Newborn Reflexes: What They Mean and Where They Actually Go

Your baby's brain at birth: the lower floors are running, the upper floors are under construction

When your baby's cheek gets touched and the head swivels toward your finger — your baby didn't decide to do that. The brainstem ran the program before the thinking brain even got the message.

That's the whole story of newborn reflexes in one sentence. A baby's brainstem and spinal cord (which handle breathing, heart rate, and basic motor responses) are largely complete at birth. The cerebral cortex, responsible for voluntary movement, memory, reasoning, anything we'd call "thinking" — is still under rapid construction. So for the first several months, the lower brain is in charge of a lot.

This is why reflexes exist at all. According to the clinical review in StatPearls, these responses are involuntary motor programs that facilitate survival in early development. They don't require conscious thought because the cortex isn't ready to provide it. Touch the cheek, rooting happens. Drop the head back slightly, and the Moro fires. The baby isn't deciding anything — the lower brain is running the software.

And this is also exactly why reflexes "disappear." As the cortex matures over the first months of life, it sends inhibitory signals down to the brainstem and spinal cord, overriding the automatic programs with voluntary control. The circuit doesn't get deleted. It gets suppressed. Which is why in adults with frontal lobe damage from a stroke or dementia, the grasp reflex can reappear — neurologists call it a "frontal release sign" — the cortex's grip has weakened and the old spinal program surfaces again — Futagi, Toribe, and Suzuki documented this hierarchy in 2012. It's a strange thought: the same circuit that makes your newborn grip your finger is lying dormant in every adult brain, waiting. It just never gets a turn again in a healthy nervous system.

This connects directly to everything happening in your baby's brain development in the first year, and the rapid myelination that makes the cortex faster and stronger is precisely what gives it the hardware to take over from the brainstem programs.

The stepping reflex and the experiment that changed the textbook

Here's the thing about the stepping reflex: you hold a newborn upright with feet touching a surface and watch those little legs lift and step in sequence. It fades at around two months. And for decades, everyone assumed that was because the cortex matured and took over.

Then Esther Thelen ran an experiment at Indiana University in 1982 that made pediatric textbooks slightly embarrassing.

She noticed that babies who were gaining weight and body fat fastest were also the ones who stopped stepping soonest. She added small ankle weights to newborns — proportional to normal fat gain — and stepping slowed down. Then she put babies' legs in warm water, reducing the effective weight they had to lift. Stepping came back. Full stepping. Same brain, same age — just legs that weren't too heavy anymore.

The reflex wasn't disappearing because the brain was maturing. It was disappearing because the legs were getting too heavy for the spinal program to lift. The cortex had nothing to do with it.

What this means in practice: absence of a reflex doesn't always mean neurological absence. The stepping reflex will reappear as voluntary walking on its own schedule, somewhere between nine and fifteen months. The two are on completely separate developmental tracks. Context always matters more than a single observation — which is what your pediatrician is trained to read.

Where these reflexes came from, and what researchers think they were originally for

Every reflex your newborn has was present in some form long before humans existed as a species. Most of these reflexes emerge in utero, some as early as 14 weeks gestation — which tells you something about how fundamental they are. Researchers studying newborns across cultures and across primate species have documented striking parallels that point toward where these programs came from.

Rooting and sucking have the most obvious rationale and the least scientific debate. Touch a newborn's cheek and the head turns toward the stimulus, mouth opening. Place something in the mouth and sucking begins automatically, from 14 weeks gestation. There's no mystery here: a baby that can't find a nipple or sustain a latch doesn't eat. The program exists because it worked.

The palmar grasp is where it gets more interesting. Futagi and colleagues reviewed observations by Brain and Curran showing infant monkeys, some only a day or two old, clinging to their mothers' flanks solely with hands and feet while she jumped between branches — receiving no support at all. The infant human palmar and plantar grasp reflexes look structurally identical, and researchers have proposed they're remnants of a function once critical for survival in a very different environment. Futagi and colleagues made this case in 2012. This remains a hypothesis based on comparative observation, not a proven chain. It's a compelling one. The reflex appears at 28 weeks gestation and fades around six months, right when voluntary reaching begins to take over.

The Moro is the most debated. Ernst Moro, who first described the reflex in 1918, drew an orangutan to illustrate what he believed was a "catch" or "cling" response — the idea being that if an infant primate felt itself falling while clinging to a caregiver, this startle-and-reach pattern would help it grab on again. A 2017 study in Infant Behavior and Development went further, arguing the Moro functions as a ritualized "pick me up" signal — the arm-spread, the cry, the orienting toward the caregiver — drawing on both the startle system and the attachment system at once. Observations of infants in nomadic communities in Botswana, carried skin-to-skin against the mother's body, showed Moro responses ending with hands gripping the mother's breast or necklace. Researchers note, however, that the phylogenetic origin of the Moro is officially still unclear, as Futagi and colleagues noted in their 2012 review. These are the strongest hypotheses available, not settled conclusions.

The Galant reflex (stroking alongside the spine produces a lateral bend toward the stimulus) is present from 25 weeks gestation. Its proposed function is birth assistance — the flexion movement may help a baby navigate the birth canal — though this too remains theoretical rather than confirmed.

The complete timeline: what to watch for and what your pediatrician is tracking

Here's what parents are actually searching for: a usable, contextual timeline for newborn reflexes. This one is based on Zafeiriou's 2004 clinical review, StatPearls, and the foundational Prechtl-Beintema neurological examination protocol, which underlies every modern newborn assessment.

The key principle running through all of it: the real diagnostic signal is timing. Not just whether a reflex is present, but whether it's present when it should be, absent when it shouldn't be, or persisting when it should have faded. Your pediatrician is reading all three.

Rooting. Appears around 32 weeks gestation, fades by one to two months during alert feeding, may persist longer when the baby is drowsy. Worth mentioning if it's absent at birth or your newborn has consistent trouble orienting to feed, as rooting absence can reflect brainstem involvement.

Sucking. Present from 14 weeks gestation. A weak or absent sucking reflex at birth is one of the signals pediatricians pay attention to in a newborn neurological exam. If feeding is going well, there's nothing to worry about here; if latching has been consistently difficult past the first few days, it's worth mentioning — a pediatrician can figure out what's going on.

Moro. Develops by 28 weeks gestation, present in all healthy full-term newborns, integrates by four to six months. It's triggered by a sudden sensation of falling — the baby extends arms wide with spread fingers, brings them back in, often cries. An absent Moro at birth is the reflex finding pediatricians take most seriously. A Moro that's consistently stronger on one side (asymmetric) is worth mentioning at your next visit — there are several straightforward explanations your pediatrician can check, including things that happened during delivery. A Moro still firing reliably after six months is a reason to mention it at the next well visit.

Palmar grasp. Present at 28 weeks gestation, fades by four to six months as voluntary grasping takes over. If your baby's hand doesn't close around your finger at all in the first weeks, or if the grasp is only consistently present on one side, that's a conversation for your pediatrician. Research by Zafeiriou and by Garfinkle and colleagues found that when multiple reflexes are absent or abnormal at the same time, it's a meaningful pattern worth bringing to your pediatrician's attention — which is exactly why the newborn neurological exam checks several reflexes together, not just one.

Plantar grasp and Babinski. Stroke the sole of the foot and the toes curl inward in young babies. That's plantar grasp. Do the same thing to an older child or adult and the toes fan outward — the Babinski sign, which is abnormal in adults. Same stimulus, completely opposite clinical meaning, depending on whether the cortex is inhibiting the spinal response. In infants under 12 to 24 months, the Babinski-type response is expected and normal. In adults, it signals upper motor neuron damage. If you've ever Googled "Babinski reflex" while your baby was asleep and alarmed yourself, that's why — adult neurology context doesn't apply here.

ATNR (the fencing reflex). Appears around 35 weeks gestation, fades by three to four months. When the baby's head turns to one side, the arm on that side extends and the opposite arm flexes — like a fencer's pose. This reflex is considered a precursor to hand-eye coordination; as it integrates, your baby can bring both hands to the midline and start the bilateral work that reaching and grasping depend on. A 2020 study in Environmental Research and Public Health found that retained ATNR in young children measurably affects gait and pelvic symmetry, which is part of why persistence well past its window matters clinically. For more on how these early reflexes feed into later hand skills, our guide to fine motor development by age covers that transition.

Galant. Present from 25 weeks gestation, integrates by three to six months. Useful clinically for assessing neurological maturity in babies born prematurely — it's one of the earliest reflexes to appear in preterm infants.

Stepping. Present at birth, fades by two to three months (Thelen's work explains why: leg weight, not cortical maturation). Completely separate from when the baby will walk voluntarily.

Our High Contrast Flashcards for 0–3 months are built around the same principle: the newborn brain arrives with specific programs already running, and the right input at the right time meets the system where it actually is, rather than pushing ahead of it.

What researchers are finding about reflexes that persist

The research on primitive reflexes has gotten more interesting in recent years, and more frequently misrepresented in parenting spaces. It's worth knowing what the evidence actually says.

A 2021 review in Current Developmental Disorders Reports — Sigafoos and colleagues — summarized evidence linking persisting primitive reflexes to cerebral palsy, ADHD, and autism spectrum disorder. This is not a causal claim. Retained reflexes may be a marker of differences in cortical inhibitory development rather than a cause of those differences, which connects to what we know about how the brain builds itself in the first year. A 2023 meta-analysis in Frontiers in Psychiatry found that children with ADHD showed significantly higher rates of retained tonic neck reflex activity than typically developing peers. Again, an association in the data, not a proven mechanism.

You'll also encounter "reflex integration therapy" in parenting spaces, a category of interventions marketed for everything from reading difficulties to behavioral issues. The scientific picture is more cautious than the marketing. A 2000 randomized trial by McPhillips and colleagues, published in The Lancet, found meaningful reading improvements in children who performed movements replicating primitive reflexes over 12 months. The result was striking enough for The Lancet, but it was one small trial of 60 children, and the finding hasn't been replicated at scale. The honest framing: there's a hypothesis worth investigating, and commercial programs have outrun the evidence considerably.

If you're noticing something about your child's development, start with your pediatrician, not a program. Systematic assessment by someone who has examined your child is more useful than any checklist, including this one. The same principle from newborn care basics applies: when in doubt, ask the person who knows your baby.

When to mention something to your pediatrician, and what can wait

The clearest reasons to contact your pediatrician promptly, not at the next scheduled visit:

A missing Moro at birth in an otherwise healthy full-term newborn. This is the newborn reflex pediatricians check most carefully, and its absence is the most reliable indicator of significant CNS involvement.

A consistently one-sided reflex (especially Moro or grasp) where one side responds clearly and the other barely does. Asymmetry is more clinically significant than weak reflexes across both sides.

A plantar grasp absent in the first weeks, identified as an early marker for spastic cerebral palsy in case series reviewed by Futagi and colleagues.

Things worth mentioning at the next well visit, but not cause for an urgent call:

Reflexes that seem to linger slightly past the textbook window. A faint Moro at seven months in an otherwise typically developing baby is worth noting, but it sits very differently from a full Moro at seven months with other developmental concerns alongside it.

Reflexes that feel very strong or easily triggered. This varies between babies and isn't inherently significant, but it's a reasonable thing to bring up.

The first week with a newborn involves a lot of watching your baby and wondering if what you're seeing is normal. Most of the time it is — normal variation in a nervous system that's running its programming in real time. The reflexes are doing exactly what they're supposed to do.