top of page

The “Four-month sleep regression:” What is it, and what can be done about it?

(Revised and updated from an earlier version)

Four month sleep regression

At Baby Sleep Science we don’t like the term “four-month regression,” because the changes in sleep that occur around 3–4 months old aren’t really a regression at all; they result from a permanent change in sleep architecture.

The Science

Sleep is made of several stages. You've probably heard of "deep" sleep and rapid eye movement (REM) sleep, and there is also Stage 1 (the light sleep that makes up the transition from awake to asleep) and Stage 2 (a light sleep that makes up half of a typical night). Each of these stages of sleep is associated with the different benefits that come from a night of sleep, including learning and memory consolidation. When a baby is born, sleep is immature. For the first few months, babies don't enter all of the sleep stages. Instead, sleep is a fairly constant state, alternating between two types of immature sleep (sometimes called "active" sleep and "quiet" sleep).

A healthy newborn (free of colic, reflux, or other medical concerns) will sleep until one of three things happens: 1) hunger, 2) another need (wet/dirty diaper), 3) sleep is no longer needed. This pattern is difficult at first, but for most families sleep unfolds as you expect it to; your baby begins life with fragmented sleep and frequent feedings. Virtually every parent is prepared for disrupted sleep at this time. Over the course of the first few months or so, the interplay between sleep and those waking factors is intuitive. As you get to know your baby, you start to learn when he or she will be sleepy, and you anticipate that need. As your baby needs less frequent feedings, he or she will naturally start to sleep longer stretches during the day and at night. These stretches might start with three hours, but over the course of a few weeks, they grow to four hours, then five hours, and so on. Some lucky parents might even see their baby naturally sleep through the entire night during this time (10–12 hours straight!). This is a beautiful progression that feels very natural and appropriate.


Need More Help?

We offer several options for optimizing your child's sleep:

SLEEP TRAINING CLASS (for well babies 6-15 months)

PHONE CONSULTATIONS (with one of our sleep experts)

THE SECRET TO NAPS (downloadable e-book)


Unfortunately, somewhere around 3–4 months everything changes. Around this time, sleep matures from being newborn sleep to a dynamic progression between sleep stages. As your baby grows, deep sleep will begin to dominate the beginning of the night, the middle of the night will be dominated by Stage 2, and REM sleep will dominate the last third of the night. However, these stages aren't continuous, they flow together in sleep cycles that contain a little bit of different stages of sleep. Sleep cycles last about 60 minutes during the night and are marked by a brief waking that happens at the end of the cycle. This brief waking is the cause of most “sleep problems.

The brief waking that happens at the end of a sleep cycle is meant to be protective (adults have it too). It allows your baby to briefly check in with the environment and ask, “Am I ok?” When we fall asleep, we expect our environment to remain constant throughout the night, so if your baby has fallen asleep in your arms and you then transferred him or her to the crib, the crib will feel wrong when s/he hits the checking wake-up. This means that after every sleep cycle, your baby will probably wake up, and you can interpret his or her cries as, “Hey, Mom! Dad! Something is wrong here!”

If you don’t know that this sleep “regression” is coming, then you’ll probably do what you normally do when your baby wakes up. For many families, the default response is to offer a feeding, but for some, it will involve rocking, bouncing, or a pacifier replacement. Since what you do at bedtime is how your baby knows how to go to sleep, s/he will fall back asleep for another hour or two and then wake again at the end of the next sleep cycle. At that point, you’ll feed/rock/bounce/replace the pacifier again. If you have been feeding your baby overnight, after a few nights of this, you’ll convince yourself that this is a growth spurt or teething or that your child getting sick, but the waking usually doesn’t resolve on its own (hence why “regression” is the wrong term). At this point, you’ll have reinforced that feeding/rocking/bouncing/pacifier replacement is needed to get to sleep. If offering a feeding is your default response, then your baby will have started to take in more calories during the night. Offering multiple "extra" feedings per night also means that the wakings now have a dual nature: 1) your baby is waking due to the sleep association, but 2) he or she is also now genuinely hungry at night and probably eating less during the day. Many babies get stuck in this reversed feeding pattern because another change that happens around 4 months is that most babies become very interested in the world around them and become very distracted during daytime feedings. This reduction in daytime feeding paired with increased night feeding means that you’re stuck in a pattern that can be difficult to fix.

During this phase, most babies also become chronic cat-nappers and start to have short naps of 30-45 minutes. We describe more about these nap changes in our series on naps.

How can you prevent or minimize the impact of the four-month regression?

Teach your baby to fall asleep without your help

If your baby is younger than 3–4 months and hasn’t yet had a sleep regression, then the best thing you can do is give your baby practice falling asleep in the place where you would like him or her to sleep all night. Start by creating a little nap and bedtime cue (e.g., singing a specific song, rocking, bouncing, comforting), and when your baby is tired, put him/her in the sleep space awake. Your baby may be confused and frustrated by this, and that’s fine; just pick him/her up and repeat your sleep cue. You can rock/nurse to sleep, or you can try to put him in the crib again, awake. At this point in a baby’s life, you shouldn’t try to force the crib if your baby isn’t ready to fall asleep there on his/her own, but you do want to teach him/her that the crib is a reasonable place to sleep. If your baby never even has the chance to fall asleep there, then there’s no reason he/she should accept the crib; from his/her perspective, in your arms is the right place to sleep. In little babies it’s all about offering the chance; you don’t have to do it every day or at every sleep bout, but it is important to think about this and start to make changes in anticipation of the four-month regression.

Learn how to distinguish true hunger from the regression

You certainly don’t want to deny your baby feedings when he or she is hungry. It can be really difficult to disentangle true feeding needs from sleep associations, but it can be done if you pay close attention to your baby’s pattern. If your baby normally sleeps 7:00 pm–1:00 am, eats and then sleeps 1:00 am–4:00 am, eats and then sleeps until 6:00 am, then you have a good sense of his or her true needs. If one day, somewhere between the 3 to 4 month mark, he or she is up at 1:00 am and 2:30 am, you can begin to suspect that this might be the beginning of the “regression.” If your baby hasn’t suddenly increased daytime feeding (which would indicate a growth spurt), then it’s probably the regression creeping in on you. Since you can feel pretty confident that your baby doesn’t need to eat at both 1:00 am and 2:30 am, this would be a good time to teach your baby to go back to sleep in new ways. For example, feed him or her at 1:00 am as usual, then at 2:30 simply rock, bounce, or hold him/her until he/she goes back to sleep, rather than feeding. The goals are to teach your baby that there isn’t just one way to fall asleep and to keep nighttime calorie intake stable. The rocking can also become a sleep association, so if your baby continues to wake up and isn’t hungry, but wants to be rocked to sleep, use the suggestions above to teach him or her that the crib is a reasonable place to be (offer the sleep cue, put your baby in the crib awake, pick your baby up and offer the sleep cue, put your baby in the crib awake, etc.).

How do you fix issues that emerge as part of the four-month regression?

Reduce "extra" night feedings, if needed

If your baby is going through the 3–4 month sleep regression, then you need to evaluate the extent of the problem. Is your baby really eating a lot more at night than before the regression? Is he or she eating a lot less during the day? If not, then move on to the next section. If so, then you will need to taper feedings at night before doing anything else (see our blog post and webinar or check out our book or sleep class to learn gentle ways to reduce night feeding). If your baby is truly hungry, then you don’t want to ask him or her to go back to sleep. He/she will be frustrated and so will you. First, determine how long your baby eats (or how many ounces he/she is taking if bottle feeding) and just reduce “extra” feedings a little each night. During the day, make sure you feed your baby in a quiet, distraction-free environment in order to ensure he/she gets nice full feedings. Consider cluster feeding at the end of the day in order to ensure he or she has had ample opportunity to eat as bedtime approaches. Once your baby is eating just a tiny bit or for a very short duration, you can start rocking him or her through those extra wakings rather than offering a feeding. For some babies, this will lead to great improvement in sleep. However, many babies will start to depend on rocking/bouncing/snuggling to get back to sleep. If that happens, then move on to the step below.

Teach your baby how to fall asleep without your help

If your baby relies on your help to fall asleep through rocking/bouncing/pacifier replacement, you will need to start putting your baby in the crib awake at bedtime (you can use the simple strategies that we described above). The suggestions that we outlined for “not yet regressed” babies will work to teach older babies to go back to sleep too. Depending on your baby’s age, you may be able to do something a little more structured at bedtime and during the night if you want the process to go faster. We generally don’t recommend really aggressive sleep training for babies going through this regression. We also don’t think there is one right way to do sleep training. You may opt for a fast approach, or you may opt for a very gradual adjustment. Your parenting style matters, as does your baby's personality. There are many ways to get to healthy sleep. If you need more help, consider our book on navigating the four-month regression, which describes several other strategies for working through the four-month sleep regression. Our sleep class is geared towards babies over six months, but the first hour covers schedules, night weaning, and developing pre-sleep routines, which are appropriate for younger babies. Lastly, if you just aren't sure where to start, then you can also book a sleep consultation with us and we will work with you to develop a plan to improve your baby's sleep.

A Final Note

If your baby is able to put him- or herself to sleep at night and is still waking several times a night to eat, then it’s probably because his/her body is expecting calories at night now (the four-month regression led to an increase in nighttime calorie intake, and now they are “stuck” in that pattern). That said, sometimes a baby can fall asleep independently and have a sleep association hang on during the night. This is why sleep is so hard! There isn’t just one thing that causes trouble. Sleep problems are usually caused by complex layers of issues, and developmental changes such as the four-month regression cause new issues like extra night feedings. Please search our other blog topics to help with your problem-solving.


Iglowstein, I., Jenni, O.G., Molinari, L. and Largo, R.H., 2003. Sleep duration from infancy to adolescence: reference values and generational trends. Pediatrics, 111(2), pp.302-307.

Jenni, O.G., Borbély, A.A. and Achermann, P., 2004. Development of the nocturnal sleep electroencephalogram in human infants. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 286(3), pp.R528-R538.

Jenni, O.G. and Carskadon, M.A., 2007. Sleep behavior and sleep regulation from infancy through adolescence: Normative aspects. Sleep medicine clinics, 2(3), pp.321-329.

Sankupellay, M., Wilson, S., Heussler, H.S., Parsley, C., Yuill, M. and Dakin, C., 2011. Characteristics of sleep EEG power spectra in healthy infants in the first two years of life. Clinical neurophysiology, 122(2), pp.236-243.

Jenni, O.G., Deboer, T. and Achermann, P., 2006. Development of the 24-h rest-activity pattern in human infants. Infant behavior and development, 29(2), pp.143-152.

Louis, J., Cannard, C., Bastuji, H. and Challamel, M.J., 1997. Sleep ontogenesis revisited: a longitudinal 24-hour home polygraphic study on 15 normal infants during the first two years of life. Sleep, 20(5), pp.323-333.

bottom of page