Olfactory communication and mate choice

From the regulation and reproduction in bacteria colonies (Bassler, 2002) to complex smell and taste systems of humans (Van Toller & Dodd, 1988), the ability of sensing chemical stimuli, known as chemosensation, is believed to be the most basic and ubiquitous of senses (Bhutta, 2007). One strain of thought places chemosensation as merely an evolved ability to detect dangerous and volatile substances – such as putrefied food (see Bhutta, 2007). Still, the notion that this ability to detect chemical stimuli, particularly in the domain of smell, serves a purpose in communication is not necessarily a contemporary concept (Wyatt, 2009).

The ancient Greeks, for instance, observed that the attraction in male dogs was influenced by secretions of female dogs in heat (ibid). Meanwhile Charles Darwin (1871), in reference to crustaceans, noted “[…]the increased number of the smelling-threads has probably been acquired through sexual selection, by the better provided males having been the more successful in finding partners and in producing offspring.” (chapter 9, online). More recently, an increasing amount of evidence suggests olfactory systems in animals are a key sensory component in mate selection; playing a fundamental role in evolution (Bhutta, 2007).

A widely studied and much discussed aspect of chemosensation in animals is the idea of pheromones (ibid). Derived from the Greek phererin (to transfer) and hormōn (to excite), pheromone was originally coined by Peter Karlson and Martin Lüscher (Wyatt, 2009) to describe “[…] substances which are secreted to the outside by an individual of the same species, in which they release a specific reaction, for example, a definite behaviour or a developmental process.” (pg. 262).

Using pheromones as a communication tool is widely observed across many animal species, particularly those showing a high degree of sociality (Bhutta, 2007). For instance, laboratory mice show many instances of pheromonal responses (ibid), including: changes in the menstrual cycles of female mice in response to the presence (Whitten, 1959) or absence (Van der lee & Boot, 1955) of males; termination of a pregnancy if a female is housed with a fertile male who is not the father (Bruce, 1960); and, female mice undergo puberty at an earlier age when housed with a male (Vanderbergh et al., 1975).

Even though we know humans produce pheromones, what about the detection of these chemical secretions?

On the basis of the animal studies, the vomeronasal organ (VNO) is the primary place of pheromone detection (Bhutto, 2007). However, some studies (Doty, 2001; Rodriguez & Mombaerts, 2002; Liman & Innan, 2003; Savic et al., 2009) argue the VNO, and associated genes, are essentially non-functional, whilst another group of studies (Monti-Bloch et al., 1998; Smith et al., 1998; Grosser et al., 2000; Grammer et al., 2005) claim the opposite; that VNO responds to pheromones in a sex-specific manner (Grammer et al., 2005). More recently though, the discovery of a pheromone receptor gene (VN1R1), transcripts of which are found in human olfactory mucosa (Bhutta, 2007), strengthens the assertion that signals from some pheromones (specifically EST) are mediated by the main olfactory bulb (MOB) (Savic et al., 2009).

Further evidence for the MOB subserving pheromone detection is from investigations into patients with Kallman’s syndrome (Dodé & Hardelin, 2009). Patients suffering from this disorder have an underdeveloped olfactory bulb and tracts alongside hypogonadism (gonad defects) and anosmia (lack of smell) (ibid). Hardelin & Dodé (2008) investigate the underlying genetic network of Kallman’s syndrome, finding the hypogonadism is due to gonadotropin-releasing hormone (GnRH) deficiency, which they argue “[…] results from a failure of the embryonic migration of neuroendocrine GnRH cells from the olfactory epithelium to the forebrain.” (pg. 181). In addition to these findings, Grammer et al. reported that “[…] preliminary research indicates that [patients with Kallman’s syndrome] show no response to pheromones.” (2005, pg. 136).

Some understanding of the underlying neurological processing in olfactory social perception is beginning to emerge. In a PET study by Lundström et al. (2008), female subjects were exposed to both body and non-body odours in an attempt to identify their respective neural activations. The primary processing differences between the two stimuli were that “[…] body odours activated a circuit including more non-olfactory than olfactory brain regions… [also] The familiarity of body odours (i.e. personal odour, odour of a friend, odour of a stranger) induced further topographical differentiation” (Brancucci et al., 2009). These findings, alongside many other studies (see Brancucci et al., 2009 for a review) suggest the overarching emphasis of pheromone processing appears to be its asymmetrical pattern of lateralisation, which is strongly dependent on sex: males demonstrate a strong right hemispheric lateralisation, whilst in females it is less pronounced (ibid).

Even though the contention surrounding our ability to detect and produce pheromones has subsided somewhat, the evolutionary benefit of such a system needs to be addressed. Of course, one suggestion is that this ability is a superfluous relic of our mammalian evolutionary heritage (Bhutto, 2007), with humans mostly being microsmatic (poor smellers) who rely more on other sensory modalities (verbal and visual, for instance) in mate selection and sexual behaviour (Grammer et al., 2005). Countering these claims are a growing number of experiments into the role of olfaction in contemporary human societies.

Much of these experiments stem from Russell (1976) and Wallace’s (1977) early discoveries that individuals are able to determine the sex of another person just through their odour (Bhutto, 2007). A common method of gathering odours from the axilla was to collect t-shirts worn for several days by the subjects providing samples (ibid). Another study by Preti et al. (2005) provide data that male axillary extracts causes “[…] neuroendocrine and mood alterations in women.” (pg. 2110). Furthermore, these extracts advance the release of a luteinizing hormone, which causes the women subjects to ovulate (ibid). This discovery perhaps explains the findings of Cutler et al. (1986), where axillary extracts encourage more regular ovulatory cycles, and Ellis & Garber’s (2000) claim that girls in homes with step-fathers experienced a faster onset of puberty than girls in single-mother homes.

MHC and Mate Selection

In mate selection, one possible purpose for olfactory communication, particularly in contemporary human societies, is to signal good genes (Grammer et al., 2005). In particular, signalling immunocompetence (Folstad & Karter, 1992) is a way in which sexual selection allows females to obtain complementary genes from males to pass on to their offspring – conferring them with an immune advantage (Grammer et al., 2005). In vertebrates, the major histocompatibility complex (MHC) is a set of protein products that play a central role in immune processes, with MHC-associated genes demonstrating “[…] extremely high intrapopulational polymorphism… and dozens of evolutionary theorists, population geneticists, immunologists and behavioral ecologists have attempted to elucidate the evolutionary pressures that have shaped it.” (Havlicek & Roberts, 2008).

Although there are many hypotheses attempting to explain the degree of MHC polymorphism (see Havlicek & Roberts, 2008), one hypothesis stresses the role of sexual selection in shaping disassortative mating preferences (ibid). That is, individuals prefer their partners to possess dissimilar MHC genotypes (ibid). The importance of a dissimilar genotype is based on the assertion of some studies (Penn et al., 2002; Piertney & Oliver, 2006) that MHC-heterozygosity is advantageous in the development of the immune system (Havlicek & Roberts, 2008).

MHC-correlated mate choice is found in many species, from sticklebacks (Reusch et al., 2001) to lemurs (Schwensow et al., 2008). However, the initial discoveries were made in the studies of mice by Yamazaki and colleagues (1976), which made the claim that MHC-associated mate choice was made on the basis of odour cues. Since these initial studies, it appears mice do not choose their mates (and avoid inbreeding) on the basis of the MHC-type, rather mice use MUP (major urinary proteins) as indicators of mate selection (Sherborne et al., 2007). Nonetheless, that MUPs are non-polymorphic in humans, whereas they are highly polymorphic in mice (ibid), is perhaps one reason not to negate the studies discussed below.

In humans, MHC is commonly termed HLA (human leuckocyte antigen), and like the various studies of animals, there appears to be evidence for MHC-correlated odour preferences (Havlicek & Roberts, 2008). Wedekind et al. (1995) provides the basis for many investigations into MHC mate preference in humans (also see: Jacob et al., 2002; Thornhill et al., 2003; Roberts et al., 2008). Essentially, women are asked to smell and then rate T-shirts worn by men for a number of nights. In Wedekind et al.’s initial study, women were divided into two groups: those using hormonal contraceptives (n=18) and those not using hormonal contraceptives (n=31). They found that “[…] On average, women not using the contraceptive pill rated as more pleasant the odors of men with MHC antigens that were dissimilar to their own.” (Havlicek & Roberts, 2008, pg. 3).

The results of Wedekind et al.’s study suggest women may have a preference for the body odour of men with MHC-dissimilar antigens, although subsequent results in repeated experiments (e.g. Roberts et al., 2008) failed to find a significant correlation, it does not necessarily mean this ability is applied when choosing a potential mate. For instance, recent studies found facial preferences correlates with both MHC-similarity (Roberts et al., 2005) and MHC heterozygosity (Lie et al., 2008). That facial preferences demonstrate opposite findings to the generally disassortative odour preferences is one instance where mate choice may rely on one modality over another in mate choice (ibid). Alternatively, neither modality may play a significant role or, as suggested by Havlicek & Roberts (2008), these two preferences, “[…] on average, might be a way of achieving an intermediate rather than maximal level of genetic variability in a preferred mate, since they could work in tandem to screen out the extreme.” (pg. 10).

As already discussed in the MHC-studies, olfaction is likely to interact with other modalities in influencing sexual behaviours and mate selection. Although assessing the degree to which olfaction interacts with or is negated by other sensory modalities is considerably difficult.

First, measuring the influence of chemical stimuli is far more tentative, unlike vision (wavelength) and audition (frequency), as there is no equivalent metric for olfaction; except for a recently generated multidimensional odour metric (Haddad et al., 2008). Second, we are still very much at the beginning in our understanding of the interaction between semiochemical (substances eliciting a conscious perception) and pheromones (substances eliciting unconscious perception) (Sbarbati & Osculati, 2006). Furthermore, there are vasana chemical signals, which “[…] are neither classifiable as odours nor as pheromones, that are not consciously detectable as odours, and do affect psychological state yet not triggering a unique set of behavioural, neural or endocrine responses.” (Brancucci et al., 2009, pg. 904).

Despite these inherent difficulties, there is neurological evidence suggesting some cross-modal integration of information processed by chemosensory, visual and auditory systems (ibid). For instance, olfactory inputs leading from the MOS/VNO are not only processed by the ipsilateral primary olfactory cortex, but subsequently at the amygdala, which plays “[…] a major role in the learning and recognition of social chemosignals as well as being a hub for visual and acoustic emotion-related information.” (ibid, pg. 904).

These neural correlates in cross-modal processing may explain some results, especially where the processing of male pheromones during the ovulation induces a stronger preference in women for symmetrical male faces (Thornhill & Gangestad, 1999). A more recent study by Havlicek and colleagues (2008) also shows how sensory reliance differs on the basis of sex: women tend to rely more on olfactory cues, and men on visual cues, in both partner choice and sexual arousal. In an interesting aside, women generally (i.e. non-sexual contexts) tended to value olfactory cues significantly more than did men (ibid).

Not only do women place more emphasis on the olfaction modality, there is further evidence to suggest it affects their behaviour and influences mate choice. One study (Wyart et al., 2007) observed that after smelling pure androstradienone, a molecule present in the sweat of men, women produced significantly higher levels of cortisol; and suggests “[…] that, like rodents, humans can influence the hormonal balance of conspecifics through chemosignals.” (pg. 1261).

Another consideration in determining the role of body odour in mate choice is to investigate the differences in olfaction between homo- and hetero-sexual individuals (Sergeant et al., 2006). One study by Sergeant et al. (2006) found heterosexual women preferred homosexual males’ scent than that of heterosexuals. Also, the brain activity and functional connectivity of homosexual individuals when exposed to pheromone-like substances shows almost a reversal of results found in heterosexuals: heterosexual men and lesbian women show larger right hemispheres, whereas homosexual men and heterosexual women show a more symmetrical hemispheric distribution (Brancucci et al., 2009).

Another consideration raised by the Havlicek et al. (2008) paper, and in the earlier section on MHC, is the role of social and cultural practices in shaping mate choice. Generally, it is reasonable to suggest that different cultures may place a greater emphasis on one aspect involved in mate choice than, say, another culture. For instance, is appears Czech high school students rate body odours (or at least the samples used) more positively than did US university students (ibid). Conversely, US university students tend to be more visually oriented in mate-choice contexts (ibid).

Modern cultural practices impinge on the relative influence of olfaction in both direct and indirect ways. Contemporary western societies use perfumes and deodorants are generally seen as methods of disguising body odour (Bhutto, 2007), which is arguably a conscious (direct) attempt to influence olfaction. However, as mentioned earlier, a study by Milinski & Wedekind (2001) found a significant correlation between HLA and scent scoring of perfumes/aftershaves for themselves (but importantly, not for others); suggesting people select perfumes on the basis of their pleasantness (conscious) and to enhance their own body odours (unconscious).

Contraceptives are another example of a cultural practice that may indirectly influence olfaction and mate choice. When compared to women with normal menstrual cycles, women using hormonal contraceptives are on average more likely to prefer the odours of MHC-similar men (Wedekind and Füri, 1997). Additional studies investigating the impact of contraceptives on mate preferences and sexual behaviour (Jones et al., 2005) support the assertions of the MHC-related findings, suggesting “[…] that pill use could change behavior towards that normally characteristic of pregnancy” (Havlicek & Roberts, 2008, pg. 12).

Differences between social structures and cultures at the population level may also account for various findings across studies into mate choice. MHC studies, for example, are often performed in western population – and in these instances, spouses tend to support the findings from odour studies, in that their respective MHCs are generally dissimilar (Raphaëlle Chaix et al., 2008; but see Graver-Apgar et al., 2006). In contrast, a study into African (Yoruba) populations show no correlation between MHC and spouses (Raphaëlle Chaix et al., 2008). Actually, the significant find Raphaëlle Chaix et al reported for Africans was that spouses tended to show genome-wide similarity. The authors suggest this observation is because Yoruban spouses are more constrained by their population structure (e.g. social hierarchies) than are Europeans (ibid).

In examining the biology and evolution of olfaction, we know olfactory communication in animals is capable of advertising an individual’s identity and their potential mate quality – all of which is honest in its presentation (Johansson & Jones, 2007). As for humans, our ability to detect and produce pheromones is gathering an ever-increasing amount of evidence (see, Grammer, 2005), but whether these abilities translate into having a practical effect on sexual behaviour and mate choice is for future studies to elucidate upon.

Despite the short-comings of our current knowledge, there are hints as to where the field is going. For instance, one challenge will be to discern the role of MHC in mate choice, which is surely on the horizon given the increasing capabilities of genetic sequencing technologies (Havlicek & Roberts, 2008). A related strand of investigation involves the potential for molecular, genetic, behavioural biology and psychological disciplines to offer a multi-discipline approach in linking together the various layers involved in olfactory communication.

If there is one clear conclusion to take away from this review, then it is this: the notion of humans being primarily optical creatures is no longer tenable.

Main References
Bhutta, M. (2007). Sex and the nose: human pheromonal responses Journal of the Royal Society of Medicine, 100 (6), 268-274 DOI: 10.1258/jrsm.100.6.268

Havlicek, J., & Roberts, S. (2009). MHC-correlated mate choice in humans: A review Psychoneuroendocrinology, 34 (4), 497-512 DOI: 10.1016/j.psyneuen.2008.10.007