How to Get a Baby to Sleep Through the Night

Most healthy babies become capable of sleeping a long stretch at night somewhere around three to four months of age, but whether they actually do depends on a mix of biology, environment, feeding, and learned habits. Research shows that by three months, roughly 45% of infants can sleep five or more hours continuously at night, up from about 10% at five weeks. That shift is driven by the emergence of a circadian rhythm, which doesn’t exist at birth and takes weeks to develop. Understanding that timeline, and the practical strategies that work with it rather than against it, is what separates restless months from a household that finally gets some sleep.

When Babies Become Physically Capable of Long Sleep

Newborns lack a circadian clock. Their internal day-night cycle has to be built from scratch, and different pieces of it come online at different times. Body temperature rhythm appears within the first week of life, but the wake rhythm doesn’t become statistically significant until around day 45, roughly when melatonin production at sunset begins to ramp up. The sleep rhythm itself lags even further, not becoming significant until after day 56.¹ In practical terms, this means a baby under about eight weeks old simply does not have the biological hardware to distinguish night from day in any reliable way. Trying to train a two-week-old to sleep through the night is working against physiology.

By three months, the picture changes. A video-based study of London infants found that 45% were sleeping five or more continuous hours at night, while about a quarter of babies at each age studied were waking but settling themselves back to sleep without parental help.² That self-settling ability is a key distinction. Many babies who “sleep through the night” are actually waking briefly and putting themselves back down. The goal isn’t necessarily to eliminate all night wakings; it’s to help the baby manage them independently.

Behavioral Strategies That Have Evidence Behind Them

The phrase “sleep training” covers a range of techniques, two of which have the strongest research support for bedtime resistance and frequent night waking. Graduated extinction involves putting the baby down awake and waiting progressively longer intervals before briefly checking in when the baby cries. Bedtime fading temporarily shifts the baby’s bedtime later, to a point where the baby falls asleep quickly, then gradually moves it earlier over days or weeks. Both are considered well-established treatments for these common sleep problems.³

A randomized trial comparing these two approaches against a control group found that both produced better sleep at a 12-month follow-up, with no differences in emotional or behavioral problems, and no differences in secure versus insecure attachment.⁴ That trial directly addressed one of the biggest fears parents have about these methods: that letting a baby cry, even briefly and with check-ins, will damage the parent-child bond. The evidence at one year did not support that concern.

If you’re choosing between the two, graduated extinction tends to work faster but involves more crying upfront. Bedtime fading is gentler but slower. Neither requires leaving a baby to cry indefinitely without any response, a common misunderstanding. Most graduated extinction protocols have parents return at set intervals to offer brief comfort. The point is to give the baby practice falling asleep without being held, rocked, or fed to sleep, so that when they naturally wake between sleep cycles at 2 a.m., they can get themselves back down.

The Cortisol Question

A small but influential study created real anxiety among parents by showing a disconnect between infant behavior and stress hormones during an extinction-based sleep program. On the first night, babies cried and their cortisol levels tracked with their mothers’ levels, as you’d expect. By the third night, the babies had stopped crying at bedtime, but their cortisol levels remained elevated even though they were no longer showing distress.⁵

This study is worth knowing about, but it has serious limitations. It involved only 25 infants in a residential program, not a typical home setting. It measured cortisol on day one and day three only, not weeks or months later, so it cannot speak to whether cortisol normalizes over time. And the larger randomized trial mentioned earlier, which followed babies for a full year after graduated extinction or bedtime fading, found no adverse effects on stress responses, attachment, or behavior.⁴ The short-term cortisol finding is real, but it sits alongside a much larger body of evidence showing no lasting harm. Parents who are uncomfortable with any crying-based approach can use bedtime fading or other gradual methods, but the fear that standard sleep training causes permanent psychological damage is not supported by the outcome data available.

How Feeding Affects Night Sleep

The relationship between feeding method and infant sleep is more complicated than the old advice of “give a bottle of formula before bed.” A large analysis found that fully breastfed infants actually had longer total nighttime sleep at multiple ages compared to formula-fed infants, but they also woke more often between six and twelve months.⁶ A Chinese population-based study found that exclusively formula-fed babies had the highest overall sleep percentage per 24 hours, but the gap between formula and breast milk narrowed as infants got older.⁷

In other words, breastfed babies may wake more frequently but don’t necessarily sleep less overall. The frequent wakings are partly because breast milk is digested faster, and partly because nighttime nursing may serve purposes beyond pure nutrition. An evolutionary biology hypothesis proposes that night waking to breastfeed extends the mother’s period of lactational suppression of ovulation, spacing out pregnancies and improving infant survival in ancestral environments.⁸ Whether or not that hypothesis fully explains night waking, it’s a useful reminder that frequent waking in a breastfed baby is biologically normal, not a sign that something is going wrong.

Does Starting Solid Food Earlier Help

One of the most common pieces of grandparent advice is to start cereal early so the baby sleeps longer. For decades, pediatricians dismissed this, and the evidence was thin. But a secondary analysis of a large randomized trial offered some support for the idea. Infants who were introduced to solid foods from three months of age, alongside continued breastfeeding, slept about 17 minutes longer per night and woke less frequently than infants who waited until six months. The differences peaked around six months, and the early-introduction group had fewer reports of serious sleep problems.⁹

Seventeen extra minutes might not sound transformative, but it was a population average, meaning some babies gained more and some less. And the reduction in severe sleep problems was more meaningful for the families who were really struggling. That said, this was one trial, and the primary outcome of the original study was allergy prevention, not sleep. Current guidelines in most countries still recommend exclusive breastfeeding until around six months, and the sleep finding alone probably isn’t enough to override that for most families. But if you’re already planning to introduce solids, the evidence suggests starting a bit earlier, around four to five months with medical guidance, isn’t harmful and may modestly help sleep.

The Sleep Environment

A few environmental factors have research support for improving infant sleep, and they’re all relatively low-effort adjustments.

Light exposure during the day is one of the strongest signals for training a developing circadian clock. A scoping review found that more daytime light exposure was associated with improved daytime wakefulness, better sleep efficiency at night, and longer nighttime sleep.¹⁰ Another study quantified this, finding that exposure to illumination above 100 lux (roughly what you’d get near a bright window, not direct sunlight) for longer periods each day was associated with stronger circadian activity patterns.¹¹ The practical takeaway: get your baby into well-lit spaces during the day and keep lights dim in the evening. This isn’t about buying a special lamp. It’s about taking walks in daylight and not keeping the nursery cave-dark during daytime naps.

White noise has consistent evidence for shortening the time it takes infants to fall asleep and improving sleep onset, based on multiple randomized controlled trials.¹² It works by masking household sounds that cause brief arousals. Keep the volume moderate, roughly at the level of a running shower, and place the machine across the room rather than right next to the crib.

Swaddling, for babies who are not yet rolling over, has been associated with increased quiet sleep duration and fewer transitions between sleep states in infants who haven’t been swaddled before.¹³ The benefit comes from reducing the startle reflex, which can jolt a baby awake. Once a baby can roll, swaddling becomes a safety risk and should be stopped.

What It Means for Parents

Infant sleep problems don’t stay contained in the nursery. Two separate randomized trials found that behavioral sleep interventions for babies led to measurable improvements in maternal depression scores and self-rated sleep quality. One trial showed that mothers in the intervention group had significantly lower depression scores at both 10 and 12 months, along with better mental health scores on a standardized measure.¹⁴ Another trial found that for mothers who started with elevated depression scores, the sleep intervention led to significantly greater improvement compared to the control group, and that by two months the intervention mothers were more likely to rate their own sleep quality as “very good.”¹⁵

This matters because parental sleep deprivation compounds. A parent running on broken sleep for months makes worse decisions, has less patience, and is at higher risk for depression. Helping the baby sleep better isn’t just about the baby. It’s a legitimate mental health intervention for the whole household. If you’re reluctant to try any sleep strategy because you feel you should just tough it out, it’s worth knowing that the research consistently shows better outcomes for both parent and child when sleep problems are addressed.

When a Baby Won’t Respond to Normal Strategies

Some babies resist all the standard approaches, and there’s often a medical reason hiding behind the sleeplessness. Gastroesophageal reflux is one of the most common culprits. Reflux episodes trigger changes in sleep state regardless of whether the baby visibly spits up; even “silent” reflux episodes that don’t reach the mouth are equally capable of causing arousals and awakenings.¹⁶ Compared to population norms, infants with diagnosed reflux disease had dramatically higher rates of night waking: about 50% of those aged three to twelve months were waking more than three times a night, versus 13% of infants without reflux. And these babies showed significantly delayed onset of sleeping through the night.¹⁷

Signs that reflux may be disrupting sleep include frequent arching of the back, fussiness after feeding, wet burps, or a pattern of waking shortly after being laid flat. If your baby consistently wakes within an hour or two of being put down, especially soon after a feed, it’s worth raising with your pediatrician. Treating the reflux often resolves the sleep problem without any behavioral intervention at all.

Mouth breathing is another underrecognized issue. A cross-sectional study found that children who breathed through their mouths had more than four times the risk of sleep-disordered breathing compared to nasal breathers.¹⁸ In infants, this can be caused by nasal congestion, enlarged adenoids, or structural issues. If your baby consistently sleeps with an open mouth, snores, or seems to struggle to breathe through the nose, it’s worth an evaluation. Nasal breathing is not just more comfortable; it produces better-quality sleep.

Temperament and Realistic Expectations

Not every baby who resists sleep has a medical problem or needs a different behavioral strategy. Temperament plays a genuine role. Research following infants across their first year found that babies rated higher in negative temperament had more sleep problems, longer sleep latency (meaning they took longer to fall asleep), and more nighttime wakefulness.¹⁹ Interestingly, the same study found that maternal stress was associated with how long the baby slept during the day, suggesting that parent and infant characteristics affect sleep through different pathways.

A baby with a more reactive temperament may simply need more time and consistency to learn self-settling, and may never be the baby who gets put down and immediately conks out. This isn’t a failure of parenting or technique. Some babies are harder to settle because of their innate wiring, and recognizing that can prevent parents from spiraling through strategy after strategy, blaming themselves each time. If your baby is generally healthy, gaining weight well, and making progress even if it’s slow, the temperament factor is worth keeping in mind.

An Evolutionary Lens on Night Waking

Modern parents tend to treat sleeping through the night as a developmental milestone, something a baby should achieve by a certain age, like rolling over or sitting up. The evolutionary picture is less tidy. For most of human history, infants slept in close contact with their mothers and breastfed throughout the night. Consolidated nighttime sleep for an infant is, in evolutionary terms, a very recent expectation tied to industrialized sleeping arrangements and parental work schedules.

The evolutionary hypothesis that night waking is an infant strategy to maintain breastfeeding and delay the mother’s return to fertility frames the behavior as adaptive rather than dysfunctional.⁸ None of this means you shouldn’t work toward better sleep for your family. But it can be reassuring to know that a four-month-old waking twice a night is not broken. The question for most families isn’t how to eliminate all night waking as fast as possible, but how to reach a pattern that lets everyone in the house function. For some families, that means full sleep training. For others, it means gradually reducing one feed at a time, or just adjusting the environment and waiting for the circadian clock to finish maturing. There’s no single correct answer, but there is a large and consistent body of evidence showing that the options available are safe and effective when matched to a baby’s developmental stage.