How to change someone's mind
You almost never change a mind by winning an argument. Lower defenses first — here is how to shift someone without triggering a fight.
Changing someone’s mind almost never happens because you won the argument. Boghossian & Lindsay (How to Have Impossible Conversations) put the mechanism plainly: self-generated ideas change minds; delivered arguments trigger defenses. Your actual job is to lower the other person’s guard enough that they can change their own mind.
Why arguments backfire (and what to do instead)
The instinct when someone believes something wrong is to correct them — clearly, calmly, with evidence. It almost never works. Brehm (1966) documented what he called psychological reactance: the moment someone senses their freedom to hold a belief is being threatened, they defend it more vigorously, not less. The content of the argument barely matters. The act of pressure is the trigger.
This is not a quirk of unreasonable people. It is how minds work under perceived threat. The belief becomes a flag — and flags get defended. Lubrano’s observation in Don’t Talk About Politics captures the practical implication: arguments don’t change minds, but friends do. The relationship — the sense that the other person is genuinely on your side — is what makes updating a belief feel safe rather than shameful.
The move that follows from this is counterintuitive: stop trying to win. Winning the exchange and shifting the view are two different objectives, and they often pull in opposite directions. Winning produces a defeated opponent who retreats to their corner and loads up counter-arguments. Shifting a view requires them to stay in the room — curious, not cornered.
For a deeper look at how to stay in disagreement without burning the relationship, see our piece on how to disagree without damaging the relationship.
Build the golden bridge
Boghossian & Lindsay introduce the image of a golden bridge — a face-saving path from a held position to a new one. The obstacle to changing someone’s mind is rarely the evidence; it is the social cost of being wrong. If updating their view means publicly admitting they were mistaken — to you, or to themselves — most people will choose to stay wrong. It is the cheaper option in the short run.
Your job is to make the update cheap. Acknowledge what was coherent or understandable about their original position. Frame any new information as something that genuinely changed your own thinking, not as a gotcha. When they do shift — even slightly — do not name the moment. Let them cross the bridge without a ceremony that forces them to wave goodbye to the old view.
McQueen’s framing in Mindstuck is useful here: most resistance is identity-based, not intellectual. People don’t hold wrong beliefs because they haven’t seen the right facts; they hold them because the belief is connected to who they are. Reframing — presenting the same underlying idea in different terms — can shift their perspective without triggering the identity defense. The goal is not to change what they believe; it is to change what they think that belief requires of them.
Listen first — actually listen
Bregman argues in Four Seconds that listening is not the warm-up act before the persuasion; it is the persuasion. When people feel genuinely heard, their defenses drop. When defenses drop, they can actually consider a different view. When you are busy loading your next counter-argument, you are not listening — and they know it.
Motivational interviewing (Miller & Rollnick) operationalizes this. The technique was developed for clinical settings — helping patients resolve ambivalence about changing behaviour — but the core mechanism applies to any conversation: ask questions that cause the other person to voice their own reasons for changing. Researchers call this “change talk.” Self-generated arguments shift views far more reliably than delivered ones, because there is no external source to dismiss.
Shell (Bargaining for Advantage) and Klaff (Flip the Script) both make the same point from a negotiation angle: frame your ask in the other person’s vocabulary. If you know they value fairness, frame it as fairness. If they value autonomy, give them choices rather than directives. The framing does not change what you are asking for; it changes whether the ask feels like an imposition or a natural fit.
The Arbinger Institute (The Outward Mindset) adds one more layer: your mindset leaks. If you enter the conversation with the goal of winning — of moving them to your position — that agenda shows, and people defend against it. Genuine curiosity about their view opens doors that tactical curiosity slams shut. Check your actual intention before you open your mouth. This is not a technique; it is a prerequisite.
For a practical read on how to spot the rhetorical moves that undermine honest dialogue, see our guide to recognizing manipulation and rhetoric.
How to apply this when it actually matters
Broockman & Kalla (2016) studied deep canvassing — extended, empathetic conversations about contested social issues — and found it produced durable attitude shifts where standard fact-delivery canvassing produced almost none. The distinguishing feature was not better arguments but a specific conversational structure: acknowledge the person’s experience, connect on shared values, and invite reflection rather than delivering conclusions. The shifts held when measured weeks later.
Three practical moves follow from all of this:
Ask before you tell. Before offering any view, ask a question that makes the other person articulate their own. “What would it take to change your mind about this?” is underused and often genuinely illuminating. It surfaces their own criteria — and those are the only criteria that will move them.
Give choices, not ultimatums. Ultimatums trigger reactance instantly. Choices preserve autonomy and make both options feel available. “I could see this going two ways — which fits better with how you see it?” keeps the conversation open in a way that “you have to agree with me on this” does not.
Plan for more than one conversation. The first exchange plants a question. If it does nothing else, that is a win. Impatience — pressing for resolution in a single sitting — is one of the most reliable ways to cause someone to dig in. Give the question time to develop on their own terms, in their own head, without you in the room.
References
-
Reference How to Have Impossible Conversations
Boghossian, P., & Lindsay, J. (2019). Lifelong Books.
-
Reference A Theory of Psychological Reactance
Brehm, J. W. (1966). Academic Press.
-
Reference Mindstuck: Mastering the Art of Changing Minds
McQueen, M. (2023). Wiley.
-
Reference Bargaining for Advantage
Shell, G. R. (2006). Penguin Books.
-
Reference Flip the Script
Klaff, O. (2019). Portfolio/Penguin.
-
Reference Four Seconds
Bregman, P. (2015). HarperOne.
-
Reference The Outward Mindset
Arbinger Institute. (2019). Berrett-Koehler.
-
Reference Motivational Interviewing: Helping People Change
Miller, W. R., & Rollnick, S. (2012). Guilford Press.
-
Reference Durably reducing transphobia: A field experiment on door-to-door canvassing
Broockman, D., & Kalla, J. (2016). Science, 352(6282).
FAQ
Why does arguing usually make people more stubborn?
Because of **psychological reactance** — when someone feels their freedom to believe something is threatened, they dig in harder to reassert it. **Brehm (1966)** documented this effect: push a view at someone and they push back, not because the evidence is weak but because the act of pushing feels like a threat. The belief becomes a flag worth defending. **Boghossian & Lindsay** put it plainly in *How to Have Impossible Conversations*: the goal is never to win; it is to open a door the other person can choose to walk through.
What is psychological reactance and how does it affect persuasion?
**Reactance** is the discomfort people feel when they sense their autonomy being constrained. The moment a conversation feels like pressure — 'you should believe X' — the listener's immediate job becomes resisting X, regardless of its merits. **Brehm (1966)** named and documented the effect. Practically, it means that the more forcefully you push a view, the more you are likely to entrench the opposite one. The antidote is to present options and invite reflection rather than deliver conclusions.
What is a 'golden bridge' and how do I build one?
The **golden bridge** is a face-saving exit from a position someone held publicly. If shifting their view means admitting they were wrong in front of others — or in front of themselves — they will often choose to stay wrong instead. **Boghossian & Lindsay** argue that your job is to make the retreat easy: acknowledge what was right about their old view, frame the shift as a natural update given new information, and never name the moment when they change their mind. Let them cross without noticing the bridge.
Does feel-felt-found actually work?
It works when used honestly and fails badly when it sounds scripted. The structure — 'I understand how you **feel**; others have **felt** the same; what they **found** was...' — works because it validates the emotion before introducing new information. **McQueen** notes in *Mindstuck* that most resistance is not intellectual but identity-based: people need to feel heard before they can hear you. The risk is that a mechanical delivery reads as manipulation immediately, which triggers more reactance than no technique at all. Use the structure as scaffolding, not a script.
What is motivational interviewing and when is it useful?
**Motivational interviewing (MI)** is a clinical technique developed by **Miller & Rollnick** for helping people resolve ambivalence about change. The core move is asking questions that cause the person to voice their own reasons for changing — 'change talk' — rather than having reasons delivered to them. Research on MI consistently shows that **self-generated arguments** shift behaviour far more effectively than delivered arguments. The technique is most useful in one-on-one conversations where the other person is already ambivalent; it is not a shortcut for someone who is fully dug in.
How do I change someone's mind on a political or values issue?
Carefully and rarely — and almost never in a single conversation. **Lubrano** argues in *Don't Talk About Politics* that direct political persuasion fails because the argument becomes the trigger for reactance, not the content. **Broockman & Kalla (2016)** found that deep canvassing — sustained, empathetic conversation that focuses on shared experience rather than facts — produced durable attitude shifts on contested social issues, where standard canvassing produced almost none. The implication: connection before content, and spread the work across multiple conversations. See also our post on [how to talk across political and value divides](/en/blog/talk-across-political-and-value-divides).
Is the backfire effect real — do corrections ever make beliefs stronger?
The original **backfire effect** (Nyhan & Reifler, 2010) suggested corrections could deepen false beliefs. More recent research — including re-examination studies from **Wood & Porter (2019)** — has found the effect is **much rarer than once thought**: corrections generally reduce false beliefs rather than entrench them, though they rarely eliminate them entirely. The lesson is nuanced: corrections work better than nothing, but they work far better when delivered through a trusted relationship than through a stranger on the internet.
How do I frame a request using someone's own values?
First, listen long enough to know what they actually value — not what you assume they value. **Shell** in *Bargaining for Advantage* and **Klaff** in *Flip the Script* both argue that requests land when they are framed in the listener's vocabulary, not the speaker's. If someone values loyalty, frame the ask in terms of loyalty. If they value autonomy, give them choices rather than directives. The frame does not change what you are asking; it changes whether the ask feels like a natural fit or an imposition.
What does changing my own mindset have to do with changing someone else's?
Everything, according to **the Arbinger Institute** in *The Outward Mindset*. When you enter a conversation with the objective of winning — moving them to your position — your body language, your choice of questions, and your responses to their points all leak that agenda. The other person senses the goal and defends against it. An outward mindset starts from genuine curiosity about their view. The practical payoff: questions asked from real curiosity open people up; the same questions asked as tactics close them down. **Bregman** makes the same point in *Four Seconds* — listening, not arguing, is the mechanism that actually shifts minds.
How long does it usually take to change someone's mind on a deeply held belief?
Longer than one conversation, almost always. **Broockman & Kalla (2016)** tracked attitude shifts over weeks after a single deep canvassing session and found they held — but a single session rarely produces instant conversion. Expect the first conversation to plant a question, not close a case. Subsequent interactions let the question develop on their own terms. Impatience is one of the most reliable ways to trigger reactance: the moment someone feels rushed toward a conclusion, they slow down or reverse. Plan for a relationship, not a debate.