At 9:40 on a Monday morning in New York City, four days before Christmas in 1982, a twenty-nine-year-old man named Reginald Andrews stood on a subway platform in Greenwich Village, waiting for an uptown train. He had been unemployed for over a year and had just come from an interview at a nearby meatpacking plant. He wasn’t especially optimistic about how it had gone. He calculated that he had applied to nearly one thousand jobs over the last year, but he was still jobless, and things were looking increasingly dire for him, his wife, and their eight children. The phone company had recently shut off his family’s service, and they were only getting by thanks to the generosity of people they knew in their community.
The train arrived and Andrews stepped toward the doors along with the rest of the passengers. As he did so, he noticed something alarming: an elderly blind man who was also moving to board the train mistook the space between two cars for the doorway to the car with his cane—and he fell down onto the tracks.
There was no time to analyze the situation before the train started moving, only seconds to act. Frantically calling out to other passengers to alert them of the situation, Andrews dropped down onto the tracks with the man.
As the crushing wheels of the train began groaning into motion, Andrews pulled seventy-five-year-old David Schnair, who was injured, into a crawl space tucked under the lip of the platform above. Had Andrews known it was there? What had he planned to do when he jumped down? It didn’t matter. They were just barely out of the way of the train when it stopped. One of the other riders, a woman, had managed to get the conductor to halt. Subway workers cut the power to the train, and soon after, the two men were lifted up to safety. Schnair had survived his near-death accident on the tracks, and Andrews had survived his near-death act of heroism. And thankfully, the down-and-out family man was rewarded for his heroism—big-time.
The day that Reginald Andrews performed his staggeringly brave feat, I was just a few blocks away from that subway station, in my NYU apartment working hard on a chapter for a book that had to be finished in a couple of weeks. The subway rescue was all over the local news that evening and then network news also picked up the story. The national coverage attracted the attention of no less than President Ronald Reagan, who mentioned Andrews the next day at his church when he was fielding inquiries from the media and then phoned Andrews. At first Andrews thought it was a prank, but as the call continued and he listened to the voice, he realized it was no joke. It really was the president of the United States on the line. Reagan congratulated Andrews for his heroic actions and wished him a merry Christmas. Later, the president called the plant where Andrews had interviewed that Monday morning. Reagan spoke to the supervisor there and suggested that he hire Andrews. Of course, the supervisor did just that.
Andrews’s gut decision had saved not only an innocent life, but his family’s economic security, too. I remember watching Reagan’s annual State of the Union address to Congress one month later and can vividly recall the moment when the president told his audience what Andrews had done and pointed to him up in the Capitol gallery, where Andrews sat as his invited guest, receiving a standing ovation from the senators, congressmen and congresswomen, and justices of the Supreme Court gathered there that evening.
Now fast-forward eighteen years, to May 11, 2010. It was a Thursday afternoon and Rose Mary Mankos stood on another Manhattan subway platform, on the Upper East Side, waiting for another train. The forty-eight-year-old lawyer from the Stuyvesant Town neighborhood a few miles south stood among the after-school crowd of students on their way home. She carried a black LeSportsac backpack, which she inadvertently let drop onto the tracks. What should she do? She jumped down to get the bag.
To many people, the distance from the floor of the tracks back up onto the platform looks easily scalable. But as transit authorities know all too well (and try to warn riders), it isn’t. Climbing up and out is difficult. This was the horrible predicament in which Mankos now found herself—she didn’t know how to get back up onto the platform—as she and the other people waiting in the station heard the ominous rumble of a train approaching the station.
Bystanders shouted for her to lie down between the tracks. They told her that the cars would pass over her, but she was too frightened to do so. The conductor of the oncoming train saw he was bearing down on a person in front of him and yanked the emergency brake while hammering hard on his horn, its earsplitting blast filling the station. It was no use. As the train slowed into the station, Mankos tried to squeeze herself against the platform, but she couldn’t get out of the way. She was killed.
Two people who leapt down onto subway tracks, two life-threatening crises, and two radically different outcomes—each defined by a gut decision made in a moment. For one, his gut decision made him a hero and changed his life for the better; for the other, it led to a horrible and premature end to her life. In hindsight, it is easy to see Andrews made the right decision and Mankos did not—it is always easy after the fact to identify the times when it was good to trust our gut and the times when it was not so good. But we need to know what the right thing to do is before we do it, not afterward. Entire books—bestsellers—have appeared in recent years that seem to give completely conflicting advice on this question: can we trust our intuitions (Blink, by Malcolm Gladwell), or not (Thinking, Fast and Slow, by Daniel Kahneman)? The answer lies in between. There are times when you can and should, and times when you can’t and shouldn’t. I will provide eight simple rules, based on the accumulated research evidence, for when you should and when you should not trust your gut.
In general, we tend to trust our intuition. In one study by decision researchers Carey Morewedge and Michael Norton and their colleagues, people reported that their intuitions and gut feelings—such as when they have a hunch, when their mind wanders onto a topic while reading, when ideas seem to just pop into their heads—revealed more about their true feelings, their real selves, than did their normal conscious thoughts, as when they are deliberately thinking about something, trying to solve a problem, or making a plan. Participants rated how spontaneously each of a variety of mental experiences occurred for them, and separately rated how much each type of experience revealed about their true beliefs and feelings. These two ratings were highly related—the more spontaneous and less intentional a mental experience, such as a dream or a Freudian slip, the more that people trusted it as a revealing insight about themselves.
Why do we trust our intuition, even more than our careful thinking? Basically, we trust our intuition for the same reason we trust our senses. Information coming into our minds easily and naturally, without our trying to figure it out or spending any effort on it, seems “true” and “out there in the world” just like when we look at that very large plant in our yard and know, immediately and without having to think about it at all, that it is a tree. I can look out my window across the lake to a low ridge, which the sunrise is lighting up against the pale blue sky, and imagine the StayPuft Marshmallow man stomping along on top of that ridge. But I can only produce a weak image of this with my imagination alone, and I know I am working hard to imagine it, so I know it is not real. If Mr. StayPuft was really out there on the ridge, the visual experience would be much stronger and clearer, and I would not have to try at all to produce it. How much I have to try to see the image (using my imagination) is a very powerful cue to whether the thing I am “seeing” is real or not. We tend to trust our intuitions for similar reasons: the more easily a particular thought appears in the mind, without our trying to produce it, the more we trust its validity, and the less we doubt it is true. We are wired to trust our senses, without questioning them; the alternative, to not trust our senses and to question them, is to be psychotic, and that is a very frightening state to be in.
What if the information we encounter out in the world is not coming into our senses so clearly and easily? What if it is getting dark, say, and we aren’t quite sure that’s our friend walking toward us, or our dog in the bushes over there, and we have to look harder and think about who or what that is? Then we are not so confident about what we think we are seeing—and this is when gut responses come in. We have to wager on what the right move is, and hope betting on ourselves will pay off.
So while we do tend to trust our intuitions, we also recognize that they can be wrong or misleading. When I began work on this chapter, I created a thread on Reddit, the social media and discussion hub, asking users about times their gut reactions turned out to be totally wrong. I noticed that their responses fell into two main categories: fear when in fact none was necessary, and overconfidence when just the opposite was needed. In the first category, a woman wrote about how, after first meeting her current romantic partner, she was convinced he was a “player.” She kept him at a distance, until she finally saw past her wary instincts and realized “he’s the sweetest and most faithful of men.” Other people wrote about times they thought someone was in danger (owing to a strange sound, or a sketchy-seeming dark street) and rushed to the rescue, only to discover a perfectly harmless situation. In the second category, of overconfidence, one man wrote that he always thought the girls he liked would come around and take notice of him, but they never did. Another Reddit user wrote about how he always thought he did well on the tests he ended up doing very poorly on. All the responses from my questions were fairly lighthearted, but they highlighted how it was not uncommon for our “blinks” to blind us.
Both of the subway jumpers, Andrews and Mankos, had to act under extreme time pressure. They had to act quickly or the blind man would have been killed or the backpack destroyed. They both took huge chances with their own lives. With the advantage of hindsight, because of the different outcomes, we know that Andrews made the right choice, and Mankos the wrong choice. But it could easily have been the other way around. Andrews and the blind man could have been killed if Andrews hadn’t had the time to tuck them into the crawl space; Mankos might have been helped up onto the platform by other passengers or the train might have stopped in time. But Andrews would still have been a hero for his selfless attempt to save another person by risking his own life, and Mankos a tragic risk-taker because her backpack was not worth the chance she took. Even if she had managed to escape to safety, she’d made a bad choice. The difference in their outcomes, life versus death, happens to dovetail with the difference in what they stood to gain. In one case, an innocent, helpless person’s life; in the other, a backpack. One is worth risking your life for; the other is not. But both Andrews and Mankos trusted their gut. What to make of this?
As with the tragic case of Rose Mary Mankos, our intuition can lead us astray if we are too quick to accept intuitive answers that would be proven wrong by just a moment’s reflection. Decision researcher Shane Frederick has developed a simple three-item quiz to measure a person’s tendency to make quick intuitive decisions without reflecting on them. For example: If it takes 5 widget-making machines 5 minutes to make 5 widgets, how many minutes would it take 100 widget-making machines to make 100 widgets? Many people quickly answer 100 because that so naturally follows the pattern of the example in the premise. It just feels right. The first one is 5, 5, 5 so the second one must be 100, 100, 100. But the correct answer is actually 5 minutes, which is the time for a single widget maker to produce a widget. No matter how many machines you have, each one will take 5 minutes, so 100 machines will make 100 in 5 minutes (one each). This reminds me of that old prank where a friend tells you to complete each sentence as fast as possible: “A funny story is a . . . ??” JOKE (you say). “To jab with your finger is a . . . ??” POKE. “A popular soft drink is a . . . ??” COKE. “The white of an egg is a . . . ??” YOLK. Gotcha!
Not questioning our gut can sometimes leave us with egg on our faces.
So right off the bat we have two basic rules for when to trust your gut. Rule #1 is to supplement your gut impulse with at least a little conscious reflection, if you have the time to do so. (Sometimes, as in Reginald Andrews’s case, we do not, but in Mankos’s case she did.) Conscious and unconscious thinking have different strengths, and different weaknesses, as we will see in a moment, and using both if you can is the best way to go. Check your work, if you can! Rule # 2 is that when you don’t have the time to think about it, don’t take big chances for small gains going on your gut alone. The blind man’s life was worth it to Andrews, God bless him. But the backpack just wasn’t worth it. Know the stakes. (Taking big risks for small rewards reminds me of all the tailgaters on the roads around where I live—right on my rear bumper at 50 mph or faster. Big risk, very small gain—I just don’t get it.)
Decision researchers generally do not like intuition and tend to portray conscious reflection as a white knight that rides to the rescue of our error-prone gut. But knights can make mistakes as well. Yes, we can underthink, but we can also overthink our choices, so that our conscious deliberations are the ones that lead us astray. Timothy Wilson and Jonathan Schooler discovered this by studying strawberry jam, college classes, and cat posters. (They did this one experiment at a time, not by focusing on all three things at once, which probably would’ve gotten messy.)
In their first study, they had participants judge the quality of different brands of jam, then compared their ratings with those of experts. They found that the participants who were asked to spend time consciously analyzing the jam had preferences that differed further from those of the experts, compared to those who responded with just the “gut” of their taste buds. In Wilson and Schooler’s second study, they interviewed hundreds of college students about the quality of a class. Once again, those who were asked to think for a moment about their decisions were further from the experts’ judgments than were those who just went with their initial feelings. And in their final study, participants got to choose a poster to take with them as a gift for being in the study. They could choose one of two types of posters: those of paintings such as Van Gogh’s irises or Monet’s water lilies, or silly posters with cartoons of cats. They either chose right away or were asked to first think about the reasons for their choice. In the “gut” reaction condition only 5 percent took the dumb cat poster, but in the “think first” condition 36 percent chose the dumb cat poster. Three weeks later the participants were contacted and asked how much they liked the poster on their wall. Those who had more spontaneously gone with their gut were happier with the gift they had chosen than were those who had decided after thinking about it first. The immediate snap judgment was a better predictor of future satisfaction than careful and patient consideration of the choice.
When this “strawberry jam” study was published in the early 1990s, Shelly Chaiken and I were in the middle of the automatic attitude research I described in Chapter 5. We found Wilson and Schooler’s findings to be very much in synch with our own conclusions. The greater the involvement of conscious and intentional evaluation processes in our studies, the harder it was to detect the unconscious attitude effect, and the weaker those effects were. It was as if the conscious evaluation processes were interfering with the more natural unconscious appraisals of the objects in our studies. So too with the strawberry jam studies—the more that people thought about their feelings about the jam, the less their stated opinions reflected their true underlying attitudes.
The different strengths and weaknesses of conscious and unconscious decision-making were revealed in a groundbreaking series of studies by Dutch researchers Ap Dijksterhuis and Loran Nordgren and their colleagues when they tested their Unconscious Thought Theory (UTT). Dijksterhuis and Nordgren were the first to extend the study of unconscious mental processes to the domain of judgment and decision-making, one of the last bastions of psychological science to accept a role for the unconscious. Psychological science had long presumed that judgments and decisions were almost exclusively conscious activities. Of course, there have been many studies over the past half century, most famously by Daniel Kahneman and Amos Tversky, showing the irrational or heuristic shortcuts that people take when making conscious decisions, but in those studies the actual making of the judgment or decision was always a conscious, deliberate process. Dijksterhuis and Nordgren’s UTT research showed that the judgments themselves could be made unconsciously, during a period of time when the conscious mind was distracted by doing something else entirely. Not only that, but also, more provocatively, they concluded that the results of the unconscious decision process were often superior to those of the consciously made judgments.
How did they test this? First, they gave participants the information needed to make a judgment such as which car was better to buy or which apartment was better to rent, out of four alternatives. They varied each of the four choices on relevant dimensions (gas mileage, price, reliability, luxury). So one car model might have the best gas mileage but a higher price and require a moderate amount of service at the garage; another might have poor mpg but need hardly any service; and so on. These four alternatives were deliberately constructed so that there was an actual objective right answer to the question of which car was the best one to buy, taking all four features into account. The same with the apartment choices: one might have the lowest rent but not the best location; another might have more space but not the best view; etc.
After the participants read all of this information about the cars or the apartments, some of them were then asked to think about which car or apartment was the best one, and others were prevented from thinking about the cars or apartments (consciously, that is) for the same period of time. Instead of being able to think about the cars or apartments, they had to do a difficult mental task that took up all of their available attention. (Imagine, for example, counting backward from 643 by sevens as fast as you can.) After this task was completed, the participants gave their decisions about the best apartment or car. Lo and behold, more of the participants in the unconscious thought condition made the best choice than did those in the conscious thought condition. The researchers then replicated this effect across many similar studies. While this was a quite surprising finding when it first appeared, it did bear out what Freud had written more than one hundred years earlier in The Interpretation of Dreams: “The most complicated achievements of thought are possible without the assistance of consciousness.”
How did the participants in the unconscious judgment condition make the best choice? Again, neuroscience research has helped explain just what was happening to the unconscious deciders during the distraction period. When David Creswell and his fellow neuroscientists at Carnegie Mellon University imaged the brains of participants during the experiment, both when they were reading about the various cars or apartments, and then later during the “unconscious thinking” (gut) time, they found that the area of the brain that had been active while the participants were consciously learning all the features of the cars or apartments remained active during the time when they were distracted by the task (and were thinking unconsciously). Furthermore, the more active that same area of the brain during the later unconscious thought period, the better the quality of the decision the participant made. In other words, the same part of the brain that was first used to acquire the important information was then used by the unconscious “gut” processes in solving the problem, while the conscious mind was elsewhere.
Dijksterhuis, Nordgren, and their colleagues continued to research the conditions under which unconscious decisions were as good or better than consciously made decisions, and the times when conscious decisions were superior. Their conclusions are very relevant to our question of when can we trust our gut, and when we can’t. Unconscious decisions tend to be better when the judgment is complex and many different dimensions or features have to be combined and integrated, as with the cars and apartments. Our conscious working memory is limited and can’t hold as much information at any given time; we can handle up to three things at once comfortably, but more than that becomes a stretch. Because our conscious mind can focus on only a few features, as in the strawberry jam or cat poster studies, other relevant features are not taken into account, and don’t have the influence they should. Conscious thought is powerful, but it is limited as to the complexity of what it can consider at any given moment. Still, conscious processes are better than unconscious ones if there is a rule to follow. For example, if you have to constrain your apartment or car choices because of your budget, and must exclude those that are too expensive, or if you have to walk to work and thus can’t live more than a mile away from your job, then conscious judgments will be better at taking those constraints into account. A natural question here is, Can these different modes of thought work together?
In their most recent studies, the UTT researchers have shown how the best decisions are made via a combination of conscious and unconscious processes, and in this order: conscious first, then unconscious. For example, you should first consciously eliminate any options that fail to meet the necessary criteria, such as too expensive or too small, too far away, and so on. Only then should you give the unconscious judgment process the options that pass the first test by doing something else and not thinking (consciously) about the choice for a while, and then seeing how you feel about it later.
Our ability to solve complex problems unconsciously, without the aid of conscious thought, makes sense from an evolutionary standpoint, given that we developed our conscious thinking abilities late in our human history. With this in mind, it would make sense if unconscious thought mechanisms worked better for those types of problems we were more likely to encounter “back in the day” of our ancient past, like judging the fair treatment of others or detecting who was harming others in a group. The ability to make such distinctions was key to harmonious social life and group solidarity. Researchers Jaap Ham and Kees van den Bos and their colleagues have applied the UTT idea to such problems as they might appear in modern life, for example, in judging guilt or innocence in a complex legal case, and judging the fairness of a company’s hiring procedures.
We are born sensitive to violations of fair treatment and with the ability to detect those who are causing harm to others, and assign blame and responsibility to them. Recent research has shown that even children three to five years old are quite sensitive to fairness in social exchanges. They preferred to throw an extra prize (an eraser) away than to give more to one child than another—even when that extra prize could have gone to themselves. Of course, the two concerns, of guilt and fairness, are related. Witness the huge public and media scrutiny paid a few years ago to the seemingly trivial matter of whether Tom Brady, the New England Patriots’ quarterback, was involved in the slight deflation of the footballs in a conference championship game. As far as world or national problems are concerned, this was a very insignificant matter, yet it consumed the American public’s attention and dominated the news cycle for weeks and even months afterward. We are still very much like our long-forgotten toddler selves who shout “Cheater!” when we see injustice in a game.
Ham and Van den Bos used the standard UTT study procedure, with an immediate three-minute conscious thought period, and a three-minute distraction (unconscious thought) period condition to see if we solve these problems unconsciously. Participants made fairness judgments about complex job application procedures. There were four application procedures described; one was objectively the most fair and another the most unfair, with the other two in between. In the fairest procedure, for example, the application process was clearly explained and all of the requested information on the application was read and considered in the hiring decision. In the least fair procedure, the application process was not very clearly described and only one of the four tests given applicants was used in the hiring decision. Participants were divided into the immediate, conscious, and unconscious decision conditions, and once again, those in the unconscious decision condition were the best at saying which of the procedures was the most fair.
In another study, participants were given the many details of an actual legal case in the Netherlands, a complicated affair in which an underage girl took a horse and carriage for a drive without the permission of their owner or her parents. Entirely coincidentally, a neighbor chose that exact moment to set off explosives to scare birds away from their crops. The explosion caused the horse to bolt, and this caused the animal to get hurt and the carriage to be damaged. It was a tricky case because there were many contributing factors, and several parties at fault. A binding arbitrator ruled that each of the four parties involved (the neighbor, the girl, her parents, the owner of the horse and carriage) was responsible in varying degrees for the damage that occurred, but the study participants didn’t know this. They had to assign guilt and responsibility on their own.
After all the evidence was presented, one group of participants made their judgments immediately, another was allowed to think about their justice judgments for three minutes before giving them, and another performed a distractor task for three minutes and then gave their judgments. How closely a participant’s judgments matched the arbitrator’s actual judgment was the measure of judgment accuracy, and the important question was what degree of responsibility each party had. Again it was the participants in the unconscious decision-making condition who made the most accurate judgment about the legal case. This finding has clear practical importance, since jurors (in the United States, at least) are not allowed to take notes or have any technological assistance when making a judgment. Often court cases are complex; many different pieces of information must be taken into consideration, different pieces of evidence point in different directions, and there can be extenuating circumstances to take into account as well. Unconscious decision processes are better at combining and integrating all this complexity.
However, for complex financial decisions or any decision to which actual quantified data is relevant, it is obviously better to use computers and the relevant data than to rely on a period of attention-distracted, unconscious thought. The bestseller Moneyball, by Michael Lewis, showed how better decisions about drafting and trading professional baseball players could be made by relying less on the intuition of scouts and more on quantifiable aspects of the game, such as an outfielder’s speed in tracking down a fly ball.
I’m a big baseball fan and have played what used to be called rotisserie baseball for nearly twenty years now. It was the original “fantasy” sports game, in which players assume the role of a professional team’s general manager and draft a starting lineup for a particular sport. Today millions of people play these games in a daily format. I play the season-long version. It is highly competitive, and one of the most important parts of the season comes well before the actual baseball season starts—the day that you and your competitors draft your teams. Preparation is the key. Starting in January, we pore over the many published guides with facts and figures on all the major-league players.
Technological advances have greatly increased the kinds of objective information available about all the players and removed much of the “gut” or intuitive aspects from the game—the fantasy game as well as the real-life general manager’s job. Major League Baseball has installed radar and other sensitive devices in its stadiums so that companies such as StatCast can measure things like how hard a ball is hit—its “exit velocity” in miles per hour leaving the bat. Pitchers’ curveballs can be measured in terms of number of spins per second. And although the fantasy leagues typically don’t use it, there is plenty of data gathered on defensive plays, such as the speed of an outfielder in tracking a fly ball, and the efficiency of the route he took to catch it. In addition to this new data about players’ performance, there are new ways of looking at the traditional kinds of data collected, such as fly ball rate, hard-contact rate, and percentage of hits after the ball is put in play (that is, not striking out), in order to develop more accurate indexes of a player’s abilities in isolation from his teammates’ performance, and also in isolation from the luck factor. (For example, a higher than average batting average on balls put into play on the field is usually just good luck, and can be expected to regress to the league average over time, predicting a lower batting average for that player in the near future.)
In the old days, before all of this technology and sophisticated data analysis, baseball teams relied on scouts—usually older scouts with much experience and a good “eye” for players. Successful scouts relied on cues that over the years had predicted major-league success. Scouting was an art form in many ways, because the scouts were relying in large part on their intuition, their fast and uncanny ability to appraise talent based often on little things that the untrained eye would never notice. The sound of the bat hitting the ball was often mentioned by scouts—a certain kind of crack! that signaled solid contact. Or, for pitchers, the pop! of the ball when it hits the catcher’s mitt.
The scouts weren’t “guessing,” however. They would not have been successful for so many years if they were. They were able to pick up the important cues and put them together well. Their success at scouting was evident in their ability to predict which young players would become stars and which would not—the successful scouts had a better track record than the less successful ones. But the intuitive nature of their personal appraisal process made it difficult for them to justify or explain their gut reactions to younger, less experienced personnel. It was their years of experience, years of close observation, that gave them their expertise. In part, they were taking advantage of what modern cognitive science calls “statistical learning”—our ability, after a good deal of experience, to detect regularities in the world, to pick up reliable patterns and sequences to determine what predicts what—without necessarily being able to explain or even being aware what those predictors or patterns are. This comes naturally from close observation—keeping both the eyes and mind open—over extended periods of time.
When I am deciding which car to buy, I go to Consumer Reports and other relevant websites to gather reliable information about miles per gallon, how much service a car needs, and features necessary for where we live, such as good handling under icy road conditions and high clearance because of all the snow in the winter. But not all of life’s important choices come with reliably measured data on the critical factors. In most aspects of our daily lives, we hardly have reliable data to use to make the best choices and decisions. Take Joe, for example. He is single and moving to a new town and wants to start dating—there are no Consumer Reports articles for that. Or for which career he is most suited and would find most fulfilling, whether he should live downtown or in the suburbs, or which of several suits or pairs of shoes he should buy. He may be able to get some objective information to help him make his decisions, but hardly a complete set of reliable data on each relevant feature or dimension. Few of our real-life choices come with the accurately measured, objective evidence, and tried-and-true predictive algorithms that investment bankers, and now baseball general managers use when deciding which stock to buy or which young baseball prospect to draft. (And even in those cases, prediction is far from perfect.)
The Unconscious Thought Theory research supports the fundamental point that evolution shaped our minds so that unconscious judgment processes produced reasonably accurate guidance for our behavior, especially for the millions of years before we had computers, algorithms, and spreadsheets (and baseball). So we come to Rule #3: When you are faced with a complex decision involving many factors, and especially when you don’t have objective measurements (reliable data) of those important factors, take your gut feelings seriously. See how you feel after a period of distraction doing something very attention-demanding to get your (conscious) mind off the decision. Or sleep on it, because the unconscious never sleeps, as we’ll see in Chapter 9.
There is one more important factor influencing our immediate gut reactions—our current goals and motivations—and that’s the topic of Chapter 8. How we feel about the people we know, and the basics of our life such as food, cigarettes, and alcohol, can change dramatically depending on whether they help or hinder what we are trying to accomplish. There are studies that show, for example, that we tend to form new friendships with those who will help us reach our personal goals, and that we are less likely to become friends with people who are very similar but who would not help us reach those goals. Who we list as our best friends in our lives changes depending on what our current goal is. Smokers who want to stop smoking but haven’t had a cigarette in many hours show implicit or automatic positive evaluations of cigarette-related items, such as an ashtray, but negative unconscious evaluations of those same items if they’ve just smoked and no longer have that need or craving.
Goals change the gut. Goals have a tremendous influence on our spontaneous appraisal of anything that is relevant to their pursuit; we have positive feelings toward something that helps us attain a goal and negative feelings toward things that do not. Like Seinfeld said, we are cheering for the clothes. A hated “dirty player” traded to our own team suddenly becomes a “wily veteran” who will do anything to help his team win. What was negative is now spun in a positive way—the gut feelings drive how the same behavior, the identical information, is spun. When we really crave a cigarette our gut says they are good (so good!); yet after we’ve had one and regret doing so, our gut reaction is that they are so bad (evil even). Your current goal changes your gut, and very often you aren’t aware of the reason for those strong immediate reactions. This brings us to Rule #4 for when you can trust your gut: Be careful what you wish for, because your current goals and needs will color what you want and like in the present.
So far we have focused on our gut feelings about important choices and decisions we’ve been mulling over. But what about our initial, immediate gut reactions to entities we encounter, especially to people we meet? Can we trust these gut instincts?
Gut instincts are something all of us experience—without understanding how they function. In the 1980s, scientists finally began looking closely at the mechanics of intuition, and two decades later, pop culture followed, most notably in the form of Malcolm Gladwell’s Blink: The Power of Thinking Without Thinking. The underlying premise of his book is that our first thought is usually our best thought, or that “blink” reactions—which don’t require conscious reflection—are more reliable and useful than ones that arise from self-questioning and rumination. As we’ll see, this is true, but only up to a point. Gladwell closed his book with a case where going with your gut was the wrong thing to do, the tragic story of Amadou Diallo, a victim of racial profiling who was shot dead by police in the South Bronx in a hail of bullets while entering his own apartment building unarmed. He was holding up his wallet to show police he had ID to prove he lived there, and in the dark night the police said they mistook the wallet in his hand for a gun. Diallo was black. Would the cops have mistaken the wallet for a gun if he had been white? For that matter, would they have even thought he was breaking into the apartment building in the first place? Such were the questions immediately raised in the public outcry that ensued.
One of the most important reasons the unconscious evolved was for “appraisal,” in particular to evaluate other people. As we saw in the previous chapter, we evolved to make snap judgments and educated hunches about people and situations, to determine whether we should stay or go. Sometimes these instantaneous appraisals go very right, and sometimes they can go very wrong. When deciding whether we should trust someone or not, one important thing to keep in mind is that our modern world is quite different from the one in which our unconscious appraisal apparatus developed. Just as with unconscious decision-making, the more a current situation resembles the conditions we faced in our ancestral world, the more accurate a guide our gut will be. But if the situation differs—and there are indeed some very notable differences—our gut is more likely to lead us astray.
We have already seen how we quickly appraise others in terms of “us” versus “them.” Even babies and young children have automatic, unconscious preferences for their own group and negative feelings for people in other social groups. In another one of our studies together, Mark Chen and I showed that subliminal presentation of smiling, attractive black faces (of models taken from popular magazines, but who were not famous) in the first part of the experiment caused greater hostility in white participants in the second part of the experiment—subliminal presentation of white faces did not. (Indeed, the officers involved in the Diallo episode certainly responded to an unarmed black man with a lot of hostility.) As we’ve seen, our hardwired us-versus-them tendencies can even lead to attempted murder among rival baseball fans. While this kind of tribe-versus-tribe gut reaction was very helpful back in Ötzti’s time, it is much less helpful today, in a world where people of various races and cultures mingle together in the same town or city. Unfortunately, it will be a long time before our innate wiring catches up with these seismic cultural shifts.
So this gives us another answer to the question of when can we trust our gut, which we will call Rule #5: When our initial gut reaction to a person of a different race or ethnic group is negative, we should stifle it. Our common negative initial gut reactions to people who are different from ourselves—and this can apply to religion or language as well as race and ethnicity—should not be trusted. These reactions are either a vestige of our evolutionary past, of Ötzti’s time and earlier, or a product of our culture through very early socialization and the mass media, as we’ve already seen. Especially when it comes to people who are clearly different from us, we need to give people a chance, look beyond their superficial aspects, and base our appraisal of them on their actual behavior instead.
The experiment I did with Mark Chen with the subliminal black faces also points to a sixth answer to the trust-your-gut question, which we will call Rule #6: We should not trust our appraisals of others based on their faces alone, or on photographs, before we’ve had any interaction with them. This is for two reasons. First, the appraisal we make regarding static faces alone, as in photographs, is not diagnostic; it is not a valid predictor of that person’s actual personality or behavior. Second, our unconscious reactions to people after we have had some experience with them, have seen them in action for even a little while, is a surprisingly valid predictor. As Mel Gibson’s character in Braveheart, William Wallace, says to his troops facing an oncoming cavalry charge: “Wait for it . . . wait for it . . .” And it turns out that, like Mel’s troops, we don’t have to wait very long at all.
As we saw in Chapter 5, we clearly perceive several different and basic personality features directly from a person’s face, in a photograph. This also happens when we first see someone in real life before even meeting her or seeing her interact with others. We read traits like competence and trustworthiness off a person’s face with such confidence that a candidate’s appearance in photographs can help determine the outcome of a political election. Even worse, studies in the courtroom have found that features of defendants’ faces determine the likelihood of their being found guilty and the length of the jail sentences they receive. Recall that “baby-faced” adults are more likely to be found innocent, and defendants with more racially prototypic faces to receive harsher sentences. But we humans did not evolve to be able to read personality from static photographs, or from facial appearances alone. Rather, we evolved to be quite sensitive to a person’s emotional expression—whether she looks sad, or disgusted, or panicky, for example—when she is in action, interacting with us or others. As Darwin was the first to point out, emotional expression is a more or less genuine marker of a person’s internal emotional state that predicts, in turn, her likely behavior toward us. We can trust these expressions to give us accurate gut readings of a person’s actual current state. However, the problem comes when we mistake a person’s resting facial expression as an indicator of one of these short-term emotions.
You may have encountered an Internet video that went viral, called “Bitchy Resting Face,” a parody of those ubiquitous pharmaceutical ads. The premise of the sketch was that there are women who aren’t able to smile well and are perceived as bitchy or hostile. As cheesy, emotional music plays in the background, people who know women with Bitchy Resting Face tell how they’ve been hurt by women with this unpleasant facial expression. A man asks a woman to marry him, only to interpret her unintentional scowl as a rejection. A female customer insults a friendly clerk by sourly saying thanks. It’s a hilarious concept, but also quite insightful. As one of the actors in the video says of women who suffer from Bitchy Resting Face, “They might not be bitchy at all.” As we saw with Old Man Marley in Home Alone or my daughter’s grade school librarian, appearances and first impressions based on faces alone can be, and are, deceiving.
Alexander Todorov, the Princeton scientist who showed how influential the faces of political candidates were to the outcome of elections, and his colleague Christopher Olivola, of University College London, looked into how accurate our fast personality assessments of a person are, based on their faces alone. Olivola and Todorov made use of an online website called “What’s My Image?” on which people can post their own photographs and have other people rate their personalities, not knowing anything else about them. They were able to obtain data from more than one million appearance-based judgments—made by about nine hundred different people—as participants guessed about things such as: the pictured person’s sexual orientation, whether they use drugs, if their parents are divorced, if they have ever been arrested or gotten into a fistfight, if they drink, if they’re a virgin. The researchers could calculate how accurate these judgments were because the person posting the photograph provided the answers to these questions. And what they found was that the participants who saw the photographs actually produced less accurate predictions than did a separate group of participants who never saw the photographs, and instead just relied on how common or expected these behaviors were in general. If you had just guessed about a given person based only on how generally common it is to be heterosexual versus homosexual, a drug user, and so on, you would have done better than the one million responders who also had the photographs to go by. So from our eyes to our guts, things can go very wrong.
In a second study, more than 1,000 participants, recruited through a link on the Scientific American website, played a “Political Guessing Game,” in which they guessed the political affiliation (Republican or Democrat) of each of nearly eight hundred political candidates, male and female, who ran in the 2002 and 2004 congressional elections—based just on photographs of the candidates’ faces. The researchers varied the proportion of Democrats that a given participant saw, and told some of the participants beforehand what that proportion was, but again, seeing the photographs caused the participants to be less accurate overall than if they had just gone with the base rate proportion. When told that 30 percent of the photographs would be of Democrats, 3 out of every 10 photographs, the respondents thought they knew which were the Democrats better than they actually did, and going with their gut reaction to the face photographs decreased, rather than improved, their predictions.
But our intuition’s dismal performance based on photographs is dramatically transformed, like a kissed frog into a prince, when it is based on the kind of information our evolutionary apparatus did have access to—a new person’s actual behavior.
In 1992, Nalini Ambady and Robert Rosenthal coined an apt term to describe the brief input the unconscious uses to generate instinctual responses: thin slices. They were studying how accurately people could assess the abilities and personalities of others, based on just a fraction of that person’s total behavior. For instance, you could sit in a classroom all day, all year, and then give your evaluation of the teacher’s ability and performance. (That would be the whole ham.) Or you could sample an hour or so each of the five days of a given week. (That would be a nice fat slice of the ham.) Or you could push the edge of the envelope really far, as Ambady and Rosenthal did, and give people just a thirty-second video clip of the teacher in the classroom—nothing more. (That would be a thin slice of the ham, the kind you get piled high on a good deli sandwich.) Ambady and Rosenthal compared the assessments that people made based on that thin slice of thirty seconds to what experts said about that teacher based on hours of observation. And over many different studies of various occupations—teachers, therapists, CEOs—and abilities, they found that we are actually quite accurate, even with these thin slices, at assessing abilities and personalities, producing judgments not all that different from those of experts who have much more evidence at their disposal.
In one of their studies, Ambady and Rosenthal videotaped thirteen Harvard graduate fellows teaching classes, and spliced together three ten-second clips of each into a thirty-second reel that was a sort of sampler of their pedagogical prowess. Next, a group of participants watched the videos and rated how good the graduate fellows were as teachers according to thirteen categories. Then Ambady and Rosenthal waited until the end of the semester, when students in the fellows’ classes filled out their usual end-of-course evaluations, and compared these “whole ham” ratings with the thin-slice ones from the experiment. Remarkably, they were highly correlated, showing a high degree of agreement between the thin-slice and whole-ham assessments. Ambady didn’t stop there, though. She went on to make the slice even thinner—now we are talking 2nd Ave. or Katz’s Deli thin—winnowing the video sampler down until it was just six fleeting seconds. Yet the participants seeing just this very short snippet of classroom performance still were able to accurately predict who the best teachers were in those semester-long classes. Ambady went on to produce other studies on thin slicing and found that humans can accurately assess other traits, such as sexual orientation, whether a CEO led a successful company, or if someone had a personality disorder. We are very lucky to have such an unconscious appraisal ability, and I have benefited firsthand from it.
After Christmas of 2012, I was in a McDonald’s south of Indianapolis, just off the interstate, with my daughter Danielle. We were driving back east from a trip visiting family in Illinois, and I was a single parent at the time. It was getting to be lunchtime, so we pulled off to get gas and some lunch. Danielle, being six at the time, wanted a Happy Meal, so we went into the McDonald’s next to the gas station. She was contentedly eating while playing with the coveted prize that she had found in the box with her hamburger and fries, when a younger child a few tables over from us started to cry, quite loudly, attracting the attention of most of us in that part of the restaurant. Danielle stopped playing and looked over at the child, too. Then she did something that I’ll never forget. She picked up her toy, walked over to the crying child, and handed her the toy. The child looked at Danielle, took the proffered plastic toy, and immediately calmed down. It seemed as if the whole restaurant was watching this little scene unfold, and you can imagine how proud I was of my daughter. She came back over to our table, and all the other patrons went back to eating, except for one person.
That person came over to our table, apologized for interrupting us, and said she wanted to tell Danielle that what she had done to make the other child feel better was such a nice and generous thing to do. She was not speaking to me or looking at me, but from that “thin slice” of her behavior I felt I knew a great deal about her. Her kind words made both my daughter and me smile and we thanked her for coming over; after her order was ready at the counter she came back to sit down with us at our table. It turned out she was on her lunch break from a nearby hospital. We kept in touch until we were able to meet again during the summer, when my daughter and I made another driving trip to the Midwest, and the rest, as they say, is history—we were married a few years later. Now when all of us make the drive back to the Midwest we often pass that same McDonald’s, which brings back memories of the day we met.
Not all first encounters with people are so positive, of course. How does our gut react to someone whose behavior shows we can’t trust him? The answer: just like Dante thought it would. Yoona Kang, Jeremy Gray, Margaret Clark, and I conducted a neuroimaging study of the brain’s immediate reactions to betrayal, back in 2011. In that study, the same insula brain region that became activated when the participant held something physically cold also became active when another participant betrayed them in an economics game, by greedily keeping all the money for themselves. That is a “cold” reaction based on actual experience with the person, and of course, because it is based on actual evidence about that person’s trustworthiness, it should be trusted. And what is more, our brains also turn off the circuitry needed to produce mimicry responses (which signal bonding and friendship to the other person) when we encounter people who have shown by their behavior that they can’t be trusted. In a study by Oriana Aragon, Michael Pineda, and myself, we measured the brain waves of participants while they played economics games with each other, and also while they watched each other’s finger movements.Watching the other person’s finger movements before the economics game caused the participants to immediately produce the brain waves associated with the start of the natural imitation process. However, after playing the economics game, if that other person had betrayed the participant—by keeping all the money for herself and not sharing any of it—then watching that other person’s finger movements no longer produced those immediate brain waves associated with imitation (and bonding and friendship). Rule #7 (it may be the most important one of all): You can trust your gut about other people—but only after you have seen them in action.
Our instincts about other people evolved in quite different times, of course, eons before the advent of social media, so what about meeting people over the Internet? Social life on the Internet is like the Wild West of American history, uncharted, somewhat lawless, often dangerous, and constantly changing. Can we trust our gut when it comes to people we meet online? Can we know who someone really is before we meet him in person?
“Can you see the real me? Can you? Can you?” sings Roger Daltrey in one of my favorite songs by the Who, written long before the Internet was invented, much less something you carried in your pocket. We’ve always packaged ourselves for public consumption, putting our best foot forward and hiding or camouflaging our faults as best we can. And we do so today in spades. Anyone who has ever been on Facebook or Instagram or any other social media knows that people spend a great deal of time carefully presenting upgraded versions of themselves that project images of lives that appear more perfect than they are. Sometimes these public personas are outright fictions, as in the practice of “catfishing.” Confiding our “real me,” who we really are inside, to another person takes a lot of trust, because doing so makes us quite vulnerable, especially if some of those parts of the “real me” are looked down upon by society or those around us.
Back in the Stone Age of the Internet—meaning the 1990s—researchers in human communications and social psychologists started to study how this new electronic way of interacting with others was affecting social life. Katelyn McKenna was one of the pioneers. In one groundbreaking series of studies she went undercover, taking the role of “participant observer,” gaining acceptance as a member of several electronic discussion boards, called newsgroups, covering a range of different topics. Back then it was not hard to participate in these anonymously, which enabled many people to join and participate in groups formed around topics that are called “stigmatized” interests. These could be political, like white supremacist groups, or sexual, such as cross-dressing or transvestism. But discussion groups also formed around more mundane specialized interests, like butterfly collecting, or Humphrey Bogart movies. People flocked to these forums because for many it was the first time they had found people who shared their interests. And especially for the stigmatized, socially frowned-upon interests—sexual proclivities such as cross-dressing or sadomasochism, political ones such as antigovernment militias or white supremacist groups—many of these people had spent their entire lives hiding that interest from not only their neighbors but in many cases also from their close friends and family, or even their spouse.
By infiltrating and participating in these groups, McKenna was able to gain the trust of the members. Then and only then, after many months and even years of participating, was she able to gather information from the members about how long they had been part of the group, how self-accepting they were of this interest—were they ashamed, okay with it, proud—and whether they had told their loved ones about it. She also kept track of whether the individuals actively participated in the group through posting and taking part in the electronic discussions, or “lurked” instead, just reading the other posts but not saying anything themselves.
What McKenna found was remarkable. In many cases, these newsgroup participants had been ashamed of or wanted desperately to keep their interest or behavior hidden. These were mainly older people—in their thirties, forties, or fifties—who had kept this part of themselves secret from others their entire lives. Many said that before they found that newsgroup, they had thought they were the only ones who had the interest. The truly remarkable effect of finding others with whom they could share their “real me” was that they came to no longer feel ashamed or bad about it. That first step of self-acceptance then, in many cases, led directly to their telling their close friends and family about it for the first time. Self-acceptance had to come first, but once that had happened, many people really wanted to come out and make that previously private part of themselves public. In some cases they did so after a lifetime, thirty or forty years, of keeping it a close secret.
I bring up McKenna’s research to stress that being able to connect to the entire world of people enables us to find and interact with others who share very important aspects of ourselves, which we often would be unable to do in face-to-face, nondigital settings. Over social media we can develop relationships with people we might not have given a moment’s notice to in real life. Over social media, those people can get by the “gating features,” as we called them, such as attraction or the chronic features of their face, that we use to screen people in real-life, face-to-face encounters. These initial filters allow certain people past the gates but they block many others. Many potentially great romantic relationships don’t get off the ground because of the importance we place on those gating features, mainly a person’s physical attractiveness or general appearance. We should all keep in mind Nietzsche’s advice to marry someone you can have conversations with, for most of your life together will be after the blush is off the rose.
Because many forms of social media (not all) enable us to bypass those gating features, then people who do not meet face-to-face, but instead through social media such as Internet discussion groups, email, blog sites, or chat rooms, might actually have just as stable and long-lasting relationships as those who meet in “real life.” Back in the 1990s, there was a pungent stigma attached to meeting on the Internet and the common wisdom was that few of these relationships would survive a couple’s first face-to-face encounter. But since then there has been a veritable explosion in online dating, and a recent national survey of nearly twenty thousand people who married between 2005 and 2012 found that fully 35 percent had first met online. About half of these people had met through online dating sites such as eHarmony and Match, the rest through their social networks (Facebook, Twitter), multiplayer game sites, chat rooms, or other online communities.
Social psychologist John Cacioppo and his colleagues, who collected and analyzed this data, reported that the couples who met online were, if anything, no more likely to have broken up than couples who had met in more traditional ways, and that they were just as satisfied in their marriages as well. Today, of course, unlike the Internet of the 1990s, you can see photos of the other person, as well as be “matched” on common interests (either by the dating platform or because you can read the content of their posts or are members of the same special interest group), so meeting online now has more gating features present than it used to. (Tinder, for example, is even more about initial attraction than are real-life encounters, involving quick yes-or-no, stay-or-go decisions based on photographs alone.) Still, meeting (and especially, getting to know) someone online can often afford you more important background information (values, political attitudes, and interests, for example) about a person than the traditional happenstance initial face-to-face meeting. And the emerging data on the quality and stability of Internet relationships is consistently disproving the initial (and somewhat snarky) skepticism of the 1990s regarding their likelihood of long-term success.
Don’t get me wrong: attractiveness is important. It is a real feature of the person. As we’ve seen, attractive faces are a literal pleasure to look at; our brain’s reward centers become active when we look at them. And, as we’ve seen, even babies prefer to look at attractive faces! It is just human nature to prefer attractive to unattractive people when it comes to close relationships. The problem comes when we use that attractiveness to make inaccurate assumptions about other qualities of the person. We tend to believe that what is beautiful is good and assume other good things, such as pleasant personality, competence, trustworthiness, when we see an attractive face. We have way too much confidence in these gut reactions based just on appearances. So that gives us Rule #8: It is perfectly fine for attraction be one part of the romantic equation, but not so fine to let it be the only, or even the main, thing. Not in the long run, anyway.
Our gut reactions served us well for many thousands, perhaps millions of years. If they had been misleading or counterproductive they would have been weeded out by natural selection. But our modern life is very different from what life was like over those thousands and millions of years. People of different races, who are different from our family and neighbors, are no longer enemies who can’t be trusted. Modern technologies such as photographs of faces can fool our gut appraisal mechanisms that were developed instead for observing people in action, in the context of how they treat us and the others around us. Our gut reactions can be quite sophisticated at combining lots of information, and should be taken seriously, but here too we need to adjust for the conditions of modern life and make use of reliable data, if we have it, and the powerful ways of analyzing it now available, especially for important choices and decisions.
Today even experts disagree on whether intuitions are accurate, and whether we can trust our gut. Those who say we can’t tend to study complex financial and business decisions, made with little or no time pressure, and based on reliable data, with powerful computers and software to analyze it. Those who say we can trust our instincts tend to be psychologists or evolutionary scientists, who study the mundane realities of daily life, where often we are under time pressure to make decisions and lack any relevant quantitative measures. So then, certainly, listen to what your gut, or heart, or other internal organ (including your brain) is telling you, take it seriously and don’t dismiss it out of hand, but also check your work, and always remember to give the other person a chance.