- Sleep Regression
- 10mins read
- Aucun commentaire
You were just starting to feel like you had figured something out.
Not perfect sleep as there is no perfect sleep in the first year, and anyone who tells you otherwise is either lying or has a very selective memory. But something. A rough pattern. A stretch of three or four hours that you had started to quietly, cautiously count on. A bedtime routine that seemed to be landing. A baby who, more nights than not, was settling.
And then, somewhere around the end of month three or the beginning of month four, all of it disappeared.
Baby is waking every hour. Every forty-five minutes sometimes. The naps that were starting to consolidate have shrunk back to twenty minutes. Bedtime, which had actually been getting easier, is suddenly a forty-five-minute operation again. You are more exhausted than you were in the newborn days, which feels impossible, but the sleep deprivation in those days was somehow easier to bear because you expected it. You did not expect this one.
You are sitting with your baby at 2am, 3am, 4am, wondering what you did wrong. Wondering if the routine you built caused this somehow. Wondering whether this is permanent. Wondering, if you are completely honest, whether you can keep doing this.
I want to tell you what is happening. Because once you understand it, like really understand it, not just the label but the actual biology of it, something shifts. It does not make you less tired but it makes you less frightened, and less frightened is a meaningful improvement at 4am.
You were just starting to feel like you had figured something out.
Not perfect sleep as there is no perfect sleep in the first year, and anyone who tells you otherwise is either lying or has a very selective memory. But something. A rough pattern. A stretch of three or four hours that you had started to quietly, cautiously count on. A bedtime routine that seemed to be landing. A baby who, more nights than not, was settling.
And then, somewhere around the end of month three or the beginning of month four, all of it disappeared.
Baby is waking every hour. Every forty-five minutes sometimes. The naps that were starting to consolidate have shrunk back to twenty minutes. Bedtime, which had actually been getting easier, is suddenly a forty-five-minute operation again. You are more exhausted than you were in the newborn days, which feels impossible, but the sleep deprivation in those days was somehow easier to bear because you expected it. You did not expect this one.
You are sitting with your baby at 2am, 3am, 4am, wondering what you did wrong. Wondering if the routine you built caused this somehow. Wondering whether this is permanent. Wondering, if you are completely honest, whether you can keep doing this.
I want to tell you what is happening. Because once you understand it, like really understand it, not just the label but the actual biology of it, something shifts. It does not make you less tired but it makes you less frightened, and less frightened is a meaningful improvement at 4am.
When does it start and how will you know
The four-month sleep regression typically occurs between three and five months of age, most commonly near four months. Every baby is different, some experience it as early as twelve weeks, and some barely seem affected at all. My baby girl was at 5.5 months and trust me, I still remember this specific night where everything changed.
The signs are specific enough that most parents recognise them immediately once they know what they are looking for.
The night waking is the most obvious as the baby who was sleeping three or four hour stretches suddenly waking every forty-five minutes to an hour, at times with a consistency that begins to feel mechanical. A new kind of tired arrives with this pattern because each waking is a full reset where you settle bavy, you lie back down, you are close to sleep, and then baby is awake again. It is the near-misses that break you rather than the total hours lost.
The naps collapse at the same time as the nights deteriorate. The forty-five-minute nap is the giveaway as your baby is waking at the end of exactly one sleep cycle, every nap, without fail. Baby is surfacing at the inter-cycle arousal and cannot bridge it in the day any more than they can at night. You spend twenty minutes settling them for twenty minutes of sleep, which is both exhausting and demoralising.
Alongside the sleep changes, you may notice an uptick in fussiness and clinginess that has nothing to do with sleep as your baby is also going through significant neurological development at this stage, becoming dramatically more alert and socially aware, and the stimulation of that can make them harder to settle even during the daytime.
Increased feeding at night is common too, partly because they are genuinely hungrier during a growth phase, and partly because the feed has become the settling tool baby calls for at every inter-cycle arousal. Both reasons are real. Neither means your supply is insufficient.
The question you are probably asking: did I cause this?
No.
I want to be direct about this because it is the question most parents arrive at in the middle of this regression, and the answer deserves to be clear rather than hedged.
The four-month sleep regression is not caused by your feeding to sleep. It is not caused by your responding to your baby’s cries. It is not caused by contact naps or cosleeping or any of the other things that parenting culture sometimes implies are creating problems. It is caused by neurological maturation: the completely healthy, completely normal, completely inevitable maturing of your baby’s brain from newborn sleep architecture to something more adult-like.
A baby’s sleep architecture significantly changes around three months, and that change happens on a developmental timeline that has nothing to do with what you did or did not do in the weeks before. Some babies who were never fed to sleep hit this regression just as hard as babies who were. Some babies who were exclusively contact-napped sail through it. The causation that parenting content often implies between early settling habits and regression severity is not supported by the research.
What is true and worth understanding, is that how a baby has learned to fall asleep does affect how they manage the inter-cycle arousals once the regression is underway. A baby who falls asleep at the breast will, on surfacing between cycles, look for the breast. A baby who falls asleep with rocking will look for rocking. This is not a moral failing or a bad habit. It is association, the same way you probably cannot sleep well in a room that is light and noisy because you associate sleep with darkness and quiet. The associations are normal. Whether they are creating a problem you want to change is a separate question; one that belongs to you, not to any parenting website.
How long does it last
Most babies improve within two to six weeks. That is the honest answer, and it is the range for a reason, it is genuinely variable.
A four-month-old who is rocked to sleep and suddenly starts frequently waking during the night may continue to see this pattern until they learn to fall asleep independently at bedtime which could be a week, or months, depending on a variety of factors.
What this means in practice: the regression itself (the acute, brutal phase of hourly waking and collapsed naps) typically settles within four to six weeks as your baby’s nervous system adjusts to its new sleep architecture. What may persist beyond that is the waking pattern that develops as a result of the regression, if the settling strategy you have been using involves significant parental input at every arousal.
Two to six weeks of difficult sleep is manageable. Especially once you understand that the end of it is not contingent on anything you need to fix, it is contingent on developmental consolidation that happens regardless.
What actually helps and what does not
The goal during a regression is not to solve it. It is to survive it while preserving the conditions your baby will return to when it passes.
Keep the bedtime routine exactly as it is. The regression is not happening because of your routine. It is happening despite your routine. Abandoning the routine during a regression means your baby has to rebuild that predictable pathway to sleep once the regression passes, which extends the disruption. The bath, the feed, the song, the dimly lit room, whatever your sequence is, keep it. Consistency is the scaffolding that remains even when everything else feels uncertain.
Watch their wake windows. At four months, most babies manage approximately ninety minutes to two hours of wakefulness between sleeps before overtiredness sets in. Overtiredness is one of the most common things that makes a regression worse, because a cortisol-flooded, overstimulated baby is harder to settle and sleeps more lightly. Watching the clock loosely and watching their cues closely like the eye rub, the glaze, the sudden loss of interest in what was engaging their moments before and acting on those cues promptly reduces the overtiredness layer on top of the regression layer.
Offer the drowsy-but-awake opportunity at bedtime. This is not sleep training. It is a gentle experiment putting them down slightly before they are fully asleep and seeing whether they can complete the journey to sleep in their own sleep space. Some nights they will. Some nights they will not, and you pick baby up and try again another time. The point is to give baby the opportunity, not to insist on an outcome. If baby finds the beginning of that skill now, the inter-cycle arousals at night will be easier to bridge as the regression eases.
Respond to the night wakings the way you can sustain. There is no version of this advice that tells you what that response should be. Some families find that a feeding-and-back-to-sleep approach at every waking is sustainable and they are fine to continue it. Some families find that the cumulative sleep loss is damaging their mental health and they want to begin thinking about gentle settling support. Both responses are valid. What is not true is that one response will end the regression faster than the other as the regression ends on its own developmental timeline. The settling approach affects your experience of it, not its duration.
Accept help. Mauritius families often have extended family close by. The four-month regression is the moment to use that proximity deliberately rather than managing in isolation. Ask someone to take the baby for three hours in the morning so you can sleep. Ask your partner to do one night waking while you wear earplugs. Sleep deprivation at this level affects your mental health, your physical health, your relationship, and your ability to parent in the ways you want to. You are not less of a mother for needing to sleep. You are a human being with physiological requirements, and right now those requirements are not being met.
What not to do: the things that make it harder
Changing your approach every three nights is one of the most common mistakes parents make during regressions and the most understandable. When something is not working, the instinct is to try something else but inconsistency during an already-disrupted period makes settling harder, because your baby is relying on predictability more than ever and getting less of it. Choose an approach, give it five days, and then assess.
Comparing your baby to other babies is the other one. The Facebook group where someone announces their four-month-old slept through the regression and was perfectly settled by week two is not a benchmark. It is one baby’s experience, almost certainly missing significant context, and it has nothing useful to tell you about yours. Your baby’s regression is theirs. It will resolve on their timeline.
The thing worth holding onto at 4am
Your baby is not broken. Your parenting has not failed. The sleep that was working last month is not gone forever, it has evolved into something more complex, and the work now is helping them build the skills to navigate that complexity.
These are predictable shifts in sleep patterns. They happen because of what is going right, not what is going wrong.
The baby waking every forty-five minutes because their brain has permanently matured into a more sophisticated form of consciousness, that is a remarkable thing, when you are not too tired to see it that way. Their brain built new architecture. Sleep cycles that did not exist before now exist. The developmental leap that is disrupting your nights is the same one that is making their laugh properly for the first time, track your face across the room, reach for your finger with something approaching intention.
The nights are hard. They will get better. The regression is temporary. The development is permanent.
You are doing fine.
Your Next Read:
References: Sleep Foundation — 4-Month Sleep Regression (updated July 2025). sleepfoundation.org. Medical News Today — Sleep Regression Stages. medicalnewstoday.com. Huckleberry — Navigate Sleep Regressions (updated March 2026). huckleberrycare.com. Gilchrist A. et al. (2025) — Maturation of infant sleep during the first six months: a mini-scoping review. Frontiers in Neuroscience. Butler B. et al. (2024) — Association between infant sleep and cognitive and psychomotor development: a systematic review. Sleep.
Disclaimer: This article is for informational purposes only and does not constitute medical advice. If you have concerns about your baby’s sleep, health, or development, contact your paediatrician. If sleep deprivation is significantly affecting your mental health, speak to your doctor. Maternal sleep deprivation is a health issue that deserves medical attention.