Anytime you hear the phrase "sleep regression," I want you to reframe it immediately: think PROgress. Your baby is progressing and growing — mentally, physically, or both — and that growth is temporarily distracting her from sleep. The term "regression" implies something is going wrong. It isn't. What you're watching is your child's brain and body doing exactly what they're supposed to do.
If you're a parent in Palo Alto, San Francisco, San Jose, Mountain View, Menlo Park, or anywhere across the Bay Area searching for answers at 3 AM, you've landed in the right place. After more than a decade working with over 1,000 families, I can tell you that understanding why these disruptions happen is the most powerful thing you can do — because it shifts you from panic mode into problem-solving mode.
Let's walk through every major sleep regression age, what's actually driving each one, and what you can do about it.
What a Sleep Regression Actually Is (And Why It Keeps Coming Back)
Think about what you need to fall asleep: the right physical conditions — a quiet, dark, cool room — and the ability to relax yourself into sleep. A simple set of cues that tell your brain it's time to wind down, finding your position, and shutting things off.
Now imagine your brain is in the middle of a major reorganization. New motor skills are coming online. New cognitive abilities are forming. Your nervous system is literally rewiring itself. Relaxing into sleep becomes significantly harder when your brain has other business to attend to.
That is what a sleep regression is. It's not a flaw. It's not a sign that something broke. It's a reflection of development.
The reason some families feel these regressions barely at all while others feel completely derailed comes down to one thing: whether or not their child has the skill of independent sleep. A baby who has learned to fall asleep on their own — without being nursed, rocked, or bounced — can get through a developmental disruption with minimal fallout, because they already know how to return to sleep on their own. A baby who has always depended on a parent to facilitate sleep will be significantly more impacted, because that crutch becomes unavailable in the middle of the night when the regression hits.
This is the foundation of The Helping Babies Sleep Method: we address sleep as a skill, not just a schedule. And understanding sleep regressions is a big part of knowing when and how to teach that skill.
The 4-Month Sleep Regression — The Big One
The most significant sleep regression happens between 3.5 and 4 months of age, and it's in a category of its own. This isn't a temporary blip. This is a permanent neurological shift.
Your baby's sleep architecture is reorganizing to more closely resemble adult sleep — moving from simple, newborn-style cycles into the lighter and deeper stages that all humans experience. Along with this comes an explosion in cognitive awareness. Your baby is developing object permanence — the understanding that you exist even when she can't see you. She's literally waking up to the world around her, and that world is fascinating and stimulating and hard to switch off from. You can read more about the 4 month sleep regression in particular in this post.
Parents frequently describe the same pattern: a 3-month-old who was sleeping 6- to 8-hour stretches is now waking every 2 to 3 hours. Naps that were running 1 to 2 hours are suddenly capping at 45 minutes. Bedtime is taking longer. Everyone is exhausted.
Here's what I tell families across the Peninsula: I personally experienced both sides of this regression. With my first child, I was already up every 2 hours nursing him back to sleep, so the regression barely registered as a change. With my daughter, who I supported in developing independent sleep skills from early on, we barely felt it — because she had the tools to return to sleep on her own when she surfaced between cycles.
The 4-month regression is also the moment when a feed-to-sleep association that felt manageable in the newborn stage becomes genuinely disruptive — because sleep cycles are now maturing and your baby is surfacing between them more frequently and more fully. If nursing or a bottle is how your baby has always transitioned to sleep, she will seek that same bridge at every cycle transition through the night. This is why addressing the feeding-sleep connection is so important at this stage, which we go into in depth in The Feeding and Sleep Connection: What Parents Need to Know.
Does the 4-month sleep regression end? For some families, night wakings naturally recede over a few weeks as the neurological reorganization settles. For others — particularly babies with strong sleep associations — the pattern established during the regression persists until the family makes a deliberate change. This is often the first moment families begin seriously considering sleep teaching, and it's a completely appropriate time to do it. We walk through exactly how that process works in The Two Phases of Sleep Training: Gentle Foundations to Behavioral Change.
One more thing worth naming here: if your 4-month-old is waking frequently and you're not sure whether it's developmental or driven by hunger, this is important to sort out before trying any sleep teaching. We cover this in detail in Why Sleep Training Fails When Feeding Isn't Addressed.
5 to 8 Month Sleep Regressions — Motor Development Takes Over
These regressions get talked about as the "6-month sleep regression" or "7-month sleep regression" or "8-month sleep regression," and I want to be honest with you: they are real, but they're driven by something different from the 4-month shift. The 4-month regression is neurological and permanent. The 5-to-8-month window is mostly about motor development, and the timing varies wildly from baby to baby.
Rolling, hovering on all fours, pulling up, standing in the crib — your baby's body is moving in all directions, and her brain is practicing these movements constantly, including at sleep time. Sleep regressions related to motor development often look different from the 4-month regression. Instead of just crying, you may find your baby awake in the crib, seemingly content, practicing her new skills at 2 AM. She's not distressed. She's busy.
There's also the concept popularized by The Wonder Weeks — mental leaps that correspond to cognitive developmental windows. While there isn't strong empirical evidence that the exact timing maps onto specific weeks the way the app suggests, anecdotally many parents find it resonates with their experience. The underlying premise — that cognitive growth temporarily disrupts sleep — is sound.
Here's the practical reality: if your baby is an independent sleeper, these regressions are largely something you can sleep through yourself. She wakes, she rolls around, she finds her comfortable position, and she goes back to sleep. If your baby depends on you to fall asleep, these months can be grueling — you're awake waiting for her to settle, repeatedly going back in, and getting less sleep than your baby is.
This is also the age range where the drowsy but awake approach, which is commonly recommended, often starts to clearly fail. If you've been placing your baby down drowsy and wondering why it's getting harder rather than easier, that post explains exactly why. You can read more about this 8 month period in this blog post.
The 9-10 -Month Sleep Regression — Separation Anxiety Arrives
Around 9 months, many parents notice a shift that feels different from the motor-driven disruptions of the previous months. Bedtime resistance increases. Babies become clingier during the day. Previously effective settling approaches start losing their power.
This is the separation anxiety regression, and it makes complete developmental sense. Your baby now has a solid grasp of object permanence — she understands you exist when you leave the room — and she has opinions about it. She wants you there. She protests when you're not.
For families in the Bay Area where both parents are often working full days, the evening hours are particularly charged. Your baby has been with a caregiver or in a daycare environment all day, and separation at bedtime hits differently when her attachment to you is actively developing.
What can you do? Give more during connection windows. Lean into floor time and eye contact and responsiveness during waking hours. And at bedtime, be warm but consistent. If your baby has learned to fall asleep independently, this regression tends to pass within two to three weeks. If she is still dependent on you to initiate sleep, the separation anxiety can make this particularly difficult — because every attempt to step back now involves her already-heightened distress about your absence. You can read more about this 9-10 month sleep regression in this post.
First Year Molars and the 12-Month Sleep Regression
First-year molars can appear anytime after 12 months, and the timing varies significantly from child to child. When molars are cutting in the first year, what looks like a sleep regression is primarily discomfort-driven. The most common pattern: good daytime naps, but bedtime resistance and one finger perpetually in the mouth at night. Some parents also see night waking in babies who were previously sleeping through reliably.
Teething disruptions tend to be more short-lived than developmental regressions — typically days rather than weeks — and respond well to comfort measures, appropriate pain management if needed, and consistent routines. If the disruption is lasting more than two to three weeks without resolution, it's worth considering whether there's a sleep association layer that the teething period activated or worsened. You can read more about toddler sleep regressions in this post.
The 14 to 18 Month Sleep Regression — Toddler Territory
I use these ages loosely because what's happening here is a range, not a precise window. Somewhere between 14 and 18 months, your baby is transitioning firmly into toddlerhood — and toddlers have preferences, opinions, and an extremely well-developed fear of missing out.
Bedtime resistance is the hallmark here. More night waking than you saw at 12 months. A child who was going down easily now protests being separated from the action of the household. This is developmentally normal. Toddlers are impulsive, boundary-testing, and motivated by whatever is most engaging in the moment — and the household in the evening hours is generally more engaging than a dark room and a crib.
What this is not a sign of is that your child needs more parental involvement to fall asleep, even though the regression may look like she does in the moment. Responding to the regression with more rocking, more nursing, more presence can inadvertently create or reinforce associations that will outlast the developmental phase by months. The regression passes. The sleep association stays.
Warmth and consistency together is the approach that works best here. Acknowledge her feelings. Hold the boundary. This is true whether you're in a two-bedroom apartment in San Francisco or a house in the Willow Glen neighborhood of San Jose — the developmental needs are the same. You can read more about these toddler months in this blog post.
22 Months — The Language Burst
Around 22 months, many toddlers go through a significant language expansion — the period when words and sentence fragments start coming rapidly. This can manifest in sleep as your child being awake but content in the middle of the night, babbling to themselves, narrating something from their day, or simply chatting into the dark. It can also look like increased bedtime resistance and a sudden desire for extended conversation at lights-out.
If your toddler is an independent sleeper, this one is mostly harmless — a strange middle-of-the-night audio experience that usually resolves on its own within a few weeks. If sleep associations are still present, the night babbling can escalate into a full request for company or feeding.
Second Year Molars
Second year molars are coming in somewhere between 20 and 30 months, and they are notoriously more uncomfortable than earlier teeth. The disruption pattern looks similar to first year molars but can feel more intense — finger in the mouth at bedtime, early morning wake-ups (often 5 AM), and occasional night waking from discomfort. Managing pain appropriately during the cutting window is the primary lever here, alongside maintaining as consistent a routine as possible.
2 to 2.5 Year Nap Resistance
Somewhere between 2 and 2.5 years, many toddlers go through a period of pronounced nap resistance — fighting nap time, taking 45 minutes to settle, or refusing entirely. Many parents interpret this as a sign that the nap is ready to be dropped.
It usually isn't. Most children genuinely need a nap until 3 to 3.5 years of age. Dropping the nap too early in response to this regression tends to create an overtired toddler who is harder to manage at bedtime, wakes earlier in the morning, and has a harder time through afternoon activities. The nap resistance in this window is typically about not wanting to stop what they're doing, not about not needing sleep.
Hold the nap. Keep the routine consistent. This too shall pass. Read more on the two year old sleep regression in this blog post.
What All of This Has in Common
Reading through every one of these regressions, the thread running through all of them is the same: children who have the skill of independent sleep navigate these disruptions significantly more smoothly than children who don't.
This is not about being a perfect parent or following a rigid method. It's about understanding that falling asleep independently is a developmental skill — one that can be taught, gently and sustainably, when the right foundations are in place. We cover exactly how that process works across the full age range in The Helping Babies Sleep Method and in The Two Phases of Sleep Training.
It’s also worth understanding that these little sleep regressions are normal and all humans experience them. In adults will have little bouts of insomnia where our sleep is disrupted when our brains are busy thinking about something exciting like a pending new baby or something stressful, like a pending new baby :)
It's also worth noting that for many of the families I work with — particularly in the 4- to 9-month range — what looks like a regression is actually a feeding pattern that has quietly been disrupting sleep for weeks or months. A snacking cycle that was manageable in the newborn stage, a feed-to-sleep association that has solidified over time, a daytime intake pattern that leaves the baby genuinely hungry at night. If you've tried everything and the regression isn't resolving on the expected timeline, The Snacking Cycle: Is Daytime Feeding Affecting Your Baby's Night Sleep? is worth reading carefully.
And if you're wondering whether the approach you're considering — Ferber, CIO, or something else — is right for your family, we go through how The Helping Babies Sleep Method compares in this post.
When Will My Baby Sleep Through the Night?
Parents ask this constantly, and the honest answer is: it depends on what's in place. A very generous benchmark is that by 12 months of age, most children are sleeping 11 or more hours through the night without needing to eat. Many children achieve this between 8 and 9 months with the right foundations — though breastfed babies may still need one night feed through that window.
The key is not waiting passively for it to happen. Sleep is a skill. The regressions will keep coming — development doesn't stop — but a child who has the tools to self-soothe will ride those waves far more smoothly than one who doesn't.
Two weeks of great sleep, a brief disruption during a developmental leap, back to great sleep. That is the pattern I see over and over again in families who have done the foundational work. That is what's possible for your family too. You can read more on sleeping through the night in this blog post.
Frequently Asked Questions About Baby Sleep Regressions
What exactly is a sleep regression, and is it different from a sleep problem?
A sleep regression is a temporary disruption to previously established sleep patterns that is directly tied to a developmental milestone — cognitive, physical, or both. It is not the same as a sleep problem, though the two can look identical on the surface. The key distinction is this: a true regression is driven by something your baby is gaining (a new motor skill, a cognitive leap, a language burst), and it typically resolves on its own within two to four weeks once that developmental phase settles. A sleep problem — like a feed-to-sleep association or a snacking cycle that's driving genuine nighttime hunger — doesn't resolve on its own because it isn't developmental. It's structural. One of the most common mistakes I see families make is waiting out what they assume is a regression when what's actually happening is a deeper pattern that has quietly been in place for months. If a disruption is lasting longer than three to four weeks, that's your signal to look more carefully at the foundations.
How many sleep regressions are there, and when do they happen?
There isn't a fixed number, because development doesn't happen on a fixed schedule. That said, the most commonly recognized windows are around 4 months, 6 to 8 months (tied to motor development), 9 months (separation anxiety), 12 months (first year molars for some children), 14 to 18 months (boundary testing and FOMO), 22 months (language burst), sometime between 20 and 30 months (second year molars), and 2 to 2.5 years (nap resistance). The 4-month regression is in a category of its own because it reflects a permanent neurological change in how your baby's sleep cycles are organized — it doesn't fully reverse. All the others are temporary disruptions that come and go as development moves through phases. The timing also varies from baby to baby, which is why your friend's 7-month sleep regression might look like your baby's 8-month regression — because their babies are hitting motor milestones on slightly different timelines.
Is the 4-month sleep regression permanent?
The regression itself is not permanent, but the neurological change that causes it is. Before 4 months, babies sleep in relatively simple cycles. After the 4-month shift, their sleep architecture reorganizes to more closely resemble adult sleep — with distinct lighter and deeper stages, and more frequent and complete surfacings between cycles. That change doesn't reverse. What this means practically is that a baby who had a relatively strong feed-to-sleep association before the 4-month regression will feel the regression far more acutely — because they're now surfacing between cycles more frequently and more fully, each time reaching for the same bridge back to sleep that was present at bedtime. For families wondering whether this is a regression or the beginning of something that needs to be addressed, The Two Phases of Sleep Training walks through how to tell the difference and what the path forward looks like.
How long does a sleep regression last?
Most true developmental sleep regressions last two to four weeks. Teething-related disruptions (first year and second year molars) tend to be shorter — often days rather than weeks — and track closely with the active cutting period. The 4-month regression can feel longer because the neurological shift is permanent and, for babies with sleep associations, the increased night surfacing doesn't improve until something about how they fall asleep changes. If you're four to six weeks into what started as a regression and things are not improving or are getting worse, the regression has likely resolved but a sleep pattern issue has been left behind in its wake. That's the moment to assess the foundations rather than continue waiting.
Will my baby's sleep regression go away on its own?
Sometimes, yes — and it depends almost entirely on one thing: whether your baby can fall asleep independently. A baby who falls asleep on their own at bedtime has a pathway back to sleep at 2 AM that doesn't require you. When a developmental disruption hits, they may have a harder time for a week or two, but they already have the skill and can use it. A baby who depends on nursing, rocking, or a bottle to fall asleep doesn't have that pathway. Every time they surface between sleep cycles during a regression, they're waiting for the same conditions that were present at bedtime — and those conditions require your involvement. For those babies, regressions tend not to fully resolve on their own because the underlying skill gap isn't developmental. It just becomes more visible when development stirs things up.
How do I know if my baby's night waking is a sleep regression or a hunger issue?
This is one of the most important questions to ask, and it doesn't always have an obvious answer in the moment. A few things I look at: How is your baby feeding during the day? Are feeds full and focused, or short and frequent? Is your baby gaining weight consistently along their own curve? How does your baby respond when fed at night — are they taking a complete, full feed with active swallowing, or a few sips and then drifting off? A baby waking from genuine hunger typically wakes with increasing urgency and takes a real feed. A baby waking from habit or a sleep association will often take a small amount and settle quickly without a complete feed. If daytime feeding has been fragmented or rushed — a pattern I see constantly with Bay Area families managing demanding work schedules — there's a good chance the nighttime picture is at least partly nutritional rather than purely developmental. The Feeding and Sleep Connection goes deep on exactly how to tell the difference and what to do about it.
Should I sleep train during a sleep regression?
Generally, I recommend completing the foundation work of Phase 1 — optimizing schedule, feeding, environment, and routines — before beginning any behavioral sleep teaching, and I would not typically start Phase 2 behavioral work in the middle of an active developmental leap or illness. That said, the 4-month regression is a special case. Because the neurological change it represents is permanent, many families find that waiting for the regression to "pass" before sleep teaching means waiting for something that isn't going to fully resolve on its own if a sleep association is present. If the other readiness markers are in place — your baby is healthy, feeds are optimized, the environment is set up well, and you have support and consistency — the 4-month window can actually be a good time to begin, not a reason to delay. The Two Phases of Sleep Training walks through the full readiness checklist so you can assess where you are.
My baby was sleep trained and now a regression hit — do I have to start over?
No, and this is one of the most relieving things I tell families. A baby who has genuinely learned to fall asleep independently — not one who was sleep trained in the behavioral sense but still has underlying foundational issues — will almost always find their way back after a developmental disruption without starting from scratch. You may have a few days to a week, as their expectations of you have changed. You may need to hold your boundaries a little more firmly at bedtime while the leap is active. But the skill is there. Development doesn't erase a learned skill; it just temporarily makes it harder to access. The families who do find themselves truly back at square one after a regression usually have something else going on underneath — a feeding pattern that reactivated, a schedule that drifted, or an association that was present before the regression and became fully entrenched during it.
Is the 9-month sleep regression related to separation anxiety?
Yes, directly. Around 9 months, your baby's cementing the understanding of object permanence which started at 4 months— the concept that you exist even when you're not in the room — is solidifying. In sum, you are their preferred person. It’s not a bad thing. It’s a sweet and flattering thing. This is a major cognitive achievement, and it brings with it a heightened awareness of your absence. Bedtime becomes emotionally charged in a way it wasn't before, because your baby now understands what it means for you to leave. Babies who have always relied on a parent's presence to fall asleep often find this regression particularly difficult. The most effective response is to increase emotional connection and responsiveness during waking hours, hold warm but consistent boundaries at bedtime, and understand that this phase is time-limited. The regression typically passes within two to four weeks for most children.
When do sleep regressions stop?
They don't stop entirely — development and thus distraction keeps moving through the toddler and preschool years and really into adulthood. But they do become more manageable and shorter-lived as your child gets older, gains more emotional regulation, and has more language to process what's happening. By age 3 to 3.5, most children have consolidated to one nap (or no nap) and have enough emotional maturity and language that sleep disruptions look more like resistance or stalling than the kind of night waking you see in infancy. The families I work with throughout the SF Bay Area who have done the foundational work early consistently describe the same thing: the regressions still come, but they ride through them in days rather than weeks because their child has the skills to self-soothe and the routine to return to on the other side.
I've been told it's a regression, but it's been two months. What's going on?
Two months is well beyond the typical regression window, and I'd encourage you to stop waiting for it to pass on its own. At this point, the developmental disruption has almost certainly resolved, but something in the sleep picture has been left behind — a reinforced association, a feeding pattern that shifted during the hard weeks, a schedule that drifted and was never corrected. The good news is that this is solvable, and it usually doesn't require starting from scratch. It requires identifying which piece is missing and addressing it systematically. If you've been in this pattern for two or more months and nothing is improving and you’re ready to invest in coaching, a Discovery Call is the fastest way to figure out what's actually going on. We work with families throughout San Francisco, Palo Alto, Menlo Park, Mountain View, San Jose, and the broader Bay Area, as well as virtually for families outside the area — and this kind of stalled progress is one of the most common reasons families reach out.
Ready to Get Ahead of the Next Regression?
If you're currently in the middle of a sleep regression and you’re tired of being the one who “makes sleep happen” and ready to help your little one develop self soothing skills — a Discovery Call is the fastest way to get clarity. We work with families throughout the Bay Area, including San Francisco, Palo Alto, Menlo Park, Mountain View, Los Altos, Sunnyvale, San Jose, and the broader Peninsula and South Bay, as well as virtually nationwide.
You can also take the free Sleep Quiz for age-specific insights, explore our online classes, or if you want to understand the full framework first, start with The Helping Babies Sleep Method.
Not sure whether you need a sleep consultant at all, or how to choose one? This post walks through exactly what to look for.
You can be loving, attached, and well-rested. All three at the same time. That's what this work is about.
Sarah Mitchell is a Certified Lactation Counselor, retired Doctor of Chiropractic, REI graduate, certified NLP Coach, and author of the #1 bestselling book The Helping Babies Sleep Method: The Art and Science of Teaching Your Baby to Sleep. She has worked with over 1,000 families across the San Francisco Bay Area and beyond, with in-home visits available in Palo Alto, Menlo Park, Atherton, Los Altos, Mountain View, Sunnyvale, Cupertino, San Jose, San Francisco, and surrounding communities. Virtual coaching is available nationwide.
