Antecedent-descriptions-change-brain-reactivity-to-emotional-stimuli-A-functional-magnetic-resonance-imaging-study-of-an-extrinsic-and-incidental-reappraisal-strategy.pdf

Neuroscience 193 (2011) 241–248

ANTECEDENT DESCRIPTIONS CHANGE BRAIN REACTIVITY TO

EMOTIONAL STIMULI: A FUNCTIONAL MAGNETIC RESONANCE

IMAGING STUDY OF AN EXTRINSIC AND INCIDENTAL REAPPRAISAL

STRATEGY

I. MOCAIBER, a T. A. SANCHEZ, b M. G. PEREIRA, c

F. S. ERTHAL, d M. JOFFILY, e,f D. B. ARAUJO, g

E. VOLCHAN d AND L. DE OLIVEIRA c *

a Polo Universitário de Rio das Ostras, Universidade Federal Flumi-

nense, Rio das Ostras, Brazil

b Faculdade de Medicina, Universidade Federal do Rio de Janeiro, Rio

de Janeiro, Brazil

c Instituto Biomédico, Universidade Federal Fluminense, Niterói, Rio de

Janeiro, Brazil

d Instituto de Biofísica Carlos Chagas Filho, Universidade Federal do

Rio de Janeiro, Rio de Janeiro, Brazil

e Center for Mind/Brain Sciences, University of Trento, Trento, Italy

Centre National de la Recherche Scientiﬁque, France

g Brain Institute, Universidade Federal do Rio Grande do Norte, Natal,

Rio Grande do Norte, Brazil

process ( Gross and Thompson, 2007 ). This regulatory

process can inﬂuence which emotions individuals have,

leading to changes to one or more aspects of emotion (e.g.

Bargh and Williams, 2007; Gross and Thompson, 2007 ). In

fact, emotion regulation can be achieved through a number

of processes, but in general, strategies can be classiﬁed as

either response or antecedent-focused ( Gross, 1998;

Ochsner and Gross, 2005 ). Response-focused strategies

occur relatively late in the emotion-generation process and

would include suppressing the behavioral expression of

emotion. This strategy, however, seems to be ineffective in

decreasing the experience of negative emotions ( Gross

and Levenson, 1993 ). On the other hand, antecedent-

focused regulation strategies occur relatively early within

the emotion-generation process and appear to be effective

in decreasing the experience of negative emotions

( Ochsner et al., 2002; Moser et al., 2006 ). One approach to

achieve the antecedent-focused regulation is reinterpret-

ing an unpleasant stimulus to be less negative, which is

called “reappraisal.” Thus, to be considered as “anteced-

ent” this strategy should be employed before the emotion

had been triggered ( Gross, 1998 ).

One unresolved issue in this area is whether emotion

regulation refers to intrinsic processes, to extrinsic pro-

cesses, or to both ( Gross and Thompson, 2007 ). The ﬁrst

process assumes that emotion regulation derives from a

person’s self-regulatory effort and the second process as-

sumes that it results from the regulatory inﬂuences of other

people ( Thompson, 2011 ). In general, researches involv-

ing adult subjects typically focus on intrinsic processes

( Thompson, 2011 ). However, even for adults, extrinsic

emotion regulation occurs in many ways. This is generally

how people intervene to manage the feelings of a spouse,

friend, or acquaintance, by dissuading them from going to

events that may be stressful or by giving them other inter-

pretation about a negative ongoing situation (for review

see Gross and Thompson, 2007 ). As pointed out by

Thompson (2011) “. . . regulatory inﬂuences arise not just

through the strategic efforts of the person to function com-

petently but also through a variety of extrinsic, noncon-

scious, implicit processes by which emotion is managed in

response to complex contextual demands.” This type of

regulatory inﬂuence can also be referred to as incidental

emotion regulation , which occurs when contextual factors

alter an affective response without one’s intentional effort

( Berkman and Lieberman, 2009 ).

Abstract—In the present study we investigated whether indi-

viduals would take advantage of an extrinsic and incidental

reappraisal strategy by giving them precedent descriptions

to attenuate the emotional impact of unpleasant pictures. In

fact, precedent descriptions have successfully promoted

down-regulation of electrocortical activity and physiological

responses to unpleasant pictures. However, the neuronal

substrate underlying this effect remains unclear. Particularly,

we investigated whether amygdala and insula responses,

brain regions consistently implicated in emotional process-

ing, would be modulated by this strategy. To achieve this,

highly unpleasant pictures were shown in two contexts in

which a prior description presented them as taken from

movie scenes (ﬁctitious) or real scenes. Results showed that

the ﬁctitious condition was characterized by down-regulation

of amygdala and insula responses. Thus, the present study

provides new evidence on reappraisal strategies to down-

regulate emotional reactions and suggest that amygdala and

insula responses to emotional stimuli are adaptive and highly

reserved.

Key words: amygdala, insula, ventrolateral prefrontal cortex

(VLPFC), extrinsic reappraisal, incidental reappraisal, mutila-

tion pictures.

Emotion regulatory processes may be automatic or con-

trolled, conscious or unconscious and may have their ef-

fects at one or more points in the emotion generative

*Corresponding author. Tel: 55-21-9182-0177.

E-mail address: ldol@vm.uff.br (L. de Oliveira).

Abbreviations: fMRI, functional magnetic resonance imaging; IAPS,

International Affective Picture System; LPP, late positive potential;

ROI, regions of interest; VLPFC, ventrolateral prefrontal cortex.

doi:10.1016/j.neuroscience.2011.07.003

241

242

I. Mocaiber et al. / Neuroscience 193 (2011) 241–248

Regarding intrinsic process, most previous studies

have investigated reappraisal strategies during which indi-

viduals are encouraged to voluntarily change how they

think about a situation, in order to decrease its emotional

impact ( Ochsner et al., 2004; Phan et al., 2005; Moser et

al., 2006 ). Several studies have shown that this strategy

can change emotional experience and its associated phys-

iological and functional magnetic resonance imaging

(fMRI) responses ( Gross and Muñoz, 1995; Gross, 1998,

2002; Ochsner et al., 2002; Ray et al., 2005; Goldin et al.,

2008 ). Specially, amygdala activity modulation has been

described as crucial to the emotion regulation process. A

number of studies have demonstrated that the use of

reappraisal, that is, cognitively changing the meaning of

emotional stimuli, affects evoked responses in the

amygdala and other brain areas ( Ochsner et al., 2002,

2004 ). For instance, fMRI studies have shown that

amygdala is sensitive to emotion regulation in that its

activity is decreased during the reevaluation of negative

affect while is increased during enhancement of negative

emotion ( Phan et al., 2005; Eippert et al., 2007 ). It is

important to note, however, that the majority of research to

date has focused on deliberate emotion regulation, which

is voluntary, volitional and driven by explicit goals ( Berk-

man and Lieberman, 2009 ). In these studies, participants

have typically been asked to generate their own alternative

interpretations after viewing an emotional stimulus. Fur-

thermore, this strategy may introduce inter- and intra-par-

ticipant variability regarding the speciﬁc way that reap-

praisals are performed by the participants.

In the present study we extend research on emotion

regulation data by employing a subtle and less deliberate

strategy. Speciﬁcally, we investigated whether individuals

would take advantage of an extrinsic and incidental emo-

tion regulation strategy by giving them precedent descrip-

tions to attenuate the emotional impact of unpleasant im-

ages. In this case, regulation starts before the emotion is

completely triggered. First of all, participants were told that

they would see very unpleasant images such as mutilated

bodies. After that, they received descriptions, which gave

them the opportunity to adjust their responses according to

each emotional context. In one context, unpleasant images

were described as taken from real situations (real) and in

the other, from movie scenes (ﬁctitious).

In fact, this kind of subtle reappraisal strategy has

successfully promoted down-regulation of electrocortical

activity and physiological responses during the viewing of

emotional pictures ( Foti and Hajcak, 2008; Oliveira et al.,

2009; Mocaiber et al., 2009, 2010 ). However, the neuronal

substrate underlying this effect remains an open question.

In the present study, we employed fMRI to investigate

whether amygdala and insula responses, brain regions

consistently implicated in emotional processing (e.g. Mor-

rison et al., 2007; Pereira et al., 2010; Pessoa and Adol-

phs, 2010 ), would be modulated by this strategy.

Thus, we hypothesized that the differential brain re-

sponses between unpleasant and neutral pictures would

be attenuated (or even eliminated) in the ﬁctitious condition

compared to the real condition. Particularly, we predicted

that the activity of amygdala and insula to highly arousing

unpleasant images would be ﬂexible and adaptive in the

context of an extrinsic and incidental reappraisal strategy.

EXPERIMENTAL PROCEDURES

Participants

Twenty-two healthy volunteers (12 male, mean age of 26.61

years, SD 4.72) participated in this study. Volunteers were se-

lected among students, and had normal or corrected-to-normal

vision. They reported no psychiatric or neurologic problems and

were not under medication with nervous system action. Partici-

pants were naive as to the purpose of the experiment. Before data

collection, all procedures were approved by the local ethics com-

mittee and participants gave informed consent.

Stimuli

Two classes of images (72 neutral and 72 unpleasant) were

employed. Neutral pictures consisted of photographs of people “in

normal life” and unpleasant images consisted of photographs of

mutilated bodies. Pictures were taken from the International Af-

fective Picture System (IAPS)—( Center for the Study of Emotion

and Attention and (CSEA-NIMH), 1999 ), from the World Wide

Web, or photographed by the authors. Additional pictures were

selected apart from the IAPS, since the IAPS database was not

large enough for the conduction of this study. We attempted to

match unpleasant and neutral stimuli in terms of both color content

and complexity (e.g. number of faces, number of body parts, etc.).

Following the protocol developed by the CSEA ( Bradley and Lang,

1994 ), all images were assessed on a 1–9 scale in terms of

valence (from negative to positive) and arousal (from low to high)

by a separate group of participants ( n

0.001)

and arousal ( M 6.6 and 3.4, respectively, t 34.43, P 0.001)

ratings. Each picture was repeated just once within the ﬁctitious

and the real contexts. Thus, the repetition of one speciﬁc picture

always occurred within the same context.

2.08 and 5.21, respectively, t

58.02, P

Procedure

Participants performed an emotional judgment task , in which they

had to decide whether the presented picture was neutral or un-

pleasant, pressing one of two buttons with the right hand. The

judgment task was performed in two contexts: ﬁctitious and real.

Initially, participants were told that they would see very unpleasant

images such as mutilated bodies. After that, they received de-

scriptions, which gave them the opportunity to adjust their re-

sponses according to each emotional context. In one context, they

were instructed that the unpleasant pictures had been taken from

movie productions (ﬁctitious condition) whereas in the other con-

text, they were instructed that the pictures corresponded to real

scenes (real condition). In the ﬁctitious condition, the precedent

descriptions represented a safety signal, which encouraged par-

ticipants to down-regulate their responses to an upcoming highly

provocative set of pictures. Neutral pictures were also presented

so that we could assess the differential responses to mutilation

stimuli in both contexts.

Therefore, the experiment consisted of a 2 (context: ﬁctitious

and real) 2 (valence: neutral and unpleasant) factorial design.

Experimental session comprised four runs (two real contexts and

two ﬁctitious contexts runs) counterbalanced across participants.

20) with ages similar to the

participants of the current study (22.3, SD 1.8). Thus, all images

employed in the present experiment (pictures from the IAPS cat-

alogue and from the web) had their valence and arousal assessed

according to the IAPS rating procedure. Unpleasant and neutral

images differed signiﬁcantly from each other in IAPS normative

valence ( M

I. Mocaiber et al. / Neuroscience 193 (2011) 241–248

243

Each run comprised four interleaved blocks of task and rest peri-

ods. Rest periods consisted of a ﬁxation cross presented for 15 s.

Task blocks were composed of nine trials (four unpleasant and

ﬁve neutral pictures in two blocks, and ﬁve unpleasant and four

neutral in the two others). Every run contained the same number

of neutral and emotional trials that were randomly presented

within each task block. Task blocks were preceded by a 3-s black

screen to warn subjects of its beginning. Every single trial was

initiated by a 500-ms ﬁxation cross, followed by a central picture

presented for 200 ms and masked by a gray-scale checkerboard

until the volunteer emitted a response. The trial duration was ﬁxed

to 3 s, corresponding exactly to one repetition time (TR). Partici-

pants were instructed to respond as quickly and as accurately as

possible, indicating whether the pictures were neutral or unpleas-

ant. Pictures were presented with Presentation software (Neu-

robehavioral Systems, http://nbs.neuro-bs.com ) running in a Win-

dows notebook and projected inside the scanner through a pro-

jection screen and mirror system ﬁxated on the head coil.

Responses were collected with an MRI-compatible button system.

Participants performed a practice session before the experiment

to ensure that they completely understood the instructions.

Participants were ﬁrst told that they would see a variety of

pictures including a very unpleasant category, mutilated bodies.

After that, and before the beginning of each run, participants had

to read a text informing the source of the forthcoming pictures. In

ﬁctitious context runs, the pictures were announced to be ﬁctitious

scenes, whereas, in real context runs, pictures were described as

real-life scenes. This experimental manipulation was done in a

very subtle way, so that participants could not suspect of the key

manipulation.

For the ﬁctitious context, participants read the following text:

“ ... The pictures that will be shown to you in the next trials were

obtained from movies with the aim of convincing the audi-

ence ... Therefore, the pictures were produced by means of di-

verse techniques such as make-up, and do not correspond to real

situations” . For the real context, participants read the following

text: “ ... The IAPS is a set of standardized colored photographs of

a wide range of situations ... All the pictures are real and were

obtained from the web, media, or taken by the group that devel-

oped the IAPS.” The aim of this manipulation was to attenuate the

emotional impact of the pictures during the ﬁctitious context com-

pared to the real context. This experimental procedure was pre-

viously validated by recent studies ( Oliveira et al., 2009; Mocaiber

et al., 2010 ).

In summary, each subject participated in one experimental

session that comprised four counterbalanced runs (two ﬁctitious

and two real runs). Each run contained four blocks comprising

nine unpleasant and neutral trials each. Unpleasant and neutral

pictures were equal in number and randomly presented within

each run. The pictures that were presented to some participants in

the real context were presented in the ﬁctitious context to others,

guaranteeing that any attenuation effect would not be associated

to a speciﬁc pool of pictures.

In order to conﬁrm the credibility of our key manipulation, we

ran a behavioral experiment with an additional sample outside the

scanner ( n 27; mean age of 19 years, SD 2.09) and asked the

participants to rate a set of pictures presented in the ﬁctitious and

real contexts. This behavioral experiment was very similar to the

fMRI experiment described before. First, participants performed

an emotional judgment task , where they had to decide whether the

picture presented was neutral or unpleasant, pressing one of two

buttons with the right hand. At the end of each block, participants

additionally rated the set of pictures just shown by means of the

Self Assessment Manikin (SAM) valence and arousal scales

( Bradley and Lang, 1994 ). This rating test was not applied in the

scanner sample to prevent participants from suspecting the key

experimental manipulation.

fMRI recording. MRI data were collected using a 1.5 T MRI

scanner (Magnetom Vision Plus, Siemens, Erlangen, Germany).

Functional images were acquired using a gradient-echo planar

imaging sequence (TR 3s;TE 60 ms; FOV 240; ﬂip an-

gle 90°; 64 64 matrix). Whole brain coverage was obtained with

25 axial slices (thickness 4 mm; in-plane resolution 3.75 3.75

mm). Echo-planar images were co-registered to a high-resolution

structural T1-weighted image obtained during the same session

(TR/TE 9.7/4.0 ms; ﬂip angle 12°; 160 slices; thickness 1 mm;

256 256 matrix; FOV 256 mm). Head movements were re-

strained with foam padding. Stimulus presentation was synchro-

nized with functional images acquisition.

fMRI data analysis

The statistical parametric mapping software package (SPM5,

Wellcome Department of Cognitive Neurology, London) was used

for preprocessing and statistical analyses. The ﬁrst three func-

tional volumes of each run were removed to eliminate non-equi-

librium magnetization effects. The remaining images were cor-

rected for head movement by realigning all the images to the ﬁrst

image via rigid body transformations. The images were then cor-

rected for differences in slice acquisition time. For each partici-

pant, functional and structural images were co-registered. Struc-

tural data were normalized by matching them to the standardized

MNI template, and the transformation parameters estimated in this

step were applied to all functional images. Functional images were

spatially smoothed with an 8-mm full width at half maximum

Gaussian kernel prior to statistical analysis.

Data analysis was performed according to the general linear

model framework, as implemented in SPM5 ( Friston et al., 1995 ).

Data obtained from the 22 participants were analyzed. Functional

data quality was checked by means of Artifact Repair (version 4)

toolbox ( Mazaica et al., 2005 ), a package of tools developed for

SPM to check and correct motion and other artifacts. This allowed

signal quality checking before reaching General Linear Model

(GLM).

The main goal of the present study was to determine the

effects of a less deliberate reappraisal of emotional stimuli on

amygdala and insula activation. However, we also performed a

whole-brain voxel-wise analysis to investigate general task-re-

lated activations, as well as to further investigate emotional brain

responses. A standard two-stage mixed-effects analysis was per-

formed.

The ﬁrst (ﬁxed) level involved determining the regression

coefﬁcients of the variables of interest, which modeled the effects

of each experimental condition: judging neutral pictures (real con-

text), judging unpleasant pictures (real context), judging neutral

pictures (ﬁctitious context) and judging unpleasant pictures (ﬁcti-

tious context). Before estimation via multiple regressions, regres-

sors of interest were convolved with a canonical hemodynamic

response function.

Second-level group analyses were conducted by means of

Student t -tests. Since random effects analyses may be fairly con-

servative in the context of fMRI data ( Worsley et al., 2002 ), we

employed a threshold of P 0.001 (uncorrected), as is commonly

employed in the literature. A clusters size threshold was estab-

lished at 10 contiguous voxels.

Critical analyses were performed within the amygdala and

insula, which have been pointed out by the literature as a crucial

part of the emotional network. Functional regions of interest (ROI)

of these structures were selected based on the contrast unpleas-

ant pictures vs. neutral pictures during the real context (a thresh-

old of P

0.001, uncorrected, was employed). These same ROIs

were used for a paired t -tests comparison between the critical

conditions (unpleasant pictures vs. neutral pictures) during the

ﬁctitious context. Functional ROIs construction and signal extrac-

tion were performed using the MarsBar toolbox (version 0.41)

244

I. Mocaiber et al. / Neuroscience 193 (2011) 241–248

Table 1. Brain regions activated for the unpleasant neutral contrast

Brain region

Hemisphere

Coordinates MNI (x y z)

Cluster size

t (score z)

Real context (unpleasant neutral)

Cerebellum

32 76 34

4.89 (3.95)

Cerebellum

34 52 34

4.26 (3.57)

Mid cingulate

14 40 50

5.09 (4.06)

Insula

34 22 18

4.06 (3.45)

Amygdala

34 2 24

4.54 (3.74)

Mid frontal gyrus

22 14 42

4.35 (3.63)

Suplementar motor area

6 22 58

3.84 (3.31)

Superior frontal gyrus

4 38 46

4.13 (3.49)

Fictitious context (unpleasant neutral)

Lingual gyrus

8 46 2

4.41 (3.67)

Caudate nucleus

20 12 16

6.44 (4.73)

Caudate nucleus

22 26 4

5.49 (4.28)

Inferior frontal gyrus (ventrolateral prefrontal cortex)

40 32 0

4.70 (3.84)

Mid frontal gyrus (ventrolateral prefrontal cortex)

30 40 4

3.88 (3.33)

Superior frontal gyrus (ventromedial prefrontal cortex)

24 50 6

4.26 (3.58)

All regions at P threshold of .001, uncorrected. L left, R right;xyz MNI coordinates of the maximally active voxel; t maximum t value.

( Brett et al., 2002 ) for SPM5. Mean values for each condition

were extracted.

( Table 1 ). Brain regions showing greater activation during

the judgment of unpleasant pictures in comparison to neu-

tral pictures in the real context are shown in Table 1 . They

include bilateral cerebellum, mid-cingulate cortex, right

amygdala, left insula, left mid-frontal gyrus, right supple-

mentary motor area (SMA) and right superior frontal gyrus.

As expected, the real context contrast revealed an “emo-

tional network” previously described by other studies.

Brain regions showing greater activation during the

judgment of unpleasant pictures in comparison to neutral

pictures in the ﬁctitious context included a group of pre-

frontal regions (inferior frontal gyrus, mid-frontal gyrus and

superior frontal gyrus), lingual gyrus and caudate nucleus

( Fig. 1 ). It is important to note that the “emotional network”

revealed in the real context contrast was not found in the

ﬁctitious context contrast, suggesting that the activation of

the emotional network was attenuated.

Our main focus of interest was to investigate the ef-

fects of emotion regulation on critical brain regions previ-

ously described in the literature as part of an emotional

network (amygdala and insula). Therefore, we conducted a

ROI analysis of these structures, deﬁned via the contrast

unpleasant pictures neutral pictures during the real con-

text. The mean parameters for unpleasant and neutral

pictures in the ﬁctitious context were extracted and com-

pared by paired t -test. This analysis showed no statistical

difference of the amygdala (x y z 34 2 24) and insula

(xyz 34 22 18) activity as a function of stimulus

valence ( P 0.5), suggesting an attenuation of emotional

effects during the ﬁctitious context ( Fig. 2 ).

RESULTS

Ratings

Data on mean ratings showed that images presented in the

real context were more arousing than the images pre-

sented in the ﬁctitious context ( M 5.55, SD 2.01 and

4.68, SD 2.48, respectively, P 0.01). Also, pictures in

the real context were marginally more unpleasant than in

the ﬁctitious context ( M 3.60, SD 1.47 and 3.93,

SD 1.24, P 0.07).

Behavioral results

Median reaction times and accuracy (number of correct

responses emitted by the participant in each experimental

condition) were determined for 19 of the 22 participants;

because of equipment malfunction, behavioral data from

three participants were not analyzed. Paired t -tests re-

vealed that the mean reaction time did not differ between

the unpleasant and neutral pictures for both the real

( M 934.5 ms, SE 28.59 vs. 935.4 ms, SE 25.35) and

the ﬁctitious context ( M 928.6 ms, SE 25.45 vs. 924.6

ms, SE 26.25). However, the overall performance to

judge unpleasant pictures ( M 73.3%, SD 1.59) was sig-

niﬁcantly higher in comparison to neutral ones ( M

68.71%, SD 1.68) in the real context ( P 0.04). This dif-

ference was not observed for the ﬁctitious context

( M 71.78%, SD 2.16 and M 71.2%, SD 1.59, for un-

pleasant and neutral pictures, respectively) ( P 0.8).

Brain imaging results

DISCUSSION

Functional MRI data were ﬁrst analyzed on a voxel-by-

voxel basis across the whole brain for each experimental

context. We will ﬁrst present the whole brain data for the

following contrasts: unpleasant neutral pictures (real con-

text) and unpleasant neutral pictures (ﬁctitious context)

In the present study we examined the effects of an extrin-

sic, incidental and less deliberate reappraisal strategy on

unpleasant image processing. Speciﬁcally, we observed

that precedent descriptions, aimed to attenuate the impact

of emotional pictures, produced changes on behavioral

I. Mocaiber et al. / Neuroscience 193 (2011) 241–248

245

Fig. 1. Prefrontal clusters activation peaks in the Fictitious context

contrast (unpleasant neutral). (A) Top panel shows right ventrolat-

eral prefrontal cortex (VLPFC) activation (xyz 40 32 0); (B) Middle

panel shows left VLPFC activation (xyz 30 40 4) and (C) Bottom

panel shows ventromedial prefrontal cortex (VMPFC) activation

(xyz 24 50 6). Activations are overlaid on subject’s average ana-

tomical image at a level of P .001, uncorrected. For interpretation of

the references to color in this ﬁgure legend, the reader is referred to

the Web version of this article.

tional stimuli involving fear and negative emotions

( Ochsner et al., 2002; Phan et al., 2002; Hariri et al., 2003;

Phelps and LeDoux, 2005; Eippert et al., 2007; Del-Ben et

al., in press ). Moreover, the visualization of the mutilation

pictures evoked increased responses in insula and cingu-

late cortex, which are regions associated with the monitor-

ing of the ongoing internal emotional state of the organism

( Craig, 2009 ) and the implementation of defensive re-

sponses ( Azevedo et al., 2005; Morrison et al., 2007; Milad

et al., 2007; Pereira et al., 2010 ), respectively.

Interestingly, during the ﬁctitious context, we did not

ﬁnd greater activation for brain regions associated with

emotional processing, such as amygdala, cingulate cortex,

insula and SMA. This ﬁnding is remarkable since we used

a very subtle strategy to down-regulate highly aversive and

provocative pictures. Actually, we chose mutilation pic-

tures because we intended to test this strategy using stim-

uli that evoke strong aversive reaction ( Pereira et al., 2004,

2006, 2010; Azevedo et al., 2005; Erthal et al., 2005;

Facchinetti et al., 2006; Oliveira et al., 2009 ). Thus, subtle

changes in the situational context (precedent descriptions)

resulted in activation of structures involved in affect

(amygdala and insula) in one condition (real) coupled with

increases in PFC activity in the other condition (ﬁctitious).

Previous studies have shown that instructions to pur-

posefully modulate emotional responding to unpleasant stim-

uli affect physiology and brain activity ( Ochsner et al., 2004;

Phan et al., 2005; Goldin et al., 2008 ). However, these stud-

ies used voluntary strategies, which are intrinsic, volitional

and driven by overt goals. In the majority of these studies,

participants have typically been asked to generate their own

alternative interpretations after viewing an emotional stimu-

lus. Here, participants were persuaded to down-regulate the

impact of aversive pictures via antecedent written instructions

that indicated that the stimuli were ﬁctitious. This key exper-

imental manipulation changed the nature of the pictures pre-

sented, in the sense that their relevance appeared to be

attenuated, resulting in diminished brain activation in struc-

tures critical to emotional processing. A potential strength of

the present study is that all participants were provided texts

that equally inﬂuenced the meaning of the upcoming pictures,

rather than being left to generate their own interpretation. In

addition, we believe that with this procedure, task difﬁculty

was more similar between participants. This procedure can

also be referred to as an incidental emotion regulation in

which changes in contextual factors alter an affective re-

sponse ( Berkman and Lieberman, 2009 ).

Another possible interpretation of the present results is

that the precedent descriptions changed the expectations

about upcoming stimuli, altering the initial processing of

emotional stimuli. In fact, adjustment of expectations about

upcoming stimuli may represent a powerful antecedent-

focused strategy for the cognitive control of emotion ( Foti

and Hajcak, 2008 ). In the present study, participants were

ﬁrst told that they would see very unpleasant images such

as mutilated bodies. Thus, in the beginning of the experi-

ment, participants were aware that they would judge

mutilation pictures. After that, they read precedent descrip-

tions, which gave them the opportunity to adjust their ex-

and brain responses, reducing amygdala and insula reac-

tivity during the viewing of highly unpleasant stimuli.

As expected, during the real condition the reactivity of

an emotional network including amygdala, SMA, cingulate

cortex and insula was modulated by stimulus valence

( Lane et al., 1999; Phan et al., 2002; Pereira et al., 2010 ).

Increased responses in the amygdala are consistent with a

considerable body of data that has documented how this

structure is involved in affective processing ( Phan et al.,

2005; Pessoa, 2008 ). A large body of studies has shown

that amygdala is implicated in responses to visual emo-

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: