Cry-it-out doesn't work the way you think — what the research really says

“Just let her cry, she’ll learn.” One of the most common and most contested sleep tips out there. After a few nights, the baby does fall asleep faster. What many parents don’t know: the child’s body is not doing this because it feels safe.

In this article we look at what really happens during cry-it-out, based on research from pediatric journals and the attachment literature.

The difference between “quiet” and “calm”

In 2012, Wendy Middlemiss and her team at the University of North Texas published a study measuring cortisol (the stress hormone) in babies during a five-day sleep training. What they found:

On day one, cortisol peaked in both baby and mother. Both were highly stressed. As expected.
By day three, the baby was quiet. Sleep training “successful”.
But the baby’s cortisol peak was just as high as on day one. The mother had calmed down. The baby had not.

The baby didn’t learn that the night was safe. She learned that crying no longer produced a response. The body stayed in alarm mode, but the outward signs disappeared.

Source: Middlemiss W. et al. (2012). Asynchrony of mother-infant hypothalamic-pituitary-adrenal axis activity following extinction of infant crying responses induced during the transition to sleep. Early Human Development.

What does the attachment literature say?

John Bowlby (British child psychiatrist) and Mary Ainsworth (Canadian developmental psychologist) built the attachment theory that nearly every modern pedagogy program rests on. One of its core points:

“A child who is consistently and sensitively responded to in moments of distress develops an internal model that says: I can rely on others when I need them.”

The British NICE guidelines and the Dutch Youth Institute (NJI) summarize it bluntly: “Sensitive responsiveness in the first year of life predicts secure attachment at twelve months, and that attachment predicts social-emotional functioning into adolescence.”

Cry-it-out training intentionally interrupts that responsiveness. The baby learns: “Sometimes someone responds, sometimes nobody does. Better to stop sending signals.”

A working mother with her baby — care and exhaustion often go hand in hand

But my friend did it and her child sleeps wonderfully

That is probably true. Cry-it-out often produces a calmly sleeping child in the short term. The question is what’s underneath:

Short term: less crying at night, parents sleep better
Medium term: mixed. Some studies see no difference, others see more sleep problems around toddler age
Long term: hard to prove because so many other factors play a role

What does come back consistently: children who are reliably soothed in their first year show fewer anxiety disorders later and better self-regulation. That is not a minor side effect.

Source: National Institute of Child Health and Human Development (NICHD), Study of Early Child Care, longitudinal cohort.

What are the alternatives?

Sleep training a baby through complete ignoring is one extreme. Picking them up day and night at every peep is the other. The middle ground supported by most pediatric associations:

Gradually responding with decreasing intensity:

First week: at every cry, sit down beside them, hand on their chest
Then: use voice from the doorway only
Then: brief wait (1-3 minutes), respond if crying continues

That is called “fading”. The child learns that the night is safe — and you are still there.

Other practical points:

Consistent evening routine: bath, feed, short story, lullaby. Predictability lowers cortisol.
Light presence: stroking the back without picking up works for many babies from 6 months.
Daylight by day, dark in the evening: the melatonin rhythm starts kicking in around 3 months.

A baby who is held feels seen — the foundation of good sleep

What if you genuinely can’t go on?

Parental sleep deprivation is not a small factor. An exhausted parent is less sensitive, less patient, and at higher risk of postnatal depression. That is also bad for the baby. A few considerations:

Take turns with your partner per night — don’t both be exhausted
Ask grandparents to take the baby for a morning, sleep four hours yourself
Consult your local clinic / health visitor: in most countries this is free and the advice is usually evidence-based
If in doubt: GP or pediatrician, especially if it lasts more than 6 weeks

One stressful night in which a baby cries for 5 minutes while you breathe does not break attachment. Weeks of systematic ignoring is something else entirely.