Every comparison of sleep training methods you will read is really an argument about tone — which one is kind, which one is harsh, which camp the writer belongs to. This one is about evidence, and the first thing the evidence says is that the question is usually asked wrong.
The naming problem
Parents choose between branded methods: Ferber, the chair method, pick up put down, the sleep lady's shuffle. Researchers test behavioural categories: unmodified extinction, graduated extinction, bedtime fading and positive routines, scheduled awakenings, preventive parent education.[1] The two lists do not line up, and almost every unhelpful argument about sleep training comes from treating them as if they did.
| Method | Research name | Evidence behind it |
|---|---|---|
| Ferber / controlled crying | Graduated extinction | Multiple RCTs; a supported category |
| Chair method / gradual retreat | Camping out | One head-to-head RCT of 91 infants |
| Bedtime fading | Bedtime fading | RCT arm; a supported category |
| Cry it out | Unmodified extinction | Strongest support in the AASM review |
| Pick up, put down | — no research name — | No trials; absent from the supported list |
| Doing nothing yet | Watchful waiting | Many problems resolve; controls caught up by 4 months in one trial |
That last row is not a joke, and it is the row most comparisons leave out. In a randomised trial of 156 infants with severe parent-reported sleep problems, the intervention group did better at two months — 53 of 76 problems resolved versus 36 of 76 — but by four months the changes in sleep problems and depression scores were similar between groups.[6] The control families got there too, more slowly.
Does any of it work?
Yes, and reliably. The American Academy of Sleep Medicine's task force reviewed 52 treatment studies: 94% reported that behavioural interventions were efficacious, with over 80% of treated children showing clinically significant improvement maintained for three to six months.[1] Those findings became formal practice parameters.[2]
94%
Does any of it harm?
This is the question the whole argument actually turns on, and it has been studied better than the tone of the debate suggests. A randomised trial of 43 infants measured infant stress directly with salivary cortisol rather than by asking parents. Cortisol showed small-to-moderate declines in both the graduated extinction and bedtime fading groups compared with controls. At twelve months there were no differences in emotional or behavioural problems, and no differences in attachment measured with the strange situation.[3]
A separate trial followed 326 children to age six. It found no evidence of differences between intervention and control families on any outcome measured — child mental health, sleep, psychosocial functioning, chronic stress, child-parent closeness and conflict, attachment, or parental depression and anxiety.[5]
So how do you actually choose?
On fit, not on virtue. The most useful trial for this question randomised 91 infants aged 9 to 18 months to checking-in — graduated extinction, with gradual separation — or camping-out, where parental presence is maintained. Both improved sleep and both held at six-month follow-up. Separation anxiety did not change under either. But separation anxiety moderated which one worked better: infants with high separation anxiety benefited more from camping-out.[4]
The authors' recommendation is that clinicians should assess separation anxiety and favour gentler approaches such as camping-out for highly anxious infants.[4] That is a genuinely useful decision rule, and it is about your baby rather than about your parenting philosophy.
Which method solves which problem
| If the problem is… | Start with | Because |
|---|---|---|
| Bedtime takes an hour | Bedtime fading | Large drop in time-to-sleep; almost no crying |
| Frequent night wakings | Graduated extinction | Its clearest advantage in the trial data |
| High separation anxiety | The chair method | Trial evidence favours it for these infants |
| Crying is a dealbreaker | Bedtime fading, then the chair | Both supported; least crying of the options |
| Nothing has been tried | Fix the schedule first | An overtired baby resists every method equally |
| A method was abandoned midway | A slower one you will finish | An abandoned method teaches nothing |
What none of them fix
- A bedtime in the wrong place. Sort the day first — wake window stacking is the usual culprit.
- Hunger, illness, teething, reflux or a house move. None of these are behavioural, and no method will out-argue them.
- Normal night waking. Waking between sleep cycles is not a disorder, and the age-appropriate expectation matters more than the method.
- A disagreement between two adults. Inconsistency is the single largest cause of failure across every method here.
- Your own exhaustion, which is a real clinical consideration and not a side issue — the trials that measured maternal mood found it improved alongside infant sleep.
The four methods, in detail
Each has its own page with the schedule, the failure modes and the evidence specific to it: the Ferber method for graduated extinction, the chair method for camping out, bedtime fading for the low-crying option, and pick up, put down — which is on this site with its evidence gap clearly labelled rather than left off because it lacks trials.
The bottom line
Behavioural sleep interventions work for most families in the short to medium term, do not appear to raise infant stress, and leave no measurable trace at five years in either direction. Several distinct methods are supported, and the trial evidence says the right one depends on your baby — particularly on separation anxiety — rather than on which philosophy you subscribe to.[2]
It is also entirely reasonable not to do any of this. Sleep problems in infancy frequently resolve without intervention, the control groups in these trials got there too, and no method is required for a child to sleep eventually. Judge the whole thing against the 24-hour total your baby is getting rather than against anyone else's schedule.[7]
