Why do I fall for the idea of someone before I know who they are?

Why Is the Picture Stronger When You Know Less?

Because you are supplying the missing parts, and you supply them from your own preferences rather than at random.

This is the mechanical heart of the whole pattern and it runs directly against how it feels. It feels like the strength of the picture is evidence about the person, as though something remarkable in them produced something remarkable in you. Read from the outside, the strength of the picture tracks the size of the gap. Thin information is the necessary condition rather than an unfortunate accident of timing.

Consider what a gap does. When a detail about somebody is unavailable, you do not leave a blank in the picture and walk around it. Something goes in. And what goes in is drawn from the material you have, which is you: your preferences, your idea of what a good person is like, and whatever you are currently short of. The result is a person assembled to your specification, from a very small number of genuine parts, presented to you by your own equipment as though it had been observed.

Which produces the strangest feature of this whole thing. The imagined version is more specific than the real one. You can describe how they would be in a crisis, how they would handle your family, what kind of quiet they keep in the mornings. You cannot describe what they think about almost anything, because they have not told you, and the parts they have told you are the parts you think about least.

This also explains why real contact is so often a small letdown. Each actual detail replaces a supplied one, and the supplied ones were selected to suit you while the real ones were not. The direction of travel is always the same. More information, weaker picture. None of that is a bad sign about them, and none of it is a failure of the connection. The whole thing is arithmetic.

A short note on how solid this ground is. The research on assumed similarity is largely correlational, and there is a genuine difficulty inside it that honest researchers name themselves: people do pair up with people somewhat like them, so a study that finds you think a partner resembles you cannot fully separate accurate perception of real similarity from projection. The effect survives that objection reasonably well, and it is not settled to the last decimal. Treat it as a mechanism you can test against your own list rather than as a law.

Does Building Someone Up Help a Relationship or Damage It?

The evidence goes in both directions, and the pages that tell you confidently either way are more confident than the research is.

Here is the honest state of it. Reviews of bias and accuracy in close relationships, including the integrative review published by Gagné and Lydon in 2004, find that positive bias about a partner is widespread and is frequently associated with better outcomes: higher satisfaction, more stability, and partners who over time grow somewhat toward the flattering picture rather than away from it. That is a real body of work and it is why any page telling you that idealization is simply a mistake to be corrected is overstating.

The same literature contains findings that cut the other way. Huston, Niehuis and colleagues reported in 2001 on a longitudinal study of newlywed couples followed over many years, and found something that people still find counterintuitive: the couples who divorced early had been more affectionate and more romantic as newlyweds than the couples who stayed married. What separated them was the slope. A steep decline in affection over the first years, which the researchers described as disillusionment, distinguished the early divorces better than the amount of conflict did.

Those two results are not in contradiction. Read together they suggest something more precise, and more useful, than either headline. Seeing somebody well appears to help. Seeing somebody well while also knowing them accurately appears to help more. The version that costs is the one where the flattering picture and the real person keep coming apart, and the gap has to be closed by repair work over and over again.

Which gives you a genuinely useful distinction to carry, and it is not about how positive your view of somebody is:

• Idealization that survives contact. You think highly of them, real information keeps arriving, and the high opinion accommodates it. They are worse at some things than you thought and you find that mildly endearing. The picture bends.
• Idealization that a real person keeps puncturing. Real information arrives and has to be reinterpreted, explained away, or held at a distance for the picture to survive. The repairs are frequent and each one is small enough not to notice.

The first is fine and possibly good for you. The second is expensive, and the cost is paid in the fourth month rather than the first week.

Two honesty notes belong here, because this is the section where it would be easy to sound more certain than anybody should. Nearly all of this research is self-report and correlational, which means it cannot settle direction. Happier couples may idealize more because they are happy, rather than being happy because they idealize. And the longitudinal work has its own limits: it followed one cohort recruited in a small number of American counties in the early 1980s, and the popular retelling of it quietly drops the fact that plenty of extremely affectionate newlyweds in that same sample stayed married.

Why Does a Short First Conversation Feel So Predictive?

Because to some degree it is, and because you then extend it enormously past what it can carry.

This is worth knowing because it stops the whole page from being a lecture about your imagination. Sunnafrank and Ramirez published work in 2004 on brief get-acquainted conversations between strangers who had never met, some of them running only a few minutes. What they found was that the forecasts people made immediately after that first conversation, about what the relationship would become, predicted where those relationships actually stood some nine weeks later.

Your instinct after three good conversations is doing something real. That deserves saying plainly, because most writing on this topic treats early conviction as pure delusion, and it is not.

The limits matter just as much. The finding is correlational and it is partly self-fulfilling, since somebody who forecasts a friendship warmly will invest more in it, which helps produce the friendship. It was a university sample doing a structured task. And most importantly, what it shows is that a forecast predicts an outcome, which is a different claim from saying the forecast was accurate about the person. A short conversation can genuinely tell you something about rapport. It cannot tell you how somebody is about money, or about being contradicted, or at two in the morning in a corridor. Those parts get supplied by you, and they arrive feeling exactly like the parts you observed, with no label attached to say which is which.

That is the actual trouble. Not that you construct, but that the constructed material and the observed material feel identical from the inside.

Why Does the Disappointment Feel Like a Betrayal When Nobody Did Anything?

Because a real loss happened and nobody was responsible for it, which is a combination the mind handles badly.

Something you had been living inside for six weeks stopped being available. That is a loss by any ordinary measure, and the body treats it as one whether or not the thing lost was ever real. The flatness is genuine. The reluctance to make plans for a while is genuine. None of that requires the picture to have been accurate.

What produces the specific sense of having been misled is the mismatch. You had a person in your head with certain qualities. The person in the room has different ones. Something feels like it was taken, and the only candidate for having taken it is them, so a low-grade grievance attaches itself to somebody who behaved reasonably throughout.

Say it plainly, because it is the sentence people need and rarely get: they never claimed to be it. They gave you an ordinary amount of true information about themselves. You built something on the far side of a gap they had no access to and could not have corrected even if they had wanted to.

That is also why this disappointment is so hard to describe to a friend. Every account of it makes you sound either unfair to them or foolish about yourself, when neither is right. The accurate sentence is that you are not disappointed in them, you are disappointed that they turned out to be a person, and a person is what everybody turns out to be.

There is a version of this where the reaction is not the slow deflation described here but a fast physical refusal, a drop in your chest at some small witnessed detail, and everything about them becomes suddenly repellent. That is a different mechanism with different roots, and the page on suddenly going off somebody you were attracted to covers it properly. The two do sit next to each other in time, and for some people the fast version is what ends the constructed one. The distinction that separates them is straightforward: this page is about the picture becoming impossible to maintain across weeks, and that page is about the body issuing a refusal in under a second at something you actually saw.

What If This Is Ordinary Early Attraction Doing Exactly What It Does?

This is the most likely answer for most people reading this, it goes first, and it gets the most space, because most people arrive here carrying a verdict they did not earn.

Early attraction assembles people. That is what it does. It takes a small number of real signals and produces a coherent, detailed, emotionally weighted picture at speed, and it does this in people whose lives are entirely in order, who have good judgment about everything else, and who have never once been described as naive by anybody who knows them. Nothing about that is a malfunction of the process. That is the process.

The reason this surprises people is that adults almost never describe it accurately to each other. What gets said out loud is a tidied version: we got on well, I thought there might be something there. What actually happened, on the inside, was a coastline in September and a hospital corridor at two in the morning. Everybody does the second thing and reports the first, so each person meets the full force of their own version believing they are unusual.

In this root, the construction is doing nothing except what early attraction does, and it corrects itself. That is the whole tell and it is worth stating carefully, because the correction is easy to miss while it is happening.

The signs of this version are specific and reassuringly dull.

The picture bends when real information arrives. They turn out to be worse at plans than you assumed, and instead of reinterpreting it, you simply update, and the update costs you nothing. The intensity is highest in the first two weeks and measurably lower by the sixth week without any effort on your part. Nothing in your life has narrowed. Your friends see you and you are actually present. And crucially, you can imagine the whole thing ending without dread, because a version of your week exists that does not have them in it.

If that is you, there is nothing here to work on, and the most useful thing this page can do is take away a verdict you were handed by other people's writing. The willingness to be moved quickly by somebody counts as one of the more generous things a person can do, and the internet has spent about eight years telling anybody who does it that they have an attachment problem.

One honest note attached to this root, since it is the one most people will land on. Being in it does not protect you from the third version further down. Circumstances change, and a construction that was pure attraction in March can be doing an entirely different job by October with no announcement of the switch.

What If the Idea Is Simply More Manageable Than a Person?

This is the version that costs the most, and it is the one people recognize with the greatest reluctance.

Everything above describes a picture that gets built by accident. This root describes a picture that is preferred. Not consciously, and not as a decision, but the preference shows up plainly in the behavior once you know what to look for.

An imagined person makes no demands. They have no mood on a Tuesday. They do not want something different from what you want on a Saturday. They never need to be accommodated, never require an apology, and above all they cannot be disappointed in you. That last one is the engine. A real person will eventually see you at close range, including the parts you do not show, and will have a reaction to what they see. The constructed version never gets to that point, so the risk never arrives.

The tells here are behavioral and they are quiet.

The intensity is highest when contact is lowest. A gap of two weeks does not weaken it and often strengthens it, which is the reverse of what happens in the first root. Real meetings produce something like mild anxiety rather than anticipation, and there is often a small relief when one gets canceled. You find reasons not to close the distance that sound practical and are always available. And when somebody does become properly available, warm, clear, and interested, the whole thing deflates in a way you cannot explain to yourself.

That deflation is the diagnostic moment, and it is worth sitting with rather than explaining away. If certainty and availability reduce the feeling rather than increasing it, then what was being pursued was never the person.

This is the version where the pattern can run for years without anybody noticing, including you, because from the outside it looks like bad luck with timing. It is also the version where nothing on this page will be enough on its own, because the thing being avoided is not a habit. Naming it accurately is a real start and it is not the finish.

What If the Speed Is the Whole Point?

This root is the least discussed and it explains a specific version of the pattern that the other four do not touch.

Here what is being pursued has nothing to do with the person or with the life. What is wanted is the evidence that you can still feel something that size.

It shows up in people who have had a long flat stretch. A few years where nothing much moved. A relationship that ended slowly enough that the ending was mostly administrative. A period of being competent and busy and not particularly alive. Somewhere in there a quiet question forms, usually never spoken: whether the capacity for a large feeling is still intact, or whether that was something that belonged to being younger.

A fast, enormous, three-conversation certainty answers that question decisively and immediately. It is proof. And the proof is the product, which is why the intensity in this root is oddly independent of the specific person, and why it can arrive twice in one year aimed at two people who have almost nothing in common.

The tells are worth knowing because this root is frequently misread as the first one.

The speed itself is the part you retell. When you describe it to a friend, the sentence that comes out first is about how fast it happened rather than about them. The feeling is somewhat detachable from the person, and if you are honest you can notice that the target could have been somebody else that month. There is a background relief in it that has nothing to do with romance, a sense of there it is, still working. And the aftermath is disproportionately flat, because what is lost when it ends is not the person but the proof, and the question quietly reopens.

Worth saying clearly: the capacity was never in doubt. Large feelings are not scarce and they are not evidence of anything about your character. It is also worth noticing that this particular proof is much easier to obtain from somebody you barely know, since a real person, by being specific, slows the whole thing down.

Why Does Deciding to Slow Down Never Do Anything?

Because the assembling is not running at a speed you have access to.

This is the standard advice, it is given constantly, and watching it fail repeatedly is one of the reasons people conclude something is wrong with them. Get to know them first. Take it slowly. Wait three months before deciding anything. The instruction is not unreasonable and it aims at the wrong layer.

By the time you could decide to slow down, the construction has already happened. It happens in the gaps between the conversations, at night, in transit, while doing something else. It uses material you did not consciously select and it presents its results as observations rather than as inventions. There is no moment where you are offered the choice to build less.

There is a measured version of this. Winter and Uleman published When Are Social Judgments Made? Evidence for the Spontaneousness of Trait Inferences in the Journal of Personality and Social Psychology in 1984, and what it established is that a judgment about what a person is like forms at the moment their behavior lands, with no intention to judge anybody and no sense afterward of having judged. The limits are plain. That was laboratory work built on written material rather than on anybody meeting a person over three conversations, and one inference about one trait sits a long way below a whole assembled life. What it settles is the timing, and the timing is earlier than any resolution to be careful can reach.

What you can do, having decided to slow down, is add a second track: a running commentary on how you are getting ahead of yourself. That does not brake anything. What it adds is company. The construction continues at full speed with a critic attached, and the critic makes the whole experience slightly worse without altering the output.

The same applies, with more force, to the checklist approach. Learning a list of warning signs about other people does nothing here, because the problem is not that you failed to detect something about them. You detected an ordinary amount. You then supplied the rest, and no checklist about their behavior has any purchase on material generated on your side.

And the worst of the standard responses is being told that none of it was real. All of it was real. You felt it, it occupied your week, and the reality of a feeling has never depended on the accuracy of the belief that produced it. Anybody who collapses those two things will lose your trust and should.

Why Do You Do It Again While Watching Yourself Do It?

Because noticing and stopping happen at different levels, and noticing has almost no braking effect on anything.

This is the part people find most disorienting about themselves, and it is worth normalizing properly. The second time, or the fifth, you can see it clearly while it happens. You know you are constructing. You can even identify which items on the right-hand column are yours. And the feeling proceeds at full strength anyway, with the commentary running alongside it like subtitles on a film you cannot pause.

Nothing has gone wrong when this happens. Insight is not a control mechanism. Knowing why you are doing something and being able to not do it are separate capacities, and they are located in different places. This is the same reason people can describe their own procrastination with complete accuracy at eleven at night while continuing to do it.

There is a specific trap that arrives at this stage, and it is the reason the phrase at the top of this page is worth taking seriously. When you catch yourself doing it again, an explanation is immediately available and it is charming: this is just how I am, I am a hopeless romantic. That sentence closes the question. It sounds like self-knowledge and it functions as a period, because a personality trait is not something you look underneath.

The distinction worth holding onto is between an explanation and a story you can tell yourself so that nothing has to be examined. The first opens something. The second closes it while feeling like the first.

The genuinely useful move at this point is smaller and less satisfying. When you catch yourself doing it again, do not try to stop, and do not adopt a verdict about your character. Write the two columns for this one too. Then put the new right-hand column beside the previous one and see how much of it is the same. That comparison is where the pattern becomes visible, and it is not visible any other way.

Where Does This Stop and Something Else Begin?

Six neighboring patterns get confused with this one, and the wrong name leads to the wrong work.

The thinking that will not stop. If the defining feature is involuntary preoccupation with one specific person, the same seven minutes rebuilt endlessly, a mood that rises and falls with what you read into their behavior, then the subject is the fixation rather than the construction. That is covered on the page about not being able to stop thinking about someone you barely know, which owns the intrusive thinking itself, including the word people use for it. This page owns the picture: what it is made of, where the material came from, and what happens when it stops holding.

The fast collapse. If attraction ends in under a second at a small witnessed detail, with a physical drop and a revulsion you cannot argue with, that has a different mechanism and its own page at suddenly going off somebody you were attracted to. The two are related in time and unrelated in cause: this page describes a picture that becomes unmaintainable across weeks, and that one describes a body issuing a refusal in a moment.

The months-long undefined arrangement. If two of you are plainly involved, seeing each other regularly, and neither has ever put a name on it, the subject is the arrangement rather than the picture, and relationships that never become relationships works it properly.

Being on the other side of it. If you are the one supplying occasional warmth to somebody you have no intention of choosing, that is the opposite chair and it is dealt with on the page about keeping somebody interested with no intention of choosing them.

Authored inner worlds. If the imagined material is built deliberately by you, populated with characters, running on music and pacing, and continuing across years as a story you are writing, that is a different pattern entirely and disappearing into daydreams covers it.

Reading your own feeling onto somebody's face. If the constructions are mostly accusations, and what you are attributing to people is your own current state rather than a hoped-for quality, the relevant page is accusing people of what you are doing yourself. And if the person occupying you is your actual partner and the material is threat rather than hope, jealousy in relationships is the right one.

What Does This Actually Cost, If Anything?

For a good number of people, close to nothing, and an honest answer has to start there rather than manufacturing a cost to justify a page.

Where a cost does appear, it appears in specific places and it is countable.

The last row is the one people underestimate. The construction itself is usually survivable and often enjoyable. The conclusion you draw about your own judgment afterward does more measurable damage than the pattern does, and it is entirely an addition. Nothing in the evidence supports it.

There is one further cost that almost everybody carries and almost nobody names, which is the silence. This is not discussed, so nobody discovers how common it is, and the concealment converts an ordinary process into a private defect. That conversion is done entirely by not saying it out loud, and it can be undone the same way.

How Long Does This Take, Honestly?

It depends entirely on which of the five you are in, and confusing them is why people conclude that nothing works.

If this is ordinary early attraction, the timeline belongs to the attraction rather than to you. Weeks, and it will correct itself whether or not you do anything. The only thing worth changing in the meantime is what you have decided it says about you, and that can change in the time it takes to read a paragraph.

If the gap is being filled with your own material, the mechanism does not go away and the labeling is available immediately. The first two-column exercise takes fifteen minutes and does most of the work. What takes longer is the effect on your choices, which shows up over a year or so, indirectly, and mostly as a reduction in how much gets built before real information turns up.

If the idea is more manageable than a person, this moves slowly, and it does not move through anything you do about one specific person. This is the version where outside help is most likely to be worth having, because what is being avoided is not a habit and it does not respond to being observed.

If somebody is standing in for a life, this moves as the life moves and not before. Months at least. People commonly notice a year later that the pattern went quiet somewhere along the way without being able to say when.

If the speed was the proof, this one can shift quickly, because it depends on a belief rather than on a structure. Once you have noticed that the size of a feeling was never in question, the specific person stops being load enough to carry it.

One thing works immediately in every version, which is separating what you observed from what you supplied. That is available today and it does not require knowing which root is yours.