Baby Sleep Regression Ages and Stages: A Complete Chart

Baby sleep regressions cluster at roughly 4, 6, 8 to 10, 12, 18 and 24 months. Each is driven by a different developmental change rather than one recurring event, and most last two to six weeks. The four-month one is not really a regression at all — it is a permanent change in how your baby sleeps, which is why it does not simply pass.

Last updated: 1 August 2026 · Written by Atikur Rahman

Prefer this as a one-page printable? Get the free sleep regression chart.

Sleep regression ages chart

Age What is actually changing Typical length What you notice at 2am
4 months Sleep architecture matures. Cycles now include lighter stages with a full arousal between them. Permanent change; 2–6 weeks to settle Waking every 45 minutes to 2 hours. Naps collapsing to one cycle.
6 months Rolling and sitting, the start of teething, and the 3-to-2 nap transition. 1–3 weeks Rolling and getting stuck. Bedtime pushed late by a third nap.
8–10 months Crawling, pulling to stand, object permanence and separation anxiety — all at once. 2–6 weeks Practising motor skills in the cot. Standing up and not knowing how to get down.
12 months Cruising and walking, first words, and a nap schedule that no longer fits. 1–3 weeks Refusing the morning nap, then falling apart before bedtime.
18 months The genuine 2-to-1 nap transition, peak separation anxiety, a language burst and molars. 2–6 weeks Bedtime protest, early waking, and a nap that works some days and not others.
24 months Cot-to-bed moves, potty training, imagination and genuinely new fears. 2–4 weeks Getting out of bed repeatedly. Calling out. Fear of the dark.

The ages are approximate and the ranges overlap. A baby who crawls at six months will usually meet the eight-month pattern early.

At what ages do babies go through sleep regression?

The six ages above are the ones most consistently reported: 4, 6, 8 to 10, 12, 18 and 24 months, with some toddlers having a further rough patch around three years. They are not a schedule. They are the points where developmental change is densest, which makes disrupted sleep more likely — not certain.

Are sleep regressions actually real?

Partly. The developmental changes are real and well documented. The idea that every baby drops off a cliff at fixed ages is not — large datasets show infant sleep improving steadily across the first year, without an obvious dip at four months.

A 2018 review of the infant sleep literature found that night wakings decrease over the first year on average. Parents report more wakings between three and six months, but wearable and app-based measurements do not show the same increase. A 2016 study using data from a sleep-tracking app found nighttime sleep sessions lengthening through the first year with no four-month dip.

Day-to-day variation is also much larger than most parents expect. Nighttime sleep can differ by around two hours from one night to the next in the same baby. A bad week is usually inside the normal range rather than a new phase.

The useful way to hold both facts: the milestones are real, the chart is a rough guide, and "regression" is a badly chosen word for normal variability that happens to land near a developmental leap.

The 4-month sleep regression

At around three to four months a baby's sleep stops being the simple two-state newborn pattern and starts cycling through lighter and deeper stages more like an adult's. At the end of each cycle there is a brief arousal. A baby who has always been put down asleep now surfaces every 45 minutes or so and has no practised way of getting back down.

This is the reason the four-month change behaves differently from the others. It is not a phase that reverses. Sleep does improve, but through the baby learning to resettle, not through the change undoing itself.

Total sleep need does not drop. The American Academy of Sleep Medicine, endorsed by the AAP, puts infants aged 4 to 12 months at 12 to 16 hours per 24 hours including naps.

This one has its own guide: The 4-Month Sleep Regression: What It Is and How Long It Lasts, including the swaddle safety change that arrives at the same time.

The 6-month sleep regression

Six months is the one most often left off the list, and it is usually a combination rather than a single cause. Rolling and sitting arrive, teething commonly begins and tends to peak between six and twelve months, solids are often starting, and the third nap becomes surplus somewhere between six and eight months.

Because the causes are mixed, this one usually resolves fastest once the nap schedule is corrected. If bedtime has crept later, the third nap is the first thing to look at.

The 8-month sleep regression

Three things arrive close together: gross motor skills, object permanence, and a nap transition.

  • Motor practice in the cot. Babies rehearse new skills during light sleep. A 2024 systematic review in SLEEP summarising this literature notes that eight-month-olds who had started crawling had more night wakings over five minutes than those who had not — and the effect was more marked in babies who crawled early.
  • Object permanence. Understanding that you still exist after you leave the room is a cognitive gain, and it makes your absence at 2am newly significant. Separation anxiety commonly surfaces around six to eight months and is a normal developmental step, not a setback.
  • Three naps to two. Usually somewhere between six and eight months. If the third nap is still in place it starts pushing bedtime too late.

This one has its own guide too: The 8-Month Sleep Regression: Crawling, Standing and Object Permanence.

Is the 12-month regression a nap transition?

Almost never, and this is the most expensive misconception in the whole chart. A twelve-month-old refusing the morning nap is usually protesting, not ready. The genuine 2-to-1 transition typically happens between 13 and 18 months.

Dropping to one nap at twelve months usually produces overtiredness, which causes shorter naps and earlier morning waking, which looks like further evidence that the nap should go. That loop is hard to climb back out of. The rule worth keeping is two consistent weeks of resistance before you change anything — see the signals that actually indicate readiness.

What is usually happening at twelve months instead: walking practice, first words, a wake window that has quietly outgrown the schedule, and often a cold picked up from nursery.

The full case, and why dropping early costs months rather than weeks: The 12-Month Sleep Regression Is Not a Nap Transition.

The 18-month sleep regression

Eighteen months is where the real 2-to-1 nap transition, peak separation anxiety, a language explosion and the first molars tend to collide. It is also the age at which a toddler can climb, protest and negotiate, which changes what bedtime resistance looks like.

The 2-to-1 transition is the long one — commonly four to six weeks of some days needing two naps and some needing one. Alternating is easier than a hard switch, and a temporarily earlier bedtime absorbs the lost daytime sleep. Toddlers aged one to two need 11 to 14 hours per 24 hours including naps.

The 2-year sleep regression

By two, the drivers are behavioural and cognitive rather than physical. Imagination develops, which produces real fears about the dark and about being alone. Potty training often lands here. So does the move from cot to bed — and a toddler who can physically leave the bed will test whether that is allowed.

The response that works is the least interesting one: identical boundaries every night, delivered calmly. Negotiating at 2am is what extends this phase; consistency is what ends it. Moving to a bed before it is necessary tends to create the problem rather than solve it.

How long does a sleep regression last?

Two to six weeks for most, with the four-month change being the exception because the underlying shift is permanent. If disrupted sleep has run past six weeks with no improvement, the cause is usually something other than a developmental leap — most often a schedule that no longer fits, a habit that formed during the disruption, or an unresolved medical issue.

What actually helps

Do not change the system on night three. This is the single most useful piece of advice available, and it is what most sleep programmes agree on. Habits form quickly. If a rough week turns into a new routine, the routine is what persists — long after the leap has passed.

Check the wake windows before anything else. An overtired baby wakes more, not less. Most sleep that looks like a regression is a timing problem that has crept in as the baby aged. Our wake windows by age chart covers the ranges.

Bring bedtime forward temporarily. Thirty to sixty minutes earlier during a disrupted patch absorbs lost daytime sleep and stops overtiredness compounding.

Give the new skill daylight practice. A baby who gets floor time to crawl, pull up and practise getting back down does less of it at 2am.

Let them try to resettle before you go in. A brief arousal is not the same as being awake. Some babies close the gap themselves within a few minutes.

When it is not a regression

Milestone timing explains a lot, but not everything. Speak to your paediatrician or health visitor if disrupted sleep comes with any of the following:

  • Fever, ear-pulling, or waking with a distinct pained cry
  • Snoring, mouth breathing, gasping or pauses in breathing during sleep
  • Feeding refusal, or a drop off the growth curve
  • Loss of a skill your baby previously had
  • Disruption running well past six weeks with no improvement at all

Frequently asked questions

At what ages do babies go through sleep regression?

Most commonly at 4, 6, 8 to 10, 12, 18 and 24 months, with some toddlers having a further rough patch around three years. The ages track developmental milestones rather than a fixed schedule, so they vary between children.

How long does a sleep regression last?

Two to six weeks for most. The four-month change is different — it is a permanent shift in sleep architecture, so it improves as the baby learns to resettle rather than by passing.

Does every baby have a sleep regression?

No. Large-scale sleep data shows considerable variation between babies, and many never show a clear dip at any of the named ages. Sleep generally improves across the first year despite rough patches.

Is the 4-month sleep regression permanent?

The change to sleep cycles is permanent. The disrupted sleep is not. Once a baby can link cycles without full assistance, the wakings reduce — which for most families takes a few weeks.

Which sleep regression is the hardest?

Most parents report the four-month one as the hardest, because it is the first, it affects every sleep cycle, and it does not resolve on its own. The 18-month one tends to last longest because the 2-to-1 nap transition runs alongside it.

Should I sleep train during a regression?

Starting a new method in the middle of an illness or a major leap is harder than waiting. Once things have settled, sleep training is an option that improves sleep for many families, and the best method is the one you can apply consistently.

Is a 12-month regression a sign to drop to one nap?

Usually not. The genuine 2-to-1 transition typically happens between 13 and 18 months. Wait for two consistent weeks of resistance before changing the schedule.

Where to go from here

The Wake Windows Guide covers the timing system underneath all of this, and the Nap Transition Guide has staged schedules for each nap drop. If you are deciding whether to sleep train after a rough patch, the Gentle Sleep Training Guide sets out the no-cry-it-out approaches.

In this series: 4 months · 8 months · 12 months · signs your baby is ready to drop a nap · why naps are only 30 minutes · when to stop swaddling.

Free printables: the sleep regression chart and the wake windows chart. Or browse all baby sleep guides.

Sources

Educational only, and not a substitute for advice from your paediatrician or health visitor. Follow current AAP safe sleep guidance.

Back to blog