Anchoring Bias: Why the First Thing Said Decides the Rest

Anchoring bias is the pull that a first value exerts on every judgment that comes after it. Put a number in front of someone and their later estimates drift toward it, even when the number was arbitrary, even when they watched it being generated at random, and even when they are told to ignore it.

Most write-ups stop at salary negotiations and price tags. The version that matters more is the one with no numbers in it at all, where the anchor is how you were treated in the first few months of a relationship, and every year afterward gets quietly measured against that instead of against anything reasonable.

What Is Anchoring Bias?

The mechanism is called anchoring and adjustment. Faced with an uncertain judgment, people start from whatever value is available and adjust from there. The adjustment is reliably insufficient, so the final answer stays closer to the starting point than it should.

What makes it more than a curiosity is that the starting point does not have to be relevant, credible, or even plausible. It only has to be present. Once a value is in the room it becomes the thing you are measuring distance from, and the question silently changes from "what is right" to "how far from this is right."

What Did the Wheel of Fortune Study Show?

Amos Tversky and Daniel Kahneman set out the classic demonstration in their 1974 paper Judgment under Uncertainty: Heuristics and Biases, published in Science, volume 185, pages 1124 to 1131. It remains one of the most cited papers in the history of psychology.

The setup was deliberately blunt. In front of each participant they spun a wheel of fortune, rigged to stop on either 10 or 65. Then they asked a question with no connection to the wheel whatsoever: what percentage of African countries are members of the United Nations?

Participants who had watched the wheel land on 10 gave a median estimate of 25 percent. Participants who had watched it land on 65 gave a median estimate of 45 percent.

Twenty points of difference, produced by a number that everyone present had just seen come out of a spinning wheel. Nobody thought the wheel knew anything about the United Nations. The irrelevance was not hidden. It was demonstrated in front of them, and the anchor worked anyway.

That last detail is the one to carry forward. Anchoring is not a failure to notice that the anchor is meaningless. People notice. The pull survives the noticing.

Why Does the First Offer Usually Win?

Because it decides what the conversation is about before the conversation starts.

Adam Galinsky and Thomas Mussweiler tested this directly in negotiations and published the results in the Journal of Personality and Social Psychology in 2001. Across three experiments, whichever side made the first offer got the better outcome. In their chemical plant negotiation, the correlation between the first offer and the final settlement price was .85, which is enormous for a social process with two motivated parties actively trying to reach their own goals.

The explanation they favored is the selective accessibility model, and it is more useful than the usual folk version. The anchor does not simply drag a number. It changes what information comes to mind. Once a high figure is on the table, you start generating reasons the thing might be valuable, because those are the thoughts the anchor makes accessible. You are not resisting a number. You are searching your own memory using their frame. That tilted search has a close relative in confirmation bias, where the evidence you gather leans toward a view you already hold.

They also found what breaks it, which matters for the rest of this article. When the person who did not make the first offer deliberately focused on information inconsistent with that offer, their own target, their own walk-away point, the other side's alternatives, the first offer advantage disappeared entirely. Not reduced. Eliminated.

How Does Early Treatment Anchor What You Accept?

This is where the bias stops being about money.

Whatever happens in the first stretch of a relationship becomes the reference point, and it does so without anyone deciding. Two people can end up in objectively identical situations and experience them completely differently depending on where they started from.

If the early months were unusually attentive, that sets a high anchor. Later coldness then reads as a departure from who they really are, a dip, a phase, something caused by work stress. The anchor supplies a story in which the good version is the true version and the current version is the anomaly. This is exactly what makes the idealization stage of the narcissist playbook so expensive later. The intensity at the start was not just pleasant. It installed a reference point that keeps explaining away everything that followed.

If the early months were already difficult, the anchor sits low. Treatment that would alarm an outside observer registers as ordinary, because you are measuring the distance from where you began rather than from any external standard. People in this position often describe realizing years afterward that something was wrong from the first month, and being unable to explain why they did not see it. Anchoring is a large part of the answer.

Either way the same thing has happened. Your sense of normal is not an assessment. It is an adjustment from a starting value you did not choose. The same pull can come from a single bad moment instead of a number, which our guide to the horn effect looks at.

Why Does a Worse Example Make Your Situation Look Fine?

Because an extreme comparison installs a new anchor, and you then measure from it.

The thought is familiar. He has never hit me. At least he does not drink. My friend's husband is far worse. Each of those sentences is doing something specific. It is placing a severe example into the room, and once it is there your own circumstances get judged as a distance from that, which makes them look mild by construction.

Notice the question that has been swapped out. You began with "is this acceptable." You are now answering "is this as bad as that." Those are different questions with different answers, and the substitution happens silently.

Once the comparison has done its work, cognitive dissonance keeps it in place, because a reference point that makes your situation tolerable is exactly the kind of belief the mind protects.

This is also why comparison to a worse case is such a common thing to be handed by other people, including well-meaning ones. It usually is not manipulation. It is a listener reaching for reassurance. But the effect is to raise the bar for what counts as a problem, and the person who benefits from a higher bar is never you.

The same structure appears in moving the goalposts, though it runs the other way. Goalpost moving changes the target after you reach it. Anchoring sets the target before you start. Together they cover both ends, which is why situations built on both are so difficult to get a fixed read on.

How Does an Anchor Get Used on You Deliberately?

Usually through an opening position that nobody expects to hold, designed so that the real request looks reasonable by comparison.

Common shapes:

• The extreme opening. An outrageous demand or accusation, followed by a retreat to something merely unfair. The retreat feels like a concession, so you accept terms you would have refused cold.
• The apology anchor. An enormous dramatic apology early in the relationship sets the standard for what remorse looks like. Later, a muttered acknowledgment gets graded against ordinary silence rather than against that first performance, and passes.
• The worst incident as reference. Once a genuinely bad episode has happened, every subsequent one gets compared to it. Anything short of the worst day starts counting as fine.
• Their stated needs first. Whoever describes what they need at the start of a conversation has anchored it. Everything you raise afterward is heard as a response to their frame.

The extreme opening is worth contrasting with the foot-in-the-door technique, which runs the opposite way. Anchoring starts high so the real request looks small beside it. Foot-in-the-door starts tiny so the real request never gets compared to anything at all. Both end up somewhere you would have refused if asked directly.

None of this requires knowledge of Tversky and Kahneman. It requires having noticed that going first works, which most people who are good at getting their way have noticed without being able to say why.

Can You Undo an Anchor Once It Is Set?

Partially, and the evidence is fairly specific about how.

What does not work is arguing against the anchor, because engaging with it keeps it in the room and keeps you generating thoughts inside its frame. Every minute spent explaining why their figure is wrong is a minute spent thinking about their figure.

What worked in Galinsky and Mussweiler's experiments was deliberately bringing to mind information inconsistent with the anchor. Your own target. What you would accept if this conversation had never happened. What your alternatives actually are. That is not a debating technique. It is a change in what you are retrieving from memory, which is the level the bias operates on.

Applied outside negotiation, the equivalent is having a standard that was decided somewhere else, at a calmer time, ideally written down. A written line from six months ago is a competing anchor with a real advantage, which is that it was set by you when nothing was at stake.

How Do You Set Your Own Anchor?

Decide before, not during. Any standard you work out inside a difficult conversation has already been shaped by that conversation. Work it out beforehand and the anchor is yours.

Write the line down. Not as a threat to deliver. As a fixed reference you can check against later, when the situation has moved and your sense of normal has moved with it.

Compare to acceptable, not to worse. When you catch yourself measuring against a severe example, stop and ask the original question instead. Would I accept this if it were the first thing that ever happened.

Treat the early period as data, not as the true version. The attentive opening months were real, but they were also the anchor. Someone's behavior under strain, over time, is a better estimate of who they are than their behavior while being chosen.

Go first when it matters. If there is a conversation where a standard is going to be set, setting it yourself is worth more than arguing well afterward. The correlation of .85 between opening and settlement is not about eloquence.

One honest limit. None of this makes you immune. Tversky and Kahneman's participants watched the wheel spin and were moved anyway. The realistic aim is not to be unanchored. It is to make sure at least one of the anchors in the room was placed by you.

What Else Do People Ask?

What is anchoring bias in simple terms?

It is the tendency for the first value you encounter to drag your later judgments toward it. Once a number or a standard is in the room, everything after it gets measured as a distance from that starting point rather than assessed on its own. It works even when the anchor is obviously random.

Does anchoring still work if you know about it?

Largely yes. In the original study the anchor came from a wheel of fortune that participants watched being spun, so its irrelevance was visible, and it still moved answers by around twenty points. Awareness reduces the effect somewhat but does not remove it, which is why it is treated as a robust bias rather than a simple mistake.

How does anchoring apply to relationships?

The anchor is how you were treated at the start. If the early months were attentive, that becomes the reference point and later coldness reads as a temporary dip. If the early months were already difficult, that sets a low baseline and genuinely bad treatment can read as normal, because you are measuring from where you started rather than from what is acceptable.

Why does hearing about someone worse off make me stay?

Because it resets your comparison point. A more extreme example becomes the anchor, and your own situation is then judged as the distance from that, which makes it look mild. The question quietly shifts from whether this is acceptable to whether this is as bad as that, and those have very different answers.

How do you protect yourself from an anchor?

Decide your own reference point before you are in the conversation, and write it down. Galinsky and Mussweiler found that focusing on information inconsistent with the other side's anchor removed its advantage, so the practical move is to bring your own standard rather than arguing against theirs.

Sources: Tversky, A., and Kahneman, D. (1974), "Judgment under uncertainty: Heuristics and biases," Science, 185(4157), 1124 to 1131. Galinsky, A. D., and Mussweiler, T. (2001), "First offers as anchors: The role of perspective-taking and negotiator focus," Journal of Personality and Social Psychology, 81(4), 657 to 669, via PubMed (pubmed.ncbi.nlm.nih.gov/11642352).

For self-reflection and educational purposes only. Nothing here is a diagnosis or a substitute for professional mental health support. Only a qualified clinician can assess or diagnose any condition. If you feel unsafe, please contact a local domestic abuse or crisis helpline.