temporal autocorrelation
in univariate linear modeling of FMRI data. NeuroImage 14, 13701386.
Yin, Y., Li, L., Jin, C., Hu, X., Duan, L., Eyler, L.T., Gong, Q., Song, M., Jiang, T., Liao,
M., Zhang, Y., Li, W., 2011. Abnormal baseline brain activity in posttraumatic
stress disorder: a resting-state functional magnetic resonance imaging study.
Neurosci. Lett. 498 (3), 185189.