Pediatrics · Part Three · How a Child Develops
Lesson 34 / 57
Tummy Time and the Curve: What Antigravity Work Builds in a Baby’s Neck
Positional flattening was found in 46.6 percent of healthy infants at 7 to 12 weeks. Time on the belly is the counterweight, and it is the half of the guidance most often missed.
Tummy time is supervised waking play in the prone position, and it is where a baby does the antigravity work that builds head control. Positional plagiocephaly was found in 46.6 percent of 440 healthy infants at 7 to 12 weeks of age, most of it mild. Infants who spend more time prone reach several motor milestones earlier, though all stay within the normal range. The Unified Model of Tone reads prone play as loading the densest movement sensors a child owns.
Positional plagiocephaly at 7 to 12 weeks
46.6 percent of infants
Of those, mild in form
78.3 percent
Right-sided flattening
63.2 percent
Spindle density, inferior oblique
242 per gram
Tummy time
Tummy time is supervised time a baby spends awake on the stomach. It is the position in which a child lifts the head against gravity, props on the forearms, and turns from one side to the other. It sits alongside back sleeping rather than in tension with it, because the two answer different needs.
Positional plagiocephaly
Positional plagiocephaly is flattening of one part of an infant skull caused by sustained pressure on that spot. Infant skull bones are soft and mobile, so steady pressure in one place reshapes them. It is common, usually mild, and it responds to varying the positions a baby spends time in.
01Two positions, one baby
Back for sleep and belly for the antigravity work
Back sleeping and tummy time answer different needs and do not compete. A newborn spine arrives as a single long curve carried from the womb, and the cervical lordosis that lets a head balance over the trunk is not present at birth. It develops as a child begins lifting and holding the head, which happens in prone.
The mechanics are straightforward. Lying face down and lifting the head loads the deep neck extensors and the suboccipital muscles, drawing the cervical spine toward extension. Repeated across weeks, that effort is what shapes the curve.
Why both, and why neither is optional
Babies sleep on the back. Nothing in the evidence below touches that guidance. What the same literature notes is that supervised prone play time is the half of the advice most commonly overlooked by families (Liao 2005). Sleep and waking positions are different decisions, and the antigravity work belongs in the waking half.
Time off the floor
The counterweight to prone play is how much of the day a baby spends held in one shape. Car seats, bouncers, swings and carriers all support the spine in flexion, and the neck rarely works against full gravity in any of them. Car seats save lives and belong in cars. The point is balance across the waking hours.
An infant skull is soft and its bones grow at the open sutures, so steady pressure on one spot flattens it while varied position keeps a head rounding evenly. Floor time, belly time, side lying and being carried different ways each give the head and neck a fresh angle to work at.
02Findings
What the research shows
The figures below come from population studies of healthy infants rather than from clinical samples.
03Flattening is common
Positional plagiocephaly affects nearly half of healthy infants
Head flattening at this age is close to a coin flip, and knowing that changes how a family hears the diagnosis. Mawji and colleagues assessed 440 healthy full-term infants at 7 to 12 weeks in a prospective community cohort. Some form of positional plagiocephaly was present in 205 of them, an incidence of 46.6 percent (Mawji 2013).
The severity distribution matters as much as the headline. In that cohort 78.3 percent of affected infants had a mild form, and 63.2 percent of flattening was right-sided.
What the right-sided skew suggests
Pressure alone would not produce a strong side bias. A consistent preference for turning one way concentrates pressure on one side, so the asymmetry of head turning and the asymmetry of the skull tend to travel together. That is why varying position across the day, and looking at how freely a head turns both ways, are the two responses that go together.
04Position changes the sequence
Time spent prone measurably shifts motor development
Where a baby spends time shows up in the milestone data. Davis and colleagues followed 351 healthy term infants and found prone sleepers reaching rolling prone to supine, tripod sitting, creeping, crawling and pulling to stand earlier than supine sleepers (Davis 1998). Walking age did not differ.
The framing in that paper is careful and worth repeating exactly. Every infant achieved every milestone within the accepted normal age range, and the authors state that this difference is not a reason to abandon the safe sleep recommendations. Reviews since have described the differences as transient (Liao 2005).
What that means for a parent
Positioning is a real input with a measurable effect, and it is also not an emergency. The finding argues for supervised waking time on the stomach as a routine part of the day rather than for anxiety about a schedule. The WHO windows are months wide in any case (WHO 2006), and the developmental sequence covers how to read them.
05What the neck is sending
Prone play loads the densest movement sensors in the body
The muscles doing the lifting are the most densely instrumented in the body. In human fetal suboccipital muscles, spindle density reaches 242 per gram in the inferior oblique, 190 in the superior oblique and 98 in rectus capitis posterior, with no tendon organs found (Kulkarni 2001). Spindles report length and stretch speed. Tendon organs report force. Muscles built this way exist to report position.
That stream feeds the brainstem and cerebellum, and the cerebellum is growing 240 percent across the same year (Knickmeyer 2008). The balance apparatus is also ready early, with semicircular canal envelopes ossifying on schedules tied to the onset of vestibular function before birth (Richard 2017).
Three streams arrive at once when a head comes up. The vestibular apparatus reports the new orientation, the deep neck proprioceptors report joint position, and the visual system begins to hold a gaze steady enough to scan a room. They converge in the brainstem and the cerebellum, and that is where postural tone is set and refined.
The reflexes that scaffold the lift
The symmetrical tonic neck reflex links neck extension to arm extension, which is part of how an infant gets up onto hands and knees. It is expected to quiet as that push becomes voluntary. Earlier patterns such as the tonic labyrinthine and neck righting reflexes shape how the head and trunk move together. Zafeiriou reviews these alongside the postural reactions as standard instruments for reading central nervous system integrity (Zafeiriou 2004).
Why active lifting is the point
A baby held upright by a device receives a very different signal from a baby lifting a head against gravity. Four-month-olds given active practice showed changed brain responses where infants who only watched showed none (Bakker 2016). The value is in the effort, not the position, which is why fussing during tummy time usually means work rather than harm.
06When to look closer
A fixed head preference is worth examining rather than waiting out
A head that consistently turns one way deserves a look. In a review of 2,047 children presenting with torticollis, congenital muscular torticollis accounted for 76.6 percent, most commonly between birth and two years, with cerebral palsy and ocular causes next (Jianqiang 2024). The differential is genuinely broad, which is the argument for examining rather than waiting.
The examination itself is ordinary. Neck rotation to both sides, symmetry of head turning in prone, head lag on gentle pull to sit, and the primitive reflexes and postural reactions read against age (Zafeiriou 2004). Head shape and torticollis covers the full picture.
The reading extends past the neck. The quality of a cry reports on vagal and glossopharyngeal function, eye movements test the third, fourth and sixth cranial nerves, and feeding draws on the fifth, seventh, ninth, tenth and twelfth. The cranial nerves of feeding works through that part of the examination.
What earns attention is a persistent one-sided preference, a head that resists turning back, or a marked head lag rather than a baby who simply dislikes the floor at first.
What the contact is like
The work that follows an examination is quiet. A clinician holds a fingertip against the upper neck, the cranial base or the pelvis for a few seconds, graded to the age and size of the child. There is no twist and no thrust, and a settled baby commonly stays asleep through it.
The aim is a head that turns evenly to both sides so the antigravity work can proceed. The model reads a held pattern of tension at the top of the neck as a standing input to the same brainstem the lift reports into. That is why freedom of rotation is the variable of interest rather than strength. Families most often arrive when one side is favored, when the belly brings tears, or when a flat spot appears.
07The model’s claim
What the Unified Model of Tone predicts about prone play
Everything above is established pediatric science. What follows is this model’s reading of it, stated as ours rather than drawn from the papers cited.
Tone is the integrated organization of the body’s interacting processes, and a body learns that organization by moving. The body cannot form an accurate model of where it is by holding still. Each lift of the head in prone sends a fresh afferent picture to be read against the brain’s predictions. And because the sensors sit inside the tissue they report on, the act of sampling rewrites the state it samples. A baby lifting a head is reading its own organization and revising it in the same motion. The model therefore reads prone play as loading the highest-density position sensors a child owns at the moment the structure that reads them is growing 240 percent. Tummy time is an information event before it is an exercise.
The prediction
From that the model predicts that the value of prone time tracks the quality of the movement it produces rather than the minutes logged. A baby who turns freely to both sides in prone is sampling symmetrically. A baby who holds one rotation is sampling a restricted set, and the model predicts the second child organizes differently even with identical time on the floor.
That prediction is measurable through the asymmetries already recorded in this literature: the 63.2 percent right-sided skew in positional flattening, head-turning range, and the symmetry of postural reactions. It also explains why the model treats freedom of head rotation as the variable of interest rather than duration. In the model’s terms these are the infant’s first readings of its canonical four, variability structure, cross-frequency coupling, reflex responsiveness, and recovery time.
Chiropractic care enters as one input aimed at that freedom of rotation, sustained and light and graded to the child, and what an adjustment is describes the method in full. This is a claim about how development is organized rather than about what treatment does. It holds that better-organized tone yields greater adaptive capacity, whichever appropriate input delivered the useful information. If head-turning range, postural symmetry, sensory reweighting and autonomic regulation are shown to move together, the unification claim is confirmed.
08The tone reading
How tummy time expresses tone
Every topic in this library expresses all of tone. In prone play three aspects carry the signature, because a baby lifting a head against gravity is loading sensors that report at 242 per gram.
Input quality
Active lifting and passive propping send different signals. What the nervous system builds depends on the fidelity of what the effort generates.
Constraint
A head that holds one rotation samples a restricted set. Flattening was right-sided in 63.2 percent, and asymmetry travels with asymmetry.
Load
Antigravity work against a head that is large for its neck is a genuine mechanical demand on a very small body.
The remaining foundations run through prone play as well. Set point: postural baseline against gravity is first established in this position. Prediction: each lift is compared against what the last one did. Gain: how heavily neck proprioception is weighted against vision is being tuned here. Coupling: head control links vestibular, visual and proprioceptive channels into one estimate. Time course: the same minutes matter more while the cerebellum is adding 240 percent. Oscillation: sleep and feeding rhythms decide when a baby has the reserve to do this work. These are readings of one organization rather than separate systems, which is the core claim of the Unified Model of Tone.
09Across the library
How this page relates to the rest of the library
Prone play touches head shape, milestones and the sensors that make it all work.
Where head control sits in the ladder, and how wide the real milestone windows are.
Why self-generated effort is the input, and what 2,368 steps an hour later looks like.
The full picture on flattening and fixed head preference, including the differential worth ruling out.
Where the neck signal goes, and why the structure reading it grows faster than anything else in year one.
The other side of positioning. What time in equipment removes from a developing nervous system.
The receptor itself, and how it encodes muscle length and the speed of stretch.
The foundation this page leans on hardest. Why fidelity changes what a system builds from a signal.
10Frequently asked
Questions families ask about tummy time
How much tummy time does my baby actually need?
Enough that it is a routine part of waking hours, built up gradually rather than forced. Start in the early weeks, even a few minutes on a parent chest. The literature supports the practice without prescribing a stopwatch figure, and short frequent sessions suit a small body better than one long one. Position does register in the data. Among 351 healthy term infants, prone sleepers reached five motor milestones earlier than supine sleepers. Every infant stayed within the normal range, and walking age did not differ at all.
Is it safe to put my baby to sleep on the stomach instead?
No. Babies sleep on the back, without qualification. The study showing earlier milestones in prone sleepers states directly that the finding is not a reason to abandon safe sleep recommendations, and every infant in it reached every milestone inside the normal range anyway. Sleep position and waking position are separate decisions, and treating them as one is the mistake worth avoiding. The antigravity work belongs entirely in supervised awake time on the floor, where an adult is watching.
My baby cries the whole time. Am I doing harm?
Almost certainly not. Lifting a head that is large relative to the neck is demanding work, and effort is the point of the position rather than a side effect. Build tolerance gradually and use short frequent sessions. What would change the picture is different from crying. Watch instead for a strong fixed preference for turning one way, a head that resists turning back, or a marked head lag when the baby is drawn gently to sitting.
How common is a flat spot, and should I worry?
Common, and usually mild. In a prospective cohort of 440 healthy full-term infants assessed at 7 to 12 weeks, 46.6 percent had some form of positional plagiocephaly, and 78.3 percent of those cases were mild. Infant skull bones are soft and mobile, so sustained pressure on one spot reshapes them and varied positioning is the usual response. Nearly half of healthy babies showing some flattening at this age puts a single flat spot in perspective, and most of those resolve as position varies.
Why do flat spots so often appear on the same side?
Because pressure follows preference. In that cohort 63.2 percent of flattening was right-sided, and pressure alone would not produce a bias that strong. A baby who consistently turns one way concentrates pressure on one side of the skull, so asymmetric head turning and asymmetric head shape tend to travel together. That is why how freely the head turns both ways matters as much as which surfaces the baby lies on, and why turning range is worth checking directly rather than assuming.
When should a fixed head preference be examined?
When it persists rather than passes. A review of 2,047 children presenting with torticollis found congenital muscular torticollis accounting for 76.6 percent, most commonly between birth and two years, with cerebral palsy and ocular causes next. The differential is broad enough that examining beats waiting. The assessment is ordinary: neck rotation both ways, symmetry in prone, head lag on pull to sit, and reflexes read against age. None of it is invasive, and it settles the question early.
What does the Unified Model of Tone say about tummy time?
That it is an information event before it is an exercise. The muscles lifting the head carry up to 242 muscle spindles per gram and no force-reporting tendon organs, so they exist to report position into a cerebellum that is growing 240 percent that year. From this the model predicts something practical: the value of prone time tracks the quality and symmetry of the movement it produces rather than the minutes logged on a timer. That is a claim about input quality, stated as ours, and it is measurable.
11The sources
References
10 primary sources, each linked to its PubMed record. Figures quoted on this page were checked against the published abstract.
Related evidence