Skip to the reader
Going Deeper

Going deeper · A reader

The pursuer-withdrawer dynamic

Why one person moves toward when something feels wrong, while the other moves away, and how both responses can keep the same problem going.

About 30 minutes Eleven sections

Why this pattern matters

Some relationship problems are difficult because two people want different things. Others become difficult because of what happens between them while they are trying to deal with the problem.

Something feels wrong. Perhaps there has been a disagreement, one person feels ignored, or an unresolved issue has been sitting between them.

One person wants to engage. They ask what is wrong, raise the issue again, seek an answer, push for a decision, or want the conversation to happen now.

The other person moves away from the interaction. They become quieter, give shorter answers, change the subject, ask for space, or disengage.

Then something predictable can happen. The more one person pursues, the more the other withdraws. The more one person withdraws, the more the other pursues.

Each person experiences themselves as reacting to the other.

The pursuer says

“I only keep pushing because you shut me out.”

The withdrawer says

“I only shut down because you won’t stop pushing.”

Both can be true. Neither is the whole picture.

The original issue is still there, but now there is another problem as well: the way the two people respond to each other once the interaction becomes difficult.

That is the pursuer-withdrawer dynamic.

The cycle

How the cycle works

Relationship researchers have studied a closely related interaction for decades under the name demand-withdraw. In this pattern, one partner presses for discussion, change or resolution while the other avoids, disengages or attempts to end the interaction.

Across many studies, greater demand-withdraw interaction has been associated with poorer relationship and communication outcomes. What makes the pattern particularly difficult is that the behaviours do not occur independently. Each person’s response changes what the other person is responding to.

Imagine a couple with something unresolved between them.

One person raises it:

“Can we talk about what happened?”

The other is reluctant:

“Not now.”

The first person experiences that response as another reason the issue needs to be addressed:

“You always say not now. When are we actually going to talk about it?”

The second person now experiences more pressure:

“I don’t want another argument.”

The first person sees further withdrawal. The second feels further pursuit. Within a few minutes, both are responding not only to the original issue but to what the other person is now doing.

The sequence begins to feed itself.

How the exchange runs

What the other person can see

  1. One · The pursuerRaises the issue again
  2. Two · The withdrawerAnswers briefly, or not now
  3. Three · The pursuerPresses harder
  4. Four · The withdrawerSays less again

And back to one

What they cannot

The pursuerWhat am I trying to get?

The withdrawerWhat am I trying to get away from?

Each person answers the top half. The bottom half varies between people, between relationships and between arguments, and the loop above can run without either person ever seeing it.

If you look only at the last move, the next response makes sense.

The pursuer really did press harder after the other person went quiet.

The withdrawer really did go quieter after the other person increased the pressure.

Each explanation is true, but each begins halfway through the story.

The same moment can carry different meanings

For some couples, the difference can be captured roughly like this.

For the pursuer, the alarm may sound like

“We are not okay.”

Distance, silence, lack of resolution or lack of movement has become the problem.

For the withdrawer, more like

“I am not okay in this.”

The interaction itself has become difficult. Pressure, criticism, helplessness, emotional intensity or the persistence of the conversation makes disengaging increasingly appealing.

These descriptions will not fit everyone, but they show how one conversation can create two different problems.

One person experiences continued engagement as necessary. The other finds continued engagement increasingly difficult.

Their attempts to deal with those different problems collide, and the cycle begins.

Demand-withdraw and pursue-withdraw

Two related ways of understanding the pattern

The research literature most commonly uses the term demand-withdraw. It focuses on observable interaction during conflict: one person pressures for discussion or change while the other resists, avoids or disengages.

In couple therapy, particularly Emotionally Focused Therapy, clinicians often describe a broader pursue-withdraw pattern. Pursuit may include demands and criticism, but it can also involve attempts to restore contact, reassurance or emotional closeness. Withdrawal can include avoidance, but it may also involve shutting down when the interaction becomes emotionally overwhelming.

The two overlap considerably without being identical.

Demand-withdraw research helps explain the structure of the interaction and the way each person’s behaviour affects the other. Pursue-withdraw language gives more room to consider what people may be experiencing within that interaction.

Using both perspectives prevents either account becoming too narrow.

Looking only at behaviour can leave pursuit appearing as pressure and withdrawal as avoidance.

Looking only for attachment fears can obscure other possibilities. A person may be pressing because they want a practical change, an answer or acknowledgement. A person may pull away because they disagree, believe the conversation is futile, need time to think, or do not want the change being requested.

Where attachment fits

Attachment can influence both positions.

Someone who is particularly sensitive to rejection, distance or uncertainty may become more activated when an important relationship feels unsettled. An unanswered message, a changed tone or an unresolved disagreement may carry more weight for them than it does for someone else, and the urge to restore contact quickly can be strong.

On the other side, research has found associations between avoidant attachment and withdrawal during conflict, and between that pattern and lower relationship satisfaction for both partners.

Neither association tells us who will take which role in a particular argument. People who generally feel secure can pursue hard about something they consider important, and someone who usually seeks closeness can withdraw when a subject feels exposing or impossible.

Attachment can form part of the explanation. The relationship, the issue and what each person wants from the conversation also influence the position they take.

Pursuit

What pursuit is trying to do

Pursuit is usually easy to see.

Someone asks again. They raise the issue repeatedly. They press for an explanation. They want an answer, reassurance, acknowledgement, a decision or some indication that something is going to change.

Calling the person demanding or needy describes how the behaviour may feel to the other person, but it tells us little about why it keeps happening.

A person may be trying to restore connection. Something feels distant or uncertain, and they want to know whether the relationship is okay.

They may instead be trying to create movement. A problem matters to them and appears to be going nowhere. They want the issue acknowledged, a decision made, behaviour changed or something resolved.

Both can be present at once.

The difficulty is that pursuit can gradually work against what the person is trying to achieve.

Someone begins with:

“I want us to sort this out.”

After repeated attempts that seem to go nowhere, it becomes:

“Why do I always have to force you to talk?”

“I need to know whether this matters to you”

becomes

“You obviously don’t care.”

“I need something to change”

becomes

“You never listen to anything I say.”

The original concern may still be legitimate. Its increasingly reactive expression can make the other person less willing or less able to engage with it.

Pursuers often feel they have already tried asking calmly. From their perspective, the urgency has a history: earlier attempts did not seem to produce enough response or movement.

Once pursuit intensifies, however, the other person is responding to more than the original concern. They are also responding to the pressure surrounding it.

Withdrawal

What withdrawal is trying to do

Withdrawal is just as visible from the outside.

The person stops talking. They look away. Their answers become shorter. They change the subject, leave the room, say there is no point discussing it, or stop engaging.

From the pursuer’s side, the meaning can seem obvious:

“You don’t care.”

“You don’t want to deal with this.”

“You would rather ignore me than work it out.”

At times that interpretation is accurate. People can withdraw to avoid responsibility, resist a change they do not want to make, punish another person, or escape the consequences of their behaviour.

Withdrawal can also serve other functions.

A person may believe continuing the conversation will make things worse. They may feel criticised and become defensive. They may see no possibility of resolution. They may need longer to think than their partner does. They may disagree with the change being requested and have no idea how to say so without escalating the conflict.

Some people become physiologically overwhelmed as conflict intensifies. Their thinking narrows, words become harder to find, and staying engaged becomes increasingly difficult. In Gottman’s work, this kind of high arousal during conflict is commonly described as flooding.

From inside that experience, stepping away may feel like an attempt to regain enough control to function rather than an attempt to leave the relationship.

Flooding is one pathway into withdrawal, among several.

What helps depends on what the withdrawal is doing

Someone who is overwhelmed may genuinely need time to settle before continuing. Someone avoiding an uncomfortable but necessary conversation needs something else. Someone who disagrees with what is being asked may need to say no rather than silently resist.

The behaviour looks similar from outside. Its function can differ substantially.

Space that allows a return

There are good reasons to pause difficult conversations. Continuing while one or both people are overwhelmed can turn a hard discussion into a damaging one.

A pause that makes return possible is different from creating distance that leaves the issue, and the other person, indefinitely suspended.

“I am done with this.”

compared with

“I want to finish this conversation, but I am not doing it well right now. I need some time and I will come back.”

The second person is still leaving the conversation. The issue remains acknowledged, the relationship remains engaged, and there is an intention to return.

Trust in that pause develops when the return actually happens.

When this is not a mutual cycle

A relational pattern should not hide responsibility

The pursuer-withdrawer framework works best when two people are caught in a reciprocal interaction in which each person’s response is affecting the other’s.

It should not make every difficult interaction equally shared.

If one person has deceived the other and the injured partner keeps asking questions, describing that simply as “pursuit” may miss why the questions exist.

If somebody repeatedly intimidates, threatens or controls their partner and the other person becomes quiet, that withdrawal cannot be understood in the same way as ordinary couple disengagement.

Silence can also be used deliberately as punishment. Persistent pursuit can become harassment, intimidation or refusal to respect another person’s reasonable need for space.

The behaviour may resemble part of the cycle while the underlying situation is very different.

Systemic language can otherwise flatten responsibility.

“We both contribute to the cycle” may describe two people genuinely caught in a reciprocal pattern. It does not adequately describe coercion, abuse, significant deception or another situation with a clear direction of harm.

Before asking how both people are maintaining a cycle, first ask whether a mutual cycle is an accurate description of what is happening.

If that is unclear

If you are unsure whether what is happening is a mutual cycle or whether there is coercion, intimidation or another clear direction of harm, take that uncertainty to an appropriately trained counsellor rather than trying to sort it out alone.

Where the pattern comes from

There is rarely one explanation

Pursuer-withdrawer patterns are sometimes explained almost entirely through childhood or attachment style.

Early relationships can contribute. People learn through experience what happens when somebody is upset, how conflict is handled, whether important people remain available, how quickly disagreements should be resolved, and how safe emotional dependence feels.

Those experiences can shape what later relationships seem to mean.

Family culture

One person may have grown up in a home where problems were discussed immediately and silence signalled that something was seriously wrong. Another may come from a family where people routinely took time apart and returned once everyone had settled.

Bring those two expectations into the same relationship and each person’s ordinary way of handling conflict can feel deeply wrong to the other.

Temperament also has a role. Some people become physiologically overwhelmed more quickly. Others tolerate prolonged emotional intensity more easily. Some need time before they know what they think. Others understand themselves by talking.

What the relationship itself builds

Previous adult relationships can change expectations.

Someone who has repeatedly experienced partners becoming unavailable during conflict may become particularly sensitive to distance.

Someone whose previous experience of conflict involved relentless criticism or arguments that continued for hours may begin disengaging earlier.

The current relationship then develops its own history.

A small tendency to pursue can become stronger after repeated experiences of withdrawal. A preference for stepping back can become more pronounced after repeated experiences of intense pursuit.

Eventually each person may be responding partly to something that has been built between them.

History, temperament, relationship experience and present circumstances can all contribute to why a particular response becomes more likely.

Tendency, role and context

You may not always be the same one

People often recognise themselves quickly in this pattern.

“I pursue.”

“He withdraws.”

That recognition can be helpful, particularly when a couple has spent years describing the same interaction as one person being needy and the other not caring.

But pursue and withdraw are roles in an interaction, not fixed personality types.

One of the most important findings in demand-withdraw research is that the subject being discussed affects who takes which role.

In classic studies, husbands and wives each became more demanding when the conversation concerned a change they wanted from their partner, and more withdrawing when their partner was the one seeking change.

Later research involving heterosexual, gay and lesbian couples found the same basic effect: the more change one person wanted relative to their partner, the more likely they were to demand while the other withdrew.

This helps explain why roles can shift.

Someone may pursue around emotional closeness and withdraw around money.

One person may repeatedly initiate conversations about parenting while avoiding conversations about sex.

Someone who pursued a distant previous partner may find themselves withdrawing from a much more persistent current partner.

There can still be tendencies. Some people more readily move toward relational difficulty, while others more readily create distance.

The role exists between two people, around something.

What about gender?

Research has repeatedly found a wife-demand/husband-withdraw pattern in heterosexual couples. Gender therefore remains part of the evidence rather than something that can be explained away.

Conflict structure has an independent effect as well: people tend to demand more when discussing changes they want and withdraw more when their partner is seeking change. Gender and conflict structure both contribute to the research picture.

The pattern also appears outside heterosexual couples. Research with gay and lesbian couples has found demand-withdraw interaction and again found that differences in desired change help predict which partner takes which role.

Cross-cultural research has identified demand-withdraw patterns among couples from different countries, although their frequency and expression can vary.

There is no sound basis for treating pursuit as inherently female or withdrawal as inherently male. Gender, power, relationship expectations, conflict structure and the particular couple can all contribute to what develops.

When the cycle becomes established

We start anticipating what the other person will do

Repeated cycles create history.

If attempts to raise difficult issues repeatedly end with the other person disengaging, the pursuer does not enter the next conversation with only the current moment in mind. They may already expect withdrawal.

That expectation can produce more urgency from the beginning.

Likewise, if previous conversations have involved prolonged pressure, repeated questioning or escalating criticism, the withdrawer may already expect another exhausting interaction.

They may begin pulling back before much has happened.

Research comparing couples with different levels of relationship distress has found greater rigidity and polarity in demand-withdraw patterns among more distressed couples. Established cycles often become more predictable and harder to leave.

The couple begins to know the script.

“I already know what you’re going to do.”

Each behaviour also acquires meaning through repetition.

Silence becomes:

“You don’t care.”

A request to talk becomes:

“Here we go again. Nothing I say will ever be enough.”

A pause becomes evidence that the issue will once again be left unresolved.

Another question becomes evidence that the conversation will never stop.

Past experience may make some of those interpretations well founded. At other times, what is happening now becomes mixed with what earlier versions of the interaction have taught each person to expect.

Once the meaning feels certain, the next response comes quickly.

The pursuer needs less withdrawal before pursuing.

The withdrawer needs less pressure before withdrawing.

Looking underneath the move

What happens just before you do it?

One way of slowing the cycle is to become interested in the short space before the familiar response.

For the pursuer

What happens when the other person becomes distant? What does the silence seem to mean, and what are you trying to get from the interaction at that point: connection, reassurance, acknowledgement, an answer, a decision, or change?

For the withdrawer

What happens as the interaction becomes more intense? Are you overwhelmed, angry, defensive, unable to think, convinced the conversation is pointless, resisting something you do not want, or trying to stop the situation getting worse?

There is no required emotion underneath either role.

A pursuer can be frightened, lonely or angry. A withdrawer can feel ashamed, resentful, overwhelmed or convinced that continuing will achieve nothing. Several things may be present together.

Rather than asking, “What am I really feeling underneath this?”, try:

“What is happening in me that the other person cannot see?”

There can be a considerable difference between the internal experience and the information the partner receives.

Inside

“I am worried that this matters to me far more than it matters to you.”

Outside

“You never care about anything that’s important to me.”

And on the other side:

Inside

“I don’t know how to answer this without another argument.”

Outside

Silence.

The outside response is what the other person has available to interpret.

Making some of the internal experience visible gives them different information.

This does not require saying everything you feel or explaining yourself for twenty minutes. One accurate sentence can change the interaction:

“I can feel myself getting scared that we’re disconnected.”

“I am angry because I feel like this keeps getting put off.”

“I need time to think, but I do want to answer you.”

“I am shutting down because I don’t know how to do this conversation right now.”

Those statements allow the other person to see more of what they are actually responding to.

Changing the cycle

Make the pattern visible

Once it is clear that this is a genuinely reciprocal cycle, both people can begin looking at the interaction itself.

“How do I get you to stop doing this to me?”

becomes

“What happens between us that keeps taking us here?”

When the other person is the entire problem, the obvious solution is to get them to behave differently.

Seeing the cycle creates another possibility. Each person can look for the point where their own response helps keep it moving.

That does not require pretending both people contributed equally to the original issue.

Someone can have a legitimate complaint and still pursue it in a way that makes discussion increasingly difficult. Someone can have a legitimate need for space and still withdraw in a way that leaves the other person stranded.

The work is finding a way for both needs to exist without immediately producing the familiar loop.

For the pursuer

The pursuer does not need to stop valuing connection or asking for legitimate change.

The question is whether increasing pressure is reducing the likelihood of getting the engagement they want.

It can help to slow down long enough to identify what is actually being asked for.

“Are you still okay with me?”

“I need you to acknowledge what happened.”

“I need an answer.”

“I want something to change.”

“I need us to make a decision.”

Those requests are different.

When they all emerge as escalating pursuit, the other person may experience pressure without understanding what would actually help.

It can also help to describe the experience without beginning with a judgement of the other person’s character.

“You always shut me out.”

compared with

“I can feel myself becoming worried that we’re disconnecting.”

“You never take this seriously.”

compared with

“This issue is important to me, and I don’t want it to disappear because the conversation is difficult.”

The second version cannot guarantee agreement or a good response. It does give the other person more accurate information.

For the withdrawer

The withdrawer does not need to remain in every conversation until the other person is satisfied.

There are times when continuing is genuinely unhelpful.

What matters is becoming clearer about what the withdrawal is doing and what needs to happen next.

If you are overwhelmed, say so.

If you need time to think, make that clear.

If the conversation has become too hostile or repetitive to continue productively, name that.

If you disagree with the change being requested, say no rather than withdrawing indefinitely.

If you have caused harm, needing time does not remove the responsibility to return and address it.

Silence leaves the other person to guess which of those is happening. A sentence gives them more to work with:

“I want to talk about this, but I’m getting overwhelmed.”

“I need time to think before I answer properly.”

“I disagree with what you’re asking for, but I don’t want to avoid the conversation.”

If you need a break, give the other person something realistic to rely on:

“I need half an hour. I’ll come back at eight.”

Then come back at eight.

Repeated return gradually changes what taking space means in the relationship.

Talk about the cycle outside the cycle

The middle of a fully developed pursue-withdraw interaction is usually a poor time to analyse it.

A calmer conversation gives both people access to more of what was happening.

“When you go quiet, I start assuming this doesn’t matter to you.”

“When you keep pressing, I start believing there is no answer I can give that will be enough.”

This begins building a shared map.

Eventually one person may notice:

“This is where I normally start chasing.”

And the other:

“This is where I normally stop talking.”

Sometimes either can simply say:

“We’re starting to do the thing.”

That recognition creates a little more room for another response.

What progress looks like

What changes first

The cycle can still appear even when both people understand it well.

These responses may have been practised over years. Knowing what is happening and being able to respond differently while it is happening are not quite the same task.

Progress often appears first in the shape of the cycle.

  • You notice it sooner.
  • The pursuit becomes less intense.
  • The withdrawal becomes less absolute.
  • A person asks for space without disappearing.
  • The other allows some space without immediately escalating.
  • A need becomes clearer.
  • A difficult conversation pauses and actually resumes.
  • A cycle that once consumed the evening lasts twenty minutes.
  • A rupture that once produced days of distance is repaired the same day.

Over time, repeated experiences like these can begin changing what each person expects.

Space becomes less threatening when it repeatedly ends in return.

Difficult conversations become less overwhelming when they repeatedly end without attack or disappearance.

Requests for change become easier to hear when they are made without contempt or relentless pressure.

Neither person has to become the opposite of who they are.

The pursuer can still value engagement.

The withdrawer can still need time and space.

The change is that those differences no longer have to set off the same sequence every time something difficult happens.

Try one thing this week

Notice the first moment when you feel the urge to make your usual move.

If you tend to pursue, ask what you are actually trying to get in that moment: connection, reassurance, acknowledgement, an answer, a decision, or change.

If you tend to withdraw, ask what you are trying to move away from: overwhelm, criticism, conflict, pressure, a decision, responsibility, or something else.

Then try to make that information a little more visible before the usual response takes over. You do not need a perfect conversation. You are looking for the point where the cycle is visible early enough that another response is still possible.

Selected sources

  1. Christensen, A., & Heavey, C. L. (1990). Gender and social structure in the demand/withdraw pattern of marital conflict. Journal of Personality and Social Psychology, 59(1), 73–81.
  2. Heavey, C. L., Layne, C., & Christensen, A. (1993). Gender and conflict structure in marital interaction: A replication and extension. Journal of Consulting and Clinical Psychology, 61(1), 16–27.
  3. Gottman, J. M. (1994). What Predicts Divorce? The Relationship Between Marital Processes and Marital Outcomes. Erlbaum.
  4. Schrodt, P., Witt, P. L., & Shimkowski, J. R. (2014). A meta-analytical review of the demand/withdraw pattern of interaction and its associations with individual, relational, and communicative outcomes. Communication Monographs, 81(1), 28–58.
  5. Christensen, A., Eldridge, K., Catta-Preta, A. B., Lim, V. R., & Santagata, R. (2006). Cross-cultural consistency of the demand/withdraw interaction pattern in couples. Journal of Marriage and Family, 68(4), 1029–1044.
  6. Eldridge, K. A., Sevier, M., Jones, J., Atkins, D. C., & Christensen, A. (2007). Demand-withdraw communication in severely distressed, moderately distressed, and nondistressed couples: Rigidity and polarity during relationship and personal problem discussions. Journal of Family Psychology, 21(2), 218–226.
  7. Holley, S. R., Sturm, V. E., & Levenson, R. W. (2010). Exploring the basis for gender differences in the demand-withdraw pattern. Journal of Homosexuality, 57(5), 666–684.
  8. Bretaña, I., Alonso-Arbiol, I., Recio, P., & Molero, F. (2022). Avoidant attachment, withdrawal-aggression conflict pattern, and relationship satisfaction: A mediational dyadic model. Frontiers in Psychology, 12, 794942.
  9. Huerta, P., Johnson, L. N., Helm, K. M., Anderson, S. R., & Wittenborn, A. K. (2023). Exploratory analysis of pursue-withdraw patterns, attachment, and gender among couples in Emotionally Focused Therapy. The American Journal of Family Therapy, 51(1), 57–75.