Paperis.

© 2026 네오쿤스. All rights reserved.

공지사항이용약관개인정보처리방침환불 정책콘텐츠·제거 요청

상호: 네오쿤스 · 대표자: 김근태 · 사업자등록번호: 751-04-03501 [사업자정보 확인] · 통신판매업 신고: 제 2026-서울서초-3586호 (서초구청)

주소: 서울특별시 서초구 서초대로 266, B103-S36호(서초동) · Tel: 010-7254-2475 · 이메일: support@paperis.app · 고객센터 운영 시간: 평일 10:00–16:00 (한국 시간, 주말·공휴일 제외) · 호스팅 서비스 제공자: Vercel Inc.

Paperis.
Why Do We Know a Lullaby the First Time We Hear It?

Paperis Articles

Why Do We Know a Lullaby the First Time We Hear It?

Why? · Lullabies

🔥Debate✦AI-generatedArts & Culture · Nov 16, 2026

Your field's papers, as audio

Pick your fields and journals — Paperis summarizes the key papers and turns them into audio. Listen on your commute.

Start Paperis free →

Why Do We Know a Lullaby the First Time We Hear It?

The child listens. The parent reads. This page helps you answer the questions your child may ask after the story. You can see the story script with the script button while it plays.

What this story is about

A lullaby is a song people sing to calm a baby and help the baby sleep. The story asks one question. Can people tell that a song is a lullaby when they do not know its language or culture? It follows how people tested this. Researchers collected songs from many societies in a fair way. They played short parts to people who did not know the songs. They used the same method with children aged 4 to 16. For babies, who cannot talk, they used the body's signals.

The answer has two layers. People guessed lullabies, dance songs, and healing songs from new foreign songs quite well, and children did too. But not every kind of song worked (love songs did not). Songs were also very different inside one society. And in one group, there was no sign of songs for babies. This page tells both sides: what is shared and what is different. Why the shared parts exist is not fully known yet.

The studies here did not measure whether lullabies make babies fall asleep. The baby study measured relaxing, through body signals. This page does not give advice about sleep.

How people found out

Collect songs fairly, and play them to people who do not know them. The first test used songs from 86 places. There were dance songs, lullabies, healing songs, and love songs. Most places were small groups, including hunter-gatherers, herders, and farmers who grew their own food. The songs were picked in a near-random way, region by region. 750 internet users in 60 countries heard 14-second parts. For each song, they gave scores on six uses (for example, "used to calm a baby"). The listeners did not know the societies, the parts were very short, and the music was very varied. Still, the scores showed correct guesses about each song's use, from the song alone. The guesses held across cultures (PMID 29395919). The test and its analysis plan were registered before it began (PMID 29395919).

Clues outside the music. In a second test, 1,000 internet users in the United States and India rated the same parts. They rated three context features, like the singer's gender. They also rated seven music features, like how complex the tune was. The context features predicted the scores from the first test. But the music features and the songs' real uses also explained their own share (PMID 29395919). So some features outside the music (not the tune or the beat, but things like the singer's gender) predicted the scores. But they did not explain everything. This is a link, not proof that those features changed the guesses. The story says these clues "went along with the guesses." The authors concluded that the results fit the idea of universal links between a song's form and its use (PMID 29395919).

Beyond English speakers. Earlier studies looked almost only at English speakers. So the next study took 5,516 people in industrial places with the internet. They spoke 28 first languages. It also took 116 people in smaller groups with little access to world media. They spoke three first languages, none of them English (DOI: 10.1073/pnas.2218593120). Listeners heard songs picked at random from four uses and rated how much each song seemed made for each use. Both groups guessed the uses of dance songs, lullabies, and healing songs, but not love songs (DOI: 10.1073/pnas.2218593120). Guesses agreed within and across groups. When the listener and the singer had closer languages or places, the guesses became only a little more correct (DOI: 10.1073/pnas.2218593120). As background, the same summary says dance songs tend to be loud and rhythmic. Lullabies tend to be quiet and full of tune (DOI: 10.1073/pnas.2218593120). The authors said the uses of three common kinds of music can be understood across cultures. They also said music's diversity, shaped by cultural change, still rests in part on shared ways of hearing (DOI: 10.1073/pnas.2218593120).

Children. 5,033 children were recruited online, from age 4 up to age 16. They heard dance songs, lullabies, and healing songs from 70 world cultures and guessed each song's first use. The children guessed reliably. Older children did only a little better than the youngest. So there was little sign that experience made the difference (PMID 36222671). The children's guesses were very close to those of 98,150 adults in a similar online test. Similar sound features predicted both groups' guesses (PMID 36222671). The authors concluded that guessing a song's use does not always need a lot of music experience (PMID 36222671). The Blackfoot lullaby example in the story comes from this paper's summary. Think of a Korean listener who knows nothing of Blackfoot music, language, or culture. This listener is far more likely to call a Blackfoot lullaby "used to calm a baby" than "used for dancing" (PMID 36222671). The summary does not say which countries or languages the children came from.

There are no percent-correct numbers. The summaries of these three studies say "accurate" or "reliably." They do not say what percent of answers were right. So the story does not give a number either.

Speech and song for babies. 410 people in 21 city, country, and small societies made 1,615 recordings. Speech and song for babies could be told apart from speech and song for adults by sound features alone. Then 51,065 people from 187 countries heard them on an English website. They guessed if each was for a baby. They did better than chance (PMID 35851843). This is not in the story. But it suggests something. What lullabies share may be part of how people sound to babies in general, not only in songs.

For babies, ask the body. 144 babies in the United States heard 8 new foreign lullabies. They also heard matched songs from other foreign societies that were not lullabies. The babies relaxed more to the lullabies. Relaxing was measured by heart rate, pupil size, and skin signals. The response was the same through the whole first year. Babies have much less music experience than adults. So the response likely does not come from experience (PMID 33077883). The parents were asked which songs they would use to calm a fussy baby. The songs were new to them too. Still, far more parents chose lullabies (PMID 33077883). The authors said babies may be born ready to respond to features that lullabies share across cultures (PMID 33077883). This study did not measure sleep. The story says, "The team measured if the babies relaxed. They did not measure if the babies fell asleep."

Another result comparing play songs and lullabies (not in the story). A review reports these lab findings. Babies 6 to 9 months old heard recorded play songs or recorded talking. With the play songs, they stayed content or calm much longer. Sometimes 10-month-old babies were fussing or crying and very stirred up. Then a mother singing, with face and touch, helped them recover better than a mother talking. Also, in Western babies, play songs lowered arousal better than lullabies (PMID 25773634). This can sound opposite to the "same song, sung two ways" study below. There, arousal went down when the song was sung to calm, and stayed the same when it was sung to play (PMID 30912725). But the two compared different things. The review compared kinds of songs (play songs and lullabies). The study changed only the way one song was sung. The review summary does not say if the songs were recorded or sung live, or what state the babies were in. So we cannot put the two together. That is why the story uses only the "same song" study. Neither one measured sleep, and neither tells parents which songs to sing.

The same song, sung two ways. 30 mothers sang "Twinkle, Twinkle, Little Star" to their babies again and again. They switched between a calming way and a playful way. Arousal was measured by skin signals. In the calming way, the arousal of mother and baby went down as the song went on. It ended below where it started. In the playful way, it stayed the same. Babies paid more attention to their mothers in the playful way. The playful way was faster, higher, and louder, with a clearer beat. The mothers also moved with more energy (PMID 30912725). The story's line, "the calm way was slower, lower, and softer," turns this comparison around. The story names the song but does not use its words.

Answers you often hear

"You can tell because it is slow and quiet." Partly true. Lullabies tend to be quiet and full of tune (DOI: 10.1073/pnas.2218593120). The same song sung to calm was slower, lower, and quieter than when sung to play (PMID 30912725). But in the first test, features outside the music, such as the singer's gender, also predicted the scores (PMID 29395919). The summaries on this page do not name one sound feature as the most important.

"Every country in the world has lullabies." "Every" goes beyond the evidence. In a study of records from many societies, what was found in every society was music. Music was linked to some uses, including caring for babies (PMID 31753969). But look at the Northern Aché of Paraguay. Researchers studied them for 122 months in all, between 1977 and 2020. They found no sign of dancing or of songs sung to babies. Besides church songs brought by missionaries, the adults sang alone and in very few styles (PMID 40306280). The authors looked at several kinds of evidence. They think dancing and baby songs were lost long ago. Back then, the number of their people became smaller, and their culture became simpler. They wrote that this fits the idea that all humans can dance and sing to babies. But it shows that passing culture on is key for these things to last (PMID 40306280). The story tells this exception inside the story.

"Music works even when words do not." Only partly true. Three of the four uses were guessed across cultures, but love songs were not (DOI: 10.1073/pnas.2218593120). Also, musical behavior differed more inside societies than between them. There were three measures: how formal, how excited, and how religious. On any one measure, one-third of societies differed clearly from the average. Half differed on at least one measure (DOI: 10.1126/science.aax0868). These numbers are in the journal's summary of the same paper.

"Babies fall asleep when they hear lullabies." The studies here did not measure sleep. The foreign lullaby study measured relaxing through body signals, not sleep (PMID 33077883). The story also says, "They did not measure if the babies fell asleep." Our collected papers include several hospital trials that played recorded lullabies, such as in newborn intensive care. They are about medical care. This story asks a different question: how we know a new lullaby. So we did not use them.

Questions your child may ask

"Why couldn't people guess love songs?" The study did not explain why. The summary reports only that love songs were not guessed across cultures (DOI: 10.1073/pnas.2218593120). It is most accurate to say, "We do not know yet."

"Did Mom and Dad sing to me too?" You could start by thinking of songs together that your child may have heard as a baby. For background: in one survey of 945 parents, most parents sang to their babies every day. How often they sang was not linked to income or ethnic group. Parents sang less as children grew older. Reports from the past 30 years were put together. Home technology changed a lot in that time. But the amount of singing changed very little (PMID 34719251). The summary does not say which country the survey was in.

"Do babies like songs or talking better?" Studies differ. 66 babies aged 6 to 10 months listened longer to singing than to speaking (PMID 27796032). But babies aged 4 to 13 months listened longer to happy speech than to hummed lullabies by the same woman. The researchers thought a happy voice, not speech or song, was the main thing that held babies' attention (PMID 23805119).

"Is it different when Mom sings?" This is a different question, so the story does not cover it. In one study, 5-month-old babies looked longer at a new person who sang a tune their parent had sung at home. This did not happen when they had learned the tune from a toy or from a friendly stranger (PMID 26917211). This shows songs carry social meaning for babies. It does not show that a mother's song calms better.

What we know and what we do not know yet

What we know. In the records studied, every society had music (PMID 31753969). A song's sound told its main use (PMID 31753969). People guessed the uses of new foreign lullabies, dance songs, and healing songs. Adults in many countries did (PMID 29395919; DOI: 10.1073/pnas.2218593120). People in small groups with little world media did (DOI: 10.1073/pnas.2218593120). Children aged 4 to 16 did too (PMID 36222671). Babies in the United States relaxed to new foreign lullabies, as shown by body signals (PMID 33077883).

The differences we know. Love songs were not guessed (DOI: 10.1073/pnas.2218593120). Musical behavior differed more inside societies than between them (PMID 31753969). The journal's summary of the same paper says that on each dimension, one-third of societies differed clearly from the average (DOI: 10.1126/science.aax0868). No songs for babies were recorded among the Northern Aché (PMID 40306280).

Not known yet: why the shared parts exist. A review asked three things about how people respond to music. Is it shared by all? Is it special to music? Does it start early in life? It found evidence that the mind's systems behind them are universal and appear early. But it said current evidence cannot show they are special to music. Instead, many results suggest these responses come from wider abilities, like emotion and hearing, that did not evolve for music. It also said singers and players make small changes over time. In this way, culture shapes music to reach shared parts of the mind. This makes both universal patterns (like the link between form and use) and styles unique to each culture (PMID 38143935). The story's line about singers and players shaping songs to reach people's hearts gives this view to the review.

Not known yet: how far the evidence goes. The baby study is one group of babies in the United States (PMID 33077883). The finding that play songs lowered arousal better than lullabies is also about Western babies (PMID 25773634). The children were found online (PMID 36222671). Most adults in the large study (5,516 people) lived in industrial places with the internet (DOI: 10.1073/pnas.2218593120).

Things to try together

  • Compare a family lullaby and a play song. Ask family members to sing one lullaby they know. Then ask for one fun play song. Compare speed, loudness, repeating parts, and high and low notes with your child. A study found that the same song had different speed, pitch, and loudness when sung to calm or to play (PMID 30912725). Hearing a lullaby from grandparents can also be a way for your child to see culture being passed on (PMID 40306280).
  • Hum and guess. Hum a song with no words, and let others guess if it is a lullaby or a play song. This is a small copy of tests where people guessed a song's use from a short part (PMID 29395919, PMID 36222671). Tell your child it is fine to be wrong. In the studies, not every kind of song was guessed either (DOI: 10.1073/pnas.2218593120).

Evidence

  • "Form and Function in Human Song." (2018) — PMID 29395919 — 750 people in 60 countries heard 14-second parts of songs from 86 societies and guessed their uses accurately; music features and real uses explained a share apart from context clues.
  • "Universal interpretations of vocal music" (2023) — DOI: 10.1073/pnas.2218593120 — 5,516 speakers of 28 languages and 116 people in small societies guessed dance songs, lullabies, and healing songs, but not love songs; closer language or place helped only a little.
  • "Children infer the behavioral contexts of unfamiliar foreign songs." (2023) — PMID 36222671 — 5,033 children aged 4 to 16 reliably guessed song uses from 70 cultures with little change by age, closely matching 98,150 adults.
  • "Acoustic regularities in infant-directed speech and song across cultures." (2022) — PMID 35851843 — In 1,615 recordings by 410 people in 21 societies, speech and song for babies were told apart by sound; 51,065 people in 187 countries guessed better than chance.
  • "Infants relax in response to unfamiliar foreign lullabies." (2021) — PMID 33077883 — 144 US babies relaxed to 8 new foreign lullabies, by heart rate, pupil, and skin signals; parents mostly chose lullabies to calm a baby.
  • "Musical affect regulation in infancy." (2015) — PMID 25773634 — A review: a mother's singing calmed fussing 10-month-olds better than talking, and play songs lowered arousal more than lullabies in Western babies.
  • "Effects of Maternal Singing Style on Mother-Infant Arousal and Behavior." (2020) — PMID 30912725 — When 30 mothers sang the same song to calm, mother and baby arousal went down; the playful way was faster, higher, and louder, with more baby attention.
  • "Universality and diversity in human song." (2019) — PMID 31753969 / DOI: 10.1126/science.aax0868 — Every society observed had music, and sound features predicted a song's main use; but music varied more within societies (the one-third figure for each dimension is from the journal's summary).
  • "Loss of dance and infant-directed song among the Northern Aché." (2025) — PMID 40306280 — In 122 months of fieldwork from 1977 to 2020, there was no evidence of dance or baby songs among the Northern Aché; they were likely lost after the population shrank.
  • "Universality, domain-specificity, and development of psychological responses to music." (2023) — PMID 38143935 — A review: responses to music are universal and appear early, but there is no evidence they are special to music; many results point to wider abilities.
  • "Across demographics and recent history, most parents sing to their infants and toddlers daily." (2021) — PMID 34719251 — Most of 945 parents sang to their babies daily, and this changed little over 30 years.
  • "Infants Prefer Infant-Directed Song Over Speech." (2017) — PMID 27796032 — 66 babies aged 6 to 10 months listened longer to singing than to speaking.
  • "Speech vs. singing: infants choose happier sounds." (2013) — PMID 23805119 — Babies listened longer to happy speech than to hummed lullabies; a happy voice mattered more than speech or song.
  • "For 5-Month-Old Infants, Melodies Are Social." (2016) — PMID 26917211 — Babies looked longer at a new person singing a tune their parent had sung; this did not happen with tunes from a toy or a stranger.
Sources (14)
  • Form and Function in Human SongPMID 29395919
  • Universal interpretations of vocal musicDOI 10.1073/pnas.2218593120
  • Children infer the behavioral contexts of unfamiliar foreign songsPMID 36222671
  • Acoustic regularities in infant-directed speech and song across culturesPMID 35851843
  • Infants relax in response to unfamiliar foreign lullabiesPMID 33077883
  • Musical affect regulation in infancyPMID 25773634
  • Effects of Maternal Singing Style on Mother-Infant Arousal and BehaviorPMID 30912725
  • Universality and diversity in human songPMID 31753969
  • Loss of dance and infant-directed song among the Northern AchéPMID 40306280
  • Universality, domain-specificity, and development of psychological responses to musicPMID 38143935
  • Across demographics and recent history, most parents sing to their infants and toddlers dailyPMID 34719251
  • Infants Prefer Infant-Directed Song Over SpeechPMID 27796032
  • Speech vs. singing: infants choose happier soundsPMID 23805119
  • For 5-Month-Old Infants, Melodies Are SocialPMID 26917211
📚 Arts and Culture for Kids
5 / 6
  1. 1.Why Won't a Song Leave My Head?
  2. 2.How Do We Know It's Still "Twinkle"?
  3. 3.Why Does a Sudden "Bang!" in Music Surprise Me?
  4. 4.Can You Tell the Animal Just by Hearing the Music?
  5. 5.Why Do We Know a Lullaby the First Time We Hear It?
  6. 6.Who Made Up the Old Fairy Tales?
← PreviousCan You Tell the Animal Just by Hearing the Music?Next →Who Made Up the Old Fairy Tales?

This article is AI-generated from research papers for general information and does not replace professional advice. Spotted an error? Let us know

© 2026 Neokuns (Paperis). Sharing links is welcome; republishing the full text requires prior permission.