7.3 Smell

Smell and taste are called chemical senses, because the stimuli for them are chemical molecules. The chemical senses are first and foremost systems for warning and attracting. They play on our drives and emotions more than on our intellects. Think of the effects produced by a valentine gift of chocolates, perfume, or fresh roses; by the aroma and taste of your favorite meal; or by the stench of feces or rotting meat.

Although the human sense of smell, or olfaction, is much less sensitive than that of many other animals (for example, humans have about 500 genes involved with olfaction, whereas mice have 1,300; Scott, 2012), it is still remarkably sensitive and useful. We can smell smoke at concentrations well below that needed to trigger even the most sensitive of household smoke detectors. We can distinguish among roughly 10,000 different chemicals by smell (Scott, 2012). Blind people regularly identify individuals by their unique odors, and sighted people can do that too when they try. And smell contributes greatly to what we call “flavor” in foods.

Anatomy and Physiology of Smell

Great progress has been made within the past decade or so in understanding the sense of smell.

Transduction and Coding for the Sense of Smell

The basic layout of the olfactory (smell) system is illustrated in Figure 7.4. The stimuli for smell are molecules that evaporate into the air, are taken with air into the nasal cavity, and then become dissolved in the mucous fluid covering the olfactory epithelium, the sensory tissue for smell, which lines the top of the nasal cavity. The olfactory epithelium contains the sensitive terminals of roughly 6 million olfactory sensory neurons (Doty, 2001). Each terminal contains many olfactory receptor sites, which are large protein molecules woven into the cell membrane that are capable of binding molecules of specific odorants (odorous substances). The binding of a molecule to a receptor site changes the structure of the cell membrane, which results in an electrical change that tends to trigger action potentials in the neuron’s axon. The greater the number of binding sites activated by odorous molecules, the greater the rate of action potentials triggered in the axon.

Figure 7.4: The anatomy of smell Molecules of odorants enter the nose through the nostrils, become dissolved in the mucous fluid covering the olfactory epithelium, and bind to receptor sites on the sensitive tips of olfactory sensory neurons, where they initiate action potentials. The sensory neurons send their axons through the cribriform plate (a small bone shelf) to form synapses on second-order olfactory neurons in the glomeruli of the olfactory bulb, directly above the nasal cavity. As illustrated in the right-hand diagram by the use of color, each glomerulus receives input from only one type of olfactory sensory neuron (defined by its type of receptor sites). Only two types of such neurons are shown here (depicted as yellow and blue), of the roughly 350 types that exist in the human olfactory system.

The olfactory nerve contains roughly 350 different types of sensory neurons, each of which is characterized by a distinctly shaped binding site on its terminals within the olfactory epithelium (Wilson & Mainen, 2006). Any given type of binding site can bind more than one odorant, but any given odorant binds more readily to some types than to others. Thus, each type of olfactory neuron differs from the other types in its degree of sensitivity to particular odorants.

The axons of the olfactory sensory neurons pass through a thin, porous bone into the olfactory bulb of the brain, where they form synapses upon other neurons in structures called glomeruli [gló-mer´-u-le] (singular glomerulus; depicted at right in Figure 7.4). The pattern of these connections is remarkably orderly. Each glomerulus in the olfactory bulb receives input from several thousand olfactory sensory neurons, but all these neurons are of the same type. For each of the 350 different types of olfactory sensory neurons, there is a different receiving glomerulus (or a set of two or three such glomeruli) in the olfactory bulb (Wilson & Mainen, 2006).

From this work, researchers have inferred the process by which qualitative and quantitative coding occurs for the sense of smell. Each odorant that we can distinguish is apparently characterized by its ability to produce a unique pattern of activity across the 350 different types of olfactory neurons and their corresponding glomeruli in the olfactory bulb. Thus, odorant A might trigger much activity in one set of glomeruli, a moderate amount in another set, and very little in others. The greater the amount of odorant A, the greater would be the total amount of activity triggered in each of the glomeruli that it affects, but the ratio of activity across glomeruli would remain relatively constant. Thus, the ratio indicates the type of odorant (quality of the smell), while the total amount of activity indicates the amount of odorant (quantity, or intensity of the smell).

Olfactory Brain Areas Beyond the Olfactory Bulb

The glomeruli in the olfactory bulb send output to various other parts of the brain. Most of this output goes to structures in the limbic system and the hypothalamus, which (as discussed in Chapters 5 and 6) are involved in basic drives and emotions. These connections, presumably, help to account for the strong and often unconscious effects that smell can have on our motivational and emotional states. The connections from the olfactory bulb to the limbic system are so strong, in fact, that the limbic system was at one time referred to as the rhinencephalon, which literally means “nose brain.” Output from the olfactory bulbs also goes to various portions of the cerebral cortex. The primary olfactory cortex is located in the underside of the temporal lobe (shown in Figure 7.1, on page 247) and wraps down underneath the rest of the brain. This area in turn sends output to a secondary olfactory area in the orbitofrontal cortex, located on the underside of the frontal lobe (Rolls, 2004). These cortical areas are crucial for the ability to experience odors consciously and identify the differences among them (Buck, 2000; Rolls, 2004).

Smell as a Component of Flavor: The Mouth-to-Nose Connection

Odorants can reach the olfactory epithelium through two different routes. The route that everyone recognizes is through the nostrils; this allows us to smell smoke, roses, skunks, and other odor sources that are outside the mouth. The other route allows us to smell substances that have entered the mouth. An opening (the nasal pharynx, at left in Figure 7.4, on page 254) connects the back of the mouth cavity with the nasal cavity. The acts of chewing and swallowing push air from the mouth up into the nose—air that carries volatile molecules of whatever you are eating. What most people call taste—and what is properly called flavor— consists not just of true taste (from taste receptors in the mouth) but also of smell that has been triggered through this mouth-to-nose, back-door route. Remarkably, you experience this sensation as coming from the mouth, where the food exists, and as indistinguishable from taste, even though it actually comes from the olfactory epithelium (Shepherd, 2006).

If air can’t flow out through the nostrils, it can’t stream into the nasal cavity from the mouth. Experiments have shown repeatedly that people’s abilities to identify foods and drinks by flavor decline markedly when their nostrils are shut. You can easily demonstrate this by tasting jelly beans one at a time with your eyes closed (suggested by Schiffman, 1996). You will probably be able to distinguish among flavors such as cherry, grape, orange, and licorice quite easily, until you try it with your nostrils pinched shut. Under that condition, you will most likely find that now all the jelly beans taste the same; all you taste is the sugar. The differences that allow you to distinguish flavors depend on smell. Smell and taste inputs converge in a certain portion of the orbitofrontal cortex, and this area appears to be critical for the psychological experience of flavor (Rolls, 2004).

An excellent bouquet This professional wine taster samples the wine’s scent through both his nose and his mouth. Through the mouth, odorant molecules reach the nasal cavity by way of a connection called the nasal pharynx. Much of what we think of as taste is actually smell.

Jupiter Images

Differences Among People in Olfactory Sensitivity

Big differences exist among individuals in general olfactory sensitivity. Women are, on average, more sensitive to odors than are men (Doty, 2001), and many women become especially sensitive to odors during pregnancy (Nordin et al., 2004). In both sexes, sensitivity to odors declines with age beginning around age 30 and more noticeably around age 65 or 70. By age 65, roughly 11 percent of women and 25 percent of men have serious olfactory impairment, and by age 85 those numbers are 60 percent and 70 percent (Murphy et al., 2002). Many elderly people complain of loss in ability to taste foods, but tests typically show that their real loss is not taste but smell (Bartoshuk & Beauchamp, 1994). The most dangerous effect of such impairment is the inability to smell smoke or toxic gases in the air. A high proportion of people who die from asphyxiation are elderly people who have lost much or all of their olfactory ability.

Many people with otherwise normal olfactory ability are completely unable to smell particular chemicals that other people can smell easily. In fact, at least 75 different chemicals have been identified that most people can smell but some cannot (Pierce et al., 2004). These differences are at least partly the result of genetic differences that affect the production of specific olfactory receptors on olfactory neurons. The most fully studied example concerns ability to smell the chemical substance androstenone, which is a derivative of testosterone. This chemical is found in human sweat, more so in males than in females, and big differences exist among individuals of both sexes in their ability to smell it. Researchers have found, in different people, three different variants of the gene that codes for the receptor protein for androstenone. People with the most common variant of the gene find androstenone’s odor to be strong and putrid; those with the second most common variant find it to be relatively weak and pleasant (“sweet and fruity”); and those with the least common variant cannot smell the chemical at all (Keller et al., 2007).

Sensitivity to specific odors is also very much affected by experience. With repeated tests, people can learn to distinguish the odors of slightly different chemicals, which initially smelled identical to them (Li et al., 2008), and can learn to detect specific odors at a much lower concentration than they could initially (Boulkroune et al., 2007). In some cases, such effects of experience have been found for both sexes, but in others it has been found to occur only in women. This was shown in a study described earlier in the chapter (refer back to Figure 7.3, p. 250). In further experiments, Dalton and her colleagues found the increased sensitivity only for women who were in their reproductive years; it did not occur for prepubescent girls or postmenopausal women. Such findings are consistent with theories that olfaction serves one or more special functions related to reproduction in women, such as choosing mates, avoiding toxins during pregnancy, or bonding with infants.

Discriminating Among Individuals by Smell

As dog owners well know, dogs greet and recognize others of their kind (and sometimes also of our kind) by smell. We humans living in a somewhat odor-phobic culture may not often admit it or even be aware of it, but we too can identify individuals of our species by smell. In a typical experiment, one set of subjects wears initially clean T-shirts for a day without washing or using deodorants or perfumes. Then, another set of subjects is asked to identify by smell alone which shirt was worn by whom. Such experiments have revealed that parents can tell which of their children wore the shirt, children can tell which of their siblings wore it, and people generally can distinguish between the odors of two strangers (Weisfeld et al., 2003).

Role of Smell in Mother-Infant Bonding

A sweet aroma Every human being has a unique, identifiable odor.

Blend Images/Shutterstock

Among some mammals, notably goats and sheep, odor recognition is a crucial part of the bond between mother and young (Kendrick et al., 1992). Might smell also play a role in human mother–infant bonding?

In one study conducted in a hospital maternity ward, 90 percent of mothers who had been exposed to their newborn babies for just 10 to 60 minutes after birth were able to identify by smell alone which of several undershirts had been worn by their own babies (Kaitz et al., 1987). In another study, breast-fed babies as young as 6 days old turned their heads reliably toward cotton pads that their own mothers had worn against their breasts, when given a choice between that and identical-looking pads that had been worn by other lactating women (Macfarlane, 1975). In still another study, babies who were exposed to a particular unusual odor (not that of their mothers) within the first hour after birth turned reliably toward that odor when given a choice between it and another unusual odor 6 days later (Varendi et al., 2002). All such evidence suggests that odor figures into the complex of stimuli that are involved in the attachment between human infants and their mothers. Unlike the case for goats and sheep, however, odor is certainly not essential for such attachment.

Possible Role of Smell in Choosing a Genetically Compatible Mate

In mice, odor has been shown to play a role in mating choices. Mice, like dogs and humans, can identify other individuals of their species by smell, and, remarkably, they prefer to mate with opposite-sex mice whose odor is most different from their own (Potts et al., 1991; Yamazaki et al., 1988). Why this preference? Researchers have found that the individual differences in odor that determine these mating preferences result from a set of about 50 highly variable genes (genes with many different alleles) referred to collectively as the major histocompatibility complex (MHC) (Yamazaki et al., 1994). These same genes also determine the precise nature of the cells used by the immune system to reject foreign substances and kill disease-producing bacteria and viruses. Thus, by choosing mates that smell most different from themselves, mice choose mates that (a) are not likely to be close relatives of themselves, and (b) will add much new genetic variation to the mix of disease-fighting cells that develop in the offspring.

The MHC also exists in human beings and contributes greatly to individual differences in odor (Brennan & Zufall, 2006). The advantages of mating with someone who has a very different MHC presumably exist in humans as much as in mice. Do humans, to any degree at all, prefer sexual mates who smell most different from themselves and differ most in MHC? At present, the answer to that is not known, but some research suggests that it might be yes (Penn, 2002). In one series of experiments, Claus Wedekind and his colleagues (1995, 1997) asked young men and women to rate the “pleasantness” (and in one study the “sexiness”) of the odors of T-shirts that had been worn by young adults of the opposite sex. All the subjects were assessed biochemically for differences in their MHCs. The result was that any given donor’s odor was, on average, rated as more pleasant by raters who differed from that person in MHC than by raters who were similar to that person in MHC.

There is some evidence that MHC differences can affect actual sexual behavior in humans. In one study, romantically involved young heterosexual couples were tested for the degree of MHC difference between them and were asked to respond to a confidential questionnaire about their sexual desires and activities. The significant findings pertained to the women’s sexual desires for their partners. On average, the more different her partner’s MHC was from hers, the greater was the woman’s interest in having sex with him and the less likely she was to have sex with someone outside of the relationship (Garver-Apgar et al., 2006). In this study, the men’s sexual desires were unrelated to the MHC differences. It is possible that this sex difference exists because, for women, the biological costs of producing children are much greater than are those for men, so selectivity for compatible genes is more important for women than for men. Other studies on effects of MHC differences on choices of sexual partners have produced mixed results, with some showing effects and others not (Penn, 2002).

The sense of smell may also play a role in incest avoidance. Most of us cringe at the thought of having sex with a family member, and, of course, society strongly reinforces that sense of disgust. But one factor that may make such aversion “natural” is the sense of smell. As we noted previously, people can identify their genetic relatives based on smell, and they find the odor of some kin more aversive than others. This was shown in a series of experiments by Glen Weisfeld and his colleagues (2003), who asked people to try to identify by smell T-shirts worn by different family members. Not only could people generally recognize family members on the basis of smell, but the odors of some genetic relatives were particularly negative. Brothers and sisters showed mutual aversion to the odor of their opposite-sex siblings, fathers were aversive to the odor of their daughters (but not their sons), and daughters to the odor of their fathers (but not their mothers). These are the pairings at most risk for incest. Moreover, these patterns were found whether or not the person smelling the T-shirt could accurately recognize who had worn the shirt.

Smell as a Mode of Communication: Do Humans Produce Pheromones?

Apheromone [fer´-e-mōn] is a chemical substance that is released by an animal and acts on other members of its species to promote some specific behavioral or physiological response. The most dramatic examples occur in insects. For instance, sexually receptive female cabbage moths secrete a pheromone that attracts male cabbage moths from as far as several miles away (Lerner et al., 1990). Most species of mammals also produce pheromones, which serve such functions as sexual attraction, territorial marking, and regulation of hormone production (Hughes, 1999; Wyatt, 2009). Most species of mammals have in their nasal cavities a structure called the vomeronasal [vo′-me-rō-nā´-zel] organ, which contains receptor cells specialized for responding to pheromones. Whereas the main olfactory epithelium is designed to distinguish somewhat imprecisely among many thousands of different odorants, the vomeronasal organ appears to be designed for very precise recognition of, and exquisite sensitivity to, a small number of specific substances—the species’ pheromones (Buck, 2000).

Do humans communicate by pheromones? We do have the structures that would make such communication possible. Like other mammals, we have specialized glands in the skin that secrete odorous substances. Such glands are especially concentrated in areas of the body where our species has retained hair—such as in the axillary region (armpits) and genital region (see Figure 7.5). One theory is that the function of hair in these locations is to retain the secretions and provide a large surface area from which they can evaporate, so as to increase their effectiveness as odorants (Stoddart, 1990). Some substances secreted by these glands, such as androstenone, are steroid molecules that resemble substances known to serve as pheromones in other mammals. We also have at least a rudimentary vomeronasal organ, but the evidence to date is inconclusive as to whether it functions in our species or is entirely vestigial (Brennan & Zufall, 2006).

Figure 7.5: Locations of maximal scent production by humans In humans, specialized scent-producing glands (apocrine glands) are concentrated most highly in the axillary region (underarms) and also exist in high concentrations in the genital area, the area around the nipples, the navel area, on the top of the head, and on the forehead and cheeks (Stoddart, 1990), as shown by the added circles.
(The statue here is Michelangelo’s Aurora, from the tomb of Lorenzo de Medici, in Florence.)

Scala/Art Resource, NY

Motivated partly by the perfume and cologne industry, most human pheromone research has centered on whether or not we produce sex-attractant pheromones. In many experiments, men and women have been exposed to various secretions taken from the other sex and have rated the attractiveness of the odor or changes in their own mood. To date, despite the often lurid claims in ads, such experiments have failed to yield convincing evidence that such pheromones exist (Brennan & Zufall, 2006; Hays, 2003). Certainly some people find some of the odorous substances secreted by other people to be pleasant, but individual differences are great, and no specific human secretion has been found to be consistently attractive to members of the opposite sex. Perhaps that should not be surprising. Sex-attractant pheromones are valuable for animals that mate only at certain times of the year or only when the female is ovulating, as a means of synchronizing the sex drives of males and females to maximize the chance of conception. As discussed in Chapter 6, humans have taken a different evolutionary route, such that sexual drive and behavior are not tied to a season, cycle, or variable physiological state. For that reason, perhaps, there is little or no need for us to advertise by scent our readiness to mate. Although there is some evidence that underarm secretions of men influence women’s reproductive cycles (Cutler et al., 1986; Preti et al., 2003), it is unclear whether pheromones or some other substance is responsible for this influence.

There is still debate about whether humans have pheromones, but that hasn’t stopped the popular media from giving the impression that they exist and influence our behavior

©The New Yorker Collection, 1993, Lee Lorenz, from cartoonbank.com. All Rights Reserved.