Content uploaded by Michael M Plichta
Author content
All content in this area was uploaded by Michael M Plichta on Dec 25, 2018
Content may be subject to copyright.
Enhancement of Activity of the Primary Visual
Cortex During Processing of Emotional Stimuli as
Measured With Event-Related Functional Near-
Infrared Spectroscopy and Event-Related Potentials
Martin J. Herrmann,
1,2,3
*Theresa Huter,
1,3
Michael M. Plichta,
1
Ann-Christine Ehlis,
1
Georg W. Alpers,
2
Andreas Mu
¨hlberger,
2
and Andreas J. Fallgatter
1
1
Department of Psychiatry and Psychotherapy, University of Wu
¨rzburg, Wu
¨rzburg, Germany
2
Department of Psychology, University of Wu
¨rzburg, Wu
¨rzburg, Germany
3
Department of Genomic Imaging, University of Wu
¨rzburg, Wu
¨rzburg, Germany
Abstract: In this study we investigated whether event-related near-infrared spectroscopy (NIRS) is suita-
ble to measure changes in brain activation of the occipital cortex modulated by the emotional content of
the visual stimuli. As we found in a previous pilot study that only positive but not negative stimuli differ
from neutral stimuli (with respect to oxygenated haemoglobin), we now measured the event-related EEG
potentials and NIRS simultaneously during the same session. Thereby, we could evaluate whether the
subjects (n¼16) processed the positive as well as the negative emotional stimuli in a similar way. During
the task, the subjects passively viewed positive, negative, and neutral emotional pictures (40 presentations
were shown in each category, and pictures were taken from the International Affective Picture System,
IAPS). The stimuli were presented for 3 s in a randomized order (with a mean of 3 s interstimulus inter-
val). During the task, we measured the event-related EEG potentials over the electrode positions O1, Oz,
O2, and Pz and the changes of oxygenated and deoxygenated haemoglobin by multichannel NIRS over
the occipital cortex. The EEG results clearly show an increased early posterior negativity over the occipital
cortex for both positive as well as negative stimuli compared to neutral. The results for the NIRS measure-
ment were less clear. Although positive as well as negative stimuli lead to significantly higher decrease in
deoxygenated haemoglobin than neutral stimuli, this was not found for the oxygenated haemoglobin.
Hum Brain Mapp 29:28–35, 2008. V
V
C2007 Wiley-Liss, Inc.
Key words: emotion; NIRS; optical topography; EPN
INTRODUCTION
The functional neuroanatomy of emotion has been
widely examined e.g. using the blood oxygen level-de-
pendent functional magnetic resonance imaging [Davis
and Whalen, 2001; Pessoa et al., 2002; Phan et al., 2002].
This research approach revealed that the processing of
emotional in comparison to neutral visual stimuli leads to
an increased activation of various cortical areas, including
the amygdala, the medial prefrontal cortex, and, in-
terestingly, also sensory areas such as the visual cortex
Contract grant sponsor: Deutsche Forschungsgemeinschaft; Con-
tract grant numbers: KFO 125/1-1.
*Correspondence to: Martin Herrmann, Genomic Imaging and
Department of Psychiatry and Psychotherapy, Fuechsleinstr. 15,
97080 Wuerzburg, Germany.
E-mail: Herrmann_M@klinik.uni-wuerzburg.de
Received for publication 3 April 2006; Revision 4 September 2006;
Accepted 21 November 2006
DOI: 10.1002/hbm.20368
Published online 21 February 2007 in Wiley InterScience (www.
interscience.wiley.com).
V
V
C2007 Wiley-Liss, Inc.
rHuman Brain Mapping 29:28–35 (2008) r
[Jungho
¨fer et al., 2001; Phan et al., 2002]. Moreover, it was
shown that emotional arousal modulates the amplitudes of
the event-related potentials (ERPs) at two time windows.
As early as 240 ms after the stimulus presentation, an early
posterior negativity (EPN) for emotional compared to neu-
tral visual stimuli was localized over the occipital cortex
[Schupp et al., 2003]. It was argued that this EPN reflects
facilitated sensory encoding of affective stimuli by natu-
rally occurring selective attention. Additionally, a second
effect starts approximately 300 ms after the stimulus pre-
sentation over the parietal cortex, with more positive
amplitudes to emotional in comparison to neutral stimuli
(slow wave, SW) [Cuthbert et al., 2000; Schupp et al.,
2000]. This effect is considered to index postsensory
(higher-order) stages of stimulus evaluation. Altogether
there is evidence that emotional stimuli, positive as well as
negative, lead to an increased activation of the occipital
cortex.
This activation can be theoretically explained by the evo-
lutionary need to quickly detect potentially meaningful
stimuli [Ohman and Mineka, 2001]. The amygdala that is
central to emotional processing has been speculated to
prime and modulate primary visual circuits [Davis and
Whalen, 2001; LeDoux, 1998]. Indeed, feedback loops from
brain regions such as the amygdala [Amaral et al., 1992] or
the anterior cingulate [Posner and Raichle, 1995] to visual
areas have been documented. We have previously shown
that this can boost the dominance of emotional pictures
over neutral pictures [Alpers and Pauli, 2006; Alpers et al.,
2005b].
A possible explanation may be natural selective attention
[Lang, 1997]. Individuals are more likely to attend to stim-
uli of evolutionary importance than to others and situa-
tions with an emotional content outrank neutral situations.
Natural selective attention led to a stronger activation of
the occipital cortex while viewing emotional pictures than
neutral pictures [Lang et al., 1998]. Furthermore, the emo-
tional content of pictures seems to enhance not only brain
activity but also the recognition of emotional pictures
[Abrisqueta-Gomez et al., 2002; Dolcos and Cabeza, 2002].
Recently near-infrared spectroscopy (NIRS), a method
using light of the infrared spectrum, has been introduced
to investigate changes of cerebral oxygenation. NIRS is
based on the facts that (1) light of the near infrared spec-
trum can penetrate biological tissue and (2) the changes in
the absorbed near infrared light over a task can be
ascribed to changes in oxygenated and deoxygenated hae-
moglobin concentrations (for a more detailed description
of the methods see [Hoshi, 2003; Obrig and Villringer,
2003]. The correspondence between neural activity and
changes in oxygenation can be described as follows: Re-
gional brain activation leads to an increasing metabolism,
which is followed by an increased regional cerebral blood
flow (rCBF). It has been well documented that the
increased rCBF exceeds the oxygenation demand of the tis-
sue, which leads to increased [O
2
Hb] and to decreased
[HHb] as a sign of activation [Fox and Raichle, 1986]. The
changes in [O
2
HB] are typically more prominent than
those in [HHb], with higher changes in amplitudes as well
as larger involved brain areas. It was argued that changes
in [O
2
Hb] are under a stronger influence of arterial com-
partments in contrast to [HHb], which comes mostly from
the venous compartments [Franceschini et al., 2003]. There-
fore, [O
2
Hb] might be more under the influence of sys-
temic contributions like heart rate and thus represent a
less localized activation than [HHb] maps. In recent years,
NIRS has been used to measure changes in brain oxygen-
ation due to cognitive tasks [Herrmann et al., 2005; Horo-
vitz and Gore, 2004; Schroeter et al., 2002; Strangman
et al., 2002] but to a lesser extent during emotional tasks
[Herrmann et al., 2003].
The aim of this study was to evaluate whether NIRS is
suitable to measure changes in brain activation of the occi-
pital cortex modulated by the emotional content of the vis-
ual stimuli. In a first pilot study [Alpers et al., 2005a] we
used a block design with positive, negative, and neutral
pictures from the International Affective Picture System
(IAPS) [Lang et al., 1999], which were passively viewed by
the individuals. In this study, we found increased brain
activation over the occipital cortex to the visual stimuli but
the emotional modulation was less clear. The effects were
rather small and significantly only for higher [O
2
Hb] for
the positive but not for the negative stimuli in comparison
to neutral stimuli. In the current study we used the same
IAPS pictures but presented them in a randomized order
to exclude any effects caused by expectancies of the sub-
jects. To include a second parameter for emotional process-
ing, we additionally measured the event-related EEG
potentials to the stimuli parallel to the NIRS measurement.
The signals of NIRS and EEG do not influence each other
and therefore a combination is possible [Ehlis et al., sub-
mitted; Horovitz and Gore, 2004] Here we measured both
the EEG and NIRS over the same brain areas (the occipital
cortex) with the EEG electrodes always placed in between
a light emitter and a light detector.
METHODS
Participants
Sixteen healthy volunteers (11 women), age ranging
from 21 to 30 years (mean age 24.2 62.4 years), partici-
pated after written informed consent was obtained. All
participants had corrected-to-normal vision and all except
one were right handed. None of them took psychoactive
medication.
Stimulus Material
One hundred and twenty pictures were selected on the
basis of their normative valence and arousal from the
IAPS, a collection of standardized photographic material
[Lang et al., 1999]. Forty of these are positive and in high
arousal, 40 negative and in high arousal, and 40 neutral
rEmotional Modulation of Primary Visual Cortex Activity r
r29 r
and in low arousal. Each picture was shown for 3 s on a
black screen located 100 cm in front of the subject in a
shaded room. Additionally, there were 40 null events
showing a black screen for 3 s. A variable interstimulus
interval (ISI) of 2–4 s appeared between the pictures. The
task of the participants was to simply view the sequen-
tially presented images; no response was required. The
sequence of the pictures and the null events was random-
ized. ERP and NIRS data were recorded for a total of 160
trials lasting 30 min.
EEG Recording
The EEG was recorded from four scalp electrodes posi-
tioned according to the international 10–20 system at the
positions O1, O2, Oz, and Pz. An electrode between Fpz
and Fz was connected to the ground, and a further elec-
trode at the root of the nose was used as the recording ref-
erence. Three additional channels recorded the electroocu-
logram from the outer canthi of both eyes and below the
right eye to monitor eye blinks. The EEG was sampled
continuously at a rate of 1000 Hz with a bandpass from
0.1 to 70 Hz. Impedances were kept at 5 kOor below.
Epochs (200 ms before stimulus onset to 800 ms) with
amplitudes or with a voltage step exceeding 6100 mV
were excluded from further analyses. The artifact-free tri-
als (at least 20 epochs) were averaged separately for each
subject and condition. Data were filtered offline with a
bandpass from 1 to 15 Hz. A baseline correction with the
baseline between 200 ms and stimulus onset was calcu-
lated.
For the EPN we calculated the mean amplitudes in the
time window between 250 and 350 ms for all three condi-
tions. The SW was determined by peak detection between
420 and 600 ms for every single subject and condition. The
EEG-data of one participant could not be analyzed due to
data loss. EEG data were analyzed using ANOVAs for
repeated measures with the within-subjects factors elec-
trode (O1/O2/Oz/Pz) and condition (positive/negative/
neutral) using SPSS 13.0.
NIRS Recording
NIRS-measurement was performed with a continuous
wave system (ETG-4000, Hitachi Medical, Japan) with a 3
5 optode probe set (consisting of 7 photodetectors and 8
light emitters, resulting in a total of 22 channels). The probe
set was placed on the occipital cortex with channel 15
placed above electrode O1, channel 16 placed above elec-
trode Oz, and channel 17 placed above electrode O2 (Fig. 3).
Two different wavelengths (695 620 nm and 830 620 nm)
are used by the system, and its frequency is modulated for
wavelengths and channels to prevent crosstalk. Reflected
light (not absorbed) leaving the tissue is received by the
photodetectors and transmitted into a set of lock-in ampli-
fiers that are limited to the particular frequencies of interest.
Both wavelengths are used to solve the modified Beer-
Lambert equation for highly scattering tissue that allows
estimating changes in [HHb] and [O
2
Hb] based on the
measurements. Since continuous wave systems cannot mea-
sure the optical path length [Hoshi, 2003] the scale unit is
the molar concentration multiplied by the unknown path
length (mmol mm). The interoptode distance was 30 mm,
which results in measuring approximately 15 mm [Okada
and Delpy, 2003] to 25 mm [Hoshi et al., 2006] beneath the
scalp. Although the exact extent of the brain region exam-
ined by each set of NIRS probes cannot be determined, the
region examined is thought to be a banana-shaped region
between the two optodes, with a depth of 0.9–1.3 cm from
the brain surface [Koizumi et al., 1999; Villringer et al.,
1997]. Sampling rate was set to 10 Hz.
NIRS Data Analysis
The data were analyzed as described before [Plichta
et al., 2006]. To remove baseline drifts and pulsation due
to heartbeat, the raw data were preprocessed by a high-
pass filter of 0.02 Hz. The preprocessed data were then an-
alyzed by the two-stage ordinary least squares (OLS) esti-
mation methodology according to the general linear
model. A Gauss function was used for the haemodynamic
response function (HRF). At the single subject level, we
included the first and second temporal derivative of HRF
in order to modulate the onset as well as the dispersion of
the HRF. A dfunction indicating the onset of sensory stim-
ulation was convolved with the predictors for each condi-
tion. Thereafter, the first-stage OLS estimation was per-
formed. We corrected our analyses for serial, correlated
errors by fitting a first-order autoregressive process to the
error term by the Cochrane-Orcutt procedure [Cochrane
and Orcutt, 1949]. Finally, the bweights were re-estimated
(second stage) and tested for statistical significance by one-
sided ttests (single subject level). To improve the power
in the event-related design with short ISI, we included
null events, which serve as a contrast condition. Therefore,
we calculated the differences between the bweights of our
active conditions (positive, negative, and neutral) and the
contrast condition (null event). These differences were
now considered as an index of activation and were tested
against zero, or they were tested for differences between
conditions (one-sided ttests). For both single subject and
group level, significant cortical activation is indicated by
positive tvalues for [O
2
Hb] and by negative t-values for
[HHb]. For one-sided t-tests, tvalues above 1.65 indicate
an aof 5%, and tvalues above 2.33 indicate aof 1%. To
account for multiple testing, all statistical inferences are
based on an adjusted alpha level of 5%. Tvalues over 2.84
indicate Bonferroni’s adjusted aof 5% for 22 single tests
(one test for each channel).
Joined EEG-NIRS Analysis
To additionally examine the associations between the
EEG and NIRS in a more direct way, we chose the
rHerrmann et al. r
r30 r
approach described by Eichele et al. [2005]. Within this
approach we tried to predict the haemodynamic responses
by the paradigm-induced amplitude modulation of the
simultaneously acquired single trial ERPs. A first stimulus
function was defined encoding the stimulus onset, irre-
spective of the condition of the stimulus. A second stimu-
lus function encoded the onsets of the null events. Two
additional stimulus functions encoded the amplitude of
the single trial EPNs, one stimulus function for the posi-
tive and one for the negative condition. To get the single
trial EPNs we calculated in a first step the mean EEG
amplitudes for every single trial in the time window
between 250 and 350 ms after stimulus presentation for all
three conditions over the electrode position Oz. As the
investigated EPN is calculated as the difference between
emotional and neutral condition, we subtracted the mean
ERP amplitudes for the neutral condition over all trials
from the single trial amplitudes for the emotional condi-
tion. The stimulus functions were decorrelated (Schmidt-
Gram orthogonalization) from each other, ensuring that
activation related to the amplitude modulated stimulus
functions was specific to the electrophysiological measure
and not to some general feature in the evoked response to
the pictures. Using these stimulus functions four regres-
sors were formed by convolving the stimulus function
with the HRF (a gauss function). All further analyze steps
were the same as described above.
RESULTS
ERP Results
For the EPN (Fig. 1) we found the main effects ‘‘Electrode’’
(F[1.6;21.8] ¼33.7, P<0.001), and ‘‘Condition’’ (F[1.6;22.4] ¼
5.2, P<0.05), as well as an interaction effect ‘‘Electrode
Condition’’ (F[2.5;34.3] ¼3.8, P<0.05). Differences between
conditions were found for all electrode positions (O1:
F[1.5;21.5] ¼5.1, P<0.05; O2: F[1.7;24.1] ¼4.6, P<0.05; Oz:
Figure 1.
Displayed are the grand-mean event-related potentials over all subjects for the three conditions
(positive, negative and neutral (thin line) stimuli) over the three electrode positions O1, O2, and Pz.
rEmotional Modulation of Primary Visual Cortex Activity r
r31 r
F[1.6;22.8] ¼5.3, P<0.05; Pz: F[1.6;22.9] ¼5.7, P<0.05). For
all of these positions, we found significantly higher ampli-
tudes in the neutral than the positive and negative conditions
(all tvalues >2.1), without any significant differences
between positive and negative condition. We calculated the
EPN as the difference between neutral and positive and neu-
tral and negative condition. The EPN amplitude did not dif-
ferforthefourelectrodepositions(allP>0.1, t<1.68). The
EPN in the positive condition also did not differ from the
EPN in the negative condition (all P>0.2, t<1.34).
For the SW amplitudes we found a significant main
effect ‘‘Electrode’’ (F[1.9,26.42] ¼14.8, P¼0.001) and a sig-
nificant interaction ‘‘Electrode Condition’’ (F[1.9,26.0] ¼
7.5, P<0.01), but no significant effect ‘‘Condition’’
(F[1.9,27.1] ¼2.1, P¼0.15). Post hoc analyses revealed
that conditions did not differ with respect to amplitudes
for the electrode position (O1, O2, Oz, all t<1.2) but only
for Pz (F[2,28] ¼9.4, P<0.001). For Pz we found signifi-
cantly higher amplitudes to negative (t[14] ¼4.1, P<
0.001) and positive pictures (t[14] ¼3.5, P<0.01) com-
pared to the neutral pictures.
NIRS Results
As expected, negative, positive, and neutral stimuli lead
to a significant (t>2.84) increase in [O
2
Hb] and corre-
sponding decrease in [HHb] within the occipital cortex
(Figs. 2, 3 and 4). The increase of [O
2
Hb] was significant in
11 of 22 channels for negative pictures (channel nos. 5, 6,
10 to 15, 17, and 18), in 18 channels for positive pictures
(no. 5 to no. 22) and in 13 channels for neutral pictures
(nos. 5, 6, 8 to 15, 17, 18, and 20).
The decrease in [HHb] in the same cortical regions
reached a significant level (t>2.84) in 16 channels for pos-
itive pictures (nos. 5, 6, 8 to 17, 19 to 22), in 18 channels
for negative pictures (no. 5 to no. 22), and in 15 channels
for neutral pictures (nos. 5, 6, 8 to 15, 17 to 20, 22).
We found significant (t>2.84) differences of [O
2
Hb]
between the positive and the neutral condition but not
between the negative and the neutral condition. In detail, in
19 channels we measured significantly higher [O
2
Hb] values
for positive in comparison to neutral stimuli (nos. 1, 4 to 22,
all t>2.84) but in no channel for the negative versus neu-
tral comparison (all t<1.0). In contrast to this, for [HHB]
we found that the decrease was significantly (t<2.84)
larger in the emotional condition compared to the neutral
condition in 4 channels for the negative (nos. 9, 16, 20, and
21) and in one channel for positive pictures (no. 21, and in 4
additional channels we found corresponding trends: channel
nos. 10, 12, and 13 had tvalues <2.54).
Joined EEG-NIRS Analysis
With the above described joined EEG-NIRS analysis we
did not find any significant associations between the am-
Figure 3.
Changes in deoxygenated haemoglobin to emotional and neutral
stimuli. Displayed are the statistical maps for deoxygenated hae-
moglobin for the second level group analyses. On the left side
the activation maps for the three conditions are displayed with
the corresponding t-value scale on the left side. On the top of
the right side we displayed the arrangement of the measured
channels. In the middle and lower right side the statistical maps
for the contrasts (middle: negative versus neutral; lower: positive
versus neutral) and the corresponding t-value scale are shown.
Figure 2.
Changes in oxygenated haemoglobin to emotional and neutral stim-
uli. Displayed are the statistical maps for oxygenated haemoglobin
for the second level group analyses. On the left side the activation
maps for the three conditions are displayed with the corresponding
t-value scale on the left side. On the top of the right side we dis-
played the arrangement of the measured channels. In the middle and
lower right side the statistical maps for the contrasts (middle: nega-
tive versus neutral; lower: positive versus neutral) and the corre-
sponding t-value scale are displayed.
rHerrmann et al. r
r32 r
plitude modulated stimulus functions and the changes in
oxygenated or deoxygenated haemoglobin. Only the stimu-
lus function encoding the onsets of the stimuli revealed
the same cortical areas over the occipital cortex to be acti-
vated as in the standard analysis described above.
DISCUSSION
In this study, we found a very clear pattern of activation
over the occipital cortex, with an increase in [O
2
Hb] and a
corresponding decrease in [HHb] during visual stimula-
tion. The centers of activation were located in the middle
of the probe set with two clear hotspots over the left and
right hemisphere. The pattern of activation was very simi-
lar to the activation pattern seen from a simple visual
stimulation using a checker board [Plichta et al., 2006].
Therefore, we can conclude that our paradigm of visually
presenting emotional stimuli leads to an activation of the
occipital cortex.
As a main result of this study, we found a modulation
of cerebral oxygenation of the occipital cortex due to the
emotional content of the presented visual stimuli. The
decrease of [HHB] during picture processing was larger
for positive as well as negative stimuli compared to neu-
tral stimuli in regional specific areas over the occipital cor-
tex. For the first time using simultaneous NIRS and EEG
recording, results underscore that emotional stimuli
increase the activity of the occipital cortex. This may due
to natural selective attention occurring while viewing pic-
tures with an emotional content [Lang, 1997; Lang et al.,
1998]. Emotional pictures are not only more interesting but
also evolutionary more important and so it is more likely
to attend to this kind of stimuli [Geday et al., 2003]. It has
to be noted that channel no. 21, in which this modulation
reached its maximum, was not in the hot spot of activation
due to picture processing, but placed in the middle
between both hot spots. This result indicates that the emo-
tional content of pictures influences the brain activity to a
maximum at slightly different areas than the hot spots. For
the positive pictures, we found three additional channels
only reaching up to a tendency, which were indeed
located exactly over the hot spots (channels nos. 10, 12,
and 13).
These NIRS results are in line with the ERPs derived
from the simultaneously measured EEG. ERPs clearly indi-
cated that both, positive as well as negative stimuli, led to
an increased activity over the occipital cortex. The ampli-
tudes to positive and negative stimuli were not as positive
(and therefore in the difference to the neutral stimuli more
negative) as those to neutral stimuli as early as 240 ms af-
ter stimulus onset. No differences in the amplitudes
between positive and negative stimuli were found. With
regard to the SW potential over Pz starting at 350 ms, we
only found differences between positive and negative pic-
tures in comparison to neutral (with higher positive ampli-
tudes for emotional pictures) but no differences between
positive and negative stimuli. The EEG results were in line
Figure 4.
Mean and standard deviations of beta weights for the single channels. Displayed are the mean values
and standard deviations of the beta weights (by the linear regression analyses) for the three condi-
tions for each channel for oxygenated (upper) and deoxygenated (lower) haemoglobin.
rEmotional Modulation of Primary Visual Cortex Activity r
r33 r
with previous studies [Cuthbert et al., 2000; Schupp et al.,
2000, 2003], although the components looked slightly dif-
ferent, due to the reference position being at the tip of the
nose in this experiment.
An alternative interpretation for the modulation of the
occipital activation could be that possibly neutral stimuli
were visually less complex than the emotional stimuli. If
that would be the case, the changes in activation might be
due to the differences in visual complexity of the stimuli.
A recent study showed that this is not the case [Jungho
¨fer
et al., 2001]. The ERPs reflecting the processing of the emo-
tional content of IAPS pictures were independent of formal
visual properties of the stimuli like complexity.
In contrast to the similar results for positive and nega-
tive stimuli for [HHB], we had divergent findings for
[O
2
Hb]. Although the increase in [O
2
Hb] over the occipital
cortex was similar to the decrease of [HHb], the modula-
tion of this increase due to the emotional content was only
significant for positive, but not for negative pictures; we
found this to be similar to our pilot study [Alpers et al.,
2005a]. In the present study we could ensure that the sub-
jects processed stimuli of both conditions in a similar way,
due to the results of the ERPs. As we found a modulation
of the decrease of [HHb] for both conditions, we also
could assume that positive and negative stimuli did not
influence the cortical activity in different areas of the occi-
pital cortex. Therefore, it seems unlikely that the regions
influenced by negative stimuli are localized deeper in the
visual cortex, which might cause an inability to measure it
with NIRS. One possible explanation might be that looking
at positive and negative stimuli changes systemic variables
in different ways. For example, it has been shown that
negative pictures significantly decrease heart rate com-
pared to neutral and positive pictures, and additionally
that positive pictures lead to an increase of systolic and di-
astolic blood pressure compared to neutral and negative
stimuli [Hempel et al., 2005; Sarlo et al., 2005]. As it has
been shown that heart rate and blood pressure correlate
with [O
2
Hb] but not with [HHb] [Mehagnoul Schipper
et al., 2000], it might be that the divergent findings for
[O
2
Hb] are caused by systemic variables. This means that
a possibly decreased heart rate during negative picture
processing reduces the increase of [O
2
Hb] compared to the
neutral and positive condition, but does not influence the
[HHb] effects. As we did not measure heart rate and blood
pressure, this interpretation is still speculative, but it
should be considered in further research.
The joined EEG-NIRS analysis did not reveal any signifi-
cant linear associations between the single trial ERP ampli-
tudes and the haemodynamic responses, although this
approach has been shown before to be successful [Eichele
et al., 2005]. One explanation might be a methodological
issue of our study design. We used very short stimulus
onset intervals of 5–7 s. The power to detect significant
activation within this design was increased by using null
events (to be able to calculate contrasts). This approach
was not possible for the amplitude modulated stimulus
function, but only for the stimulus function encoding the
stimulus onset (or in the first analysis). A second problem
might be that we have to calculate the difference of the
amplitudes between the emotional conditions and the neu-
tral condition to get the EPN. Therefore, the EEG ampli-
tudes used to predict the haemodynamic responses were
modulated not only by the single trial EEG response but
also by the mean EEG response to the neutral stimuli.
Summing up, in this study we found indications for
increased activation of the occipital cortex due to emo-
tional stimulus processing, as indicated by ERPs as well as
by a larger decrease of [HHb] measured with event-related
NIRS. In the future, it might be possible to use NIRS as a
noninvasive and little constraining technology to investi-
gate the haemodynamic responses during the processing
of emotional stimuli, which might be helpful in investigat-
ing the processing of disorder specific visual cues in emo-
tional disorders such as anxiety or addiction.
REFERENCES
Abrisqueta-Gomez J, Bueno OFA, Oliveira MGM, Bertolucci PHF
(2002): Recognition memory for emtotional pictures in Alzhei-
mer’s patients. Acta Neurol Scand 105:51–54.
Alpers GW, Pauli P (2006): Emotional pictures predominate in bin-
ocular rivalry. Cognit Emot 20:596–607.
Alpers GW, Herrmann MJ, Pauli P, Fallgatter AJ (2005a): Emo-
tional arousal and activation of the visual cortex: A near infra-
red spectroscopy analysis. J Psychophysiol 19:106.
Alpers GW, Ruhleder M, Walz N, Mu
¨hlberger A, Pauli P (2005b):
Binocular rivalry between emotional stimuli: A validation using
conditioned stimuli and EEG. Int J Psychophysiol 57:25–32.
Amaral DG, Price JL, Pitkanen A, Carmichael S (1992): Anatomical
organization of the primate amygdaloid complex. In: Aggleton
JP, editor. The Amygdala. New York: Wiley. pp 1–66.
Cochrane D, Orcutt G (1949): Application of least squares regres-
sion to relationships containing autocorrelated error terms.
Am Stat Assoc 44:32–61.
Cuthbert BN, Schupp HT, Bradley MM, Birbaumer N, Lang PJ
(2000): Brain potentials in affective picture processing: Covaria-
tion with autonomic arousal and affective report. Biol Psychol
52:95–111.
Davis M, Whalen PJ (2001): The amygdala: Vigilance and emotion.
Mol Psychiatry 6:13–34.
Dolcos F, Cabeza R (2002): Event-related potentials of emotional
memory: Encoding pleasant, unpleasant, and neutral pictures.
Cognit Affect Behav Neurosci 2:252–263.
Eichele T, Specht K, Moosmann M, Jongsma ML, Quiroga RQ,
Nordby H, Hugdahl K (2005): Assessing the spatiotemporal
evolution of neuronal activation with single-trial event-related
potentials and functional MRI. Proc Natl Acad Sci USA 102:
17798–17803.
Fox PT, Raichle ME (1986): Focal physiological uncoupling of cere-
bral blood flow and oxidative metabolism during somatosen-
sory stimulation in human subjects. Proc Natl Acad Sci USA
83:1140–1144.
Franceschini MA, Fantini S, Thompson JH, Culver JP, Boas DA
(2003): Hemodynamic evoked response of the sensorimotor
cortex measured noninvasively with near-infrared optical
imaging. Psychophysiology 40:548–560.
rHerrmann et al. r
r34 r
Geday J, Gjedde A, Boldsen A-S, Kupers R (2003): Emotional va-
lence modulates activity in the posterior fusiform gyrus and in-
ferior medial prefrontal cortex in social perception. Neuro-
image 18:675–684.
Hempel RJ, Tulen JH, van Beveren NJ, van Steenis HG, Mulder
PG, Hengeveld MW (2005): Physiological responsivity to
emotional pictures in schizophrenia. J Psychiatr Res 39:509–
518.
Herrmann MJ, Ehlis AC, Fallgatter AJ (2003): Prefrontal activation
through task requirements of emotional induction measured
with NIRS. Biol Psychol 64:255–263.
Herrmann MJ, Ehlis A-C, Scheuerpflug P, Fallgatter AJ (2005): Op-
tical topography with near-infrared spectroscopy during a
verbal-fluency task. J Psychophysiol 19:100–105.
Horovitz SG, Gore JC (2004): Simultaneous event-related potential
and near-infrared spectroscopic studies of semantic processing.
Hum Brain Mapp 22:110–115.
Hoshi Y (2003): Functional near-infrared optical imaging: Utility
and limitations in human brain mapping. Psychophysiology
40:511–520.
Hoshi Y, Shimada M, Sato C, Iguchi Y (2006): Reevaluation of near-
infrared light propagation in the adult human head: Implica-
tions for functional near-infrared spectroscopy. J Biomed Opt
10:064032.
Jungho
¨fer M, Bradley MM, Elbert TR, Lang PJ (2001): Fleeting
images: A new look at early emotion discrimination. Psycho-
physiology 38:175–178.
Koizumi H, Yamashita A, Maki T, Ito Y, Itagaki H, Kennan R
(1999): High-order brain function analysis by trans-cranial
dynamic near-infrared spectroscopy imaging. J Biomed Opt
4:403–413.
Lang PJ (1997): Motivated attention: Affect, activation and action.
In: Lang PJ, editor. Attention and Orienting: Sensory and Moti-
vational Processes. Mahwah, NJ: Erlbaum. pp 97–135.
Lang PJ, Bradley MM, Fitzsimmons FR, Cuthbert BN, Scott JD,
Moulder B, Nangia V (1998): Emotional arousal and activation of
the visual cortex: An fMRI analysis. Psychophysiology 35:199–
210.
Lang PJ, Bradley MM, Cuthbert BN (1999): International Affective
Picture System (IAPS): Technical Manual and Affective Rat-
ings. Gainesville, FL: Center for Research in Psychophysiol-
ogy.
LeDoux J (1998): Fear and the brain: Where have we been, and
where are we going? Biol Psychiatry 44:1229–1238.
Mehagnoul Schipper DJ, Vloet LC, Colier WN, Hoefnagels WH,
Jansen RW (2000): Cerebral oxygenation declines in healthy el-
derly subjects in response to assuming the upright position.
Stroke 31:1615–1620.
Obrig H, Villringer A (2003): Beyond the visible–imaging the
human brain with light. J Cereb Blood Flow Metab 23:1–18.
O
¨hman A, Mineka S (2001): Fears, phobias, and preparedness: To-
ward an evolved module of fear and fear learning. Psychol
Rev 108:483–522.
Okada E, Delpy DT (2003): Near-infrared light propagation in an
adult head model. I. Modeling of low-level scattering in the
cerebrospinal fluid layer. Appl Opt 42:2906–2914.
Pessoa L, Kastner S, Ungerleider LG (2002): Attentional control of
the processing of neural and emotional stimuli. Brain Res Cogn
Brain Res 15:31–45.
Phan KL, Wager T, Taylor SF, Liberzon I (2002)): Functional neu-
roanatomy of emotion: A meta-analysis of emotion activation
studies in PET and fMRI. Neuroimage 16:331–348.
Plichta MM, Herrmann MJ, Baehne CG, Ehlis A-C, Richter MM,
Pauli P, Fallgatter AJ (2006): Event-related functional near
infrared spectroscopy (fNIRS): Are the measurements reliable?
Neuroimage 31:116–124.
Posner MI, Raichle ME (1995): Precis of images of mind. Behav
Brain Sci 18:327–383.
Sarlo M, Palomba D, Buodo G, Minghetti R, Stegagno L (2005):
Blood pressure changes highlight gender differences in emo-
tional reactivity to arousing pictures. Biol Psychol 70:188–196.
Schroeter ML, Zysset S, Kupka T, Kruggel F, Yves von Cramon D
(2002): Near-infrared spectroscopy can detect brain activity
during a color-word matching Stroop task in an event-related
design. Hum Brain Mapp 17:61–71.
Schupp HT, Cuthbert BN, Bradley MM, Cacioppo JT, Ito T, Lang
PJ (2000): Affective picture processing: The late positive poten-
tial is modulated by motivational relevance. Psychophysiology
37:257–261.
Schupp HT, Junghofer M, Weike AI, Hamm AO (2003): Emotional
facilitation of sensory processing in the visual cortex. Psychol
Sci 14:7–13.
Strangman G, Boas DA, Sutton JP (2002): Non-invasive neuroi-
maging using near-infrared light. Biol Psychiatry 52:679–693.
Villringer K, Minoshima S, Hock C, Obrig H, Ziegler S, Dirnagl U,
Shwainger M, Villringer A (1997): Assessment of local brain
activation. A simultaneous PET and nearinfrared spectroscopy
study. Adv Exp Med Biol 413:149–153.
rEmotional Modulation of Primary Visual Cortex Activity r
r35 r
