An argument may concern money, parenting or a phone, while the interaction follows much the same sequence each time. One partner raises the issue with growing urgency as the other explains, becomes quiet or leaves, and each response increases the other's distress.
Researchers call this a demand-withdraw pattern, although the label provides only a starting point for understanding what each person is trying to manage. One may be seeking connection or resolution while the other tries to avoid criticism or escalation, with neither intention readily visible to their partner.
How the sequence reinforces itself
When withdrawal feels like abandonment, the partner seeking a response may become more urgent or critical, which can increase the other's wish to retreat. Both can leave the conversation with an account that makes sense from their own position, while missing how their response contributed to the next turn.
Demand-withdraw patterns have long been associated with poorer relationship outcomes. A newer observational study recorded 151 couples discussing a sexual conflict and found more observed demand-withdraw behaviour associated with lower relationship and sexual satisfaction and higher sexual distress at the time, as well as lower relationship satisfaction twelve months later.
These associations cannot establish that the interaction caused the later difficulties, and one recorded discussion captures only part of a relationship. They do support paying attention to what happens between partners, alongside what each reports about the problem.
Understanding the response without excusing harm
A partner who becomes quiet may be experiencing intense arousal, shame or difficulty organising their thoughts, even when they appear calm. Silence may have been learned as a way to prevent conflict worsening, while repeated questions can become an attempt to find out whether the issue or relationship still matters.
Understanding those possible meanings helps only when responsibility for behaviour remains clear. Contempt, intimidation or coercion require their own attention, and an explanation about fear does not make those responses acceptable.
Couple communication strategies are not appropriate substitutes for individual safety planning where there is violence, coercive control, fear or ongoing abuse. Assessment of those conditions needs to shape whether and how joint work proceeds.
Taking and returning from a break
A pause can interrupt escalation when both people know why it is happening and when the conversation will resume. Naming the need to settle, acknowledging that the issue remains important and agreeing on a realistic return time makes the break more predictable.
The partner seeking a response may then need to allow the pause without following or continuing the discussion by text. That becomes easier when the agreement to return has been kept, so the reliability of what happens afterwards contributes to whether a break feels possible next time.
When the discussion resumes, describing the experience underneath the response can give the other person more to work with. Feeling unimportant and panicking, or becoming afraid of making things worse and shutting down, offers a different opening from another accusation about what the partner always does.
Practising a different response
Couple therapy can help slow the sequence so that each person can recognise and take responsibility for their part. The work may draw on emotionally focused, behavioural, cognitive or Gottman-informed methods, according to the couple's needs.
A recent systematic review and meta-analysis found that couple interventions improved relationship distress, although approaches and study quality varied substantially. Sessions need to translate into changes outside the room, such as noticing escalation earlier or keeping an agreement to return to an unfinished discussion.
Reviewing a recent disagreement in detail can show whether those changes are occurring, including where the familiar sequence resumed and what either person managed differently. Continued disagreement can then be examined without making the absence of conflict the only measure of progress.
Research drawn on
- Rosen NO, Dubé JP, Bosisio M, Bergeron S. Do demand-withdrawal communication patterns during sexual conflict predict couples’ relationship satisfaction, sexual satisfaction, and sexual distress?. Journal of Sex Research. 2025.
- Schrodt P, Witt PL, Shimkowski JR. A meta-analytical review of the demand/withdraw pattern of interaction and its associations with outcomes. Communication Monographs. 2014.
- Ebrahim S, et al. Effectiveness of couple interventions in marital distress: a systematic review and meta-analysis. BMC Psychology. 2025.
AI disclosure
Brent Munro researched, wrote and reviewed this insight, with ChatGPT assisting in literature searches, limited research comparisons, editing and image generation. He retains full responsibility for the published content.