Importantly, pheromones signal fear. In one study researchers got armpit swabs from volunteers under two conditions—either after contentedly sweating during a comfortable run, or after sweating in terror during their first tandem skydive (note—in tandem skydives you’re yoked to the instructor, who does the physical work; so if you’re sweating, it’s from panic, not physical effort). Subjects sniffed each type of sweat and couldn’t consciously distinguish between them. However, sniffing terrified sweat (but not contented sweat) caused amygdaloid activation, a bigger startle response, improved detection of subliminal angry faces, and increased odds of interpreting an ambiguous face as looking fearful. If people around you smell scared, your brain tilts toward concluding that you are too.25
Finally, nonpheromonal odors influence us as well. As we’ll see in chapter 12, if people sit in a room with smelly garbage, they become more conservative about social issues (e.g., gay marriage) without changing their opinions about, say, foreign policy or economics.
Interoceptive Information
In addition to information about the outside world, our brains constantly receive “interoceptive” information about the body’s internal state. You feel hungry, your back aches, your gassy intestine twinges, your big toe itches. And such interoceptive information influences our behavior as well.
This brings us to the time-honored James-Lange theory, named for William James, a grand mufti in the history of psychology, and an obscure Danish physician, Carl Lange. In the 1880s they independently concocted the same screwy idea. How do your feelings and your body’s automatic (i.e., “autonomic”) function interact? It seems obvious—a lion chases you, you feel terrified, and thus your heart speeds up. James and Lange suggested the opposite: you subliminally note the lion, speeding up your heart; then your conscious brain gets this interoceptive information, concluding, “Wow, my heart is racing; I must be terrified.” In other words, you decide what you feel based on signals from your body.
There’s support for the idea—three of my favorites are that (a) forcing depressed people to smile makes them feel better; (b) instructing people to take on a more “dominant” posture makes them feel more so (lowers stress hormone levels); and (c) muscle relaxants decrease anxiety (“Things are still awful, but if my muscles are so relaxed that I’m dribbling out of this chair, things must be improving”). Nonetheless, a strict version of James-Lange doesn’t work, because of the issue of specificity—hearts race for varying reasons, so how does your brain decide if it’s reacting to a lion or an exciting come-hither look? Moreover, many autonomic responses are too slow to precede conscious awareness of an emotion.26
Nonetheless, interoceptive information influences, if not determines, our emotions. Some brain regions with starring roles in processing social emotions—the PFC, insular cortex, anterior cingulate cortex, and amygdala—receive lots of interoceptive information. This helps explain a reliable trigger of aggression, namely pain, which activates most of those regions. As a repeating theme, pain does not cause aggression; it amplifies preexisting tendencies toward aggression. In other words, pain makes aggressive people more aggressive, while doing the opposite to unaggressive individuals.27
Interoceptive information can alter behavior more subtly than in the pain/aggression link.28 One example concerns how much the frontal cortex has to do with willpower, harking back to material covered in the last chapter. Various studies, predominantly by Roy Baumeister of Florida State University, show that when the frontal cortex labors hard on some cognitive task, immediately afterward individuals are more aggressive and less empathic, charitable, and honest. Metaphorically, the frontal cortex says, “Screw it. I’m tired and don’t feel like thinking about my fellow human.”
This seems related to the metabolic costs of the frontal cortex doing the harder thing. During frontally demanding tasks, blood glucose levels drop, and frontal function improves if subjects are given a sugary drink (with control subjects consuming a drink with a nonnutritive sugar substitute). Moreover, when people are hungry, they become less charitable and more aggressive (e.g., choosing more severe punishment for an opponent in a game).* There’s debate as to whether the decline in frontal regulation in these circumstances represents impaired capacity for self-control or impaired motivation for it. But either way, over the course of seconds to minutes, the amount of energy reaching the brain and the amount of energy the frontal cortex needs have something to do with whether the harder, more correct thing happens.
Thus, sensory information streaming toward your brain from both the outside world and your body can rapidly, powerfully, and automatically alter behavior. In the minutes before our prototypical behavior occurs, more complex stimuli influence us as well.
Unconscious Language Effects
Words have power. They can save, cure, uplift, devastate, deflate, and kill. And unconscious priming with words influences pro- and antisocial behaviors.
One of my favorite examples concerns the Prisoner’s Dilemma, the economic game where participants decide whether to cooperate or compete at various junctures.29 And behavior is altered by “situational labels”—call the game the “Wall Street Game,” and people become less cooperative. Calling it the “Community Game” does the opposite. Similarly, have subjects read seemingly random word lists before playing. Embedding warm fuzzy prosocial words in the list—“help,” “harmony,” “fair,” “mutual”—fosters cooperation, while words like “rank,” “power,” “fierce,” and “inconsiderate” foster the opposite. Mind you, this isn’t subjects reading either Christ’s Sermon on the Mount or Ayn Rand. Just an innocuous string of words. Words unconsciously shift thoughts and feelings. One person’s “terrorist” is another’s “freedom fighter”; politicians jockey to commandeer “family values,” and somehow you can’t favor both “choice” and “life.”*30
There are more examples. In Nobel Prize–winning research, Daniel Kahneman and Amos Tversky famously showed word framing altering decision making. Subjects decide whether to administer a hypothetical drug. If they’re told, “The drug has a 95 percent survival rate,” people, including doctors, are more likely to approve it than when told, “The drug has a 5 percent death rate.”*31 Embed “rude” or “aggressive” (versus “considerate” or “polite”) in word strings, and subjects interrupt people more immediately afterward. Subjects primed with “loyalty” (versus “equality”) become more biased toward their team in economic games.32
Verbal primes also impact moral decision making.33 As every trial lawyer knows, juries decide differently depending on how colorfully you describe someone’s act. Neuroimaging studies show that more colorful wording engages the anterior cingulate more. Moreover, people judge moral transgressions more harshly when they are described as “wrong” or “inappropriate” (versus “forbidden” or “blameworthy”).
Even Subtler Types of Unconscious Cuing
In the minutes before a behavior is triggered, subtler things than sights and smells, gas pain, and choice of words unconsciously influence us.
In one study, subjects filling out a questionnaire expressed stronger egalitarian principles if there was an American flag in the room. In a study of spectators at English football matches, a researcher planted in the crowd slips, seemingly injuring his ankle. Does anyone help him? If the plant wore the home team’s sweatshirt, he received more help than when he wore a neutral sweatshirt or one of the opposing team. Another study involved a subtle group-membership manipulation—for a number of days, pairs of conservatively dressed Hispanics stood at train stations during rush hour in predominately white Boston suburbs, conversing quietly in Spanish. The consequence? White commuters expressed more negative, exclusionary attitudes toward Hispanic (but not other) immigrants.34
Cuing about group membership is complicated by people belonging to multiple groups. Consider a famous study of Asian American women who took a math test.35 Everyone knows that women are worse at math than me
n (we’ll see in chapter 9 how that’s not really so) and Asian Americans are better at it than other Americans. Subjects primed beforehand to think about their racial identity performed better than did those primed to think about their gender.
Another realm of rapid group influences on behavior is usually known incorrectly. This is the “bystander effect” (aka the “Genovese syndrome”).36 This refers to the notorious 1964 case of Kitty Genovese, the New Yorker who was raped and stabbed to death over the course of an hour outside an apartment building, while thirty-eight people heard her shrieks for help and didn’t bother calling the police. Despite that being reported by the New York Times, and the collective indifference becoming emblematic of all that’s wrong with people, the facts differed: the number was less than thirty-eight, no one witnessed the entire event, apartment windows were closed on that winter’s night, and most assumed they were hearing the muffled sounds of a lover’s quarrel.*
The mythic elements of the Genovese case prompt the quasi myth that in an emergency requiring brave intervention, the more people present, the less likely anyone is to help—“There’s lots of people here; someone else will step forward.” The bystander effect does occur in nondangerous situations, where the price of stepping forward is inconvenience. However, in dangerous situations, the more people present, the more likely individuals are to step forward. Why? Perhaps elements of reputation, where a larger crowd equals more witnesses to one’s heroics.
Another rapid social-context effect shows men in some of their lamest moments.37 Specifically, when women are present, or when men are prompted to think about women, they become more risk-taking, show steeper temporal discounting in economic decisions, and spend more on luxury items (but not on mundane expenses).* Moreover, the allure of the opposite sex makes men more aggressive—for example, more likely in a competitive game to punish the opposing guy with loud blasts of noise. Crucially, this is not inevitable—in circumstances where status is achieved through prosocial routes, the presence of women makes men more prosocial. As summarized in the title of one paper demonstrating this, this seems a case of “Male generosity as a mating signal.” We’ll return to this theme in the next chapter.
Thus, our social environment unconsciously shapes our behavior over the course of minutes. As does our physical environment.
Now we come to the “broken window” theory of crime of James Q. Wilson and George Kelling.38 They proposed that small signs of urban disarray—litter, graffiti, broken windows, public drunkenness—form a slippery slope leading to larger signs of disarray, leading to increased crime. Why? Because litter and graffiti as the norm mean people don’t care or are powerless to do anything, constituting an invitation to litter or worse.
Broken-window thinking shaped Rudy Giuliani’s mayoralty in the 1990s, when New York was turning into a Hieronymus Bosch painting. Police commissioner William Bratton instituted a zero-tolerance policy toward minor infractions—targeting subway fare evaders, graffiti artists, vandals, beggars, and the city’s maddening infestation of squeegee men. Which was followed by a steep drop in rates of serious crime. Similar results occurred elsewhere; in Lowell, Massachusetts, zero-tolerance measures were experimentally applied in only one part of the city; serious crime dropped only in that area. Critics questioned whether the benefits of broken-window policing were inflated, given that the approach was tested when crime was already declining throughout the United States (in other words, in contrast to the commendable Lowell example, studies often lacked control groups).
In a test of the theory, Kees Keizer of the University of Groningen in the Netherlands asked whether cues of one type of norm violation made people prone to violating other norms.39 When bicycles were chained to a fence (despite a sign forbidding it), people were more likely to take a shortcut through a gap in the fence (despite a sign forbidding it); people littered more when walls were graffitied; people were more likely to steal a five-euro note when litter was strewn around. These were big effects, with doubling rates of crummy behaviors. A norm violation increasing the odds of that same norm being violated is a conscious process. But when the sound of fireworks makes someone more likely to litter, more unconscious processes are at work.
A Wonderfully Complicating Piece of the Story
We’ve now seen how sensory and interoceptive information influence the brain to produce a behavior within seconds to minutes. But as a complication, the brain can alter the sensitivity of those sensory modalities, making some stimuli more influential.
As an obvious one, dogs prick up their ears when they’re alert—the brain has stimulated ear muscles in a way that enables the ears to more easily detect sounds, which then influences the brain.40 During acute stress, all of our sensory systems become more sensitive. More selectively, if you’re hungry, you become more sensitive to the smell of food. How does something like this work? A priori, it seems as if all sensory roads lead to the brain. But the brain also sends neuronal projections to sensory organs. For example, low blood sugar might activate particular hypothalamic neurons. These, in turn, project to and stimulate receptor neurons in the nose that respond to food smells. The stimulation isn’t enough to give those receptor neurons action potentials, but it now takes fewer food odorant molecules to trigger one. Something along these lines explains how the brain alters the selective sensitivity of sensory systems.
This certainly applies to the behaviors that fill this book. Recall how eyes carry lots of information about emotional state. It turns out that the brain biases us toward preferentially looking at eyes. This was shown by Damasio, studying a patient with Urbach-Wiethe disease, which selectively destroys the amygdala. As expected, she was poor at accurately detecting fearful faces. But in addition, while control subjects spent about half their face-gazing time looking at eyes, she spent half that. When instructed to focus on the eyes, she improved at recognizing fearful expressions. Thus, not only does the amygdala detect fearful faces, but it also biases us toward obtaining information about fearful faces.41
Psychopaths are typically poor at recognizing fearful expressions (though they accurately recognize other types).42 They also look less at eyes than normal and improve at fear recognition when directed to focus on eyes. This makes sense, given the amygdaloid abnormalities in psychopaths noted in chapter 2.
Now an example foreshadowing chapter 9’s focus on culture. Show subjects a picture of an object embedded in a complex background. Within seconds, people from collectivist cultures (e.g., China) tend to look more at, and remember better, the surrounding “contextual” information, while people from individualistic cultures (e.g., the United States) do the same with the focal object. Instruct subjects to focus on the domain that their culture doesn’t gravitate toward, and there’s frontal cortical activation—this is a difficult perceptual task. Thus, culture literally shapes how and where you look at the world.*43
CONCLUSIONS
No brain operates in a vacuum, and over the course of seconds to minutes, the wealth of information streaming into the brain influences the likelihood of pro- or antisocial acts. As we’ve seen, pertinent information ranges from something as simple and unidimensional as shirt color to things as complex and subtle as cues about ideology. Moreover, the brain also constantly receives interoceptive information. And most important, much of these varied types of information is subliminal. Ultimately, the most important point of this chapter is that in the moments just before we decide upon some of our most consequential acts, we are less rational and autonomous decision makers than we like to think.
Four
Hours to Days Before
We now take the next step back in our chronology, considering events from hours to days before a behavior occurs. To do so, we enter the realm of hormones. What are the effects of hormones on the brain and sensory systems that filled the last two chapters? How do hormones influence our best and worst behaviors?
While this chapter examines various hormone
s, the most attention is paid to one inextricably tied to aggression, namely testosterone. And as the punch line, testosterone is far less relevant to aggression than usually assumed. At the other end of the spectrum, the chapter also considers a hormone with cult status for fostering warm, fuzzy prosociality, namely oxytocin. As we’ll see, it’s not quite as groovy as assumed.
Those who are unfamiliar with hormones and endocrinology, please see the primer in appendix 2.
TESTOSTERONE’S BUM RAP
Testosterone is secreted by the testes as the final step in the “hypothalamic/pituitary/testicular” axis; it has effects on cells throughout the body (including neurons, of course). And testosterone is everyone’s usual suspect when it comes to the hormonal causes of aggression.
Correlation and Causality
Why is it that throughout the animal kingdom, and in every human culture, males account for most aggression and violence? Well, what about testosterone and some related hormones (collectively called “androgens,” a term that, unless otherwise noted, I will use simplistically as synonymous with “testosterone”)? In nearly all species males have more circulating testosterone than do females (who secrete small amounts of androgens from the adrenal glands). Moreover, male aggression is most prevalent when testosterone levels are highest (adolescence, and during mating season in seasonal breeders).
Behave: The Biology of Humans at Our Best and Worst Page 10