Toddler sleep regressions: 18 months and 2 years Happyly baby sleep coaching app toddler sleep regression guide

If you survived the 4-month sleep regression, you might expect the toddler regressions to look the same. They do not.

Infant regressions are about sleep architecture — the brain is physically reorganizing how it sleeps. The result is frequent night wakings. Toddler regressions are about cognitive and emotional development — the brain is learning about separation, autonomy, cause and effect, and the power of language. The result is bedtime battles.

Avi Sadeh and Jodi Mindell's large web-based study of 5,000 children aged 0-3 documented this shift: the 18-month and 24-month sleep disruptions were characterized by bedtime resistance and early morning waking rather than the frequent night wakings typical of infant regressions. Separation anxiety was a significant predictor of bedtime resistance in the 18-24 month window.

The 18-month regression typically arrives alongside a cluster of developmental milestones: walking consolidation or running, a language explosion (vocabulary often doubles between 18 and 24 months), the emergence of separation anxiety, and the beginning of autonomous will.

Anat Scher's motor milestone research documented how major physical skills — crawling, standing, walking — temporarily fragment sleep. At 18 months, many toddlers are consolidating walking or learning to climb, and the practice does not stop at night. Parents describe toddlers standing in the cot at 2 AM unable or unwilling to lie back down.

Separation anxiety adds a layer that was not present in infant regressions. The attachment research from John Bowlby and Mary Ainsworth established that separation anxiety peaks between 18 and 24 months — the toddler understands that the parent can leave but does not yet fully trust that they will return. This is why bedtime suddenly becomes a crisis: you are leaving the room, and the toddler's developing brain reads that as abandonment.

The good news: this is a phase. Pennestri and colleagues' data reassures that night waking and sleep disruption at this age do not predict worse developmental outcomes. The regression resolves when the developmental surge integrates.

The 2-year regression is similar in flavor to the 18-month one but with a new ingredient: language. Your 2-year-old can now articulate demands, fears, and preferences. 'I do not want to sleep.' 'I am scared.' 'One more story, please.' The negotiation skills are genuinely impressive and genuinely exhausting.

Additional drivers at 2 years:

• The [nap transition](/guides/nap-transitions) — many toddlers are transitioning from 2 naps to 1 around this age, and the overtiredness from the transition can compound bedtime resistance.
• Potty training — if toilet training is underway, the nighttime potty need becomes a legitimate and sometimes strategic request.
• Imagination development — 2-year-olds develop the capacity for fear of the dark, monsters, and shadows. These fears are real to them even if they seem invented to you.
• Crib escape — many toddlers learn to climb out of the crib between 18 and 30 months. If this happens during a regression, the combination of physical freedom and emotional dysregulation makes bedtime very difficult.

The 2-year regression can last 2 to 6 weeks but may extend longer if boundaries are inconsistent. Sadeh's data suggests that clear, predictable responses to bedtime resistance shorten the disruption.

1. Protect the bedtime routine. This is the single most important thing you can do. Jodi Mindell's research consistently shows that routine consistency is the strongest protective factor against bedtime resistance. Same steps, same order, same time. The routine is the anchor. See our bedtime routines guide for the full breakdown.

2. Validate the feeling, hold the boundary. When your toddler says 'I am scared' or 'do not leave,' acknowledge it: 'I know you want me to stay. I love you. It is time to sleep.' The acknowledgement matters. The boundary holds. Dismissing the feeling ('there is nothing to be scared of') does not help — they are not rational yet.

3. Use the one-return rule. After goodnight, you will come back once if called. The return is brief, warm, and boring: a pat, a quiet 'I love you, sleep time.' After the one return, no further engagement. This gives the toddler the reassurance that you respond while setting a clear limit.

4. Do not start a new sleep training method mid-regression. The regression will resolve on its own for most families. Introducing graduated extinction or a new approach during the peak of separation anxiety can backfire. Wait until the acute phase passes (2-3 weeks), then reassess.

5. Check the nap. If the regression coincides with overtiredness from a nap transition, the bedtime resistance may be partly sleep-pressure-driven. An earlier bedtime often helps more than any behavioral strategy.

6. Delay the crib-to-bed transition if you can. If your toddler has not yet climbed out of the crib, keep them in it through the regression. Adding physical freedom to emotional dysregulation doubles the problem. If they have climbed out, make the room safe and use a gate or closed door — the boundary needs to be physical when the toddler's impulse control is not yet there.

The most demoralising part of a toddler regression for parents who sleep trained is the feeling that all that work was for nothing. It was not.

The sleep training foundation is still intact. Your toddler still knows how to fall asleep independently — they are choosing to resist because a developmental imperative is temporarily overriding the habit. When the developmental surge integrates (usually 2-6 weeks), the old pattern re-emerges without re-training.

What matters now is not re-training. It is not backsliding. Do not add new settling methods (rocking, feeding, lying with them) that were not part of the pre-regression routine, because those CAN become new habits. The instruction is simple: hold the routine, hold the boundaries, ride it out.

If you never sleep trained and are managing with your current approach, a toddler regression is not the right time to start. Let the regression pass. Then, if you want to make changes, see our gentle sleep training guide.