Skip to content
My Baby Sleep Schedule

Pillar guide · 4, 6, 8 and 12 months

Sleep regressions: what is actually happening

Start with the part almost nobody leads with: “sleep regression” is not a medical term. It appears in no pediatric guideline. The American Academy of Pediatrics does not define it, the AASM does not measure it, and there is no test for it. It is a word parents invented for a pattern parents kept noticing, and it stuck because the pattern is real even though the label is informal.

That matters practically, not pedantically. Because there is no clinical definition, there is also no clinical treatment, which means every confident protocol you find online for “fixing the 4-month regression” is somebody’s opinion with good formatting. What does exist is solid evidence about the underlying development — sleep cycle maturation, motor milestones, object permanence — and reasonable, low-risk things to do while it passes.

It is also the wrong word. Regression implies going backwards, and what is happening at four months in particular is the opposite: your baby’s sleep is maturing toward the structure adults use. It feels like a loss because the newborn sleep it replaced was, from a parent’s perspective, easier. Nothing has gone wrong. Something has moved forward and taken your evenings with it.

Jump to the timeline Check the schedule still fits

The four rough patches, in order

Ages are approximate and the ordering matters more than the numbers. Plenty of babies hit these earlier, later, mildly, or barely at all.

  1. Around 4 months

    Sleep architecture matures

    The one with a real mechanism behind it. Newborn sleep has two broad states; somewhere around three to four months it reorganises into the cycled structure adult sleep uses, with lighter phases between cycles. Babies begin surfacing at those junctions the way we do — the difference is that we roll over and forget it, and they need help. This is why 30-to-45 minute naps appear almost overnight and why night wakings multiply. It is a permanent developmental change, not a phase that reverses, which is why 'regression' is such a poor word for it.

    Typically lasts: Two to six weeks of visible disruption; the underlying change is permanent

  2. Around 6 months

    Rolling, sitting, and a nap count that no longer fits

    Less a regression than a collision. New motor skills get practised at 2am because the brain rehearses what it has just learned, and at the same time the third nap starts falling out of a day that has run out of room for it. Families who read this purely as a regression sometimes wait it out when what the day actually needed was a schedule change.

    Typically lasts: One to three weeks, often resolving as soon as the nap count is adjusted

  3. 8 to 10 months

    Separation anxiety and crawling

    Object permanence firms up: your baby now understands that you continue to exist after you leave the room, which is developmentally excellent and socially inconvenient at 7pm. Combined with crawling or pulling to stand — both of which get practised in the crib — this is often the most disruptive stretch of the first year after the four-month change.

    Typically lasts: Two to six weeks

  4. Around 12 months

    Walking, words, and a false nap alarm

    First steps and early language both land here, and both are associated with disturbed sleep. This period also produces the single most common scheduling mistake of the first two years: a baby refuses the second nap for a fortnight, everyone concludes it is time to drop to one, and the drop turns out to be a year early. Most babies are closer to 14 to 18 months for that transition.

    Typically lasts: Two to four weeks, though the nap confusion can drag on longer

Why four months is the one that breaks people

Because it is the only one on that list with a permanent structural change underneath it, and because it usually arrives at the exact moment a family has started to believe they have figured this out. Newborn sleep, for all its chaos, has a forgiving property: babies drop into deep sleep quickly and stay there. What replaces it around three to four months is cycled sleep, with lighter passages between cycles where a baby can surface fully.

Adults surface at those junctions too, several times a night. We simply do not remember it, because we have decades of practice drifting back down. A four-month-old has none. So she surfaces at the end of the first nap cycle, roughly 30 to 45 minutes in, finds herself awake and alone in a room that has changed since she fell asleep, and calls for you. That is the entire mechanism behind the infamous 45-minute nap, and it is why nothing you do makes it stop on demand.

The genuinely useful responses are unglamorous. Keep wake windows current for her actual age — under-tiredness makes the surfacing worse, not better. Make the room properly dark, because a room that is dim rather than dark at 5:30am is a common contributor to early waking. Accept that some naps will simply be short and that a short-nap day is best repaid with an earlier bedtime rather than a late catch-up nap. And know that it does settle, usually within a month or two, without any specific intervention.

Regression, or a schedule that no longer fits?

This is the distinction most worth making, because one of these requires patience and the other requires a fifteen-minute change you can make today. Wake windows grow considerably during the second half of the first year — a baby who needed two hours awake at five months may need close to three by eight months. Schedules do not update themselves, and an under-tired baby produces almost exactly the symptom set of a regression: refusing naps, taking forever to settle, waking early.

A rough test: if she is cheerful and alert while resisting sleep, the window is probably too short. If she was fine and then fell apart within about ninety seconds, it ran long. Both produce short naps, which is why nap length alone is uninformative and the behaviour going in is the tell. Before concluding you are in a regression, run the current age through the wake window calculator and compare its windows to what you are actually doing. A surprising proportion of “regressions” turn out to be a schedule that is six weeks out of date.

What actually helps while you wait it out

Nothing here is dramatic, and that is the point. The evidence base for regression-specific interventions does not exist, so the defensible advice is limited to things that are low-risk and address the aggravating factors rather than the development itself.

  • Update the wake windows. The most frequently productive single change, and the most frequently overlooked.
  • Protect the morning. Darkness until a reasonable hour, and daylight soon after waking, both support the circadian rhythm that is still consolidating through the first year.
  • Bring bedtime earlier on bad days. Thirty to forty-five minutes. Lost daytime sleep is more reliably repaid at night than by a late afternoon nap that steals from bedtime.
  • Keep the routine identical. During a period when everything else is changing, predictability is one of the few things you control.
  • Change one thing at a time. Give it three to five days. Two simultaneous changes tell you nothing about either.
  • Lower the bar for a fortnight. This one is not a technique, but the families who come through these periods least damaged are usually the ones who stopped optimising and started surviving.

A note on where this site stops

Everything above is about scheduling and normal development. Safe sleep — surfaces, positioning, bedding, room sharing — is a separate subject with real stakes, and we do not write about it, because we are not qualified to. Go straight to NICHD Safe to Sleep and the AAP’s own guidance. And if a difficult stretch comes with fever, poor feeding, poor weight gain, unusual lethargy, heavy snoring or breathing pauses, that is a call to your pediatrician today, not a scheduling problem to solve with a chart.

Straight answers

Is 'sleep regression' a medical term?
No. It is a parenting term, not a clinical diagnosis, and you will not find it in the AAP or AASM literature. That does not make it useless — it names something real that a great many parents experience at broadly similar ages. But it does mean nobody can give you an authoritative definition, a diagnosis, or a treatment, and anyone selling one of those should be treated with suspicion.
How long does a sleep regression last?
Commonly quoted figures are two to six weeks, which matches most parents' experience. The four-month change is different in kind: the sleep-cycle maturation behind it is permanent, so what settles is your baby's ability to handle the new structure, not a return to how things were. If a rough patch has run past six weeks with no improvement at all, it is worth considering whether something else is going on — a schedule that no longer fits, teeth, illness, or a room that gets light at 5am.
Should I change anything during a regression?
Change as little as possible, but do check the schedule. The two things most worth verifying are that wake windows still match your baby's current age — they grow faster than most parents update them — and that the last nap is not ending so late that it steals from bedtime. Beyond that, introducing a brand-new sleep approach during the most chaotic fortnight of the quarter usually adds a variable rather than solving one.
Can I prevent a sleep regression?
Not really, and it is worth being blunt about that because a lot of content implies otherwise. These disruptions accompany normal neurological and motor development. You can reduce the amount of additional overtiredness stacked on top — by keeping wake windows current, protecting the room's darkness, and moving bedtime earlier on rough days — which genuinely makes the period more bearable. You cannot skip the development itself.
When is it not a regression?
When there are other signs. Fever, ear-pulling with distress, a change in feeding, poor weight gain, unusual lethargy, heavy snoring, or pauses in breathing during sleep are not developmental sleep disruption and should go to your pediatrician rather than a search engine. The same applies to any change that arrives suddenly in a baby who was previously settled and comes with a shift in daytime behaviour.

Deeper on each regression

Sources

Sukie Gao, founder of My Baby Sleep Schedule

Written by

Sukie Gao

Mom of a daughter, and the person behind every tool on this site.

Sukie Gao has a daughter, and when her daughter was a newborn she was exhausted and completely out of her depth. She spent months piecing together sleep advice from scattered corners of the internet, and built My Baby Sleep Schedule so the next tired mother would not have to search as hard. She writes as a parent from lived experience, not as a clinician.

More about SukieYouTubeHow this was researched

Last reviewed and updated . Sukie is a parent, not a clinician — check anything that matters with your pediatrician.