Rather than lesioning or electrically stimulating parts of the brain or implanting gonadal hormones with the purpose of determining the involvement of these parts in the control of sexual behavior, more indirect procedures can be employed. Some years ago it became a kind of fashion to study the expression of the c-fos gene after exposing animals to sexually relevant events, such as the odor or the sight of a conspecific of the opposite sex, or allowing the subjects different amounts of sexual activity, ranging from performing or receiving one mount to several ejaculations. There is, as I mentioned somewhere, abundant evidence showing that the immediate early gene c-fos is activated whenever neural activity is heightened.
Enhanced expression of the c-fos gene product is, then, an indicator of neural activity. Results from the large amount of c-fos studies have essentially confirmed what we already knew, perhaps adding some detail here and there. Among the areas activated by sexual activity or sexually relevant stimuli are the olfactory bulbs, different parts of the amygdala, the bed nucleus of the stria terminalis, the medial preoptic area and many other sites (see e.g. Dudley et al., 1992; Wersinger et al., 1993; Bressler and Baum, 1996; Erskine and Hanrahan, 1997; Greco et al., 1998). As we will see when we turn to the female, sex-induced fos expression in a structure does not necessarily mean that the structure is important for sexual behavior, so the functional as opposed to statistical significance of the fos data needs to be established by other methods, such as lesion or stimulation.
We have seen many examples of the usefulness of studying changes in genital blood flow in response to sexual incentives. Instead of that, some researchers have preferred to study changes in blood flow in the brain during exposure to such incentives. Changes in blood flow or in the oxygenation levels of hemoglobin are supposed to depend on altered nervous activity. In the marmoset, exposure to the odor of periovulatory females increased the signal from the preoptic area and the anterior hypothalamus more than exposure to the odor of ovariectomized females did. Furthermore, while the response to the periovulatory female odor outlasted the presence of the odor, the small response to the odor of ovariectomized female disappeared as soon as the odor was withdrawn (Ferris et al., 2001). This study confirms that the preoptic area is activated by sexually relevant stimuli. Data from the human, strangely enough, do not systematically do so. Young men watching a pornographic movie while inside a magnetic resonance imaging machine showed enhanced activation in the inferior frontal lobe, cingulate and insular gyri, corpus callosum, thalamus, caudate nucleus, globus pallidus and the inferior temporal lobe. No such activation was found while watching a neutral movie. Not unexpectedly, subjects displayed activation of the occipital lobe when watching the pornographic as well as the neutral movie (Park et al., 2001b). An interesting detail in this study is that penile erection was quantified during exposure to the movies in order to assure that the pornographic movie indeed succeeded in enhancing sexual arousal. It did. However, two hypogonadal men failed to respond with erection to that movie. They also showed less activation of the relevant brain sites than the responders did. Treatment with testosterone brought their level of brain activation as well as of erection in response to the pornographic movie up to that of the other men. Although data from two men are far from sufficient for a conclusion, it might be tentatively suggested that the pattern of brain activation observed is dependent on gonadal hormones. The fact that testosterone treatment of these hypogonadal men enhanced activation indicates certainly that it is androgen dependent, exactly as any human male sexual response should be. Again, this proposal goes far beyond what can be ascertained by data from two men, but it should at least be considered as worthwhile of further investigation.
A somewhat more recent study went even further and correlated the degree of erection with the hemodynamic brain response. The anterior cingulate and insular cortices, amygdala and secondary somatosensory cortices showed activation associated with sustained erection. The data from the hypothalamus were unclear. There was some activation at the onset of erection, but there was no difference in signal between no erection and sustained erection (Ferretti et al., 2005). Erection was also monitored in another imaging study employing pornographic movies. The neutral movie used as control depicted sports. An overall comparison of activation between the pornographic and sports movies failed to detect any difference, but when erection was included as a regressor in the analysis, activation was found in the insular cortex and in the middle occipital and temporal gyri as well as in the cingulum, caudate-putamen and hypothalamus. All these data suggest that hemodynamic changes in response to visual sexual incentives are reliably obtained in the cingular and insular cortices. The hypothalamus, and by extension the pre-optic area, is sometimes included among the sites responding to visual sexual stimulation. However, a study employing mechanical stimulation of the penis (masturbation) rather than pornographic movies failed to detect activity changes in the hypothalamus/preoptic area. The only structures showing enhanced blood flow were the insular cortex and the secondary somatosensory cortex (Georgiadis and Holstege, 2005).
While experimental studies in non-human animals uniformly have demonstrated the fundamental role of the preoptic area for male sexual behavior, the human imaging studies have only rarely found activation of that area during sexual arousal. In contrast, the only non-human primate imaging study I know of, performed in the marmoset, did, as I mentioned above, find activation of the pre-optic area in response to sexually relevant odors. How can we explain the annoying discrepancy between non-human and human studies? One possibility that occurs to me is that the preoptic area, and its human equivalent, is a rather small structure compared to the cortical areas usually found lit up in imaging studies. It is likely that any signal generated in the small preoptic area is smaller than signals generated in the large cortical areas. This small signal may be lost when artifacts caused by minute head movements or other factors are eliminated. The marmoset study, showing preoptic area activation, employed a head-restraining device and a very strong magnetic field (9.5 tesla compared to fields of 1.5-2 tesla used in the human studies). A consequence of these differences may be that movement artifacts were smaller and resolution higher. This may have allowed for the detection of preoptic area activation. I recognize that this explanation is pure speculation, but I find it most unlikely that the human should be fundamentally different from other primates. This belief forces me to find an explanation that can convince myself, at least.
To my consolation, a review paper published after I wrote the above paragraph supports my speculation. When discussing the failure of many imaging studies to find activity in the preoptic area during sexual arousal, it is stated that the very low spatial extension of this region and the relatively low spatial resolution of neu-roimaging techniques may be the cause of this failure (Mouras, 2006). A few studies purportedly finding activation of the hypothalamus are cited in the Mouras
(2006) paper in order to make the human imaging data consistent with the large amount of non-human animal data that have firmly established a role for the pre-optic area. However, one of these studies found low, unilateral hypothalamic activation in response to a pornographic movie, while far larger activations were observed in several areas of the cerebral cortex (Arnow et al., 2002). Moreover, what is not mentioned in the review is that other recent studies fail to find hypo-thalamic response to sexual stimuli causing erection (Georgiadis and Holstege, 2005) or find only a transient response (Ferretti et al., 2005) as discussed above.
I have not reviewed all the published imaging studies. Rather than doing that, I have essentially limited the discussion to those including an objective measure of sexual arousal, erection. This seems essential if we are to talk about changes in brain activity caused by sexual arousal and not by stimuli that some researcher imagines as sexually arousing. In conclusion, it seems that imaging studies of sexual arousal only coincide with regard to enhanced activity in the insular cortex. In all studies employing visual stimuli, the cingular cortex is also activated. The meaning or interest of these observations remains obscure for the moment.
Was this article helpful?