Ferber Method: What Richard Ferber's Chart Actually Says

Who Richard Ferber is

Richard Ferber is a pediatrician and sleep physician who spent his career at Boston Children's Hospital. Harvard Medical School's magazine describes him as the co-founder and former director of the hospital's Center for Pediatric Sleep Disorders, which he dates to 1978, and notes that he is now retired. You can read that profile in Harvard Medicine magazine.

His book, Solve Your Child's Sleep Problems, was first published in 1985 and revised and expanded in 2006. It is a book about children's sleep problems in general, and the progressive-waiting approach is only one part of it. The problem it targets is described in the same Harvard profile: when parents step in at every waking, a child can come to depend on that intervention rather than settling back to sleep alone after the ordinary arousals everyone has during the night. The method is aimed at that dependence, not at crying as such.

That framing is worth holding on to, because it is where most of the online argument goes wrong. Read closely, the approach is a response to a specific pattern a family has already identified as a problem: a child who needs to be rocked, fed or held back to sleep every time, and parents who can no longer keep that up. It was never a rule that all children should be put to bed this way, and a family whose nights are working has no reason to reach for it.

One more piece of context. Ferber is a physician writing for parents, not a commercial sleep consultant, and the book predates the paid courses and apps that now sell versions of the same idea. If you are weighing one of those, our page on choosing a sleep training program explains what you are and are not paying for.

The progressive-waiting chart, night by night

Here is the chart. It is reproduced below from the Sleep Foundation's page on the method, which sets out the waiting times from Ferber's book and is medically reviewed by a board-certified sleep medicine physician. Each figure is the number of minutes a parent waits before going back in, if the child is still crying. The first column is the wait after leaving at bedtime; the second is the wait after the first check-in; the last column applies to the third check-in and every one after it that night.

Two things about the shape matter more than any single number. The waits get longer within a night, and each new night starts higher than the one before. By night seven the first wait is twenty minutes. The chart stops there, and so does the reproduction.

Now the disagreement, because there is one. Healthline's summary, which is also medically reviewed, agrees on nights one and two but gives night three as 12, 15 and 17 minutes, which is what the Sleep Foundation's reproduction lists for night four. Commercial sleep sites publish further variants. We have used the Sleep Foundation figures because they present a full seven nights and match the night-one and night-two numbers that every reproduction shares, but the book itself is the authority. If an exact minute matters to you, read Ferber's chart in his own words rather than anyone's copy of it, including this one.

And keep the numbers in proportion. Nobody has shown that waiting twelve minutes rather than fifteen on night three changes the outcome. The chart is a scaffold for consistency, which is the part that does the work.

What a check-in is for

The check-in is the part parents most often get wrong, and it is wrong in an understandable direction. When you walk back in, every instinct says the job is to stop the crying. It is not. In the reproductions of the method the visit is brief, the child is not picked up or fed back to sleep, and the parent leaves again while the child is still awake, whether or not the crying has stopped. Healthline's summary puts the purpose plainly: the check-in is reassurance, for the child and for you, not a cure.

Seen that way, the rules make sense. A check-in that ends in rocking to sleep has quietly reinstated the very condition the method exists to change, and the next waking will ask for it again. A check-in that runs long becomes a new kind of company to cry for. A short, dull, predictable visit says the same thing every time: I am here, you are safe, it is still bedtime.

Things that are commonly described as fine during a check-in:

• A few calm words, the same few each time
• A brief pat or hand on the back or chest
• Straightening a twisted sleep sack or a leg caught against the cot rail
• Checking for anything genuinely wrong: a dirty nappy, a fever, vomit, a trapped limb
• Leaving while the child is still awake and still upset, which is the hardest part

Graduated extinction, and what that label rules out

In the research literature the Ferber method sits under the heading graduated extinction. The "extinction" part comes from behavioural psychology: a response that is no longer reinforced, in this case crying that no longer produces rocking or feeding to sleep, fades over time. The "graduated" part is the check-ins. The parent still responds, on a predictable timetable, rather than not at all.

That makes it different from unmodified extinction, the approach most people mean when they say "cry it out", in which the parent puts the child down awake and does not return until a set morning time except for genuine need. The American Academy of Sleep Medicine's 2006 review of behavioural treatments by Mindell and colleagues treats these as separate interventions and gives both supportive ratings, alongside bedtime fading, positive routines and parent education.

So it is fair to say Ferber's approach involves crying, and unfair to say it means leaving a baby to cry all night. Both halves of that sentence matter. If the idea of any crying you are choosing not to end immediately is unacceptable in your family, then this method is unacceptable too, and that is a legitimate place to land. The gentler alternatives, and the case for not sleep training at all, are on the main sleep training page.

What graduated extinction is not, either, is a feeding plan. Nothing in the label says night feeds must stop. Plenty of families run check-ins at bedtime while keeping a feed overnight that their pediatrician has agreed is still needed.

Age: what the AAP says, and what it does not

Here the honest answer is narrower than most pages admit. The American Academy of Pediatrics does not publish a Ferber-specific start age on its parent site. What its HealthyChildren guidance on getting your baby to sleep does say, under a heading for babies four months and older, is to put babies to bed drowsy rather than asleep, not to rush in to soothe a crying baby, and that babies need time to learn to fall back asleep on their own. The same page notes that babies do not have regular sleep cycles until about four months of age, and that it is normal for a six-month-old to wake in the night and go back to sleep after a few minutes.

Other sources go further. The Sleep Foundation's Ferber page says sleep training can start once a child is six months old. The trials below enrolled babies aged six to sixteen months, and delivered their intervention at eight to ten months. Nobody has studied the method in young infants, and the existing guidance gives no basis for using it in the newborn months, when frequent night feeding is expected.

Put together, the published floor is not one number. Four months is when the AAP starts framing advice around self-settling; around six months is where the method-specific advice and the research begin. Your pediatrician, who knows your baby's weight, feeding and health, is the right person to say whether your child is there yet.

Three studies people cite, and what each actually measured

Three papers come up in almost every argument about Ferber's approach. Each is real, each has been verified against PubMed for this page, and each answers a narrower question than the way it is usually quoted.

Mindell and colleagues, 2006, in the journal Sleep. A task force appointed by the American Academy of Sleep Medicine reviewed 52 treatment studies of bedtime problems and night waking in young children. It concluded that behavioural therapies produce reliable and durable changes, with the strongest support for unmodified extinction and preventive parent education, and support also for graduated extinction. It is a review of many small studies, not a test of one family's experience. (PubMed 17068979)

Gradisar and colleagues, 2016, in Pediatrics. Forty-three infants aged six to sixteen months were randomised to graduated extinction, bedtime fading, or sleep education as a control. Graduated extinction produced large decreases in time to fall asleep, number of awakenings and time awake after sleep onset. Infant stress was measured through saliva cortisol, which fell slightly in both active groups rather than rising. Twelve months later the researchers found no differences in emotional or behavioural problems or in attachment between the groups. (PubMed 27221288)

Price and colleagues, 2012, also in Pediatrics. This was a five-year follow-up of an Australian trial in which families of infants with parent-reported sleep problems at seven months received behavioural sleep techniques through one to three nurse consultations at eight to ten months, or usual care. At age six, the researchers found no evidence of differences on any outcome they measured, including emotional and conduct behaviour, stress hormones and the child–parent relationship. Their summary: no marked long-lasting effects, positive or negative. (PubMed 22966034)

Read together, they say something modest and useful: in the families studied, graduated extinction improved sleep in the short term, and the researchers did not find harm at one or five years. They do not say it is the only way, or that every child will respond the same.

The limits of that evidence

It would be easy to stop at "no harm found". It would also be overstating it, so here are the limits, stated plainly.

The samples are small. Gradisar's trial split forty-three babies three ways, so each arm had about fourteen. Small trials can miss modest effects in either direction.

The families chose to take part. People who volunteer for a sleep study are motivated, often already leaning towards trying a method, and not a picture of every household.

The Price follow-up did not test the Ferber chart as such. Its nurses taught behavioural techniques in general, so it supports the family of approaches more than this particular timetable.

And much of the outcome data, especially on sleep and behaviour, comes from parent report, which is the kind of measure that is vulnerable to what parents hope they are seeing.

None of that makes the research worthless. It makes it research: the best available answer to a hard question, with real gaps. If you have read a confident claim in either direction, that the method is proven harmless for every child or that it damages attachment, it is going further than the studies do.

When the chart stays in the drawer

Some weeks are the wrong weeks, and some children are the wrong age, and none of that is a failure. The commonly cited reasons to hold off are practical rather than philosophical.

Illness is the obvious one. A child with a fever, an ear infection or a cough is crying for a reason that a timetable cannot fix, and the AAP's symptom guide for a crying child lists crying that gets worse during an illness as a reason to call your doctor. Teething discomfort, a recent vaccination, travel, a house move, a new childcare arrangement, or a nap transition already underway all make a result impossible to read, and the site's nap transitions hub is worth a look if one of those is in play.

Age is the second. Below the guidance described above, the method has no evidence base and night feeding is still expected.

Health signals are the third, and they override everything else on this page. Snoring, laboured or noisy breathing, pauses in breathing, poor weight gain, reflux symptoms, or a child who seems unwell in daylight as well as at night are reasons to speak to your pediatrician before any behavioural approach. A sleep problem with a medical cause will not answer to check-ins, and pushing on would be unfair to the child.

And safe sleep does not change because a method has started. Back to sleep, on a firm flat surface, with nothing loose in the cot. The AAP's safe sleep guidance and NICHD Safe to Sleep cover it properly; we link rather than paraphrase.

Where this sits among the other sleep-training pages

If you have read this far and decided to try it, the method is the smaller half of the work. The other half is the day around it, because check-ins cannot rescue a bedtime that lands when the child is not tired, or one that lands an hour after they were. The baby sleep training schedule page sets out how to fix the wake time, derive bedtime from it, and hold the day still for a week so that the nights can be compared with each other.

If you are still deciding between approaches, how to sleep train a baby compares graduated extinction with full extinction, the chair method, pick-up-put-down and bedtime fading, and treats not sleep training as a complete answer rather than a footnote. If you are wondering whether to pay someone for a version of this, the sleep training program page is the one to read first.

And if you would rather start somewhere calmer, the homepage lays out the whole map: schedules by month, wake windows, nap transitions and regressions, so you can check whether the problem in front of you is a training question at all. Quite often it turns out to be a timing question wearing a training costume, and a fifteen-minute change to bedtime does more than any chart.