Sleep training is the topic where parenting advice is most polarized and least useful. Internet forums divide families into “cry it out monsters” and “co-sleeping zealots” and contribute approximately nothing to anyone’s evening. Here’s an attempt at a fair walkthrough of the four methods we see in clinical practice, what each one actually requires, and how to think about which one fits your family.

1. Extinction (“cry it out”) — the original Ferber-adjacent

What it is. Put baby down drowsy but awake, leave the room, do not return until morning. Most strict versions allow no check-ins.

What the evidence says. The empirical literature on full extinction is the strongest of the four methods — multiple RCTs show it produces independent sleep within 5–7 nights in roughly 80% of healthy babies aged 6+ months, with no detectable long-term harm. The 2012 Australian Infant Sleep Study followed 326 families to age 6 with no measurable differences in attachment, behavior, or stress markers.

When it fits. Parents who can tolerate 3–5 hard nights, healthy babies 6+ months, and families where one consistent caregiver can implement it.

When it doesn’t. Families with a history of trauma around crying, parents who will fold at night 2 (which extends the process), or any baby with a medical reason to wake (reflux, sleep apnea, weight gain concerns).

2. Graduated extinction (“Ferber method”)

What it is. Put down, leave, return at increasing intervals (3 min, 5 min, 10 min, etc.). Brief check-ins, no picking up.

What the evidence says. Slightly slower than full extinction (7–10 nights average) but with similar outcomes. The check-ins reduce parent distress without measurably extending crying duration.

When it fits. Most families who want sleep training but can’t tolerate complete silence. This is the version we recommend most often in practice.

3. Chair method (“camping out”)

What it is. Parent sits in a chair in the room as baby falls asleep. Each night the chair moves further from the crib. After 2 weeks, parent is at the door, then outside.

What the evidence says. Slower than the extinction methods (3–4 weeks average), with higher abandonment rates. When followed through, similar outcomes.

When it fits. Families with toddlers (not infants), families where the child has high separation anxiety, or co-sleeping families transitioning to a crib.

4. The one nobody names: “responsive bedtime”

What it is. Not formal sleep training. The parent stays available, responds to wake-ups, but works gradually on independent sleep skills over months rather than weeks. Often combined with bed-sharing or room-sharing.

What the evidence says. Slower to consolidate sleep (most babies on this approach don’t sleep through the night until 12–18 months), but no measurable harm. The risk is that some of these families end up with permanent disrupted parent sleep, which is its own clinical issue.

When it fits. Families with cultural traditions of co-sleeping, families with health constraints that require night feeding, or parents who simply find structured training intolerable.

A note on which is “best”

None of the methods is best. The method with the strongest evidence (extinction) is the one most families can’t tolerate. The method with the highest parent comfort (responsive) takes the longest. The choice should be driven by what your family can actually sustain — a method you abandon at night 4 is worse than one you don’t start.

If you’re undecided, graduated extinction is the safest middle ground for most healthy babies 6+ months.