- 01No method has strong evidence of long-term harm — but evidence is reassuring, not conclusive
- 02Start no earlier than 4 months corrected age; 6 months is the most widely cited window
- 03Ferber and cry-it-out have the strongest RCT evidence base; gentle methods take longer but cause less parental distress
- 04The Ferber interval table gives the exact night-by-night waiting protocol
- 05Consistency between caregivers matters more than which method you pick
- 06Illness, teething, and active regressions always override sleep training — pause and address first
Which method fits your baby? Ask Lunara's AI.
Lunara's AI can help you decide which sleep training method fits your baby's age and temperament in minutes.
Ask Lunara →What Is Sleep Training?
Sleep training is a behavioural approach to helping a baby learn to fall asleep without external settling — feeding, rocking, holding, or bouncing — and to resettle between sleep cycles independently.
What sleep training is NOT:
- Ignoring a baby's needs
- Forcing a baby to stop feeding at night
- A universal standard every family must follow
- The same as leaving a baby to cry indefinitely without any framework
The term covers a wide spectrum — from full extinction (the least interventionist approach during the night) to gentle fading (the most gradual). All methods share the same underlying goal: giving the baby the opportunity to practise falling asleep without a specific external condition. The methods differ only in how much parental intervention is offered during the learning process.
Why sleep associations matter: a sleep association is any condition present at sleep onset. Because babies wake between every 45-minute sleep cycle, they instinctively look for the same conditions when they rouse. A baby who falls asleep on the breast will wake at the end of each cycle expecting the breast. A baby who falls asleep in their cot will wake at the end of each cycle and can return to sleep without external help. Sleep training replaces the first pattern with the second.
Is Sleep Training Safe? What the Research Actually Says
The safety of sleep training is the most-searched question in this topic cluster and the least clearly answered by existing content. Here is what the current evidence actually says — including the limitations.
Price et al. (2012) — INFANT Trial, Pediatrics [1]
Finding: no significant differences in emotional behaviour, mental health, stress, or parent-child attachment between sleep-trained and control groups at 5-year follow-up.
What it doesn't tell us: the study did not extend beyond age 5; adult outcomes are unknown.
Gradisar et al. (2016) — RCT, Pediatrics [2]
Finding: both sleep training groups showed improved sleep. Cortisol levels were not elevated compared to controls. Parent wellbeing improved significantly.
What it doesn't tell us: sample size was small; the study lasted 12 months, not longer.
Mindell et al. (2006) — Meta-analysis, Sleep [3]
Finding: behavioural interventions are effective and well-tolerated, with no reported adverse events across studies.
What it doesn't tell us: studies varied widely in methodology and measurement; publication bias may skew results toward positive findings.
The weight of current evidence does not support long-term harm from sleep training methods used after 4-6 months of age. The claim that sleep training causes attachment damage or psychological harm is not supported by the available data. However, the evidence base is not large — most RCTs have small samples and follow-up periods of 1-5 years. The honest position is: current evidence is reassuring, but not conclusive in every direction. If a family chooses not to sleep train, that is a valid choice with its own evidence base around parental-presence settling. If a family chooses to sleep train, the current evidence supports doing so without guilt.
When Is Your Baby Ready for Sleep Training?
| Method | Minimum Age | Recommended Age |
|---|---|---|
| Extinction (CIO / Weissbluth) | 4 months corrected | 6 months |
| Ferber / Graduated Extinction | 4 months corrected | 5-6 months |
| Chair / Sleep Lady Shuffle | 4 months corrected | 6 months |
| Pick Up, Put Down | 4 months | 4-8 months (less effective after 8 months) |
| Fading / Gentle / No Tears | 3-4 months | Any age from 4 months |
Medical readiness:
- Gaining weight appropriately and feeding well in daytime
- No active illness, ear infection, or viral illness in the past 2 weeks
- If premature: 4 months corrected age (not chronological)
- Paediatrician has no specific concerns about night feeds at this age
Developmental readiness:
- Shows some ability to be briefly calm independently (bouncer, play mat)
- Not in the peak of a major developmental leap (4-month architecture shift, 8-10 month separation anxiety peak)
- If rolling: free from swaddle — free arm movement required for safe arousal
Family readiness:
- All primary caregivers have agreed on the method and can apply it consistently
- No major household disruptions in the next 2 weeks (travel, house move, visitor)
- At least one caregiver has 1-2 hours of protected time in the evening
- Active illness (teething flare, cold, ear infection, fever)
- Medical concerns about growth or feeding
- Under 4 months corrected age
- Family crisis (bereavement, major life stress) — parental mental bandwidth matters
All Sleep Training Methods Compared
This is the biggest content gap across competitors — the only side-by-side comparison table for these methods.
| Method | Crying Involved | Age Range | Duration | Best For |
|---|---|---|---|---|
| Ferber (Graduated Extinction) | Moderate — between timed check-ins | 4-18 months | 5-7 nights | Parents wanting results within a week but who want check-ins |
| Cry-It-Out (Extinction) | High initially — no check-ins | 6+ months | 3-5 nights | Families wanting the fastest results; consistent caregivers |
| Chair Method (Sleep Lady Shuffle) | Moderate — parent visible | 6 months+ | 2-3 weeks | Parents who want to stay present; high separation anxiety |
| Pick Up, Put Down (PUPD) | Moderate — pick-up at cry | 4-8 months | 5-14 nights | Responsive parents; younger babies |
| Fading | Low to moderate — gradual withdrawal | 4 months+ | 2-4 weeks | Gentle transition; families with time and flexibility |
| No Tears / Attachment-Based | Minimal — no targeted crying | Any age | Months, not nights | Attachment-parenting philosophy families |
Speed vs. distress tradeoff:
- No Tears / Attachment-based: Slow, low distress
- Fading: Slow, low distress
- Chair / PUPD: Moderate speed, moderate distress
- Ferber: Fast-moderate, moderate distress
- Cry-It-Out: Fastest, highest initial distress
Method Deep-Dives: Step-by-Step Protocols
The Ferber Method (Graduated Extinction)
Named for Dr. Richard Ferber, Director of the Center for Pediatric Sleep Disorders at Boston Children's Hospital [5]. The method is more accurately called "graduated extinction" — progressive waiting with brief, non-stimulating check-ins. A brief parental check-in every few minutes tells the baby the parent exists and the environment is safe, while still giving the baby the opportunity to fall asleep independently.
Who it works for: parents who want meaningful results within a week but cannot tolerate the absence of any check-ins. Works best for babies 5 months and older.
The Ferber Interval Table
| Night | First Wait | Second Wait | Third Wait | Subsequent Waits |
|---|---|---|---|---|
| 1 | 3 min | 5 min | 10 min | 10 min |
| 2 | 5 min | 10 min | 12 min | 12 min |
| 3 | 10 min | 12 min | 15 min | 15 min |
| 4 | 12 min | 15 min | 17 min | 17 min |
| 5 | 15 min | 17 min | 20 min | 20 min |
| 6 | 17 min | 20 min | 25 min | 25 min |
| 7 | 20 min | 25 min | 30 min | 30 min |
During check-ins: enter briefly (1-2 minutes), speak calmly and reassuringly, do not pick up, do not feed, then leave. Check-ins are optional — some babies settle faster without them. Apply the same schedule to night wakings as to initial bedtime settling.
Step-by-Step
- Complete the bedtime routine; place baby in cot drowsy-but-awake
- Say goodnight phrase; leave the room
- If crying: start your timer per the Night 1 column
- At each interval: brief check-in (1-2 min, no pick-up), leave
- Repeat until asleep
- For each subsequent waking: repeat from step 3, using Night 1 intervals until back on the same-night row
- Advance to the next night's column the following evening
What to Expect
- Night 1: often the hardest — 45-90 min of crying possible
- Nights 2-3: typically shorter periods
- Nights 5-7: most babies settling in under 10 minutes
Cry-It-Out (Full Extinction / Weissbluth)
Full extinction involves placing the baby in the cot after the bedtime routine and not returning until morning (or a designated feed time), regardless of crying. Named for Dr. Marc Weissbluth [6]. The baby learns that crying will not result in parental attendance, which removes the reinforcement cycle that maintains night waking. Because there are no check-ins, there is also no partial reinforcement — the signal is clear and consistent.
Who it works for: families who want the fastest results and have consistent caregivers who can hold the boundary through intense early nights. Not suited to parents for whom the crying causes extreme distress.
Step-by-Step
- Ensure the baby meets all readiness criteria above (6 months minimum recommended for extinction)
- Complete the bedtime routine; place in cot awake; say goodnight
- Do not return until a pre-agreed time (e.g., 11pm feed if still nutritionally needed, then again at 7am)
- Apply consistently for a minimum of 3 nights before assessing
What to Expect
- Night 1: may involve 1-2 hours of crying
- Night 2: typically shorter
- Nights 3-5: most babies substantially improved or fully settled
The Chair Method (Sleep Lady Shuffle)
Developed by Kim West [7]. The parent remains visibly present throughout the settling process but gradually withdraws over 2-3 weeks. Also known as the camping-out method.
Who it works for: babies with high separation anxiety; parents who need to be present to manage their own anxiety; families who have more time than they have stamina for cry-based methods.
Step-by-Step
- Nights 1-3: chair placed next to the cot. Parent sits as baby falls asleep — no picking up, feeding, or rocking.
- Nights 4-6: chair moved to the middle of the room.
- Nights 7-9: chair moved to the doorway.
- Nights 10-12: parent briefly visible from outside the doorway.
- Nights 13+: parent no longer in the room; standard check-in protocol if needed.
Common Challenge
Some babies are more stimulated by the parent's visible presence and take longer to settle. If settling takes longer than 45-60 minutes per session after night 5, consider switching to Ferber.
Pick Up, Put Down (PUPD)
Developed by Tracy Hogg [7]. The parent picks the baby up at each sustained cry until crying stops, then immediately puts them back down. Repeat until baby falls asleep in the cot.
Age note: works best at 4-8 months. After 8 months, picking up can become stimulating rather than comforting — switching to chair or Ferber is typically more effective.
Step-by-Step
- Place baby in cot drowsy-but-awake after bedtime routine
- Wait — at sustained cry (not a grizzle), pick up
- Hold until crying stops (not until asleep — this is key)
- Immediately put back down
- Repeat until asleep
Limitation: sessions can be physically demanding — 20-50 pick-ups in the first night is not unusual for younger babies.
Gentle Methods: Fading and No Tears
Fading (also called "fading out" or "gradual retreat") involves identifying the current settling method and incrementally withdrawing it. No Tears / attachment-based methods prioritise avoiding crying through parental presence, feeding, and co-sleeping.
Fading Protocol — Rocking Association
- Rock the baby until 80% asleep, place in cot
- Once successful: rock to 60% asleep, place
- Once successful: rock to 40% asleep, place
- Continue reducing until the baby is placed fully awake
Fading Protocol — Feeding Association
- Feed, then burp, then hold for 5 minutes before placing in cot
- Progress to: feed, burp, 3-minute hold
- Progress to: feed, burp, immediate placement
- Progress to: feed earlier in routine (before bath), then room-dark placement
Timeline: typically 2-4 weeks. Each step takes 3-5 days to consolidate before moving to the next.
No Tears approach: these methods do not have a fixed protocol — they are a philosophy. The core principle is that sleep develops naturally when the baby's needs for closeness, feeding, and comfort are fully met, and that encouraging independence before a baby is developmentally ready is unnecessary. Both fading and no-tears approaches are valid. Families with the flexibility and temperament for a 2-4 week timeline and no tolerance for crying report high satisfaction with fading. Families who need faster results are better suited to Ferber or extinction.
Track your sleep training progress night by night.
Lunara shows settling time trends so the improvement you can't feel is visible in the data.
Start Free →How to Choose the Right Method for Your Family
The most common mistake in method selection is choosing a method the family finds emotionally unmanageable and abandoning it on night 2. An incomplete attempt — starting CIO, breaking at 3am, feeding to sleep — teaches the baby that sustained crying eventually brings the parent. This partial reinforcement makes the next attempt harder. It is better to choose a method that is sustainable for at least 7 nights than to choose the "fastest" method and stop early.
Starting Sleep Training: The First Week
- The hardest night. Expect initial settling to take 45-90 minutes. Night wakings will likely be frequent. This is the peak of protest — it does not predict the final outcome.
- Usually similar to or slightly shorter than Day 1. Some babies show improvement; some have a "second night surge" — a brief intensification before improvement. Hold the method.
- Most families using Ferber or CIO report noticeably shorter settling times. 5-20 minutes is typical by this night for babies over 6 months.
- A step-change improvement for most families. Night wakings reducing significantly; settling time often under 10 minutes.
- Continued improvement. Most Ferber/CIO families report 5-10 minute settling with minimal protest by now.
- Some babies no longer protest at all. Chair and PUPD families will still be working through the method — progress is more gradual.
- Assessment point. Most families reach their new baseline by this night. Chair/PUPD families continue the gradual process beyond this week.
Pre-training checklist:
- All caregivers agreed on the method and the response to crying
- Bedtime routine in place for 5+ nights
- Feeding complete before the final settling step
- Safe sleep conditions confirmed (firm flat surface, no soft objects, correct room temp)
- White noise on — helps the baby settle and reduces awareness of parental sounds
- Blackout blind in place — light prolongs waking
- Phone timer app ready for Ferber intervals
For schedule and timing guidance, see the baby sleep schedule guide. For the 4-month regression that often triggers sleep training, see the sleep regression guide.
Sleep Training Troubleshooting
No improvement after 1 week
Rule out in order: illness (ear infection is a common hidden cause — settles fine then screams 45 min into sleep); inconsistency between caregivers; overtiredness (wrong bedtime or wake window causing a cortisol spike); or an ongoing developmental leap/regression. If all four are clear: extend the method by 1 more week with fully consistent application before switching.
Baby vomits from crying
This is more common than most guides acknowledge, particularly in the first 2 nights of extinction-based methods. Vomiting does not indicate distress beyond normal crying — it is a physiological response in some babies. Enter calmly, clean and change with minimal interaction, replace in cot, and continue the method. Entering with high distress or extra comfort will reinforce the vomiting behaviour.
Sleep training worked but stopped working 2 months later
A solved sleep issue that recurs almost always has a cause: a sleep regression, teething, illness, or a schedule drift. Rule out each in order. Resume the settled method for 3-5 nights; most families report a faster response the second time.
One parent wants to sleep train, the other doesn't
This is one of the most common reasons sleep training fails. Inconsistency between caregivers — one holding the method, the other rescuing — teaches the baby that sustained crying achieves rescue. A shared decision is essential. If agreement is impossible, fading (the least confrontational method) is more compatible with mixed caregiver responses.
Naps not improving even though night sleep is
Night sleep typically improves before nap sleep during sleep training. This is normal — the homeostatic sleep drive is stronger at night, making it easier to learn. Apply the same method to naps from the second week. Nap settling often takes 2-4 weeks longer than night settling to consolidate.
Safe Sleep During Sleep Training
- Back to sleep — every sleep, every time, until 12 months
- Firm, flat sleep surface — cot, Moses basket, or bassinet; no sofas, armchairs, or adult beds
- No soft objects in the sleep space until 12 months (bumpers, blankets, toys, pillows)
- Room temperature 16-20°C / 60-68°F
- Room share (not bed share) recommended for at least the first 6 months
| Situation | Guidance |
|---|---|
| Baby rolling during sleep training | Stop swaddling at the first roll attempt — arms must be free for safe arousal. Use a sleep bag/sack. |
| Baby standing in cot | Safe if cot side is still raised to correct height. Do not lower the mattress until the baby can pull to stand. |
| Lowering cot mattress | Lower to the lowest position once the baby can sit independently (typically 6-8 months). |
| White noise | Safe at ≤50 dB and ≥200 cm from the baby's head — equivalent to rainfall, not loud music or machinery. |
| Dummy/pacifier | Offer at sleep onset during sleep training; do not replace if it falls out during the night. |
How Lunara Supports Sleep Training
Sleep training works best with data. Subjectively, a baby who settled in 45 minutes on night 1 and 30 minutes on night 3 feels like a failure — the data shows progress. Lunara captures what exhausted parents misremember.
- Settling time per night — log when you placed baby down and when they were asleep; the trend chart shows the improvement that feels invisible in the moment
- Wake window compliance — correct timing is the most overlooked variable in sleep training failure
- Night waking frequency — weekly average, so temporary increases (regression, illness) don't obscure the underlying trend
- Nap durations — cumulative daytime sleep against the age target, shown live
- AI question answering — ask "Is it normal that my baby vomited on night 2 of Ferber?" and get guidance calibrated to your specific situation
Partner co-tracking: with Lunara's shared baby tracker, the partner taking the night shift logs wakings and settling times without waking the sleeping partner. Both see the same weekly progress view.
Frequently Asked Questions
There is no single "best" method - effectiveness depends on the baby's age and temperament, the family's parenting philosophy, and how consistently the method is applied. The Ferber method and graduated extinction show the strongest evidence base in RCTs. Gentle methods (fading, chair) have strong parental adherence rates because they cause less parental distress, though they typically take longer. The best method is the one the whole family can apply consistently for at least 1-2 weeks without abandoning it.
Most paediatric sleep experts recommend no earlier than 4 months corrected age. Before 4 months, the circadian rhythm is not fully developed and the baby cannot biologically regulate sleep the way sleep training requires. The AAP and most clinical guidelines suggest 6 months as the optimal starting age. Earlier start (4-5 months) is appropriate for some families; some gentle methods can be trialled from 4 months.
The current body of evidence does not support long-term harm. A landmark 5-year follow-up (Price et al., Pediatrics 2012) found no differences in stress, emotion regulation, behaviour, or parent-child attachment between sleep-trained and non-sleep-trained children. A 2016 RCT (Gradisar et al., Pediatrics) compared graduated extinction and bedtime fading to control; both groups showed improvements in sleep and parental wellbeing with no increase in cortisol over time. Consult a paediatrician before proceeding if your baby has any medical conditions.
In Dr. Ferber's original protocol: Night 1 - wait 3 min, then 5 min, then 10 min. Night 2 - 5/10/12 min. Night 3 - 10/12/15 min. Night 4 - 12/15/17 min. Night 5 - 15/17/20 min. Night 6 - 17/20/25 min. Night 7 - 20/25/30 min. During check-ins, enter briefly (1-2 min), use a reassuring phrase, but do not pick up.
The key difference is check-ins. Ferber (graduated extinction) uses timed intervals of increasing length before a brief parental check-in. Cry-It-Out (full extinction / Weissbluth) involves placing the baby in the cot at bedtime and not returning until morning or a designated feed time, regardless of crying. CIO typically produces faster results (3-5 nights) but involves more crying. Ferber typically takes 5-7 nights and causes less parental distress.
Yes. Sleep training and breastfeeding are fully compatible. The key distinction is between feeding as nutrition and feeding as the settling method. A breastfed baby who is developmentally ready for sleep training can continue all daytime and dream feeds - the change is that feeding is moved to the beginning of the bedtime routine rather than as the last step before sleep.
The chair method (also called the Sleep Lady Shuffle, developed by Kim West) involves placing a chair next to the cot on night 1 and sitting in it while the baby falls asleep - without picking up, feeding, or rocking. Every 2-3 days, the chair is moved progressively further from the cot until the parent is no longer in the room. The process typically takes 2-3 weeks.
Depends on the method. Extinction (CIO): typically 3-5 nights before significant improvement. Ferber / graduated extinction: typically 5-7 nights for substantial improvement. Chair method: typically 2-3 weeks for full completion. Fading / gentle methods: typically 2-4 weeks, with gradual improvement rather than a step-change. All methods require consistency - abandoning and restarting resets the learning.
Developed by Tracy Hogg ("Baby Whisperer"), pick up, put down (PUPD) involves placing the baby in the cot awake. When the baby cries, the parent picks them up briefly until crying stops, then immediately puts them back down. This cycle repeats until the baby falls asleep in the cot. PUPD works best between 4-8 months. After 8 months, picking up can become counterproductive.
First check: is there an underlying cause (teething, illness, ear infection)? Is the method being applied consistently by all caregivers? Is the baby at a developmental leap or regression? Is the wake window correct? If all four checks pass and there is no improvement after 2 weeks, consider switching to a less stimulating method or consulting a paediatric sleep specialist.
Yes, with one caveat. A dummy is fine to offer at bedtime through sleep training. The challenge arises if the baby cannot yet replace the dummy independently when it falls out during the night - this creates a new settling association. If the baby can replace the dummy themselves (typically 6-8 months), dummy use is compatible with all sleep training methods.
Starting sleep training during an active regression is not recommended - the biological disruption is happening regardless of the method, making it harder to isolate cause and effect, and consistency is harder to maintain under extreme sleep deprivation. However, if you are already mid-way through sleep training when a regression hits, pausing is often counterproductive - many families find holding the method through a 2-4 week regression faster than pausing and restarting.
Related Articles
References
- Price, A. M. H., Wake, M., Ukoumunne, O. C., & Hiscock, H. (2012). Five-year follow-up of harms and benefits of behavioral infant sleep intervention: Randomized trial. Pediatrics, 130(4), 643-651.
- Gradisar, M., Jackson, K., Spurrier, N. J., et al. (2016). Behavioral interventions for infant sleep problems: A randomized controlled trial. Pediatrics, 137(6).
- Mindell, J. A., Kuhn, B., Lewin, D. S., Meltzer, L. J., & Sadeh, A. (2006). Behavioral treatment of bedtime problems and night wakings in infants and young children. Sleep, 29(10), 1263-1276.
- Moon, R. Y., & AAP Task Force on SIDS. (2022). Safe sleep and a healthy sleep environment. Pediatrics, 150(1).
- Ferber, R. (2006). Solve Your Child's Sleep Problems (revised ed.). Fireside.
- Weissbluth, M. (2003). Healthy Sleep Habits, Happy Child (3rd ed.). Ballantine Books.
- Hogg, T., & Blau, M. (2001). Secrets of the Baby Whisperer. Ballantine Books. (Also source for the Chair / Sleep Lady Shuffle developer, Kim West.)
Keep a log this week. See whether it's working.
Lunara's sleep tracker captures settling times, night wakings, and nap data — so you can see progress even when you're too tired to tell.
Open Lunara →