← Back to blog

The 4-Month Sleep Regression: What's Actually Happening

·3 min read

It's 2am. Your baby slept a six-hour stretch last month. Last night you were up at 11, 1, 3, and 5. You are lying there doing the math on how many total hours you've slept this week and the number is not okay.

You haven't done anything wrong. This is the 4-month regression, and it is real, it is brutal, and it happens to babies who were previously sleeping beautifully.

What actually changed

Around 3-4 months, a baby's sleep architecture permanently shifts. Before this, infants spend a disproportionate amount of time in deep sleep, which is why a 6-week-old can get drowsed on a boob and stay out for hours. After this shift, they cycle through lighter and deeper sleep stages the way adults do, surfacing briefly between cycles every 45-60 minutes or so.

The problem is they don't know how to get themselves back down. If they fell asleep in your arms, or at the breast, or in motion, they wake up in a still, dark crib and the conditions are completely wrong. So they cry. You come. You rock them back to sleep. You put them down. Eyes open. Repeat until sunrise.

The NHS notes that babies at this age are still developing sleep patterns and will often need help settling. This isn't a discipline problem. It's a developmental one.

It is not because you did something wrong

The parent who nursed to sleep, the one who used a pacifier, the one who rocked in a swing for 45 minutes then did it eight times and fell asleep mid-swing - none of them caused this. The regression happens on its own schedule regardless of what you've been doing.

What the previous sleep associations do affect is how hard the regression hits. If your baby needs specific conditions to fall asleep, they'll need those same conditions at each waking. That's the mechanic. Not a moral failure.

What helps (and what doesn't)

The actual solution, if there is one, is teaching your baby to fall asleep independently so that when they surface at 1am they can get back down without you. The AAP's sleep guidance recommends placing babies drowsy but awake as a way to help them learn to self-settle - which is easier to say at noon than to execute at 3am when you have work in four hours.

Sleep training is the structured version of this. The right age, method, and timing depends on your baby's weight, your own situation, and how much more of this you can take. If you are tracking feeds (TandemBaby, or a notebook, or whatever), look at whether your baby is eating enough during the day - genuine hunger at night is different from habit waking, and the data helps you tell them apart.

What does not help: waiting for it to resolve on its own without any change. Some babies do cycle out. Many don't, or not for months.

The question nobody asks directly

Several of the posts floating around about this right now are from people who are beyond exhausted and starting to feel something darker than tired. One person wrote "I just feel so damn done." Another started their work day at 3am because they couldn't get the baby down and had to shower.

That level of depletion is worth taking seriously on its own. Postpartum Support International has resources specifically on postpartum anxiety, which often spikes during sleep disruption and looks like lying awake catastrophizing even when the baby is finally down. If that sounds familiar, it's worth looking at.

Tonight

You can't fix the regression in one night. But if your baby is older than 4 months and at a healthy weight, putting them down drowsy but awake once, tonight, and seeing what happens - that's a start. It may go badly. It may surprise you.

The sleep you had last month is not gone forever. It just moved.

Sources

← Back to blog

Parenting in rhythm.