James V. Kohl

The Mind's Eyes: Human Pheromones, Neuroscience,
and Male Sexual Preferences

Originally published in: Journal of Psychology & Human Sexuality. 18;4: 313-369. 2006.
Reproduced here by permission of the author.
Author Bio: James V. Kohl is a clinical laboratory scientist and independent researcher and Director of Research and Development for Stone Independent Research, Inc., Phoenix, New York.  Please address correspondence to jvkohl@bellsouth.net



The across-species genetic conservation of intercellular and extracellular chemical communication enables unicellular and multicellular organisms to functionally distinguish between self and non-self.  Non-self olfactory/pheromonal input from the social environment elicits a vertebrate neuroendocrine response.  The organization and activation of this neuroendocrine response modulates the concurrent maturation of the mammalian neuroendocrine system, the reproductive system, and the central nervous system during the development of sexual preferences that may be expressed in sexual behavior.  Psychophysiological mechanisms for the development of these sexual preferences include focus on unconscious affects that are detailed in reciprocal cause and effect relationships.  Olfactory/pheromonal conditioning elicits neuroendocrine effects accompanied by unconscious affects on the development of sexual preferences.  Integrating these unconscious affects extends to humans a developmental model of behavior that includes the development of male sexual preferences for other males.

KEY WORDS: Developmental model, olfaction, pheromones, sexual preferences, homosexual

“A homogeneity of sexual preferences inevitably masks a heterogeneity of desired

emotional productions . . . . [E]ven the most conventional may find their sources of sexual excitement fueled by the slightest ‘whiff’ of the unthinkable.”

William Simon (1994)


    Human physical attraction may not cognitively equate with definitive indications of sexual preferences or definitive sexual behavior, because sexual preferences can be cognitively denied and sexual behavior can be suppressed.  Thus, when comparing human and non-human animal behavior, it is difficult to separate cognitive effects, like thoughts, from unconscious affects, like neuroendocrine changes, that may be manifest as human emotions.  Therefore, sex researchers cannot be sure whether they are sampling some vague unconscious affect of human behavior that is not cognitively considered by their subjects—and not considered in the study design or the data analysis.

    The failure to fully consider unconscious affects that may be manifest as opportunistic behavioral tendencies leads to a lack of clear findings, and the underlying biological underpinnings of sexual preferences can readily be missed.  Accordingly, the reporting of incongruous and ill-defined results that are, nonetheless, meaningfully interpreted, is problematic.  Meaningful results require an important consideration in any scientific endeavor; we must first get the model right!

    No model is consistently used in the scientific study of human sexuality.  For example, the effect of auditory stimuli in songbirds or visual stimuli like the colorful plumage of the peacock’s tail are used as examples of sensory input from the social environment that somehow influences sexual behavior in some species.  In contrast, the effect of olfactory/pheromonal input from the social environment is more typically used as an example of sensory input that influences levels of hormones and sexual behavior in mammals.

    A consistent mammalian model that links olfactory/pheromonal input from the social environment to hormonal influences on sexual behavior should help to reduce disparate findings and debate over inconsistent results from studies of human sexual behavior.  Currently, disparate findings and debate tend to weakly support a false nature versus nurture dichotomy.  This false dichotomy might well be eliminated from further consideration if individual studies began to more fully address a causal link between nature and nurture.  Such a link can be addressed within the context of a developmental model of how olfactory/pheromonal input influences sexual preferences and how these sexual preferences influence sexual behavior.

    In non-human animals, a causal relationship must exist among the development of sexual preferences for attractive physical features and how these preferences are manifest in sexual behavior.  This causal relationship must develop before sexual preferences or sexual behaviors are expressed.  Whether or not it is acknowledged, such a causal relationship appears to exist before human sexual preferences are fully developed and long before adult sexual behavior is expressed.  Extension to humans of the mammalian olfactory/pheromonal model presented here addresses a causal relationship that includes the unconscious affect of olfactory/pheromonal input from the social environment on hormones and the development of sexual preferences manifest in the expression of sexual behavior.


    A general rule for any model that includes the influence of hormones on behavior is that neuroendocrine responses (i.e., from hormone-secreting nerve cells) must first be elicited by sensory input from the social environment.  Mammalian pheromones are olfactory input from the social environment that elicit a typical neuroendocrine response.  This typical neuroendocrine response to mammalian pheromones correlates with unconscious affects on levels of other hormones during the development of mammalian male and female sexual preferences.

    Kohl, Atzmueller, Fink, and Grammer (2001) have integrated research findings that link mammalian pheromones to unconscious affects on human behavior.  They proposed that sexual preferences for attractive human physical features are readily influenced by the effect of pheromones on gonadotropin releasing hormone (GnRH)-directed, gonadotropin-modulated androgenic and estrogenic changes in the brain.  Because androgenic and estrogenic changes in the brain are not consciously perceived, the hormonal changes associated with pheromones and their effect on GnRH are considered to be unconscious affects.  These unconscious affects supposedly occur in the brain during conditioning of the human visual response to olfactory/pheromonal input.  Thus, hormone-dependent putative human pheromone production and hormone-dependent pheromone distribution from conspecifics (other members of the species) are what condition sexual preferences for hormone-associated characteristics, such as facial symmetry, facial attractiveness, waist-to-hip ratio, and most, if not all, hormone-associated physical features.

    Because estrogen/androgen ratios vary in males and females, androgenic and estrogenic hormone-dependent physical features are considered to be sexually dimorphic physical features, and estrogenic or androgenic pheromone production is considered to be sexually dimorphic.  Conditioning of a sexually dimorphic human visual response to sexually dimorphic pheromone production and distribution explains how psychophysiological effects of sexually explicit visual stimuli elicit elevations in luteinizing hormone (LH) and/or testosterone (T) in men (Hellhammer, Hubert, & Schurmeyer, 1985; LaFerla, Anderson, & Schalch 1978; Redoute, Stoleru, Gregoire, Costes, Cinotti, Lavenne et al., 2000; Stoleru, Ennaji, Cournot, & Spira, 1993).  Conditioning of the sexually dimorphic human visual response to pheromones also incorporates the “learned sexuality” motivational theory of behavior  (Woodson, 2002).

    In the olfactory/pheromonal model presented here, the learned sexuality motivational theory of behavior accounts for the species-specific stimulus involved—namely, putative human pheromones and their unconscious affects on the development of sexual preferences.  Even in the absence of cognitive assessment and, clearly in the absence of intent, the neuroendocrine response elicited by putative human pheromones could be important in the development of non-cognitively or cognitively based sexual preferences that are associated with other sensory input, like visual input from the social environment.  Succinctly stated, Kohl et al. (2001) have posited that mammalian pheromones, including putative human pheromones, elicit unconscious affects via hormonal changes and that these hormonal changes are largely responsible for the unintentional development of sexual preferences and sexual attraction, which are associated with, if not always manifest in, sexual behavior.

    Unconscious affects of pheromones on hormone al changes might help to explain the development of a male sexual preference for the natural body odor of other males.  Accordingly, GnRH-directed, gonadotropin-modulated, androgenic and estrogenic changes, which occur in the brain during conditioning of the human visual response to olfactory/pheromonal input, may extend to human males a mammalian olfactory/pheromonal model for the development of a variety of sexual preferences.  A GnRH-directed mammalian olfactory/pheromonal model for the development of sexual preferences also meets other conditional requirements for a model that links the social environment to the genetic underpinnings of human sexual behavior.  In addition to abiding by the general rule that neuroendocrine responses must first be elicited by sensory input from the social environment, this model incorporates a logical pathway that directly links sensory input from the social environmental to the development of sexual preferences and to the development of sexual behavior.

    The five-step logical pathway that directly links sensory input from the social environment to the development of sexual preferences and sexual behavior is gene→cell→tissue→organ→organ system.  Failure to incorporate any of the steps in this pathway makes it impossible to directly connect to sexual preferences and sexual behavior, a logical sequence of events that must somehow and somewhere begin with the effect of sensory input from the social environment.  For example, activation of tissue in and of itself does not detail what genes and in what cells of the tissue are activated by the sensory input from the social environment.  And, merely stating that the brain, or a part of the brain, dictates or influences behavior neither details the organ system that is participating nor does it detail how this organ system develops and coordinates its genetically-directed function.  In contrast, mammalian olfactory/pheromonal input activates early gene expression in hormone-secreting cells of tissue in the brain, which is an organ that is part of an organ system involved in the development of sexual preferences and sexual behavior.  This five-step pathway also has its molecular biochemical origin in unicellular organisms.


    The molecular biochemical elements of intercellular communication systems are phylogenetically conserved throughout the plant and animal kingdomsFindings from unicellular organisms that multiply through asexual or sexual reproduction suggest that the origins of vertebrate intercellular communication evolved from a system of intercellular communication in unicellular organisms (LeRoith, Shemer, & Roberts, 1992)

    Intercellular communication systems in any organism or any species are self-limiting because communication occurs within genetically similar individuals.  Communication among other members of the same species requires the presence of a non-self signal from the social environment, which includes conspecifics.

    The social environment of bacteria contains chemical signals that enable the species survival mechanism of quorum sensing.  Quorum sensing incorporates mechanisms that allow bacterium living in the presence of other bacteria to distinguish self from a collective non-self.  An example of quorum sensing is when colonization ceases if a colony becomes too large to sustain itself on the available resources of food and water that are required to satisfy bacterial “hunger” and “thirst.”  Molecular biochemical mechanisms of bacterial hunger and thirst have biological and, therefore, evolutionary origins.  Chemical communication is sufficient to drive the self-serving mechanistic behaviors that are required for individual survival and for species survival.  Chemical signals from the social environment provide the appropriate cues for successful bacterial colonization.

    Chemical signals from the social environment, which influence the sexual preferences and sexual behavior of conspecifics, characterize a focal change from self-serving mechanistic survival behaviors to more highly developed social behaviors.  These more highly developed social behaviors benefit the species through the interaction of unicellular organisms that are individually, rather than collectively, able to distinguish self from non-self.

    The unicellular organisms of brewers’ yeast (Saccharomyces cerevisiae) reproduce asexually and sexually.  Sexual reproduction requires yeast alpha mating factor.  The yeast alpha mating factor is part of a highly conserved intercellular chemical communication system, which functionally depends on an early form of the mammalian decapeptide, GnRH.  In mammals, GnRH, elicits a pituitary LH response.

    Conservation of the intercellular GnRH-directed, LH-facilitated chemical communication system across species and its extension to function in an extracellular chemical communication system among different species are exemplified by the fact that the yeast alpha mating factor binds to GnRH receptors in cultured cells from the pituitary gland of a multicellular mammal: the rat.  The binding of natural yeast alpha mating factor or synthetic yeast alpha mating factor to pituitary GnRH receptors causes the release of LH (Loumaye, Thorner, & Catt, 1982).

    Conservation of another intercellular chemical communication system across species and its extension to function as an extracellular chemical communication system among species also occur in brewer’s yeast mating types.  Several studies suggest that the polymorphic Major Histocompatibility Complex (MHC) locus of genes in non-human animals—referred to as the Human Leukocyte Antigen (HLA) locus of genes—is conserved by natural selection and by sexual selection and enabled by the yeast alpha mating factor.

    MHC class I molecules are cell surface glycoproteins that trigger and mediate immune responses by functioning as receptors for peptide antigens, like GnRH.  The MHC/HLA locus of genes provides chemical cues of immunological function and is believed to be responsible for discrimination of self and non-self within the immune system.  Nearly all somatic tissues express these molecules, although levels of expression vary considerably among tissues.  In lay terminology, genes at the MHC/HLA locus are said to determine “tissue type.”  The recipient of an organ transplant may reject any organ from a donor whose genes at the MHC/HLA locus are too dissimilar.

    Unlike the well-conserved mammalian decapeptide GnRH, the protein products that the MHC/HLA genes encode have completely diverged from those of the yeast mating-type genes to those of a mammal: the pig.  However, despite this divergence, the molecular biochemical mechanisms that regulate MHC expression have been conserved.  For example, in pigs, MHC genes encode a MHC antigen that is involved in self and non-self recognition.  In yeast, similar regulatory genes determine mating type and mating-type recognition—a self and non-self recognition system (Weissman & Singer, 1991).

    Concisely put, the across-species conservation of a neuroendocrine chemical communication system and an immune chemical communication system enable genetic diversity in many organisms that functionally distinguish between self and non-self.  This duality in the conservation of neuroendocrine system and immune system chemical communication enables mammals to respond to immunological non-self signals with a GnRH-directed LH (i.e., neuroendocrine) response, which facilitates mammalian sexual reproduction.


    The alpha mating factor of brewers’ yeast is commonly referred to as a yeast mating pheromone.  The traditional definition of a pheromone originated with Karlson and Luscher (1959) who wrote: “Pheromones are defined as substances which are secreted to the outside by an individual and received by a second individual of the same species, in which they release a specific reaction, for example, a definite behaviour, or a developmental process (p. 55).”  The yeast mating pheromone signals a reproductive state and causes a reaction that enables extended contact between two cells.  This reaction facilitates the exchange of genetic material.  In mammals, the chemical signal-facilitated fusion of two haploid cells, ovum and spermatozoa, allows for the same result.

    Activities like signaling and enabling extended contact/fusion may not appear to address any specific behavioral component that fit the traditional definition of a pheromone.  Clearly, however, the original definition of a pheromone required that a means for communication among conspecifics be established and maintained by natural selection.  In yeast, this occurred through sexual selection.  Sexual selection requires that communication contribute to the evolutionary “fitness” of both the sender and the receiver (Meredith, 2001).  The yeast mating pheromone—an early form of mammalian GnRH—signals and enables sexual reproduction and contributes to the evolutionary fitness of the sender and receiver by helping to ensure genetic diversity through self/non-self signals and by appropriate responses to non-self signals.

    From an evolutionary perspective, it is apparent that the molecular biochemical properties of the intercellular and extracellular chemical signal, GnRH, and its receptor must continue to contribute to the across-species fitness of the sender and the receiver.  Genetic conservation of these signaling mechanisms across-species is apparent.  For example, the GnRH molecule in coral stimulates in vitro LH release from the pituitary cells of fish.  Coral and fish have cells that are differentiated into tissues and organs.  Additional data from studies of coral strongly suggest that the phylogenetically complex mammalian GnRH-directed hypothalamic-gonadal-sex hormone pathway developed early in evolution from invertebrates.  Reports of the direct action of GnRH on gonadal hormone activity represent the role of GnRH and its receptor in the control of reproduction with the emergence of vertebrates (Twan, Hwang, Lee, Jeng, Yueh, Tung et al., 2006).  Long before the appearance of the pituitary as a relay, long before the emergence of vertebrates, and long before the emergence of the mammalian hypothalamic-pituitary-gonadal (HPG) or hypothalamic-pituitary-adrenal (HPA) axes, it was the GnRH and its receptor that played the primary role in sexual reproduction.

    Barran and colleagues (2005) have reviewed the evolutionary significance of the GnRH molecule and receptor signaling from protochordates to vertebrates.  Though its evolutionary origin is in unicellular organisms, GnRH became the biological core of vertebrate reproduction.  The mammalian form of GnRH, which is found in humans, has a wide distribution in vertebrate species.  This form of GnRH can be detected in primitive bony fish, but not in species that are thought to have evolved earlier.  Thus, the human form of GnRH arose some 400 million years ago (Sherwood, Lovejoy, & Coe, 1993), although its evolutionary origins date back much further, to unicellular organisms like yeast.

    In biologically based evolutionary theory, the more closely that natural selection guards a particular gene through time, the more important is its function.  The gene for mammalian GnRH is unequivocally required for pituitary LH release (Hoffman, Lee, Attardi, Yann, & Fitzsimmons, 1990).  And, the “phenomenon” of LH release in response to contact with olfactory/pheromonal input from mammalian conspecifics is well known (Meredith & Fernandez-Fewell, 1994).

    The overall influence of mammalian olfactory/pheromonal input on hypothalamic GnRH release and on LH release from the pituitary represent a highly conserved effector system that includes genes that code for the peptide ligand (i.e., GnRH) and the cell-surface receptor: the GnRH receptor (GnRHR).  The result of this genetic conservation of GnRH and its receptor is the physiological regulation of sexual reproduction across species.  The transition from asexual to sexual reproduction involves two organisms that incorporate the equally well-conserved GnRH-directed, GnRHR-enabled neuroendocrine system signaling mechanism and the MHC/HLA-directed immune system signaling mechanism.  These signaling mechanisms are involved in the chemical discrimination of self from non-self, genetic diversification, and the neuroendocrine response to non-self chemical signals, like olfactory/pheromonal input from the social environment.


    Mammals use the olfactory/pheromonal chemical signals in natural body odor to discriminate between self and non-self.  In rodents, natural body odor is significantly influenced by the MHC and can be discriminated by members of different species (Brennan, 2004; Wysocki, Yamazaki, Curran, Wysocki, & Beauchamp, 2004).  In humans, the HLA influences both natural body odors and preferences for natural body odors.  Women rate natural body odors from men with differing HLA as more pleasant than natural body odors from men with similar HLA’s.  Further, scorings of the pleasantness of natural body odors are correlated negatively with the degree of genetic similarity (Wedekind, Seebeck, Bettens, & Paepke, 1995; Wedekind & Furi, 1997).  Simply put, women sometimes prefer the scent of genetic diversity.

    In an early leap from MHC/HLA-related molecular biochemical reactions to how olfactory/pheromonal input might influence behavior, Ober reportedly said that “. . . sweat and breast milk could be odor cues governing behavior” (Choosing the perfect mate, 1993, p. 22A).  Four years later, Ober, Weitkamp, Cox, Dytch, Kostyu, and Elias (1997) published findings showing that isolated human populations avoid mating with individuals with same HLA haplotypes more than expected by chance.  This mating avoidance is considered evidence that HLA-linked genes influence human sexual preferences.  Moreover, it is apparent that MHC/HLA genes elicit unconscious affects that are manifest as preferences for natural body odors that vary by the degree of MHC/HLA similarity.  Indeed, MHC/HLA proteins may be expressed as pheromones that elicit unconscious affects on sexual preferences and on sexual behavior.

    Shepherd (2006) has summarized recent olfactory/pheromonal research findings and clarified concerns with regards to an ongoing debate about the presence or absence of a functional human vomeronasal organ (VNO).  Both the main olfactory pathway and VNO-enabled accessory olfactory pathway process pheromones, so there is no need for a human VNO.  Further, he stated that “We have much more to learn about how intimately neuroendocrine functions, controlled by pheromones, acting through our noses, interact with other operations within the brain to control human behaviour and cognition (p. 151).”  Terms like “odor cues governing behaviour” (Choosing the perfect mate, 1993) or neuroendocrine functions “controlled by pheromones” that may “control human behaviour and cognition” (Shepherd, 2006), when referring to the role of olfactory/pheromonal sensory input in human behavior and cognition, represent a paradigm shift with regard to the likely influence of human chemical communication on sexual preferences and sexual behavior.  Clearly, olfactory/pheromonal input is intimately involved in human behavior.


    In this olfactory/pheromonal model of the development of sexual preferences that may be manifest in sexual behavior, the genetic/immunological (e.g., MHC/HLA) and neuroendocrine/hormonal (e.g., GnRH-directed LH-facilitated response) regulatory systems involved in mammalian chemical communication have their origins in yeast.  Although the mammalian protein products that the MHC/HLA genes encode have completely diverged from those of the yeast mating-type genes, the molecular biochemical mechanisms that regulate their expression and their GnRH-directed neuroendocrine effect have been conserved.  Thus, it seems likely that these two regulatory chemical communication systems have a common evolutionary origin.  The evolutionary origins of the MHC/HLA gene loci and the GnRH-directed LH-facilitated response to mammalian pheromones from a conspecific are the clearest indicators that these evolved molecular biochemical mechanisms are central to the reproductive physiology of all organisms.

    Despite clear indications of its evolutionary importance, the molecular biochemical basis of the GnRH-directed LH-facilitated response to olfactory/pheromonal input from a conspecific has not yet helped researchers arrive at a consensus with regard to use of the term pheromone.  This lack of consensus derives from discussion of pheromones in yeast, to use of the term pheromone when describing human behavior (for review, see Meredith, 2001).  Through recent reports it has become clearer that mammalian pheromones and putative human pheromones elicit a GnRH-directed LH-facilitated response in conspecifics as a reaction to social-environmental changes that are accompanied by olfactory input, as evidenced by several studies cited later.  Therefore, the term pheromone is used here in a manner that bridges the transition from unicellular asexual reproduction to unicellular sexual reproduction.  The term pheromone continues to bridge this evolutionary transition by including GnRH-directed, LH-facilitated multicellular mammalian sexual reproduction.

    As used here, mammalian pheromones, including putative human pheromones, are chemical signals from the social environment that elicit a measurable change in vertebrate levels of GnRH-directed LH in conspecifics.  Mammalian pheromones also enable discrimination of self from non-self, although the mechanisms for this discrimination are not yet clear, and the unconscious affect on sexual preferences of this discriminatory ability is unlikely to be directly measurable.

    What is most clear is that a change in mammalian GnRH-directed LH levels occurs in males and in females and that this change is typically most pronounced when males are exposed to pheromones from females, or vice versa.  Because the GnRH-directed LH response is “sex-typical,” discussions of pheromones sometimes typify them as sex pheromones.  Unlike MHC/HLA linked pheromones, sex pheromones are more likely to be by-products of hormone metabolism.  Androgens, estrogens, and progesterone are considered to be “sex hormones” that metabolize to androgenic, estrogenic, and progesteronic “sex pheromones.”  However, data from recent studies show that the term “sex pheromone” is misleading.  Further, any association between the term sex and the term pheromone may be ambiguous.  Although these terms appear to fit a mammalian olfactory/pheromonal model for the development of sexual preferences associated with hormone dependent physical features, caution is advised with regard to any assumption that incorporates what may appear to be a logical step to use of human sex-typical designations like “homosexual” and “heterosexual” odors or pheromones.  Inadvertent use of a sex-typical or sexual orientation-specific designation might lead to labeled characterization of idiosyncratic olfactory/pheromonal preferences.  A male who prefers pheromones from another male might inadvertently and incorrectly be labeled a “homosexual.”  

    There is no proof as yet that any compound is a human pheromone.  So there is no proof that any putative human pheromone is characteristically associated with sexual orientation.  Although associations between olfactory/pheromonal input and sexual preferences are apparent, these associations do not extend well to individual labels like homosexual or heterosexual.  However, the use of sex-typical designations is pervasive in the literature, and results from relatively ambiguous classifications that are frequently associated with ill-defined aspects of sexual orientation (for review, see Kauth, 2005).  Thus, for concision, sex-typical designations or designations that are associated with sexual orientation have sometimes been incorporated in this review.  If, for example, research findings have already included labels like homosexual and heterosexual, these labels are used here in their original context.  Integration of these labels is not meant to validate their use.  


    Boehm, Zou, and Buck (2005) have detailed the feedback loops that link olfactory/pheromonal input to sexual behavior via the GnRH neuronal pathway.  They reviewed the important link between olfactory/pheromonal input from the social environment to activation of early gene expression in cells of brain tissue that secretes GnRH.  For example, a widely accepted marker of early gene activation is c-fos, which is mentioned 27 times.  As mentioned earlier, without a direct link between sensory input from the social environment and gene activation, any link from the social environment to hormones and their effect on behavior would be missing a crucial component for a model of behavior.  In contrast, according to Boehm, Zou & Buck (2005):

“The fact that GnRH neurons are the master regulators of reproductive endocrine status indicates that pheromone effects on reproductive hormone levels are ultimately mediated by these neurons. Indications that GnRH peptide plays an important role in the control of sexual behaviors suggest that pheromone effects on these behaviors might also involve GnRH neurons (p. 683).”


    Kohl (1992) may have been the first to focus on the importance of the GnRH-directed, LH-facilitated response to mammalian pheromones and its likely influence on human behavior.  Kohl and Francoeur (1995; 2002) reviewed the evolutionary significance of olfactory/pheromonal communication and its effect on LH and follicle stimulating hormone (FSH), on the production of other hormones, on neurotransmission associated with changes in behavior and, thus, on the likely affect of olfactory/pheromonal input on mammalian behavior, including human behavior.  The mammalian LH response to pheromones from conspecifics changes LH/FSH ratios and levels of hormones like androgens and estrogens.  GnRH-directed, LH-facilitated change in levels and ratios of androgens and estrogens hormones, like the androgen, T, and the estrogen, estradiol (E2), modulate neurotransmission via their effects on synaptogenesis, synaptolysis, and apoptosis during the development of sexual preferences and sexual behavior.

    The LH and T response to olfactory/pheromonal input has been repeatedly indicated or reported in findings from non-human animal studies and from human studies.  In mammals, short-term exposure of males to females is linked to a T increase in men, as well as in rats, mice, rabbits, bulls, rams, and monkeys.  The T increase in non-human mammals is believed to be due to the effect of pheromonal conditioning of an LH response that precedes the T increase (Graham & Desjardins, 1980).  (An ovulatory phase, LH-facilitated, E2 and T response to pheromones is indicated but less well-detailed due to its cyclic nature in females).

Putative Human Pheromones Effect Levels of Hormones

    Although LH was not measured, an aqueous mixture of five ovulatory phase rhesus monkey and human female-associated fatty acids, referred to as copulins, evoked increased saliva T levels in men (Jutte,1996).  Additional reports indicate that hormone responses in mammalian males and females vary with exposure to the putative pheromones of either males or females.  For example, Berliner, Monti-Bloch, Jennings-White, and Diaz-Sanchez (1996) indicated that a progesteronic putative human pheromone alters LH pulsatility and T levels in men, and Stern and McClintock (1998) have shown that the putative pheromones of women regulate ovulation in other women, presumably by influencing the LH/FSH ratio.  The putative human pheromone, androstenol, elicits changes in LH pulse frequency in women (Shinohara, Morofushi, Funabashi, Mitsushima, & Kimura, 2000).   Shinohara, Morofushi, Funabashi, and Kimura (2001) have shown that axillary putative pheromones from women who were either in the follicular or in the ovulatory phase of the menstrual cycle differentially modulate pulsatile LH pulse frequency in other women.  Preti, Wysocki, Barnhart, Sonheimer, and Leyden (2001) have suggested that the LH response may be used to determine precisely what compound is involved in an effect on LH exhibited by women who were exposed to the axillary secretions of men.  In addition, Preti, Wysocki, Barnhart, Sondheimer, and Leyden (2003) have demonstrated that male axillary extracts increase LH levels and that they also elevate mood in female recipients.  This mood elevation might be manifest in preferences that are manifest in the sexual behavior of women.  The putative human pheromone, androstadienone, has also been shown to elicit hormonal and mood changes (Grosser, Monti-Bloch, Jennings-White, & Berliner, 2000; Jacob & McClintock, 2000).

Mammalian Pheromones Affect Sexual Behavior

    The ability of mammalian pheromones, including putative human pheromones, to influence GnRH-directed levels of LH in conspecifics implies that in general pheromones elicit postnatal unconscious affects.  Such affects alter levels of hormones, including androgens and estrogens like T and E2 that influence mood and perhaps the development of sexual preferences and sexual behavior.

    Unconscious affects of pheromones can be powerful influences on mammalian sexual behavior.  For example, pheromones that male rats learn to visually associate with sexual activity can be used to condition LH release.  After minimal association with the natural odor of a female, an arbitrary odor will subsequently elicit a conditioned male LH response, even in the absence of odor previously associated with a female.  It follows that mammalian pheromones may initially be responsible for the LH response but that other odors associated with the pheromone-induced LH response may also play a role in the conditioning of the visual response or other sensory responses to olfactory/pheromonal cues.

    According to Graham and Desjardins (1980), the functional significance of the pheromone-induced conditioned change in LH secretion lies principally in the unequivocal demonstration that olfactory cues can activate the GnRH-directed male HPG axis in a way that mimics in every respect the activation achieved by exposure to a female.  After the LH response has been conditioned to pheromones, associated visual or other associated non-olfactory sensory input from the social environment can activate the HPG axis, even in the absence of olfactory cues.  From an endocrine perspective, given the link between mammalian LH and levels of hormones like T, the female olfactory/pheromonal cues that condition LH release presumably also condition T release.

    Exposure to sexually dimorphic putative human pheromones that elicit an LH response is the most likely explanation for the recent finding that saliva T levels in men increase with exposure to a young woman, but do not increase with exposure to a young man (Roney, Mahler, & Maestripieri, 2003).  Further, olfactory/pheromonal conditioning of a sexually dimorphic human visual response to pheromones helps to explain how psychophysiological effects of sexually explicit visual stimuli elicited elevations in LH and T in men in studies cited earlier.

    It seems likely that putative human pheromones via their influence on LH release also have the ability to condition T responses to non-olfactory sensory input, like visual, auditory, or tactile input.  Because change in levels of T have repeatedly been linked to male sexual behavior, this biologically based unconscious affect of pheromones on hormones links the postnatal social environment to olfaction and to the GnRH-directed neuroendocrinology of male sexual behavior.  However, this unconscious affect of pheromones does not require the ongoing presence of olfactory/pheromonal sensory input, and it does not require cognition.  Therefore, the role that pheromones play in the development of sexual preferences that may be manifest in sexual behavior might be compared to the role that odors, which are associated with food choice, play in the development of preferences that are manifest in food choice.


    There may be a uniquely human bias that favors reports of the consciously processed effect of visual input like erotic imagery on sexual preferences and on sexual behavior over reports of olfactory/pheromonal input and its unconscious affect on sexual preferences and behavior.  However, this bias is not logical because it has been shown conclusively that chemical signals that may or may not be manifest as consciously detected odors are associated with the psychophysiological responses that are involved in the development of food preferences and ingestive behavior.

    Gottfried and Dolan (2003) have shown that appetitive neural responses to a particular food can be generated by pairing odor with the sight of a computer-generated image.  The ability to make neural connections between appetitive or aversive odors and relevant or irrelevant visual stimuli is a part of learning that is probably common to all animals. This type of learning is how picture advertisements might effectively enroll the development of human food preferences.  For example, a picture of a steak may positively or negatively influence appetite.  The unconscious affect of the picture on appetite involves a form of classical conditioning of the visual response to olfactory input.

    In classical conditioning, a previously neutral item, the conditioned stimulus (CS), gains behavioral significance after being paired with a biologically active unconditioned stimulus (UCS).  The efficacy of conditioning depends on establishing CS:UCS links.  Typically, biologically based CS:UCS links are formed during multiple biologically active UCS representations that allow association of the CS with sensory features: a reward value (e.g., food reward) or a general effect (e.g., satiety).

    In an olfactory model for the development of food preferences and ingestive behavior, odors are the biologically active UCS that elicits a hormone response, which are associated with a food reward and with the temporary satiation of appetite.  In this olfactory/pheromonal model for the development of sexual preferences and sexual behavior, pheromones are the biologically active UCS that elicits a hormone response, which are associated with a sexual reward and with the temporary satiation of sexual interest.  Whether the hormone response is elicited by odors or pheromones, it is integrated with the CS, which cognitively appears to be visual input, during olfactory-visual integration of the dual olfactory and visual sensory input.  Thus, it is olfactory/pheromonal input that conditions the hormone response to visual input.

    Obviously, no mammal cognitively chooses to eat food that lacks olfactory appeal, and humans do not eat pictures of food.  On the other hand, even though the same visual stimulus is processed, a picture of a steak is unlikely to look good to someone whose classically conditioned odor-associated food preferences and ingestive behavior make them a vegetarian.  Therefore, it is reasonable to consider individual comments on how food looks as little more than a means to describe the food’s chemical or olfactory appeal.

    People who have developed different food preferences in accord with variations in the reward value of ingestive behavior may not agree on olfactory/odor appeal; the food might not look good to everyone.  Similarly, preferences for erotic imagery could be expected to correlate with one’s conditioned visual response to olfactory/pheromonal input and its reward value or sexual effect.  Thus, erotic visual imagery of a female, and most likely the female herself, will lack any classically conditioned olfactory/pheromonal (e.g., hormonal) association and chemical appeal for a male who has developed a sexual preference for other males.  This may explain why some erotic imagery of human females does not always attract, arouse, or look good to all human males, why erotic imagery has no value to other mammals, and why humans develop sexual preferences for specific erotic imagery.


    Neural responses to olfactory/pheromonal input demonstrate that olfactory classical conditioning of the visual response can differentiate appetitive and aversive conditioning to pleasant or unpleasant odors that have been paired with faces.  Recent studies of neural responses that occur in the orbitofrontal cortex (OFC) and amygdala (AMY) also link conscious choice to unconscious affects generated from the limbic system.  These reports help to establish the means by which consciously perceived odor input or the unconscious affect of pheromonal input can influence how we respond to visual input from faces (Gottfried, O’Doherty, & Dolan, 2002).  Preferences associated with emotions seem likely to reflect olfactory classical conditioning of the consciously perceived visual appeal of food or faces.  Another person may not look like he or she smells.

    Consciously perceived differences in the visually appealing physical features of men and women seem to largely depend on estrogen and androgen levels.  High estrogen/androgen ratios are associated with feminine facial characteristics, such as a small jaw, full lips, youth, and fertility.  Low estrogen/androgen ratios are associated with masculine facial characteristics, which may be visual cues that signal reproductive fitness.  The androgen, T, stimulates the growth of the jaw, cheekbones, brow ridges, center of the face from the brow to the bottom of the nose, and facial hair.  Men with higher T have more masculine faces.

    Estrogen-dependent facial femininity increases attractiveness of female faces.  Women with higher estrogen levels also have a body shape that is most attractive to most men: large breasts and narrow waists (Jasienska, Ziomkiewicz, Ellison, Lipson, & Thune, 2004).  Women with higher estrogen levels have higher pitched voices, and men prefer the faces of women with higher pitched voices.  In contrast, higher androgen levels in men provide their characteristically masculine features: taller, darker, more muscular, with lower pitched voices, and deposits of adipose tissue that increase the waist to hip ratio (WHR) compared to desirable females (Singh, 1993).

     Personal preferences for androgenic and estrogenic hormone-dependent physical characteristics of masculinity and femininity strongly suggest that such preferences are hormonally and, thus, biologically rather than culturally based.  Further, human sexual preferences for hormone-dependent physical characteristics emerge before cultural standards of beauty are likely to be assimilated.  Perhaps more relevant to a biological approach for the development of sexual preferences is that human preferences for hormone-dependent physical characteristics diverge with the hormonal changes that begin to occur during adrenarche and continue through puberty.  Androgen and estrogen-dependent natural body odor production changes with the onset of adrenarche.

    Young boys and girls do not typically have well-developed preferences either for hormone-dependent physical characteristics or for androgen and estrogen-dependent natural body odors.  Besides, adult sex differences in preferences are typical for some hormone-dependent characteristics and atypical for others.  Men and women, for example, tend to agree on the sexually dimorphic characteristics of facial beauty.  However, the preferences of many men and women also indicate that sex-typical sexual preferences develop for some physically attractive features that are sex-typical, like increased male height and darker male complexion.

    In accord with the learned sexuality motivational theory of behavior  of, preferences for extremely masculine or extremely feminine physical features have been attributed to learning mechanisms that allow extreme examples to generate stronger responses.  For example, the increased androgenic natural body odor production of a dominant male might be involved in a learning mechanism that is associated with his visibly perceived masculine physical features (Havlicek, Roberts, & Flegr, 2005).  Similarly, cyclic androgenic, estrogenic, and progesteronic influences on the natural body odor production of women might be involved in a learning mechanism associated with visual cues during fertility cycles (Havlicek, Dvorakova, Bartos, & Flegr, 2006).  In other mammals, pheromones associated with reproductive fitness and fertility coordinate properly timed reproductive sexual behavior.  This would help to explain how a man’s sexual preference for the scent of high-estrogen ovulatory phase women (Singh & Bronstad, 2001) could acquire functional significance in properly timed human reproductive sexual behavior.

    In mammals, including humans, sexual preferences are most likely to be by-products of the way the brain processes information.  Experience cognitively influences what men and women find attractive, but experience may also elicit unconscious affects on emotions.  Experience with detectable or undetectable olfactory/pheromonal cues to reproductive hormones, which may have shaped sexual preferences, could provide an adaptive evolutionary explanation for the cross-cultural tendency for feminine female faces to be found most attractive by men and for average or masculine faces to be found most attractive by women, whose preferences vary with menstrual cycle phase.

    Typically, the mechanisms that may be involved in the human development of preferences for attractive physical features, like human facial characteristics, are not described or mentioned in literature on mammalian mate preferences.  And, only recently has it been suggested that future human studies should address the nature of facial cues related to hormone levels.  (For a review of preferences for physical features and indications from research, see Law-Smith, Perrett, Jones, Cornwell, Moore, Feinberg et al., 2005; Rhodes, 2006).

Androgen/Estrogen Ratios Link Biology and Beauty

    The nature of attractive hormone-dependent facial cues can readily be addressed in this olfactory/pheromonal model.  Pheromones are by-products of hormone metabolism.  Some hormones, like the androgen T metabolize to pheromones that are associated with physical characteristics of males.  Other hormones like the estrogen E2 and progesterone metabolize to pheromones that are associated with the physical characteristics of females and with their fertility cycles.  Apart from their metabolism to pheromones and the role that olfactory/pheromonal input might play in the development of sexual preferences, these hormones have no inherent value as sensory stimuli from the social environment, because there is no non-olfactory link from these hormones to any sensory stimuli from the social environment that might be directly involved in the development of sexual behavior.

    The metabolism of androgens, estrogens, and progesterone into pheromones involves several factors.  One important factor is the initial production level of a particular hormone and how its production level compares to that of other hormones.  Men produce more androgens than women, whereas women produce more estrogens and progesterone than men.  Estrogen levels peak just before the most fertile phase of the menstrual cycle; progesterone levels peak with the relatively infertile phase of the menstrual cycle.  The cyclical hormone changes of the menstrual cycle compared to the tonic, relatively unchanging levels of androgens in reproductively fit males explains why men tonically produce different concentrations of androgenic, estrogenic, and progesteronic putative human pheromones than do women, whose hormone concentrations vary with fertility cycles.

    Cornwell and colleagues (2004) have proposed that both facial characteristics and putative human pheromones signal mate quality and that preferences for cues to mate quality should co-vary across the visual and olfactory domains.  For example, women who prefer more masculinized faces should have an increased preference for the androgenic pheromones of men.  Two studies found concordance in the preferences for facial characteristics and for putative human pheromones.  One of these putative human pheromones is linked to masculinity via androgen metabolism.  The other putative human pheromone is linked to femininity via estrogen metabolism.

    Kovacs and colleagues (2004) have found that smelling either an androgenic compound or an estrogenic compound biased the face preference of men, when female faces were relatively ambiguously masculinized.  With exposure to the androgenic compound, men more rapidly judged faces with less masculinized features to be men.  With exposure to an estrogenic compound, men took longer to judge faces with more masculinized features to be men.

    The combined findings from studies of putative human androgenic or estrogenic pheromones and facial characteristics show that male-associated and female-associated olfactory/pheromonal cues can affect the judgment of sexually dimorphic facial characteristics that are consciously perceived to be visually attractive.  Therefore, one could surmise that putative human pheromones and sexually dimorphic facial characteristics convey common information about the quality of potential mates.

    Cornwell et al. (2004) and Kovacs et al. (2004) did not attempt to elucidate the molecular biochemical mechanisms that would enable the concordance of visual and olfactory/pheromonal cues with regard to mate quality.  However, they noted other studies that discuss a likely mechanism, namely integration of visual and olfactory cues in neuroanatomical structures of the brain where sex-typical facial cues are believed to be processed (for a review, see Savic, Berglund, Gulyas, & Roland, 2001; Sobel, Prabhakaran, Hartley, Desmond, Glover, Sullivan et al., 1999).

    Integration of olfactory and visual cues in neuroanatomical structures of the brain is an important ability that is probably required for men and women to develop sexual preferences for facial features or other physical features.  Minimally, if male sexual preferences for the physical features of females involves mechanisms that enable concordance among olfactory/pheromonal and visual cues, current findings suggest that the neural mechanisms that enable the concordance of this integration and perhaps even pair bonding are very different between males and females, as indicated by Fischer, Sandblom, Herlitz, Fransson, Wright, & Backman (2004).  It also seems likely that concordance among olfactory/pheromonal and visual cues varies with sexual preferences that have developed for the physical features of males or females.

    Predictably, in males, olfactory/pheromonal conditioning of the visual response to estrogenic pheromones would increase a female’s visual appeal and make a female with a high estrogen/ androgen ratio look better at the same time her fertility level was relatively high.  Similarly, physically appealing androgenic characteristics like height, darker complexion, masculine facial features, and aspects of bilateral symmetry might explain a preference for a tall, dark, handsome, and dominant male, since these characteristics are associated with androgenic pheromones (see Kohl & Francoeur, 2002, for review).  Thus, it is appropriate to consider the role of pheromones in males who have developed preferences for other males.


    Kohl et al. (2001) have indicated that future consideration should be given to the role of pheromones in sexual orientation: “Predictably, we will soon address other aspects of human attraction, and social confounds such as the paraphillias—and even sexual orientation in future discourse (p. 319).”  Intuitively, any paraphillic sexual arousal that might involve a sexual preference for rubber, leather, shoes, undergarments, etc., may also include odor associations.  These odor associations may elicit unconscious affects that are associated with sexual arousal during early relevant experiences, such as masturbation in the presence of such odors and in the presence of other sensory stimuli that are not necessarily from the social environment but that are associated with the paraphillic sexual arousal.

    Recent studies provide significant evidence of an olfactory/pheromonal link to mammalian sexual orientation, including human sexual orientation.  For example, there are differences in the production of human odor profiles and differences in the preferences for human odor profiles that appear to correlate with sexual orientation.  In addition, Martins, Preti, Crabtree, Runyan, Vainius, and Wysocki (2005) have found that homosexual men distinguish between the odor of heterosexual men and homosexual men and that homosexual men prefer the natural body odor of other homosexual men.  Their report is the first to directly link the uniquely human conscious perception of human natural body odor production to male sexual preferences for other men.  There are, however, mammalian correlates that suggest this link.

Mammalian Neuroendocrine Correlates

    Unlike other sensory systems, the mammalian olfactory system is sexually dimorphic at birth.  Accordingly, an olfactory/pheromonal model for the development of mammalian male sexual preferences for other males can be expected to include sexually dimorphic neuroanatomical features that correlate with a sexually dimorphic neuroendocrine response to pheromones that begins at birth.  Indeed, in all mammals that have been studied, this innate prenatal neuroanatomical sexual dimorphism (i.e., the organization of the mammalian olfactory system) establishes sexual dimorphism in the activation by pheromones of the postnatal GnRH-directed LH response.

    The specialized role that GnRH plays in prenatal and in postnatal sexual differentiation of the brain, which includes sexual differentiation of how differences in physical features are consciously perceived by humans and elicit unconscious affects on sexual preferences in all mammals, is central to the concept that is modeled here.  Namely, by influencing hypothalamic GnRH pulsatility, mammalian pheromones can also play an important role in postnatal sexual differentiation.  During postnatal sexual differentiation, the effect of pheromones on hypothalamic GnRH pulsatility can be expected to correlate with unconscious affects on emotions, as proposed by Kohl et al. (2001):

“The ‘affective primacy hypothesis’ asserts that positive and negative affective reactions can be evoked with minimal stimulus input and virtually no cognitive processing.  Olfactory signals seem to induce emotional reactions whether or not a chemical stimulus is consciously perceived.  We theorize that the importance of human non-verbal signals is based upon information processing, which occurs in the limbic system, and without any cognitive (cortical) assessment.  Affect thus does not require conscious interpretation of signal content.  Underlying this fact is that affect dominates social interaction and it is the major currency in social interactions.  Affective reactions can occur without extensive perceptual and cognitive encoding.  They are made with greater confidence than cognitive judgments, and can be made sooner.  Olfactory input from the social environment is well adapted to fit such assertions.  For example, chemical cues allow humans to select for, and to mate for, traits of reproductive fitness that cannot be assessed simply from visual cues (p. 310).”


An example of a sexually dimorphic postnatal affective reaction is the GnRH-directed, LH-facilitated response to olfactory/pheromonal input from the social environment.

    GnRH neurons share a common origin with olfactory neurons during human fetal development (Barni, Maggi, Fantoni, Granchi, Mancina, Gulisano et al., 1999).  Prenatal organization of the sexually dimorphic GnRH-directed, LH facilitated response to olfactory/pheromonal input occurs when embryonic migration of GnRH neurosecretory neurons from the olfactory placode to the hypothalamus establishes the hypothalamic GnRH pulse.  The hypothalamic GnRH pulse modulates the prenatal and postnatal concurrent maturation of the neuroendocrine system, the reproductive system, and the central nervous system (CNS).  In both animal and human studies, two things seem clear: 1) When hypothalamic GnRH pulse frequency decreases, levels of FSH increase; 2) when hypothalamic GnRH pulse frequency increases, levels of LH and of either T or E2 increase (for review, see Everett, 1994; Grumbach & Styne, 1992; Silverman, Roberts, Dong, Miller, & Gibson, 1992).

    The GnRH-directed, LH-facilitated prenatal secretion of androgens, like T, and of estrogens, like E2, are primarily responsible for organizing and establishing traditionally conceived prenatal and postnatal sexual dimorphism in the classical gonad-to-hormones-to-behavior (G-H-B) model.  In this G-H-B model, sex differences in the brain and behavior of mammals, like rodents, develop under the influence of estrogens, which are derived from the conversion of T to estrogens via the neural aromatization of testosterone.  In the presence of T-derived estrogens, the brain develops as male, and as female in their absence.  T secreted by the testes causes masculinization of the brain and behavior and an increase in male-typical characteristics.  The absence of T secreted by the testes allows for defeminization and a decrease in female-typical characteristics, of the brain and behavior.

    In this G-H-B model, there are differences between effects of T and E2 on sexual differentiation in rodents compared to primates, including humans.  For example, estrogenic metabolites of androgens may not be critical for masculinization and defeminization in rhesus monkeys or humans (Wallen, 2005).  Despite mammalian species differences in masculinization and defeminization, this olfactory/pheromonal G-H-B model includes sexual dimorphism in the olfactory system and in many other neuronal pathways and in the neuroendocrine systems of all terrestrial mammals (for review, see Diamond, Binstock, & Kohl, 1996).

    Throughout postnatal development, the hypothalamic GnRH pulse continues to develop the neural substrates that enable mammalian olfactory pathways to exhibit sexually dimorphic specificity to pheromones.  The hypothalamic GnRH pulse also promotes the general ability of these neural substrates to transduce sexually dimorphic pheromones into a neuroendocrine response.  In a reciprocal relationship, this neuroendocrine response to pheromones alters the hypothalamic GnRH pulse, the release of the gonadotropins, LH, and FSH, as well as levels of androgens and estrogens, like T and E2.  Levels of T and E2 correlate with sex-typical pheromone production during sexual maturation and sexual differentiation (Kloek, 1961; Michael & Keverne, 1971; Nixon, Mallet, & Gower 1988; Preti & Huggins, 1975).

    Perhaps the best evidence that unconscious affective reactions to olfactory/pheromonal input do not require human cognition comes from non-human mammalian models, since the affect of mammalian pheromones on hormones and behavior is unlikely to be cognitively processed in any non-human mammal.  Kohl (2005) has proposed that three mammalian models link mammalian pheromones to neural pathways and to limbic system structures that exert significant cumulative effects on the HPG axis and on behavior, effects on the HP-adrenal (HPA) axis notwithstanding.  These cumulative effects on the HPG axis include development of a sexually dimorphic GnRH-directed, LH-facilitated hormone response to the pheromones of conspecifics, subsequent effects on E2 receptor (E2R) content in brain structures, and unconscious affects on sexual preferences and sexual behavior that have been associated with mammalian male sexual orientation.

Mammalian Models and Neuroendocrinology

    In the first mammalian model, pheromonal induction of GnRH release influences the HPG axis and levels of E2, resulting in higher levels and occupancy rates of E2Rs in the hypothalamus of hormone-treated male rats that exhibit lordosis—a female-like behavioral response (for review, see Samama & Aron, 1989).  In the second mammalian model, sex differences in odor preference are most likely related to functional T and E2-determined sex differences in the olfactory pathway of rats.  Neonatal treatment with 1,4,5-androstatriene-3,17-dione (ATD), which blocks the aromatization of T to E2, alters the sexual differentiation of olfactory pathways in rats.  Male rats that are treated with ATD show a female-like mammalian pheromone-processing system and exhibit lordosis when they are exposed to the odors of sexually active males.  However, ATD-treated males also exhibit mounting behavior when exposed to the odors of estrus female rats.  It appears that ATD treatment causes male rats to respond to the presence of pheromones from females or from other males (for review, see Bakker, Baum, & Slob, 1996; Bakker, van Ophemert, & Slob, 1996).  Lordosis in males has been challenged, along with most claims of demonstrable behaviors that have been linked to males or to females and, therefore, to sexual orientation in mammals.  A third mammalian model linking pheromones, hormones, and sexual behavior may persuade others to reconsider such challenges.

    In this third model, the sex-typical mammalian male LH response does not occur in homosexual rams when they are exposed to estrus females (Perkins Fitzgerald, & Price, 1992).  Absence of the male LH response to estrus female pheromones indicates that an atypical GnRH-directed response to pheromones might be involved in the forced-choice homosexual behavior of rams.  To help explain the existence of what may be exclusive homosexual orientation in domesticated rams, Perkins, Fitzgerald, and Moss (1995) discussed correlates among perception of olfactory input from potential forced-choice mates and levels of LH, T, and estrogens, as well as correlates with E2R content in the AMY: “Estradiol receptors present in the AMY could . . . translate information about peripheral concentration of testosterone, thereby integrating the AMY with other centers important for copulation (p. 38).”  Simply put, E2 receptor content in the AMY could “tell” a male he is male, or a female that she is female. 

    Pheromonal induction of the sexually dimorphic LH response can be expected to alter levels of E2, which ultimately effect E2R content in the AMY of rams.  The LH response, or lack thereof, to the pheromones of estrus ewes, as well as E2R content in the AMY, appear to be measurable factors that correlate with sexual orientation in rams.  Rams that exhibit a full range of proceptive homosexual behaviors, lack the LH response.  These rams also have less E2R content in the AMY, which is a characteristic of ewes.  E2R content in the AMY of homosexual rams and in ewes was similar, but less than the E2R content in the AMY of heterosexual rams.  The apparent link among pheromones, LH, T and E2, E2R content in the AMY, and homosexual orientation in another mammal incorporates the first two mammalian models and provides a reason to look for additional clues that might link olfactory/pheromonal input from the social environment to hormones and to human sexual preferences for other males.

An Equivocal Neuroendocrine Correlate

    The link from sexual dimorphism of the olfactory system to the sexually dimorphic LH response to pheromones and to E2R content in the AMY of homosexual rams is important because it may also link olfactory/pheromonal input to the development of adult human cyclic hormone secretion, which may vary with the development of sexual preferences and human male sexual orientation.  The adult cyclic hormone secretion that is characteristic of women includes an estrogen-induced ovulatory LH surge.  In homosexual men, an LH response to estrogen priming may be intermediate to the response of heterosexual males and females (Gladue, Green, & Hellman, 1984; for review, see Dörner, 1988).  The equivocal LH response to estrogen priming appears to exemplify incomplete sexual differentiation of the homosexual male GnRH-directed neuroendocrine system.  Some believe that this intermediate response indicates that homosexual men have higher levels of E2 during development (e.g., Dörner, Docke, Gotz, Rohde, Stahl, & Tonjes, 1987).

    Perhaps, levels of T and factors that influence the aromatization of T to E2 correlate with E2R content in specific brain tissues like the AMY during human development.  There also may be other mechanisms involved in the determination of T and E2 levels, as well as E2R content in the AMY during human and non-human animal development.  For example, Bakker, De Mees, Douhard, Balthazart, Gabant, Szpirer, and Szpirer (2006) have shown that alpha-fetoprotein binds prenatal E2 in the developing female rat brain and prevents the masculinizing effect of T conversion to E2 in a manner similar to that of ATD inhibition.

    Typically, higher T levels during male reproductive maturation somehow neutralize the mechanism for adult cyclic hormone secretion in men.  In males who develop preferences for other males, levels of T and E2 or the mechanisms involved in the aromatization of T to E2 may fail to neutralize the E2-induced LH surge during development.  Theoretically, this indicates that a neuroendocrine predisposition may exist for the development of male sexual preferences for other males.  There should be little doubt that this neuroendocrine predisposition could be manifest in the GnRH-directed, LH facilitated production of E2 and T, which are modulated by the hypothalamic GnRH pulse and, thus, by the likely influence of olfactory/pheromonal input on the GnRH pulse during development.

    The aforementioned scenario offers some insight about how incomplete sexual differentiation, including sexual differentiation of the olfactory system, could affect E2R content in the AMY and sexual preferences for odors.  The AMY is important to the processing of olfactory input (Zald & Pardo, 1997).  During postnatal development, olfactory/pheromonal input could influence sexual differentiation of E2R content in the AMY via their influence on GnRH, LH secretion and, thus, on estrogen or androgen secretion, like E2 and T.  And, as noted above, E2 induced patterns of hormone secretion and E2 receptor in the AMY have been linked, albeit equivocally, to mammalian male sexual preferences for the odors of other males.  Therefore, incomplete sexual differentiation of the olfactory system and the influence of olfactory/pheromonal input could individually or collectively influence sexual differentiation and sexual preferences for the odor of other males.

    With regard to GnRH-directed LH secretion and its potential association with classically conditioned odor preferences, it is worth noting that an atypical LH response to GnRH also appears to correlate with human sexual orientation in both male-to-female and female-to-male transsexuals.  Sensitivity of LH secretion to GnRH in transsexual women is decreased, but increased in transsexual men.  Kula, Dulko, Pawlikowski, and Imielinski (1986) have proposed that these correlates might reflect the effect of the environment on the down-regulation of the pituitary LH response, although the social environment was not specified.  The fact remains that the LH response to olfactory/ pheromonal sensory input from the social environment could be the best adult indicator of the endocrine milieu during postnatal development.

    To reiterate, it appears that mammalian pheromones are likely to influence postnatal brain development via their GnRH-directed effect on the HPG axis and, thus, their effect on LH, T, and E2.  Neuroanatomical organization and neuroendocrine activation are manifest in these hormonal responses, and these hormone responses parallel neuroendocrine responses seen in animal studies in which sex differences in hypothalamic activation and in the LH response to pheromones have been correlated with the influence of mammalian pheromones on sexual behavior.  It is extremely likely that during the GnRH-directed concurrent maturation of the neuroendocrine system, the reproductive system, and the CNS the influence of pheromones on GnRH-directed perturbations could allow olfactory/pheromonal conditioning of the LH response to other sensory input that is associated with the development of sexual preferences and sexual behavior.

Neuroanatomical Correlates of the Neuroendocrine Response to Pheromones

    Early in life, neuroendocrine mechanisms are active for the control of pituitary-gonadal hormone production and in part for the control of adrenal hormone production.  For example, Jakacki, Kelch, Sauder, Lloyd, Hopwood, and Marshall (1982) have shown that prepubertal children secrete GnRH-directed LH in a pulsatile manner, well before sexual maturation.  Because mammalian pheromones alter the GnRH-directed neuroendocrine mechanisms that control LH/FSH ratios and levels of androgens and estrogens, olfactory/pheromonal input also is capable of influencing the development of behavior during sexual maturation and, notably, long before most signs of human sexual maturation appear.

    In any model that addresses the development of sexual preferences and sexual behavior, consideration must be given not only to neuroendocrine activation of the behavior by sensory input from the social environment but also to the initial and perhaps ongoing organization of the physiological processes that allow for this activation.  Activation by sensory input from the social environment occurs throughout postnatal life, and activation is paired with a response to the event that activated it (Black & Prokasy, 1972).  Once a neuroendocrine response is established, it is considered to be an organized response.  However, a classically conditioned neuroendocrine response to pheromones might initially be organized by olfactory input from the social environment that is later also paired to elicit further organization.  Mammalian pheromones, therefore, might first organize and then better organize a classically conditioned GnRH-directed neuroendocrine response.

    In this olfactory/pheromonal model of the development of sexual preferences and sexual behavior, the obvious neuroanatomical basis for mammalian neuroendocrine function predicts neuroanatomical links that vary with genetic sex and with the development of sexual preferences.  Neuroanatomical links to sexual preferences add substantial detail to GnRH-directed regulatory aspects of androgens and estrogens levels, as well as the involvement of these hormones in the E2-induced LH surge and regulation of E2R content in various tissues, like the AMY.  Innate neuroanatomical differences that vary both with genetic sex and with the development of sexual preferences also strongly suggest that there are genetically determined neuroanatomical (e.g., organizational) aspects of sexual orientation.

Genetically Determined Neuroanatomy, Pheromones, and Sexual Preferences

    In mice that lack GnRH, very few GnRH neurons are required to induce a hypothalamic GnRH pulse that is capable of activating the GnRH-directed LH response from the pituitary, which changes levels of androgens and estrogens like T and E2 (Silverman et al., 1992).  Evidence of neonatal sexual dimorphism in the GnRH neuronal system of rats strongly attests to the likelihood that more GnRH neurons (Tobet & Fox, 1992) or their sexually dimorphic connectivity collectively allow for a more frequent hypothalamic GnRH pulse in males than in females.  A more frequent hypothalamic GnRH pulse is likely to increase LH/FSH ratios and androgen/estrogen ratios in males.  Males also have greater numbers of neurons at sites of synaptic transmission that are involved with the processing of olfactory input in the limbic system (for review, see Segovia & Guillamon, 1993).  Perhaps fewer GnRH neurons or their sexually dimorphic connectivity correlate with a less frequent hypothalamic GnRH pulse and with the normal occurrence of comparatively increased production of FSH and estrogens in mammalian females, including human females.

    Whether or not sex differences in the number or the connectivity of GnRH neurons are involved in hypothalamic GnRH pulse frequency, there are consistent differences among the neuronal density of hypothalamic nuclei and heterosexual activity in non-human primates and in homosexual and heterosexual men (Byne, Tobet, Mattiace, Lasco, Kemether, Edgar et al., 2001).  This pattern suggests that neuroanatomical differences in hypothalamic nuclei might lead to alterations in GnRH pulse frequency and to a more female-type pattern of LH and FSH secretion in males who have developed sexual preferences for other males.  Consistent neuroanatomical differences in the medial preoptic area of the anterior hypothalamus (MPOA/AH) are related to development of sexual preferences.

    The MPOA/AH is critical for the expression of male sexual behavior in many mammals (Hull, Meisel, & Sachs, 2002).  It is also essential for LH release and, thus, is very likely to play a role in GnRH pulse frequency (for review, see Meisel & Sachs, 1994).  Within the MPOA/AH, sexually dimorphic cell groups that are larger in males than in females have been identified in several species, including humans.  Experiments in several species indicate that the development of these sexually dimorphic cell groups (i.e., nuclei) within the MPOA/AH are the direct result of exposure to T or to its metabolites, which probably include aromatized E2, during a critical period in prenatal or early neonatal life (Cooke, Hegstrom, Villeneuve, & Breedlove, 1998).

    A sexually dimorphic nucleus (SDN) exists in the sheep MPOA/AH, which has been linked to sexual preferences and to sexual behavior.  The SDN is three times larger in volume and contains more nerve cells in rams than in ewes (Roselli, Larkin, Resko, Stellflug, & Stormshak, 2003).  It is believed that SDN volume in sheep and the number of cells in the SDN somehow bias the processing of sexually relevant sensory cues, which include olfactory/pheromonal cues in most if not all mammals.  Sexual dimorphism in cells of the SDN also correlates with E2R content in the AMY of sheep and with sexual partner preference.  Further, an SDN in male rats treated with ATD correlates positively with male-typical behavior and with female-directed partner preferences (Houtsmuller, Brand, De Jonge, Joosten, Van De Poll, & Slob, 1994).  Thus, it would be difficult to overestimate the importance to sexual behavior of neuronal number and density in the MPOA/AH nuclei, or in any other brain tissue.  For example, male rats have greater numbers of neurons at sites of synaptic transmission that include the MPOA/AH and the AMY (Bressler & Baum, 1996).  To the degree that cross-species comparisons are valid, a sexually dimorphic response to mammalian pheromones in hypothalamic nuclei, like those in the MPOA/AH that regulate GnRH-directed LH release, could help to explain the development of sex differences in mammalian behaviors, including sexual preferences.

    The likely role of olfactory/pheromonal effects on the neurons in the MPOA/AH and the relevance of these neurons to the development of male sexual preferences was indicated by Oomura, Yoshimatsu, and Aou (1983), who suggested that MPOA/AH neurons integrate visual and olfactory cues from the receptive female, which lead to sexual arousal and initiation of mating by the male monkey.  With some additional focus on the role olfactory/pheromonal input, Ishai, Ungerleider, Martin, Schouten, & Haxby (1999) have shown that putative human pheromones are associated with neocortical aspects of facial recognition via the comparative study of activation of brain areas by presenting odors paired with pictures of houses, chairs, and faces.  Continued focus on the neuroanatomy that might be associated with the processing of putative human pheromones led Savic et al. (2001) to show that the human male and female brain responds differently to the androgenic compound, AND, as compared to the estrogenic compound, EST, and as compared to common odors.  A sex difference in brain activation by the putative human pheromones AND and EST strongly suggests that there is a neuroanatomical basis for the sex differences in brain activation that are manifest with exposure to androgenic or estrogenic compounds, which may be human pheromones.

    Genetically determined neuroanatomical aspects of sex differences that play a role in the development of sexual preferences also are the most likely reason that hypothalamic activation by EST is greater in men, while hypothalamic activation with AND is significantly greater in women.  The effect of the putative human pheromone EST on men was concentrated in the dorsomedial hypothalamic nucleus.  The effect on women of the putative human pheromone AND was in the MPOA/AH.  Given the likely role that the MPOA/AH plays in the integration of visual and olfactory/pheromonal cues, it is not surprising that AND and EST also elicited sexually dimorphic responses in neuroanatomical regions, like the fusiform and lingual gyrus that have been linked to the visual imagery of faces.  As noted earlier, visual imagery of faces appears to be consciously associated with estrogen/androgen ratios, although this conscious association may result from an unconscious affect of olfactory/pheromonal input.

    Savic et al. (2001) have briefly touched on the possible reasons for sexually dimorphic activation of the MPOA/AH by putative human pheromones—namely, that hormones, like androgens and estrogens, modulate sexually dimorphic neuronal density.  Androgens and estrogens also modulate their hormone receptor content in the hypothalamus, in the AMY (e.g., E2R), and more globally in the limbic system.  Additional insight from this study may further support extension to humans of a mammalian olfactory/pheromonal model for sexual orientation.  For example, while addressing sexually dimorphic neuroanatomy, Savic et al. (2001) specifically mentioned the third interstitial nucleus of the anterior hypothalamus (INAH-3), which may link pheromones to male preferences for other males.

    Arguably, the size of the INAH-3 in homosexual men is intermediate between human heterosexual males and females (LeVay, 1991).  A sex difference in INAH-3 volume, but not a sexual orientation difference in volume, has been attributed to a sex difference in neuronal number.  Instead, variation in INAH-3 size with male sexual orientation appears to correlate best with neuronal density, which is likely to be influenced by environmental cues during postnatal brain development (Byne et al., 2001).  The developmental effects of social-environmental cues like human pheromones on GnRH-directed LH/FSH ratios and, thus, on levels of T and E2 might cause sex differences both in INAH-3 size and in INAH-3 neuronal density during postnatal brain development.  Indeed, based primarily on mammalian models, Byne et al (2001) have acknowledged that the most likely reason for a sex difference in the INAH-3 is that it depends in part upon sex differences in developmental exposure to gonadal hormones, like T and E2.

    Given this neuroanatomical explanation for a correlation between hormones like T and E2 and sexual differentiation, the production of these gonadal hormones might be expected to vary with the outcome of sexual differentiation and with male and female heterosexual orientation.  However, neuroanatomical correlates of heterosexual male and of homosexual male orientation do not sufficiently explain why neuronal density in the INAH-3 co-varies with male sexual orientation.  Accordingly, it is important to address the most likely neuroendocrine substrate that could be involved in the development of this neuroanatomical correlate—one that allows olfactory/pheromonal input to alter postnatal levels of T and E2 and, thus, to possibly alter neuronal density in the INAH-3.

    The INAH-3 is located in the MPOA/AH.  As noted above, this area is essential for GnRH release, which modulates LH/FSH ratios, androgen/estrogen ratios and, thereby, the sexual differentiation of hormones and behavior in mammals.  In contrast to heterosexual men and in congruence with heterosexual women, homosexual men displayed greater hypothalamic activation in response to the androgenic compound AND.  In congruence with heterosexual women, the MPOA/AH of homosexual men was activated by AND (Savic, Berglund, & Lindstrom, 2005).  The difference in the specific area of the hypothalamus that was activated by putative human pheromones suggests that olfactory/pheromonal activation of these areas may be involved in the development of sexual preferences and sexual behavior. 

    The MPOA/AH is a fundamental part of the olfactory pathways which process pheromones.  The presence of a female rat is sufficient to activate MPOA/AH neurons in sexually inactive males.  Sniffing and pursuing the female increase the firing rate of MPOA/AH neurons.  Mating induces MPOA/AH changes in gene activation and in neuronal immunoreactivity in males within 30 minutes of exposure to a female.  Lesions of the MPOA/AH inhibit pursuit of females by male rats.  Neurotoxic or electrolytic MPOA/AH lesions in male ferrets and electrolytic MPOA/AH lesions in male rats change their preferred stimulus from estrous female to male.  Male rhesus monkeys stopped mating after MPOA/AH lesions.  Thus, evidence from several mammalian species strongly suggests that the functional integrity of MPOA/AH neurons is crucial for the male ability to identify a female partner, determine sexual partner preference, and eventually mate with the female.  Further, T conversion to E2 by aromatase may facilitate male sexual behavior via E2 receptors on neurons in the MPOA/AH.  Finally, lower aromatase activity in the MPOA/AH is present in rams with less E2R content in the AMY that prefer mounting other males instead of estrus females (for review, see Kindon, Baum, & Paredes, 1996; Paredes, 2003; Paredes, Tzschentke, & Nakach 1998).

    Regarding the development of male sexual preferences, there is no non-olfactory/pheromonal means to so clearly pinpoint specific cells in tissue of the brain (e.g., in the MPOA/AH) that can be linked to the effects of sensory input from the social environment, to effects on hypothalamic GnRH pulsatility, and to effects of GnRH on levels and ratios of other hormones like T and E2 that appear to influence sexual preferences.  The effect of putative human pheromones on the MPOA/AH, on hypothalamic GnRH-directed, LH/FSH ratios, and on T and E2 during postnatal development may help to explain environmentally effected co-variation in the neuronal density of the human male INAH-3 and, thus, more globally, in the MPOA/AH, and perhaps throughout the limbic system.  This environmentally effected co-variation could be related both to GnRH pulsatility and to genetically predisposed, GnRH-directed, LH-facilitated unconscious affects on the development of male sexual preferences for other males.


    With regard to environmentally influenced co-variation, Byne et al (2001) have noted that the major expansion of the human brain occurs postnatally, while the individual is in constant interaction with the environment.  The social environment was not specified.  However, the effect of olfactory/pheromonal input from the social environment on GnRH and on other hormone levels like LH, T, and E2 may best explain how the social environment can unconsciously affect sexually dimorphic postnatal expansion of the human brain during development.

    In this olfactory/pheromonal model, the adult brain responds to environmental and internal challenges that induce significant functional and anatomical modifications in the brain during the development of odor preferences that are associated with sexual preferences.  This concept of neural plasticity has been well documented for the perinatal period—called a critical period.  A critical period is a time during which sensory-driven activity patterns are able to induce long-term changes in specific neuronal circuits that last throughout adulthood (Berardi, Pizzorusso, & Maffei, 2000).  For example, Bakker, van Ophemert, andSlob (1996) have exemplified a critical period in the development of the olfactory system using the aromatase inhibitor ATD.  When used in the perinatal critical period, ATD alters odor-associated sexual partner preferences in mature male rats.

    Olfactory/pheromonal-associated sexual partner preferences are susceptible to change during development due to important inherent features of the mammalian olfactory system.  For example, the main olfactory bulb is the first central relay of the olfactory system and is also one of the few structures in the mammalian CNS that is continuously supplied with newly generated nerve cells (neurogenesis).  The effect of odor stimulation on activity-dependent neurogenesis provides a means by which changes in newly generated nerve cells might be related to changes in behavior that are associated with olfactory/pheromonal input from the social environment or other sensory input.  During development, inhibitory or excitatory signals from olfactory/pheromonal input are paired with the self-renewing ability of the olfactory bulb neuronal network.  This ongoing remodeling of the main olfactory bulb may mean that there is no critical period in mammals that are relatively slow to reach sexual maturity.  At any time during the process of sexual maturation, any developmental perturbation that might influence either inhibitory or excitatory sensory signals could also result in the reorganization of cortical circuitry, which could be positively or negatively influenced by olfactory/pheromonal experience that is consciously associated with other sensory experience.

    Ongoing neurogenesis also occurs in the mammalian hippocampus, a brain structure that is believed to be important to learning, memory, and behavior.  In the mammalian hippocampus, genetic underpinnings and hormone levels are among several factors that influence proliferation, differentiation, and survival of mature stem cells.  Recent studies suggest that changes in olfactory bulb neurogenesis occur during events that are relevant to behavioral development in rodents.  For example, pair bond formation in monogamous prairie voles involves olfactory learning for mate recognition, which is associated with an increase of neurogenesis in the olfactory system.  However, ovariectomized female prairie voles with resultant estrogen deprivation do not show increased neurogenesis when exposed to males.  Because estrogen treatment partly reinstates neurogenesis, it is likely that estrogen or perhaps the GnRH-directed, LH facilitated estrogen/androgen ratio mediate the increase of neuronal proliferation following exposure to a male in this species.  Odor-associated experience also appears to change brain function in the olfactory bulb of mice.  It is likely that an odor enriched environment influences olfactory bulb and hippocampal neurogenesis and, therefore the learning, memory, and behavior of all mammals (for review, see Lledo & Gheusi, 2003).

    Barni et al. (1999) have shown GnRH-release responses to odor in cultures of human GnRH neurons, which also respond to hormones like T and E2, with GnRH release.  This data represents additional evidence of a direct link between the human olfactory system and the human neuroendocrine systems.  It also suggests that odor-associated experiences have the ability to alter levels of hormones that influence neurogenesis during development.  In addition, the hormone concentrations that elicited GnRH secretion in vitro were similar to those found in human blood in vivo at the time of ovulation, which is when a physiological surge in GnRH secretion takes place in women.  This link to both the influence of odor and to the influence of hormones in women is not surprising, since there is substantial data from non-human mammals to suggest the reciprocal nature that olfactory/pheromonal input shares with hormone levels like E2 and T in males and in females.  To reiterate, primary evidence of this reciprocity is the increase in LH that occurs with exposure to mammalian pheromones from conspecifics of the opposite sex, since the LH/FSH ratio alters levels of hormones like E2 and T.


    A mammalian olfactory/pheromonal model suggests that the LH response to pheromones is incompletely sexually differentiated in males who prefer other males.  It follows that these males might respond neuroanatomically and neuroendocrinologically to the pheromones of other males as they might otherwise respond to the pheromones of females.  This was indicated in homosexual rams that respond to the odors of other males and do not show an LH response to the odors of estrus ewes.

    An atypical sexual preference for the pheromones of other males might lead to additional neuroanatomical and, therefore, neuroendocrine differences that correlate with sexual preferences and sexual behavior.  During postnatal expansion of the human brain, these neuroanatomical and neuroendocrine differences could be expected to correlate with the LH response to pheromones, with organizational and activational effects of hormones like E2 and T on E2R content in the AMY and, via neurogenesis, with neuronal size and density in the MPOA/AH.

    A noteworthy proposal by Hamer and Copeland (1994) concerning the genetics of sexual orientation may further extend this mammalian olfactory/pheromonal model to humans: “Where might a ‘gay gene’ fit into LeVay’s analysis?  The most simple hypothesis would be that Xq28 makes a protein that is directly involved in the growth or death of neurons in the INAH-3.  Alternatively, the gene could encode a protein that influences the regulation of this region by hormones (p. 163).”  Given this early speculation about the genetics of neuronal number and perhaps of neuronal density in the INAH-3, about regulation of this structure by hormones, and about male sexual orientation it is interesting to note that the presence of the X-linked Kalig 1 gene affects olfaction, GnRH, LH, androgen/estrogen ratios, and sexual behavior as a result of Kal protein deficiency.  Afflicted patients are genetically predisposed by a single gene to be both anosmic and hypogonadal.  In a small cohort, their sexual behavior is not unlike that of anosmic mammals of other species that lack sexual motivation.  However, Bobrow, Money, and Lewis (1971) have also indicated that these individuals even lack the uniquely species-specific human experience of falling in love.  It is possible, therefore, that a single gene might have dramatic effects on the development of human odor-associated sexual preferences.

    Even before Hamer and Copeland’s proposal with regard to Xq28, William J. Turner (personal communication, 1993) was interested in a possible link between genetics, GnRH neuronal system development, and olfaction and had this to say:

“. . . I am only now beginning to study the literature on olfaction.  I want to assure you that I shall not ever attempt to claim any primacy in the concept that the gene for Homosexuality Type 1 functions through olfaction, though I did come to the idea quite independently.  No, I did not even come to it that way.  In 1952 I was told by an aging gay patient that he had never made a mistake in choosing a partner.  He said, ‘I smell them.’  I would claim that the gene lies at Xq28.”


    In addition to the sex-linked loci, Xq28, several identified autosomal loci suggest a multi-gene regulation of a pathway involved in sexual orientation (Mustanski, Dupree, Nievergelt, Bocklandt, Schork, & Hamer, 2005).  This multi-gene regulation of the sexual orientation pathway is expected for any complex behavioral trait.  Notably, the region near 8p12 contains several candidate genes that might be linked to alterations in GnRH synthesis.  If true, these candidate genes lend support to the olfactory/pheromonal model of male preferences for other males.

    Despite the likely multi-gene regulation of any sexual orientation pathway, Bocklandt, Horvath, Vilain, and Hamer, (2005) have hypothesized that one central neuronal pathway establishes sexual attraction to either males or females, usually towards the opposite sex.  If sexual attraction is associated with homosexual and heterosexual orientations, the same neuronal pathway should be involved in either outcome.  From an ontogenetic and phylogenetic perspective, the GnRH neuronal pathway is the most likely central neuronal pathway involved in the development of sexual preferences.  Alterations in the development of the GnRH neuronal pathway or in the development of neuronal pathways that impact GnRH synthesis and pulsatile secretion could be expected to accompany alterations in sexual preferences for attractive physical features that are manifest in sexual orientation.


Pharmacological Influences on Hypothalamic GnRH Pulsatility and Behavior

    The number of potential influences on the development, maintenance, and function of the GnRH neuronal system allow for many possible unconscious affects on behavior.  For example, as noted above, Silverman et al. (1992) have found that only a few GnRH neurons implanted into the MPOA/AH are required to induce a mammalian GnRH pulse that is capable of activating the pituitary-gonadal axis in hypogonadal mice that lacked the GnRH gene.  If only a few GnRH neurons are required to activate the mammalian pituitary-gonadal axis, GnRH neuronal number/density are likely to make a difference in hypothalamic GnRH pulse frequency.

    Also, the GnRH neuronal number or neuronal density in the MPOA/AH may be critical to hypothalamic GnRH pulse frequency and to the unconscious affect of pheromones on sexual orientation.  It may also be that other connecting neurons and neuronal systems—many of which are sexually dimorphic in their structure and function—are very important to the regulation of the hypothalamic GnRH pulsatility that directs the HPG axis during development and the unconscious affect of the HPG axis on sexual behavior.  For example, noradrenergic, dopaminergic, serotoninergic, and opiotergic pathways, as well as inhibitory neurotransmitters like gamma­aminobutyric acid and excitatory amino acids like glutamic and aspartic acids and other brain peptides including pineal secretions like melatonin and corticotrophin ­releasing hormone and the complex interactions among them, are subtle but functional species-­specific influences on the electrochemical transmission of neuronal signals that the hypothalamus translates to the chemical signal GnRH (Grumbach & Styne, 1992).  Individually, many of these influences on the frequency and amplitude of the GnRH pulse are linked through pharmacology and therapeutic drugs to reproductive function, sexual behavior, and various neurodegenerative diseases, some of which are manifest with olfactory deficits.  Collectively, these influences also make it difficult to establish whether, or how, reciprocal relationships among hypothalamic GnRH pulsatility, LH/FSH ratios, secretion of other hormones like E2 and T, and factors that alter neurotransmission affect human sexual behavior.  However, when it comes to mammalian pheromones, including putative human pheromones, their affect on sexual behavior is indicated by a measurable change in LH, which may vary with a number of other likely influences on hypothalamic GnRH pulse frequency and covary with sexual preferences.

Other Influences on Hypothalamic GnRH Pulsatility and Their Relevance to Other Theories about Sexual Orientation

    Alterations in the immune system seem likely to influence the development of the GnRH neuronal system.  For example, various cytokines produced during an immune reaction can modulate the HPG axis, probably by inducing changes in the activity of hypothalamic GnRH neurons (Igaz, Salvi, Rey, Glauser, Pralong, & Gaillard, 2006).  In addition, the GnRH receptor gene is a potential multi-system target gene that enables feedback between the neuroendocrine, immune, and stress-response systems in multiple tissues via autocrine, paracrine, and endocrine signaling (Hapgood, Sadie, van Biljon, & Ronacher, 2005).  Thus, the development of the GnRH neuronal system might be altered by two mechanisms of the immune system—effects on GnRH synthesis or effects on GnRH receptors.

    It seems likely that development of the GnRH neuronal system of a fetus could be influenced by the maternal immune system, perhaps in a manner somewhat comparable to the way that the GnRH-directed neuroendocrine system is influenced by Kal protein deficiency.  A maternal immune system influence on GnRH might explain the link to homosexual orientation from the number of human male offspring, such as the fraternal birth order effect.  In the fraternal birth order effect theory, fetal cells from the male are recognized as foreign by the mother’s immune system, and she produces antibodies to the male cells.  In subsequent male fetuses, these antibodies cross the placental barrier and enter the brain of the male.  In theory, these antibodies in the male brain somehow divert sexual differentiation from the male-typical pathway.  Thus, with each subsequent male birth, it is more likely that an individual male will later be attracted to men rather than women (Blanchard & Bogaert, 2004).  If the increasing number of antibodies with each male birth also impacts the GnRH neuronal system or if they impact GnRH receptors, they could influence the organizational sexual differentiation of the olfactory system.  This scenario would lend further support to an olfactory/pheromonal model of male sexual preferences for other males.  However, this hypothesis is not the only immune system hypothesis that supports an olfactory/pheromonal model.

    The immune system probably developed only after organisms developed the ability to recognize self and non-self (Boehm, 2006).  Immune system cells use sexually dimorphic cell-surface molecules to recognize self and non-self.  Immune cell subsets (ICS) with the capability to recognize self and non-self exist in mucosal epithelia.  Further, immune system-mediated events in mucosal epithelia can be transduced into neuronal signals that activate the CNS.  Binstock (2001) has suggested that sexual orientation might be associated with an activated response encoded within mucosal ICS.  If these mucosal ICS are found in the olfactory system they might differentially transduce MHC/HLA-associated, pheromone-generated neuronal signals to the CNS.  This differential signal transduction could enable the discrimination of body odor that is more similar to self than to non-self.  Accordingly, this hypothesis lends psychophysiological credibility to the likelihood of psychoneuroimmunological unconscious affects of olfactory/pheromonal input on homosexuals who prefer the natural body odor of other homosexuals (Martins et al., 2005).  This preference may in part be determined by the activational effect of MHC/HLA-associated pheromones on ICS in the olfactory mucosa that influence the electrical transmission of neuronal signals that the hypothalamus translates to the chemical signal GnRH.  As with mammalian pheromones that are associated with hormone metabolism, the activity of MHC-associated pheromones is not restricted to activity generated from the VNO.  It has recently become clearer that MHC-associated pheromones, like other mammalian pheromones, can be processed in the main olfactory system (Spehr, Kelliher, Li, Boehm, Leinders-Zufall, & Zufall, 2006).  HLA-associated putative human pheromones are also expected to be processed in the main olfactory system of humans.

    The activational effect of olfactory/pheromonal input on ICS in the olfactory mucosa of humans might explain differences in the way men and women process olfactory signals that convey information about the immune system, like HLA-associated odors.  Pause, Krauel, Schrader, Sojka, Westphal, Müller-Ruchholtz, and Ferstl (2006) have found that the unconscious affect of HLA-related body odors occurs more rapidly when men and women are exposed to HLA-similar donors and that subsequent evaluative processing of these odors activates more neurons than does processing body odor from HLA-dissimilar donors.  Further, when men and women were exposed to odors from the same sex, HLA-associated brain responses were processed in different parts of the brain: male, frontal; female parietal, which correlated with odor ratings.  Pause et al. (2006) indicated that differences in the way that men and women process HLA-associated odor signals point to the likelihood that same sex odor plays a different role in behavior than odor from the opposite sex.  Same sex odor may be most important to male-male competitive behaviors and to female-female communal behavior.  This speculation correlates with results that show male and female adolescents recognize the odor of platonic friends (Olsson, Barnard, & Turri, 2004).  Taken together, the ability to recognize olfactory/pheromonal cues, either from communal friends or from competitors, and the ability to differentially process or respond to these cues could play a role in establishing long-term platonic alliances that mimic HLA-associated kinship alliances, like those found in the Israeli kibbutzim.  Olfactory/pheromonal input could bias the development and maintenance of platonic male-male, female-female, female-male, male-female, and group alliances, which might be adaptive for cooperative efforts (for review, see Kohl & Francoeur, 1995, 2002).


    Sex differences in pheromone production lend further support to an olfactory/pheromonal model for the development of preferences that may be manifest in human male sexual behavior.  Although no human pheromone has been isolated, effects on hormones of putative human pheromones are readily apparent, as evidenced by the GnRH-directed LH response.  Clearly, putative human pheromones are sexually dimorphic and their effects on hormone levels are sexually dimorphic.  If, as Martins et al. (2005) indicated, males distinguish among different pheromones and some of these males establish a preference for the pheromones of other males, the question arises, “How might they do this?”  An investigation of adrenal hormone metabolism, which may be important to putative human pheromone production and distribution, might provide some clues.

    A focus on adrenal hormone metabolism, rather than on gonadal hormone metabolism, with regard to pheromone production and distribution is reasonable because mammalian pheromones are largely species-specific, and humans are unique in having adrenals that secrete large amounts of the prohormone, dehydroepiandrosterone (DHEA) and its sulfate, DHEA-S, into the bloodstream of males and females.  Even non-human primates produce only ~10% of the DHEA found in humans (see Celec & Starka, 2003).  The relative enzymatic activities of aromatase and other enzymes and relative levels of DHEA or its derivatives determine whether DHEA will be preferentially converted in peripheral tissues into androstenedione or androstenediol and then into potent androgens and estrogens, which is how androstenedione and androstenediol maintain a close correlation between the concentration of androgens and estrogens in the blood (Adams, 1985).

    Androstenedione is the principle metabolite of DHEA.  The C19 steroids, androsterone (A) and etiocholanolone (E), are the enzymatically reduced metabolites of androstenedione.  Early studies noted that various characteristics of A and E are sexually dimorphic.  For example, human blood concentration, metabolism, and localization of A and E are sexually dimorphic.  Thus, the A/E ratio may vary between males and females.

    The A/E ratio also may vary with sexual orientation.  Massion-Verniory (1957) predicted that a by-product of hormone metabolism found in urine would be found to differentiate homosexuals from heterosexuals.  Subsequently, Margolese (1970) has reported that A/E ratios in urine samples could be used to determine whether a particular urine sample came from a heterosexual male or a homosexual male.  His finding that homosexual males have decreased urinary A/E ratios was confirmed by Evans (1972), Margolese and Janiger (1973), and Friedman, Dyrenfurth, Linkie, Tendler, & Fleiss (1977).

    Although the production of DHEA and DHEA-S and changes in A/E ratios are very sensitive to other influences, including influences from the immune system, Margolese and Janiger (1973) have speculated that the decreased urinary A/E ratios in homosexual males indicate a shift in a metabolic pathway toward the female side.  They also raised the possibility of possible enzyme induction by prenatal hormone conditions, which include GnRH-directed, LH-facilitated prenatal hormone conditions.  They proposed “. . . that the metabolic pathway which results in a relatively high androsterone value is associated with sexual preference for females by either sex, whereas a relatively low androsterone value is associated with sexual preference for males by either sex (p. 210).”

    Typically, aromatic compounds found in urine can also be found in different concentrations in the peripheral blood, saliva, axillary, and other secretions of men and women.  Accordingly, the A/E ratio is likely to also contribute to odor components in salivary secretions and in axillary or other bodily secretions and, thus, to sex differences in putative human pheromone production.  Manifestation of the A/E ratio in putative pheromone production could explain the ability of males who prefer the natural body odor of other males to distinguish among odors that are representative of the A/E ratio in other males and to establish a sexual preference for male body odor, as indicated by Martins et al., (2005).


    The evolutionary adaptation of any complex behavioral trait might be better addressed in theory rather than through a model that integrates biological facts, which are more rigid.  Either the biological facts support the model, or not.  The facts integrated here allow little room for assumptions about how sexual preferences develop.  These facts also show that chemical communication is adaptive across species.  However, this does not mean that sexual preferences, which appear to develop in accord with this olfactory/pheromonal model, are an evolutionary adaptation.

    Instead, the biological facts simply support the notion that a genetic propensity evolved to predispose complex behaviors via molecular biochemical mechanisms. These mechanisms are influenced by olfactory/pheromonal input from the social environment.  It is likely that classical conditioning of a mammalian genetically predisposed GnRH-directed, LH facilitated sexual response cycle results in the development of sexual preferences and sexual behavior.  However, all instances throughout life when conditioning occurred would need to be considered in attempts to predict or correctly classify the conditioned sexual preferences or the sexual behaviors that result from this conditioning.

    In this olfactory/pheromonal model, genetically predisposed sexual preferences are influenced by sensory input from the social environment, and these sexual preferences may be manifest in the sexual behavior of individual human males.  Individual human males who manifest sexual preferences in their sexual behavior are collectively classified by others as homosexual, heterosexual, bisexual, transsexual, asexual, or as experiencing an odor-associated paraphillia.  In this regard, there appears to be a common olfactory/pheromonal theme that is employed, although such classifications do not mean that any individual male fits into a specific category of sexual behavior.

    The wide variety of sexual preferences manifest in sexual behavior indicates that categories of behavior might need to be redefined to include or exclude individuals whose sexual preferences do not always fit.  A better approach might be to eliminate these categories or better understand their limitations.

    The role that olfactory/pheromonal conditioning plays in the development of sexual preferences might be helpful in considering such limitations, as might be the GnRH-directed, LH facilitated response to the natural body odor of conspecifics.  Minimally, the use of biological facts to classify individuals might show how idiosyncratic preferences are developed and how these preferences are manifest in the sexual behavior among individuals and in each species.  In this regard, the development of olfactory/pheromonal preferences associated with sexual behavior might be as idiosyncratic as odor-associated food preferences and ingestive behavior.

     Sexual preferences might even compare well with categories of carnivore and herbivore ingestive behavior.  Further classification of the animalistic behavior of human “carnivores” and “herbivores” might include preferences for steak or broccoli, leading to the categories of steak-lover or chicken-lover, and broccoli-lover versus broccoli-hater, ad infinitum due to the nearly infinite number and combinations of categories that could be associated with food preferences.

    In contrast, this olfactory/pheromonal model for male attraction to other males focuses only on the development of a few human sexual preferences.  Consideration for how any human sexual preference develops probably should be included before any individual preference is classified or categorized.  Without the development of personal preferences, there would be no human sexual behavior.  Clearly, unicellular organisms and other non-human multicellular organisms exhibit sexual preferences and sexual behaviors that are functionally based in the molecular biochemical aspects of chemical communication.  However, many organisms are limited to species-wide categories of relatively limited behaviors.

    We can expect that with further human research, results will continue to show a common theme that will permit more focus on various aspects of chemical communication, whether the results of chemical communication are theoretically adaptive and whether the associated sexual preferences and sexual behaviors are reproductively adaptive.


    Here, I have detailed an olfactory/pheromonal model for the evolved development of male sexual preferences (see Kohl, 2005 for a less detailed presentation).  Other models pertinent to the evolved development of mammalian male sexual preferences can now be compared to this one. If no comparable model exists, this model probably is the right one.

    A comparable model would incorporate a proximate theory of human sexual attraction that extends well to male-male sexual attraction.  However, the evolutionary history of any theory of human sexual attraction should be detailed before any attempt is made to determine whether the proximate mechanisms are adaptive.  Similarly, the evolutionary history of sexual orientation should be detailed (e.g., using animal models) before labels are associated with human sexual preferences or sexual behaviors.

    Animal models support the likelihood that sexual preferences and sexual behaviors associated with human sexual orientation may be manifestations of unconscious affects.  Unconscious affects include a continuum of subtle response gradations that remain largely outside the realm of current consideration.  Studies either initially or ultimately label participants based on singular factors and responses in the absence of any model.  In this olfactory/pheromonal model, unconscious affects are not likely to be completely differentiated, dimorphic, or dichotomous.  Thus, except for concision, it would be a mistake to consider either sexual preferences or sexual behaviors to be completely differentiated, dimorphic, or dichotomous.

    In this olfactory/pheromonal model, sexual preferences develop under the influence of what might be metaphorically termed “the mind’s eyes.”  The mind’s eyes consist of a distributed neuronal network that collectively integrates responses to olfactory and visual input.  Studies of non-human mammals indicate that visual input has relatively little impact on the development of sexual preferences.  Accordingly, this model focuses on olfactory input.  It incorporates well-conserved molecular biochemical mechanisms for the identification of conspecifics and for the development of sexual preferences.  This model also provides a detailed gene–cell–tissue–organ–organ system pathway from sexually dimorphic hormone-associated olfactory input and from MHC/HLA-associated olfactory input to sexually dimorphic behavior.

    Conditioning of a sexually dimorphic visual response to androgenic, estrogenic, progesteronic, and MHC/HLA-associated pheromones appears to play the primary role in mammalian mate choice.  This conditioning integrates the effects of olfactory/pheromonal input on hormones with the unconscious affects of hormones on behavior.  The ability to form neural connections between appetitive or aversive food odors and visual input explains the psychophysiological effects of olfactory/pheromonal conditioning.  Innate sexual dimorphism in the mammalian olfactory system and ongoing olfactory/pheromonal effects on sexual differentiation of the brain and of behavior allow the inclusion of specific examples from neuroscience that support this model.

    The common origin of olfactory and GnRH neurons during mammalian fetal development and the sexually dimorphic MPOA/AH activation of these hormone-secreting neurons by mammalian pheromones, including putative human pheromones, attest to the likelihood that established findings in other species extend well to humans.  This includes findings on the role that MPOA/AH activation may play in the development of mammalian sexual preferences.

    Broad coverage of the olfactory/pheromonal influence on GnRH during sexual differentiation includes downstream effects of GnRH on gonadotropin secretion, androgen secretion, and estrogen secretion.  Also considered are T conversion by aromatase to E2, E2 receptor content in sexually dimorphic brain tissues, sexually dimorphic size or neuronal density of specific hypothalamic nuclei, and the likelihood of X-linked and autosomal links to the genetic, including immunological, underpinnings of sexual preferences.  This extended coverage integrates interdisciplinary research to help further a congruent explanation of how males develop sexual preferences for other males.

    Animal-tested therapeutic and psychotropic drug effects on the GnRH neuronal system and on behavior indicate that this olfactory/pheromonal model for the development of male sexual preferences for other males has more explanatory power than other theories.  Isolation of putative human pheromones through their demonstrable effects on the GnRH-directed secretion of LH may facilitate a paradigm shift that will focus more interest in the role that olfactory/pheromonal input plays in the development of human sexual preferences.

    Minimally, the details of this olfactory/pheromonal model serve as a reminder that olfaction and pheromones determine, without conscious thought, aspects of mate choice and sexual behavior in other mammals and in most, if not all, non-mammalian species that sexually reproduce (for review, see Kohl & Francoeur, 1995, 2002; Wyatt, 2003).  The evidence presented here also suggests that pheromone-affected psychophysiological responses are associated with sexual preferences of males for other males.


    In the two months of preparation that preceded the submission of this review, interdisciplinary research findings have continued to support an olfactory/pheromonal model for the development of sexual preferences that may be manifest in sexual behavior.  Publication dates of several referenced citations indicate that support is increasing.  A growing body of evidence affirms that brain activity mirrors actions and reactions and that behavior develops from substrates in the brain.  Neuroanatomical substrates in the brain that respond to actions with neuroendocrine reactions are readily linked to the development of behavior.  GnRH-directed neuroendocrine reactions are primary indicators of olfactory/pheromonal cause, unconscious affects, and cognitive effects on sexual behavior.  Olfactory/pheromonal conditioning allows unconscious affects and cognitive effects on behavior to elicit reciprocal actions and reactions.  This reciprocity is unparalleled in attempts to explain how sexual preferences develop.

    In the month of this review’s submission, Kranz and Ishai (2006) have shown that activation in the reward circuitry of the brain by faces of males or females varies with human sexual orientation.  Specifically noted was activation in the OFC, which, as noted earlier, is a brain area that links conscious choice to unconscious affects on emotions generated from the limbic system.  The OFC response of males who had a sexual preference for other males was more typical of a female viewing a male face.  The OFC response of females who had a sexual preference for other females was more typical of a male viewing a female face.  These authors posited that male and female faces evoke activity in the OFC, which may play a more general, less sex-typical, role in representing the neural reward value of the faces of potential sexual partners.  The mind’s eyes approach to the interpretation of OFC responses suggests that face-evoked OFC activation may largely be due to olfactory/pheromonal conditioning of sexual preferences and that OFC activation has minimal, if any, association with reproductive function.  Thus, the effect of olfactory/pheromonal input on hormones and the unconscious affect of hormones on consciously perceived physically attractive features may be more associated with sexual preferences than with reproductive function.

    In the week of this review’s submission, Vogt (2006) provided a perspective on the molecular biochemical mechanisms of vision and olfaction that “. . . offer a rich comparative landscape for the evolution of what is ultimately a genetic interface between animals and their environment (p. 3).”  Although visual input is processed more rapidly, genetic divergence in coding for olfactory receptor proteins ensures increased sensitivity and specificity to olfactory input.  Also, while both olfactory and visual receptors detect sensory input from the environment, olfactory receptors are more finely attuned to respond to olfactory input from the social environment—a more specific function.  Visual receptors are finely attuned to respond to variations in light.  Subtle variations in visual input are less likely to influence the development of mammalian sexual preferences than are subtle variations in olfactory/pheromonal input, which are associated with levels of hormones and MHC/HLA diversity.  However, subtle variations in visual input that are associated with olfactory/pheromonal input could influence sexual preferences via olfactory/pheromonal conditioning of the visual response.  This conditioning provides the only known molecular biochemical means by which variations in visual input could acquire biological significance during the development of sexual preferences.

    It is not likely that subtle variations in visual input are inherently processed differently in males than in females and not likely that these variations elicit sexually dimorphic variability in reciprocal effects on hormones and their unconscious affects on the development of sexual preferences.  Further, to directly affect behavior in the same manner that olfactory/pheromonal input affects behavior, visual input would need to activate genes in cells of hormone-secreting brain tissue that have been linked to the development of sexual preferences.  The processing of visual input would also need to be somewhat permanently altered by hormones but allow for neuronal plasticity during the ongoing development of sexual preferences.  Thus, the most recent perspective on the molecular biochemical mechanisms of a genetic interface between animals and their environment favors an olfactory/ pheromonal interface.


    Rarely do sex researchers address the ongoing philosophical debate between canonical neo-Darwinism and Biblical creation.  Perhaps this is because any debate between scientific theory and religion arises from distinctly different domains of cognitive thought.  Does the acceptance of Darwin’s theory represent the glorification of Science pitted against religion, or is it a means by which Science and religion might be integrated?  Integration of Science and religion might be achieved by recognizing that the key components of this olfactory/pheromonal model appear to be as irreducibly complex as the basic tenets of evolution and the basic tenets of religion.

    From an evolutionary perspective, highly conserved GnRH peptide ligand/receptor signaling mechanisms are the molecular biochemical mechanisms for sexual reproduction in all organisms.  These signaling mechanisms also appear to play an integral role in the development of sexual preferences.  From a religious perspective, these signaling mechanisms dictate that the creation of life, which begets life, also allows for the creation of diversified life through the same mechanisms.  These mechanisms allow life to recognize the difference between self and non-self and to respond to this difference.

    Perhaps the creation of diversified human life gave us the ability to recognize differences between our sexual behavior and the sexual behavior of others.  Since all life does not beget diversified life, those who judge sexual preferences that do not seem to result in diversified life may be judging creation itself.

    It is easy to understand how someone could judge a particular sexual preference, without thought.  Unconscious affects that are manifest in the development of human sexual preferences are, by their nature, a part of diversified life that few people think about.  What we think about human sexual preferences becomes less meaningful when we realize that most of sexual behavior is not what we cognitively think it should be.  Indeed, the largest contributor to sexual preferences that are manifest in the sexual behavior of any species appears to be unconscious affect.  This also appears to be the basis for diversified life.


This paper is dedicated to Lee Ellis, Lyn Ebertz, and Eugene Garfield for their roles in the International Behavioral Development Symposiums held in Minot, North Dakota in 1995, 2000, and 2005.

The author declares his commercial interest in the formulation and in the marketing of fragrance products containing putative human pheromones.


Brush, F. R., & Levine, S. (1989). Psychoendocrinology. San Diego: Academic Press.

Fisher, H. (1992). Anatomy of Love: The Natural History of Monogamy, Adultery, and Divorce. New York: W.W. Norton.

McEwen, B., & Schmeck Jr., H. M. (1994). The Hostage Brain. New York: Rockefeller University Press.

Moir, A., & Jessel, D. (1991). Brain Sex: The Real Difference Between Men and Women. New York: Carol Publishing Group.

Persky, H. (1987). Psychoendocrinology of Human Sexual Behavior. New York: Praeger.



Adams, J. B. (1985). Control of secretion and the function of C19-delta 5-steroids of the human adrenal gland. Molecular and Cellular Endocrinology, 41, 1-17.

Bakker, J., Baum, M. J., & Slob, A. K. (1996). Neonatal inhibition of brain estrogen synthesis alters adult neural Fos responses to mating and pheromonal stimulation in the male rat. Neuroscience, 74, 251-260.

Bakker J., van Ophemert, J., & Slob, A. K. (1996). Sexual differentiation of odor and partner preference in the rat. Physiology & Behavior, 60, 489-494.

Bakker, J, De Mees, C., Douhard, Q., Balthazart, J., Gabant, P., Szpirer, J., & Szpirer, C. (2006). Alpha-fetoprotein protects the developing female mouse brain from masculinization and defeminization by estrogens. Nature Neuroscience, 9, 220-226.

Barni, T., Maggi, M., Fantoni, G., Granchi, S., Mancina, R., Gulisano, M., Marra, F., Macorsini, E., Luconi, M., Rotella, C., Serio, M., Balboni, G. C., & Vannelli, G. B. (1999). Sex steroids and odorants modulate gonadotropin-releasing hormone secretion in primary cultures of human olfactory cells. The Journal of Clinical Endocrinology and Metabolism, 84, 4266-4273.

Barran, P. E., Roeske, R. W., Pawson, A. J., Sellar, R., Bowers, M. T., Morgan, K., Lu, Z. L., Tsuda, M., Kusakabe, T., & Millar, R. P. (2005). Evolution of constrained gonadotropin-releasing hormone ligand conformation and receptor selectivity. The Journal of Biological Chemistry, 280, 38569-38575.

Bobrow, N. A., Money, J., & Lewis, V. J. (1971). Delayed puberty, eroticism and sense of smell: A psychological study of hypogonadotropinism, osmatic and anosmatic (Kallmann’s syndrome). Archives of Sexual Behavior, 1, 329-344.

Berardi, N., Pizzorusso, T., & Maffei, L. (2000). Critical periods during sensory development. Current Opinion in Neurobiology, 10, 138–145.

Berliner, D. L., Monti-Bloch, L., Jennings-White, C., & Diaz-Sanchez, V. (1996). Functionality of the human vomeronasal organ (VNO): Evidence for steroid receptors. The Journal of Steroid Biochemistry and Molecular Biology, 58, 259-265.

Binstock, T. (2001). An immune hypothesis of sexual orientation. Medical Hypotheses, 57, 583-590.

Black, A. H., & Prokasy, W. F. (1972). Classical conditioning II: Current research and theory. New York: Appleton-Century-Crofts.

Blanchard, R, & Bogaert, A. F. (2004). Proportion of homosexual men who owe their sexual orientation to fraternal birth order: An estimate based on two national probability samples. American Journal of Human Biology, 16, 151-157..

Bocklandt, S., Horvath, S., Vilain, E., & Hamer, D. H. (2005). Extreme skewing of X chromosome inactivation in mothers of homosexual men. Human Genetics, 21, 1-4.

Boehm, T. (2006) Co-evolution of a primordial peptide-presentation system and cellular immunity. Nature Reviews. Immunology  6, 79-84.

Boehm, U., Zou, Z., & Buck, L.B. (2005). Feedback loops link odor and pheromone signaling with reproduction. Cell, 123, 683-695.

Brennan, P. A. (2004). The nose knows who’s who: Chemosensory individuality and mate recognition in mice. Hormones and Behavior, 46, 231-40.

Bressler, S. C., & Baum, M. J. (1996). Sex comparison of neuronal Fos immunoreactivity in the rat vomeronasal projection circuit after chemosensory stimulation. Neuroscience, 71, 1063-1072.

Byne, W., Tobet, S., Mattiace, L. A., Lasco, M. S., Kemether, E., Edgar, M. A., Morgello, S., Buchsbaum, M. S., Jones, L. B. (2001). The interstitial nuclei of the human anterior hypothalamus: an investigation of variation with sex, sexual orientation, and HIV status. Hormones and Behavior, 40, 86-92.

Celec, P., & Starka, L. (2003). Dehydroepiandrosterone - is the fountain of youth drying out? Physiological Research, 52, 397-407.

Choosing the perfect mate a matter of who smells best. (1993, October 10). Las Vegas Review-Journal/Sun, p. 2A.

Cooke, B., Hegstrom, C. D., Villeneuve, L. S., & Breedlove, S. M. (1998). Sexual differentiation of the vertebrate brain: principles and mechanisms. Frontiers in Neuroendocrinology, 19, 253-286.

Cornwell, R. E., Boothroyd, L., Burt, D. M., Feinberg, D. R., Jones, B. C., Little, A. C., Pitman, R., Whiten, S., & Perrett, D. I. (2004). Concordant preferences for opposite-sex signals? Human pheromones and facial characteristics. Proceedings of the Biological Sciences /The Royal Society, 271, 635-640.

Diamond, M., Binstock, T., & Kohl, J. V. (1996). From fertilization to adult sexual behavior. Hormones and Behavior, 30, 333-353.

Dörner G. (1988). Neuroendocrine response to estrogen and brain differentiation in heterosexuals, homosexuals and transsexuals. Archives of Sexual Behavior, 17, 57-75.

Dörner, G., Docke, F., Gotz, F., Rohde, W., Stahl, F., & Tonjes, R. (1987). Sexual differentiation of gonadotrophin secretion, sexual orientation and gender role behavior. Journal of Steroid Biochemistry, 27, 1081-1087.

Evans, R. B. (1972). Physical and biochemical characteristics of homosexual men. Journal of Consulting and Clinical Psychology, 39, 140-147.

Everett, J. W. (1994). Pituitary and hypothalamus: perspectives and overview. In E. Knobil and J. D. Neill (Eds.), The Physiology of Reproduction, Volume 1, (pp. 1509-1526). New York: Raven Press.

Fischer, H., Sandblom, J., Herlitz, A., Fransson, P., Wright, C. I., & Backman, L. (2004). Sex-differential brain activation during exposure to female and male faces. Neuroreport, 15, 235-8.

Friedman, R. C., Dyrenfurth, I., Linkie, D., Tendler, R., & Fleiss, J. L. (1977). Hormones and sexual orientation in men. The American Journal of Psychiatry, 134, 571-572.

Gladue, B. A., Green, R., & Hellman R. E. (1984) Neuroendocrine response to estrogen and sexual orientation. Science, 225, 1496-1498.

Gottfried, J. A., & Dolan, R. J. (2003). The nose smells what the eye sees: Crossmodal visual facilitation of human olfactory perception. Neuron, 39, 375-386.

Gottfried, J. A., O’Doherty, J., & Dolan, R. J. (2002). Appetitive and aversive olfactory learning in humans studied using event-related functional magnetic resonance imaging. The Journal of Neuroscience, 22, 10829-10837.

Graham, J. M., & Desjardins, C. (1980). Classical conditioning, induction of luteinizing hormone and testosterone secretion in anticipation of sexual activity. Science, 210, 1039-1041.

Grosser, B. I., Monti-Bloch, L., Jennings-White, C, & Berliner, D. L. (2000). Behavioral and electrophysiological effects of androstadienone, a human pheromone. Psychoneuroendocrinology, 25, 289-99.

Grumbach, M. M., & Styne, D. M. (1992). Puberty: Ontogeny, neuroendocrinology, physiology & disorders. In J. D. Wilson and D. F. Foster (Eds.), Textbook of endocrinology, 8th edition, (pp. 1139-1222). Philadelphia: W. B. Saunders.

Hamer, D., & Copeland, P. (1994). The science of desire. New York: Simon and Schuster.

Hapgood, J. P., Sadie, H., van Biljon, W., & Ronacher, K. (2005). Regulation of expression of mammalian gonadotrophin-releasing hormone receptor genes. Journal of Neuroendocrinology, 17, 619-638.

Havlicek, J., Dvorakova, R., Bartos, L., & Flegr, J. (2006). Non-advertized does not mean concealed: Body odour changes across the human menstrual cycle. Ethology, 112, 81-90.

Havlicek, J., Roberts, S. C., & Flegr, J. (2005). Women’s preference for dominant male odour: Effects of menstrual cycle and relationship status. Biology Letters (Online), 1, 256-259.

Hellhammer, D. H., Hubert, W., & Schurmeyer, T. (1985).  Changes in saliva testosterone after psychological stimulation in men. Psychoneuroendocrinology, 10, 77-81.

Hoffman, G. E., Lee, W. S., Attardi, B., Yann, V., & Fitzsimmons, M. (1990). Luteinizing hormone-releasing hormone neurons express c-fos antigen after steroid activation. Endocrinology, 126, 1736–1741.

Houtsmuller, E. J., Brand, T., De Jonge, F. H., Joosten, R. N. J. M. A., Van De Poll, N. E., & Slob, A. K. (1994). SDN-POA volume, sexual behavior, and partner preference of male rats affected by perinatal treatment with ATD. Physiology & Behavior, 56, 535-541.

Hull, E. M., Meisel, R. L., Sachs, B. D. (2002). Male sexual behavior. In D. W. Pfaff,  A. P. Arnold, A. M. Etgen, S. E. Fahrbach, and R. T. Rubin (Eds.), Hormones, Brain and Behavior, Volume 1, (pp. 3-137). San Diego: Academic Press.

Igaz, P., Salvi, R., Rey, J. P., Glauser, M., Pralong, F. P., & Gaillard, R. C. (2006). Effects of cytokines on GnRH gene expression in primary hypothalamic neurons and in GnRH neurons immortalized conditionally. Endocrinology, 147, 1037-1043.

Ishai, A., Ungerleider, L. G., Martin, A., Schouten, J. L., & Haxby, J. V. (1999). Distributed representation of objects in the human ventral visual pathway. Proceedings of the National Academy of Sciences, 96, 9379-9384.

Jacob, S., & McClintock, M. K. (2000). Psychological state and mood effects of steroidal chemosignals in women and men. Hormones and Behavior, 37, 57-78.

Jakacki, R. I., Kelch, R. P., Sauder, S. E., Lloyd, J. S., Hopwood, N. J., & Marshall, J. C. (1982). Developmental changes in neuroendocrine regulation of gonadotropin secretion in gonadal dysgenesis. The Journal of Clinical Endocrinology and Metabolism, 55, 453-458.

Jasienska, G., Ziomkiewicz, A., Ellison, P. T., Lipson, S. F., & Thune, I. (2004). Large breasts and narrow waists indicate high reproductive potential in women. Proceedings of the Biological Sciences /The Royal Society, 271, 1213-1217.

Jutte, A. (1996, August). Cognitive and physiological responses of men to female pheromones. International Society for Human Ethology, 13th Conference, Vienna, Austria.

Karlson, P., & Luscher, M. (1959). Pheromones: A new term for a class of biologically active substances. Nature, 183, 55-56

Kauth, M. R. (2005). Revealing assumptions: Explicating sexual orientation and promoting conceptual integrity. Journal of Bisexuality, 5, 79-105.

Kindon, H. A., Baum M. J., & Paredes, R. J. (1996) Medial preoptic/anterior hypothalamic lesions induce a female-typical profile of sexual partner preference in male ferrets. Hormones and Behavior, 30, 514-527

Kloek, J. (1961). The smell of some steroid sex-hormones and their metabolites. Reflections and experiments concerning the significance of smell for the mutual relation of the sexes. Psychiatria, Neurologia, Neurochirurgia, 64, 309-344.

Kohl, J. V. (1992, November). Luteinizing hormone: The link between sex and the sense of smell? Annual Meeting of the Society for the Scientific Study of Sexuality, San Diego, CA

Kohl J. V., Atzmueller, M., Fink, B., & Grammer, K. (2001). Human pheromones: Integrating neuroendocrinology and ethology. Neuroendocrinology Letters, 22, 309-321.

Kohl, J. V., & Francoeur, R. (1995). The scent of Eros: Mysteries of odor in human sexuality. New York: Continuum.

Kohl, J. V., & Francoeur, R. (2002). The scent of Eros: Mysteries of odor in human sexuality. iUniverse.

Kohl, J. V. (2005, August). Human pheromones, neuroscience, and male homosexual attraction. International Behavioral Development Symposium, Minot, ND and Entelechy: Mind and Culture, 6. <entelechyjournal.com/kohl.html>.

Kovacs, G., Gulyas, B., Savic, I., Perrett, D. I., Cornwell, R. E., Little, A. C., Jones, B. C., Burt, D. M., Gal, V., & Vidnyanszky, Z., (2004). Smelling human sex hormone-like compounds affects face gender judgment of men. Neuroreport, 15, 1275-1277.

Kula, A., Dulko, S., Pawlikowski, M., & Imielinski, K. (1986). A nonspecific disturbance of the gonadostat in women with transsexualism and isolated hypergonadotropism in the male-to- female disturbance of gender identity. Experimental and Clinical Endocrinology, 87, 8-14.

Kranz, F., & Ishai, I. (2006) Face perception is modulated by sexual preference. Current Biology: CB, 16, 63-68.

LaFerla, J. J., Anderson, D. L., & Schalch, D. S. (1978). Psychoendocrine response to sexual arousal in human males. Psychosomatic Medicine, 40, 166-172.

Law-Smith, M. J., Perrett, D. I.,  Jones, B. C., Cornwell, R. E., Moore, F. R., Feinberg, D. R., Boothroyd, L. G., Durrani, S. J., Stirrat, M. R., Whiten, S., Pitman, R. M., & Hillier, S. G. (2005). Facial appearance is a cue to oestrogen levels in women. Proceedings of the Royal Society of London. Series B, 273, 135-140.

LeRoith, D., Shemer, J., & Roberts, C. T., Jr. (1992). Evolutionary origins of intercellular communication systems: Implications for mammalian biology. Hormone Research, Supplement2, 38, 1-6.

LeVay, S. (1991). A difference in hypothalamic structure between heterosexual and homosexual men. Science, 253, 1034-1037.

Lledo, P. M., & Gheusi, G. (2003). Olfactory processing in a changing brain. Neuroreport, 14, 1655-1663.

Loumaye, E., Thorner, J., & Catt, K. J. (1982).Yeast mating pheromone activates mammalian gonadotrophs: Evolutionary conservation of a reproductive hormone? Science, 218, 1323-1325.

Margolese, M. S. (1970). Homosexuality: A new endocrine correlate. Hormones and Behavior, 1, 151-155.

Margolese, M. S., & Janiger, O. (1973). Androsterone-etiocholanolone ratios in male homosexuals. British Medical Journal, 3, 207-210.

Martins, Y., Preti, G., Crabtree, C. R., Runyan, T., Vainius, A., & Wysocki, C. J. (2005) Preference for human body odors is influenced by gender and sexual orientation. Psychological Science,16, 694-701.

Massion-Verniory, L. (1957). Level of 17-ketosteroids in various male sexual abnormalities. Acta Neurological et Psychiatrica Belgica, 57, 890-897.

Meisel, R. L., & Sachs, B. D. (1994). The physiology of male sexual behavior. In E. Knobil and J. D. Neill (Eds.), The physiology of reproduction, volume 2, (pp. 3-105). New York: Raven Press.

Meredith, M. (2001). Human vomeronasal organ function: a critical review of best and worst cases. Chemical Senses 26, 433-445.

Meredith, M., & Fernandez-Fewell, G. (1994). Vomeronasal system, LHRH, and sex behaviour. Psychoneuroendocrinology, 19, 657-672.

Michael, R. P., & Keverne, E. B. (1971) Pheromones: Isolation of male sex attractants from a female primate. Science, 172, 964-966.

Mustanski, B. S., Dupree, M. G., Nievergelt, C. M., Bocklandt, S., Schork, N. J., & Hamer, D. H. (2005). A genomewide scan of male sexual orientation. Human Genetics, 116, 272–278

Nixon, A., Mallet, A. I., & Gower, D. B. (1988). Simultaneous quantification of five odorous steroids (16-androstenes) in the axillary hair of men. Journal of Steroid Biochemistry, 29, 505-510.

Ober, C., Weitkamp, L. R., Cox, N., Dytch, H., Kostyu, D., & Elias, S. (1997). HLA and mate choice in humans. American Journal of Human Genetics, 61, 497–504.

Olsson, S. B., Barnard, J., & Turri, L. (2004, April). The scent of friendship: High school students research the mysteries of human odor recognition. 26th Annual Meeting of the Association for Chemoreception Sciences, Sarasota, FL.

Oomura, Y., Yoshimatsu, H., & Aou, S. (1983). Medial preoptic and hypothalamic neuronal activity during sexual behavior of the male monkey. Brain Research, 266, 340-343.

Paredes, R. G. (2003). Medial preoptic area/anterior hypothalamus and sexual motivation. Scandinavian Journal of Psychology, 44, 203-212.

Paredes, R. G., Tzschentke, T., & Nakach, N. (1998). Lesions of the medial preoptic area/anterior hypothalamus (MPOA/AH) modify partner preference in male rats. Brain Research, 30, 1-8.

Pause, B. M., Krauel, K., Schrader, C., Sojka, B., Westphal, E., Müller-Ruchholtz, W., & Ferstl, R. (2006). The human brain is a detector of chemosensorily transmitted HLA-class I-similarity in same- and opposite-sex relations. Proceedings of the Royal Society of London. Series B, 273, 471-478.

Perkins, A., Fitzgerald, J. A., & Moss, G. E. (1995). A comparison of LH secretion and brain estradiol receptors in heterosexual and homosexual rams and female sheep. Hormones and Behavior, 29, 31-41.

Perkins, A., Fitzgerald, J. A., & Price, E. O. (1992). Luteinizing hormone and testosterone response of sexually active and inactive rams. Journal of Animal Science, 70, 2086-2093.

Preti, G., & Huggins, G. R. (1975). Cyclical changes in volatile acidic metabolites of human vaginal secretions and their relation to ovulation. Journal of Chemical Ecology, 1, 316-326.

Preti, G., Wysocki, C. J., Barnhart, K., Sonheimer, S. J., & Leyden, J. J. (2001, April). Male axillary extracts effect lutenizing hormone (LH) pulsing in female recipients. 23rd Annual Meeting of the Association for Chemoreception Sciences, Sarasota, FL.

Preti, G., Wysocki, C. J., Barnhart, K. T., Sondheimer, S. J., & Leyden, J. J. (2003). Male axillary extracts contain pheromones that affect pulsatile secretion of luteinizing hormone and mood in women recipients. Biology of Reproduction, 68, 2107-2113.

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. Human Brain Mapping, 11, 162-77.

Rhodes, G. (2006). The evolutionary psychology of facial beauty. Annual Review of Psychology, 57, 199-226.

Roney, J. R., Mahler, S. V., & Maestripieri, D. (2003). Behavioral and hormonal responses of men to brief interactions with women. Evolution and Human Behavior, 24, 365-375.

Roselli, C. E., Larkin, K., Resko, J. A., Stellflug, J. N., & Stormshak, F. (2004). The volume of a sexually dimorphic nucleus in the ovine medial preoptic area/anterior hypothalamus varies with sexual partner preference. Endocrinology, 145, 478-483.

Samama, B., & Aron, C. (1989). Changes in estrogen receptors in the mediobasal hypothalamus mediate the facilitory effects exerted by the male's olfactory cures and progesterone on feminine behavior in the male rat. Journal of Steroid Biochemistry, 32, 525-529.

Savic, I., Berglund, H., Gulyas, B., & Roland, P. (2001). Smelling of odorous sex hormone-like compounds causes sex-differentiated hypothalamic activations in humans. Neuron, 31, 661-668.

Savic, I., Berglund, H., & Lindstrom, P. (2005). Brain response to putative pheromones in homosexual men. Proceedings of the National Academy of Sciences, 102, 7356-7361.

Segovia, S., & Guillamon, A. (1993). Sexual dimorphism in the vomeronasal pathway and sex differences in reproductive behaviors. Brain research. Brain Research Reviews, 18, 51-74.

Shepherd, G. M. (2006). Behaviour: Smells, brains, and hormones. Nature, 439, 149-151.

Sherwood, N. M., Lovejoy, D. A., & Coe, I. R. (1993). Origin of mammalian gonadotropin-releasing hormones. Endocrine Reviews, 14, 241-54.

Shinohara, K., Morofushi, M., Funabashi, T., Mitsushima, D., & Kimura, F. (2000). Effects of 5alpha-androst-16-en-3alpha-ol on the pulsatile secretion of luteinizing hormone in human females. Chemical Senses, 25, 465-467.

Shinohara, K., Morofushi, M., Funabashi, T., & Kimura, F. (2001). Axillary pheromones modulate pulsatile LH secretion in humans. Neuroreport, 12, 893-895.

Silverman, A. J., Roberts, J. L., Dong, K. W., Miller, G. M., & Gibson, M. J. (1992). Intrahypothalamic injection of a cell line secreting gonadotropin -releasing hormone results in cellular differentiation and reversal of hypogonadism in mutant mice. Proceedings of the National Academy of Sciences, 89, 10668-10672.

Simon, W. (1994). Deviance as history: The future of perversion. Archives of Sexual Behavior, 23, 1-20.

Singh, D. (1993). Adaptive significance of female physical attractiveness: role of waist-to-hip ratio. Journal of Personality and Social Psychology, 65, 293-307.

Singh, D., & Bronstad, P. M. (2001). Female body odour is a potential cue to ovulation. Proceedings of the Royal Society of London. Series B, 268, 797-801.

Sobel, N., Prabhakaran, V., Hartley, C. A., Desmond, J. E., Glover, G. H., Sullivan, E. V., & Gabrieli, J. D. (1999). Blind smell: Brain activation induced by an undetected air-borne chemical. Brain, 122(2), 209-217.

Spehr, M., Kelliher, K. R., Li, X. H., Boehm, T., Leinders-Zufall, T., & Zufall, F. (2006). Essential role of the main olfactory system in social recognition of major histocompatibility complex peptide ligands. The Journal of Neuroscience, 26, 1961-1970.

Stern, K., & McClintock, M.K. (1998). Regulation of ovulation by human pheromones. Nature, 392, 177-179.

Stoleru, S. G., Ennaji, A., Cournot, A., & Spira, A. (1993). LH pulsatile secretion and testosterone blood levels are influenced by sexual arousal in human males. Psychoneuroendocrinology, 18, 205-18.

Tobet, S. A., & Fox, T. O. (1992). Sex differences in neuronal morphology influenced hormonally throughout life. In A. A. Gerall, H. Moltz, and I. L. Ward (Eds.), Handbook of behavioral neurobiology, volume 11, (pp. 41-84). New York: Plenum Press.

Twan, W. H., Hwang, J. S., Lee, Y. H., Jeng, S. R., Yueh, W. S., Tung, Y. H., Wu, H. F., Dufour, S., & Chang, C. F. (2006). The presence and ancestral role of gonadotropin-releasing hormone in the reproduction of scleractinian coral, Euphyllia ancora. Endocrinology, 47, 397-406.

Vogt, R. G. (2006). How sensitive is a nose? Science's STKE: Signal Transduction Knowledge Environment, 322, Retrieved February 16, 2006 from <stke.sciencemag.org/cgi/content/abstract/sigtrans;2006/322/pe8>.

Wallen, K. (2005). Hormonal influences on sexually differentiated behavior in nonhuman primates. Frontiers in Neuroendocrinology, 26, 7-26.

Wedekind, C., Seebeck, T., Bettens, F., & Paepke, A. J. (1995). MHC-dependent mate preferences in humans. Proceedings of the Royal Society of London. Series B, 260, 245–249.

Wedekind, C., & Furi, S. (1997). Body odour preferences in men and women: Do they aim for specific MHC combinations or simply heterozygosity? Proceedings of the Royal Society of London. Series B, 264, 1471–1479.

Weissman, J. D., & Singer, D. S. (1991). Striking similarities between the regulatory mechanisms governing yeast mating-type genes and mammalian major histocompatibility complex genes. Molecular and Cellular Biology, 11, 4228-4234.

Woodson, J. C. (2002). Including “learned sexuality” in the organization of sexual behavior. Neuroscience and Biobehavioral Reviews, 26, 69-80.

Wysocki, C. J., Yamazaki, K., Curran, M., Wysocki, L. M., & Beauchamp, G. K. (2004). Mice (Mus musculus) lacking a vomeronasal organ can discriminate MHC-determined odortypes. Hormones and Behavior, 46, 241-6

Wyatt, T. D. (2003). Pheromones and animal behaviour: Communication by smell and taste. Cambridge, UK: Cambridge University Press.

Zald, D. H., & Pardo, J. V. (1997). Emotion, olfaction, and the human amygdala: Amygdala activation during aversive olfactory stimulation. Proceedings of the National Academy of Sciences, 94, 4119-4124.