Baby and Toddler Sleep: What Is Normal at Each Age and the Methods With Evidence

Scroll any parenting feed as of September 2026 and two claims collide within seconds. One video insists that letting a baby fuss for even a few minutes rewires the stress system for life. The next sells a weighted sleep sack as the secret to twelve uninterrupted hours. Both clips rack up millions of views, and both leave exhausted parents at 3 a.m. feeling they are failing a test nobody explained.

The science of baby sleep is calmer than the comment sections. Safe-sleep guidance was substantially revised in 2022, long-term follow-up of sleep training trials has now reached school age, and acoustic engineers have measured what white noise machines actually do beside a crib. The picture that emerges is specific, reassuring in places and cautionary in others.

This guide walks through what normal looks like from the first week to the third birthday, which methods have real evidence behind them and how strong that evidence is, and the handful of warning signs that should send you to a pediatrician rather than a search bar.

What changed recently in baby sleep guidance

Three developments explain why baby sleep is suddenly everywhere again.

First, the guidance itself moved. In June 2022 the American Academy of Pediatrics released its first full update to safe sleep recommendations since 2016. It kept the core advice (back to sleep, firm flat surface, no soft bedding) but added an explicit statement that weighted blankets, weighted sleep sacks and weighted swaddles are not recommended for infants, and tightened the language on inclined sleepers. The NIH Safe to Sleep campaign and the NHS pages linked below reflect the same principles. Later that year the US Safe Sleep for Babies Act took effect, removing inclined infant sleepers and padded crib bumpers from the market as of November 2022. Secondhand versions still circulate online, which is why the warnings keep resurfacing.

Second, the long-term data on sleep training matured. The five-year follow-up of a large Australian trial, published in Pediatrics in 2012, and a 2016 randomized trial that measured infant cortisol and attachment directly are now the backbone of most clinical advice. They did not find harm, and that finding sits uneasily beside viral claims, so the debate reignites every few months.

Third, the market changed. Sleep trackers, smart bassinets and app-based coaching programs multiplied, and each new product arrives with marketing that outruns its evidence. A neutral reading of the literature helps here: products are tested against safety standards, not usually against outcomes such as how many hours a baby sleeps.

One thing has not changed. Sudden unexpected infant death still claims about 3,500 infants a year in the United States, according to CDC surveillance, and back-sleeping campaigns since 1994 are credited with roughly halving the rate of sudden infant death syndrome. That is the number worth remembering when a trend tempts you to bend a safety rule.

How do babies typically sleep in the first months?

A newborn’s sleep looks nothing like an adult’s, and that mismatch is the root of most early worry.

Adults cycle through light sleep, deep sleep and dreaming (REM) sleep roughly every 90 minutes. Newborns run cycles of about 50 to 60 minutes, and around half of each cycle is active sleep, the infant version of REM. Active sleep is noisy: twitching limbs, fluttering eyelids, grunts, sucking movements and the occasional cry that stops on its own. Parents often rush in during active sleep and wake a baby who was never awake. Waiting a minute to see whether the sounds settle is one of the most useful habits of the first six weeks.

Timing is scattered because the circadian rhythm, the internal clock that ties sleepiness to darkness, has not matured. Melatonin production in the baby’s own body becomes rhythmic at roughly 2 to 3 months. Until then, a newborn does not know night from day and may sleep as soundly at noon as at midnight.

Hunger sets the rest of the pattern. A newborn stomach holds very little, so most babies wake to feed every 2 to 3 hours, and the NHS notes that breastfed newborns in particular may feed even more often in the early weeks. Total sleep adds up to roughly 14 to 17 hours a day, but in pieces, rarely longer than 2 to 4 hours at a stretch.

By 6 to 8 weeks many babies begin to show a longer stretch, often in the first part of the night. By 3 to 4 months, deeper non-REM sleep becomes more adult-like, cycles lengthen and the first real pattern of a longer night and three or four daytime naps tends to appear. None of this requires teaching. It is maturation, and it arrives on the baby’s timetable rather than the family’s.

How much sleep does a baby need? A baby sleep schedule by age

Sleep need is a range, not a target, and healthy babies at the same age can differ by several hours a day. The figures below combine the American Academy of Sleep Medicine consensus (which Mayo Clinic and MedlinePlus cite) with typical nap patterns seen in observational studies.

Age	Total sleep per 24 hours	Typical daytime naps	What is usually normal
0–3 months	About 14–17 hours	Many short naps, no fixed count	Waking every 2–3 hours to feed, day and night
4–11 months	12–16 hours including naps	3 naps shrinking to 2 by about 9 months	One longer night stretch emerging; night feeds still common
1–2 years	11–14 hours including naps	2 naps moving to 1 between 12 and 18 months	Bedtime resistance, occasional night waking
3–5 years	10–13 hours including naps	1 nap, often dropped between 3 and 5 years	Nightmares and night terrors peak

A schedule is a description, not a prescription. A 5-month-old who sleeps 12 hours and seems content is not short of sleep because a chart says 14. The clinically meaningful signs are how the baby behaves awake: feeding well, gaining weight, alert for stretches of play and recovering from fussiness with comfort.

Wake windows, the stretches a baby can comfortably stay awake, lengthen from roughly 45 to 90 minutes in the newborn period to 2 to 3 hours by late infancy. Watching the window helps parents catch sleepy cues before overtiredness, but strict minute-counting is not something trials have tested. Treat the numbers as a rough compass.

When do babies sleep through the night, honestly?

The phrase hides a definition problem. Researchers usually count a 6 to 8 hour stretch as sleeping through, which may run from 7 p.m. to 2 a.m. Parents mean 7 p.m. to 7 a.m. The gap between those two meanings explains a lot of frustration.

Observational data are humbling. A Canadian cohort published in Pediatrics in 2018 followed several hundred infants and found that at 6 months, 38 percent had not yet slept 6 hours in a row on a typical night, and 57 percent had not managed 8 hours. At 12 months, 28 percent still were not sleeping 6 consecutive hours. Those infants were developing normally. The study found no link between consolidated sleep at 6 or 12 months and later mental or motor development.

Mayo Clinic puts the practical milestone this way: most babies are physically capable of sleeping a longer stretch without feeding by around 6 months, but capable is not the same as doing so. Night waking in the second half of the first year is often about sleep associations rather than hunger. A sleep association is whatever a baby has learned to need in order to fall asleep, such as feeding, rocking or a parent’s presence. Because every child surfaces briefly between sleep cycles, a baby who falls asleep only while being rocked will tend to call for rocking at each surfacing.

The so-called 4-month sleep regression is the moment this becomes visible. Sleep architecture matures, light sleep increases and the baby notices that the conditions at 2 a.m. do not match the conditions at bedtime. It is not a regression in development, and it is not caused by anything a parent did.

Night feeds remain appropriate for many babies well into the first year, particularly for breastfed infants and those with slower weight gain. Dropping them is a conversation with your pediatrician, not a milestone to chase.

What are the signs that my baby is fussy or overtired?

Babies broadcast tiredness in a fairly consistent sequence, and the earlier a parent catches it, the easier settling tends to be.

Early cues are quiet. The baby’s gaze drifts away from faces and toys, the eyelids droop, and movements slow. A yawn, a slight redness around the eyebrows and a loss of interest in a previously fascinating ceiling fan are classic. This is the ideal moment to begin winding down.

Mid-stage cues get louder: rubbing eyes or ears, pulling at hair, arching the back, clenching fists, and a whinier, more insistent cry. Many babies become clingy or want to feed for comfort rather than hunger.

Late cues look paradoxically like energy. An overtired baby may become wide-eyed, flail, laugh in a high-pitched way and then crash into inconsolable crying. Stress hormones rise when sleep is delayed, which is why an overtired infant is often harder to settle than one caught at the first yawn.

Fussiness is not always about sleep. The NHS lists hunger, a wet or dirty diaper, being too hot or cold, wind, overstimulation and simply wanting to be held among the common reasons a baby cries. Normal crying peaks at around 6 to 8 weeks of age, often totaling 2 to 3 hours a day, and declines by 3 to 4 months. Late-afternoon and evening fussing, sometimes called the witching hour, is common and does not indicate a problem with the day’s naps.

Two situations deserve a different response. Crying that is high-pitched or unusual for your baby, and crying accompanied by fever, vomiting, poor feeding or lethargy, is not a sleep issue. Those signs belong in the section on when to see a doctor, and they should not be waited out.

Safe baby sleep: the rules that reduce the risk of SIDS

Sudden infant death syndrome (SIDS) is the unexplained death of a seemingly healthy baby under one year, usually during sleep. It is rare, it is not anyone’s fault, and the sleep environment is the single most studied way to lower the risk. The evidence here is observational (case-control studies across many countries) but consistent and large, which is why NIH, CDC, the NHS and the WHO give near-identical advice.

• Back to sleep, every sleep, until the first birthday. Side and stomach positions roughly double to triple the risk in pooled studies. Once a baby can roll both ways independently, it is fine to leave them in the position they choose.
• A firm, flat surface with a fitted sheet and nothing else. No pillows, quilts, bumpers, positioners, loose blankets or soft toys. Sleep sacks replace blankets.
• Room sharing without bed sharing for at least the first 6 months. Sharing a room is associated with a lower risk; sharing a sleep surface raises it, especially under 4 months, with a parent who smokes or has had alcohol or sedating medicines, or on a sofa or armchair.
• No inclined sleepers, swings or car seats for routine sleep. If a baby falls asleep in one, move them to a flat surface when practical.
• Offer a pacifier at sleep times once breastfeeding is established; the association with lower risk is consistent across studies, though the mechanism is uncertain.
• A smoke-free pregnancy and home, avoiding overheating, breastfeeding where possible and routine immunizations are each associated with lower risk.

Weighted sleep products deserve a specific mention because they trend. No controlled trial shows they help infants sleep, and safety authorities advise against them because added weight on a small chest can impair breathing and a weighted garment makes rolling harder to recover from. Monitors that track breathing or oxygen are marketed for peace of mind; no evidence shows they prevent SIDS, and they are not a substitute for a safe surface.

What the evidence actually says about baby sleep methods

Grading the evidence is the only honest way through this topic, so here is the hierarchy applied to baby sleep.

Strongest: randomized controlled trials. Behavioral sleep interventions for infants over about 6 months have been tested in more than 50 trials. A 2006 review in the journal Sleep found that 94 percent of studies reported improvement, with roughly 80 percent of treated children showing clinically meaningful change. The 2016 Australian trial of 43 infants aged 6 to 16 months randomized families to graduated extinction (timed check-ins that lengthen), bedtime fading (temporarily moving bedtime later to match when the baby actually falls asleep) or an education-only control. Both active methods cut the time to fall asleep and the number of night wakings compared with control. Infant afternoon cortisol, a stress hormone measured in saliva, was lower in the treatment groups, not higher, and at 12 months there were no differences in attachment security or emotional and behavioral problems.

Strong for long-term safety: the 2012 Pediatrics follow-up of 326 children, five years after a trial of graduated extinction and camping out, found no differences between intervention and control groups in child mental health, sleep, stress regulation measured by cortisol, or the parent-child relationship. Benefits for maternal depression seen at 2 years had evened out by 6 years, which tells you the method is safe rather than magical.

Moderate: observational data. Bedtime routines, room sharing, and early sleep-wake patterns come mostly from large cohorts. Associations are consistent but cannot prove cause.

Weakest: expert opinion and tradition. Specific wake-window tables, nap transition timings, lullaby playlists and most app-based schedules fall here. They may be sensible, but they have not been tested against doing nothing.

Two caveats run through everything. Trials have mostly enrolled infants 6 months and older, so there is no trial evidence to support formal sleep training before that age. And trials measure averages; a method that works for 80 percent of families leaves one in five still looking for another approach.

Sleep training methods compared: what each involves and how strong the evidence is

The methods differ far less than their reputations suggest. All of them teach a baby to fall asleep under the same conditions that will be present at 2 a.m. The table summarizes what trials and reviews support.

Method	What a parent does	Evidence strength	Notes
Bedtime routine plus consistent timing	Same 20–30 minute sequence nightly, dim light, bed at a regular time	Strong observational, some trials	Foundation for every other method; safe from birth
Graduated extinction (checks at increasing intervals)	Put baby down awake, return at timed intervals to reassure briefly without picking up	Multiple randomized trials; 5-year safety follow-up	Best-studied method; typically improves within 1–2 weeks
Unmodified extinction	Put baby down awake and do not return until morning unless there is a safety concern	Randomized trials show effectiveness	Hard for many parents to sustain; same outcomes as graduated versions
Bedtime fading	Move bedtime later to the time the baby actually falls asleep, then shift earlier in small steps	One randomized trial, several smaller studies	Involves little crying; useful when a baby lies awake for long stretches
Camping out (chair method)	Sit beside the crib and move the chair farther away over successive nights	Randomized trial evidence	Slower, gentler option; part of the 2012 follow-up
Pick up, put down	Lift a crying baby, settle, return to crib awake, repeat	Expert opinion only	Widely recommended, not formally tested
Scheduled awakenings	Wake baby shortly before the usual waking time, then stretch intervals	Small trials	Rarely used; parents find it counterintuitive

What this table does not say matters as much as what it does. No trial has shown that one effective method produces better child outcomes than another, so the right choice is the one a family can follow consistently for two weeks without resentment. Starting before 6 months is not supported by trials, and the decision to begin at all is reasonable to discuss at a well-child visit, especially if a baby was premature, has reflux or is gaining weight slowly.

Is sleep training harmful? Attachment, cortisol and the viral claim

The claim circulating online is precise: letting a baby cry floods the brain with cortisol, damages attachment and causes lasting harm. It deserves a precise answer.

Cortisol is a stress hormone that rises with any demand, including hunger, a bath or an exciting game, and falls again. Chronic, unrelieved high cortisol in early life is linked to later problems, mainly in studies of severe neglect and institutional care. Those settings involve months or years of unresponsive caregiving, not a loving family running timed check-ins for a week.

The 2016 randomized trial is the only study to measure infant cortisol directly during sleep training. Afternoon salivary cortisol was lower in the graduated extinction and bedtime fading groups than in controls after the intervention, consistent with the babies being better rested. Attachment was assessed at 12 months with the Strange Situation procedure, the standard observational test of infant-parent attachment, and showed no difference between groups. Parent-reported emotional and behavioral problems were also no different.

A smaller observational study often cited in the viral videos found that, during the first nights of a sleep program, infant cortisol stayed elevated while mothers’ cortisol fell. That study had no control group and measured only three nights, so it cannot tell us whether the pattern differs from a baby’s ordinary bedtime protest or whether it persists.

The five-year follow-up adds the longest view available: no differences in behavior, emotional health, sleep problems, cortisol or the quality of the parent-child relationship at age 6.

Honesty cuts both ways. Trials are of modest size, they enrolled mainly families who sought help, and they cannot exclude small effects. They do show, with reasonable confidence, that the dramatic harms described online have not materialized in the children actually studied. Parents who prefer to avoid any crying have evidence-based options too, such as bedtime fading and camping out. Choosing not to sleep train is not a failure either; the Canadian cohort shows many babies consolidate sleep on their own.

Baby sleep music, lullabies and white noise: what age to start and what the evidence shows

Search results for baby sleep are dominated by twelve-hour videos of Brahms and rainfall, so it is fair to ask whether sound helps at all.

Lullabies have the oldest evidence of the lot: every studied culture sings to infants, and small studies of live singing in neonatal units show calmer heart rates and better feeding. Live singing is not a playlist, though. Trials of recorded music for settling healthy infants at home are few and small, and they mostly show modest short-term calming rather than longer sleep. There is no age too early to sing to a baby, and a soft, repeated song is a reasonable cue inside a bedtime routine from the first weeks.

White noise, a steady broadband sound that masks sudden changes, has slightly more structured evidence. A frequently cited 1990 study found that 80 percent of newborns fell asleep within five minutes with white noise compared with 25 percent without. The study was tiny and the effect has not been robustly replicated, so grade it as suggestive.

The safety data are the part parents rarely hear. A 2014 study in Pediatrics tested 14 infant sound machines at maximum volume. All exceeded 50 decibels, the level recommended for hospital nurseries, when placed close to a crib, and three exceeded 85 decibels, the threshold at which workplace regulators require hearing protection for adults over an eight-hour shift. Prolonged exposure at those levels could plausibly affect developing hearing, and animal studies suggest constant noise may disrupt auditory development.

The practical reading: if you use sound, place the device across the room rather than in or beside the crib, keep the volume at about the level of a quiet conversation or a running shower heard from the next room, and consider a timer so the noise does not run all night. Babies also need to hear ordinary speech during waking hours, which is where language is learned.

None of this is a method on its own. Sound can be one gentle cue inside a routine. It cannot replace a safe surface or a consistent bedtime.

Toddler sleep: what is normal from 1 to 3 years

Toddler sleep problems look different from infant ones. The biology is mostly settled; the psychology has arrived.

Total need falls to 11 to 14 hours, and the nap schedule reorganizes twice. Most children drop from two naps to one between 12 and 18 months, and the single nap usually disappears between ages 3 and 5. The transitions are bumpy: a child who is not quite ready for one nap may be overtired by late afternoon, and a child who still naps at 4 may resist bedtime until 9 p.m. Capping or moving the nap earlier is a reasonable first adjustment.

Bedtime resistance peaks around age 2. Curtain calls for water, one more story and a sudden interest in the meaning of life are normal expressions of autonomy and separation anxiety, not defiance. The approaches with evidence are the same behavioral ones used in infancy, adapted: a predictable routine with a clear end point, limited choices offered before bed rather than after, and calm, brief, boring returns to the room. A toddler who climbs out of the crib should move to a low bed or a mattress on the floor for safety.

Night terrors appear in this age range and alarm parents more than children. A night terror is a partial arousal from deep sleep, usually in the first third of the night, during which the child may scream, thrash and look terrified while not actually awake and not recognizing you. It ends on its own, the child does not remember it, and the right response is to keep them safe without trying to wake them. Nightmares are different: they happen later in the night, the child wakes fully, remembers the dream and wants comfort.

Screens deserve one evidence-based sentence. Observational studies consistently link evening screen use with later bedtimes and shorter sleep in toddlers, so keeping the last hour before bed screen-free is a low-cost change with a plausible mechanism in light exposure and arousal.

Common myths about baby sleep that the evidence does not support

Viral advice is often the exact opposite of what studies show. These are the claims clinicians hear most.

Myth: keeping a baby awake longer during the day makes them sleep better at night. The reverse is more often true. Overtired infants produce more stress hormones and settle worse. Trials of behavioral methods work by changing how a baby falls asleep, not by restricting daytime sleep.

Myth: adding cereal to a bottle helps a baby sleep through. A randomized study in the 1980s and later work found no difference in night waking. Cereal in a bottle also raises choking risk and is not recommended before solids are introduced around 6 months.

Myth: weighted sleep sacks are a proven sleep aid. No controlled trial supports this for infants, and safe-sleep authorities advise against any weighted product for babies because of breathing and rolling concerns.

Myth: sleep training causes attachment damage. The only randomized trial measuring attachment found none, and the five-year follow-up found no differences in the parent-child relationship.

Myth: never wake a sleeping baby. Newborns who have not regained their birth weight, usually by about 2 weeks, generally need to be woken to feed at least every 3 to 4 hours. After that point, with good weight gain, a longer stretch is fine; check with your pediatrician.

Myth: a baby who sleeps on the stomach sleeps better, so it is worth the trade. Stomach sleeping does produce deeper sleep, and that is precisely the problem. Reduced arousability is one proposed mechanism for the two- to threefold increase in SIDS risk.

Myth: melatonin gummies are a harmless fix for toddler sleep. Melatonin is a hormone, not a vitamin. It has not been tested for safety in infants, pediatric poison center calls involving melatonin rose more than 500 percent in the decade to 2021, and any use in young children is a decision for the treating clinician, never a self-treatment.

Bedtime routines that work: the simplest evidence-backed habit

If one change deserves to come before every other, it is the routine. The evidence is observational but unusually large and consistent.

A 2015 analysis of more than 10,000 mothers in 14 countries found that children with a nightly bedtime routine fell asleep faster, woke less often and slept longer, and the relationship was dose-dependent: the more nights a week the routine happened, the better the sleep. Children whose routines began in infancy had the best outcomes. Because the data are observational, calmer families may simply be more likely to keep routines, but the biology makes the link plausible. Repeated cues in dim light help entrain the circadian rhythm, and predictability lowers arousal at the moment a child needs it lowest.

A routine does not need to be elaborate. Mayo Clinic and MedlinePlus describe the same core: a consistent sequence of 3 to 4 calm activities over 20 to 30 minutes, in the same order, ending in the sleep space with the baby drowsy but awake. A bath, a feed, a book or song, a few quiet words and lights out covers it. The feed belongs earlier in the sequence once a baby is a few months old so that feeding does not become the only route into sleep.

Light is the lever most families underuse. Bright daylight in the morning and through the day, then dim, warm light in the hour before bed, is the pattern that strengthens the maturing clock from about 2 to 3 months onward. Overnight feeds and diaper changes go better with the dimmest light that lets you see safely and minimal talking.

Consistency between caregivers matters more than any single detail. A routine that happens one way with one parent and another way at grandparents’ house is two routines, and toddlers notice.

Finally, protect your own sleep where you can. The advice to sleep when the baby sleeps is sound in the newborn weeks, when night sleep is fragmented for everyone. Splitting the night into shifts, accepting help with daytime tasks and keeping your own caffeine and light exposure sensible are not luxuries; parental exhaustion is linked to lower mood and less consistent routines, so it feeds back into the baby’s sleep.

When to see a doctor about your baby's or toddler's sleep

Most sleep struggles are developmental and resolve with time and routine. A smaller set of patterns point to something medical, and these should prompt a call to your pediatrician rather than another week of trying.

Seek urgent medical attention, including emergency services where appropriate, if a baby:

• Has pauses in breathing, turns blue or gray around the lips, or goes limp and is hard to rouse.
• Is under 3 months old and has a rectal temperature of 100.4 F (38 C) or higher.
• Is unusually drowsy, will not wake to feed, or has a high-pitched or weak cry unlike their usual one.
• Has a bulging soft spot, a stiff neck, a rash that does not fade under pressure, or repeated vomiting.

Book a non-urgent appointment if you notice:

• Loud snoring most nights, gasping or mouth breathing during sleep, or restless sleep with unusual positions. These can signal enlarged tonsils or adenoids and sleep-disordered breathing, which is treatable.
• Poor weight gain, fewer wet diapers than expected, or a baby who seems too sleepy to feed effectively.
• Sleep that has clearly worsened alongside a change in development, such as loss of skills already gained.
• Excessive daytime sleepiness in a toddler despite adequate night sleep, or sleep problems alongside frequent ear infections, eczema that itches at night, or symptoms of reflux such as arching and pain with feeds.
• Night terrors or sleepwalking that are frequent, violent or begin in a child who was previously settled.
• Any situation where you are considering a medicine, supplement or sleep product for your child. Decisions about melatonin, antihistamines, reflux medicines or anything else belong with the prescribing clinician, who will weigh your child’s age, weight and health history. Do not start, stop or adjust any medicine on the basis of an article or a video.

One more reason to call is about you. Persistent low mood, anxiety, intrusive thoughts or an inability to sleep even when the baby does are features of postpartum depression and anxiety, which affect roughly 1 in 7 mothers and a meaningful number of fathers. They are common, treatable and worth raising at any visit. A rested, supported parent is part of the baby’s sleep plan, not separate from it.

Frequently asked questions

How do babies typically sleep in the first weeks?

Newborns sleep around 14 to 17 hours a day in stretches of 2 to 4 hours, waking to feed every 2 to 3 hours around the clock. About half of their sleep is active sleep, with twitching, grunting and fluttering eyelids. They have no day-night rhythm yet; that emerges at roughly 2 to 3 months as their own melatonin production becomes rhythmic.

What are the signs that my baby is fussy or overtired?

Early tired cues include staring away, drooping eyelids, yawning and slowing movements. Later cues are eye rubbing, ear pulling, back arching and a whinier cry. An overtired baby may paradoxically seem wired and giggly before melting down. Fussiness also comes from hunger, wind, temperature or overstimulation, and normal crying peaks at 6 to 8 weeks of age.

What is a good age to start sleep music or white noise?

Singing or a soft repeated song can be part of a routine from birth. Recorded music and white noise have only small, short-term evidence for settling. If you use a sound machine, place it across the room, keep it at quiet-conversation volume and use a timer, because tested machines at full volume exceeded safe levels near the crib.

Should I sleep when my baby sleeps?

Yes, in the newborn weeks it is sensible advice, because night sleep is fragmented for everyone and parental exhaustion worsens mood and consistency. Many families manage better by splitting the night into shifts and accepting daytime help. If you cannot sleep even when the baby does, or feel persistently low or anxious, mention it to your clinician.

When do babies sleep through the night?

Researchers count 6 to 8 hours as sleeping through, and many babies reach that between 4 and 6 months, but a 2018 cohort found 38 percent of 6-month-olds and 28 percent of 12-month-olds had not yet done so. Most babies can physically go longer without feeding by about 6 months; whether they do depends on sleep associations and temperament.

Is sleep training safe for babies?

The best available evidence says yes for infants over about 6 months. A randomized trial found lower infant cortisol and no difference in attachment at 12 months, and a five-year follow-up of 326 children found no differences in behavior, emotional health or the parent-child relationship. Trials have not studied training before 6 months, so it is not recommended earlier.

What is normal toddler sleep at 1 to 3 years?

Toddlers need about 11 to 14 hours a day including naps, drop to one nap between 12 and 18 months, and usually give up napping between ages 3 and 5. Bedtime resistance peaks around age 2, and night terrors, which are partial arousals the child does not remember, are common and harmless. A consistent routine is the most evidence-backed tool.

What is the 4-month sleep regression?

It is a maturation point, not a setback. Around 3 to 4 months a baby’s sleep develops adult-like stages with more light sleep, so they surface more often between cycles and notice when conditions differ from bedtime. Babies who fall asleep being fed or rocked tend to call for the same help at night. It passes, and routines help.

Can my baby sleep on their stomach once they roll over?

Always place a baby on their back to sleep until age 1. Once a baby can roll confidently both ways, usually around 4 to 6 months, you do not need to turn them back if they roll onto their stomach, provided the surface is firm, flat and free of soft bedding. Stop swaddling as soon as rolling begins.

Can I give my toddler melatonin to help with sleep?

Not without talking to your child’s clinician. Melatonin is a hormone, it has not been tested for safety in infants, and poison center calls involving children and melatonin rose sharply over the past decade. For most toddlers, behavioral approaches and a consistent routine are the first-line options. Any decision about a supplement or medicine belongs with the treating clinician.

References

• MedlinePlus: Bedtime habits for infants and children
• NHS: Helping your baby to sleep
• NHS: Reduce the risk of sudden infant death syndrome (SIDS)

This article is for general information only and is not a substitute for professional medical advice. Please consult a qualified doctor about your individual situation.