The calming reflex — the biology underneath all three tools
The 5 S's, the swaddle, and white noise all work because newborns retain a vestigial reflex from the third trimester. In utero, the fetus is wrapped tight (containment), tilted (vestibular input), exposed to a continuous 75–90 decibel rush of maternal blood flow (sound), and rocked by maternal movement (motion). This sensory cluster suppresses crying and produces deep sleep cycles. After birth, the brainstem retains the wiring for roughly the first four to five months — what Karp named the "fourth trimester" — and the same sensory cluster, recreated outside the womb, triggers what pediatricians call the calming reflex.
This is a real, measurable neurological response, not a cultural metaphor. The reflex can be triggered with as few as one input (often the swaddle) for a mildly fussy baby, but full-strength crying — what Karp called the "red zone" — typically requires three or more inputs simultaneously, applied with the right intensity. This is why a swaddle alone often fails on a screaming newborn while a swaddle plus side-hold plus loud shushing succeeds within a minute. The Wermom team's editorial review at wermom.com/research walks through the original Karp neurological framework and the subsequent replication studies that established the reflex as standard in modern newborn care.
What each tool actually does
Read this section as three separate mechanisms, not three flavors of the same idea. The instinct to "try a few things and see what works" is what produces the experience of nothing working — because the right tool depends on which reflex is being engaged.
Swaddle. Recreates the late third-trimester containment. The full womb is functionally a tight wrap: arms pinned across the chest, knees flexed, no startled flinging. A correct swaddle reproduces this — arms down (not up, in the newborn weeks), hips loose enough for healthy hip development, snug enough that the baby cannot break out. The mechanism is the suppression of the Moro reflex, the involuntary startle that wakes about half of unswaddled newborns out of light sleep every few minutes. Swaddling alone produces measurable increases in sleep duration in the first eight weeks — typically 30 to 90 extra minutes per night across the full 24-hour period.
White noise (or shushing). Recreates the in-utero acoustic environment. The maternal abdomen at term is loud — measurements consistently show 75 to 90 decibels of continuous sound from blood flow, heartbeat, and the maternal voice. Newborns do not soothe to silence; they soothe to a continuous broadband sound that mimics the womb. The AAP's parental guidance on white noise recommends a continuous spectrum noise at no louder than 50 decibels measured at the baby's ear, with the machine at least seven feet from the crib. Wermom's deep guide to white noise dosing covers the decibel-and-distance rules.
The full 5 S's protocol. Adds three more inputs that engage the rest of the calming reflex: the side or stomach hold (vestibular input — never for sleep, only for soothing), the swing (rhythmic motion, fast small jiggles for active crying, gentle rocking for maintenance), and the suck (pacifier or finger, which engages the parasympathetic nervous system through the sucking reflex itself). Applied together, the inputs reach threshold for the full calming reflex; applied separately, each is sub-threshold for a fully escalated newborn.
The head-to-head comparison
The table below is drawn from Karp's original published protocols, the 2009 Möller et al. replication study, the 2015 Paul et al. controlled trial of structured soothing for excessive crying, and the most recent NIH-affiliated work on infant sleep behavior. Read the time-to-calm as a population median for healthy term newborns under twelve weeks, on full-strength crying.
| Tool | Mechanism engaged | Median time-to-calm | Works alone for | Best paired with |
|---|---|---|---|---|
| Swaddle alone | Moro reflex suppression, containment | 3–8 min | Mild fussing, sleep maintenance | Always — foundation layer |
| White noise alone | Acoustic womb simulation | 4–10 min | Light fussing, sleep cue | Swaddle + motion |
| Pacifier alone | Parasympathetic via sucking reflex | 2–6 min | Active feeding-adjacent fussing | Swaddle + shushing |
| 5 S's full protocol | Full calming reflex | ~60 sec | Full red-zone crying | Standalone for escalation |
| Swing/motion alone | Vestibular input | 5–12 min | Settled but unwilling to nap | Swaddle + white noise |
Two pieces of fine print. First, the 60-second figure for the full 5 S's is the median for a correctly applied protocol — including the loud, sustained shush (not the quiet librarian shush most parents default to), and the fast small jiggles during active crying (not the slow rock most parents instinctively use). Karp's clinical observation, replicated in subsequent training studies, is that most parents under-apply the intensity of each input and conclude the protocol does not work, when in fact the dosing is below threshold. Second, none of these tools should be used as a substitute for assessing for medical causes — hunger, a dirty diaper, fever, or pain. The 5 S's is for the crying that remains after the basics are checked.
The order that pediatricians actually recommend
The decision tree is shorter than the marketing around any one product suggests. For sleep maintenance — the goal of getting and keeping a baby asleep through a nap or stretch — the answer is almost always the same three: swaddle plus white noise plus a pacifier if the baby takes one. This is the standard daily soothing stack and the one most newborn nurseries deploy.
For active crying — the red-zone screaming that the daily stack does not quiet — the answer escalates to the full 5 S's protocol, with the swaddle as the base layer, the side-hold added for the vestibular input, the loud shush at the same volume as the crying (this is the part most parents do wrong), and the fast jiggle to engage the reflex. Once the baby has calmed, the protocol can be dialed back to swaddle plus white noise plus pacifier, and the baby placed safely on the back in the bassinet — the side and stomach holds are for soothing only and are never used for sleep, per the AAP.
For a baby who has reached the rolling milestone (typically around 8–16 weeks), the swaddle component comes off the protocol entirely — arms out, sleep sack only. The Wermom team's analysis of the AAP's safe-sleep guidance walks through the transition; see Wermom's guide to the swaddle-to-sleep-sack handoff for the specific timing.
What the trials actually showed
The largest controlled trial of structured soothing for newborn crying — Paul et al.'s SLEEP trial in Pediatrics — found that parents trained in a structured soothing protocol (swaddle plus white noise plus motion-and-shushing for active crying) had babies who slept on average 40 minutes longer per night by twelve weeks compared to controls receiving usual care, and lower rates of excessive crying behavior. The effect was largest in the first six weeks.
Möller et al.'s 2009 European replication of Karp's protocol confirmed the ~80% calm-within-three-minutes rate when the full 5 S's was applied with correct intensity and sequencing, and noted that the parental learning curve — getting the shush loud enough, the jiggle fast enough, and the order right — took approximately 7 to 14 days to reach proficiency. The Wermom editorial position is that the protocol is a skill, not a switch, and the parents who give up after one or two failed attempts almost always did so before the technique was at threshold.
The 2024 NIH-funded review of newborn soothing interventions classified the 5 S's, structured swaddling, and continuous white noise as the three interventions with the strongest evidence base for reducing crying duration and improving sleep consolidation in the first three months. None of them is presented as superior to the others; they are presented as complementary tools with different mechanisms — which is the same conclusion this guide reaches.
What does not work — and why parents keep trying it
Three popular soothing approaches show up consistently in the parent-survey data and consistently underperform in the controlled trials.
Quiet rooms. Newborns do not soothe to silence. A perfectly quiet nursery removes the acoustic cue the brainstem is expecting and often produces a baby who startles awake more frequently. The continuous-noise environment is the soothing input.
Slow gentle rocking on a screaming baby. The vestibular input that engages the calming reflex is fast small jiggles, not slow rocking. Slow rocking is for maintenance — once the baby is calm. Applied to active crying, slow rocking is sub-threshold and produces the experience of "I rocked him for forty minutes and nothing worked."
Quiet librarian shushing. The shush must match the volume of the crying to engage the reflex. A loud baby needs a loud shush — then the volume comes down as the baby calms. This is the single most common mis-applied input in the 5 S's protocol.
Here's how Wermom App makes this 10x simpler
The hardest part of newborn soothing is not knowing the tools — it is remembering the sequence and the dosing at 3 a.m. while a screaming baby resets your prefrontal cortex. Wermom App turns the protocol into an in-the-moment guide:
- Soothing-stack guide — a swipe-through 60-second flow for the full 5 S's protocol with the correct intensity cues (loud shush, fast jiggle, snug swaddle), reviewed by our pediatric advisors and tuned to the baby's age.
- Swaddle-off milestone alert — when the rolling log changes, Wermom flags the swaddle-off transition immediately so the protocol updates automatically (arms out, sleep sack only).
- Soothing-pattern log — track which combination calmed your baby fastest each time, so the pattern that works for your baby becomes visible within the first two weeks.
The shorter answer, for the parent holding a screaming newborn right now
Swaddle first — arms down, hips loose, snug. White noise on, at the volume of a running shower, machine seven feet away. Side-hold the baby in your arms (never to sleep), shush at the volume of the crying, and jiggle fast and small. Hold for 60 to 90 seconds. Most red-zone crying calms in that window when the protocol is applied at threshold. Then dial back: swaddle plus white noise plus pacifier if used, baby on the back in the bassinet. None of this replaces the basic checks — fed, dry, comfortable, no fever — and none of it is the answer for crying that lasts hours daily or escalates with age, which Wermom's guide to the PURPLE crying window covers in detail. For the editorial mission behind the Wermom team's newborn coverage and the advisors who shaped it, see our editorial principles.