Sleep deprivation is one of the hardest parts of early parenthood, and the question of whether and how to sleep train is one of the most searched, most debated topics in parenting — generating strong opinions on all sides. If you have found yourself reading conflicting advice at 3am, you are in good company. The aim of this article is to cut through the noise with a clear, evidence-based account of what sleep training is, what the main approaches involve, and what the research actually shows about their safety and effectiveness.
The short version is that multiple methods work, none of the well-researched ones harm babies, and the most important factor is consistency. Understanding the options in detail will help you choose the approach that fits your baby's temperament and your own capacity, which dramatically increases your chances of succeeding.
If you want to track your baby's sleep patterns and see whether your chosen approach is working over time, Healthbooq has a dedicated sleep log that makes it easy to spot progress across weeks rather than focusing only on individual difficult nights.
What Sleep Training Is — and Is Not
Sleep training refers to the process of helping a baby learn to fall asleep independently at the start of the night and, with time, to settle themselves back to sleep when they wake between sleep cycles. It does not mean eliminating night feeds for young babies who still need them, and it is not about forcing a baby to sleep. What it does address is the distinction between a baby who can fall asleep alone and one who has learned that falling asleep requires a specific input from a caregiver — feeding to sleep, rocking, or being held — and therefore needs that same input every time they surface from a light sleep during the night.
Most sleep training methods are designed for babies from around four to six months of age, when the circadian rhythm is sufficiently developed and the neurological capacity for self-settling is established. Before four months, babies' sleep patterns are governed largely by hunger and developmental stage rather than learned associations, which is why most sleep experts do not recommend structured sleep training earlier than this.
Graduated Extinction (the Ferber Approach)
The most well-known sleep training method, often called the Ferber method or graduated extinction, involves putting your baby down awake and leaving the room, then returning at progressively longer intervals to briefly reassure them without picking them up. The intervals increase gradually over successive nights — starting at three minutes, then five, then ten — giving the baby increasing windows in which to practise settling independently while still receiving parental reassurance.
Research on this method is extensive. A 2012 randomised controlled trial published in the journal Pediatrics found that graduated extinction improved infant sleep, reduced maternal stress, and showed no evidence of harm to child emotional development, cortisol levels, or the parent-child attachment relationship. Follow-up studies at age five found no measurable differences in behaviour or emotional wellbeing between sleep-trained and non-sleep-trained children. This is among the strongest evidence bases of any sleep approach.
Full Extinction (Cry It Out)
Full extinction, sometimes called "cry it out", involves placing the baby in their cot awake and not returning until morning — or until a pre-set time. It is a faster method than graduated extinction, typically producing results in two to four nights, but it requires a higher tolerance for crying in the initial nights. The same body of evidence that supports graduated extinction applies here: long-term studies consistently show no negative outcomes for emotional development, attachment, or cortisol regulation.
The method is most effective when the baby is healthy, when the room environment is consistent (dark, cool, with white noise), and when both caregivers are fully committed, since any inconsistency tends to prolong the process.
Chair Method (Sleep Lady Shuffle)
The chair method — popularised as the "Sleep Lady Shuffle" — takes a more gradual approach to parental withdrawal. On night one, you sit in a chair next to the cot while your baby settles; every two to three nights, you move the chair a little further away from the cot until eventually you are outside the room. You provide verbal reassurance but do not pick the baby up or help them fall asleep. The process takes two to three weeks and involves less acute crying than the extinction methods, though the process is longer and requires sustained parental presence at bedtime each night.
Pick Up, Put Down and No-Cry Approaches
Several methods take a softer approach, involving picking the baby up when they cry and putting them back down when calm, or using a combination of patting, shushing, and very gradual withdrawal of sleep associations. These approaches have a smaller evidence base but are well-regarded by many families. They tend to work more slowly than extinction methods and may not be effective for all babies, particularly older infants or those with well-established sleep habits. However, for parents who cannot tolerate any crying, they represent a genuine path forward.
Preparing for Success
Regardless of which method you choose, a consistent and calming pre-sleep routine is the single most important preparation. A routine of 20 to 30 minutes — a bath, a feed, a story or song, and then into the cot — creates a reliable signal for the body that sleep is coming. Babies learn associations very quickly, and a predictable routine does much of the heavy lifting before the method itself even begins.
Timing matters too. Putting a baby down to sleep at the first signs of tiredness — yawning, glazed eyes, quieting down — rather than waiting until they are overtired produces far better results. An overtired baby's cortisol levels have already risen, making it harder for them to settle even with help.
Key Takeaways
All major sleep training methods — including those that involve some crying — have been shown in research to be safe and effective, with no evidence of long-term psychological harm. The best method is whichever one you can apply consistently for at least two weeks. Most babies show significant improvement within five to seven nights. Sleep training is generally not recommended before four months of age. A predictable pre-sleep routine is the single most effective preparation, regardless of which method you choose.
