Pediatrics · Part Two · The Newborn Nervous System

25FEEDING

Lesson 25 / 57

The Cranial Nerves of Feeding: What Each Nerve Actually Does

Five nerve pairs run a feed. Of 388 swallows imaged during breastfeeding, 379 matched a pause in breathing, and the average swallow lasted 0.63 seconds.

The cranial nerves of feeding are the trigeminal, facial, glossopharyngeal, vagus and hypoglossal. They carry sensation from the lips and pharynx, close the jaw, move the tongue and supply the larynx. A brainstem network sequences them into one repeating pattern, so a swallow and a breath alternate rather than overlap. The Unified Model of Tone reads a feed as one organized output rather than five separate skills.

Nerves that run a feed

V, VII, IX, X and XII

Mean duration of an infant swallow

0.63 seconds

Swallows matched by a pause in breathing

379 of 388

Air pulse needed to close the larynx

6.2 mm Hg with apnea against 4.3 mm Hg

What each nerve supplies

The trigeminal serves sensation from the lips, gums and tongue and drives the muscles that close the jaw. The facial nerve moves the lips and carries taste from the front of the tongue. The glossopharyngeal senses the pharynx. The vagus supplies the larynx and pharynx. The hypoglossal moves the tongue.

Where the timing is produced

Sensory fibers from the mouth and pharynx end in the nucleus of the solitary tract. Motor commands leave through the nucleus ambiguus and the hypoglossal nucleus. Reticular formation neurons connect the two and drive cranial motoneuron pools on both sides. The order of muscle activation therefore runs from the perioral muscles down to the cricopharyngeal sphincter without varying.

01Which nerve does what

Five nerve pairs divide the work of one feed

Five cranial nerve pairs do the work of a feed. The trigeminal, the fifth, carries sensation from the lips, gums and tongue. It also drives the muscles that close the jaw. The facial nerve, the seventh, moves the lips and carries taste from the front of the tongue.

The glossopharyngeal, the ninth, senses the pharynx. The vagus, the tenth, supplies the larynx and pharynx. The hypoglossal, the twelfth, moves the tongue. A review of swallowing control assigns the pharyngeal phase to the plexus formed by the ninth, tenth and eleventh nerves (Costa 2018).

Where they leave the skull

The hypoglossal nerve leaves the occipital bone through the hypoglossal canal. The glossopharyngeal, vagus and accessory nerves travel together at the jugular foramen, between the occipital and temporal bones. The jugular foramen and the vagus holds that anatomy in full.

The hypoglossal also joins fibers from the first and second cervical nerves to form the ansa cervicalis. From there a branch runs to the geniohyoid, which elevates the hyoid and larynx (Costa 2018). Control of the tongue is not purely cranial.

02Findings

What the research shows

The figures below come from two swallowing reviews, an ultrasound study of breastfeeding infants, a laryngeal sensation study and a cadaver series.

Five nerves, one pharyngeal phase
The pharyngeal phase begins through the pharyngeal plexus, formed by the glossopharyngeal, vagus and accessory nerves, with the trigeminal, facial and hypoglossal nerves involved (Costa 2018). Feeding is one circuit.
Pressure triggers the swallow, not touch
The stimulus that triggers the pharyngeal phase is pharyngeal pressure distension rather than the contact produced by the bolus passing (Costa 2018). How milk arrives matters.
The order never varies
Muscle activation runs from the perioral muscles down to the cricopharyngeal sphincter, and that sequence does not alter (Ertekin 2003). Invariance is the evidence for a central pattern generator.
Swallowing fails at the controller
Three-fourths of oropharyngeal dysphagia is caused by neurological diseases (Ertekin 2003). The throat is rarely the problem.
A swallow and a breath alternate
Of 388 swallows imaged by ultrasound, 379 matched a swallow apnea detected by plethysmography, and the mean swallow lasted 0.63 seconds (Geddes 2010). The two outputs take turns.
Sensation speeds the pattern up
Entrainment of trigeminal primary afferents accelerates pattern genesis for the suck generator and the transition to oral feeding in preterm infants (Barlow 2009). What the lips receive changes what the tongue does.
Laryngeal sensation varies between babies
Infants with apnea needed 6.2 mm Hg of air pulse to trigger the laryngeal adductor reflex against 4.3 mm Hg in comparison infants (Thompson 2005). Airway protection is not uniform.
A wider canal does not mean a larger nerve
In cadavers there was no apparent correlation between hypoglossal canal size and either the size of the nerve or the number of axons it contains (DeGusta 1999). The bone measurement is not the working variable. The living tissue around the nerve is.

03The brainstem program

The sequence is generated centrally and the mouth tunes it

The order in which feeding muscles fire is not assembled fresh at each feed. Activation runs from the perioral muscles caudally to the cricopharyngeal sphincter, and that sequence does not alter (Ertekin 2003). Invariance under changing sensory and cortical input is why a central pattern generator is inferred.

The network has an address. It includes the nucleus tractus solitarius and the nucleus ambiguus, with the reticular formation linking synaptically to cranial motoneuron pools on both sides. Descending input from the cortex can trigger a swallow and modulate the brainstem sequence.

The parts mature at different rates

The generators do not come online together. Relevant networks and their neuromuscular targets attain functional status at different rates, which shapes how the systems interact (Barlow 2009). Links between suck and swallow, and between swallow and respiration, pass through transitional forms from late gestation through the first year.

This is why a feed can look uncoordinated in a baby whose parts are all intact. The suck works, the swallow works, breathing works, and the joins between them settle last. Suck, swallow, breathe follows that timetable.

04The tongue nerve

The hypoglossal nerve reaches the tongue long before the tongue is used

The twelfth nerve is built early, and the embryology is dated precisely. Serial sections of 105 human embryos, from stage 11 at 24 days to stage 22 at 54 days, tracked its formation (O'Rahilly 1984). The hypoglossal nucleus is evident at stage 12 and becomes isolated from other efferent nuclei at stage 14.

The first nerve fibers appear at stage 12 as well. The roots unite at stage 14, and the main trunk arrives in the tongue at stage 15. Four occipital myotomes grow toward the tongue as the hypoglossal cord and arrive before the nerve does. The same occipital material forms the basioccipital and exoccipital chondrocranium.

One neighborhood, one system

Tongue muscle, tongue nerve and the bone the nerve passes through come from a single embryonic neighborhood. The material that forms the basioccipital and exoccipital chondrocranium is the material whose myotomes become the tongue. Head, neck and mouth are assembled together from the start.

A nerve is not measured by the width of the hole it travels through. That question was settled in cadavers, in work dating the origin of speech in fossil hominids. No apparent correlation appeared between hypoglossal canal size and the size of the nerve or the number of axons it contains (DeGusta 1999).

What the Unified Model of Tone reads at this junction is living tissue instead. The variable is the tension held by the muscles and membranes around the nerve, and the position information those tissues send into the brainstem that sequences a feed.

05Airway protection

The vagus closes the airway for every swallow, and it does so unevenly

Milk and air use the same corridor, so every swallow requires the larynx to shut. Ultrasound imaging of breastfeeding infants makes the alternation measurable. Of 388 swallows detected, 379 correlated with a swallow apnea recorded by respiratory inductive plethysmography, and the mean swallow lasted 0.63 seconds (Geddes 2010).

How readily that closure fires is not fixed. The laryngeal adductor reflex was induced in 20 infants with apnea by 50 millisecond air pulses to the aryepiglottic fold. Those infants needed 6.2 mm Hg against 4.3 mm Hg in comparison infants, and they cleared secretions more poorly (Thompson 2005). The difference held after adjustment for postconceptional age.

The other half of the tenth nerve

The vagus at the throat is a laryngeal and pharyngeal nerve. It supplies the muscles that close and open the airway, and it carries the sensation reporting what sits in the pharynx. The same trunk carries the parasympathetic traffic that settles heart rate and gut, which is one reason an easy feed and a settled baby so often arrive together. The vagus and the calm holds that half in full.

Coughing, choking, color change or wet and gurgling breathing during feeds are signs that airway protection is not keeping up. Those belong with a pediatrician promptly.

06When a feeding nerve is injured

Newborn facial palsy usually resolves, and two cohorts disagree on how common it is

The seventh nerve is the one most often injured at birth, and two cohorts disagree on how often. Among 44,292 infants born over five years at one hospital, 81 acquired cases of facial palsy were recorded, an incidence of 1.8 per 1,000 (Falco 1990). Forceps were involved in 74 of the 81, while forceps were used in 19 percent of all deliveries.

A later cohort disagreed. Among 83,067 infants delivered over 12 years, 29 were diagnosed with traumatic facial palsy, an incidence of 0.03 percent (Al Tawil 2010). In that series 22 of the 29 cases, or 75.9 percent, arrived without any forceps application.

What the disagreement means

The two incidence figures differ roughly fivefold, and the forceps association reverses. Both stand here because a family reading one alone would be misled. Outcome is where they agree. Recovery was complete in 89 percent of the first series, and 93 percent of the second recovered spontaneously within two months.

The home observation is asymmetry. A mouth pulling to one side when crying, an eye that does not close fully, or a lip seal leaking on one side is worth showing a pediatrician.

07Sense before movement

The mouth organizes around what it feels, under a skull that is still settling

Feeding begins with sensation rather than with strength. Before a drop of milk moves, the trigeminal and facial nerves register where the breast or bottle meets the lips and tongue. The mouth then organizes around what it feels. A latch is the tongue, jaw and lips arranging themselves around a moving target, and the order of operations is always sense first, then move.

The measurements follow that order. Entrainment of trigeminal primary afferents accelerates pattern genesis for the suck generator and the transition to oral feeding in preterm infants (Barlow 2009). A brainstem cannot build a clean motor pattern out of muddy input, so rich sensation from the lips and tongue is what gives the twelfth nerve accurate information to shape the tongue against the palate. The mouth is a sensory map before it is a milk pump.

A skull that is still settling

A newborn skull is a set of separate plates joined by sutures and fontanelles rather than one fused vault. Those plates move during birth and settle across the weeks that follow. The nerves running the tongue and the swallow leave through openings in bones still finding their resting relationship. The muscles of the jaw, the floor of the mouth and the upper neck all pull on that same arrangement.

The Unified Model of Tone reads the whole of it as one tensioned system rather than a set of separate passages. On that reading, tension held at the occiput and the upper cervical tissues is information arriving at the brainstem that has to sequence five nerves. The model expects a feed to be among the first behaviors to show it. That is the model’s own claim, stated as ours.

What the work involves

The clinician watches a feed first, then palpates the floor of the mouth, the jaw, the occiput, the upper cervical tissues and the cranial base, feeling where one side resists and where the tissue gives. The contact that follows is a sustained, light hold, no more than the pressure a person could rest on a closed eyelid, graded to the age and size of the child. There is no twist, no thrust and no audible release, and a settled baby commonly stays asleep through it.

The findings families bring are specific. Clicking at the latch, milk escaping at one corner of the mouth, a jaw that opens further one way than the other, a head that turns more easily to one side. The model reads those as cranial nerve findings, which is why the examination goes after the whole feeding circuit rather than the lip.

A systematic review of manual therapy in children screened 3,563 articles and included 50 studies, 18 of them judged high quality (Parnell Prevost 2019). Adverse events across the 20 studies reporting them were transient and mild to moderate. Laryngeal sensory threshold, swallow timing and cervical rotation range have never been recorded together in one infant cohort. The model expects them to share one organization, compensation decides which reading moves first, and the feeding team sets out who reads which part today.

08The model’s claim

What the Unified Model of Tone predicts about the feeding nerves

The measurements above come from the swallowing and developmental literature. What follows is this model’s reading of them, stated as ours rather than drawn from the papers cited.

The model separates the site where trouble shows from the source where it starts. A disturbance surfaces at the site holding the least compensatory reserve, so "the loudest tissue is the weakest link rather than the origin." In a struggling feed, the mouth is the loud tissue. Five nerves, one brainstem network and the respiratory rhythm are visible in a single behavior, and the order of muscle activation is fixed (Ertekin 2003). What varies between babies is the state of the network running the sequence, which is why a latch that fails at the lips can be reporting a strain that lives elsewhere.

The prediction

From that follows a claim the swallowing literature does not make. Feeding coordination is not a local skill of the mouth, so the mouth’s difficulty should have measurable company across the network. The model predicts that a baby whose suck and swallow join poorly is more likely to differ on laryngeal sensory threshold, on head and neck symmetry, and on autonomic regulation, while compensation can hold any one of those readings quiet. Those are reflex responsiveness, postural organization and variability structure wearing feeding names, readings of one variable rather than four.

That is testable, and separate from any claim about treatment. 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 swallow timing, laryngeal sensory threshold, postural symmetry and autonomic regulation are shown to move together, the unification claim is confirmed.

09The tone reading

How the feeding nerves express tone

Every topic in this library expresses all of tone. In the feeding nerves three aspects carry the signature, because five cables and one rhythm are visible in a single behavior.

Coupling

A swallow and a breath cannot share a moment. In 388 imaged swallows, 379 matched a pause in breathing, so the two alternate.

Input quality

Sensation leads. Entrainment of trigeminal afferents accelerates pattern genesis for the suck generator, so what the lips receive changes what the tongue does.

Time course

The hypoglossal nucleus appears at stage 12 and the nerve reaches the tongue at stage 15, many weeks before the tongue is used to feed.

The remaining foundations run through this topic as well. Constraint: one corridor carries both milk and air, which forces the timing. Gain: the air pulse needed to close the larynx differed by nearly 2 mm Hg between groups. Set point: a drowsy baby and an alert one bring different resting jaw posture to the same breast. Prediction: a mature feeder places the tongue for the next bolus rather than the last. Load: fast flow, a blocked nose or a tired baby changes what the circuit delivers. Oscillation: suck, swallow and breathe are three rhythms sharing one clock. These are readings of one organization rather than separate systems, which is the core claim of the Unified Model of Tone.

10Across the library

How this page relates to the rest of the library

The feeding nerves sit between the newborn examination, the airway and the breast.

Suck, Swallow, Breathe

The three rhythms these nerves carry, and the timetable on which the joins settle.

Tongue and Lip Ties

What the twelfth nerve moves, and what happens when that structure is restricted.

The Jugular Foramen and the Vagus

The opening the ninth, tenth and eleventh nerves share, in full detail.

Rooting and Sucking

The sensory search that precedes a latch, and what an examination reads in it.

A Brainstem-Run Newborn

The maturation schedule of the network that sequences these five nerves.

The Vagus and Calm

The autonomic half of the tenth nerve, where the heart rate figures live.

Coupling

How two outputs come to share one clock, as a measurable state in adults.

11Frequently asked

Questions families ask about the cranial nerves of feeding

Which cranial nerves does a baby use to feed?

Five pairs. The trigeminal carries sensation from the lips, gums and tongue and closes the jaw. The facial nerve moves the lips and carries taste from the front of the tongue. The glossopharyngeal senses the pharynx. The vagus supplies the larynx and pharynx. The hypoglossal moves the tongue. A review of swallowing control assigns the pharyngeal phase to the plexus formed by the ninth, tenth and eleventh nerves, with the fifth, seventh and twelfth all involved in the act.

What actually triggers the swallow?

Pressure rather than touch. The stimulus that triggers the pharyngeal phase is pharyngeal pressure distension, with or without contents, rather than the pharyngeal contact produced by a bolus passing through. That distinction matters at the breast and the bottle, because it means the swallow answers how milk is delivered into the pharynx and not merely that milk has touched it. Flow rate and position change the stimulus the circuit receives. The glossopharyngeal nerve carries that pharyngeal sensation, and the brainstem network decides when to fire.

Can my baby breathe and swallow at the same time?

No, and the alternation has been measured. Ultrasound imaging of breastfeeding infants detected 388 swallows, and 379 of them correlated with a pause in breathing recorded by plethysmography. The two methods agreed closely, with an R squared of 0.98, and the mean swallow lasted 0.63 seconds. Milk and air share one corridor, so the larynx closes while a bolus passes and breathing resumes after it. A feed is therefore a sequence of very short breath holds strung between breaths.

Is the swallow reflex fully reliable in a newborn?

It is protective and it varies, which is different from reliable. The laryngeal adductor reflex was induced in 20 infants with apnea using brief air pulses to the aryepiglottic fold. Those infants needed 6.2 mm Hg to trigger it, against 4.3 mm Hg in a comparison group, and they cleared secretions more poorly. The comparison infants were being evaluated for upper airway anomalies rather than healthy controls. The difference held after adjusting for postconceptional age. Coughing, choking or color change during feeds needs prompt medical review.

My baby’s face looks uneven when crying. What is that?

It may be a facial nerve palsy, which is the birth nerve injury seen most often. One cohort of 44,292 infants found 81 acquired cases, an incidence of 1.8 per 1,000, with forceps involved in 91 percent. A later cohort of 83,067 infants found 29 cases, and 75.9 percent of those arrived without forceps. Forceps, a birth weight of 3,500 grams or more and a first pregnancy were risk factors. Recovery was complete in 89 percent of one series and 93 percent of the other. Show it to your pediatrician.

My baby clicks at the latch and loses milk at one corner. What is that?

Those are cranial nerve findings. A lip seal is held by the seventh nerve, the tongue is shaped by the twelfth, and the swallow is timed by the ninth and tenth against a breath. A click is usually suction breaking, and milk escaping at one corner points to an uneven seal. The model reads both as reports from the circuit rather than as habits, so the examination covers jaw, tongue, lip and neck together rather than the lip alone.

What does a visit for a feeding problem involve?

The feed is watched first. Then the clinician palpates the floor of the mouth, the jaw, the occiput, the upper cervical tissues and the cranial base, feeling where one side resists and where the tissue gives. The contact is a sustained light hold, no heavier than the pressure you could rest on a closed eyelid, graded to your baby’s size. There is no twist and no thrust, and settled babies commonly sleep through it. Adverse events reported in the pediatric manual therapy literature were transient and mild to moderate.

12The sources

References

1
Costa MMB. Neural control of swallowing. Arq Gastroenterol. 2018. PMID 30156597
2
Ertekin C, Aydogdu I. Neurophysiology of swallowing. Clin Neurophysiol. 2003. PMID 14652082
3
Barlow SM. Central pattern generation involved in oral and respiratory control for feeding in the term infant. Curr Opin Otolaryngol Head Neck Surg. 2009. PMID 19417662
4
Geddes DT, Chadwick LM, Kent JC, Garbin CP, et al. Ultrasound imaging of infant swallowing during breast-feeding. Dysphagia. 2010. PMID 19626366
5
Thompson DM, Rutter MJ, Rudolph CD, Willging JP, et al. Altered laryngeal sensation: a potential cause of apnea of infancy. Ann Otol Rhinol Laryngol. 2005. PMID 15895779
6
O'Rahilly R, Muller F. The early development of the hypoglossal nerve and occipital somites in staged human embryos. Am J Anat. 1984. PMID 6720613
7
DeGusta D, Gilbert WH, Turner SP. Hypoglossal canal size and hominid speech. Proc Natl Acad Sci U S A. 1999. PMID 9990105
8
Falco NA, Eriksson E. Facial nerve palsy in the newborn: incidence and outcome. Plast Reconstr Surg. 1990. PMID 2293714
9
Al Tawil K, Saleem N, Kadri H, Rifae MT, et al. Traumatic facial nerve palsy in newborns: is it always iatrogenic?. Am J Perinatol. 2010. PMID 20387190
10
Parnell Prevost C, Gleberzon B, Carleo B, Anderson K, et al. Manual therapy for the pediatric population: a systematic review. BMC Complement Altern Med. 2019. PMID 30866915

10 primary sources, each linked to its PubMed record. Figures quoted on this page were checked against the published abstract.

Related evidence

← All 57 lessons