How Minds Actually Change

Most arguments fail before the first fact lands.

Not because one side is dumb. Not because the evidence is weak.

They fail because identity threat shows up first — and curiosity disappears.

When someone hears “your view is wrong” as “you are bad,” their brain switches from exploration to defense. Your logic is now competing with social survival. Social survival wins every time.

Two books changed how I think about this problem. Adam Grant’s Think Again examines how we update our own beliefs. David McRaney’s How Minds Change examines how others update theirs. They arrive at the same core insight from opposite directions: facts alone almost never change minds. What changes minds is the quality of the conversation surrounding those facts and the emotional safety of the person hearing them.

Both books shaped the design philosophy behind 9takes. But more importantly, they reveal something most people get wrong about persuasion itself.

The Conversation That Goes Wrong

Picture a dinner table. Someone brings up a political topic — immigration, gun policy, school curriculum, whatever lights up your particular nervous system.

You have data. Good data. You’ve read the studies. You know the counterarguments. You open your mouth, and you’re precise, organized, factual.

And the other person’s eyes glaze over. Or worse — they harden. Their jaw sets. You can see them mentally drafting their rebuttal before you’ve finished your second sentence. They’re not listening anymore. They’re defending.

You didn’t say anything wrong. You just triggered something deeper than logic. You threatened their identity — the web of beliefs that tells them who they are, what tribe they belong to, and whether they’re a good person. Once that threat registers, the conversation is over. Everything after that point is performance.

This is the moment Adam Grant calls the shift out of scientist mode. And it happens faster than you think.

The Four Modes We Get Trapped In

Grant opens Think Again with an observation from psychologist Phil Tetlock that’s so accurate it stings. When we think and talk, we unconsciously slip into the mindset of one of three professions: preacher, prosecutor, or politician.

In preacher mode, your beliefs become sacred. You deliver sermons. You’re not exploring — you’re protecting. Changing your mind feels like moral weakness, like betraying something essential about who you are.

In prosecutor mode, you’re building a case. You’re scanning for flaws in the other person’s reasoning, marshaling arguments to prove them wrong. You’re not trying to understand — you’re trying to win.

In politician mode, you’re performing. You’re reading the room, adjusting your message to whatever gets approval. You’re not interested in truth — you’re interested in applause.

All three modes feel productive. The preacher feels righteous. The prosecutor feels sharp. The politician feels savvy. But none of them are actually thinking. They’re running on autopilot with different emotional fuel.

Grant illustrates this with Mike Lazaridis, the inventor of BlackBerry. In 2009, BlackBerry owned nearly half the smartphone market. Lazaridis was brilliant — he’d built a genuinely revolutionary device. But his brilliance trapped him. He was convinced the BlackBerry was for email, period. People would never want a full computer in their pocket. His own engineers told him otherwise. He overruled them.

He’d slipped into preacher mode. His pride in what he’d built became inseparable from his beliefs about what it should become. By 2014, BlackBerry’s market share had collapsed to less than 1%.

The lesson isn’t that Lazaridis was stupid. The lesson is that intelligence doesn’t protect you from the preacher trap. If anything, it makes the trap more convincing. Smart people are better at rationalizing positions they arrived at for emotional reasons.

This is the first layer of how minds change: you have to be willing to rethink your own.

What Actually Changes Someone Else’s Mind

But personal rethinking only gets you halfway. What about the person across the table — the one whose eyes glazed over? How do you help them update?

David McRaney spent years investigating this question, and the most surprising answer he found came from a small team working out of the Los Angeles LGBT Center.

The Leadership LAB developed a technique called deep canvassing. Canvassers would go door to door in Los Angeles neighborhoods and have conversations about contentious issues — specifically, transgender rights. Not scripted talking points. Not pamphlets. Real conversations.

Here’s what a typical exchange looked like: A canvasser asked a woman how she felt about laws allowing transgender women to use women’s restrooms. She rated herself a six out of ten on support. She said she worried about children’s safety. Then the canvasser asked a simple question: “Do you know anyone who is transgender?”

She paused. She did — a nephew she’d helped raise, until he began transitioning. They hadn’t spoken in a long time. It bothered her.

That was the moment. Not data. Not argument. A question that connected the abstract policy to a real person in her real life. The canvasser didn’t lecture. Didn’t correct. Just listened, and asked one more question: “Have you ever been treated unfairly for being different? What did that feel like?”

The results stunned political scientists. A study by researchers David Broockman and Joshua Kalla, published in Science, found that a single 10-minute deep canvassing conversation changed approximately one in ten people’s attitudes toward transgender rights. The change wasn’t superficial — it lasted at least three months, with no signs of backsliding. For context, that single conversation produced a shift larger than the average decrease in American homophobia from 1998 to 2012.

One conversation. Ten minutes. No coercion.

The mechanism is the opposite of what most people try. Most people dump evidence, challenge competence, and demand concession. Deep canvassing does the reverse: it asks how someone arrived at their belief, reflects that reasoning back without judgment, and invites them to connect their lived experience to the topic at hand. People don’t change because they’re overpowered. They change because they feel safe enough to examine their own thinking.

Why Knowing This Isn’t Enough

If you’ve read both books and absorbed the lessons, you now know two powerful things: how to rethink your own beliefs (scientist mode), and how to help others rethink theirs (identity-safe dialogue). That should be enough, right?

It isn’t. And the reason reveals something neither book fully explores.

Think about what happens when a political thread goes viral on social media. The first comment frames the entire conversation. Everyone who comes after is either agreeing with that frame or reacting against it. The most provocative response gets the most engagement. The most nuanced response gets buried. The platform is selecting for prosecutor and preacher mode, regardless of what any individual person intends.

Intentions matter less than structure.

This is the insight that connects Grant’s work, McRaney’s work, and what we’re building with 9takes. You need three layers working together:

Your own willingness to rethink (the scientist mindset from Think Again). Without this, you’re just performing — preaching, prosecuting, or politicking your way through conversations that were dead on arrival.

The skill to create safe conversations (the deep canvassing insight from How Minds Change). Reflect before rebutting. Ask “how” questions, not trap questions. Show genuine curiosity about how someone arrived at their view. Let them do their own updating.

A system that doesn’t sabotage the first two layers. This is the piece most conversations about polarization miss entirely. You can have the right mindset and the right technique, but if the platform rewards certainty theater over genuine curiosity, you’re fighting the architecture.

What Curiosity Theater Actually Looks Like

On 9takes, the core mechanic is simple: you answer first, then you see everyone else’s answers.

That one structural decision changes everything downstream. You can’t mimic the top comment because you haven’t seen it yet. You can’t perform for the crowd because you’re writing before you know what the crowd thinks. You’re forced into something uncomfortably close to honesty — your actual take, uncontaminated by social proof.

Then something interesting happens. When you unlock everyone else’s responses, you’re reading them differently than you would on a normal platform. You’re not scanning for who to agree with or who to attack. You’re comparing. You already committed to your position, so you’re not defensive. You’re genuinely curious about what others said.

Say a polarizing question gets posted. On most platforms, the default pattern is identity signal, outrage spiral, side-taking, zero movement. On 9takes, the pattern shifts: independent first response, slower comparison, less imitation, and better odds of arriving at something like “I still disagree, but I see your logic.”

That sentence — “I still disagree, but I see your logic” — is what progress actually looks like. Not conversion. Not capitulation. Just the acknowledgment that another person’s reasoning makes sense from where they’re standing. That’s the crack through which genuine understanding enters.

Why the Enneagram Makes This Work Better

Here’s where the Enneagram layer comes in, and probably not in the way you’d expect.

The Enneagram isn’t a replacement for evidence. It’s a translation layer for emotional logic. Two people can hear the exact same sentence and react in completely different ways — not because one is smarter or more informed, but because they’re protecting different core concerns.

One person hears a challenge to their competence. Another hears a threat to their relationships. A third hears an attack on their autonomy. Same words, three different nervous systems lighting up. And if you don’t understand which concern is driving the reaction, your perfectly logical argument is hitting the wrong target.

When you spot the emotional driver, you can keep a disagreement from escalating into something that feels personal. You stop assuming bad faith and start recognizing different wiring. That’s not soft. That’s strategic communication.

On 9takes, this context travels with every response. You’re not just reading nine different opinions — you’re reading nine different emotional architectures processing the same question. And that makes the comparison exponentially richer than a standard comment thread.

The Rethinking Challenge

If you’ve read this far, you probably recognize yourself in at least one of Grant’s four modes. Maybe you caught yourself preaching during a recent argument. Maybe you realized you were prosecuting a colleague’s idea when you could have been asking better questions. Maybe you’ve been politicking — performing agreement you don’t feel because the social cost of honesty seemed too high.

Here’s a challenge worth trying this week. Next time you find yourself in a disagreement, pause and ask yourself three questions:

Which mode am I in right now — preacher, prosecutor, or politician?

What is the other person protecting — and can I name it without judgment?

What evidence would actually change my mind about this?

That last question is the hardest. If you can’t answer it honestly, you’re not in scientist mode. You’re somewhere else — and the conversation is already dead.

The deep canvassers in Los Angeles changed minds one doorstep at a time. Not with data. Not with force. With questions that invited people to connect their own experience to someone else’s reality. It worked because the structure of the conversation protected dignity.

That’s the bet 9takes is making. Not that people are too polarized to listen. Not that the internet is broken beyond repair. Just that the architecture of most conversations is optimized for the wrong outcome — and that better rails produce better results.

One question. Nine perspectives. A better shot at understanding.