Cry it out is not harmful to babies, based on the best evidence available. Multiple studies have tracked infants through sleep training and found no negative effects on stress levels, emotional development, or parent-child attachment. That said, the term “cry it out” gets applied to several different methods, and understanding the distinctions can help you choose an approach that feels right for your family.
What “Cry It Out” Actually Means
The phrase “cry it out” is used loosely, but it typically refers to one of two specific techniques. Unmodified extinction means putting your baby down awake and not returning until morning (or the next scheduled feed). This is the strictest version. Graduated extinction, better known as the Ferber method or check-and-console, is essentially cry it out with check-ins. You leave the room but return at set intervals to briefly soothe your baby, and those intervals get longer each night.
Both methods work by helping a baby learn to fall asleep without being held, rocked, or nursed. The crying isn’t the goal. It’s a temporary protest that fades as the baby develops the ability to self-soothe. Most pediatricians recommend waiting until a baby is at least 4 months old and around 14 pounds before starting any form of sleep training.
What the Research Shows About Stress
The biggest concern parents have is that prolonged crying floods a baby’s system with cortisol, the body’s primary stress hormone, and causes lasting damage. This worry traces largely to a 2012 study by Middlemiss and colleagues, which measured cortisol in 25 mother-infant pairs during a five-day, hospital-based sleep training program. The study found that by the third day, mothers’ cortisol levels had dropped while infants’ levels stayed the same. The authors framed this as a problem: the mother and baby were “out of sync.”
But the actual data told a simpler story. The babies’ cortisol did not increase during the program. Their stress hormone levels stayed flat across the first three days of training. The mothers, meanwhile, became less stressed. The study also had significant limitations: it didn’t report baseline cortisol levels and didn’t track what happened after day three. In short, the most commonly cited study against cry it out doesn’t actually show that it raises infant stress.
A later study published in Pediatrics strengthened the case. Researchers in Australia randomly assigned 43 families to graduated extinction, another gentle sleep method, or no sleep training at all. Saliva tests showed that babies in the sleep training groups had slightly lower cortisol levels than the control group, not higher. Twelve months later, there were no differences among the groups in emotional health, behavioral problems, or parent-child attachment.
The Difference Between Harmful and Temporary Stress
Not all stress is the same. Harvard’s Center on the Developing Child distinguishes three categories. Positive stress involves brief increases in heart rate and mild hormone spikes, like a child experiencing a new situation. It’s normal and necessary for development. Tolerable stress is more intense but time-limited, and the child has a supportive caregiver to buffer the experience. Toxic stress is strong, frequent, and prolonged, occurring without the protection of a caring adult relationship. This is the kind linked to lasting harm.
Sleep training falls squarely into the tolerable category at most. The crying happens in a safe environment, lasts minutes to an hour per session, continues for a matter of days, and occurs in the context of a baby who is otherwise loved, fed, and cared for around the clock. It looks nothing like the neglect, abuse, or chronic deprivation associated with toxic stress. A baby who cries for 40 minutes at bedtime and then spends the rest of the day being held, talked to, and responded to is not experiencing the kind of stress that disrupts brain development.
How Long the Crying Lasts
One reason sleep training sounds worse than it is: parents imagine weeks of screaming. The reality is much shorter. A 2018 study tracking real-world sleep training found that crying peaked on the first night and resolved within a week. Most experts say you should see clear improvement by the end of the first week, with crying largely done by two weeks. If things aren’t getting better in that window, it’s worth pausing and checking in with your pediatrician rather than pushing through.
The first night or two are usually the hardest. After that, most babies cry for progressively shorter stretches before falling asleep on their own. Many parents are surprised at how quickly the process moves once they commit to consistency.
The Effect on Parents
Sleep deprivation is one of the strongest risk factors for postpartum depression and anxiety, and this is where sleep training has some of its clearest benefits. A study published in BMJ Open followed 80 mothers through a sleep intervention and found striking improvements. Depression scores dropped by 66%. Stress scores fell by 42%. Anxiety decreased by 44%. Before the intervention, about a third of the mothers met criteria for some degree of depression. Afterward, only 5% did.
These numbers matter because a parent’s mental health directly shapes a baby’s environment. A mother who is severely sleep-deprived and depressed is less responsive, less patient, and less emotionally available during all the waking hours she spends with her child. If sleep training helps a parent function better during the day, the baby benefits from that improved caregiving far more than they’re harmed by a few nights of crying.
Why the Debate Persists
If the evidence is this reassuring, why does the controversy continue? Part of it is that “letting your baby cry” triggers a deep biological alarm in parents. Infant crying evolved to demand a response, and ignoring it feels wrong on a visceral level. That instinct is healthy and normal, but it doesn’t mean every cry requires immediate intervention.
There’s also a gap between how sleep training is discussed online and what the research actually measures. Critics often extrapolate from studies on severe institutional neglect (Romanian orphanages, for example) to argue that any unattended crying causes brain damage. But the conditions in those studies bear no resemblance to a loved baby crying in a crib for 20 minutes while a parent watches the monitor in the next room. Conflating the two misrepresents the science.
Finally, parenting identity plays a role. For families practicing attachment parenting, sleep training can feel like a philosophical betrayal. That’s a valid personal preference, but it’s not an evidence-based safety concern. Families who choose not to sleep train aren’t doing anything wrong, and families who do choose it aren’t either.
Choosing the Right Approach
If you’re considering sleep training, graduated extinction (Ferber) is often easier for parents to tolerate because the periodic check-ins provide reassurance that the baby is okay. Some babies, however, get more upset when a parent appears and then leaves again. For those babies, unmodified extinction can actually result in less total crying.
Whichever method you choose, consistency matters more than the specific technique. Going in to pick up your baby after 45 minutes of crying teaches them that 45 minutes of crying gets results, which makes the next night harder. The families who struggle most with sleep training are usually the ones who start and stop repeatedly.
Sleep training also works best when the basics are in place: a consistent bedtime routine, a dark and cool room, age-appropriate wake windows during the day, and no underlying medical issues like reflux or ear infections that could be causing pain. If your baby is in genuine discomfort, addressing that comes first.