The term "sleep regression" gets applied to almost any period of disrupted baby sleep, which makes it less useful than it should be. In its precise sense, a sleep regression is a temporary period of increased night waking or difficulty settling driven by developmental change — cognitive, physical, or neurological — rather than illness, environment, or routine disruption.

Knowing which regression you're in — if you're in one at all — matters, because different causes call for different responses.

The 4-month regression

The most significant developmental regression, and the only one that represents a permanent change. At around 3.5–5 months, infant sleep cycles mature to more closely resemble adult sleep architecture. Babies who previously slept through cycle transitions begin waking at each cycle end (every 40–50 minutes at night, 20–30 minutes for naps).

This is permanent: sleep doesn't revert to the "newborn" pattern after this regression. The path through it is helping your baby learn to transition between cycles independently, which often involves a gradual settling approach. Most families are through it within 4–6 weeks of applying a consistent method.

The 8–10 month regression

Often coincides with major physical development: crawling, pulling to stand, cruising. Babies practise new motor skills during sleep — their brains rehearse motor sequences — which disrupts rest. Cognitive leaps in object permanence also peak here; they now understand you're gone when they can't see you, making night separation more anxiety-provoking.

Duration: typically 2–6 weeks. Extra physical practice during wake windows helps discharge the overnight motor drive.

The 12-month regression

Often misidentified as a signal to drop a nap. The 12-month regression can look like readiness to drop a nap but is usually temporary. Key distinction: a genuinely nap-transition-ready baby is comfortable staying awake 4–5 hours. A regression baby shows tired cues but fights sleep.

The 15–18 month regression

Linked to the vocabulary explosion of this period. Language acquisition is cognitively demanding, and the processing involved disrupts sleep. This also coincides with the genuine 2-to-1 nap transition for many babies — which is why tracking is especially valuable here to separate the two.

The 2-year regression

Less universal. When it occurs, it's often tied to major life changes — new sibling, starting childcare, moving — or another cognitive expansion. Toddler bedtime resistance is the most common presentation.

What actually helps

Consistency. The most evidence-backed recommendation across every regression. Whatever routine you use — same sequence, same timing, same response to night waking — matters more during a regression than at any other time. Inconsistency extends regressions by preventing the settling habits that help babies through them.

Earlier bedtime. Sleep debt accumulates quickly during regressions. A 30–45 minute earlier bedtime reduces overtiredness, which paradoxically reduces night waking rather than increasing it.

Extra physical activity during wake windows. For physical development regressions (8–10 months, 12 months), extra floor time helps discharge the drive to practise motor skills at night.

What doesn't help

Starting new sleep associations — feeding or rocking to sleep when you weren't previously doing so — typically extends regressions. When it resolves, you'll have the new association to manage as well.

How tracking helps

One of the most valuable things a sleep log does during a regression is show you it's progressing — even when individual nights still feel terrible. Night waking reducing from 6 per night to 4 to 3 looks like real progress on a chart, and knowing you're approaching the end changes the experience of living through it.

Lullwise tracks sleep sessions and wake events, so you can see your regression arc over time and know when you're coming out the other side rather than just enduring it.