How to Apply Think Again: Build a Rethinking Habit

Why Rethinking Beats Being Smart

In August 1949, a crew of fifteen smokejumpers parachuted into Mann Gulch, Montana, to fight what looked like a routine fire. Within minutes the fire jumped the gulch and started racing uphill toward them faster than a person can run. Their foreman, Wagner Dodge, did something that made no sense to anyone watching. He stopped running, lit a match, and burned a patch of grass in front of him, then lay down in the ashes and told his crew to do the same. They thought he'd lost his mind and kept running. Dodge survived in his "escape fire." Thirteen of the others did not.

Adam Grant opens Think Again: The Power of Knowing What You Don't Know (2021) with that story because it's about a very specific kind of failure. The crew wasn't lacking courage or fitness. They were lacking the ability to abandon their training, their instinct to run, and their heavy tools, fast enough to survive. Later research on wildland firefighters found that many who died were found still carrying their chainsaws and packs. Dropping your tools feels like admitting the fight is lost. Grant's point is that this reluctance, the refusal to let go of what we know, is a much wider human problem than a rare fire.

His thesis lands in one sentence: "Intelligence is traditionally viewed as the ability to think and learn. Yet in a turbulent world, there's another set of cognitive skills that might matter more: the ability to rethink and unlearn." Being smart can even work against you here. The sharper you are at building arguments, the better you get at defending conclusions you should be questioning. Rethinking is the counterweight, and unlike raw IQ, it's trainable.

This is why the book pairs so well with other guides to judgment. Grant is describing the mental posture; how to apply Thinking in Bets gives you the machinery for holding beliefs as probabilities instead of certainties. Read together, they turn "keep an open mind" from a platitude into a practice.

The Four Modes: Preacher, Prosecutor, Politician, Scientist

Grant borrows a framework from the psychologist Philip Tetlock. When our beliefs are threatened, most of us reflexively drop into one of three mindsets, and none of them help us find the truth.

In preacher mode, we treat our convictions as sacred and deliver sermons to protect them. In prosecutor mode, we hunt for flaws in the other person's case, because the point is to win, not to learn. In politician mode, we campaign for approval, telling our audience what it wants to hear and bending our stated views to court their favor. Each mode feels like thinking. None of it is rethinking.

The fourth mode is the one to build. In scientist mode, you treat your beliefs as hypotheses to be tested, not truths to be defended. You get curious about what would prove you wrong and you look for it on purpose. Grant cites a study in which Italian entrepreneurs were trained to treat their business strategy as a hypothesis and run experiments on it. Compared with a control group, the scientist-trained founders brought in revenue more than twice as fast, because they pivoted on evidence instead of clinging to a plan.

The tell is how you feel when someone disagrees. If your first instinct is to convince, condemn, or campaign, you've left scientist mode. The practical move is to catch which chair you're sitting in, then ask a scientist's question instead: what evidence would change my mind, and have I actually looked for it?

Confident Humility and the Overconfidence Trap

The usual advice is to be more confident or more humble, as if they sit on opposite ends of a slider. Grant argues they're independent, and the sweet spot needs both. He calls it "confident humility": faith in your ability to figure something out, paired with real doubt about whether you already have. Confidence is believing in yourself. Humility is knowing you don't have all the answers. You can hold both at once, and the people who learn fastest do.

The failure mode is armor-plated confidence with no humility, and the psychology has a name. The Dunning-Kruger effect describes how people with the least competence in an area often overestimate their ability the most, precisely because they lack the knowledge to see what they're missing. The cruel twist is that a little learning inflates confidence faster than competence, so the moment you feel like an expert is often the moment you should be most suspicious.

Grant's business example is BlackBerry. Its founder, Mike Lazaridis, built the device that put email in everyone's pocket and dominated the smartphone market. When Apple showed the iPhone, Lazaridis was convinced people didn't want a touchscreen keyboard, a browser hungry for bandwidth, or a phone that doubled as a computer. He was defending a belief that had made him a fortune. BlackBerry's share of the US smartphone market fell from around 50% in 2009 to less than 1% by 2014. The knowledge that built the company became the anchor that sank it, because nobody rethought it in time.

Confident humility is what keeps you from becoming a Lazaridis. It sounds like "I'm good at learning this" rather than "I already know this," and that one swap keeps the door open long enough to notice when the world has moved.

The Joy of Being Wrong

Why is rethinking so hard? Because we tie our opinions to our identity, so an attack on the idea feels like an attack on us. Grant's fix is to detach the two. Detach your present from your past, and detach your opinions from your identity. Your beliefs are things you have, not things you are. Once that link is cut, being wrong stops feeling like a defeat.

Grant admits he's trained himself to feel something stranger in that moment: delight. When he discovers he was wrong, it means the version of him from this morning was less accurate than the version that exists now. He learned something. "If knowledge is power, knowing what we don't know is wisdom," he writes. The scientist's reflex is not "I knew it" but "here's something I didn't know," and the second one is where growth lives.

There's a discipline underneath the good feeling. Grant suggests holding your identity loosely and holding your reasons for believing something even more loosely than the belief itself. Rethink how you came to a conclusion, and you can update the conclusion without any drama. This is the same muscle that Carol Dweck's growth mindset trains, applied to specific beliefs instead of general ability. A fixed mindset says being wrong exposes a flaw. A growth mindset, aimed at your opinions, says being wrong is how the flaw gets fixed.

The practical version for readers is small but powerful: keep a record of what you used to think. When you can look back at a highlight from a year ago and see that you now disagree with the version of you that saved it, that's not embarrassing. That's the receipt for your own learning.

How to Argue So People Actually Rethink

If rethinking is hard for you, it's harder for the person across the table. Grant devotes much of the book to persuasion that respects that difficulty, and his star example is Harish Natarajan, one of the most successful competitive debaters alive. Facing an audience that had already made up its mind, Natarajan doesn't overwhelm people with a flood of arguments. He picks a few strong points, concedes what's true in the other side, and asks questions that lead the audience to reexamine their own view.

The counterintuitive lesson: more reasons often persuade less. Pile on ten arguments and a few weak ones dilute your strongest, while your opponent picks off the easy targets and feels justified dismissing the rest. Grant also warns against "binary bias," the urge to sort a messy issue into two camps. Complicating the picture, showing that there are more than two positions, actually lowers people's defenses because it signals you're reasoning rather than recruiting.

Inside teams, Grant separates two kinds of conflict that people constantly confuse. Relationship conflict is personal and emotional, the kind that curdles into grudges, and it wrecks performance. Task conflict is disagreement about ideas and decisions, and the best teams have more of it, not less. The trick is to fight about the work without making it about the person. You want a "challenge network," a circle of people you trust to tell you you're wrong, which is very different from a support network that tells you you're great.

The most useful reframe for everyday disagreements: approach them like a scientist testing a hypothesis together, not a prosecutor building a case. Ask what evidence would change the other person's mind, and be honest about what would change yours. That question alone shifts a fight into an investigation.

Motivational Interviewing: Helping Others Change Their Minds

Sometimes you want to help someone rethink something important, and every direct argument makes them dig in harder. Grant's answer is a technique from clinical psychology called motivational interviewing. Instead of pushing, you help people find their own reasons to change by asking open questions, listening for their doubts, and reflecting their words back to them.

The story he tells involves a pediatrician known as a "vaccine whisperer" who works with hesitant parents. He doesn't bombard them with data or shame them. He asks what worries them, listens without judgment, and lets them talk their way toward the decision. Programs built on this approach measurably increased vaccination rates, not by winning arguments but by declining to have them. The counterintuitive core of motivational interviewing is that people are more persuaded by what they hear themselves say than by what you tell them.

Three moves make it work. Ask open-ended questions rather than leading ones. Practice reflective listening, so the person feels understood before you offer anything. And affirm their freedom to decide, because the pressure to comply is usually what triggers the resistance. You're not steering to a preset destination. You're a guide who trusts the other person to get somewhere better if you clear the path.

This matters for anyone who shares ideas online, mentors, or leads a team. The instinct is to preach the right answer. The skill is to ask the question that lets someone reach it themselves. You can practice the listening half by reading generously: when you meet a view you find wrong, try to summarize it so fairly its holder would nod before you decide what you think.

Turn Your Highlights Into a Rethinking System

Grant's ideas are easy to nod along to and easy to forget. The way to keep them is to wire rethinking into how you read, and a highlighting workflow is well suited to it. Reading is where beliefs form, so it's the natural place to build the habit of questioning them.

Start by changing what you highlight. Most people mark the sentences that confirm what they already think, which quietly builds a preacher's scrapbook. Flip it. When you use Glasp's web highlighter, reserve one highlight color for claims that challenge you, contradict a note you saved earlier, or make you a little uncomfortable. Over time that color becomes a challenge network on the page, a running collection of the strongest cases against your current views. The same discipline works for video: pull the transcript points that push back on you with YouTube Summary, not just the ones that ring true.

Then close the loop by revisiting. Rethinking only happens if you go back. Set a monthly habit of rereading old highlights and asking a single question of each: do I still believe this? When the answer is no, don't delete the old note. Annotate it with what you think now and why you changed. That trail of updated beliefs is the tangible proof of learning that Grant celebrates, and it turns your notes into a rethinking log instead of a museum of past certainties.

You can also recruit AI as a sparring partner. Use Glasp's AI chat to interrogate your own saved highlights: ask it to steelman the opposite of a note, list the strongest objections to a claim you've collected, or point out where two of your highlights contradict each other. That's scientist mode on demand, a way to run the test Grant wants you to run without waiting for someone to disagree with you.

Finally, widen the input. Preachers read to confirm; scientists read to be surprised. Browse Glasp's community feed to see what people who don't share your assumptions are highlighting in the same articles and books. Seeing which line a stranger marked as essential, when you skimmed right past it, is one of the cheapest ways to notice your own blind spots. Building this kind of second brain is less about hoarding and more about keeping your beliefs in motion; our guide to the digital commonplace book covers the collecting habit that feeds it.

A 7-Day Rethinking Playbook

You don't rebuild a thinking style by reading about it. You rebuild it with small reps. Here's a week-long starter plan that turns each chapter's idea into a single concrete action you can actually do.

The pattern under all seven days is the same loop: surface a belief, test it against evidence or another mind, and update on purpose. Run it for a week and you'll feel the shift from wanting to be right to wanting to get it right. That's the whole book compressed into a habit, and the compounding is real; a mind that updates weekly ends the year in a very different place than one that hardened in January.

Frequently Asked Questions

What is the main idea of Think Again by Adam Grant?

The book argues that in a fast-changing world, the ability to rethink and unlearn matters more than raw intelligence. Grant shows that we default into "preacher," "prosecutor," and "politician" modes that defend our beliefs, and he makes the case for a fourth mode, thinking like a scientist, where you treat your opinions as hypotheses to be tested rather than truths to be protected.

What are the four modes of thinking in Think Again?

Preacher (defending sacred beliefs), prosecutor (attacking the other side to win), politician (seeking approval by telling people what they want to hear), and scientist (testing beliefs as hypotheses and looking for evidence you're wrong). Grant, drawing on Philip Tetlock's research, argues we should spend far more time in scientist mode, even about topics we know well.

What does "confident humility" mean?

It's the combination of believing in your ability to figure things out with genuine doubt about whether you already have the right answer. Grant treats confidence and humility as partners rather than opposites. Confident humility keeps you learning, while unearned certainty, the overconfidence behind failures like BlackBerry, keeps you stuck.

How is Think Again different from Thinking in Bets or Superforecasting?

They're complementary. Thinking in Bets focuses on treating decisions as probabilistic wagers, and Superforecasting studies people who predict the future accurately by updating in small steps. Think Again is broader and more personal: it's about the mindset and habits of rethinking itself, across arguments, identity, teams, and relationships, not just decisions and forecasts.

How can I practice rethinking while reading?

Highlight the passages that challenge your beliefs, not only the ones that confirm them, and revisit your old notes on a schedule to see what you'd now revise. Using a tool like Glasp, you can reserve a highlight color for challenging ideas, annotate beliefs you've changed instead of deleting them, and use AI chat to argue the opposite of your own saved highlights.

Conclusion

Think Again reframes what it means to be a good thinker. It isn't about how much you know or how fast you learn it. It's about how willingly you let go of a belief when the evidence turns, how comfortable you are saying "I was wrong," and how often you choose the scientist's chair over the preacher's, prosecutor's, or politician's. Those are habits, which means they're trainable, and reading is one of the best places to train them.

The move is small and repeatable. Highlight what unsettles you, revisit what you saved, and treat every strong opinion as a hypothesis with a shelf life. Glasp's web highlighter, AI chat, and community feed give you a place to collect the challenges, argue with your past self, and see what people who think differently noticed that you missed. Start with one highlight color for the ideas that make you uncomfortable. That's where the rethinking begins.