The Folly of Fools: The Logic of Deceit and Self-Deception in Human Life - Robert Trivers (2011)

Chapter 6. The Immunology of Self-Deception


So far we have concerned ourselves with an individual’s relationship to the outside world—his or her competitors, friends, mates, and family. How does success or failure in each of these relationships involve deceit and self-deception? What kinds of self-deception are special to each realm, and what are their costs? But there is also an inner world that has strong effects on the costs and benefits of self-deceptive behavior (costs and benefits, as usual, are ultimately defined and measured by their effects on survival and reproduction). This inner world consists of a very large number of parasites (which cause disease)—invading organisms bent on eating us from the inside—and a very complex immune system of our own arrayed against them.

The importance of this world to self-deception comes primarily from the fact that the immune system is very expensive. It can act as an immense reservoir of energy and proteins and is very flexible—benefits and costs can be transferred to other functions at the flick of a molecular switch. Divert resources to attacking another male for possible immediate reproduction? Let’s deal with disease later. Such decisions have very important downstream effects on health, freedom from disease, and ultimately survival and reproduction. And many of these decisions, as we shall see, involve choices between psychological states with differing degrees of self-deception. Put differently, self-deception may have strong negative or, less often, positive effects on the immune system and therefore survival and reproduction—in short, reproductive success (RS).

The inner world is populated by a series of antagonistic actors, mostly parasites—that is, species specialized to attack and devour us from the inside but also including cancer cells, mutated forms of one’s own cells now replicating out of control. Parasites come in such major categories as viruses, bacteria, fungi, protozoa, and worms. They cause an enormous array of diseases: malaria, AIDS, rheumatic fever, tuberculosis, pneumonia, dysentery, smallpox, mumps, whooping cough, and elephantiasis, to name only some of the deadlier forms. Indeed, it is a sobering thought that more than half of all species on earth are parasitic on the other half—and this is a gross underestimate of the relative frequency of the two, since species of parasites are usually much smaller and harder to detect than are their host species. Most parasites have relatively mild effects, but in aggregate effects on RS, the inner world of parasites is almost as important as the outer, causing perhaps as much as 30 percent of total mortality every generation. This huge selective force has generated a very large, complex, and highly diverse system to counter the internal enemies—our immune system.

The immune system sends many cellular types to detect, disable, engulf, and kill invading organisms. One part, the innate immune system, is automatic, acts as the first line of defense, and does not rely heavily on learning. The second is based on experience and learning, the preferential production of defenses against parasites one has already encountered. This system produces as many antiparasite defenses (antibodies) as there are parasites. It has been called our “sixth sense,” directed inward to spot invaders as well as cancer cells and stop them. This kind of defense, with a detailed memory of past parasitic attacks, is so important it is found even in bacteria (whose parasites are viruses).

So disease is important and we invest heavily in protecting ourselves from it—nothing surprising there. What does this have to do with deceit and self-deception? Surprisingly enough, the answer is “a lot.” As we shall see, hiding one’s sexual orientation (or HIV status) is costly—not just in social relations and identity but in impaired immune function and associated early death. Shame, guilt, and depression are all associated with depressed immune function, but shame has greater effects than does guilt. Sharing thoughts about a trauma—even with a private journal—is associated with improved immune function. Good marriages appear to be associated with immune benefits and bad ones with immune costs. Meditation that improves mood also improves immune function. Religiosity is associated with better immune function, as is optimism. And so on. In short, there seems to be a general rule that suppressing the truth is costly to immune function and health, as is negative affect. The key is to understand why. Why should psychological suppression of reality be associated with immune costs and sharing reality or facing it, with immune benefits? And why should an upbeat personality be associated with immune benefits, and depression with immune costs?

Perhaps the most important aspect of the immune system in this regard is its enormous cost, measured in energy and protein consumption. These resources can easily be diverted for other purposes. No one has figured out yet how to estimate the aggregate cost of the immune system, whether in energy or in other critical units, but there can be no doubt that it is large, probably on the order of the brain itself (20 percent of resting metabolic energy). We turn first to this key point.


The beginning of wisdom about our immune system is to understand that it is extremely costly, both in energy and in the building blocks of life, proteins. It is ongoing and active twenty-four hours a day, seven days a week. To keep it running, every two weeks (roughly the maximum life span of many white blood cells), the body produces a set of cells greater in volume than two grapefruits. Some immune cells are among the most metabolically active cells in the body. Each of several thousand B cells specialized to produce antibodies grinds out about two hundred antibodies per second. Put differently, in one day’s time, they generate their own weight in antibodies, the proteins that bind to parasites and disable them. Of course, they can manage this feat for only about a day and a half and must be continually replenished. Because the immune system employs a bewildering array of cell types in a very complex manner, nobody has come close to estimating its total metabolic cost, though survival costs of heightened immune activity have been measured in several bird species. Mice lacking an immune system have been created in the lab, but these animals are prone to infections of every sort and must be maintained in sterile or near-sterile conditions, where they do not thrive, in part because they are not exposed to the useful bacteria we depend upon (for digestion and skin health, for example).

Scientists have been able to show that the short-term immune response to an immediate parasite attack typically is costly in energy. Fever is often a response because it is harder on the parasite than on the host, but for every 1 degree C increase in human temperature due to fever, there is about a 15 percent increase in metabolic rate (roughly translated: the rate at which we consume energy), so the response is costly. Immunizations, which merely mimic parasite attack, commonly elevate metabolic rate by about 15 percent for several days, while real attacks impose twice the metabolic cost per unit time. This is measured not only in energy but also protein consumed—as much as 20 percent loss in total body protein in sick humans, while in some sick rats more than 40 percent of muscle protein is broken down and new synthesis is sharply reduced. Chickens reared in germ-free environments enjoy about a 25 percent gain in body weight compared to those raised in conventional environments. Of course, this reflects absence of immune costs as well as those of the parasites themselves. The metabolic requirements of mammals raised in germ-free environments drops by as much as 30 percent. Supplying antibiotics in food is associated with growth gains in birds and mammals on the order of 10 percent. The take-home message should be clear. Inside us is a system of which we are mostly unconscious that is vast, powerful, and very expensive. As we shall see, it has numerous psychological correlates, cause and effect often go in both directions, and processes of self-deception produce striking effects.

It is also striking that about one-tenth of all the proteins our cells produce are promptly degraded and their peptides recycled—a wasteful process involving largely two cell organelles specialized for this purpose (the proteosome and lysosome). Some of this involves regulating proteins that are being produced at too high numbers or are misshapen, but the rest consists of grinding up proteins made by viruses, bacteria, and cancerous cells, both to mediate their effects and to recognize them for future attack.

Thus the immune system is expensive in both energy expended and proteins consumed. But this also means that it is an energy and protein reservoir that can be drawn on for other purposes—and this is probably the key to understanding many of its behavioral and psychological correlates.

One piece of evidence for how expensive (and important) the immune system is comes from “sickness behavior”—the cost the immune system imposes on the rest of the body when it needs to repair itself. Right after the immune system has fought off a parasitic invader—let us say a virus or bacteria—it is physiologically exhausted. It has drawn down heavily on its own resources to deal with the invader, and it now needs to rebuild itself to be ready for the next one. To do this, it induces a state of torpor, apathy, and lack of interest in life in the larger organism—the “blahs.” This is achieved by releasing a hormone (a particular cytokine) that acts on the brain to make the person anhedonic, that is, not taking pleasure in anything. In rats, this can be shown experimentally by releasing into healthy individuals the immune cytokine that targets the brain—the rat simply will not work as hard (on a treadmill) for sugar or other rewards.

To me, this finding was especially striking because I had always thought you felt bad after the initial attack of parasites (disease) because you were still fighting them, perhaps just mopping up operations but still enough to keep the immune system busy. Now I see that the immune system—fresh from heroic work on the barricades—merely wants to rebuild itself, and can we kindly help out by becoming inactive? To redirect energy to itself, the immune system makes other activities unrewarding so they will no longer be sought out. Internally you experience this as akin to depression. Would we suffer it better if we understood its purpose and went along with the program? Stay in bed; do not try to eat or have sex or pursue other activities that are usually fun but that make demands on the immune system and its regeneration—be satisfied with a “vacation from pleasure.” Preserve your energy and be humble. Things will soon get better.


A profound role for sleep and immune replenishment is emerging from a variety of studies. The simple logic goes as follows—more sleep is more time for immune system regeneration (which occurs preferentially at low activity levels, such as during sleep). But self-deception often interferes with sleep. It causes internal conflict and dissatisfaction—tossing and turning mentally and physically. Since active suppression of thoughts and repression of emotions may cause a rebound effect—people may think more about what they are trying to suppress than if they didn’t even try—it may directly interfere with sleep. Other things being equal, one predicts better sleep—and, therefore, better health—with less self-deception.

What the immune work shows is that there is a direct, strong, and positive relationship between sleep, immune function, and health: the more the better. Mammals generally respond more strongly to infection with increased sleep, while those rabbits that sleep more following artificial infection survive better. Meanwhile, totally sleep-deprived rats soon die from systemic bacterial infections. It is probably wise to be conscious of this connection. If you find yourself sleeping more, you may already be infected. You should probably indulge the sleep and “go with the flow.”

Within a species, the more time individuals can spend sleeping, the higher are their white blood cell counts for most cell types, while red blood cells, which originate from the same tissues but are not part of the immune system, are unaffected. This correlation applies to both REM sleep (with dreaming) and non-REM sleep. Perhaps the most striking fact about the hidden benefits of sleep comes from comparing different species of mammals. Individuals from species that spend more time asleep are less likely to be infected by parasites. Mammal species range from those that sleep as little as three hours a night to those that sleep more than twenty-one. Across this range, species with ten more hours of sleep per night have rates of parasitism twenty-four times lower. In short, for long-sleeping species, life may be dull, but it sure is healthy. It is worth noting, however, that sleep and dreaming play complementary roles in consolidating memories acquired during wakefulness. Both are required for initial memory storage and then several days later, spreading the memories to the neocortex—the more social part of the brain. So for all we know, small species of mammals (with long sleep) may have superb memories.

We should also note that deliberate sleep deprivation, as practiced in various penal colonies and torture centers around the world, is expected to increase parasite attack on the victims (on top of its other negative effects).


Trade-offs appear to explain major hormonal correlates of immune activity. For example, testosterone suppresses immune function in males. Since increases in testosterone are associated with both sexual opportunities and aggressive threats, the body faced with either one appears in effect to be saying, “I will deal with my tapeworm later; right now I’ll use some of those immune resources to defeat a rival male, or perhaps enjoy an extra copulation.” Consistent with this, among the lowest testosterone levels are those found in men living monogamously and with children; next higher, monogamously and without children; higher still, monogamously with outside sexual activity; and highest of all, no children, no partner, in full competition. In fact, some homosexual men show the highest levels of testosterone of any, perhaps for just this reason: no parental investment, minimal marital ties, maximum male-male competition.

Health maps inversely on testosterone. Marriage tends, for example, to increase life span in men. As expected, work on monkeys, apes, and humans shows that males with higher testosterone are more likely to become infected (with such diseases as malaria) and that disease itself lowers testosterone levels—in other words, the body lowers testosterone levels to shift investment to its immune system. There is nothing magic about testosterone. It is only a signal, not a source of potency. Some of the same correlations are found in insects, in which testosterone is not involved: males have a weaker immune system than females and suffer higher parasite loads and lower survival, just as in most mammals. This difference is probably general to most animals—certainly males typically suffer higher mortality. A testosterone-associated trait—degree of fat-free muscle mass—is associated with greater self-reported sexual activity in men and earlier age at first sexual experience. The trait is also associated with higher energy consumption and lower immune function.

Likewise, corticosteroids—produced in response to stress and associated with anxiety and fear—are immune suppressors. For example, subordinate monkeys who are harassed by dominants are often high in corticosteroids and low in immune function. The immune correlation suggests that the immune system is making resources available for dealing with whatever is causing the stress and, in any case, for maintenance in the face of anxiety and fear—even if doing so temporarily increases risk of disease. (Of course, the effects of prolonged stress are another matter.) In short, whether we are pumped up on testosterone or empowered by a corticosteroid such as cortisol, we sacrifice our long-term internal defenses for short-term gains. We shall soon see that this may be yet another cost of self-deception, hyping the aggressive or the threatened, with adverse immune consequences.

The brain is also a very costly organ. Although representing only 3 percent of total body weight, the brain consumes 20 percent of all resting metabolic energy. When a person is awake, this price seems to be invariant. In the 1950s, it was shown that doing arithmetic did not require additional mental energy, a finding that now seems quaint, given that the 20 percent energy cost is known to be constant whether you are happy, depressed, schizophrenic, or on an LSD trip. The cost is slightly diminished during nondreaming sleep but slightly elevated during dreaming. Thus throughout the full twenty-four-hour cycle, the brain’s resting energy cost remains virtually constant. In our species, 20 percent is the price of poker—the price to play life with a functioning brain. You must pay it or else. Indeed, not paying it for five minutes typically leads to death or, at the least, irreversible brain damage. This is just a fact of life—and an extraordinary one at that.

The invariant cost is important because one might easily imagine that different psychological functions have different energetic costs. Perhaps part of the benefit of depression is that the brain thereby saves energy. No—depression appears to have no effect on the 20 percent of energy the brain extracts. If depression lowers energy demands, it does so by lowering overall activity and metabolic rate. Likewise, if repression (suppression of truth from the conscious mind) lowers immune function, as it appears to, this is unlikely to mean that repression itself requires extra energy over and above normal function, the energy being supplied by the immune system. Instead, we must look for other changes associated with repression—which the immune system then pays for.

It has also been known for some time that the brain is the most genetically active tissue in the human body. In other words, a higher percentage of genes are active in the brain than in all other tissues, almost twice as high as in the liver and in muscle, the nearest competitors. A good one-third of all genes are so-called housekeeping genes, useful in running most kinds of cells, so they are widely shared, but the brain is unique both in the total number of genes expressed and in the number expressed there and nowhere else. By some estimates, more than half of all genes express themselves in the brain: that is, more than ten thousand genes. This means that genetic variation for mental and behavioral traits should be especially extensive and fine-grained in our species—contra decades of social science dogma. This includes, of course, such traits as degree of honesty and degree and structure of deceit and self-deception.

What we do not know is what the parallel facts are for our immune system. How much of our genes are also activated there? Are there important chemicals common to both the brain and immune system so that depletion in one system causes problems in the other? Certainly we would expect there to be, and if there are we would expect to see immune/psychological correlates we would not otherwise imagine. An analogy may help. Beginning in 1982, it was shown that female birds choose brightly colored males as a way of getting parasite-resistant genes for their offspring. This result has been documented many times since then—both that females like brightly colored males and that such males are relatively low in parasite number. It seems to be difficult to be brightly colored and sick at the same time, but why? Only in the 1990s was it shown that carotenoids—which give us orange, yellow, and red and which are not manufactured by any vertebrate but must come from their diet—play a vital role in immune function. This means that a more active immune system—for example, in response to infection—must draw carotenoids from surrounding tissues to help fight the invaders, as indeed it does. Those that are strong and healthy have color to spare, which they move to the body’s exterior as an advertisement.

Are there important brain function genes that also have immune correlates? A possible example was first described in a honeybee. When the bee is given a harmless antigen to which it mounts a response, the response interferes with associative learning but not with perception or discrimination. Since it is unlikely that any of these activities increases the brain’s energy budget, the explanation must lie elsewhere. In honeybees we know that associative learning depends on octopamine, a chemical that happens to be important in their immune system. In vertebrates we know that cytokines produced by the immune system can directly affect the hippocampus and reduce memory consolidation, but the functional meaning is obscure. We know that parasitic infection has a dramatic and negative effect on learning abilities. This effect must result because the activated immune system deprives the brain of other chemicals vital for learning—or has other effects, such as a decrease in sleep or dreaming, both known to be vital in consolidating learning in various species.

In birds there is clearly an intimate relationship between the immune system and the brain, one that appears to be heightened by the action of sexual selection. Two organs are intimately involved in immune function (mostly B cell production and storage)—the bursa of Fabricius of juvenile birds and the spleen of adults. The relative size of these two organs is positively associated with relative brain size across a range of species: the bigger the brain, the greater the investment in the immune system.

This may in part be due to big brains’being associated with long life span (which places a premium on parasite defense), but the correlation is especially strong when the sexes differ in brain size. That is, the bigger the relative size of the male’s brain compared to the female’s, the greater the relative size of the two key antiparasite organs in the species. The assumption is that males are especially likely to suffer from parasite load and its associated cognitive impairment (shown numerous times for birds), so that selection, especially in big-brained birds, will favor heavier investment in immune functions the better to protect against cognitive impairment. In this view, the two systems are complementary—the greater the investment in one (the immune system), the better the functioning of the other (the brain), presumably because the brain is especially vulnerable to parasite damage. For example, river otters that are parasitized by nematode worms show brain damage and reduction in brain size, but the effects are more prominent in males. In humans it has recently been shown that national averages in adult intellectual development are lower the greater the average parasite load.


In a series of important experiments from the 1980s to the 2000s, scientists showed that writing about trauma produced clear immune benefits. Although most of this writing was done in English, the same effect holds for Spanish, Italian, Dutch, and Japanese, that is, broadly. In one set of experiments, people were asked to imagine the most traumatic event in their lives. They were then split into two groups—those who spent twenty minutes each day for four consecutive days writing in a private diary about their trauma and those who wrote for twenty minutes each day on superficial topics (for example, what they had done that day). Blood was drawn before the experiment began, after the last day of writing, and six weeks later. Although those writing on their trauma said they felt worse at the end of the writing than those who wrote on innocuous topics, their immune system already showed improvement, which was still detectable six weeks later, at which time they also reported feeling better (than those who had not written about their traumas). In summary, the immediate feeling of confronting trauma is negative but the immune effects tend to be positive, and the longer-term effects on mood and immune system are both positive.

Note that the positive immune effect precedes the positive effect on mood, and how little writing is necessary to beget a measurable immune effect some weeks later. A recent review of about 150 studies confirms that there is a general pattern in which emotional disclosure, even in the form of occasional autobiographical writings, is often associated with consistent immune benefits.

Writing about trauma in a private journal in a lab is obviously an evolutionarily recent event, but it probably acts as a substitute for sharing this information with others. Certainly rituals of confession are common in most religions, whether public, as in many New World Amerindian religions, or private, as in the Catholic confessional. Indeed, the injunction to confess one’s sins to God herself in prayer may serve a similar disclosure function. The benefits of the “talking cure,” psychotherapy, may also arise in part from disclosing traumatic or shameful information that one is, in fact, hiding from others. When traveling, we will often tell secrets to complete strangers, people we have never met before and, crucially, do not expect to see again. The more that people talk in small groups, the more they claim to have learned from the group. As one psychologist drily notes, sharing our thoughts is apparently “a supremely enjoyable learning experience.” For this reason, particular theories of human development—say, Freud’s psychosexual stages—may be as valid as astrology, yet talking to one’s analyst may provide benefits for the same reason that writing in a journal does.

One important possibility is that some of these positive correlations may in fact be caused by effects on sleep. If disclosing trauma to others results in fifteen more minutes of sleep, or at least less fitful sleep, this alone could induce the known immune benefits. A striking effect of disclosure is how quickly the benefit kicks in, as would happen if it immediately led to less troubled sleep. One final feature of the work on expressive writing is worth emphasizing. Computer-based analysis has isolated three aspects of the writing that produce beneficial effects: emotion words, cognitive words, and pronouns. The more people use positive emotion words, the more their health improves. Even writing “not happy” is better than writing “sad,” perhaps because the focus in the first remains on the positive emotion. Using lots of negative emotion words and none at all are both associated with no benefit, while a moderate number is. Perhaps one is overwhelmed in the first case and in complete denial in the second. The value in taking alternative perspectives on a problem is suggested by the fact that changing back and forth from the first person (“I,” “me,” “my”) to all other pronouns (“they,”“she,”“we”) is associated with improvement, while remaining in one or the other perspective is not.

Conversely, there is evidence that inhibition is associated with health problems. Consistent with this, those with undisclosed childhood traumas (sexual, physical, or emotional abuse, parental death or divorce) show more illness as adults, including cancer, high blood pressure, flu, headaches, and so on. In one study, 10 percent reported sexual trauma before age seventeen, and these people had the greatest health problems of any group—fewer than half had ever discussed the problem. From this, one might easily imagine that a spouse’s suicide, for example, would be talked about less than spousal death by other causes and would be expected to be more traumatic. But in fact suicide support groups permit more talking about these kinds of deaths, a nice example of a cultural invention that permits sufferers to come together to enjoy the benefits of sharing and disclosure.

One striking effect of writing about recent traumas is not immunological but still important: writing about job loss improves one’s chance of reemployment. This sort of writing appears to be cathartic—people immediately feel better. More striking, at least in one study, is a sharply increased chance of getting a new job. After six months, 53 percent of writers had found a new job, compared with only 18 percent of nonwriters. One effect of writing is that it helps you work through your anger so it is not displaced onto a new, prospective employer or, indeed, revealed to the employer in any form. This presumably makes you more attractive to them.


Given the global importance of HIV and AIDS, it is hardly surprising that the effects of disclosing or suppressing information have been well studied in those who are infected with HIV. Here disease progression itself can be taken as a sensitive measure of immune function, and the main findings above have been replicated almost exactly. Even relatively modest writing interventions improve apparent health status (immune chemicals per viral load). A form of “expressive” group therapy also lowers viral counts while boosting an immune measure. As has been discovered more generally, the writing/disclosure benefits tend to occur only when the writing includes increasing insight/causation and social words. Whether this is cause and effect or merely diagnostic is not known, but the correlation is strong.

Homosexuality and HIV status also turn out to be especially useful in studying deceit and self-deception because each invites a form of denial that, unlike the experimental work, occurs almost daily over a long period of time. Homosexual men often differ in the number of people to whom they reveal their sexual identity (degree to which they are “out of the closet”)—from only a few heterosexual close friends, to those plus one’s family, to all of those plus one’s workmates, to the whole world. Likewise, it is possible to deny HIV-positive status to others and to attempt to deny it to self. All of these efforts bring negative immune and health effects, which may be substantial.

Relative to HIV-positive men who are mostly or completely out of “the closet,” those who were at least half in the closet enjoyed 40 percent less time before they suffered from AIDS itself and 20 percent lower survival rate overall. Three separate studies show that denying one’s HIV-positive status to others or even to self (“I am not really sick”) is associated with lower immune function and/or more rapid progression of the eventually fatal HIV infection. In HIV-positive women, evidence of emotional support was not associated with immune change but evidence of psychological inhibition (use of inhibition words in daily speech) was associated—more inhibition, faster immune decay.

One study of the progression of HIV in gay men as a function of the degree the men were in the closet also controlled for unprotected sex of the dangerous kind (anal receptive). Sure enough, those in the closet practiced more of this kind of sex (being in denial, they probably prepared less for the sex likely to occur later that night). This factor had a positive effect on the rate at which their HIV progressed (probably due to the addition of competing HIV strains), but independently, being in the closet was much more harmful for resistance to HIV. At least in this respect, truth appears to be healthy for the organism expressing it: your immune system is stronger, and at the same time you are more conscious—in this case, less likely to act in obviously self-destructive ways. The US government’s recent policy on service by homosexuals, “don’t ask, don’t tell,” is an immunological disaster. You are asked to deny your sexual identity, which will invite a host of unwanted and unnecessary immune problems for you, all in order to keep everyone else relaxed.

Here is one vivid account of what it would be like to hide your heterosexual identity if this were required (as in the US military):

Try never mentioning your spouse, your family, your home, your girlfriend or boyfriend to anyone you know or work with—just for one day. Take that photo off your desk at work, change the pronoun you use for your spouse to the opposite gender, guard everything you might say or do so that no one could know you’re straight, shut the door in your office if you have a personal conversation if it might come up. Try it. Now imagine doing it for a lifetime. It’s crippling; it warps your mind; it destroys your self-esteem. These men and women are voluntarily risking their lives to defend us. And we are demanding they live lives like this in order to do so.

The ill effects of concealing one’s homosexual orientation are not limited to HIV-positive men. In a sample of twenty-two HIV-negative gay men studied for five years, those who concealed their homosexual identity were about two times as likely to suffer cancer and infectious diseases, such as bronchitis and sinusitis, as those who did not. These results are independent of a variety of potentially confounding factors such as age, socioeconomic status, drug use, exercise, anxiety, depression, and so on. What is especially striking is that for both cancer and infectious diseases, the effect is strictly dose-dependent—the more you are in the closet, the worse for you. Recent evidence suggests that disclosing homosexual orientation may bring correlating cardiovascular benefits as well.

Not all homosexual men are alike, of course; some are more sensitive to rejection than are others and this can have important effects. Those who are more rejection-sensitive are more likely to remain in the closet, where they avoid rejection and benefit from this immunologically. Apparently there is a general cost to remaining in the closet, but a variable benefit when one is rejection-sensitive, and this benefit can overwhelm the cost.

Have you heard of the latest twist in this saga? There are gay men who are said to be living in a glass closet. They project heterosexuality to their friends, because they believe they would be rejected if people knew about their homosexuality, but in fact the friends know about it and merely go along with the charade. It would be interesting to know where these men lie along the immune continuum. I would guess they are healthier than those in conventional closets, but not by much.


Direct experimental tests confirm a strong association between positive affect and immune function but are unclear regarding the correlates of negative affect. Challenging people who have never been exposed to hepatitis B with a hepatitis B vaccine shows a clear positive association between positive affect and a strong, positive immune response, no matter whether the measure of positive affect emphasizes calm, well-being, or vigor. Although negative affect has the opposite effect, this was not significant when corrected for positive affect. In general, it seems as if positive affect is not merely the absence of negative and vice versa. In some cases negative and positive affect act as independent variables and in others as only partly independent ones.

The activity of neurotransmitters such as dopamine and serotonin provides a partial explanation. Dopamine shows a phasic spike in single neurons in response to the anticipation of a reward. If the reward equals expectation, the spikes continue apace; if it exceeds, the spikes increase in rate, and if it is less than anticipated, the spikes shrink to less than the spiking baseline rate for negative rewards. Positive affect increases both dopamine and serotonin production, but negative affect has no direct effect on dopamine (though it may indirectly do so via serotonin production). Dopamine modulates immune functioning and there is an asymmetry between positive and negative affect—positive having stronger effects than negative—both on cognitive and immune function. The deeper reason for this asymmetry remains unclear.

Measures of positive affect are also associated with better survival in relatively healthy elderly people who are living independently in their communities, but curiously enough, positive affect appears to be associated with reduced survival among those already institutionalized. Likewise, those with terminal conditions, such as malignant melanoma and metastatic breast cancer, are worse off with positive affect, but in diseases with higher long-term survival, such as AIDS and non-metastatic breast cancer, positive affect is beneficial.

A possible functional explanation for these anomalies comes again from considering the rate of reward necessary to maintain positive affect and positive immune function. If your body is deteriorating quickly and you feel bad because your illness is proceeding so fast, then the expected reward of positive dopamine spikes dies down rapidly, as do the dopamine spikes. This reduces the positive cognitive and immune benefits of enhanced dopamine production. If, on the other hand, a person is caught in a long-term degenerative condition, the rate of deterioration may be slow enough that dopamine spikes and positive affect are capable of generating the cycle of positive feedback necessary to sustain improved mental and immune function.


By choosing to listen to music, people can alter their mood and their immune system. Some of the music experiments are almost too good to be true. For example, Musak (bland, peaceful music designed to calm people in a claustrophobic situation, such as an elevator) produced an increase in output of an important immune chemical by 14 percent, while jazz did so by only 7 percent. No sound had no effect, and simple noise had a 20 percent negative effect. Melodic music may suggest a happy and harmonious structure to the immediate world, while noise is cacophonous and connotes disorder, uncertainty, even danger. Music composed to match the pitch and tempo of natural monkey (tamarin) sounds but not using the monkey sounds themselves induced behavioral changes in the lab in tamarins similar to those observed in our own species. Though tamarin music based on threat vocalizations induced more anxious activity, music based on positive social interactions had positive effects: less surveillance, less sociality, and more foraging—exactly what one finds in other animals when external threat is reduced. Almost certainly there were parallel immune changes, negative to threat and positive to warm affect, so that the human response to music must have a very long past.

Two recent results stand out. Injecting about five hundred cancer cells into mice that have been stressed by exposure to noise at midnight results in much less cancerous growth if the mice then enjoy five hours of melodious music each morning. An equally dramatic example comes from humans. People undergoing bronchial physiotherapy (aspirating medicine, breathing, coughing) while listening to Bach’s music (in a major key) recover much more quickly than those enjoying the therapy without music. (Minor keys show neutral or negative effects.) The point is that the right kind of music can induce positive feelings that are in turn associated with positive immune and health effects.

Certainly we know that female choice has forced a cognitive burden on males, the better to keep the females entertained. Song repertoire size in birds, which is favored by females, is controlled in males by a substantial set of neurons in the brain that completely regress during the nonbreeding season (clear evidence of the cost of running the show). We would expect pleasing male song to be both sexually arousing in females and immunologically positive. The same thing might be said for human courtship and for relations between a pair—surely there are many immunologically positive interactions possible on both sides, including good sex, and many negative ones, such as conflict, anger, suppressed feelings, and bad sex.


I suggest that an old-age positivity effect operates in a similar fashion to choosing to listen to pleasing music. By age sixty (if not earlier), a striking bias sets in toward positive social perceptions and memories. The original experiment had people looking at two faces next to each other on a screen, one with a neutral expression and one with either a positive or a negative one. After one second, the faces are removed from the screen and a dot appears where one of the faces was located. The person must hit a button as soon as the dot is perceived, one button for left side, one for right side. At ages twenty to thirty, people are equally quick to spot the dot no matter what face it was associated with. But by age sixty, a bias appears: the dot is perceived more quickly if it succeeds the positive face and more slowly if it succeeds the negative one. Study of eye movements shows that the older people spend more time inspecting faces with positive expressions than negative, and the positive ones are remembered later more often. Young people show none of these biases. These results are true among Asians, Europeans, and Americans. They appear to involve a measurable effect in the amygdala, where positive faces evoke a stronger response than negative ones in older people but not in younger people. Finally, older people tend to respond to a negative mood induced by unpleasant music by preferentially looking at positive faces, as if attempting actively to counter the negative and maintain or induce a positive mood. Young people tend, if anything, to be mood congruent—if made to feel bad, they look more at negative faces.

Why show such a positivity bias? Young people would be wise to pay attention to reality—both positive and negative—the better to make the appropriate responses later. Avoiding negative information seems risky on its face—negative events may have as big an effect on one’s interests (inclusive fitness) as positive ones. By contrast, in old age it hardly matters what you learn, but greater positive affect is associated with stronger immune response, so you may be selected to trade a grasp of reality for a boost in dealing with your main problem, that of internal enemies, including cancer. A positivity bias sacrifices attention to and learning from negative stimuli the better to enjoy strong immune function now. If you haven’t learned to spot an external enemy by now, chances may be low that you will learn to, and in the meantime you can enjoy a positive mood and immune response. Grandchildren may admire Gramps and Grandma because nothing seems to faze them, but Gramps and Grandma are living in positivity land—they may scarcely know the difference.

It is an interesting coincidence that although people’s implicit bias in favor of youth over old age hardly changes with age (as measured by an IAT)—from twenty to seventy, they favor young over old—by our forties, our explicitbias in favor of youth (what we say we care about) declines until at exactly sixty, people start to say they think older is better than younger. Like everyone else, they implicitly associate youth with positive features, but they start preaching the opposite at roughly the same time they display the old-age positivity bias.

Note that the positivity effect requires no suppression of negative information or affect. The bias occurs right away. People simply do not attend to the negative information, do not look at it, and do not remember it. Thus, the possible negative immune effects of affect suppression do not need to arise. This must be a general rule—the earlier during information processing that self-deception occurs, the less its negative downstream immunological effects. At the same time, there may be greater risk of disconnect from reality, since the truth may be minimally stored or not at all.

Given what I have just said, the question arises of why old people are often perceived as being cranky or grumpy. This appears to result from an entirely independent mechanism, which sometimes cancels out or overwhelms the positivity bias. With increasing age, for reasons that are not entirely clear, people suffer greater deficits in their inhibitory abilities, that is, their ability to stop behavior under way that they may wish to stop. Since people often wish to inhibit behavior that will be seen as socially inappropriate, it is not surprising that with increasing age comes exactly that, increasingly socially inappropriate behavior. This includes discussion of private material in public, more frequent overt expressions of prejudice and stereotype, greater difficulty taking the perspective of another, and more off-target verbosity (“Don’t get me started!”). Perhaps many of these traits are later described or rationalized by saying that Gramps sure is “cranky” today.


All of this work is consistent with an immunological theory of human happiness in which a finely tuned immune system purring along at near-peak efficiency with hardly a target in sight would be experienced internally as a highly enjoyable state. Even such variables as absence of food (hunger) or water (thirst) must be at least partly aversive because of their negative effects on the immune system. At the very least, it must be true that as the brain looks outward and acts to increase inclusive fitness in part by increasing happiness, then surely the same must be true when looking inwardly.

According to this view, the brain is split between outward-directed and inward-directed activity. In the outside world, many features are stationary and predictable—the shape of your bedroom, the location of food in your refrigerator, the way to work, etc. Within this world, of course, there is important variation: a predator appears, a food source, a possible mating opportunity, a hole in the street, for all of which you are selected to make appropriate responses. You have an internal reward/punishment system that goads you in appropriate directions.

Now imagine the whole thing all over for the internal system. Your brain looks inward and sees many constant features—feet and hands farther from it than the trunk, a particular circulatory system through which almost all chemicals must ultimately pass, including those produced by the brain to regulate downstream chemical activity. But in this world also live (in principle) hundreds and even thousands of species of parasites, at the moment just a few, perhaps, but taking particular configurations that need to be countered. The brain may receive or note signals that a major infection is under way in the lower left abdomen but miss the fact that a core of parasitic cells resides in the right big toe and are capable of generating the primary attack.

One important distinction concerns consciousness. We are highly conscious of interactions outside our bodies but highly unconscious of interactions within the body. Why? Part of it is that many signals to self need no consciousness, but one wonders why we are so unconscious of parasitic interactions—for example, failing to appreciate the meaning of “sickness behavior” or the value of more sleep.

Despite its importance, almost no attention has been directed toward measuring the correlates of immune function with such major components of individual fitness—or reproductive success—as survival, fecundity, physical attractiveness, and so on. The comparative work has all been done in birds. Here the pattern is clear. A greater natural immune response to some kind of challenge is positively associated with survival in nature and in the lab, and the effect size is relatively large—18 percent of variation in survival is explained by immune variation, while the closest competitor, degree of bodily symmetry, explains only 6 percent of variance in survival.

How is optimism related to immune function? A number of studies have shown a positive correlation between optimism and health outcomes, immune function, and survival. A recent study is especially striking. Law students were assayed five times throughout the year both for optimism regarding their studies and for a major immune parameter. Within a student’s year, high optimism was associated with high immune function, but when comparing students, there was no effect; that is, optimistic students were not more likely to have stronger immune systems. Although psychologists almost uniformly assume that mood affects immune system, the reverse is equally plausible. With your immune system at near-top efficiency, you should feel happy, positive, and optimistic.

The psychological and immune systems are deeply intertwined, cause and effect go in either direction, and it is hardly possible for one system to react without affecting the other. For reasons that are not always obvious, self-deception appears to have strong immune effects, usually according to the rule more self-deception, lower immune strength, but occasionally, more self-deception, better immune function.

This field is still in its infancy. Some interesting things are known, but much more remains to be found out. Which levels of information suppression are associated with what immune effects? And what chemicals are common to the brain and the immune system, leading to important trade-offs between the two? And what questions do we not even know enough to ask?