European Journal of Neuroscience, Vol. 24, pp.

33053316, 2006

doi:10.1111/j.1460-9568.2006.05206.x

Regional cerebral blood ow changes associated with clitorally induced orgasm in healthy women
Janniko R. Georgiadis,1 Rudie Kortekaas,1 Rutger Kuipers,1 Arie Nieuwenburg,2 Jan Pruim,3 A. A. T. Simone Reinders3 and Gert Holstege4
1

Department of Anatomy and Embryology, University Medical Center Groningen, University of Groningen, Antonius Deusinglaan 1, bldg 3215 room 729, 9713 AV Groningen, the Netherlands 2 Department of Urology, University Medical Center, University of Groningen, Hanzeplein 1, P.O. Box 30.001, 9700 RB Groningen, The Netherlands 3 Department of Nuclear Medicine and Molecular Imaging, University of Groningen, Hanzeplein 1, P.O. Box 30.001, 9700 RB Groningen, The Netherlands 4 Department of Uroneurology, University Medical Center, University of Groningen, Antonius Deusinglaan 1, bldg 3215 room 505, 9713AV Groningen, The Netherlands Keywords: cerebellum, human, orbitofrontal cortex, PET, sexual behaviour

Abstract
There is a severe lack of knowledge regarding the brain regions involved in human sexual performance in general, and female orgasm in particular. We used [15O]-H2O positron emission tomography to measure regional cerebral blood ow (rCBF) in 12 healthy women during a nonsexual resting state, clitorally induced orgasm, sexual clitoral stimulation (sexual arousal control) and imitation of orgasm (motor output control). Extracerebral markers of sexual performance and orgasm were rectal pressure variability (RPstd) and perceived level of sexual arousal (PSA). Sexual stimulation of the clitoris (compared to rest) signicantly increased rCBF in the left secondary and right dorsal primary somatosensory cortex, providing the rst account of neocortical processing of sexual clitoral information. In contrast, orgasm was mainly associated with profound rCBF decreases in the neocortex when compared with the control conditions (clitoral stimulation and imitation of orgasm), particularly in the left lateral orbitofrontal cortex, inferior temporal gyrus and anterior temporal pole. Signicant positive correlations were found between RPstd and rCBF in the left deep cerebellar nuclei, and between PSA and rCBF in the ventral midbrain and right caudate nucleus. We propose that decreased blood ow in the left lateral orbitofrontal cortex signies behavioural disinhibition during orgasm in women, and that deactivation of the temporal lobe is directly related to high sexual arousal. In addition, the deep cerebellar nuclei may be involved in orgasm-specic muscle contractions while the involvement of the ventral midbrain and right caudate nucleus suggests a role for dopamine in female sexual arousal and orgasm.

Introduction
Female orgasm is an intriguing part of sexual behaviour: whereas in men orgasm is tightly coupled to reproduction through ejaculation, in women no such obvious and direct link exists. Nevertheless, proreproductive functions, including sperm retention (Fox & Fox, 1967), mate selection (Dunn et al., 2005) and strengthening of the pair-bond (Young & Wang, 2004) have been proposed for orgasm in women. Most scientic research of female orgasm has focused on physiological parameters. Heart rate, blood pressure and respiratory rate (Fox, 1976; Carmichael et al., 1994), but also plasma concentrations of oxytocin (Carmichael et al., 1987; Carmichael et al., 1994) and prolactin (Exton et al., 1999) increase dramatically. In addition, patterns of rectal and vaginal pressure change markedly (Bohlen et al., 1982), probably due to involuntary rhythmical contractions of perineal striated musculature and or the smooth muscle of the uterus (Masters & Johnson, 1966; Bohlen et al., 1982). Experiencing orgasm is a function of the brain and this is conrmed by clinical studies: brain trauma, epileptic seizures and psychopharmaca may strongly inuence the orgasmic ability and or experience (Lundberg & Brattberg, 1992; Aloni & Katz, 1999). Indeed, orgasm induces a profound change in mental state: women report that reaching orgasm requires conscious action and concentration (Sholty et al., 1984) but that experiencing orgasm involves emotions like intense pleasure, release of tension, feeling of inevitability, and loss of conscious behavioural control (for review, see Mah & Binik, 2001). EEG was used rst to investigate cerebral activity during ejaculation and associated orgasm in men (Cohen et al., 1976; Graber et al., 1985), but without conclusive ndings. In contrast, Tiihonen et al. (1994), using single photon emission tomography (SPECT), reported ejaculation-related increased activation in the right prefrontal cortex and decreased activation in all other cortical areas. However, no deep or subcortical brain regions were analysed. Later, a positron emission tomography (PET) study by our group showed various activations during ejaculation, mainly in subcortical parts of the brain (Holstege et al., 2003). In women only one study exists: using functional magnetic resonance imaging (fMRI), Komisaruk et al. (2004) reported brain activition related to orgasm induced by

Correspondence: Dr Janniko R. Georgiadis, as above. E-mail: j.r.georgiadis@med.umcg.nl Received 30 May 2006, revised 6 September 2006, accepted 2 October 2006

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd

3306 J. R. Georgiadis et al. vaginocervical self-stimulation in three spinal cord-injured women. Unfortunately, the statistical power in this study was low, which makes the activations, which were scattered throughout the brain in each of the three subjects, very difcult to interpret. Taken together, there is a severe lack of knowledge regarding the brain regions involved in orgasm in general, and female orgasm in particular. We aimed to investigate orgasm-related brain activity in healthy women. PET was used because it is more robust to motion artifact than fMRI. Based on prior studies on the orgasmic brain, we hypothesised that female orgasm would involve decreased activation of the neocortex and increased activation of subcortical parts of the brain. In particular, because of the rewarding nature of an orgasm, we expected involvement of dopaminergic cell groups in the ventral midbrain and their targets in the striatum. We also expected activation of the hypothalamus because it is the site where oxytocin (Poulain & Wakerley, 1982) and prolactin are produced (Quadri et al., 1977) and because it is renowned for its role in female reproductive behaviour (Pfaff et al., 1994). intervals of  8 min between completion and commencement. A scanspecic calculated attenuation correction was performed to minimize interscan displacement-induced variance (Reinders et al., 2002). Experimental tasks The experimental setup consisted of four conditions, which were divided over eight scans (Fig. 2A). 1. Rest: passive nonsexual resting state. 2. Imitation: voluntary repetitive contractions of hip, buttock, abdominal and pelvic oor muscles in a rhythmic orgasm-like fashion, while receiving stimulation to the clitoris. This condition was included to control for the motor output during orgasm. 3. Stimulation: receiving clitoral stimulation without executing bodily movements. The subjects were encouraged to enjoy the stimulation, but were also explicitly instructed not to have an orgasm. This condition was included to control for the clitoral stimulation and high sexual arousal during orgasm. 4. Orgasm: orgasm induced by clitoral stimulation. The subject tried to reach orgasm in the rst minute after tracer injection. After each orgasm attempt, volunteers reported whether they had reached orgasm and, if so, when they thought it had occurred relative to the moment of tracer injection. Data analysis Verication of reported orgasms Reported orgasms were included if they met the following criteria: (i) according to the subject, orgasm was unambiguus and occurred in the rst minute of the scan; (ii) the orgasm report was supported by the rectal pressure pattern of the same scan; and (iii) within a subject the rectal pressure pattern of a reported orgasm and that of an imitation (condition 3) were clearly distinct, because both involved contraction of the same muscles, but probably with different frequency characteristics (Bohlen et al., 1982). Extracerebral measurements Fluctuations in rectal pressure appear to be an important feature of an orgasm (Bohlen et al., 1980; Bohlen et al., 1982). Therefore, we calculated the SD in the rectal pressure signal (RPstd) over the rst minute of each scan. Subsequently, for each subject these values were normalised (value average value over eight scans). Ratings of perceived level of sexual arousal (PSA) were not normalised. A KruskalWallis test was performed on both parameters to investigate the inuence of the experimental conditions on the variance. In the case of a signicant condition effect, post hoc analysis (Wilcoxon rank sum test) was performed to explore differences between experimental conditions, which were Bonferroni-corrected for multiple comparisons. PET measurements Statistical parametric mapping software (SPM99; Friston et al., 1995) was used for spatial preprocessing (spatial normalisation, afne only; smoothing, 10 mm full width at half maximum), global blood ow normalisation (proportional scaling) and statistical analysis of the PET data. The rst six image frames (1 min) were summed and analysed. Small-volume correction was carried out for brain regions hypothesised to be involved in female orgasm, but with subsignicant cluster P-values (P > 0.05, corrected for the whole brain). The stereotactic atlases of the human brain by Talairach & Tournoux (1988) and Schmahmann et al. (1999) were used for localisation of regions in the cerebral cortex and the cerebellum, respectively. For data presentation

Materials and methods

Participants and protocol Twelve heterosexual women (mean age 34, range 2147 years) were included in this study. All women (subjects) volunteered for the study together with their male partners. According to the Declaration of Helsinki, all participants gave written informed consent. The experimental procedures were approved by the internal ethics committee of the University Medical Center Groningen. The subjects were healthy and righthanded, and without any history of psychiatric or sexual disorders. Genital stimulation was provided by the male partner and was centred on the clitoris, because clitoral stimulation most easily leads to orgasm (Mah & Binik, 2001; Lloyd, 2005). An adhesive band was used to restrain the subjects head. During scans the subject had her eyes closed. The couples were allowed to communicate verbally between successive scans. Data acquisition Extracerebral measurements Rectal pressure was measured continuously with a rectal probe containing a microtip transducer (B. Braun Melsungen AG , type 501002 7; Melsungen, Germany). The probe was positioned in the rectum by an experienced urologist. The signals were sampled via the MMS system (Medical Measurements Systems B.V.; Enschede; the Netherlands) and 50 Hz data were stored for analysis. In addition, after each scan subjects were asked to rate their perceived level of sexual arousal on a 10-point scale (0, not sexually aroused; 10, highest level of sexual arousal ever experienced). PET measurements PET scans were made in 3-D mode (63 planes; axial eld of view of 15.5 cm) using an Ecat Exact HR+ (CTI Siemens, Knoxville, TN, USA) camera with a spatial resolution of 45 mm full width at half maximum in all three directions. The radiotracer [15O]-H2O was used to measure regional cerebral blood ow (rCBF). Per scan, 500 MBq of activity, dissolved in 32 mL of 0.9% saline, was administered intravenously into the right forearm at 8 mL s. After injection of the radiotracer, data were collected for 120 s. Eight dynamic scans were made: each comprised one injection of radiotracer and the subsequent acquisition of a total of eight 3-D image frames, seven of 10 s duration and the eighth of 50 s duration. Consecutive scans were made with

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

Brain activation during clitoral stimulation and orgasm 3307 we used the medical imaging tool AMIDE (http://amide.sourceforge.net). The following analyses were carried out. Main effects. The parameters of the General Linear Model were estimated with volunteer and condition as factors, leaving 64 degrees of freedom. A voxel-wise F-test (P < 0.05, corrected for multiple comparisons) was performed to investigate which were the primary brain regions involved in any of the experimental conditions. Comparisons between conditions. Using the same experimental design, differences in rCBF between conditions were tested by a voxel-wise two-tailed t-test. First, we separately compared stimulation and orgasm with rest. Then, we compared orgasm with stimulation (sexual arousal control) and imitation (motor output control). A less stringent signicance threshold (P < 0.001, uncorrected) was used in order to be more sensitive to smaller brain regions that may be involved in female orgasm, such as the hypothalamus and ventral midbrain.

Fig. 1. As an example of orgasm verication, three typical cases of rectal pressure patterns during reported orgasm (red line) and imitation (blue line) are depicted. Despite the fact that pelvic muscles were contracted in both instances, the difference between the two patterns is clearly appreciable.

Fig. 2. (A) Scan order and time span of the experiment. Scans (black lines) lasted 2 min. Consecutive scans were made with an 8 min interval (grey line). (B) Main effects of the present study. The glass brain shows an F-map (P < 0.05, corrected), giving an overview of the brain regions involved in any of the experimental conditions. This F-map is also depicted in colourscale, superimposed on sagittal sections of a standard single-subject T1-weighted MR image (SPM99). The location of the section relative to the midline is indicated in the bottom right corner of each section (negative value for x, left hemisphere). Bar graphs in the top right corner of each section indicate how the response in the centre of signicant brain clusters varied during the four conditions. This average signal change for each parameter relative to the mean normalised activition over all scans is expressed as a percentage. Abbreviations: F, F-value; Im, imitation of orgasm; LH, left hemisphere; Or, orgasm; Re, nonsexual passive rest; RH, right hemisphere; St, clitoral stimulation. The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

3308 J. R. Georgiadis et al.

Table 1. Main effect of experimental condition Side R L L R L L Cluster Precentral gyrus (MI) postcentral gyrus (SI) paracentral lobule Inferior temporal gyrus Middle inferior frontal gyrus (lateral OFC) Superior temporal gyrus (anterior temporal pole) Deep cerebellar nuclei Inferior temporal gyrus BA 345 20 11 47 38 () 20 x 12 )52 )38 32 )10 )42 y )30 )28 44 8 )48 )12 z 74 )30 )16 )44 )22 )38 Z-value 5.96 5.96 5.91 5.65 5.50 5.02 F-value 21.97 21.94 21.48 19.41 18.23 14.93

The signicance threshold of the F-test was P < 0.05, corrected for multiple comparisons. Brain regions are listed with the stereotactic coordinates (Montreal Neurological Institute) of the most signicant voxel of the cluster and the corresponding Z- and cluster P-values. Abbreviations: BA, Brodmanns area; L, left; R, right; x, leftright; y, anteriorposterior; z, dorsalventral.

Regression analyses. PSA ratings (sexual arousal) and RPstd values (rectal pressure variability) entered the General Linear Model as covariate (no interaction with subject) and were correlated with rCBF in each voxel of the brain. Because there were eight instead of 12 subjects who provided PSA ratings, there were separate designs for PSA and RPstd. In the case of RPstd there were 58 degrees of freedom and in case of PSA 43. For the same reason as before, we explored possible correlations at an uncorrected P < 0.001.

MI), but also including the postcentral gyrus (primary somatosensory cortex: SI) and, ventrally, the paracentral lobule on the mesial surface of the brain; (ii) bilateral inferior temporal gyrus and anterior temporal pole; (iii) the ventral border of the left middle and superior frontal gyrus, corresponding to the lateral orbitofrontal cortex; and (iv) the left deep cerebellar nuclei. The rCBF response plots in Fig. 2B depict the activation over the course of the experiment for each of these brain regions. Stimulation vs. rest (Table 2 and Fig. 3) Stimulation > rest. A signicant rCBF increase was found in the left inferior parietal lobule, corresponding to the secondary somatosensory cortex (SII), and in the right dorsal SI. Trends of activation were in the left dorsal SI, the right frontal dorsal gyrus, the left dorsolateral prefrontal cortex and lateral orbitofrontal cortex, and, posteriorly, in the posterior cingulate, the precuneus, and the superior parietal lobule. Stimulation < rest. Trends of decreased rCBF were found in the ventrolateral aspect of the left temporal lobe (inferior and middle temporal gyrus) and in the left amygdala. Orgasm vs. rest (Table 2 and Fig. 3) Orgasm > rest. A large signicant cluster of increased rCBF was found surrounding the dorsal aspect of the central sulcus. The peak of the activation was in SI, but the cluster also included the dorsal primary motor cortex and the paracentral lobule. Additional signicant activation was observed in the deep cerebellar nuclei and adjacent anterior lobe of the vermis. Trends of activation were in the left posterior insula and the right cerebellar hemisphere. Orgasm < rest. Widespread and signicant rCBF decreases were found on the ventro- and posterolateral aspect of the left temporal lobe (inferior + middle temporal gyrus), the left amygdala and, more anteriorly, the gyrus rectus (ventromedial prefrontal cortex). On the right side there was signicant deactivation of the anterior temporal pole. In the right ventro- and posterolateral temporal lobe, bilateral fusiform gyrus, left anterior insula and bilateral lingual gyrus trends of deactivation were observed. Orgasm vs. control tasks (Table 3; Fig. 4) Orgasm > stimulation. rCBF was signicantly increased in the region of the left deep cerebellar nuclei. In addition, trends of activation were present in bilateral cerebellar vermis, the right cerebellar hemisphere, and in the paracentral lobule. Orgasm < stimulation. The primary rCBF decrease was in the left lateral orbitofrontal cortex. Additional signicant deactivations were

Results
Reported orgasms Six women reached one orgasm, four women reached two orgasms and two women reached three orgasms each. Six orgasms were ill-timed and two had an indistinct rectal pressure pattern. Therefore, of the 28 orgasms reported, 20 were included in the analysis. There was high intersubject variability with regard to the orgasmic rectal pressure patterns but in multiorgasmic women rectal pressure patterns looked consistent between orgasms. Figure 1 demonstrates that orgasmic rectal pressure patterns were clearly distinct from those during imitation. Extracerebral measurements The median (1st, 3rd quartile) of PSA, a measure of sexual arousal, was for rest, 0 (0, 0); for imitation, 4 (3, 5); for stimulation, 6 (6, 7); for orgasm, 9 (8.1, 10). There was a signicant effect of experimental condition on the PSA ratings (P < 10)8). For all sexual conditions, the PSA was signicantly higher than during rest. The PSA after stimulation was signicantly higher (P < 10)4) than after imitation, and signicantly lower (P < 10)6) than after orgasm. The median (1st, 3rd quartile) of RPstd, a measure of rectal pressure variability, was for rest, 0.31932 (0.14852, 0.55158); for imitation, 0.91827 (0.5398, 1.4363); for stimulation, 0.45855 (0.3753, 0.65063); for orgasm, 1.811 (1.3366, 2.1299). There was a signicant effect of experimental condition on RPstd (P < 10)7). In particular, RPstd during orgasm was signicantly greater than during rest (P < 0.001), imitation (P < 0.003) and stimulation (P < 10)5). RPstd during imitation was signicantly greater than during rest (P < 0.007), but not than during stimulation (P 0.054). Pet measurements In this section changes in rCBF are called signicant in the case of a corrected cluster P < 0.05. Subsignicant rCBF changes (P > 0.05 corrected) are called trends. Main effects (Table 1 and Fig. 2B) Signicant effects of interest were found in: (i) the region of the dorsal central sulcus, centred on the precentral gyrus (primary motor cortex:

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

Brain activation during clitoral stimulation and orgasm 3309

Table 2. Clitoral stimulation and orgasm vs. rest Side Centre of cluster BA x )52 16 )20 )36 )44 )28 0 )30 38 6 )28 )62 )44 )58 )32 16 )8 36 26 24 )50 32 )14 )28 60 )20 )12 56 52 y )28 )40 )44 )14 34 46 )72 )56 )4 )34 )6 )54 )22 )58 )54 )36 )48 )18 )44 )60 )26 8 )90 )52 )60 26 )60 )10 )44 z Z-value t-value Pcorrected

Stimulation > Rest (activations) L Inferior parietal lobule (SII) R Postcentral gyrus (SI) L Postcentral gyrus (SI) L Postcentral gyrus (SI) L Middle frontal gyrus (dorsolateral prefrontal cortex) L Middle inferior frontal gyrus (lateral OFC) M Precuneus L Superior parietal lobule R Middle frontal gyrus (premotor cortex) R Posterior cingulate Stimulation < Rest (deactivations) L Amygdala L Inferior middle temporal gyrus L Fusiform gyrus L Inferior temporal gyrus L Fusiform gyrus Orgasm R L R R R > Rest (activations) Postcentral gyrus (SI) precentral gyrus (MI) paracentral lobule Deep cerebellar nuclei anterior vermis cerebellar hemisphere Posterior insula Cerebellar hemisphere Cerebellar hemisphere

40 57 57 3 46 10 11 7 7 6 23 () 21 37 20 37 37 345 () () () () 10 20 21 37 38 18 37 37 () 18 21 21 37

40 68 66 58 30 )16 38 50 60 30 )22 )6 )28 )20 )12 66 )22 14 )28 )26 )30 )44 )12 )10 )6 )6 6 )12 )16

4.67 4.63 4.45 4.02 3.86 3.72 3.71 3.69 3.56 3.55 3.85 3.73 3.73 3.71 3.60 5.8 5.42 3.71 3.69 3.63 5.93 5.92 4.33 4.04 3.76 3.72 3.58 3.56 3.54

5.17 5.12 4.88 4.33 4.14 3.97 3.96 3.94 3.78 3.77 4.13 3.99 3.98 3.96 3.83 6.79 6.22 3.96 3.94 3.86 6.99 6.97 4.73 4.36 4.02 3.97 3.81 3.78 3.76

0.015* 0.000* 0.415 0.581 0.268 0.909 0.618 0.864 0.903 0.792 0.344 0.693 0.857 0.438 0.914 0.001* 0.000* 0.670 0.829 0.716 0.000* 0.001* 0.296 0.421 0.663 0.936 0.807 0.770 0.936

Orgasm < Rest (deactivations) L Inferior temporal gyrus middle temporal gyrus amygdala ventromedial prefrontal cortex R Superior temporal gyrus (temporal pole) L Lingual gyrus L Fusiform gyrus R Inferior temporal gyrus L Anterior insula L Lingual gyrus R Middle temporal gyrus R Fusiform gyrus middle temporal gyrus

The signicance threshold was P < 0.001, uncorrected. Brain regions are listed with the stereotactic coordinates (Montreal Neurological Institute) of the most signicant voxel of the cluster and the corresponding t- and cluster P-values (corrected for the whole brain). *Clusters with signicant rCBF change, i.e. with a corrected cluster P < 0.05. Abbreviations: BA, Brodmanns area; L, left; R, right; M, middle; x, leftright; y, anteriorposterior; z, dorsalventral.

found in the left ventrolateral temporal lobe and fusiform gyrus, the left anterior temporal pole, the left posterior insula and bilateral dorsomedial prefrontal cortex. Trends of deactivation were observed in the right lateral orbitofrontal cortex, the right temporal pole and fusiform gyrus, the left superior and inferior (SII) parietal lobules, and the precuneus. Orgasm > imitation. A trend of increased rCBF was found in the left deep cerebellar nuclei. Orgasm < imitation. Signicant rCBF decreases were found in bilateral ventrolateral temporal lobe, fusiform gyrus and anterior temporal pole, the ventromedial prefrontal cortex and the left lateral orbitofrontal cortex. The right lateral orbitofrontal cortex, the left parahippocampal gyrus, the right paracentral lobule and the left middle frontal gyrus showed a trend towards deactivation. Regression analyses (Table 4; Fig. 5) RPstd. RPstd correlated positively (P < 0.05) with rCBF in the left deep cerebellar nuclei. A positive trend was observed in the right cerebellar hemisphere. Signicant negative correlations between RPstd and rCBF were found bilaterally in dorsolateral, dorsomedial and ventromedial prefrontal cortices, lateral orbitofrontal cortex and anterior cingulate gyrus. In the posterior brain there were signicant

negative correlations in the left lingual gyrus and precuneus dorsally, and the parahippocampal gyrus ventrally. Negative trends were observed in the right lingual gyrus and precuneus, and the left middle occipital gyrus, ventrolateral temporal lobe and inferior parietal lobule. PSA. A positive correlation between PSA and rCBF was found in the ventral midbrain and right caudate nucleus. After small-volume correction (sphere with 10 mm radius) both regions reached a corrected P-value of 0.05. In the left deep cerebellar nuclei and cerebellar hemisphere the correlation was subsignicant. PSA correlated negatively with rCBF in large parts of the temporal lobe (ventrolateral, ventral and anterior temporal pole) bilaterally.

Discussion
The present results provide insight into the brain regions involved in clitoral stimulation and clitorally induced orgasm in women. We discovered that decreased rCBF in the neocortex was prominent during female orgasm, which was consistent with our hypothesis. Active limbic and subcortical involvement in female orgasm was less than we anticipated. However, signicant orgasm-related increased rCBF was found in the deep cerebellar nuclei, while the perceived level of sexual arousal was positively correlated with rCBF in the ventral midbrain and right caudate nucleus.

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

3310 J. R. Georgiadis et al.

Fig. 3. t-Contrasts: clitoral stimulation and orgasm vs. rest. The signicance threshold in this gure is P < 0.001, uncorrected. For each comparison, a t-map of rCBF changes is depicted (glass brain: left side of the gure). These rCBF increases (red scaling) and decreases (blue scaling) are also superimposed on horizontal sections of a standard single-subject T1-weighted MR-image (SPM99). The top row shows the result for stimulation vs. rest, the botom row for orgasm vs. rest. The location of each section relative to the anterior commissure (AC) is indicated in the bottom right corner of each section (negative value for z, ventral to AC). Abbreviations: MI, primary motor cortex; OFC, orbitofrontal cortex; PFC, prefrontal cortex; rCBF, regional cerebral blood ow; SI, primary somatosensory cortex; SII, secondary somatosensory cortex.

Fig. 4. t-Contrasts: orgasm control comparisons. The signicance threshold in this gure is P < 0.001, uncorrected. For each comparison, a t-map of rCBF changes is depicted (glass brain: left side of the gure). These rCBF increases (red scaling) and decreases (blue scaling) are also superimposed on horizontal sections of a standard single-subject T1-weighted MR image provided by SPM99. The top row shows the result for orgasm vs. stimulation (sexual arousal control comparison), the bottom row for orgasm vs. imitation (motor output control comparison). The location of the section relative to the anterior commissure (AC) is indicated in the bottom right corner of each section (negative value for z, ventral to AC). Abbreviations: MI, primary motor cortex; OFC, orbitofrontal cortex; PFC, prefrontal cortex; rCBF, regional cerebral blood ow; SI, primary somatosensory cortex.

Extracerebral measurements As expected, orgasm was the condition with the highest PSA. Furthermore, PSA was signicantly higher after stimulation than after imitation. During debrieng most of the subjects stated that the

voluntary motor act required for imitation precluded sexual arousal, despite the clitoral stimulation that they had received. In contrast, they reported to have enjoyed the clitoral stimulation during stimulation, which often meant that they had to restrain themselves to prevent

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

Brain activation during clitoral stimulation and orgasm 3311

Table 3. Orgasm vs. control tasks Side Orgasm L L R R R Orgasm L L L R L R R R L L L L R R L Centre of cluster > Stimulation (activations) Deep cerebellar nuclei anterior vermis cerebellar hemisphere Posterior vermis vermis Paracentral lobule (MI SI) Cerebellar hemisphere < Stimulation (deactivations) Middle inferior frontal gyrus (lateral OFC) Inferior temporal gyrus middle temporal gyrus fusiform gyrus Posterior insula Superior frontal gyrus (dorsomedial PFC) Superior frontal gyrus (dorsomedial PFC) Fusiform gyrus Superior temporal gyrus (temporal pole) Middle inferior frontal gyrus (lateral OFC) Superior parietal lobule Inferior parietal lobule (SII) Precuneus posterior cingulate Inferior temporal gyrus Precuneus Superior frontal gyrus Middle frontal gyrus BA x )10 )4 10 14 26 )38 )52 )46 18 )12 42 32 38 )36 )56 )6 )60 6 20 )34 )24 )52 2 30 )44 40 2 )22 )20 )50 y )50 )80 )66 )24 )44 44 )26 )12 32 36 )12 8 42 )66 )28 )52 )38 )66 58 14 )50 )30 42 8 34 40 )16 )40 )16 16 z )20 )44 )30 58 )28 )16 )30 6 28 26 )34 )44 )14 56 40 42 )24 34 26 40 )38 )32 )28 )42 )14 )14 66 )12 )24 24 Z-value t-value Pcorrected

() () () 5 () 10 11 47 20 () 8 8 20 38 11 47 7 40 31 20 7 9 10 6 () 20 38 11 20 38 11 47 11 47 5 35 27 28 44

5.13 4.16 3.72 3.71 3.64 6.48 4.67 4.36 4.08 4.00 4.77 4.76 4.27 4.14 3.92 3.71 3.53 3.52 3.49 3.40 3.81 5.94 4.29 5.36 4.46 4.50 4.33 4.17 3.64 3.60

5.80 4.52 3.97 3.96 3.87 7.90 5.17 4.77 4.42 4.31 5.30 5.29 4.65 4.48 4.22 3.96 3.75 3.74 3.70 3.59 4.08 7.00 4.68 6.13 4.89 4.95 4.73 4.52 3.88 3.83

0.000* 0.903 0.800 0.663 0.857 0.000* 0.002* 0.010* 0.034* 0.050* 0.513 0.546 0.067 0.857 0.519 0.670 0.950 0.871 0.936 0.931 0.936 0.000* 0.000* 0.001* 0.002* 0.083 0.079 0.101 0.444 0.610

Orgasm > Imitation (activations) L Cerebellar hemisphere Orgasm < Imitation (deactivations) L Inferior temporal gyrus fusiform gyrus superior temporal gyrus (anterior temporal pole) R gyrus rectus R Superior temporal gyrus (temporal pole) fusiform gyrus L Middle inferior frontal gyrus (lateral OFC) R Middle inferior frontal gyrus (lateral OFC) R Paracentral lobule (MI) L Parahippocampal gyrus L Parahippocampal gyrus L Inferior frontal gyrus

The signicance threshold was P < 0.001, uncorrected. The top section lists the results of the sexual arousal control comparison (orgasm vs. stimulation), the bottom section of the motor output control comparison (orgasm vs. imitation). Brain regions are listed with the stereotactic coordinates (Montreal Neurological Institute) of the most signicant voxel of the cluster and the corresponding t- and cluster P-values (corrected for the whole brain). *Clusters with signicant rCBF change, i.e. with a corrected cluster P < 0.05. Abbreviations: BA, Brodmanns area; L, left; R, right; M, middle; t, t-value; x, leftright; y, anteriorposterior; z, dorsalventral.

orgasm. We designed this study so that stimulation could be used to control for the clitoral stimulation and high sexual arousal during orgasm and the PSA ratings conrmed that stimulation was the best condition for this purpose. Rectal pressure (RP) can be used as an indicator of muscular contractions in the rectal vicinity. RP patterns change markedly during orgasm in men and women (Bohlen et al., 1980; Bohlen et al., 1982), an important feature being increased RP variability. We tried to capture these uctuations by calculating the RPstd. Indeed, RPstd was signicantly greater during orgasm than during the other conditions, despite the fact that during imitation there were strong rhythmical contractions of muscles in the pelvic region. No signicant difference was found between imitation and stimulation (P 0.054). However, the relatively low P-value is indicative of more muscular activity during imitation. The RP measurements conrmed that RPstd may be a an indicator of orgasmic muscular contractions.

were not signicantly related to the experimental conditions in the present study. Instead, the effects of interest were found in the region of the dorsal central sulcus, the ventrolateral aspect of the temporal lobe, the left orbitofrontal cortex and the left deep cerebellar nuclei. The rCBF response plots in Fig. 2B illustrate how these regions behaved during the four conditions. For instance, rCBF in the temporal regions decreased from rest to orgasm, while the reverse occurred in the cerebellum. In the remainder of this paper we will discuss the results from comparisons between conditions (t-contrasts) in an effort to more precisely determine the role of these brain regions in female sexual performance.

rCBF changes during stimulation of the clitoris The importance of the clitoris for female sexual pleasure is undisputed. However, this is the rst account of brain regions involved in the experience of clitoral stimulation. The strongest rCBF increase was in the left secondary somatosensory cortex (SII). In male volunteers, penile tumescence was positively correlated with the level of activity in the left and right SII (Bocher et al., 2001; Ferretti et al., 2005) and stimulation of the erect penis elicited increased rCBF in the right SII (Georgiadis & Holstege,

PET measurements We rst performed an F-test to determine in which brain regions the variance could be explained by the experimental conditions. Contrary to our hypothesis, the hypothalamus, the brainstem and the striatum

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

3312 J. R. Georgiadis et al. 2005). Similarly, SII is consistently activated in response to visual erotica (Redoute et al., 2000; Bocher et al., 2001; Stoleru et al., 2003; Ferretti et al., 2005). In addition, nonsexual stimulation of the dorsal penile nerve (Makela et al., 2003), penile skin (Kell et al., 2005) and, in women, the vaginal vestibulum (Pukall et al., 2005) also increased activation in SII. The latter study showed that when vestibular stimulation was perceived as painful, SII was activated more strongly. Perhaps SII assigns a conscious label to the salience of a somatosensory stimulus, e.g. in the present study perception of the genital stimulation as sexual. Dorsally on the right primary somatosensory cortex (SI) there was additional signicant rCBF increase. The same region was activated on the left side at a subsignicant threshold. Our result adds to the existing controversy regarding the location of the external genitals on SI. In his classical experiments, Peneld found the genital SI to be located in the paracentral lobule on the interhemispheric surface (Peneld & Rasmussen, 1950), and this location has recently been conrmed (Nakagawa et al., 1998; Makela et al., 2003). However, other researchers have proposed a revision of the Peneld homunculus, providing evidence that the genital (penis) SI is located on the dorsal convexity of the postcentral gyrus (Bradley et al., 1998; Kell et al., 2005). We concur with the latter studies, showing increased rCBF on SIs dorsal convexity when women received clitoral stimulation. Of particular interest was a trend of increased rCBF that was observed in the left lateral orbitofrontal cortex (OFC), because it has been suggested that increased activity in this region may inhibit sexual behaviour (Stoleru et al., 1999; Beauregard et al., 2001; Redoute et al., 2005). Decreases of rCBF were not statistically signicant. Nevertheless, deactivation of these regions (amygdala, inferior temporal gyrus and fusiform gyrus) has consistently and strongly been related to increased sexual arousal (Stoleru et al., 1999; Redoute et al., 2000; Bocher et al., 2001; Arnow et al., 2002; Montorsi et al., 2003). Indeed, as the PSA ratings show, sexual arousal was signicantly elevated during stimulation.

Orgasm-related rCBF changes Orgasm vs. rest Despite liberal statistical thresholding, there was no orgasm-related increased rCBF in the hypothalamus, the brainstem or the striatum, regions known to be key players in sexual and or rewarding behaviours. Instead, there was a large cluster surrounding the dorsal central sulcus, centred on the dorsal convexity of SI. As the stimulation > rest comparison demonstrated, this is the clitoral part of SI. This large cluster also clearly included dorsal and interhemispheric parts of MI, corresponding with abdominal, hip and leg (Peneld & Rasmussen, 1950) and pelvic oor (Blok et al., 1997; Seseke et al., 2006) muscle control, respectively. Primary activation was also found in the anterior part of the cerebellum, centred on the left deep cerebellar nuclei and extending to the anterior lobe of the vermis. We have previously shown activation in this area of the cerebellum when men ejaculated and experienced orgasm (Holstege et al., 2003). The anterior lobe of the vermis is crucial for axial sensorimotor control (Schmahmann,

Table 4. Regression analyses Side Centre of cluster BA x )8 30 )44 34 38 )20 )62 )16 )38 )12 46 2 )46 0 4 14 10 26 40 )54 )62 )56 y )48 )38 4 6 12 )60 )52 )62 )54 )88 )72 )68 )56 )54 )18 )4 )52 )50 )12 )30 )54 )60 z )22 )28 36 )44 32 )2 )2 54 54 )12 2 60 )6 42 )12 18 )22 )38 )36 )30 )6 )22 Z-value t-value Pcorrected

Positive correlation with RPstd L Deep cerebellar nuclei R Cerebellar hemisphere Negative correlation with RPstd R L Superior temporal gyrus (temporal pole) inferior temporal gyrus fusiform gyrus gyrus rectus anterior cingulate superior frontal gyrus R Superior temporal gyrus (temporal pole) amygdala fusiform gyrus R Inferior middle frontal gyrus L Lingual gyrus parahippocampal gyrus precuneus L Middle temporal gyrus L Precuneus L Inferior parietal lobule L Lingual gyrus R Middle occipital gyrus L Precuneus L Inferior temporal gyrus M L Precuneus Positive correlation with PSA R ventral midbrain R Caudate nucleus L Deep cerebellar nuclei L Cerebellar hemisphere Negative R L L L correlation with PSA Inferior temporal gyrus fusiform gyrus uncus Inferior temporal gyrus middle temporal gyrus fusiform gyrus gyrus rectus Inferior temporal gyrus middle temporal gyrus fusiform gyrus Inferior temporal gyrus

() () 811 2025 32 20 8 9 44 19 31 21 7 40 18 19 7 37 31 () () () () 20 38 11 20 21 20 37 37

4.56 3.66 6.27 5.29 5.54 3.84 3.87 3.86 3.86 3.85 3.82 3.73 3.65 3.43 3.21 3.28 3.59 4.00 5.34 5.00 4.38 3.77

5.03 3.90 5.46 6.02 4.94 4.12 4.15 4.15 4.14 4.13 4.09 3.98 3.89 3.63 3.44 3.52 3.90 4.43 6.42 5.87 4.95 4.13

0.010* 0.858 0.000* 0.001* 0.011* 0.049* 0.348 0.822 0.815 0.404 0.739 0.287 0.546 0.777 0.050* 0.050* 0.547 0.645 0.000* 0.000* 0.020* 0.684

The signicance threshold was P < 0.001, uncorrected. The top section of this table lists correlations with rectal pressure standard deviation (RPstd) and the bottom section with perceived level of sexual arousal (PSA). Brain regions are listed with the stereotactic coordinates (Montreal Neurological Institute) of the most signicant voxel of the cluster and the corresponding t- and P-values (corrected for the whole brain). *Clusters with signicant correlation, i.e. with a corrected cluster P < 0.05. Abbreviations: BA, Brodmanns area; L, left; R, right; M, middle; t, t-value; x, leftright; y, anteriorposterior; z, dorsalventral. After smallvolume correction. The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

Brain activation during clitoral stimulation and orgasm 3313

Fig. 5. Positive correlations between the extracerebral measurements and rCBF. The signicance threshold in this gure is P < 0.001, uncorrected. Correlations are shown superimposed on horizontal sections of a standard single-subject T1-weighted MR-image provided by SPM99.

2004), and the deep cerebellar nuclei (dentate nucleus) have a tight functional relationship with the primary motor cortex (MI; Penhune & Doyon, 2005), which was also activated in the present study. Similar coactivation of the anterior cerebellum and MI was found with voluntary contraction of the pelvic oor (Blok et al., 1997; Seseke et al., 2006), indicating that the involvement of the cerebellum and the MI in female orgasm is probably related to axial motor control. Widespread rCBF decreases were found, particularly in the left inferior temporal gyrus, the left medial temporal lobe (including the amygdala), the right anterior temporal pole and the ventromedial prefrontal cortex. Similar decreases in activity have been found in men who were sexually aroused (Redoute et al., 2000; Bocher et al., 2001) and who achieved orgasm (Tiihonen et al., 1994). This orgasm-related perfusion decrease is also corroborated by clinical evidence: damage to the temporal and or frontal lobes frequently affects elements of sexual behaviour in men and women, including orgasm (Aloni & Katz, 1999). For instance, temporal lobe epilepsy may induce orgasmic auras (Janszky et al., 2004; Fadul et al., 2005), while temporal lobe resection aimed to resolve the epileptic seizures frequently causes hypersexuality (Baird et al., 2002). Frontal lobe deciency is also linked to increased sexual behaviour (Aloni & Katz, 1999), including hyperactive pelvic thrusting as a result of frontal lobe epilepsy (Leutmezer et al., 1999). In men suffering from erectile dysfunction (ED), administration of apomorphine resolved the ED and, most importantly, caused a down-regulation of activity in frontal and temporal lobes (Montorsi et al., 2003). Collectively,

these sources provide strong evidence that the frontal and temporal lobes have inhibitory control over sexual behaviour. Decreased perfusion of these regions would release the inhibition and enable sexual behaviour. Orgasm vs. control tasks Using a passive resting baseline often induces deactivation in mesial frontal and temporal regions when compared to a task of goal-directed nature, exposing the so-called default mode of resting brain activity (Gusnard & Raichle, 2001; Raichle et al., 2001). To avoid this pitfall, we compared rCBF during orgasm with rCBF during the other sexual conditions. More specically, stimulation served as a control for the genital stimulation and heightened sexual arousal during orgasm whereas imitation served as a control for the orgasm-related motor output. The sexual arousal control comparison (orgasm > stimulation) resulted in signicant rCBF increase in the left deep cerebellar nuclei, but the motor output control comparison (orgasm > imitation) did not leave signicantly activated brain regions. We therefore conclude that rCBF increases in the deep cerebellar nuclei were related to axial motor output, regardless of the behavioural and emotional context. The most marked rCBF decrease resulted from orgasm < stimulation and was found in the left lateral OFC. It is well known that hypoactivity of or damage to the OFC may result in antisocial, impulsive and disinhibited behaviour, including promiscuity and

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

3314 J. R. Georgiadis et al. sexual disinhibition (Damasio et al., 1994; Lapierre et al., 1995; Yang et al., 2005). Conversely, hyperactivity of the OFC is a feature of disorders such as obsessivecompulsive disorder (Saxena & Rauch, 2000) and these patients exhibit excessive self-control and reduced sexual behaviour, including anorgasmia (Aksaray et al., 2001). In healthy men, decreased rCBF in the left lateral OFC was found in response to visual sexual stimuli (Stoleru et al., 1999). Moreover, greater degrees of left lateral OFC deactivation were associated with higher plasma levels of testosterone (Redoute et al., 2005). On the other hand, the same region showed a trend of increased rCBF in response to clitoral stimulation in the present study (Figs 2B and 3), despite increased sexual arousal. This paradoxical opposite activation of the left lateral OFC during stimulation and orgasm may be explained as follows: using a very interesting paradigm, Beauregard et al. (2001) demonstrated that activity in the OFC increased when male participants who watched erotica voluntarily suppressed emerging feelings of sexual arousal. Likewise, in the present study participants were explicitly instructed not to have an orgasm during stimulation. Interestingly, most participants reported (off the record) that during stimulation they often had to hold back in order to prevent orgasm. This conscious control over a basic drive like sexual behaviour may be reected by the activation level of the left lateral OFC: increased rCBF during stimulation may represent high inhibition, whereas decreased rCBF during orgasm may represent sexual disinhibition. The latter possibly reects the loss of conscious control or the release of tension that women report when they describe an orgasmic experience (Mah & Binik, 2001). Although the rCBF difference in the left lateral OFC was largest for stimulation > orgasm, it was also present when orgasm was compared to rest (Fig. 3) and to imitation (Fig. 4). There was additional signicant deactivation in the dorsomedial prefrontal cortex (PFC), a region best known for its involvement in moral reasoning and social judgement (Mitchell et al., 2002). Activation of the dorsomedial PFC has been reported in at least two studies on human sexual arousal (Redoute et al., 2000; Arnow et al., 2002), perhaps reecting moral conicts when watching erotica in a scanner environment. The signicant deactivation of this region during orgasm implies absence of moral judgement and selfreferential thought. Although the left lateral OFC might be key in the release of orgasm, the dorsomedial PFC may also modulate sexual behaviour. During orgasm, blood ow was profoundly decreased in the temporal lobe, especially when compared to imitation. This decrease was most pronounced in the left temporal lobe, resembling that of orgasm < rest. Far less widespread rCBF decreases in the left temporal lobe resulted from orgasm < stimulation. As mentioned earlier, increased sexual arousal has consistently been associated with decreased activition of the temporal lobe (Stoleru et al., 1999; Redoute et al., 2000; Bocher et al., 2001; Arnow et al., 2002; Montorsi et al., 2003) and in the present study sexual arousal was higher during stimulation than during imitation. Thus, temporal lobe deactivation was most profound in the comparison with the greatest difference in sexual arousal, orgasm < imitation. From these control comparisons a very interesting distinction emerged regarding the rCBF decrease during orgasm: rCBF decreases were predominantly prefrontal for the sexual arousal control comparison (orgasm < stimulation) and mainly temporal for the motor output control comparison (orgasm < imitation). These control comparisons provide evidence that the temporal lobe deactivation is related to the level of sexual arousal, but that the prefrontal cortex deactivation causes the behavioural release that characterises orgasm. Regression analyses Rectal pressure variability RPstd reects activity of muscles in the rectal vicinity. RPstd correlated positively with rCBF in the left deep cerebellar nuclei, conrming the results from the t-contrasts that this region is related to axial motor output. However, because RPstd values were highest during orgasm, the left deep cerebellar nuclei may be involved in orgasm-specic pelvic muscular contractions. This result is supported by the rCBF response plot for the left deep cerebellar nuclei (Fig. 2B), which mirrors the RPstd during these conditions. Also, we found increased rCBF in this part of the cerebellum during ejaculation and orgasm in men (Holstege et al., 2003). Similar to all t-contrasts involving orgasm, there was much more profound negative than positive perfusion change. These negative correlations covered large areas of the frontal and temporal lobes. Perceived sexual arousal When we explored the brain to investigate positive correlations between rCBF and PSA, subsignicant correlations were found in the medial ventral midbrain and the caudate nucleus. Because these regions were included in our a priori hypothesis we performed small-volume correction, which resulted in signicant corrected cluster P-values for both clusters. The ventral tegmental area, which contains the A10 dopaminergic cell group, is located in the ventral midbrain. This cell group plays a crucial role in a wide range of rewarding behaviours (McBride et al., 1999), including euphoric states induced by drugs of abuse (Breiter et al., 1997; Sell et al., 1999), pleasurable music (Blood & Zatorre, 2001) and even eating chocolate (Small et al., 2001). In the present study PSA was highest during orgasm and because PSA correlated positively with rCBF in the ventral midbrain, this nding underlines the reinforcing nature of orgasm in women. While dopaminergic projections from the A10 cell group might be important for the reward aspect of orgasm in women, the nigrostriatal dopaminergic system has been linked to sexual behaviour itself (Hull et al., 1986). This might explain the positive correlation between PSA and rCBF in the right caudate nucleus. Highly signicant and widespread negative correlations between PSA and rCBF were found in both temporal lobes, again conrming that temporal lobe deactivation is related to increased sexual arousal. Cortical deactivation during orgasm and, in the case of the temporal lobe, a negative correlation between perfusion and sexual excitement appear to be involved in autoerotic asphyxiation. This is a sexual practice where the blood supply to the brain is restricted in order to achieve a level of orgasmic euphoria that has been described as addictive (Sinn, 1993).

Limitations and considerations In this study the subjects report of orgasm was crucial, leaving the possibility for false orgasm reports. However, the subjects knew that the rectal pressure measurements were visible for the experimenters throughout the experiment, which may have discouraged them from reporting falsely. Two reported orgasms were rejected on the basis of the rectal pressure data, because no obvious differences could be detected between rectal pressure patterns during reported orgasm and imitation. Therefore, the orgasms included in the present study are likely to be real. Orgasm is known to induce large increases in blood pressure, but we have no suspicion about artefactual activation in the anterior cerebellum, the only region with increased rCBF during orgasm. First, this same activation was also present during imitation of orgasm, a

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

Brain activation during clitoral stimulation and orgasm 3315 normotensive task, and second, the literature describes activations in this same area during voluntary motion of pelvic oor muscles. We have attributed the profound orbitofrontal rCBF decrease to behavioural disinhibition. It should be noted that we collected no measure of cognitive control other than the informal reports of the subjects. The hypothalamus is well known for its role in female reproductive behaviour (Pfaff et al., 1994) and during orgasm it may release oxytocin (Carmichael et al., 1994) and especially prolactin (Exton et al., 1999). However, no sign of hypothalamic involvement was observed in the present study. This absence can be explained in several ways: (i) synthesis of these hormones is unaltered and release requires no metabolic increase and hence no rCBF increase; (ii) the temporal resolution of our PET experiment may have been too limited to detect short-lasting events occurring in the hypothalamus; (iii) the results from Komisaruk et al. (2004) suggests that vaginocervical stimulation may be more effective in activating the hypothalamus; (iv) there was high intersubject variability in orgasmic RP patterns and, although no objective proof exists for orgasmic typology (Mah & Binik, 2001), these different RP patterns might still reect a qualitative difference in the orgasmic experience. Together this may be an explanation for the absence of activity in the hypothalamus. Conclusion We conclude that decreased rCBF in the prefrontal cortex and left lemporal lobe is the main feature of orgasm in women. In particular, activation in the left lateral OFC seems to reect the level of behavioural inhibition during sexual behaviour. The deactivation of the temporal lobe is directly related to the level of sexual arousal. As the regression analyses revealed, the deep cerebellar nuclei appear to be involved in orgasm-specic muscle contractions while the ventral midbrain and right caudate nucleus suggest a role for dopamine in orgasm. This novel nding might have important implications for the treatment of anorgasmia, a prevalent female sexual disorder which probably originates in the brain. Epilogue In French, orgasm is called Le Petit Mort, or Little Death, refering to the experience of timelessnes and release. To some degree, the present results seem to be in accordance with this notion, because female orgasm is associated with decreased blood ow in the orbitofrontal cortex, a part of the brain that is crucial for behavioural control.
Aloni, R. & Katz, S. (1999) A review of the effect of traumatic brain injury on the human sexual response. Brain Inj., 13, 269280. Arnow, B.A., Desmond, J.E., Banner, L.L., Glover, G.H., Solomon, A., Polan, M.L., Lue, T.F. & Atlas, S.W. (2002) Brain activation and sexual arousal in healthy, heterosexual males. Brain, 125, 10141023. Baird, A.D., Wilson, S.J., Bladin, P.F., Saling, M.M. & Reutens, D.C. (2002) Hypersexuality after temporal lobe resection. Epilepsy Behav., 3, 173181. Beauregard, M., Levesque, J. & Bourgouin, P. (2001) Neural correlates of conscious self-regulation of emotion. J. Neurosci., 21, RC165. Blok, B.F., Sturms, L.M. & Holstege, G. (1997) A PET study on cortical and subcortical control of pelvic oor musculature in women. J. Comp. Neurol., 389, 535544. Blood, A.J. & Zatorre, R.J. (2001) Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion. Proc. Natl Acad. Sci. USA, 98, 1181811823. Bocher, M., Chisin, R., Parag, Y., Freedman, N., Meir Weil, Y., Lester, H., Mishani, E. & Bonne, O. (2001) Cerebral activation associated with sexual arousal in response to a pornographic clip: a 15OH2O PET study in heterosexual men. Neuroimage, 14, 105117. Bohlen, J.G., Held, J.P. & Sanderson, M.O. (1980) The male orgasm: pelvic contractions measured by anal probe. Arch. Sex Behav., 9, 503521. Bohlen, J.G., Held, J.P., Sanderson, M.O. & Ahlgren, A. (1982) The female orgasm: pelvic contractions. Arch. Sex Behav., 11, 367386. Bradley, W.E., Farrell, D.F. & Ojemann, G.A. (1998) Human cerebrocortical potentials evoked by stimulation of the dorsal nerve of the penis. Somatosens. Mot. Res., 15, 118127. Breiter, H.C., Gollub, R.L., Weisskoff, R.M., Kennedy, D.N., Makris, N., Berke, J.D., Goodman, J.M., Kantor, H.L., Gastfriend, D.R., Riorden, J.P., Mathew, R.T., Rosen, B.R. & Hyman, S.E. (1997) Acute effects of cocaine on human brain activity and emotion. Neuron, 19, 591611. Carmichael, M.S., Humbert, R., Dixen, J., Palmisano, G., Greenleaf, W. & Davidson, J.M. (1987) Plasma oxytocin increases in the human sexual response. J. Clin. Endocrinol. Metab., 64, 2731. Carmichael, M.S., Warburton, V.L., Dixen, J. & Davidson, J.M. (1994) Relationships among cardiovascular, muscular, and oxytocin responses during human sexual activity. Arch. Sex Behav., 23, 5979. Cohen, H.D., Rosen, R.C. & Goldstein, L. (1976) Electroencephalographic laterality changes during human sexual orgasm. Arch. Sex Behav., 5, 189 199. Damasio, H., Grabowski, T., Frank, R., Galaburda, A.M. & Damasio, A.R. (1994) The return of Phineas Gage: clues about the brain from the skull of a famous patient. Science, 264, 11021105. Dunn, K.M., Cherkas, L.F. & Spector, T.D. (2005) Genetic inuences on variation in female orgasmic function: a twin study. Biol. Lett., 1, 14. Exton, M.S., Bindert, A., Kruger, T., Scheller, F., Hartmann, U. & Schedlowski, M. (1999) Cardiovascular and endocrine alterations after masturbationinduced orgasm in women. Psychosom. Med., 61, 280289. Fadul, C.E., Stommel, E.W., Dragnev, K.H., Eskey, C.J. & Dalmau, J.O. (2005) Focal paraneoplastic limbic encephalitis presenting as orgasmic epilepsy. J. Neurooncol., 72, 195198. Ferretti, A., Caulo, M., Del Gratta, C., Di Matteo, R., Merla, A., Montorsi, F., Pizzella, V., Pompa, P., Rigatti, P., Rossini, P.M., Salonia, A., Tartaro, A. & Romani, G.L. (2005) Dynamics of male sexual arousal: distinct components of brain activation revealed by fMRI. Neuroimage., 26, 10861096. Fox, C.A. (1976) Some aspects and implications of coital physiology. J. Sex Marital Ther., 2, 205213. Fox, C.A. & Fox, B. (1967) Uterine suction during orgasm. BMJ, 4, 300 301. Friston, K.J., Worsley, K.J., Holmes, A.P., Poline, J.-B., Frith, C.D. & Frackowiak, R.S. (1995) Statistical parametric mapping in functional imaging: a general linear approach. Hum. Brain Mapp., 2, 189210. Georgiadis, J.R. & Holstege, G. (2005) Human brain activation during sexual stimulation of the penis. J. Comp. Neurol., 493, 3338. Graber, B., Rohrbaugh, J.W., Newlin, D.B., Varner, J.L. & Ellingson, R.J. (1985) EEG during masturbation and ejaculation. Arch. Sex Behav., 14, 491 503. Gusnard, D.A. & Raichle, M.E. (2001) Searching for a baseline: functional imaging and the resting human brain. Nat. Rev. Neurosci., 2, 685 694. Holstege, G., Georgiadis, J.R., Paans, A.M., Meiners, L.C., van der Graaf, F.H. & Reinders, A.A. (2003) Brain activation during human male ejaculation. J. Neurosci., 23, 91859193. Hull, E.M., Bitran, D., Pehek, E.A., Warner, R.K., Band, L.C. & Holmes, G.M. (1986) Dopaminergic control of male sex behavior in rats: effects of an intracerebrally-infused agonist. Brain Res., 370, 7381.

Acknowledgements
The authors would like to thank the staff at the PET facilities for their kind cooperation. We would also like to thank Rob Meijer and Anita Kuiper for their help in performing this study.

Abbreviations
OFC, orbitofrontal cortex; PET, positron emission tomography; PFC, prefrontal cortex; PSA, perceived level of sexual arousal; PET, positron emission tomography; rCBF, regional cerebral blood ow; RP, rectal pressure; RPstd, rectal pressure standard deviation.

References
Aksaray, G., Yelken, B., Kaptanoglu, C., Ou, S. & Ozaltin, M. (2001) Sexuality in women with obsessive compulsive disorder. J. Sex Marital Ther., 27, 273277.

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316

3316 J. R. Georgiadis et al.

Janszky, J., Ebner, A., Szupera, Z., Schulz, R., Hollo, A., Szucs, A. & Clemens, B. (2004) Orgasmic aura a report of seven cases. Seizure., 13, 441444. Kell, C.A., von Kriegstein, K., Rosler, A., Kleinschmidt, A. & Laufs, H. (2005) The sensory cortical representation of the human penis: revisiting somatotopy in the male homunculus. J. Neurosci., 25, 59845987. Komisaruk, B.R., Whipple, B., Crawford, A., Liu, W.C., Kalnin, A. & Mosier, K. (2004) Brain activation during vaginocervical self-stimulation and orgasm in women with complete spinal cord injury: fMRI evidence of mediation by the vagus nerves. Brain Res., 1024, 7788. Lapierre, D., Braun, C.M. & Hodgins, S. (1995) Ventral frontal decits in psychopathy: neuropsychological test ndings. Neuropsychologia, 33, 139 151. Leutmezer, F., Serles, W., Bacher, J., Groppel, G., Pataraia, E., Aull, S., Olbrich, A., Czech, T. & Baumgartner, C. (1999) Genital automatisms in complex partial seizures. Neurology, 52, 11881191. Lloyd, E.A. (2005) The Case of the Female Orgasm: Bias in the Science of Evolution. Harvard University Press, Cambridge, MA. Lundberg, P.O. & Brattberg, A. (1992) Sexual dysfunction in selected neurologic disorders: hypothalamopituitary disorders, epilepsy, myelopathies, polyneuropathies, and sacral nerve lesions. Semin. Neurol., 12, 115 119. Mah, K. & Binik, Y.M. (2001) The nature of human orgasm: a critical review of major trends. Clin. Psychol. Rev., 21, 823856. Makela, J.P., Illman, M., Jousmaki, V., Numminen, J., Lehecka, M., Salenius, S., Forss, N. & Hari, R. (2003) Dorsal penile nerve stimulation elicits lefthemisphere dominant activation in the second somatosensory cortex. Hum. Brain Mapp., 18, 9099. Masters, W.H. & Johnson, V.E. (1966) Human Sexual Response. Little, Brown, Boston. McBride, W.J., Murphy, J.M. & Ikemoto, S. (1999) Localization of brain reinforcement mechanisms: intracranial self-administration and intracranial place-conditioning studies. Behav. Brain Res., 101, 129 152. Mitchell, J.P., Heatherton, T.F. & Macrae, C.N. (2002) Distinct neural systems subserve person and object knowledge. Proc. Natl Acad. Sci. USA, 99, 1523815243. Montorsi, F., Perani, D., Anchisi, D., Salonia, A., Scifo, P., Rigiroli, P., Deho, F., De Vito, M.L., Heaton, J., Rigatti, P. & Fazio, F. (2003) Brain activation patterns during video sexual stimulation following the administration of apomorphine: results of a placebo-controlled study. Eur. Urol., 43, 405411. Nakagawa, H., Namima, T., Aizawa, M., Uchi, K., Kaiho, Y., Yoshikawa, K., Orikasa, S. & Nakasato, N. (1998) Somatosensory evoked magnetic elds elicited by dorsal penile, posterior tibial and median nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 108, 5761. Peneld, W. & Rasmussen, T. (1950) The Cerebral Cortex of Man. A Clinical Study of Localization of Function. The Macmillan Comp., New York. Penhune, V.B. & Doyon, J. (2005) Cerebellum and M1 interaction during early learning of timed motor sequences. Neuroimage, 26, 801 812. Pfaff, D.W., Schwartz-Giblin, S., McCarthy, M.M. & Kow, L.-M. (1994) Cellular and molecular mechanisms of female reproductive behaviors. In: Knobil, E. & Neill, J.D. (eds), Physiology of Reproduction. Raven Press, New York, pp. 107220. Poulain, D.A. & Wakerley, J.B. (1982) Electrophysiology of hypothalamic magnocellular neurones secreting oxytocin and vasopressin. Neuroscience, 7, 773808. Pukall, C.F., Strigo, I.A., Binik, Y.M., Amsel, R., Khalife, S. & Bushnell, M.C. (2005) Neural correlates of painful genital touch in women with vulvar vestibulitis syndrome. Pain, 115, 118127. Quadri, S.K., Norman, R.L. & Spies, H.G. (1977) Prolactin release following electrical stimulation of the brain in ovarectomized and ovariectomized estrogen-treated rhesus monkeys. Endocrinology, 100, 325330. Raichle, M.E., MacLeod, A.M., Snyder, A.Z., Powers, W.J., Gusnard, D.A. & Shulman, G.L. (2001) A default mode of brain function. Proc. Natl Acad. Sci. USA, 98, 676682. Redoute, J., Stoleru, S., Gregoire, M.C., Costes, N., Cinotti, L., Lavenne, F., Le Bars, D., Forest, M.G. & Pujol, J.F. (2000) Brain processing of visual sexual stimuli in human males. Hum. Brain Mapp., 11, 162177. Redoute, J., Stoleru, S., Pugeat, M., Costes, N., Lavenne, F., Le Bars, D., Dechaud, H., Cinotti, L. & Pujol, J.F. (2005) Brain processing of visual sexual stimuli in treated and untreated hypogonadal patients. Psychoneuroendocrinology, 30, 461482. Reinders, A.A.T.S., Willemsen, A.T.M., Georgiadis, J.R., Hovius, M., Paans, A.M.J. & den Boer, J.A. (2002) Interscan displacement-induced variance in PET activation data is excluded by a scan-specic attenuation correction. Neuroimage, 17, 18441853. Saxena, S. & Rauch, S.L. (2000) Functional neuroimaging and the neuroanatomy of obsessive-compulsive disorder. Psychiatr. Clin. North Am., 23, 563586. Schmahmann, J.D. (2004) Disorders of the cerebellum: ataxia, dysmetria of thought, and the cerebellar cognitive affective syndrome. J. Neuropsychiatry Clin. Neurosci., 16, 367378. Schmahmann, J.D., Doyon, J., McDonald, D., Holmes, C., Lavoie, K., Hurwitz, A.S., Kabani, N., Toga, A., Evans, A. & Petrides, M. (1999) Threedimensional MRI atlas of the human cerebellum in proportional stereotaxic space. Neuroimage, 10, 233260. Sell, L.A., Morris, J., Bearn, J., Frackowiak, R.S., Friston, K.J. & Dolan, R.J. (1999) Activation of reward circuitry in human opiate addicts. Eur. J. Neurosci., 11, 10421048. Seseke, S., Baudewig, J., Kallenberg, K., Ringert, R.H., Seseke, F. & Dechent, P. (2006) Voluntary pelvic oor muscle control-an fMRI study. Neuroimage, 31, 13991407. Sholty, M.J., Ephross, P.H., Plaut, S.M., Fischman, S.H., Charnas, J.F. & Cody, C.A. (1984) Female orgasmic experience: a subjective study. Arch. Sex Behav., 13, 155164. Sinn, L.E. (1993) The silver bullet. Am. J. Forensic Med. Pathol., 14, 145147. Small, D.M., Zatorre, R.J., Dagher, A., Evans, A.C. & Jones-Gotman, M. (2001) Changes in brain activity related to eating chocolate: from pleasure to aversion. Brain, 124, 17201733. Stoleru, S., Gregoire, M.C., Gerard, D., Decety, J., Lafarge, E., Cinotti, L., Lavenne, F., Le Bars, D., Vernet-Maury, E., Rada, H., Collet, C., Mazoyer, B., Forest, M.G., Magnin, F., Spira, A. & Comar, D. (1999) Neuroanatomical correlates of visually evoked sexual arousal in human males. Arch. Sex Behav., 28, 121. Stoleru, S., Redoute, J., Costes, N., Lavenne, F., Bars, D.L., Dechaud, H., Forest, M.G., Pugeat, M., Cinotti, L. & Pujol, J.F. (2003) Brain processing of visual sexual stimuli in men with hypoactive sexual desire disorder. Psychiatry Res., 124, 6786. Talairach, J. & Tournoux, P. (1988) Co-Planar Atlas of the Human Brain. Thieme Verlag, New York. Tiihonen, J., Kuikka, J., Kupila, J., Partanen, K., Vainio, P., Airaksinen, J., Eronen, M., Hallikainen, T., Paanila, J. & Kinnunen, I.& (1994) Increase in cerebral blood ow of right prefrontal cortex in man during orgasm. Neurosci. Lett., 170, 241243. Yang, Y., Raine, A., Lencz, T., Bihrle, S., LaCasse, L. & Colletti, P. (2005) Volume reduction in prefrontal gray matter in unsuccessful criminal psychopaths. Biol. Psychiatry, 57, 11031108. Young, L.J. & Wang, Z. (2004) The neurobiology of pair bonding. Nat. Neurosci., 7, 10481054.

The Authors (2006). Journal Compilation Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 24, 33053316