Toddler regression after separation

Module 02 · Toddlers & potty training · Article 05 · 0–3

She was reliably using the potty by April. Six weeks of clean undies, the dinosaur ones, the proud announcements at the table. By June, six months after the separation, she's having two accidents a day. Sometimes in undies, sometimes on the rug, sometimes in the bath. She also asks to be carried up the stairs she used to climb. She's started asking for the bottle she gave up at fifteen months.

Your first thought is that you've damaged her. Your second is that something must be happening at her co-parent's. Your third is that you should never have separated.

None of these is what's actually happening.

This article is about regression. What it is, what it isn't, why it shows up around major family changes, what helps and what hurts, and when the regression is a sign of something larger.

What regression actually is

A toddler's developing nervous system has finite resources. Sleep, attachment, language, motor skills, social skills, self-regulation, toileting, eating, autonomy. All of these are being built at the same time, on a single energy budget that's also doing the daily work of growing.

When a major change happens in the toddler's environment, the body redirects energy toward integrating that change. The integration work is invisible but expensive. To pay for it, the body temporarily withdraws investment from skills that were already in progress. Skills already mastered can wobble. Skills that were nearly mastered can fully come undone for a few weeks or months.

This is not the toddler going backward. This is the toddler reallocating. The skills come back, usually in a similar order to how they first arrived, once the integration is further along.

A separation is one of the larger things a toddler's environment can do. The household structure splits. The number of bedrooms doubles. The cast of adults, the rhythms of meals, the sequence of mornings, the geography of the day, all reorganise. The toddler's body is doing months of integration work to build a single internal map of two homes. While that's happening, the easier-to-mess-up skills wobble.

This is why regression after separation is so common. It's not a bug. It's a feature of how toddlers process change.

The forms regression takes

What you might see, in a toddler 1 to 6 months after a major family change:

Toileting. Accidents in a child who'd been reliable. Wanting nappies again. Pushing the potty away. Constipation. Holding for hours.

Sleep. Earlier waking. Later settling. Wanting to be held to sleep again after self-settling. Night waking with no clear cause. The 4am wake-up that wasn't there last month.

Language. Reverting to baby words. Less talking. Whining where they used to make sentences. Sometimes a brief stutter that resolves in weeks.

Eating. Pushing away foods that were favourites. Wanting only the same three things. Eating less overall, sometimes for a week or two. Wanting to be fed when they were feeding themselves.

Comfort objects. Wanting the bottle they gave up. Wanting the dummy back. Reattaching to a soft toy that had moved to the back of the cot.

Separation anxiety. Crying at handovers when they'd settled. Crying when you go to the bathroom. Following you from room to room. Wanting to sleep in your bed.

Motor skills. Wanting to be carried for short distances they'd been walking. Trouble with stairs. Less interest in physical activity that excited them a month ago.

Self-regulation. Bigger meltdowns over smaller things. Less tolerance for waiting. Crying instead of using words.

You might see one of these. You might see several at once. The pattern is usually that the most-recently-acquired skills wobble first. Toileting was new. Now it's gone. The vocabulary explosion of the last month thins out. The reliable handover protest comes back.

What it isn't

Several misreadings, all of which make the situation harder than it needs to be.

She's broken. No. Her body is doing developmental work that all toddlers do around major change. The skills aren't gone. They're temporarily de-prioritised.

Something must be happening at the second home. Sometimes. Often not. Most regression is the integration work itself, not a sign of something specific at one of the two places.

I should restart everything. The temptation to launch a fresh potty training campaign, sleep training program, or behaviour chart in the middle of regression is strong. Don't. Adding new variables in the middle of integration work makes the work harder. Hold what you've been doing. Let the regression run its course.

This means we should reconsider the separation. Almost always not. Regression happens around the separation transition. It's not evidence that the transition itself is wrong. The transition is over; the integration is what's happening now. Reversing it would create a new transition and a new round of integration.

She's manipulating me. A two-year-old isn't manipulating. A two-year-old is asking for what their body needs in the language they have. Letting them be carried up the stairs for two months while they go through this isn't spoiling them. It's giving them the regulatory support that lets the integration finish.

What helps

Most regression resolves on its own with time and steady support. A few things that accelerate the resolution:

Don't escalate consequences. A child who's having accidents doesn't need to clean it up themselves to teach a lesson. The accident isn't a behaviour. It's a body event. Clean it up neutrally. Move on. Same with everything else on the list above.

Lower the bar across the day. Cut activities. Cut homework standoffs (where they exist). Cut the morning expectations. The toddler is using more energy than usual on the integration work. Give them less to do with the energy they have.

Hold the rituals you've been holding. The bedtime ritual, the comfort object, the same dinner times, the same pickup person. The architecture of the days is the thing the child is reorganising around. If you change the architecture during the reorganisation, the work has to start over.

Meet the regression at face value. If she wants the bottle, give her the bottle for a few weeks. If she wants to be carried, carry her for the short distances. If she wants to be in your bed at 3am, let her come. The regression is asking for a known regulatory state. Provide it. The pull toward independence will return on its own when the integration work is further along.

Watch your own state. Toddlers borrow regulation from the adult around them. A parent who's panicking about the regression is producing a household nervous-system state that itself slows the integration work. The most useful thing you can do during regression is keep your own breathing slow. (See Toddlers 01 fifth pillar.)

Talk to her, lightly, once. I know things are different now. We have two homes. It's a lot. I'm not in a hurry for you to be a big girl right now. Said quietly, once or twice across the regression period. Don't keep raising it. The over-naming becomes its own pressure.

What hurts

A few moves that are tempting and counterproductive.

Punishing accidents. This makes the next accident more likely. The body that's holding regulation work doesn't have spare resources for shame management.

Comparing to siblings, cousins, or how she was a year ago. Your sister was potty trained at this age hurts and doesn't help.

Adding incentive systems mid-regression. Sticker charts, bribes, threats, withdrawal of privileges. None of these address what regression is. They add cognitive load to a system that's already at capacity.

Pressing her to perform skills she's regressed on. Insisting she walk up the stairs. Insisting she use the potty. Insisting she sleep in her own bed. The insistence prolongs the regression because it adds dysregulation to the integration work.

Bringing it up at handovers. She's been having accidents all week, you'll need to know. True information for the co-parent, sometimes. But not in front of the child at the doorway. Not in a way that signals to her that her regression is a problem the adults are now talking about.

The conversation with your co-parent

This is one place where shared information genuinely helps.

A toddler regressing across both homes, where both homes are doing the right things, is doing developmental work. Both parents can hold steady. The conversation is informational, not blame-finding.

The framing: I'm seeing X over here. Are you seeing it on your end? Let's hold the protocols and not make any new changes for a few weeks. Let's compare notes again in a fortnight.

Where things get harder is when one home is panicking and one isn't. The panicking home is sometimes pressuring the calmer home. You need to crack down on this. She's getting away with it at your place. This rarely helps. Cracking down adds dysregulation to a system that's regulating itself slowly.

The other harder version is when one home thinks the second home's protocol is the cause. She wasn't having accidents until she started staying with you four nights a fortnight. This may correlate, but the causation is usually the integration work itself, not the second home's parenting. Treating the second home as the problem makes the integration work harder, not easier.

If both parents can hold the framing as integration-work-not-failure, the regression usually resolves over weeks to a few months.

When to consult someone

Most toddler regression resolves on its own. Some warrants professional eyes.

Worth attention:

The regression has been deepening, not improving, after 8 to 12 weeks of steady support
Multiple skill areas are regressing simultaneously and severely (sleep, eating, toileting, language all wobbling at once)
New behaviours emerge that weren't there before: head-banging, severe self-injury, complete withdrawal, prolonged inconsolability
The child seems flat or absent rather than emotionally expressive
Significant weight drop or eating that's becoming unsafe
Constipation that's persistent or distressing
Sleep collapse that's affecting both child and parental functioning across both homes

Any of these is worth a conversation with a paediatrician, ideally one with experience in family transitions.

The longer arc

The regression that shows up at month two is usually starting to ease by month four. The toileting comes back. The vocabulary returns. The bottle gets put down again. The carrying up the stairs becomes climbing. The rate isn't always linear; some weeks are better than others. The direction is, almost always, toward re-integration.

What the toddler builds during this period is more durable than what they had before. The skills that come back after regression have been re-acquired in a more complex environment. The toilet training that holds across two homes is a sturdier toilet training than the one that only had to work in one. The self-settling that comes back after the integration is a stronger self-settling. The vocabulary that re-emerges has a richer texture, with the words from both homes layered into it.

This isn't recovery. It's rebuilding at a higher level of complexity.

Closing

The two-year-old in the dinosaur undies, having accidents in June after being reliable in April, is not damaged. She's doing the work her body needs to do to make a single internal map out of two homes. The accidents will pass. The bottle she's asking for will go back in the cupboard once she doesn't need it. The carrying up the stairs will become climbing again.

What she needs from you is steadiness. The same dinner. The same bath. The same bed. The same comfort object. The acceptance of where she is right now, without the pressure to be where she was two months ago.

By Christmas, the dinosaur undies will be back in rotation. The bottle will be in a box at the back of a cupboard. She'll be telling you, in full sentences, about the rabbit she saw at her co-parent's. The integration will have happened, the rebuilding will be further along, and the regression you watched in June will be one of the things you'll one day, much later, have forgotten.

Tonight, change her clothes. Don't say anything about it. Read the book. Let her be small. The body does the work in its own time.