Think Again
by Adam Grant
The smokejumpers landed near the top of Mann Gulch late on a scorching August afternoon in 1949
— Same anecdote as Keith
With all due respect to the lessons of experience, I prefer the rigor of evidence. When a trio of psychologists conducted a comprehensive review of thirty-three studies, they found that in every one, the majority of answer revisions were from wrong to right.5 This phenomenon is known as the first-instinct fallacy.
We laugh at people who still use Windows 95, yet we still cling to opinions that we formed in 1995. We listen to views that make us feel good, instead of ideas that make us think hard. At some point, you’ve probably heard that if you drop a frog in a pot of scalding hot water, it will immediately leap out. But if you drop the frog in lukewarm water and gradually raise the temperature, the frog will die. It lacks the ability to rethink the situation, and doesn’t realize the threat until it’s too late. I did some research on this popular story recently and discovered a wrinkle: it isn’t true.
Tossed into the scalding pot, the frog will get burned badly and may or may not escape. The frog is actually better off in the slow-boiling pot:12 it will leap out as soon as the water starts to get uncomfortably warm. It’s not the frogs who fail to reevaluate. It’s us. Once we hear the story and accept it as true, we rarely bother to question it.
Consider a group of students who built what has been called Harvard’s first online social network. Before they arrived at college, they had already connected more than an eighth of the entering freshman class in an “e-group.”21 But once they got to Cambridge, they abandoned the network and shut it down. Five years later Mark Zuckerberg started Facebook on the same campus. From time to time, the students who created the original e-group have felt some pangs of regret. I know, because I was one of the cofounders of that group. Let’s be clear: I never would have had the vision for what Facebook became. In hindsight, though, my friends and I clearly missed a series of chances for rethinking the potential of our platform. Our first instinct was to use the e-group to make new friends for ourselves; we didn’t consider whether it would be of interest to students at other schools or in life beyond school. Our well-learned habit was to use online tools to connect with people far away; once we lived within walking distance on the same campus, we figured we no longer needed the e-group. Although one of the cofounders was studying computer science and another early member had already founded a successful tech startup, we made the flawed assumption that an online social network was a passing hobby, not a huge part of the future of the internet
Progress is impossible without change;1 and those who cannot change their minds cannot change anything. —GEORGE BERNARD SHAW
Most of us take pride in our knowledge and expertise, and in staying true to our beliefs and opinions. That makes sense in a stable world, where we get rewarded for having conviction in our ideas. The problem is that we live in a rapidly changing world, where we need to spend as much time rethinking as we do thinking.
Unfortunately, when it comes to our own knowledge and opinions, we often favor feeling right over being right.
Two decades ago my colleague Phil Tetlock discovered something peculiar. As we think and talk, we often slip into the mindsets of three different professions:13 preachers, prosecutors, and politicians. In each of these modes, we take on a particular identity and use a distinct set of tools. We go into preacher mode when our sacred beliefs are in jeopardy: we deliver sermons to protect and promote our ideals. We enter prosecutor mode when we recognize flaws in other people’s reasoning: we marshal arguments to prove them wrong and win our case.14 We shift into politician mode when we’re seeking to win over an audience: we campaign and lobby for the approval of our constituents. The risk is that we become so wrapped up in preaching that we’re right, prosecuting others who are wrong, and politicking for support that we don’t bother to rethink our own views.
But being a scientist is not just a profession.17 It’s a frame of mind—a mode of thinking that differs from preaching, prosecuting, and politicking. We move into scientist mode when we’re searching for the truth: we run experiments to test hypotheses and discover knowledge.
Over the following year, the startups in the control group averaged under $300 in revenue. The startups in the scientific thinking group averaged over $12,000 in revenue. They brought in revenue more than twice as fast—and attracted customers sooner, too. Why? The entrepreneurs in the control group tended to stay wedded to their original strategies and products. It was too easy to preach the virtues of their past decisions, prosecute the vices of alternative options, and politick by catering to advisers who favored the existing direction. The entrepreneurs who had been taught to think like scientists, in contrast, pivoted more than twice as often. When their hypotheses weren’t supported, they knew it was time to rethink their business models.
Sometimes even great scientists need to think more like scientists.
THE SMARTER THEY ARE, THE HARDER THEY FAIL Mental horsepower doesn’t guarantee mental dexterity. No matter how much brainpower you have, if you lack the motivation to change your mind, you’ll miss many occasions to think again. Research reveals that the higher you score on an IQ test, the more likely you are to fall for stereotypes, because you’re faster at recognizing patterns.21 And recent experiments suggest that the smarter you are,22 the more you might struggle to update your beliefs.
Being a quant jock makes you more accurate in interpreting the results—as long as they support your beliefs. Yet if the empirical pattern clashes with your ideology, math prowess is no longer an asset; it actually becomes a liability. The better you are at crunching numbers, the more spectacularly you fail at analyzing patterns that contradict your views. If they were liberals, math geniuses did worse than their peers at evaluating evidence that gun bans failed. If they were conservatives, they did worse at assessing evidence that gun bans worked. In psychology there are at least two biases that drive this pattern. One is confirmation bias:23 seeing what we expect to see. The other is desirability bias:24 seeing what we want to see. These biases don’t just prevent us from applying our intelligence. They can actually contort our intelligence into a weapon against the truth. We find reasons to preach our faith more deeply, prosecute our case more passionately, and ride the tidal wave of our political party. The tragedy is that we’re usually unaware of the resulting flaws in our thinking. My favorite bias is the “I’m not biased”25 bias, in which people believe they’re more objective than others. It turns out that smart people are more likely to fall into this trap.26 The brighter you are, the harder it can be to see your own limitations. Being good at thinking can make you worse at rethinking.
After all, the purpose of learning isn’t to affirm our beliefs; it’s to evolve our beliefs.
Although Jobs knew how to “think different,” it was his team that did much of the rethinking.
Ignorance more frequently begets confidence than does knowledge.1 —CHARLES DARWIN
When we lack the knowledge and skills to achieve excellence, we sometimes lack the knowledge and skills to judge excellence.
If you don’t know a thing about football, you probably don’t walk around believing you know more than the coach.
Still, it leaves me wondering if we’ve been misjudging impostor syndrome by seeing it solely as a disorder.
When a core belief is questioned, though, we tend to shut down rather than open up. It’s as if there’s a miniature dictator living inside our heads,8 controlling the flow of facts to our minds, much like Kim Jong-un controls the press in North Korea. The technical term for this in psychology is the totalitarian ego, and its job is to keep out threatening information.
Being wrong is the only way I feel sure I’ve learned anything.”
Danny isn’t interested in preaching, prosecuting, or politicking. He’s a scientist devoted to the truth. When I asked him how he stays in that mode, he said he refuses to let his beliefs become part of his identity. “I change my mind at a speed that drives my collaborators crazy,” he explained. “My attachment to my ideas is provisional. There’s no unconditional love for them.” Attachment. That’s what keeps us from recognizing when our opinions are off the mark and rethinking them. To unlock the joy of being wrong, we need to detach. I’ve learned that two kinds of detachment are especially useful: detaching your present from your past and detaching your opinions from your identity.
In psychology, one way of measuring the similarity between the person you are right now and your former self is to ask: which pair of circles best describes how you see yourself?
As Bridgewater founder Ray Dalio told me, “If you don’t look back at yourself and think,21 ‘Wow, how stupid I was a year ago,’ then you must not have learned much in the last year.”
I’m guessing you wouldn’t want to see a doctor whose identity is Professional Lobotomist, send your kids to a teacher whose identity is Corporal Punisher, or live in a town where the police chief’s identity is Stop-and-Frisker. Once upon a time, all of these practices were seen as reasonable and effective.
I consider all of my opinions tentative. When the facts change, I change my opinions.”
I’ve noticed a paradox in great scientists and superforecasters: the reason they’re so comfortable being wrong is that they’re terrified of being wrong. What sets them apart is the time horizon
Jean-Pierre Beugoms has a favorite trick for catching himself when he’s wrong. When he makes a forecast, he also makes a list of the conditions in which it should hold true—as well as the conditions under which he would change his mind. He explains that this keeps him honest, preventing him from getting attached to a bad prediction.
— Board room predictions
In the early 1990s, the British physicist Andrew Lyne published a major discovery in the world’s most prestigious science journal.39 He presented the first evidence that a planet could orbit a neutron star—a star that had exploded into a supernova. Several months later, while preparing to give a presentation at an astronomy conference, he noticed that he hadn’t adjusted for the fact that the Earth moves in an elliptical orbit, not a circular one. He was embarrassingly, horribly wrong. The planet he had discovered didn’t exist. In front of hundreds of colleagues, Andrew walked onto the ballroom stage and admitted his mistake. When he finished his confession, the room exploded in a standing ovation. One astrophysicist called it “the most honorable thing I’ve ever seen.”40
“Taking responsibility is taking your power back.”
When we find out we might be wrong, a standard defense is “I’m entitled to my opinion.” I’d like to modify that: yes, we’re entitled to hold opinions inside our own heads. If we choose to express them out loud, though, I think it’s our responsibility to ground them in logic and facts, share our reasoning with others, and change our minds when better evidence emerges.
The teams that performed poorly started with more relationship conflict than task conflict. They entered into personal feuds early on and were so busy disliking one another that they didn’t feel comfortable challenging one another
It took months for many of the teams to make real headway on their relationship issues, and by the time they did manage to debate key decisions, it was often too late to rethink their directions.
As they resolved some of their differences of opinion, they were able to align on a direction and carry out their work until they ran into new issues to debate.
Relationship conflict is destructive in part because it stands in the way of rethinking. When a clash gets personal and emotional, we become self-righteous preachers of our own views, spiteful prosecutors of the other side, or single-minded politicians who dismiss opinions that don’t come from our side. Task conflict can be constructive when it brings diversity of thought, preventing us from getting trapped in overconfidence cycles. It can help us stay humble, surface doubts, and make us curious about what we might be missing. That can lead us to think again, moving us closer to the truth without damaging our relationships.
When students fill out course evaluations, one of their most common complaints is that I’m “too supportive of stupid comments.”
When he refused, they fired him.
We learn more from people who challenge our thought process than those who affirm our conclusions. Strong leaders engage their critics and make themselves stronger. Weak leaders silence their critics and make themselves weaker. This reaction isn’t limited to people in power. Although we might be on board with the principle, in practice we often miss out on the value of a challenge network.
Some organizations and occupations counter those tendencies by building challenge networks into their cultures. From time to time the Pentagon and the White House have used aptly named “murder boards”32 to stir up task conflict, enlisting tough-minded committees to shoot down plans and candidates. At X, Google’s “moonshot factory,”33 there’s a rapid evaluation team that’s charged with rethinking proposals: members conduct independent assessments and only advance the ones that emerge as both audacious and achievable. In science, a challenge network is often a cornerstone of the peer-review process. We submit articles anonymously, and they’re reviewed blindly by independent experts. I’ll never forget the rejection letter I once received in which one of the reviewers encouraged me to go back and read the work of Adam Grant. Dude, I am Adam Grant.
I’m looking for disagreeable people who are givers, not takers. Disagreeable givers often make the best critics: their intent is to elevate the work, not feed their own egos.
— Radical Candor
It’s common for people who lack power or status to shift into politician mode,36 suppressing their dissenting views in favor of conforming to the HIPPO—the HIghest Paid Person’s Opinion. Sometimes they have no other choice if they want to survive.
The notion of a spirited debate captures something important about how and why good fights happen. If you watch Brad argue with his colleagues—or the pirates fight with one another—you can quickly see that the tension is intellectual, not emotional. The tone is vigorous and feisty rather than combative or aggressive. They don’t disagree just for the sake of it; they disagree because they care. “Whether you disagree loudly, or quietly yet persistently put forward a different perspective,” Nicole explains, “we come together to support the common goal of excellence—of making great films.”
How can we avoid that slippery slope?
GETTING HOT WITHOUT GETTING MAD A major problem with task conflict is that it often spills over into relationship conflict.38 One minute you’re disagreeing about how much seasoning to put on the Thanksgiving turkey, and the next minute you find yourself yelling “You smell!”
Although the Wright brothers had a lifetime of experience discovering each other’s hot buttons, that didn’t mean they always kept their cool. Their last grand challenge before liftoff was their single hardest problem: designing a propeller. They knew their airplane couldn’t take flight without one, but the right kind didn’t exist. As they struggled with various approaches, they argued back and forth for hours at a time, often raising their voices. The feuding lasted for months as each took turns preaching the merits of his own solutions and prosecuting the other’s points. Eventually their younger sister, Katharine, threatened to leave the house if they didn’t stop fighting. They kept at it anyway, until one night it culminated in what might have been the loudest shouting match of their lives. Strangely, the next morning, they came into the shop and acted as if nothing had happened. They picked up the argument about the propeller right where they had left off—only now without the yelling. Soon they were both rethinking their assumptions and stumbling onto what would become one of their biggest breakthroughs. The Wright brothers were masters at having intense task conflict without relationship conflict. When they raised their voices, it reflected intensity rather than hostility. As their mechanic marveled, “I don’t think they really got mad, but they sure got awfully hot.”
“I see that you are back at your old trick of giving up before you are half beaten in an argument. I feel pretty certain of my own ground but was anticipating the pleasure of a good scrap before the matter was settled. Discussion brings out new ways of looking at things.”
When they argued about the propeller, the Wright brothers were making a common mistake. Each was preaching about why he was right and why the other was wrong. When we argue about why, we run the risk of becoming emotionally attached to our positions and dismissive of the other side’s. We’re more likely to have a good fight if we argue about how. When social scientists asked people why they favor particular policies on taxes,40 health care, or nuclear sanctions, they often doubled down on their convictions. Asking people to explain how those policies would work in practice—or how they’d explain them to an expert—activated a rethinking cycle. They noticed gaps in their knowledge, doubted their conclusions, and became less extreme; they were now more curious about alternative options.
They appeared less assertive, but much like in a dance, they led by letting their partners step forward.
When I asked Harish how to improve at finding common ground, he offered a surprisingly practical tip. Most people immediately start with a straw man, poking holes in the weakest version of the other side’s case. He does the reverse: he considers the strongest version of their case, which is known as the steel man.10
There are times when preaching and prosecuting can make us more persuasive. Research suggests that the effectiveness of these approaches hinges on three key factors: how much people care about the issue, how open they are to our particular argument, and how strong-willed they are in general. If they’re not invested in the issue or they’re receptive to our perspective, more reasons can help: people tend to see quantity as a sign of quality.11 The more the topic matters to them, the more the quality of reasons matters.12 It’s when audiences are skeptical of our view, have a stake in the issue, and tend to be stubborn that piling on justifications is most likely to backfire.13 If they’re resistant to rethinking, more reasons simply give them more ammunition to shoot our views down.
In a matter of seconds I’d devolved from logic bully to playground bully.
Having a conversation about the conversation shifts attention away from the substance of the disagreement and toward the process for having a dialogue. The more anger and hostility the other person expresses, the more curiosity and interest you show. When someone is losing control, your tranquility is a sign of strength. It takes the wind out of their emotional sails. It’s pretty rare for someone to respond by screaming “SCREAMING IS MY PREFERRED MODE OF COMMUNICATION!” This is a fifth move that expert negotiators made more often than average negotiators. They were more likely to comment on their feelings about the process and test their understanding of the other side’s feelings: I’m disappointed in the way this discussion has unfolded—are you frustrated with it? I was hoping you’d see this proposal as fair—do I understand correctly that you don’t see any merit in this approach at all? Honestly, I’m a little confused by your reaction to my data—if you don’t value the kind of work I do, why did you hire me? In a heated argument, you can always stop and ask, “What evidence would change your mind?” If the answer is “nothing,” then there’s no point in continuing the debate. You can lead a horse to water, but you can’t make it think.
In politics, you probably know some Democrats who view Republicans as being greedy, ignorant, heartless cretins, and some Republicans who regard Democrats as lazy, dishonest, hypersensitive snowflakes. As stereotypes stick and prejudice deepens, we don’t just identify with our own group; we disidentify with our adversaries,8 coming to define who we are by what we’re not. We don’t just preach the virtues of our side; we find self-worth in prosecuting the vices of our rivals.
When students answered this question, most felt strongly enough that they were willing to sacrifice a dollar to those who agreed with them to make sure those who disagreed got less.
Psychologists find that many of our beliefs are cultural truisms:38 widely shared, but rarely questioned. If we take a closer look at them, we often discover that they rest on shaky foundations. Stereotypes don’t have the structural integrity of a carefully built ship. They’re more like a tower in the game of Jenga—teetering on a small number of blocks, with some key supports missing. To knock it over, sometimes all we need to do is give it a poke. The hope is that people will rise to the occasion and build new beliefs on a stronger foundation.
As a general rule, it’s those with greater power who need to do more of the rethinking, both because they’re more likely to privilege their own perspectives and because their perspectives are more likely to go unquestioned.41 42 In most cases, the oppressed and marginalized have already done a great deal of contortion to fit in.
Daryl doesn’t do this by preaching or prosecuting. When he begins a dialogue with white supremacists, many are initially surprised by his thoughtfulness. As they start to see him as an individual and spend more time with him, they often tap into a common identity around shared interests in topics like music. Over time, he helps them see that they joined these hate groups for reasons that weren’t their own—it was a family tradition dating back multiple generations, or someone had told them their jobs were being taken by Black men. As they realize how little they truly know about other groups, and how shallow stereotypes are, they start to think again. After getting to know Daryl, one Imperial Wizard didn’t stop at leaving the KKK. He shut down the chapter. Years later, he asked Daryl to be his daughter’s godfather
It’s a rare person who wants to hear what he doesn’t want to hear. —ATTRIBUTED TO DICK CAVETT
Instead of attacking or demeaning his clients, Miller started asking them questions and listening to their answers. Soon afterward, he published a paper on his philosophy, which found its way into the hands of Stephen Rollnick, a young nurse trainee working in addiction treatment. A few years later, the two happened to meet in Australia and realized that what they were exploring was much bigger than just a new approach to treatment. It was an entirely different way of helping people change. Together, they developed the core principles of a practice called motivational interviewing.11 The central premise is that we can rarely motivate someone else to change. We’re better off helping them find their own motivation to change. Let’s say you’re a student at Hogwarts, and you’re worried your uncle is a fan of Voldemort. A motivational interview might go like this: You: I’d love to better understand your feelings about He Who Must Not Be Named. Uncle: Well, he’s the most powerful wizard alive. Also, his followers promised me a fancy title. You: Interesting. Is there anything you dislike about him? Uncle: Hmm. I’m not crazy about all the murdering. You: Well, nobody’s perfect. Uncle: Yeah, but the killing is really bad. You: Sounds like you have some reservations about Voldemort. What’s stopped you from abandoning him? Uncle: I’m afraid he might direct the murdering toward me. You: That’s a reasonable fear. I’ve felt it too. I’m curious: are there any principles that matter so deeply to you that you’d be willing to take that risk?
Overall, motivational interviewing has a statistically and clinically meaningful effect on behavior change in roughly three out of four studies, and psychologists and physicians using it have a success rate of four in five. There aren’t many practical theories in the behavioral sciences with a body of evidence this robust.27
In motivational interviewing, there’s a distinction between sustain talk and change talk.29 Sustain talk is commentary about maintaining the status quo. Change talk is referencing a desire, ability, need, or commitment to make adjustments. When contemplating a change, many people are ambivalent—they have some reasons to consider it but also some reasons to stay the course. Miller and Rollnick suggest asking about and listening for change talk, and then posing some questions about why and how they might change.
Salespeople, fundraisers, and entrepreneurs have a vested interest in getting to yes.
As the writer E. M. Forster put it, “How can I tell what I think till I see what I say?”38 That understanding made Forster an unusually dedicated listener. In the words of one biographer, “To speak with him was to be seduced by an inverse charisma,39 a sense of being listened to with such intensity that you had to be your most honest, sharpest, and best self.” Inverse charisma. What a wonderful turn of phrase to capture the magnetic quality of a great listener. Think about how rare that kind of listening is
For two decades, the psychologist who runs the lab, Peter T. Coleman, has been bringing people together to talk about polarizing issues. His mission is to reverse-engineer the successful conversations and then experiment with recipes to make more of them. To put you in the right mindset before you begin your conversation about abortion, Peter gives you and the stranger a news article about another divisive issue: gun control. What you don’t know is that there are different versions of the gun control article, and which one you read is going to have a major impact on whether you land on the same page about abortion. If the gun control article covers both sides of the issue, making a balanced case for both gun rights and gun legislation, you and your adversary have a decent chance at reaching consensus on abortion. In one of Peter’s experiments, after reading a “both-sides” article, 46 percent of pairs were able to find enough common ground to draft and sign a statement together. That’s a remarkable result. But Peter went on to do something far more impressive. He randomly assigned some pairs to read another version of the same article, which led 100 percent of them to generate and sign a joint statement about abortion laws. That version of the article featured the same information but presented it differently. Instead of describing the issue as a black-and-white disagreement between two sides, the article framed the debate as a complex problem with many shades of gray,3 representing a number of different viewpoints.