Your goal is to summarize the article that you have chosen. In your summary you need to address the following:
•State the hypothesis
•Describe the aims of the research project
•Note: The aims and hypothesis are different
•How was the hypothesis tested?
•Describe the experimental design
•What are the independent and dependent variables?
•Describe the levels of the variables
•Describe the subjects
•Mention the relevant demographic characteristics
•Describe experimental conditions.
•Within or Between Subjects Design
•Survey
•Describe the experimental procedure in enough detail so that the reader will understand the basics of the study.
•What are the results?
•Was the hypothesis supported?
•Are the conclusions justified?
•Any confounding variables?
•Lastly, if you had to replicate this experiment what would you do to improve it?
•That is, how would you design a better experiment?
•If you are reviewing a “review article,” then describe the “gaps” in the literature review.
Point Allocation
•Describe the aims of the research project
•Note: The aims and hypothesis are different
•How was the hypothesis tested?
•Describe the experimental design
•What are the independent and dependent variables?
•Describe the levels of the variables
•Describe the subjects
•Mention the relevant demographic characteristics
•Describe experimental conditions.
•Within or Between Subjects Design
•Survey
•Describe the experimental procedure in enough detail so that the reader will understand the basics of the study.
•What are the results?
•Was the hypothesis supported?
•Are the conclusions justified?
•Any confounding variables?
•Lastly, if you had to replicate this experiment what would you do to improve it?
•That is, how would you design a better experiment?
•If you are reviewing a “review article,” then describe the “gaps” in the literature review.
Point Allocation
Distinguishing specific sexual and general emotional effects in
fMRI—Subcortical and cortical arousal during erotic picture viewing
Martin Walter,a,? Felix Bermpohl,b Harold Mouras,c Kolja Schiltz,a,d Claus Tempelmann,d
Michael Rotte,d Hans Jochen Heinze,d Bernhard Bogerts,a and Georg Northoff a
a
Department of Psychiatry, University Hospital, Magdeburg, Germany
b
Department of Psychiatry and Psychotherapy, Charité-University Medicine, Berlin, Campus Mitte, Berlin, Germany
c
Laboratoire de Neurosciences Fonctionnelles et Pathologies, Faculté de Médecine, Amiens, France
d
Department of Neurology II, University Hospital, Magdeburg, Germany
Received 5 October 2007; revised 8 January 2008; accepted 24 January 2008
Available online 7 February 2008
Sexual activity involves excitement with high arousal and pleasure as pertinent to an ongoing debate on different accounts of emotion
typical features of emotions. Brain activations specifically related to erotic processing: Proponents of the “basic emotion theory” would suggest
feelings and those related to general emotional processing are therefore on the basis of animal findings and translational observations, that a
hard to disentangle. Using fMRI in 21 healthy subjects (11 males and 10 limited set of hard wired, distinct basic emotions exists across human
females), we investigated regions that show activations specifically related and non-human animals (Panksepp, 1998). These encompass dif-
to the viewing of sexually intense pictures while controlling for general ferent executive systems such as rage, fear and lust. Such primary
emotional arousal (GEA) or pleasure. Activations in the ventral striatum emotive systems would be based on hierarchically organized circuits
and hypothalamus were found to be modulated by the stimulus’ specific
that are genetically imprinted and orchestrate certain affective
sexual intensity (SSI) while activations in the anterior cingulate cortex
neurodynamic responses (Panksepp, 1992). While this view would
were associated with an interaction between sexual intensity and emotional
valence. In contrast, activation in other regions like the dorsomedial try to discern systems mediating SSI from other basic systems, the
prefrontal cortex, the mediodorsal thalamus and the amygdala was dimensional approach relies on processes of cognitive appraisal
associated only with a general emotional component during sexual arousal. attributing a general emotional arousal (GEA) and hedonic valence to
No differences were found in these effects when comparing females and an emotional experience. While some have questioned the existence
males. Our findings demonstrate for the first time neural differentiation or applicability of the “basic emotion” approach (Ortony and Turner,
between emotional and sexual components in the neural network 1990), there is evidence that both views may apply on different system
underlying sexual arousal. levels (Panksepp, 1992).
© 2008 Elsevier Inc. All rights reserved. Recent studies investigating the neural correlates of human sexual
arousal revealed a broad network of cortical and subcortical brain
regions that are activated during the processing of visual erotic stimuli.
Cortical regions comprise of a parieto-occipito-temporal network,
mainly consisting of the superior and inferior parietal lobes (SPL,
Introduction
IPL), the lateral occipital cortex (LOC), as well as the medial prefrontal
cortex (MPFC), anterior cingulate cortex (ACC) and insula. Sub-
Sexual excitement is characterized by feelings of specific sexual
cortical regions include the ventral and dorsal striatum, the amygdala
intensity as well as by a strong general emotional arousal and feeling
and the hypothalamus (Arnow et al., 2002; Ferretti et al., 2005;
of pleasure. Any neural investigation that targets the neural network
Karama et al., 2002; Moulier et al., 2006; Mouras et al., 2003; Paul
underlying this specific sexual intensity (SSI) is confronted with the
et al., in press; Ponseti et al., 2006; Redoute et al., 2000; Stoleru et al.,
problem of differentiating it from other structures mediating general
1999; see also Mouras, 2007, and Mouras and Stoleru, 2007, for an
emotional processing. This distinction of separate structures is highly
extensive overview).
To specify the role of each of these regions during sexual arousal
(SA), prior investigations attempted to relate them to distinct com-
? Corresponding author. Otto von Guericke University, University ponents of a multidimensional model of SA. According to this model,
Hospital, Department of Psychiatry, Leipziger Strasse 44, 39112 Magde- cognitive, motivational, autonomic and emotional components con-
burg, Germany. Fax: +49 391 6713167. tribute to SA (Redoute et al., 2000; Stoleru et al., 1999). Cognitive
E-mail address: martin.walter@med.ovgu.de (M. Walter). components include processes of early appraisal of a stimulus as sexual
Available online on ScienceDirect (www.sciencedirect.com). incentive and direction of attentional focus to a sexual target. Related
1053-8119/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.neuroimage.2008.01.040
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1483
neural activity was assumed to be mediated mainly by activations in 2007). Since SA induced by visual erotic stimuli inherently involves
temporo-occipito-parietal regions as well as in the orbitofrontal cortex the processing of bodies and body parts, neural investigation needs
(Mouras, 2004; Mouras et al., 2003). On the basis of correlational to distinguish activations mediating SSI from those related to the
findings between markers of penile tumescence and concurrent neural processing of body parts.
activity, the autonomic component of SA was related in particular to
activations in the insula, hypothalamus and anterior cingulate (Arnow Aims and hypotheses
et al., 2002; Ferretti et al., 2005; Moulier et al., 2006; Redoute et al.,
2000). On the other hand, motivational components including adap- Based on previous imaging findings, we tried to elicit neural
tation of motor responses were related to activations in the basal activity in the subcortical–cortical network modulated by SSI as op-
ganglia (Redoute et al., 2000). The involvement of the medial thala- posed to GEA. Sexual stimuli can be considered strong incentives and
mus, ventral striatum (VS) and amygdala was related to the emotional sexual activity a rewarding situation (Fisher et al., 2006). Considering
dimension of hedonic value, i.e., pleasantness or valence, or to general its pivotal role in reward mechanisms (Knutson and Wimmer, 2007;
emotional arousal (GEA) during SA (Ferretti et al., 2005; Karama Yacubian et al., 2007), the ventral striatum was hypothesized to show
et al., 2002; Redoute et al., 2000). an effect for SSI independent of GEA or emotional valence. A further
One should, however, be aware that all the above regions are also specific effect of sexually intense pictures was considered for the
activated by non-erotic emotional stimuli. Recruitment of attentional hypothalamus given the well described effects of both hypothalamic
resources leads to higher activations in temporo-parieto-occipital lesions and electric stimulation on sexual behavior in animals
regions for more emotionally arousing stimuli (Ishai et al., 2004; Lang (Karama et al., 2002; McKenna, 1999) as well as its role in initiating
et al., 1998; Pessoa et al., 2002b). This recruitment is mainly modulated specific autonomous responses during SA (Ferretti et al., 2005;
by the amygdala and prefrontal cortex (Desimone and Duncan, 1995; Georgiadis and Holstege, 2005).
Lane et al., 1998; Pessoa et al., 2002a; Vuilleumier et al., 2001; In addition to those regions being specific to SSI, we investigated
Vuilleumier and Pourtois, 2007; Vuilleumier et al., 2004). The regions of the subcortical–cortical network that allow for interaction
involvement of the MPFC and the ACC in non-erotic emotional between processes of SSI and GEA and those regions whose ac-
processing is also well documented (Grimm et al., 2006; Phan et al., tivation during SA might be modulated only by GEA. Using func-
2002). Therefore, their attribution to a distinct and specific autonomic tional magnetic resonance imaging (fMRI), we presented carefully
component of SA together with the anterior insula and hypothalamus matched erotic and emotional, non-erotic, as well as non-erotic, non-
may be questionable. Although measures of penile tumescence emotional (neutral) visual stimuli. In addition, subjects evaluated
suggested the involvement of these particular brain structures in the these stimuli in a post-scanning session according to their subjective
erectile response of sexual arousal (Arnow et al., 2002; Ferretti et al., experience of SSI, GEA and emotional valence. Subjective ratings
2005; Moulier et al., 2006), concurrent neural activations could still be were then used to guide analyses of fMRI data.
attributable to somatosensory integration of erection as well as to the
pleasant feeling associated with its perception (Mouras, 2004). Methods
Given that the very same regions are activated during both erotic
and non-erotic emotional processing, control of GEA during SA is Experimental design
particularly important in the identification of neural activity spe-
cifically related to SSI during SA. Neural specificity in particular brain Twenty-one heterosexually oriented subjects (11 males and 10
regions would be indicated by occurrence of higher activations in females) were asked to view and passively experience a total of 256
stimuli inducing higher SSI and the independence of this activation bodily-erotic, emotional and neutral photographs taken from the
from dimensions of valence or GEA. In order to achieve such neural International Affective Picture System (Lang et al., 2005). Age range
differentiation, neuroimaging studies attempted to reduce differences was 21–36 years (mean: 25.73, median: 24) for males and 22–28 years
between erotic and non-erotic visual control stimuli in these general (mean: 23.9, median: 23) for females. Erotic (bodily) pictures
emotional dimensions by introducing control conditions of humorous depicted partially or completely naked people of both sexes alone or in
or sports clips (Arnow et al., 2002; Redoute et al., 2000). However, couples of men and women in a context of positive or negative
these stimuli still differed in general emotional dimensions like in- valence. Mean number of people per picture was 1.94 (±0.70) and did
duced feeling of pleasure (Mouras et al., 2003). not differ between positive and negative bodily pictures. Non-bodily-
The challenge persists to examine which parts of the described emotional pictures showed one or more humans in sports scenes or
subcortical–cortical network display additional activity during other social interactions as well as emotionally arousing non-human
processing of erotic stimuli without being modulated by emotional motives in a positive or negative context.
valence or GEA. Past investigations often used long dynamic video Bodily erotic and non-bodily, emotional picture categories were
clips to induce sexual excitement. Although the long duration ensures matched for standard ratings of arousal, dominance and valence as
induction of sexual feelings, it also allows several emotional states to provided by the IAPS catalogue (Lang et al., 2005). Neutral and
occur within one clip, which makes it rather difficult to efficiently emotional pictures were matched for dominance and mean valence
control for GEA or valence (Janssen et al., 2003). Short presentations ratings, while emotional pictures had higher arousal values. Mean
of still pictures may be matched more easily and more accurately for valence of emotional stimuli was very low for negative and very high
induced emotional states. Furthermore, the bodily content of erotic for positive stimuli while neutral stimuli always had intermediate
pictures has to be manipulated independently of SSI to account for valence levels.
activations related to the processing of human body parts. This would Runs consisted of either non-bodily-emotional and neutral (4 runs)
require the introduction of both sexual-bodily and non-sexual-bodily or non-bodily-emotional and bodily (erotic) stimuli (4 runs) and were
stimuli. Recent investigations demonstrated that human body parts by presented in a pseudorandomized order. Each run consisted of a total
themselves activate regions which are implicated in part in SA of 32 images and had a total length of 8 min 32 s. Stimulus duration
(Downing et al., 2001; Downing et al., 2007; Peelen and Downing, was intended to address two main concerns. While SA in earlier
1484 M. Walter et al. / NeuroImage 40 (2008) 1482–1494
studies was induced either by short movies or pictures of at least 20– sequence 60 slices sagittal, 2.8 mm thickness), (b) T1 weighted
30 s, the shortest duration shown to evoke penile tumescence was 10 s anatomic images coplanar with the functional images (23 slices,
(McConaghy, 1989). Aiming to be especially sensitive to early aligned to the plane connecting the anterior and posterior commissure
components of SA such as cognitive and emotional evaluation, axis covering the whole head in oblique axial orientation), (c) inversion
imaging studies investigating effective emotional stimulation use recovery T1 weighted echo planar images coplanar with the functional
shorter stimulus durations which reliably induce an emotional re- images and (d) echo planar functional images sensitive to BOLD
sponse and support sufficient specificity of the investigated emotional contrast (257 sequential acquisitions, 23 slices with 3.125 mm in-plane
process. Consequently, a presentation duration of 5 s was chosen, resolution, 5 mm thickness, 1 mm gap; T2* weighted gradient echo
which has been shown to reliably invoke subjective experience of SA sequence: TR 2 s, TE 40 ms). The first seven images were discarded
(measured on a 1–9 item scale) and found to be short enough to due to T1 saturation effects. During image acquisition, stimuli were
maintain a certain emotional state (Heinzel et al., 2006). Pictures were projected on a mirror mounted on the headcoil using a LCD projector.
presented in a randomized order across runs and subjects were asked
to promptly press a button at the appearance of each picture to monitor Image analysis
their vigilance in the scanner, as assessed by reaction times.
In order to control for the possible influence of preceding attention Image preprocessing and statistical analyses were carried out using
on the experience of the presented stimuli, half of the pictures were MATLAB 6.5.1 and SPM2 (SPM2 software package; Wellcome
preceded by an expectancy period of 4.5 s indicating the type of the Department of Cognitive Neurology, London, UK; see http://www.fil.
subsequently presented picture. Expected and unexpected pictures ion.ucl.ac.uk/spm2.html). A total of 2000 (8 × 250) volume images
(see below) were balanced for the number of positive and negative were realigned to the first image to correct for head movement
erotic and emotional or neutral pictures, the total number of pictures, between scans, mean-adjusted by proportional scaling, resliced and
and IAPS values. Following an approach described by Bermpohl et al. normalized into standard stereotactic space (resulting in an isotrophic
(2006), the expectancy period was indicated by the presentation of 3 mm resolution). The cut-off for subjects showing translational head
a white arrow on a dark background. Downward arrows indicated movement was set to 2 mm and for rotational movement to 1°. No
sexual, upward arrows emotional and rightward arrows neutral ex- subject had to be excluded. Image normalization was performed using
pectancy. After the picture presentation, a fixation cross (intertrial the MNI (Montreal Neurological Institute) template provided by
interval) was presented for a duration of 8 s. Arrows and fixation cross SPM. Spatial transformation included both linear and non-linear
were of equal size, color and luminance and were centered on a black dimensions and used the non-linear sampling algorithm provided with
background. Prior to the experimental session, subjects were SPM2. Data were thereafter expressed in terms of standard
familiarized with the paradigm by completing a test run consisting stereotactic coordinates in the x, y and z axes. Transformed functional
of 32 trials. data sets from each subject were smoothed with a Gaussian kernel of
8 mm (full-width half-maximum) for the group analysis to meet the
Behavioral and subjective evaluation statistical requirements of the General Linear Model and to
compensate for normal variation in individual brain size, shape and
Reaction times were defined as the time between picture onset and sulcal/gyral anatomy across subjects. Subject-specific low frequency
subsequent button press during the fMRI experiment. Subjective drifts in signals were removed by a high pass filter of 128 s.
ratings of the pictures were conducted outside the scanner after the For each subject, a design matrix modeling of emotional and erotic
fMRI session using a visual analogue scale. This approach was pre- or emotional and neutral IAPS pictures was defined. For the analysis
ferred to an intra scan assessment to avoid judgment related cognitive of effects of preceding expectancy, unexpected and expected viewing
influence which can confound neural activity during emotional of the stimulus categories were modeled as separate events. This
perception (Blair et al., 2007; Grimm et al., 2006; Mitchell et al., 2007; enabled us to compare the main contrasts for stimuli presented in an
Northoff et al., 2006; Taylor et al., 2003). Ratings of experienced SSI, “expected mode” (after expectancy cues) or in an unexpected mode
GEA and valence were acquired. Subjects were asked to rate their (after fixation).
own personal experience instead of cognitively evaluating the erotic Besides these experimental conditions, regressors for the baseline
and emotional properties of the stimuli. During analysis of behavioral condition (fixation cross) and expectancy periods were entered into a
data, average scores of ratings were calculated for the whole group. design matrix including a total of 7 regressors. To investigate effects
For intra-scan reaction times and reports of GEA, SSI and valence, of positive and negative subcategories of erotic and emotional stimuli
analyses of covariance (ANOVA) containing the factors stimulus and to make explicit use of their sexual and emotional properties,
category (bodily or non-bodily-emotional), stimulus subcategory these were modeled separately in a third design matrix on single
(positive or negative) and mode of presentation (unexpected or subject level. Specific effects were tested by applying appropriate
expected) were calculated for main effects and interactions. Gender of linear contrasts to the parameter estimates for each condition, resulting
subject was entered as a between subjects factor. Since many stimuli in a t-statistic for each voxel. Analysis of effects for the whole group of
depicted couples that induce high sexual intensities (Hamann et al., all subjects and for groups of males and females was performed in a
2004), ratings of all stimuli were regarded together and no further second level, random effects model on a whole brain level (Friston
analysis was performed on gender effects on the viewing of opposite et al., 1999).
sex stimuli. We report, if not explicitly mentioned otherwise, differences in
activated regions that survived a threshold of p b 0.001, corrected for
Imaging–scanning procedures multiple comparisons using FDR correction (false discovery rate;
Genovese et al., 2002) for clusters consisting of at least 10 contiguous
Data acquisition was conducted on a 1.5 Tesla General Electric voxels (k = 10) for main contrasts, and a conjoint p-threshold of
Signa scanner using a standard headcoil. Imaging procedures included p b 0.05 (FDR-corrected) for conjunction analyses or serial subtraction
collection of (a) structural high resolution images (rf-spoiled GRASS contrasts (interactions) of more than one contrast at the second level.
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1485
Table 1 analysis (Nichols et al., 2005), with a contrast of the type [S N s] that
Mean ratings (±SD) are reported for all bodily, non-bodily-emotional and compared two stimuli which differed in their ratings of SSI but not in
non-bodily neutral picture categories, as well as for positive and negative GEA or valence. Activations in common areas revealed by this con-
subcategories junction analysis were interpreted as being related to greater sexual
Stimulus category SSI GEA Valence intensity common in both contrasts but not as (common) effects of
Mean SD Mean SD Mean SD greater GEA or valence. Characterization of a stimulus subtype as
more or less emotionally arousing or sexually intense was done
Bodily-emotional 4.06 1.12 6.38 1.31 4.70 0.30
according to the behavioral results and was always relative to the other
Positive 5.93 1.27 6.07 1.5 6.95 0.76
stimulus subtype entered in the contrast (Fig. 4a).
Negative 2.19 1.27 6.70 1.17 2.46 0.76
Non-bodily-emotional 1.16 0.46 5.75 1.32 4.41 0.36 The linear relationship between subjective ratings and neural
Positive 1.25 0.76 5.00 1.56 6.32 0.73 activations was then tested adding subjects’ ratings of SSI, GEA and
Negative 1.07 0.18 6.56 1.23 2.50 0.74 valence for each stimulus as parametric regressors (Lewis et al., 2007;
Neutral 1.12 0.23 2.61 1.03 5.14 0.30 Nagai et al., 2004). This new design matrix consisted of regressors of
Subjects were asked to rate “How sexually intense” (1–9), “How emotionally interest for bodily and non-bodily stimuli which were then modulated
intense” (1–9) and “What kind of emotion” (valence with 1: very negative to 9: by three parametric regressors of first polynomial order for individual
very positive) they experienced the picture. subjective ratings of SSI, GEA and valence. Effects of all parametric
regressors were then tested in a second level random effects model
The strategy of analysis aimed to first reveal several regions (Friston et al., 1999). This analysis aimed to reveal a parametric
involved in SA and then specify their involvement in more specific (linear) modulation by SSI, GEA or valence in regions showing ef-
aspects as SSI or GEA. First, BOLD effects of all positive bodily fects either for SSI, GEA or valence.
pictures greater than those of all positive non-bodily stimuli were Effects of greater GEA within areas activated during sexual
assessed by a linear contrast. This contrast was further performed arousal were tested by a conjunction of [bodily N non-bodily pictures]
separately for expected and unexpected stimuli and contrast images of and [non-bodily N neutral pictures], identifying common activations
both contrasts were compared on a second level, random effects, paired for higher GEA but not SSI. The interaction of specific sexual
t-test. Gender effects in the main contrast were tested on the second intensity and valence was tested by a serial subtraction search for
level, performing a two sample t-test of subjects’ contrast images. higher effects of valence in bodily stimuli than in non-bodily-
To investigate specific neuronal effects of differences in SSI be- emotional stimuli.
tween stimulus categories, specific contrasts were chosen according to Second level analyses of more than one contrast per subject were
effects revealed by behavioral results in a next step. One contrast corrected for non-sphericity. For resulting regions of interest, neural
could compare a stimulus type rated higher for SSI (S = higher SSI), responses were extracted and transformed to percentage signal changes
higher for GEA (E = higher GEA) and higher for valence (P= more for our regressors of interest in each subject using the MarsBar toolbox
pleasant) to a less sexually intense (s = lower SSI), less emotionally (Brett et al., 2002). Temporal patterns were investigated by time course
arousing (e = lower GEA) and less pleasant (p = less pleasant) stimulus analyses applying a finite impulse response model (FIR) which does
([SEP N sep]). This contrast was then entered into a conjunction not make an assumption on the resulting signal changes after stimulus
typical features of emotions. Brain activations specifically related to erotic processing: Proponents of the “basic emotion theory” would suggest
feelings and those related to general emotional processing are therefore on the basis of animal findings and translational observations, that a
hard to disentangle. Using fMRI in 21 healthy subjects (11 males and 10 limited set of hard wired, distinct basic emotions exists across human
females), we investigated regions that show activations specifically related and non-human animals (Panksepp, 1998). These encompass dif-
to the viewing of sexually intense pictures while controlling for general ferent executive systems such as rage, fear and lust. Such primary
emotional arousal (GEA) or pleasure. Activations in the ventral striatum emotive systems would be based on hierarchically organized circuits
and hypothalamus were found to be modulated by the stimulus’ specific
that are genetically imprinted and orchestrate certain affective
sexual intensity (SSI) while activations in the anterior cingulate cortex
neurodynamic responses (Panksepp, 1992). While this view would
were associated with an interaction between sexual intensity and emotional
valence. In contrast, activation in other regions like the dorsomedial try to discern systems mediating SSI from other basic systems, the
prefrontal cortex, the mediodorsal thalamus and the amygdala was dimensional approach relies on processes of cognitive appraisal
associated only with a general emotional component during sexual arousal. attributing a general emotional arousal (GEA) and hedonic valence to
No differences were found in these effects when comparing females and an emotional experience. While some have questioned the existence
males. Our findings demonstrate for the first time neural differentiation or applicability of the “basic emotion” approach (Ortony and Turner,
between emotional and sexual components in the neural network 1990), there is evidence that both views may apply on different system
underlying sexual arousal. levels (Panksepp, 1992).
© 2008 Elsevier Inc. All rights reserved. Recent studies investigating the neural correlates of human sexual
arousal revealed a broad network of cortical and subcortical brain
regions that are activated during the processing of visual erotic stimuli.
Cortical regions comprise of a parieto-occipito-temporal network,
mainly consisting of the superior and inferior parietal lobes (SPL,
Introduction
IPL), the lateral occipital cortex (LOC), as well as the medial prefrontal
cortex (MPFC), anterior cingulate cortex (ACC) and insula. Sub-
Sexual excitement is characterized by feelings of specific sexual
cortical regions include the ventral and dorsal striatum, the amygdala
intensity as well as by a strong general emotional arousal and feeling
and the hypothalamus (Arnow et al., 2002; Ferretti et al., 2005;
of pleasure. Any neural investigation that targets the neural network
Karama et al., 2002; Moulier et al., 2006; Mouras et al., 2003; Paul
underlying this specific sexual intensity (SSI) is confronted with the
et al., in press; Ponseti et al., 2006; Redoute et al., 2000; Stoleru et al.,
problem of differentiating it from other structures mediating general
1999; see also Mouras, 2007, and Mouras and Stoleru, 2007, for an
emotional processing. This distinction of separate structures is highly
extensive overview).
To specify the role of each of these regions during sexual arousal
(SA), prior investigations attempted to relate them to distinct com-
? Corresponding author. Otto von Guericke University, University ponents of a multidimensional model of SA. According to this model,
Hospital, Department of Psychiatry, Leipziger Strasse 44, 39112 Magde- cognitive, motivational, autonomic and emotional components con-
burg, Germany. Fax: +49 391 6713167. tribute to SA (Redoute et al., 2000; Stoleru et al., 1999). Cognitive
E-mail address: martin.walter@med.ovgu.de (M. Walter). components include processes of early appraisal of a stimulus as sexual
Available online on ScienceDirect (www.sciencedirect.com). incentive and direction of attentional focus to a sexual target. Related
1053-8119/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.neuroimage.2008.01.040
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1483
neural activity was assumed to be mediated mainly by activations in 2007). Since SA induced by visual erotic stimuli inherently involves
temporo-occipito-parietal regions as well as in the orbitofrontal cortex the processing of bodies and body parts, neural investigation needs
(Mouras, 2004; Mouras et al., 2003). On the basis of correlational to distinguish activations mediating SSI from those related to the
findings between markers of penile tumescence and concurrent neural processing of body parts.
activity, the autonomic component of SA was related in particular to
activations in the insula, hypothalamus and anterior cingulate (Arnow Aims and hypotheses
et al., 2002; Ferretti et al., 2005; Moulier et al., 2006; Redoute et al.,
2000). On the other hand, motivational components including adap- Based on previous imaging findings, we tried to elicit neural
tation of motor responses were related to activations in the basal activity in the subcortical–cortical network modulated by SSI as op-
ganglia (Redoute et al., 2000). The involvement of the medial thala- posed to GEA. Sexual stimuli can be considered strong incentives and
mus, ventral striatum (VS) and amygdala was related to the emotional sexual activity a rewarding situation (Fisher et al., 2006). Considering
dimension of hedonic value, i.e., pleasantness or valence, or to general its pivotal role in reward mechanisms (Knutson and Wimmer, 2007;
emotional arousal (GEA) during SA (Ferretti et al., 2005; Karama Yacubian et al., 2007), the ventral striatum was hypothesized to show
et al., 2002; Redoute et al., 2000). an effect for SSI independent of GEA or emotional valence. A further
One should, however, be aware that all the above regions are also specific effect of sexually intense pictures was considered for the
activated by non-erotic emotional stimuli. Recruitment of attentional hypothalamus given the well described effects of both hypothalamic
resources leads to higher activations in temporo-parieto-occipital lesions and electric stimulation on sexual behavior in animals
regions for more emotionally arousing stimuli (Ishai et al., 2004; Lang (Karama et al., 2002; McKenna, 1999) as well as its role in initiating
et al., 1998; Pessoa et al., 2002b). This recruitment is mainly modulated specific autonomous responses during SA (Ferretti et al., 2005;
by the amygdala and prefrontal cortex (Desimone and Duncan, 1995; Georgiadis and Holstege, 2005).
Lane et al., 1998; Pessoa et al., 2002a; Vuilleumier et al., 2001; In addition to those regions being specific to SSI, we investigated
Vuilleumier and Pourtois, 2007; Vuilleumier et al., 2004). The regions of the subcortical–cortical network that allow for interaction
involvement of the MPFC and the ACC in non-erotic emotional between processes of SSI and GEA and those regions whose ac-
processing is also well documented (Grimm et al., 2006; Phan et al., tivation during SA might be modulated only by GEA. Using func-
2002). Therefore, their attribution to a distinct and specific autonomic tional magnetic resonance imaging (fMRI), we presented carefully
component of SA together with the anterior insula and hypothalamus matched erotic and emotional, non-erotic, as well as non-erotic, non-
may be questionable. Although measures of penile tumescence emotional (neutral) visual stimuli. In addition, subjects evaluated
suggested the involvement of these particular brain structures in the these stimuli in a post-scanning session according to their subjective
erectile response of sexual arousal (Arnow et al., 2002; Ferretti et al., experience of SSI, GEA and emotional valence. Subjective ratings
2005; Moulier et al., 2006), concurrent neural activations could still be were then used to guide analyses of fMRI data.
attributable to somatosensory integration of erection as well as to the
pleasant feeling associated with its perception (Mouras, 2004). Methods
Given that the very same regions are activated during both erotic
and non-erotic emotional processing, control of GEA during SA is Experimental design
particularly important in the identification of neural activity spe-
cifically related to SSI during SA. Neural specificity in particular brain Twenty-one heterosexually oriented subjects (11 males and 10
regions would be indicated by occurrence of higher activations in females) were asked to view and passively experience a total of 256
stimuli inducing higher SSI and the independence of this activation bodily-erotic, emotional and neutral photographs taken from the
from dimensions of valence or GEA. In order to achieve such neural International Affective Picture System (Lang et al., 2005). Age range
differentiation, neuroimaging studies attempted to reduce differences was 21–36 years (mean: 25.73, median: 24) for males and 22–28 years
between erotic and non-erotic visual control stimuli in these general (mean: 23.9, median: 23) for females. Erotic (bodily) pictures
emotional dimensions by introducing control conditions of humorous depicted partially or completely naked people of both sexes alone or in
or sports clips (Arnow et al., 2002; Redoute et al., 2000). However, couples of men and women in a context of positive or negative
these stimuli still differed in general emotional dimensions like in- valence. Mean number of people per picture was 1.94 (±0.70) and did
duced feeling of pleasure (Mouras et al., 2003). not differ between positive and negative bodily pictures. Non-bodily-
The challenge persists to examine which parts of the described emotional pictures showed one or more humans in sports scenes or
subcortical–cortical network display additional activity during other social interactions as well as emotionally arousing non-human
processing of erotic stimuli without being modulated by emotional motives in a positive or negative context.
valence or GEA. Past investigations often used long dynamic video Bodily erotic and non-bodily, emotional picture categories were
clips to induce sexual excitement. Although the long duration ensures matched for standard ratings of arousal, dominance and valence as
induction of sexual feelings, it also allows several emotional states to provided by the IAPS catalogue (Lang et al., 2005). Neutral and
occur within one clip, which makes it rather difficult to efficiently emotional pictures were matched for dominance and mean valence
control for GEA or valence (Janssen et al., 2003). Short presentations ratings, while emotional pictures had higher arousal values. Mean
of still pictures may be matched more easily and more accurately for valence of emotional stimuli was very low for negative and very high
induced emotional states. Furthermore, the bodily content of erotic for positive stimuli while neutral stimuli always had intermediate
pictures has to be manipulated independently of SSI to account for valence levels.
activations related to the processing of human body parts. This would Runs consisted of either non-bodily-emotional and neutral (4 runs)
require the introduction of both sexual-bodily and non-sexual-bodily or non-bodily-emotional and bodily (erotic) stimuli (4 runs) and were
stimuli. Recent investigations demonstrated that human body parts by presented in a pseudorandomized order. Each run consisted of a total
themselves activate regions which are implicated in part in SA of 32 images and had a total length of 8 min 32 s. Stimulus duration
(Downing et al., 2001; Downing et al., 2007; Peelen and Downing, was intended to address two main concerns. While SA in earlier
1484 M. Walter et al. / NeuroImage 40 (2008) 1482–1494
studies was induced either by short movies or pictures of at least 20– sequence 60 slices sagittal, 2.8 mm thickness), (b) T1 weighted
30 s, the shortest duration shown to evoke penile tumescence was 10 s anatomic images coplanar with the functional images (23 slices,
(McConaghy, 1989). Aiming to be especially sensitive to early aligned to the plane connecting the anterior and posterior commissure
components of SA such as cognitive and emotional evaluation, axis covering the whole head in oblique axial orientation), (c) inversion
imaging studies investigating effective emotional stimulation use recovery T1 weighted echo planar images coplanar with the functional
shorter stimulus durations which reliably induce an emotional re- images and (d) echo planar functional images sensitive to BOLD
sponse and support sufficient specificity of the investigated emotional contrast (257 sequential acquisitions, 23 slices with 3.125 mm in-plane
process. Consequently, a presentation duration of 5 s was chosen, resolution, 5 mm thickness, 1 mm gap; T2* weighted gradient echo
which has been shown to reliably invoke subjective experience of SA sequence: TR 2 s, TE 40 ms). The first seven images were discarded
(measured on a 1–9 item scale) and found to be short enough to due to T1 saturation effects. During image acquisition, stimuli were
maintain a certain emotional state (Heinzel et al., 2006). Pictures were projected on a mirror mounted on the headcoil using a LCD projector.
presented in a randomized order across runs and subjects were asked
to promptly press a button at the appearance of each picture to monitor Image analysis
their vigilance in the scanner, as assessed by reaction times.
In order to control for the possible influence of preceding attention Image preprocessing and statistical analyses were carried out using
on the experience of the presented stimuli, half of the pictures were MATLAB 6.5.1 and SPM2 (SPM2 software package; Wellcome
preceded by an expectancy period of 4.5 s indicating the type of the Department of Cognitive Neurology, London, UK; see http://www.fil.
subsequently presented picture. Expected and unexpected pictures ion.ucl.ac.uk/spm2.html). A total of 2000 (8 × 250) volume images
(see below) were balanced for the number of positive and negative were realigned to the first image to correct for head movement
erotic and emotional or neutral pictures, the total number of pictures, between scans, mean-adjusted by proportional scaling, resliced and
and IAPS values. Following an approach described by Bermpohl et al. normalized into standard stereotactic space (resulting in an isotrophic
(2006), the expectancy period was indicated by the presentation of 3 mm resolution). The cut-off for subjects showing translational head
a white arrow on a dark background. Downward arrows indicated movement was set to 2 mm and for rotational movement to 1°. No
sexual, upward arrows emotional and rightward arrows neutral ex- subject had to be excluded. Image normalization was performed using
pectancy. After the picture presentation, a fixation cross (intertrial the MNI (Montreal Neurological Institute) template provided by
interval) was presented for a duration of 8 s. Arrows and fixation cross SPM. Spatial transformation included both linear and non-linear
were of equal size, color and luminance and were centered on a black dimensions and used the non-linear sampling algorithm provided with
background. Prior to the experimental session, subjects were SPM2. Data were thereafter expressed in terms of standard
familiarized with the paradigm by completing a test run consisting stereotactic coordinates in the x, y and z axes. Transformed functional
of 32 trials. data sets from each subject were smoothed with a Gaussian kernel of
8 mm (full-width half-maximum) for the group analysis to meet the
Behavioral and subjective evaluation statistical requirements of the General Linear Model and to
compensate for normal variation in individual brain size, shape and
Reaction times were defined as the time between picture onset and sulcal/gyral anatomy across subjects. Subject-specific low frequency
subsequent button press during the fMRI experiment. Subjective drifts in signals were removed by a high pass filter of 128 s.
ratings of the pictures were conducted outside the scanner after the For each subject, a design matrix modeling of emotional and erotic
fMRI session using a visual analogue scale. This approach was pre- or emotional and neutral IAPS pictures was defined. For the analysis
ferred to an intra scan assessment to avoid judgment related cognitive of effects of preceding expectancy, unexpected and expected viewing
influence which can confound neural activity during emotional of the stimulus categories were modeled as separate events. This
perception (Blair et al., 2007; Grimm et al., 2006; Mitchell et al., 2007; enabled us to compare the main contrasts for stimuli presented in an
Northoff et al., 2006; Taylor et al., 2003). Ratings of experienced SSI, “expected mode” (after expectancy cues) or in an unexpected mode
GEA and valence were acquired. Subjects were asked to rate their (after fixation).
own personal experience instead of cognitively evaluating the erotic Besides these experimental conditions, regressors for the baseline
and emotional properties of the stimuli. During analysis of behavioral condition (fixation cross) and expectancy periods were entered into a
data, average scores of ratings were calculated for the whole group. design matrix including a total of 7 regressors. To investigate effects
For intra-scan reaction times and reports of GEA, SSI and valence, of positive and negative subcategories of erotic and emotional stimuli
analyses of covariance (ANOVA) containing the factors stimulus and to make explicit use of their sexual and emotional properties,
category (bodily or non-bodily-emotional), stimulus subcategory these were modeled separately in a third design matrix on single
(positive or negative) and mode of presentation (unexpected or subject level. Specific effects were tested by applying appropriate
expected) were calculated for main effects and interactions. Gender of linear contrasts to the parameter estimates for each condition, resulting
subject was entered as a between subjects factor. Since many stimuli in a t-statistic for each voxel. Analysis of effects for the whole group of
depicted couples that induce high sexual intensities (Hamann et al., all subjects and for groups of males and females was performed in a
2004), ratings of all stimuli were regarded together and no further second level, random effects model on a whole brain level (Friston
analysis was performed on gender effects on the viewing of opposite et al., 1999).
sex stimuli. We report, if not explicitly mentioned otherwise, differences in
activated regions that survived a threshold of p b 0.001, corrected for
Imaging–scanning procedures multiple comparisons using FDR correction (false discovery rate;
Genovese et al., 2002) for clusters consisting of at least 10 contiguous
Data acquisition was conducted on a 1.5 Tesla General Electric voxels (k = 10) for main contrasts, and a conjoint p-threshold of
Signa scanner using a standard headcoil. Imaging procedures included p b 0.05 (FDR-corrected) for conjunction analyses or serial subtraction
collection of (a) structural high resolution images (rf-spoiled GRASS contrasts (interactions) of more than one contrast at the second level.
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1485
Table 1 analysis (Nichols et al., 2005), with a contrast of the type [S N s] that
Mean ratings (±SD) are reported for all bodily, non-bodily-emotional and compared two stimuli which differed in their ratings of SSI but not in
non-bodily neutral picture categories, as well as for positive and negative GEA or valence. Activations in common areas revealed by this con-
subcategories junction analysis were interpreted as being related to greater sexual
Stimulus category SSI GEA Valence intensity common in both contrasts but not as (common) effects of
Mean SD Mean SD Mean SD greater GEA or valence. Characterization of a stimulus subtype as
more or less emotionally arousing or sexually intense was done
Bodily-emotional 4.06 1.12 6.38 1.31 4.70 0.30
according to the behavioral results and was always relative to the other
Positive 5.93 1.27 6.07 1.5 6.95 0.76
stimulus subtype entered in the contrast (Fig. 4a).
Negative 2.19 1.27 6.70 1.17 2.46 0.76
Non-bodily-emotional 1.16 0.46 5.75 1.32 4.41 0.36 The linear relationship between subjective ratings and neural
Positive 1.25 0.76 5.00 1.56 6.32 0.73 activations was then tested adding subjects’ ratings of SSI, GEA and
Negative 1.07 0.18 6.56 1.23 2.50 0.74 valence for each stimulus as parametric regressors (Lewis et al., 2007;
Neutral 1.12 0.23 2.61 1.03 5.14 0.30 Nagai et al., 2004). This new design matrix consisted of regressors of
Subjects were asked to rate “How sexually intense” (1–9), “How emotionally interest for bodily and non-bodily stimuli which were then modulated
intense” (1–9) and “What kind of emotion” (valence with 1: very negative to 9: by three parametric regressors of first polynomial order for individual
very positive) they experienced the picture. subjective ratings of SSI, GEA and valence. Effects of all parametric
regressors were then tested in a second level random effects model
The strategy of analysis aimed to first reveal several regions (Friston et al., 1999). This analysis aimed to reveal a parametric
involved in SA and then specify their involvement in more specific (linear) modulation by SSI, GEA or valence in regions showing ef-
aspects as SSI or GEA. First, BOLD effects of all positive bodily fects either for SSI, GEA or valence.
pictures greater than those of all positive non-bodily stimuli were Effects of greater GEA within areas activated during sexual
assessed by a linear contrast. This contrast was further performed arousal were tested by a conjunction of [bodily N non-bodily pictures]
separately for expected and unexpected stimuli and contrast images of and [non-bodily N neutral pictures], identifying common activations
both contrasts were compared on a second level, random effects, paired for higher GEA but not SSI. The interaction of specific sexual
t-test. Gender effects in the main contrast were tested on the second intensity and valence was tested by a serial subtraction search for
level, performing a two sample t-test of subjects’ contrast images. higher effects of valence in bodily stimuli than in non-bodily-
To investigate specific neuronal effects of differences in SSI be- emotional stimuli.
tween stimulus categories, specific contrasts were chosen according to Second level analyses of more than one contrast per subject were
effects revealed by behavioral results in a next step. One contrast corrected for non-sphericity. For resulting regions of interest, neural
could compare a stimulus type rated higher for SSI (S = higher SSI), responses were extracted and transformed to percentage signal changes
higher for GEA (E = higher GEA) and higher for valence (P= more for our regressors of interest in each subject using the MarsBar toolbox
pleasant) to a less sexually intense (s = lower SSI), less emotionally (Brett et al., 2002). Temporal patterns were investigated by time course
arousing (e = lower GEA) and less pleasant (p = less pleasant) stimulus analyses applying a finite impulse response model (FIR) which does
([SEP N sep]). This contrast was then entered into a conjunction not make an assumption on the resulting signal changes after stimulus
Fig. 1. Main regions of sexual stimulus perception controlling for emotional stimulus perception. (a) The sagittal brain section shows activity in hypothalamus,
VS, pgACC and DMPFC to be higher in positive bodily (erotic) than positive non-bodily (emotional) pictures at a corrected threshold of p b 0.001 and a voxel
threshold of k N 10. (b) Bar diagrams in the upper panel show percentage signal changes in these regions for all four stimulus conditions (positive (+) and negative
(-) bodily and non-bodily stimuli). The lower panel shows corresponding time courses for each condition for the next 8 acquisition time points (16 seconds) after
stimulus onset. For DMPFC, see also Fig. 5.
1486 M. Walter et al. / NeuroImage 40 (2008) 1482–1494
Table 2
Effects of positive bodily (i.e., erotic) pictures when compared to positive non-bodily (emotional) pictures ( p b 0.001, corrected, k N 10) and when additionally
compared to negative bodily pictures
Region [positive bodily N positive Conjunction: [positive Signal correlation
non-bodily] bodily N positive non-bodily] ^ with sexual intensity
[positive N negative bodily]
(x, y, z) Z-values (x, y, z) Z-values (x, y, z) Z-values
VS – 9, 9, 3; 4.53; 0, 12, – 3; 3.98; – 3, 15, – 3; 4.82;
9, 6, 6; 5.85;
Hypothalamus – 3, 12, – 9; 4.54 – 2, 11, – 11; 3.98;
6, 0, – 12; 4.88;
ACC 0, 39, 6; 4.84; 0, 39, 6; 4.84; – 6, 48, – 9; 6.05;
VMPFC 3, 57, 3; 4.94; 3, 57, 3; 4.90;
DMPFC 6, 51, 30; 4.85;
LOC – 51, – 78, 0; 7.36;
54, – 72, 0; 7.69; 54, – 72; – 6; 3.93;
SPL – 36, – 60, 57; 4.84; 33, – 39, 48; 5.05;
27, – 60, 60; 5.05;
Precuneus 3, – 66, 36; 5.71; 3, – 66, 48; 4.43;
FFA 42, – 45, 24; 6.32;
Ventral tegmentum – 3, – 6, – 9; 4.53;
Amygdala 18; – 3; – 15; 4.43;
The conjunction revealed common regions in both contrasts at p b 0.05 (corrected, k N 10). Correlation of activations with individually reported sexual intensity
for each bodily picture as revealed by parametric analysis is reported at p b 0.001 (corrected, k N 10).
VS: ventral striatum, ACC: anterior cingulate cortex, VMPFC: ventral medial prefrontal cortex, DMPFC: dorsal medial prefrontal cortex, LOC: lateral occipital
complex, SPL: superior parietal lobe, FFA: fusiform face area.
presentation. Parameter estimates were calculated for 8 time bins of in Table 1, positive bodily, i.e., erotic, pictures where rated not only
1TR (=2 s) length for each regressor of our design matrix. higher for specific sexually intensity than positive non-bodily
pictures, but also more pleasant and more emotionally arousing,
while negative bodily and non-bodily pictures only differed in their
Results sexual intensity ratings (see also Fig. 4a). No effects were found for
gender when entered as a between subjects factor in the analysis.
Subjective ratings
1TR (=2 s) length for each regressor of our design matrix. higher for specific sexually intensity than positive non-bodily
pictures, but also more pleasant and more emotionally arousing,
while negative bodily and non-bodily pictures only differed in their
Results sexual intensity ratings (see also Fig. 4a). No effects were found for
gender when entered as a between subjects factor in the analysis.
Subjective ratings
A (2 × 2 × 2) ANOVA for reported SSI revealed significant main Behavioral performance
effects for stimulus category (bodily or non-bodily-emotional) and
stimulus subcategory (positive or negative stimuli) but not for the An ANOVA on reaction times (RT) revealed significant effects for
mode of presentation (expected or unexpected) at p b 0.001. the mode of presentation, with significantly faster RTs for expected
Although balanced by IAPS standard ratings of pleasantness and than for unexpected pictures ( p b 0.001), while no significant effects
arousal, ANOVAs on reported GEA and valence showed significant ( p N 0.05) were found for the factors stimulus category and stimulus
main effects of all three factors at a lower level of p b 0.01. As shown subcategory or subjects’ gender or for any interaction. Effects of general
effects for stimulus category (bodily or non-bodily-emotional) and
stimulus subcategory (positive or negative stimuli) but not for the An ANOVA on reaction times (RT) revealed significant effects for
mode of presentation (expected or unexpected) at p b 0.001. the mode of presentation, with significantly faster RTs for expected
Although balanced by IAPS standard ratings of pleasantness and than for unexpected pictures ( p b 0.001), while no significant effects
arousal, ANOVAs on reported GEA and valence showed significant ( p N 0.05) were found for the factors stimulus category and stimulus
main effects of all three factors at a lower level of p b 0.01. As shown subcategory or subjects’ gender or for any interaction. Effects of general
Fig. 2. Specific effects of positive bodily picture viewing. Regions revealed by the conjunction of the contras [positive bodily N positive non-bodily pictures] and
[positive bodily N negative bodily pictures] at a corrected threshold of p b 0.05, k N 10. Common activations, reflecting common effects of positive erotic
experience controlled for the general display of naked people, were found in pgACC, VS, anterior hypothalamus and precuneus (see also Table 2).
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1487
attentiveness for the main picture conditions (positive and negative stimuli with and without preceding expectancy. The above mentioned
bodily and non-bodily-emotional pictures) could thus be ruled out. target regions of sexual arousal (VS, Hypothalamus, pgACC) did not
show stronger effects in the expected modes when compared to the
fMRI data unexpected mode ( p N 0.05, corrected, for detailed analyses of
expectancy effects in other regions; see Walter et al., in press).
Distinction between sexual and emotional processing Finally, none of these regions showed a significant gender dif-
ference ( p N 0.05 corrected) for neuronal effects in the main contrasts
Categorical approach. Positive bodily, i.e., erotic, pictures in- [bodily N non-bodily-emotional stimuli] or [positive bodily N positive
duced higher activity than positive non-bodily-emotional pictures in non-bodily stimuli]. On an exploratory level, at a lower threshold of
the ventral striatum (VS) and hypothalamus as well as in the peri- p b 0.05 uncorrected, neither the amygdala nor the hypothalamus
genual anterior cingulate (pgACC), dorsal (DMPFC) and ventral showed differential activation between males and females.
(VMPFC) medial prefrontal cortex, precuneus, left and right occipital
cortex (LOC) including the right fusiform face area (FFA) and Behaviorally driven approach. Analysis of parametric modulation
bilateral superior parietal lobes (SPL) ( p b 0.001, corrected, k N 10; of regional signal intensities during bodily picture viewing by sub-
see Fig. 1 and Table 2 for z-values and peak coordinates as well as for jective ratings of specific sexual intensity identified the pgACC, the
regions significant at a corrected p b 0.05, k N 10). The contrast be- VS, extending into the medial preoptic area of the anterior hypo-
tween positive and negative bodily stimuli yielded similar regions and thalamus (MPOA) and further into the right SPL ( p b 0.001 corrected,
the conjunction of both contrasts, identifying brain regions that are k N 10, Fig. 3 and Table 2). The higher sexually intense a bodily
specific for positive bodily, e.g., erotic, stimuli as distinguished from stimulus was rated by a subject, the higher the signals were detected in
negative bodily and from positive non-bodily stimuli, revealed these regions. In contrast to these regions, we did not observe para-
common activations in the VS and the Hypothalamus as well as in metric modulation in the posterior hypothalamus by SSI. Further-
pgACC, VMPFC, precuneus and right LOC ( p b 0.05, corrected, more, we did not observe any correlating regions for emotional
k N 10; Fig. 2 and Table 2). valence or GEA at this level of statistical significance ( p b 0.001,
To exclude effects of preceding attention, we separately tested the corrected). While at a lower level of p b 0.05, corrected, we observed
main contrast [positive bodily N positive emotional pictures] for parametric modulation of signal intensities in the pgACC by emo-
Fig. 3. Correlation of neural responses and subjective ratings. (a) Brain sections show effects of bodily stimuli in the VS and the pgACC, extending into the VMPFC,
that were found to be positively modulated by reported ratings of sexual intensity when entered as parametric regressors ( p b 0.001, corrected, k N 10). The lower panel
in (b) shows the parametric modulation of responses for one representative subject to illustrate this finding. The modeled changes in hemodynamic responses function
are therefore plotted as a function of peristimulus time (PST, stimulus onset at t = 0) and individually reported sexual intensity. The anterior cingulate responded with
strongest signal decreases for bodily stimuli with low sexual intensity while highly sexually intense stimuli were correlated with near resting state activity. In the VS in
contrast, stimuli with highest sexual intensity ratings elicited highest signal increases (compare also Figs. 1 and 4).
1488 M. Walter et al. / NeuroImage 40 (2008) 1482–1494
Fig. 4. Conjunction analysis on the basis of intensity ratings. (a) Comparison of subjective effects: Bar diagrams plot mean ratings for SSI, GEA and valence for
all four stimulus conditions. Both positive (+) and negative (-) bodily picture sets were rated more sexually intense than respective non-bodily pictures sets but
only positive bodily pictures were also rated more pleasant (higher valence) and more emotionally arousing than positive non-bodily-emotional pictures. For
exact values, compare Table 1 and behavioral results. The comparisons between stimulus categories are color coded, and the same colors indicate the comparison
of BOLD responses between these stimulus categories. Underlying differences in subjective ratings for the conjoined contrasts below are indicated by red and
blue color bars and indexed for the two contrasts c1 and c2 that compare the same stimuli for (a) and (b). (b) Comparison of neural effects: The commonality of
higher SSI in behavioral ratings was reflected by the conjunction of both functional stimulus contrasts, looking for common differential activations at a corrected
p b 0.05, k N 10. Common differences in activations, which can be traced back to differences in SSI but not in GEA or valence, were found in the VS,
hypothalamus, DMPFC and precuneus. Differences in GEA or valence were not present in conditions in either contrast and are thus not represented by this
conjunction.
tional valence, but no modulation by GEA. Similar patterns were positive and negative) compared to non-bodily-emotional (i.e.,
found for males and females when analyzed separately (Supplemen- positive and negative) picture conditions and a higher activity in
tary Fig. 1). non-bodily-emotional compared to neutral conditions were re-
Control for emotional valence and intensity in erotic stimuli. In
view of our behavioral data, we conducted a further conjunction Table 3
analysis where we conjoined the contrast [positive bodily N positive Effects specific for differences in sexual intensity (p b 0.05, corrected, k N 10)
non-bodily pictures] with the one [negative bodily N negative non- Region Conjunction: [positive bodily N positive
bodily pictures]. As can be seen in our behavioral data, both non-bodily] ^ [negative bodily N negative
contrasts showed significant effects of SSI whereas this was not the non-bodily]
case for valence or GEA, which did not differ in negative pictures Coordinates (x, y, z) Z-values
(Fig. 4a). Resulting common regions thus mirror the specific
effects of sexual intensity so that we were able to specifically VS – 9, 9, 6; 3.44;
9, 9, 3; 4.64;
control for effects of emotional valence and GEA in positive
Hypothalamus – 9, 3, – 13; 3.73;
bodily, e.g., erotic, stimuli. This analysis revealed foci in the DMPFC – 6, 51, 27; 3.93;
hypothalamus and VS as well as in the DMPFC, precuneus, SPL, LOC – 51, – 75, – 6; 5.33;
FFA and LOC ( p b 0.05, corrected; Fig. 4b and Table 3). 54, – 66, 0; 5.03;
SPL – 33, – 57, 54; 4.79;
Common and interacting regions between sexual and general 33, – 57, 57; 4.86;
emotional processing Precuneus 9, – 63, 39; 4.21;
FFA 45, – 42, – 24; 3.81;
Common regions in sexual and general emotional processing. VS: ventral striatum, DMPFC: dorsomedial prefrontal cortex, LOC: lateral
Regions that showed higher activity in bodily-emotional (i.e., occipital complex, SPL: superior parietal lobe, FFA: fusiform face area.
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1489
view of our behavioral data, we conducted a further conjunction Table 3
analysis where we conjoined the contrast [positive bodily N positive Effects specific for differences in sexual intensity (p b 0.05, corrected, k N 10)
non-bodily pictures] with the one [negative bodily N negative non- Region Conjunction: [positive bodily N positive
bodily pictures]. As can be seen in our behavioral data, both non-bodily] ^ [negative bodily N negative
contrasts showed significant effects of SSI whereas this was not the non-bodily]
case for valence or GEA, which did not differ in negative pictures Coordinates (x, y, z) Z-values
(Fig. 4a). Resulting common regions thus mirror the specific
effects of sexual intensity so that we were able to specifically VS – 9, 9, 6; 3.44;
9, 9, 3; 4.64;
control for effects of emotional valence and GEA in positive
Hypothalamus – 9, 3, – 13; 3.73;
bodily, e.g., erotic, stimuli. This analysis revealed foci in the DMPFC – 6, 51, 27; 3.93;
hypothalamus and VS as well as in the DMPFC, precuneus, SPL, LOC – 51, – 75, – 6; 5.33;
FFA and LOC ( p b 0.05, corrected; Fig. 4b and Table 3). 54, – 66, 0; 5.03;
SPL – 33, – 57, 54; 4.79;
Common and interacting regions between sexual and general 33, – 57, 57; 4.86;
emotional processing Precuneus 9, – 63, 39; 4.21;
FFA 45, – 42, – 24; 3.81;
Common regions in sexual and general emotional processing. VS: ventral striatum, DMPFC: dorsomedial prefrontal cortex, LOC: lateral
Regions that showed higher activity in bodily-emotional (i.e., occipital complex, SPL: superior parietal lobe, FFA: fusiform face area.
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1489
Fig. 5. Common and interacting regions in sexual and emotional processing. (a) Red voxels indicate regions with a significant interaction in processing of sexual
arousal and valence, as revealed by the contrast [positive N negative bodily-emotional pictures] N [positive N negative non-bodily-emotional pictures] at p b 0.05 ,
corrected, k N 10. Significantly stronger effects in the first contrast were found in the pgACC and bilateral occipital cortex (LOC) (see also Table 4 and bar
diagrams in Figs. 1 and 4 for comparison of neural and behavioral effects in pgACC). Blue voxels indicate significant regions for the conjunction of the contrasts
[bodily N non-bodily-emotional pictures] and [non-bodily-emotional N non-bodily neutral pictures] at p b 0.05, corrected, k N 10. In resulting regions, including
amygdala, DMPFC, LOC, tectum and thalamus, activity during non-bodily emotion processing was found to be lower than during processing of bodily stimuli
but higher than during neutral picture presentation, reflecting rather general effects of emotional intensity. (b) Bar diagrams plot mean percentage signal changes
for bodily, non-bodily-emotional and neutral conditions in these common regions.
vealed by the contrast [all bodily-emotional N all non-bodily- and Table 4 as well as % signal changes in Fig. 1 for the
emotional pictures] conjoined with the contrast [all non-bodily- pgACC).
emotional stimuli N all neutral stimuli]. Significant commonalities
indicating common effects of GEA were found for the DMPFC,
bilateral amygdala, right medial thalamus, dorsal midbrain and Discussion
bilateral FFA and LOC ( p b 0.05 corrected, k N 10, see the blue
colored voxels in Fig. 5 and Table 4). To demonstrate signal The present study aimed to investigate the neural correlates of
increases and decreases, we show mean percentage signal changes specific sexual intensity independent of confounding variables like
for all conditions in DMPFC, amygdala, right medial thalamus and involvement of general emotional dimensions and bodies. This was
dorsal midbrain. achieved by carefully controlling for these factors in both design and
analysis. We replicated patterns of activations in a large set of
regions previously reported during SA, and we disentangled
Interaction between sexual and general emotional processing neuronal mechanisms mediating SSI from activity related to
Regions showing a higher difference between positive and processes of GEA or valence. Neuronal effects in the VS and
negative stimuli in the bodily-emotional conditions when hypothalamus were independent of differences in emotional valence
compared to the non-bodily-emotional conditions, with positive or GEA, while activation in the pgACC depended on an interaction
stimuli yielding higher signal intensities than negative stimuli, of SSI and valence.
were tested with the serial subtraction [positive N negative bodily This distinction was further supported by a correlation between
stimuli] N [positiveNnegative non-bodily stimuli]. Significant dif- neural activity in the VS and MPOA and SSI ratings alone, and a
ferences were found in the pgACC (extending into the VMPFC) correlation between activity in the pgACC and SSI as well as valence
and LOC ( p N 0.05 corrected, see red colored voxels in Fig. 5a ratings. The interaction between SSI and emotional valence was
1490 M. Walter et al. / NeuroImage 40 (2008) 1482–1494
Table 4 blocked designs (Mouras, 2006). However, as we did not directly
Common regions for bodily and non-bodily-emotional picture viewing compare both long and short stimulus types, this remains to be tested
( p b 0.05, corrected, k N 10) by further studies.
Region Conjunction: [bodily-emotional N The relation of hypothalamic activation to reported specific sexual
non-bodily-emotional] ^ intensity was also suggested by Karama et al. (2002). They further
[non-bodily-emotional N non-bodily found this behavioral measure to sufficiently account for neural
neutral] differences in activations between males and females, when the
Coordinates (x, y, z) Z-values female group of their study reported erotic stimuli being less sexually
intense. We could confirm this finding as, even on a considerably
Amygdala – 15, 3, – 18; 3.72;
18, 0, – 18; 4.00; liberal statistical threshold, we did not find different hypothalamic
Thalamus 9, 0, 0; 4.48; activations when no differences of reported SSI existed between
Dorsal midbrain 3, – 24, 0; 4.00; genders.
LOC – 48, – 81, 0; 6.88; Considering the rewarding nature of sexual incentives, our
51, – 69, 0; 6.17; findings, especially in the ventral parts of striatum, may be interpreted
FFA 42, – 45, – 24; 4.78; as mirroring implicit reward processing. The nucleus accumbens has
Inferior frontal gyrus 39, 33, – 9; 4.08; been found to be essential during reward processing (Knutson and
DMPFC – 6, 54, 30; 3.54; Wimmer, 2007; Wrase et al., 2007; Yacubian et al., 2007). In contrast
LOC: lateral occipital cortex, FFA: fusiform face area, DMPFC: to other regions like the anterior insula, activations in the VS, i.e.,
dorsomedial prefrontal cortex. nucleus accumbens, were found to be involved in reward processing
irrespective of its emotional value (Jensen et al., 2007). Erotic stimuli
could therefore present a highly salient incentive in their own right.
restricted to the pgACC and LOC, while a main effect of GEA This is supported by our finding of a strong, isolated correlation of
independent of SSI was found for the DMPFC, amygdala, medial signal intensities with sexual intensity as rated by our subjects.
thalamus and midbrain. Our findings support earlier models of SA Ventral striatal activity is strongly influenced by dopaminergic
favoring involvement of certain brain regions in distinct components transmission. While antidopaminergic effects of antipsychotics result
of sexual arousal. in reduced libido and erectile impairment in males, the opposite effect
can be observed during pharmacological administration of dopamine
Specific involvement of hypothalamus and VS agonists (see also Arnow et al., 2002). Dopaminergic inputs originate
mainly in the ventral tegmental area (VTA), which was recently found
Recent imaging findings of hypothalamic activity during SA and to be involved in the mediation of expectancy of novel visual or
particularly during penile erection led researchers to discuss its somatosensory stimuli (Fairhurst et al., 2007; Wittmann et al., 2007).
involvement especially in autonomous aspects of behavioral This finding suggests an attentional modulation by stimulus salience
responses to erotic stimuli (Karama et al., 2002; Mouras, 2004; dependent on the context of presentation. However, in our target
Redoute et al., 2000). However, given its fundamental role in regions of sexual intensity, namely the hypothalamus and VS, no
mediating all kinds of autonomous, i.e., hormonal responses, it could differences were observed for effects of SSI, when relevant contrasts
not be excluded that hypothalamic activation during SA reflected a of expected and unexpected picture conditions were compared. It
rather more general emotional arousal. should also be noted that the reported regions receive also non-
Our findings demonstrate for the first time that the hypothalamus dopaminergic inputs.
is specifically involved in processing sexual intensity in humans Both the hypothalamus and VS thus have been previously
independent of general emotional dimensions or bodies. This hypothesized to be key players of specific erotic processing as
complements recent behavioral and neurohormonal findings. The opposed to mediating structures through elevated but unspecific
hypothalamus has been shown to be specifically involved in emotional arousal that normally accompanies sexually arousing
copulatory behavior and erection in many animals (McKenna, situations. To our knowledge, this study is the first to demonstrate
1999; Meisel, 1983; Meisel and Mullins, 2006). In humans, this specificity of neural activity on the basis of behavioral and
paraventricular hypothalamic secretion of oxytocin was reported to subjective data. This was accomplished by carefully matching for
be increased during SA in males and females (Carmichael et al., 1987, effects of emotions and body during sexual arousal.
1994). In a recent study, Ferretti et al. (2005) found highest hy- Still, our findings should not be interpreted as suggesting that
pothalamic responses during the onset of penile erection in males, activity in the VTA-VS continuum would be restricted to sexual
suggesting its particular involvement in triggering a behavioral processes. Instead, it should be pointed out that other affective
response to a condition that is specifically sexually intense. In contrast processes handling, e.g., rewarding or stressful, situations have been
to other regions like the insula and somatosensory cortex, hy- shown to activate these regions. The framework of our study only
pothalamic activity was not elevated during sustained erection when allows to relate activations during sexually arousing conditions to SSI
compared to phases were no erection was recorded (Ferretti et al., and not to unspecific emotional dimensions such as GEA or valence.
2005). Georgiadis and Holstege (2005) related hypothalamic activity However, these dimensions describe aspects that are conceptualized
to a specific neural process of sexual intensity rather than to actual in cognitive-attributional emotion theories while SSI would share
sexual performance. This involvement of the hypothalamus in the more characteristics with primary emotive systems such as the
initial phase of sexual arousal (Ferretti et al., 2005; Moulier et al., expectancy command system (Panksepp, 1982) or the behavioral
2006), as opposed to the course of sexual performance, would favor its activation system (Gray, 1987). In addition to the distinct effects of
detection in designs using short-term visual stimulation. This was the sexual intensity in the VS and hypothalamus, we found other cortical
case in our study, while other studies focusing on longer periods of areas that mediate either general emotional arousal within SA or an
sexual arousal, did not find reliable hypothalamic activations using interaction of SSI and valence. In a more speculative way, our findings
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1491
might allow basic emotion views to primarily prevail in some stimuli and disappeared when SSI was controlled for differences in
subcortical regions while more general dimensional views may better GEA or valence. Consistent with earlier findings (Canli et al., 2002;
reflect the function of higher cognitive systems and related cortical Hamann and Mao, 2002), amygdala activations could be related to
networks. GEA which normally occurs in sexually arousing situations (Canli
and Gabrieli, 2004). This would be supported by the finding of
Interaction of sexual and emotional processes in the ACC Hamann et al. that pictures of opposite-sex nudes did not lead to
different amygdala responses, in contrast to pictures of sexual inter-
In contrast to the above regions, an interaction of specific sexual actions of couples, which were reported to be also more physically
intensity and its hedonic value can be proposed for the pgACC. On the arousing (Hamann et al., 2004).
one hand, the pgACC showed the strongest parametric modulation of In the same vein, the medial thalamus could be related to GEA
neural activity by reports of sexual arousal. On the other, isolated during SA as suggested by Karama et al. (2002). In their study, the
sexual intensity alone could not promote a higher neural activity in our authors compared explicitly erotic film excerpts to neutral ones, the
conjunctional analysis. Further parametric effects of valence were latter lacking the potential to induce emotional or erotic reactions.
found in the pgACC when the statistical threshold was lowered to They found activations in the mediodorsal thalamic nucleus. For a
p b 0.05, corrected. The rostral anterior part of the ACC, mainly comparable peak voxel localization, we did not only replicate their
covering BA 24, receives specific projections from the shell of the finding, but further confirmed their interpretation of this region as
nucleus accumbens in primates (Devinsky et al., 1995; Kunishio and being especially related to general emotional processes during SA.
Haber, 1994). This region is considered the “affective” subdivision This interpretation is supported by anatomical interconnections of
(Bush et al., 2000; Vogt et al., 1992). Support for this view comes from the medial and midline nuclei in corticolimbic loops with the MPFC
a large set of imaging studies (George et al., 1995; Lane et al., 1997; (Siwek and Pandya, 1991), ventral pallidum (Haber et al., 1985) and
Phan et al., 2002), as well as projections from the accumbens to the amygdala (Aggleton and Mishkin, 1984) in human and non-human
medial thalamus and to limbic as well as paralimbic areas (Devinsky primates. All these structures could also be related to GEA during
et al., 1995; Russchen et al., 1985; Vogt, 2005). SA. These interconnections further include the dorsal midbrain
We consider the pgACC to be a key structure in mediating (Velayos and Reinoso-Suarez, 1982) which was also related to GEA
previously reported interactions of emotional valence and sexual by the same conjunction analysis (Fig. 5).
arousal (Bancroft et al., 2003; Lykins et al., 2006; Peterson and Higher signal intensities in the DMPFC during erotic as compared
Janssen, 2007). The pgACC was found to be the only region with to non-erotic emotional situations could in the same way be
overlapping correlations of reported sexual intensity and emotional interpreted as representing mainly higher GEA. Contrary to this
valence with neural responses, supporting its role in maintaining an interpretation, activations were also observed when no difference in
emotional tone in response to internal or external stimuli and emotional valence or GEAwas found (Table 3). It should be noted that
generating adapted behavior (Bechara and Naqvi, 2004; Ferretti et al., this conjunction of contrasts comparing positive and negative bodily
2005). Activity in the pgACC was recently further related to personal picture conditions relative to non-erotic conditions of positive or
relevance of stimuli (Phan et al., 2004, 2002) or more explicitly to negative valence (Fig. 4) could not be controlled for display of naked
self-relatedness (Frith and Frith, 1999; Northoff and Bermpohl, humans. Furthermore, sexual interaction of couples has to be regarded
2004). Heinzel et al. (2006) report the ventral part of the medial as a highly specific social constellation. This has to be taken into
prefrontal cortex including the pgACC to be specifically related to account as pictures of persons and social interactions have been
subjective reports of self relatedness during visual presentation of reported to elicit strong effects in the DMPFC (Amodio and Frith,
bodily erotic stimuli. 2006; Iacoboni et al., 2004; Mitchell et al., 2002, 2005). This
interpretation would be supported by the absence of effects in the
Effects of general emotional processing DMPFC when the comparison between positive and negative bodily
stimuli was entered into the conjunction (Fig. 2 and Table 2).
Considering that sexually arousing stimuli are also emotionally Combining findings from these two conjunctions, we suggest that
arousing, one may expect additional effects in regions which have DMPFC activations are related to higher order processing of socially
been previously found to be activated during general emotional relevant stimuli, which could also be of particularly high self-
processing. relevance. This last notion would be supported by studies reporting
Extensive research revealed the human amygdala to be a core effects in the DMPFC for either explicit self relatedness (Phan et al.,
region of emotional processing. However studies of SA have been less 2004) or implicit self relatedness during emotional conditions
consistent in reported effects, including several positive (Ferretti et al., (Heinzel et al., 2006). This early effect of still erotic pictures might,
2005; Hamann et al., 2004; Karama et al., 2002) and negative findings however, be substantially different from DMPFC activations which
(Arnow et al., 2002; Bocher et al., 2001; Moulier et al., 2006; Stoleru were specifically correlated with penile tumescence in a later phase of
et al., 1999) as well as evidence of deactivations (Georgiadis and SA (Moulier et al., 2006).
Holstege, 2005; Holstege et al., 2003). Inconsistencies have In both DMPFC and ACC, effects were found to modulate the
previously been assumed to be related to either low magnitudes of extent of deactivations. This is consistent with previous studies that
BOLD responses (Moulier et al., 2006), late deactivations common find robust but unspecific signal decreases in the medial prefrontal
for consummatory phase of erection (Hamann et al., 2004) or even cortex (including the rostral ACC) during a large variety of tasks,
contrary effects of vigilance and sexual responses (Georgiadis and while showing highest activity at rest (Gusnard and Raichle, 2001).
Holstege, 2005; Georgiadis et al., 2006). Due to the specific setup of These negative BOLD responses have previously been found to
our experiment, we cannot contribute to the discussion of amygdala correlate with local concentrations of the inhibitory transmitter
responses during later phases of SA. However, in the early phase, we GABA (Northoff et al., 2007). Consistent with recent findings
found robust activations. These were unspecific in that they also (Heinzel et al., 2006), the degree of self relatedness of an external
appeared in non-erotic control conditions when compared to neutral stimulus would thus be correlated with the amount of deactivation,
1492 M. Walter et al. / NeuroImage 40 (2008) 1482–1494
with highest self-related stimuli leaving activations closest to found for the rostral anterior cingulate, while activation during SA
baseline (Northoff et al., in press). in other regions could be related to non-specific components of SA
Finally, emotional stimuli also included non-human stimuli and including GEA or more complex features of emotional scenes. Due
therefore also depicted fewer human faces than the bodily stimuli. At to our methodological design using subjective ratings to guide
the same time, faces in neutral pictures showed low emotional neural analysis, we were able, for the first time, to neurally dif-
expressions. While the present study focused on the distinction of ferentiate distinct effects of SSI from those related to emotional
processes related to SSI or GEA, we have to assume that FFA ac- dimensions or body parts that are inherently involved in sexual
tivations during SA may be related to an interaction of face perception arousal.
and emotions. There is increasing evidence that this interaction might
be mediated by connections between the amygdala and visual areas, Acknowledgments
the FFA in particular (see also Vuilleumier, 2005). However, due to
our design, we cannot investigate such processes. Selected IAPS We would like to thank Dr. G. Makropoulos for his help in pre-
pictures included different views on human faces, and subjects were paring the revised manuscript as well as two anonymous reviewers for
further free to orient their spatial attention on the whole scene. Thus, their important suggestions.
face perception as well as possible modulating effects of emotions on
face perception cannot be ruled out or disentangled from other
motivational aspects attributed to the FFA such as internally generated Appendix A. Supplementary data
motivation like hunger or thirst (LaBar et al., 2001; Parsons et al.,
2000). Still, according to our results, FFA activations during SA were Supplementary data associated with this article can be found, in
not observed when controlling for confounding effects of body or the online version, at doi:10.1016/j.neuroimage.2008.01.040.
emotion (Table 2 and Fig. 4).
At least an interaction of SSI, emotional dimensions and human References
bodies have to be taken into account for a small cluster of 38 voxels in
the right LOC. Activation here remained in both conjunctions which
should rule out confounding effects of bodies, GEA, or valence. Aggleton, J.P., Mishkin, M., 1984. Projections of the amygdala to the
thalamus in the cynomolgus monkey. J. Comp. Neurol. 222 (1), 56–68.
However, in contrast to the pgACC, no correlation was found for
Amodio, D.M., Frith, C.D., 2006. Meeting of minds: the medial frontal
reported GEA, SSI or valence. Furthermore, the interaction analysis of
cortex and social cognition. Nat. Rev. Neurosci. 7 (4), 268–277.
SSI and valence as well as the conjunction of common effects of GEA Arnow, B.A., Desmond, J.E., Banner, L.L., Glover, G.H., Solomon, A.,
yielded significant effects in LOC. Bearing in mind the complexity of Polan, M.L., Lue, T.F., Atlas, S.W., 2002. Brain activation and sexual
such an interaction, this finding should be discussed in terms of an arousal in healthy, heterosexual males. Brain 125 (Pt 5), 1014–1023.
“extrastriate body area” (Downing et al., 2001) and reported at- Bancroft, J., Janssen, E., Strong, D., Carnes, L., Vukadinovic, Z., Long, J.S.,
tentional top-down modulation of visual association areas by 2003. The relation between mood and sexuality in heterosexual men.
emotional relevance (Corbetta et al., 1993; Mouras et al., 2003; Arch. Sex Behav. 32 (3), 217–230.
Stoleru et al., 1999). This emphasizes the distinction of sexual-bodily Bechara, A., Naqvi, N., 2004. Listening to your heart: interoceptive
and non-sexual-bodily processes in the brain, with mere perception of awareness as a gateway to feeling. Nat. Neurosci. 7 (2), 102–103.
Bermpohl, F., Pascual-Leone, A., Amedi, A., Merabet, L.B., Fregni, F.,
bodies per se leading to activation rather in visual association cortex
Gaab, N., Alsop, D., Schlaug, G., Northoff, G., 2006. Attentional
than in primary “sex regions”.
modulation of emotional stimulus processing: an fMRI study using
It should, however, be noted that in our study we cannot make any emotional expectancy. Hum. Brain Mapp. 27 (8), 662–677.
inference about a specific relation of any of these regions to other Blair, K.S., Smith, B.W., Mitchell, D.G., Morton, J., Vythilingam, M.,
aspects of sexual arousal such as penile tumescence in males or Pessoa, L., Fridberg, D., Zametkin, A., Sturman, D., Nelson, E.E., et al.,
comparable genital responses in females. To allow subjects to 2007. Modulation of emotion by cognition and cognition by emotion.
experience erotic and non-erotic stimuli as undisturbed as possible, Neuroimage 35 (1), 430–440.
we did not obtain measures of erectile responses. In contrast, we Bocher, M., Chisin, R., Parag, Y., Freedman, N., Meir Weil, Y., Lester, H.,
concentrated on an early phase, which did not result in full erection but Mishani, E., Bonne, O., 2001. Cerebral activation associated with sexual
could be matched to GEA or valence more easily. Furthermore, by arousal in response to a pornographic clip: A 15O-H2O PET study in
heterosexual men. Neuroimage 14 (1 Pt 1), 105–117.
focusing on this early component of SA, we revealed global mech-
Brett, M., Anton, J.L., Valabregue, R., Poline, J.B., 2002. Region of interest
anisms found in both genders. Although we did not find differences in
analysis using an SPM toolbox. Conference abstract—8th International
either subjective ratings or neural responses, we acknowledge that this Conference on Functional Mapping of the Human Brain, Sendai, Japan,
observation has to be limited to this early phase of SA. Possible supplement on CD-ROM. Neuroimage 16 (2).
gender-specific differences in responses during the course of ongoing Bush, G., Luu, P., Posner, M.I., 2000. Cognitive and emotional influences in
sexual arousal, including orgasm, could therefore not be covered by anterior cingulate cortex. Trends Cogn. Sci. 4 (6), 215–222.
our study. Canli, T., Gabrieli, J.D., 2004. Imaging gender differences in sexual arousal.
Nat. Neurosci. 7 (4), 325–326.
Conclusion Canli, T., Sivers, H., Whitfield, S.L., Gotlib, I.H., Gabrieli, J.D., 2002.
Amygdala response to happy faces as a function of extraversion. Science
296 (5576), 2191.
Focusing on subjective erotic experience in males and females,
Carmichael, M.S., Humbert, R., Dixen, J., Palmisano, G., Greenleaf, W.,
we compared patterns of differential activation between several Davidson, J.M., 1987. Plasma oxytocin increases in the human sexual
regions previously related to a brain network of sexual arousal. response. J. Clin. Endocrinol. Metab. 64 (1), 27–31.
Using stimuli of short duration, we could relate activations in the Carmichael, M.S., Warburton, V.L., Dixen, J., Davidson, J.M., 1994.
ventral striatum and hypothalamus to SSI independent of induced Relationships among cardiovascular, muscular, and oxytocin responses
GEA or valence. An interaction of SSI and emotional valence was during human sexual activity. Arch. Sex Behav. 23 (1), 59–79.
M. Walter et al. / NeuroImage 40 (2008) 1482–1494 1493
Corbetta, M., Miezin, F.M., Shulman, G.L., Petersen, S.E., 1993. A PET Ishai, A., Pessoa, L., Bikle, P.C., Ungerleider, L.G., 2004. Repetition suppression
study of visuospatial attention. J. Neurosci. 13 (3), 1202–1226. of faces is modulated by emotion. Proc. Natl. Acad. Sci. U. S. A. 101 (26),
Desimone, R., Duncan, J., 1995. Neural mechanisms of selective visual 9827–9832.
attention. Annu. Rev. Neurosci. 18, 193–222. Janssen, E., Carpenter, D., Graham, C.A., 2003. Selecting films for sex
Devinsky, O., Morrell, M.J., Vogt, B.A., 1995. Contributions of anterior research: gender differences in erotic film preference. Arch. Sex Behav.
cingulate cortex to behaviour. Brain 118 (Pt 1), 279–306. 32 (3), 243–251.
Downing, P.E., Jiang, Y., Shuman, M., Kanwisher, N., 2001. A cortical area Jensen, J., Smith, A.J., Willeit, M., Crawley, A.P., Mikulis, D.J., Vitcu, I.,
selective for visual processing of the human body. Science 293 (5539), Kapur, S., 2007. Separate brain regions code for salience vs. valence
2470–2473. during reward prediction in humans. Hum. Brain Mapp. 28 (4), 294–302.
Downing, P.E., Wiggett, A.J., Peelen, M.V., 2007. Functional magnetic Karama, S., Lecours, A.R., Leroux, J.M., Bourgouin, P., Beaudoin, G.,
resonance imaging investigation of overlapping lateral occipitotemporal Joubert, S., Beauregard, M., 2002. Areas of brain activation in males and
activations using multi-voxel pattern analysis. J. Neurosci. 27 (1), 226–233. females during viewing of erotic film excerpts. Hum. Brain Mapp. 16 (1),
Fairhurst, M., Wiech, K., Dunckley, P., Tracey, I., 2007. Anticipatory 1–13.
brainstem activity predicts neural processing of pain in humans. Pain Knutson, B., Wimmer, G.E., 2007. Splitting the difference: how does the
128 (1–2), 101–110. brain code reward episodes? Ann. N. Y. Acad. Sci. 1104, 54–69.
Ferretti, A., Caulo, M., Del Gratta, C., Di Matteo, R., Merla, A., Montorsi, Kunishio, K., Haber, S.N., 1994. Primate cingulostriatal projection: limbic
F., Pizzella, V., Pompa, P., Rigatti, P., Rossini, P.M., et al., 2005. striatal versus sensorimotor striatal input. J. Comp. Neurol. 350 (3),
Dynamics of male sexual arousal: distinct components of brain 337–356.
activation revealed by fMRI. Neuroimage 26 (4), 1086–1096. LaBar, K.S., Gitelman, D.R., Parrish, T.B., Kim, Y.H., Nobre, A.C.,
Fisher, H.E., Aron, A., Brown, L.L., 2006. Romantic love: a mammalian Mesulam, M.M., 2001. Hunger selectively modulates corticolimbic
brain system for mate choice. Philos. Trans. R. Soc. Lond., B Biol. Sci. activation to food stimuli in humans. Behav. Neurosci. 115 (2), 493–500.
361 (1476), 2173–2186. Lane, R.D., Reiman, E.M., Bradley, M.M., Lang, P.J., Ahern, G.L., Davidson,
Friston, K.J., Holmes, A.P., Price, C.J., Buchel, C., Worsley, K.J., 1999. R.J., Schwartz, G.E., 1997. Neuroanatomical correlates of pleasant and
Multisubject fMRI studies and conjunction analyses. Neuroimage 10 (4), unpleasant emotion. Neuropsychologia 35 (11), 1437–1444.
385–396. Lane, R.D., Reiman, E.M., Axelrod, B., Yun, L.S., Holmes, A., Schwartz, G.E.,
Frith, C.D., Frith, U., 1999. Interacting minds—a biological basis. Science 1998. Neural correlates of levels of emotional awareness. Evidence of an
286 (5445), 1692–1695. interaction between emotion and attention in the anterior cingulate cortex.
Genovese, C.R., Lazar, N.A., Nichols, T., 2002. Thresholding of statistical J. Cogn. Neurosci. 10 (4), 525–535.
maps in functional neuroimaging using the false discovery rate. Neuro- Lang, P.J., Bradley, M.M., Fitzsimmons, J.R., Cuthbert, B.N., Scott, J.D.,
image 15 (4), 870–878. Moulder, B., Nangia, V., 1998. Emotional arousal and activation of the
George, M.S., Ketter, T.A., Parekh, P.I., Horwitz, B., Herscovitch, P., Post, visual cortex: an fMRI analysis. Psychophysiology 35 (2), 199–210.
R.M., 1995. Brain activity during transient sadness and happiness in Lang, P.J., Bradley, M.M., Cuthbert, B.N., 2005. International affective
healthy women. Am. J. Psychiatry 152 (3), 341–351. picture system (IAPS): Affective ratings of pictures and instruction
Georgiadis, J.R., Holstege, G., 2005. Human brain activation during sexual manual. Technical Report A-6. University of Florida, Gainesville, FL.
stimulation of the penis. J. Comp. Neurol. 493 (1), 33–38. Lewis, P.A., Critchley, H.D., Rotshtein, P., Dolan, R.J., 2007. Neural
Georgiadis, J.R., Kortekaas, R., Kuipers, R., Nieuwenburg, A., Pruim, J., correlates of processing valence and arousal in affective words. Cereb.
Reinders, A.A., Holstege, G., 2006. Regional cerebral blood flow Cortex 17 (3), 742–748.
changes associated with clitorally induced orgasm in healthy women. Lykins, A.D., Janssen, E., Graham, C.A., 2006. The relationship between
Eur. J. Neurosci. 24 (11), 3305–3316. negative mood and sexuality in heterosexual college woman and men.
Gray, J.A., 1987. The psychology of fear and stress. Cambridge University J. Sex Res. 43 (2), 136–143.
Press, Cambridge. McConaghy, N., 1989. Validity and ethics of penile circumference measures
Grimm, S., Schmidt, C.F., Bermpohl, F., Heinzel, A., Dahlem, Y., Wyss, M., of sexual arousal: a critical review. Arch. Sex Behav. 18 (4), 357–369.
Hell, D., Boesiger, P., Boeker, H., Northoff, G., 2006. Segregated neural McKenna, K., 1999. The brain is the master organ in sexual function: central
representation of distinct emotion dimensions in the prefrontal cortex—an nervous system control of male and female sexual function. Int. J. Impot.
fMRI study. Neuroimage 30 (1), 325–340. Res. 11 (Suppl 1), S48–S55.
Gusnard, D.A., Raichle, M.E., 2001. Searching for a baseline: functional Meisel, R.L., 1983. Recovery of masculine copulatory behavior from lesions
imaging and the resting human brain. Nat. Rev. Neurosci. 2 (10), 685–694. of the medial preoptic area: effects of age versus hormonal state. Behav.
Haber, S.N., Groenewegen, H.J., Grove, E.A., Nauta, W.J., 1985. Efferent Neurosci. 97 (5), 785–793.
connections of the ventral pallidum: evidence of a dual striato pallidofugal Meisel, R.L., Mullins, A.J., 2006. Sexual experience in female rodents:
pathway. J. Comp. Neurol. 235 (3), 322–335. cellular mechanisms and functional consequences. Brain Res. 1126 (1),
Hamann, S., Mao, H., 2002. Positive and negative emotional verbal stimuli 56–65.
elicit activity in the left amygdala. Neuroreport 13 (1), 15–19. Mitchell, J.P., Heatherton, T.F., Macrae, C.N., 2002. Distinct neural systems
Hamann, S., Herman, R.A., Nolan, C.L., Wallen, K., 2004. Men and women subserve person and object knowledge. Proc. Natl. Acad. Sci. U. S. A.
differ in amygdala response to visual sexual stimuli. Nat. Neurosci. 7 (4), 99 (23), 15238–15243.
411–416. Mitchell, J.P., Neil Macrae, C., Banaji, M.R., 2005. Forming impressions of
Heinzel, A., Walter, M., Schneider, F., Rotte, M., Matthiae, C., Tempelmann, people versus inanimate objects: social-cognitive processing in the medial
C., Heinze, H.J., Bogerts, B., Northoff, G., 2006. Self-related processing prefrontal cortex. Neuroimage 26 (1), 251–257.
in the sexual domain: parametric event-related fMRI study reveals neural Mitchell, D.G., Nakic, M., Fridberg, D., Kamel, N., Pine, D.S., Blair, R.J.,
activity in ventral cortical midline structures. Social Neuroscience 1 (1), 2007. The impact of processing load on emotion. Neuroimage 34 (3),
41–51. 1299–1309.
Holstege, G., Georgiadis, J.R., Paans, A.M., Meiners, L.C., van der Graaf, F.H., Moulier, V., Mouras, H., Pelegrini-Issac, M., Glutron, D., Rouxel, R.,
Reinders, A.A., 2003. Brain activation during human male ejaculation. Grandjean, B., Bittoun, J., Stoleru, S., 2006. Neuroanatomical correlates
J. Neurosci. 23 (27), 9185–9193. of penile erection evoked by photographic stimuli in human males.
Iacoboni, M., Lieberman, M.D., Knowlton, B.J., Molnar-Szakacs, I., Moritz, Neuroimage 33 (2), 689–699.
M., Throop, C.J., Fiske, A.P., 2004. Watching social interactions produces Mouras, H., 2004. Identifying distinct components in the cerebral treatment
dorsomedial prefrontal and medial parietal BOLD fMRI signal increases of visual sexual information through functional neuroimaging: new
compared to a resting baseline. Neuroimage 21 (3), 1167–1173. approaches about male sexuality in man. J. Soc. Biol. 198 (3), 247–253.
1494 M. Walter et al. / NeuroImage 40 (2008) 1482–1494
Mouras, H., 2006. Neuroimaging techniques as a new tool to study the Phan, K.L., Wager, T., Taylor, S.F., Liberzon, I., 2002. Functional neuroanatomy
neural correlates involved in human males sexual arousal. Current of emotion: a meta-analysis of emotion activation studies in PET and fMRI.
Medical Imaging Reviews 2 (1), 71–77. Neuroimage 16 (2), 331–348.
Mouras, H., 2007. Central role of somatosensory processes in sexual arousal Phan, K.L., Taylor, S.F., Welsh, R.C., Ho, S.H., Britton, J.C., Liberzon, I.,
as identified by neuroimaging techniques. Behav. Brain Sci. 30 (2), 217. 2004. Neural correlates of individual ratings of emotional salience: a
Mouras, H., Stoleru, S., 2007. Functional neuroanatomy of sexual arousal. In: trial-related fMRI study. Neuroimage 21 (2), 768–780.
Kandeel, F., Lue, T., Pryor, J., Swerdloff, R. (Eds.), Male Sexual Dysfunction: Ponseti, J., Bosinski, H.A., Wolff, S., Peller, M., Jansen, O., Mehdorn, H.M.,
Pathophysiology and Treatment. New York: Informa Healthcare. Buchel, C., Siebner, H.R., 2006. A functional endophenotype for sexual
Mouras, H., Stoleru, S., Bittoun, J., Glutron, D., Pelegrini-Issac, M., Paradis, orientation in humans. Neuroimage 33 (3), 825–833.
A.L., Burnod, Y., 2003. Brain processing of visual sexual stimuli in Redoute, J., Stoleru, S., Gregoire, M.C., Costes, N., Cinotti, L., Lavenne, F.,
healthy men: a functional magnetic resonance imaging study. Neuro- Le Bars, D., Forest, M.G., Pujol, J.F., 2000. Brain processing of visual
image 20 (2), 855–869. sexual stimuli in human males. Hum. Brain Mapp. 11 (3), 162–177.
Nagai, Y., Critchley, H.D., Featherstone, E., Fenwick, P.B., Trimble, M.R., Russchen, F.T., Bakst, I., Amaral, D.G., Price, J.L., 1985. The amygdalos-
Dolan, R.J., 2004. Brain activity relating to the contingent negative triatal projections in the monkey. An anterograde tracing study. Brain
variation: an fMRI investigation. Neuroimage 21 (4), 1232–1241. Res. 329 (1–2), 241–257.
Nichols, T., Brett, M., Andersson, J., Wager, T., Poline, J.B., 2005. Valid con- Siwek, D.F., Pandya, D.N., 1991. Prefrontal projections to the mediodorsal
junction inference with the minimum statistic. Neuroimage 25 (3), 653–660. nucleus of the thalamus in the rhesus monkey. J. Comp. Neurol. 312 (4),
Northoff, G., Bermpohl, F., 2004. Cortical midline structures and the self. 509–524.
Trends Cogn. Sci. 8 (3), 102–107. Stoleru, S., Gregoire, M.C., Gerard, D., Decety, J., Lafarge, E., Cinotti, L.,
Northoff, G., Grimm, S., Boeker, H., Schmidt, C., Bermpohl, F., Heinzel, A., Lavenne, F., Le Bars, D., Vernet-Maury, E., Rada, H., et al., 1999.
Hell, D., Boesiger, P., 2006. Affective judgment and beneficial decision Neuroanatomical correlates of visually evoked sexual arousal in human
making: ventromedial prefrontal activity correlates with performance in males. Arch. Sex Behav. 28 (1), 1–21.
the Iowa Gambling Task. Hum. Brain Mapp. 27 (7), 572–587. Taylor, S.F., Phan, K.L., Decker, L.R., Liberzon, I., 2003. Subjective rating
Northoff, G., Walter, M., Schulte, R.F., Beck, J., Dydak, U., Henning, A., of emotionally salient stimuli modulates neural activity. Neuroimage 18
Boeker, H., Grimm, S., Boesiger, P., 2007. GABA concentrations in the (3), 650–659.
human anterior cingulate cortex predict negative BOLD responses in Velayos, J.L., Reinoso-Suarez, F., 1982. Topographic organization of the
fMRI. Nat. Neurosci. 10 (12), 1515–1517. brainstem afferents to the mediodorsal thalamic nucleus. J. Comp. Neurol.
Northoff, G., Schneider, F., Rotte, M., Matthiae, C., Tempelmann, C., Wiebking, 206 (1), 17–27.
C., Bermpohl, F., Heinzel, A., Danos, P., Heinze, H.J., et al., in press. Differ- Vogt, B.A., 2005. Pain and emotion interactions in subregions of the cingulate
ential parametric modulation of self-relatedness and emotions in different gyrus. Nat. Rev. Neurosci. 6 (7), 533–544.
brain regions. Hum. Brain Mapp. (Electronic publication ahead of print). Vogt, B.A., Finch, D.M., Olson, C.R., 1992. Functional heterogeneity in
Ortony, A., Turner, T.J., 1990. What’s basic about basic emotions? Psychol. cingulate cortex: the anterior executive and posterior evaluative regions.
Rev. 97 (3), 315–331. Cereb. Cortex 2 (6), 435–443.
Panksepp, J., 1982. Toward a general psychobiological theory of emotions. Vuilleumier, P., 2005. How brains beware: neural mechanisms of emotional
Behav. Brain Sci. 5, 407–467. attention. Trends Cogn. Sci. 9 (12), 585–594.
Panksepp, J., 1992. A critical role for “affective neuroscience” in resolving Vuilleumier, P., Pourtois, G., 2007. Distributed and interactive brain
what is basic about basic emotions. Psychol. Rev. 99 (3), 554–560. mechanisms during emotion face perception: evidence from functional
Panksepp, J., 1998. Affective Neuroscience: The Foundations of Human and neuroimaging. Neuropsychologia 45 (1), 174–194.
Animal Emotions. Oxford University Press, New York. Vuilleumier, P., Armony, J.L., Driver, J., Dolan, R.J., 2001. Effects of
Parsons, L.M., Denton, D., Egan, G., McKinley, M., Shade, R., Lancaster, J., attention and emotion on face processing in the human brain: an event-
Fox, P.T., 2000. Neuroimaging evidence implicating cerebellum in related fMRI study. Neuron 30 (3), 829–841.
support of sensory/cognitive processes associated with thirst. Proc. Natl. Vuilleumier, P., Richardson, M.P., Armony, J.L., Driver, J., Dolan, R.J.,
Acad. Sci. U. S. A. 97 (5), 2332–2336. 2004. Distant influences of amygdala lesion on visual cortical
Paul, T., Schiffer, B., Zwarg, T., Kruger, T.H., Karama, S., Schedlowski, M., activation during emotional face processing. Nat. Neurosci. 7 (11),
Forsting, M., Gizewski, E.R., 2007. Brain response to visual sexual 1271–1278.
stimuli in heterosexual and homosexual males. Hum. Brain Mapp. Walter, M., Matthia, C.,Wiebking, C., Rotte, M., Tempelmann, C., Bogerts,
(Electronic publication ahead of print). B., Heinze, H.J., Northoff, G., 2007. Preceding attention and the
Peelen, M.V., Downing, P.E., 2007. The neural basis of visual body dorsomedial prefrontal cortex: Process specificity versus domain
perception. Nat. Rev. Neurosci. 8 (8), 636–648. dependence. Hum. Brain Mapp. (Electronic publication ahead of print).
Pessoa, L., Kastner, S., Ungerleider, L.G., 2002a. Attentional control of the Wittmann, B.C., Bunzeck, N., Dolan, R.J., Duzel, E., 2007. Anticipation of
processing of neural and emotional stimuli. Brain Res. Cogn. Brain Res. novelty recruits reward system and hippocampus while promoting
15 (1), 31–45. recollection. Neuroimage 38 (1), 194–202.
Pessoa, L., McKenna, M., Gutierrez, E., Ungerleider, L.G., 2002b. Neural Wrase, J., Kahnt, T., Schlagenhauf, F., Beck, A., Cohen, M.X., Knutson,
processing of emotional faces requires attention. Proc. Natl. Acad. Sci. B., Heinz, A., 2007. Different neural systems adjust motor be-
U. S. A. 99 (17), 11458–11463. havior in response to reward and punishment. Neuroimage 36 (4),
Peterson, Z.D., Janssen, E., 2007. Ambivalent affect and sexual response: 1253–1262.
the impact of co-occurring positive and negative emotions on subjective Yacubian, J., Sommer, T., Schroeder, K., Glascher, J., Braus, D.F., Buchel,
and physiological sexual responses to erotic stimuli. Arch. Sex Behav. C., 2007. Subregions of the ventral striatum show preferential coding of
36 (6), 793–807. reward magnitude and probability. Neuroimage 38 (3), 557–563.
FOR MORE INFORMATION ON THIS TOPIC CLICK HERE
No comments:
Post a Comment