Sleep regressions during the early weeks of separation

Module 01 · Sleep & bedtime · Article 10 · 0–3

Three weeks ago, your two-year-old was sleeping seven to seven. He had been doing that since he was eighteen months old. Last night was the eighth night in a row he woke at 1am. He cried for forty minutes. He wanted you. He wouldn't take water. He didn't want the bear. When he finally fell asleep, it was on the floor, between your bed and the wall.

Three weeks ago, you and your co-parent moved into separate homes.

This article is about that. The sleep that was working and isn't anymore. What's happening, why it's happening now, how long it usually lasts, and what to do in the meantime.

What's happening, in the body

Children in their first three years are still building the architecture of sleep. They have a baseline rhythm. They have favoured ways of falling asleep, favoured ways of waking, favoured ways of being soothed back. That architecture is held together by predictability. Same room. Same people. Same shape of the day. Same shape of the evening.

When the architecture changes, sleep destabilises. Not because the child is sad in the abstract sense (although they may also be sad). Because the cues the body uses to know when to drop into sleep, how deep to drop, when to surface, and how to settle through a surfacing, those cues are now mixed with new cues. The body is trying to map a new pattern while still running the old one.

This is what is sometimes called a separation sleep regression. It is not a developmental regression in the classical sense (the 8-month-old, 18-month-old, 2-year-old waves), although it can coincide with one. It is a regression caused by a change in the environment that the child's nervous system uses to regulate sleep.

The signs of it:

Falling asleep takes much longer than it used to
Waking multiple times a night
Crying inconsolably at the wakings, beyond the usual brief settle
Wanting one specific parent, often the parent the child is not with that night
Earlier or later wake times than the established norm
Pushing away the comfort object, the bedtime book, or the song that used to work
Coming into the parent's bed when that wasn't the established pattern
Daytime irritability beyond what the child's normal range would explain
Eating less than usual, especially at dinner

Most of these are not pathological. They are the surface of an underlying nervous-system reorganisation. The body is recalibrating around the new shape of life.

Why now, specifically

Separation is one of the most disruptive events a young child's nervous system can experience. The reasons are physiological, not philosophical.

The smell of the home changes. A toddler navigates the world significantly through smell. The house smells different now. There may be a new bedroom, new bedding, a new kitchen, a new car. Even at the same physical address, the smell composition shifts as one parent moves their things out. The nervous system reads this as something fundamental has changed.

The sound of the home changes. The ambient sound profile of a home is specific. Two adults talking quietly in the next room. The shower running at 7am. The kettle. The footsteps. When the household composition changes, the sound profile changes. The body, which uses these sounds as cues for I am safe, the adults are nearby, has to relearn the new soundscape.

The body misses the absent parent. The child has a felt sense of who is in the house right now. When the second parent isn't there, the body registers an absence even before the conscious mind does. The bedtime ritual was being held by both parents in some configuration. Now it's being held by one. The body notices.

The child can sense the parents' state. Even without understanding what has happened, the child reads their parents' nervous systems. A parent who is grieving, anxious, or angry, even if they're holding it well in front of the child, transmits a different baseline. The child borrows that baseline. Sleep is harder when the borrowed baseline is activated.

The bedtime ritual is in flux. The very ritual that was holding sleep together is itself changing. The book at one home, not the other. The song with one parent, not the other. The exact rhythm of the wind-down, not yet established at the new home.

These are five reasons. There may be more in any specific family. Together, they describe why this is happening now and why it's not a sign of something being wrong with the child or the parenting.

How long it usually lasts

This is the question parents ask most. The answer, in its honest form:

The acute phase, two to six weeks. This is the period during which sleep is markedly worse than the pre-separation baseline. Most families come through it within a month, though some take longer. The first week is usually the hardest. By week three, most children have begun to find a new pattern at one of the homes, even if not yet both.

The transition phase, six to twelve weeks. Sleep is recovering but not yet stable. There will be a string of three or four normal nights, then a regression, then more normal nights. Pattern, not steady state. The body is still consolidating the new architecture.

Stabilisation, three to six months. A new normal. Not the same as the pre-separation normal. Sleep often looks slightly different in two-home life than it did in one-home life. There may be one home where the child sleeps more easily. There may be a recurring soft regression around handovers. The architecture is stable, but it's a new architecture.

Persistent disruption, beyond six months. When sleep is still significantly disrupted at six months, it's worth asking different questions. The arrangement may not be working for this specific child. The schedule may be too long for the developmental window. There may be unresolved emotional load in the household that the child is carrying. (Sleep 17 covers persistent bedtime difficulty in more depth.)

These are typical ranges, not guarantees. Some children come through the acute phase in a week. Some take three months. The variation is wide. The direction, in most cases, is toward stabilisation.

What helps

Some things that consistently help families through these weeks.

Hold the ritual. The bedtime ritual that travels (Sleep 02) is doing more work right now than it normally does. Don't let it slip during the regression. The exact same shape every night, even when you're exhausted, even when the child is pushing back on it. The ritual is the bridge. Don't burn it.

Co-sleep if you need to. This is a regression. Established rules can be temporarily relaxed. If the child is calmer in your bed than in their own, and you can sleep that way, do it for these weeks. You can re-establish the pre-separation pattern once the regression passes. Don't think of it as a step backwards. Think of it as the body needing extra holding through a hard period.

Stay calm in the night. The middle-of-the-night cries are when your own grief is most likely to surface. The exhausted parent at 2am, with a crying toddler, is the parent most likely to cry themselves. The child reads this. Slow your breathing first. Pick the child up only when your own body is steady. Your calm is the regulation they need to borrow.

Talk to your co-parent if you can. Compare notes on what's happening at each home. He's been waking at 1am at my place. What are you seeing? This is information-sharing, not blame. If you both have data, you can both adjust. (See Co-parent communication 01 if you're not sure how to have these conversations.)

Reduce other change. Don't start potty training in week three. Don't move bedrooms. Don't introduce a new sibling caregiver. Don't begin a sleep-training method you haven't used before. The body is processing as much as it can already. New variables make it harder.

Lower the bar for the day. A child sleeping badly will be a child eating less, learning less, and tolerating less during the day. Don't take it as a separate problem. The day is downstream of the night. Plan less, expect less, hold the routine but don't try to extend it. Days are for survival in this period.

Hold the comfort object firmly. Sleep 05 covers this in full. During the regression, the comfort object is doing more work than usual. It must travel reliably. If it doesn't go in the bag, the night it's missing will be much worse than usual.

Tell the receiving parent honestly what's happening. Don't perform that things are fine if they aren't. The receiving parent needs to know what they're walking into. Bedtime has been hard the last four nights. He's been waking at 1am. He's been crying for the bear. Information helps them do their job.

What doesn't help

A few things that look reasonable but tend to make this harder.

Treating the regression as a behaviour problem. It isn't. It's a body finding its way through a structural change. Punishments, time-outs, and reward charts don't work for this kind of waking. The child isn't choosing to wake. The child is being woken by their own dysregulating system.

Cycling through new sleep methods. Trying a new method each week, looking for the one that fixes it. The child cannot adapt to a new method during a regression. Whatever you were doing before, hold that. If you weren't doing anything specific, hold what you were already doing. This is not the time to introduce a new system.

Bringing the child into your big feelings. The child can feel that you're sad. They cannot, and should not, hold the explanation for why. Mama is sad because Daddy and I aren't living together anymore is too much for a two-year-old. Mama loves you. It's bedtime now. Bear is here. That's enough.

Comparing nights between homes as a verdict. He sleeps fine at my place. He must be doing something wrong over there. This is rarely true and almost always corrosive. The two homes have different conditions, different rooms, different rituals, different rhythms. A child sleeping better at one home in week three doesn't mean the other home is failing. It means the child has settled into one place a little earlier than the other.

Pushing the child to talk about it. Some children at this age don't have the language. Even those who do may not be ready. Imposed conversations about how the child is feeling about the separation make sleep worse, not better. The child will bring it up when they're ready, often months later, often in fragments. (Module 11 article 01 covers how to talk about separation when the child does ask.)

When to ask for help

Most families navigate this without professional support. A few signals that you might want it.

Sleep that hasn't measurably improved by 8–10 weeks, despite the architecture being held
Daytime regression in skills the child had previously mastered (potty training, words, eating)
Persistent night terrors, panic awakenings, or self-soothing behaviours that look new and intense
Either parent struggling significantly to function, with the child reading the strain
Domestic violence, addiction, or a major mental health concern in either home

A paediatrician is the first call. They can rule out medical causes and refer onward if needed. A child psychologist who works with separated families is worth seeking when the picture is complex. A family mediator may be relevant if the arrangement itself is part of what isn't working.

Asking for help is not a sign that the separation has failed the child. It's a sign that the parents have noticed something hard and are responding to it.

Closing

The early weeks of separation are some of the hardest sleep your family will go through. The toddler who was sleeping is no longer sleeping. The architecture is being rebuilt. The body is doing the work, mostly underground.

Hold the ritual. Co-sleep if you need to. Stay calm in the night. Don't start anything new. Send the comfort object every time. Tell your co-parent honestly what's happening at your home. Lower the bar for the day.

Most of this passes within six to twelve weeks. Some of it lingers in soft form for months. The child you knew before the separation is still there, underneath. Sleep is the body's way of re-finding them.

It will get better. Not tomorrow. But it will.