Why do I feel more attached to people who treat me badly than to people who are good to me?

Why Does a Kind, Steady Partner Feel Boring or Flat to Me?

Because a steady partner produces almost no movement, and movement is the thing you have been reading care off. Underneath that answer sits the question this whole section is built on, so it is worth asking straight: are intensity and closeness the same measurement?

They are two measurements, and the whole pattern runs on treating them as one.

Intensity is how much a relationship moves you. Closeness is how much of you is known, and how safely. Those come apart far more easily than it feels like they should. A relationship that swings between cold and warm generates enormous intensity, because it supplies stakes, uncertainty and contrast in quantity, and it can do that while almost nothing about you is actually known to the other person, because a person who goes cold for thirteen days is a person you learn to show less to. Meanwhile a steady relationship can carry a great deal of closeness with very little intensity, for the simple reason that nothing in it is at stake from one day to the next.

Fireworks are brighter than a porch light, and nobody has ever read a book by them.

If the only gauge you have for a relationship is how much it moves you, then the difficult person will win every comparison you run, because they have been supplying movement, and the kind person will read as almost nothing, because steadiness produces almost no movement by definition. A speedometer reads speed. It cannot read direction. You can drive very fast toward nowhere, and the dial will report that you are doing extremely well.

There is research on the cost of the mixed relationship specifically, and it points somewhere uncomfortable. Holt-Lunstad, Uchino and colleagues studied what they called ambivalent ties, people in a person's life who are a source of both support and upset, and found that time spent with those people was associated with higher physiological reactivity, measured as ambulatory blood pressure, than time spent with people who were simply negative. A relationship that is sometimes very warm and sometimes very cold, in other words, appears to cost the body more than one that is reliably bad. One note on the citation: those names belong to one research group, so this is one line of evidence rather than two, and blood pressure is a modest proxy for what a relationship costs, gathered in fairly small samples. The finding is suggestive rather than settled. It does line up with what the intensity itself has been telling you, which is that this relationship moved you more than any other. Movement is expensive. That was never in doubt. What was in doubt is whether movement was ever the thing you were measuring for.

If most of what you can list about the difficult person lands in the left column, you have been running a very accurate speedometer. It has simply never been asked the right question. The page on when the early feeling fades covers the same two measures from the other end, inside a relationship that started intense and settled.

What Are You Actually Attached To?

To the ending of the distress. The person is where the ending came from, and that is a different thing to be attached to than the person, even though from the inside they feel identical.

Here is the test that separates them. Think about the moment the warmth returned, the most recent time. What did you feel first? For most people running this pattern the honest answer is release rather than joy: the unclenching, the breath, the feeling of a weight coming off. Joy comes later, if it comes, and it is quieter. What you were most attached to, in that moment, was the weight coming off. And the weight was put there by the same person who took it off, which is the whole structure of the thing, because it means the only person who can produce your strongest feeling is the person producing your worst.

How can you tell whether you miss the person or the unpredictable relief they provided? Ask the question above and watch which one arrives first, then check the arithmetic afterward: if the size of the missing tracks how long the last cold stretch ran, what you are missing is the ending of a wait. There is measured work behind why the two can be pulled apart at all. Laboratory research on reward established that wanting and liking come apart, that how hard something is pursued and how much pleasure it delivers rest on separable systems, so the size of a pull carries no reading of how good the thing pulled toward was. I am naming the model rather than the researchers here, because the page on buying things when upset already carries their names and the detail of the work, and the limit is the usual one: that research was done on rewards in a laboratory, and nobody has run it on somebody you love.

Follow that through and something important falls out of it. If what you are attached to is the ending of a distress, then a person who never causes you distress cannot compete, and it has nothing to do with how good they are. They have no weight to lift. They can be kind for years and never once produce the feeling of the weight coming off, because they never put it on. The comparison you have been running, between how much the difficult person moves you and how little the steady one does, was a comparison between a lifted weight and no weight at all rather than between two people, and no weight always loses that comparison, and always will, until the measure changes.

None of this says the feeling was not love. Love and a schedule can run inside the same relationship at the same time, and often do, and this page has no way of knowing which parts of yours were which. What it can say is that the size of the pull settles nothing either way, because unpredictability alone produces pulls that size in every learner ever tested.

There is a version of this pattern where the poor treatment matches a verdict you already hold about yourself, and staying with it is easier than the open question of being alone. That is a different root with a different page, feeling unworthy of love, and if the phrase "a match is easier to hold than a contradiction" lands harder than anything here, that page is where to go.

Why Am I More Attached to Someone Who Treats Me Badly Than to Someone Kind?

Because the two of them were never running the same schedule, and a schedule is what builds a pull. That is the whole answer, and it belongs in one place, because the sections above build it a piece at a time.

The person who treats you badly supplies cold stretches of unpredictable length, followed by warmth that arrives on no rhythm you can learn. That is the exact arrangement that produces persistent reaching in every learner it has been tried on. The kind person supplies warmth that arrives when you reach for it, which is the arrangement that produces a weaker pull and a quicker fade. Neither of those facts describes how good either person is. They describe two delivery patterns, and the delivery pattern is doing most of the work.

The clearest human demonstration of the difference is old and narrow, and both of those words matter. Humphreys ran it in 1939, in the Journal of Experimental Psychology, conditioning an eyelid response and finding that when the reinforcement had arrived on only some of the trials, the response took markedly longer to fade after reinforcement stopped than when it had arrived on all of them. Read that for what it is. An eyeblink, in a laboratory, in a study older than almost everybody reading this, and later human work found the effect real but more variable and sensitive to what participants had been told about the schedule. Nobody has run it on a relationship and nobody could. What it gives you is a demonstration that the schedule effect reaches people at all, which is exactly as much as this page claims from it.

There is a second thing loading the comparison, and it has nothing to do with either person. Memory keeps the returns and loses the cold. The returns were events, with a shape and a feeling and often a place attached, and the cold stretches were long, featureless, and made of nothing that files itself. So when you set the two people side by side, the difficult one is scored on the returns, which is the best material anybody could ask for, and the steady one is scored on the whole run of days, because a run of days with nothing wrong in it leaves memory nothing to pick out. A fair judge would return the wrong verdict on that evidence. The comparison was decided before you ran it, by what got kept rather than by what either person did.

None of that settles which of the two is better for you. That question is real and it is yours, and the rest of this page is about the five different reasons a pull like this forms, because what to do next depends entirely on which of them is yours.

What If This Was Simply an Intense Relationship That Ended Badly?

Then that is what it was, and it belongs first, before any of the mechanisms, because not every painful relationship was a pattern and a page like this one can make everything look like one.

Two people can be intense together and fail. They can have a real thing, with real warmth that arrived mostly when you reached for it, and still end badly, through circumstance, through a mismatch neither of them could fix, through one hard year that neither of them survived as a pair. What that leaves is grief, and grief for an intense relationship is intense. It produces missing that feels disproportionate. It produces a pull. And it can make a kind person who arrives afterward feel flat, for the simple reason that you are still in the middle of something else.

Here is how this root reads differently from the ones below, so that you can check. The warmth in a relationship like this was mostly predictable. You knew, more or less, where you stood, and the cold stretches, if there were any, were rare and had reasons. The cruelty, if there was any, was occasional rather than rhythmic. And the pull is fading on its own, unevenly, but in one direction, with no returns of warmth to feed it. If that description fits, you are grieving a real thing that ended, and what you need is time and probably company, and none of the machinery below applies to you. The page on what happens after it ended covers that territory properly, including how long the body takes.

The reason this root goes first is that it is the generous reading, and generous readings deserve to be tested before suspicious ones. Somebody who has just been through a hard ending deserves to be asked whether they were running a schedule, and then believed when they say no, rather than told.

What If Intensity Is the Only Measure of Care You Have Ever Used?

Some people arrive at this pattern with no schedule in their history at all. What they have instead is a gauge.

Somewhere along the way, and it can be as simple as the first serious relationship having been a dramatic one, the amount a relationship moved you became the measure of how much it mattered. Big feelings meant a big thing. Small feelings meant a small thing. And that gauge was never checked against anything else, because it did not seem like a gauge. It seemed like the feeling itself.

A person carrying that gauge will read the difficult person as more significant on first contact, with no training by unpredictable warmth required, because the difficult person produces more movement, and they will read the kind person as less significant for the same reason. No schedule is needed. The pull here is a measurement error rather than a conditioning effect, and it feels identical from inside.

Here is how it reads differently. The timeline test above comes back weak: the warmth was fairly predictable and the pull is still large. Intensity has been the deciding factor in every relationship you have chosen, and not only the romantic ones. The friendships you rate highest are the dramatic ones. You have described a calm stretch, in any relationship, as something being missing. And a steady person has felt flat to you before this one, more than once, with different steady people.

That last item is the tell. A schedule produces flatness toward one particular kind person, the one who arrived after a particular difficult one. A gauge produces flatness toward all of them.

If this is yours, the mechanism section of this page is interesting and beside the point. What needs work is the gauge, and the honest news is that a gauge changes through use rather than through insight. You start measuring by something other than movement, badly at first, over months, with the old gauge complaining the whole time, until the new one becomes readable. The table of intensity and closeness markers above is one place to start reading from. The page on attracting the same people covers a different route into the same flatness, where the calm person feels flat because a familiar shape is missing rather than because the gauge only reads movement, and the two are worth telling apart.

What If the Shape Was Already Familiar Before You Met Them?

Some people met this schedule long before they met this person.

If warmth, early in your life, arrived on a rhythm you could not predict, if the people whose care you depended on were sometimes fully present and sometimes gone, with no signal in advance and no explanation after, then you learned the schedule as the shape of care itself. You learned that reaching sometimes works. You learned that a gap proves nothing and that the return, when it comes, is the best feeling there is. You learned it thoroughly, at an age when learning is thorough, and you learned it about care in general rather than about any particular person.

An adult who runs the same schedule then feels, from inside, less like a warning and more like recognition. This is what care feels like. The unpredictability itself registers as familiar, and familiar registers as right, and a kind person whose warmth is reliable registers as foreign, because reliability was never part of the shape you know. Their steadiness is a language you were never taught, and it reads as an absence of language.

Two things need saying about this root, and both matter. The first is that having it makes you neither damaged nor unusual. It makes you somebody who learned early what many people learn later, and who therefore had the pattern ready before an adult arrived to run it. The page on growing up with an unpredictable parent covers the learning itself in depth, from the child's side, and this page is about the adult pull that results, so I will hand the early part to it rather than repeat it. The second is that this root does not have to be present for the pull to form. The schedule works on everybody. Plenty of people with entirely predictable childhoods walk into an unpredictable adult and come out with the same pull, because the principle was demonstrated on creatures that had no childhood to speak of. Familiarity makes the pull faster and the steady person more foreign. On its own it creates nothing.

Did growing up with an unpredictable parent teach you to read inconsistency as love? Partly, on the evidence, and less completely than the question assumes. Fraley pooled the longitudinal studies available at the time, in 2002, in Personality and Social Psychology Review, under the title Attachment Stability From Infancy to Adulthood, and found continuity between early attachment and later attachment that was real but modest, closer to a tendency that can be revised than to a setting fixed at the age of one. The limits are large enough to state alongside it. There were few studies to pool, the samples in them were small, and what gets measured in an infant and what gets measured in an adult are not the same measure, so the continuity figure underneath that conclusion is soft in both directions.

Here is how it reads differently. The timeline test comes back strong, and it would come back strong for earlier relationships too. The steady person feels foreign rather than flat: less boring than somehow wrong, as though they were doing care incorrectly. And the return of warmth after cold has a quality of homecoming to it that is hard to explain to anybody who did not grow up inside that rhythm.

If this is yours, the pull toward this particular person is the least of it, and the page can only point. What needs finding is the earlier schedule, and then the slow business of learning that reliability is a form of care rather than the absence of one, which takes longer than anybody wants and works. The page on anxious attachment covers the standing forecast that often grows out of the same early rhythm, and the two travel together more often than either travels alone.

Why Does Your Explanation Get Thinner the Longer You Talk?

Because most of what you had was a feeling of understanding rather than an understanding, and feelings of understanding do not survive being asked for details.

Rozenblit and Keil showed this in 2002, and the study is worth describing because it has nothing to do with relationships, which is why it is trustworthy here. They asked people to rate how well they understood ordinary objects, zippers, flush toilets, cylinder locks. People rated themselves fairly high. Then they asked them to write out, step by step, how the object actually worked, and to rate their understanding again. The ratings dropped, consistently, and the drop was the finding: people carry a sense of understanding that is considerably larger than the understanding itself, and the gap only becomes visible when they are made to produce the explanation. They named it the illusion of explanatory depth.

Applying that to explaining a person to your friend is my extension, and the original researchers were working on zippers, so weigh it accordingly. But the shape matches too well to leave out. You have a strong sense that you understand this person and why the relationship is what it is. That sense has been built mostly from the returns of warmth, which were vivid, and from a great deal of thinking in the gaps, which felt like understanding. When you are made to say it out loud, in order, to somebody who was not there, the sense has to become an account, and the account is thinner than the sense. You can hear it thinning. Your friend can hear it too, and is usually kind enough to say nothing.

What that thinning is telling you is that the account came second. The pull was built on the schedule, on the relief, on the gauge, on the shape, on whichever of the roots above is yours, and none of those are the kind of thing that fits into an explanation to a friend, because none of them are about the person's qualities. The account was assembled afterward, to make the pull make sense. When you try to say it, you find out how much of it was assembled.

That is worth knowing for its own sake, and it is also useful. A pull that gets thinner when described is a pull that was built by something other than what it claims. That is the first honest reading you can take, and it costs one conversation. The page on not being able to stop thinking about someone covers the version of this that forms on thin information, before there is any relationship to explain, and the two are different in what the friend is being asked to believe.

Which Part of This Do You Actually Choose?

Not the pull. That needs saying without qualification, because people running this pattern spend a great deal of energy blaming themselves for a feeling that a schedule produced, and the blame changes nothing while costing a lot.

The pull is produced. Unpredictable warmth builds persistent reaching in every learner ever tested, and it built it in you, and the fact that you can now describe the mechanism does nothing to the mechanism. Relief will still be louder than contentment. The return of warmth will still land as the best you have felt in weeks. The kind person will still, for a while, read as almost nothing. None of that is chosen, and treating it as chosen is a way of adding a second weight to the first.

What is chosen is smaller and it is the whole workable surface. It is what happens in the hour after the warmth returns.

In that hour there are three decisions, and they are small enough to actually make. Whether you respond in the first minute or wait until the relief has finished being the loudest thing in the room. Whether you read the return as a verdict on the relationship, proof that it was real all along, or as one more unpredictable reward arriving on schedule, which is what the timeline says it is. And whether you tell anybody, because a return that gets described out loud to a friend, in the hour it arrives, is a return that has to survive being made into an account, and you already know what that does to accounts.

None of those three tell you to leave, or to stay, or to cut off contact. This page prescribes none of that, and would distrust itself if it did; the question of whether to leave has its own page, with its own honesty about how little anyone else can decide it for you. What the three decisions do is put a gap between the relief and the response, and the gap is where the schedule starts to lose its grip, because a reward that has to wait an hour to be collected is a reward that gets examined on the way. The pull arrives on its own. The hour after it is yours.

What Does This Look Like Once It Has Stopped Costing You?

Less dramatic than the pattern would have you expect, which is itself part of the answer.

Intensity still registers. A person who is warm on no schedule still produces a pull, because the mechanism has not gone anywhere and never will. What changes is what the pull is read as. It stops being the verdict and becomes a reading, the way a speedometer reading stops being an opinion about the journey once you know what a speedometer measures. You notice the pull, you know roughly what built it, and you have a fair idea of how much of it is the person and how much is the schedule, and the knowing changes what you do in the hour even while it changes nothing about the feeling.

The relief when somebody difficult is warm again gets smaller, and the way it gets smaller is instructive: mostly by your no longer waiting for it. A strain you were not counting the days of ends less loudly. The cold stretch stops being a cold stretch and becomes, slowly, just a stretch, and a stretch that ends produces a much smaller signal than a wait that ends.

And the steady person becomes readable. Not intense, because they were never going to be, and not suddenly thrilling, because that would be the old gauge reporting again. Readable. On the other measure, the one that runs on what you can say to them and what happens after you say it, they start returning a signal, and the signal grows over months rather than weeks, with the old gauge complaining the whole time that nothing is happening. Sometimes what emerges is a real fit. Sometimes it is a genuine no, arrived at honestly, on the right measure. Both count as this pattern having stopped costing you.

None of it requires the difficult relationship to have been fake, or the feeling to have been wrong, or you to have been foolish. It requires a second gauge, read for long enough to trust.

How Long Does This Take, Honestly?

Longer than the pattern's own timeline, and that is the schedule again rather than a failure of yours.

Two clocks run here, and they run at different speeds. The first is the pull toward the specific person, and this is the clock the reinforcement principle describes directly. Behavior learned under unpredictable reward is, by the principle's own definition, the behavior that takes longest to fade once the reward stops, because a gap never proved anything before and does not start proving anything now. Worse, it usually intensifies briefly first. The weeks right after the last return of warmth are, for most people, the worst of the whole pull, arriving exactly when they expected relief, and that intensification is one of the most reliable findings in the field rather than a sign that they cannot live without the person. It passes. Then the decline is uneven, over months, with sudden returns triggered by anything that resembles the old signal, a name, a place, the particular hour of the evening when the messages used to come. It fades. It fades slowly and in steps, and each step down feels like a plateau until the next one.

The second clock is the gauge, and this one runs slower still, because a gauge changes through use rather than through insight. Reading a steady person on the closeness measure instead of the intensity one is something you do badly for a long time before you do it well, and the first months of it are mostly the sensation of nothing happening. People who have done it describe somewhere between a season and a year before the second measure feels like it is reporting anything, and longer before they trust it over the first.

I would rather give you those numbers than the ones the pattern hands out, which are that it should have faded by now and that the fact it has not means something. It means the schedule was doing exactly what schedules do. That is all it means.

What Is Worth Doing This Week, Before You Decide Anything About Anybody?

The timeline, and nothing else.

Ten returns of warmth, in order, in writing. For each one: how long the cold before it lasted, whether you could have predicted the day, and how big the relief was, on a scale you would not show anybody. Change no behavior while you do it. Do not decide anything about the relationship, do not go quiet, do not make any announcement, and above all do not use the timeline as ammunition in a conversation with the person, because it is a reading of you rather than a case against them.

At the end you will have a piece of paper that says one of two things. Either the returns land on no pattern and the relief scaled with the wait, in which case the schedule was doing most of the work, and you now know the size of the pull was measuring the wait and never the person. Or the warmth was fairly predictable and the pull is still this large, in which case the schedule is weak and one of the other four roots is where the pull lives, and the timeline has just saved you from working on the wrong thing.

Then take the second reading, the one that costs a conversation. Explain the person to a friend, out loud, and listen for the thinning. If the account holds, that is information. If it thins, that is information too, and it is the more common result.

If both readings come back saying the schedule and the relief, the page has done what it can, and what remains is the hour after the next return, which is yours. If they come back pointing underneath, at the gauge, at the relief you have come to need, at a shape you knew before you met them, then what needs work sits below the relationship rather than in it, and no amount of managing your behavior toward one person will reach it.

Our Sacred Valley seminar works at that layer. We do not start with the person, because you can already describe them, and you have just heard what happens to that description under pressure. We start with which of the roots on this page is yours, because a principle can be looked up in a minute and the root cannot, and as Florian puts it, "once you know the real reason, you can actually get the solution based on that reason." A solution built on the wrong root is the one you have probably already tried. Then we look at what stands in the way of the second gauge, and build a plan you are held accountable to afterward, back in your own week, with your own phone, on the evenings when the old signal used to arrive. I will say the boundary one more time because it belongs here as much as at the top: if you are being frightened, threatened or hurt, that belongs with a licensed professional or a domestic abuse service today rather than with a seminar or a page, and nothing here asks you to stay or to understand it first. Everybody is screened before arriving, which keeps the group small enough that somebody who has spent years reading care off the size of a relief gets worked with properly rather than handed a worksheet. One limitation worth stating plainly: exact dates, the price, and the final cap on group size are still being settled, so anybody comparing programs on those numbers this week cannot yet compare ours. Ask every program you are weighing the same questions, this one included, and decide from what comes back.

The timeline costs an evening and a piece of paper, and it is worth having whichever way it comes out.