Sleeping newborns can anticipate the timing of notes in music within days of birth, but they do not appear to predict melodic pitch changes, according to a study published Feb. 5, 2026, in the peer-reviewed journal PLOS Biology. The findings add evidence that the human brain arrives prepared to process rhythm, a building block that also underlies movement and speech, while melodic tracking may rely more on learning and exposure after birth.
Researchers recorded electroencephalography, or EEG, from 49 sleeping newborns while the infants listened to monophonic piano pieces by Johann Sebastian Bach through earphones, the study said. The team also played “shuffled” versions of the same music, rearranging tones and inter-onset intervals to disrupt structure while keeping basic sound features.
To test whether babies’ brains tracked musical structure, the researchers used computational modeling to estimate how predictable each note was based on what came before. They examined two dimensions: timing (rhythm) and pitch (melody). The researchers then looked for neural signatures that aligned with those predictions, using a method called temporal response functions that can estimate how different features of a continuous sound stream relate to EEG activity over time.
They found evidence that newborns’ brains encoded “probabilistic rhythmic expectations” while listening to real Bach pieces, but not while listening to shuffled music. In other words, when the music retained meaningful structure, babies’ neural activity reflected when a note’s timing came as more or less of a surprise given the preceding pattern.
The team did not find comparable evidence that newborns tracked melodic expectations based on pitch relationships in those same musical excerpts.
The paper argues that the contrast matters because rhythm and melody rely on different kinds of information. Rhythm depends on patterns in timing, while melody depends on patterns in pitch. The study also attempted to separate higher-level expectations from simpler acoustic cues by including low-level features such as acoustic onsets, spectral flux and local changes in timing and pitch in its modeling.
“Overall, this study provides neurophysiological evidence that tracking rhythmic statistical regularities is a capacity present at birth, whilst melodic tracking might not be,” the authors wrote.
Roberta Bianco, the study’s first author, suggested the results may connect to how these abilities emerged over evolutionary time. “From an evolutionary point of view, rhythm is maybe…rooted in more ancient brain functions and melody may be more young,” she said in an interview with NPR’s “Short Wave.”
The idea fits with some cross-species research cited by the PLOS Biology authors, who wrote that sensitivity to rhythmic patterns appears in nonhuman primates, while sensitivity to melodic pitch relations is more variable and may be weaker outside humans. The study did not attempt to test animals, but it framed the newborn findings as consistent with the possibility that rhythm rests on older, broadly shared auditory capacities.
Outside experts cautioned that the work answers one question while raising others.
Lauren Trainor, a developmental neuroscientist at McMaster University who studies music perception, said rhythm permeates infant behavior and physiology. “Babies crawl rhythmically. They flail their arms rhythmically. Their heartbeats are rhythmic,” Trainor told NPR. “So in biological systems, rhythms are just fundamental to everything from movement to perceiving things like speech or music to thinking.”
The researchers acknowledged limits. Newborns have had months of prenatal exposure to rhythmic cues such as a mother’s heartbeat and speech prosody, and the study cannot cleanly separate what the brain may have learned before birth from what is genetically prepared, a point noted in coverage of the work. The study also tested a specific kind of music: monophonic classical piano lines, which may not capture every feature of melody that infants could learn to track in different settings.
Even so, the study’s approach offers a detailed look at how early brains process structure in real-world sound streams, rather than relying only on simplified tones or obvious “oddball” interruptions. The paper also puts a date stamp on a long-running debate in music cognition: which parts of musical understanding require experience and which parts come early.
For parents, the results do not mean newborns “understand” music in the everyday sense. But for researchers, the data suggest something more basic and potentially broader: from the start, the human brain may treat timing as a central organizing principle for sound, with melody building later through exposure to speech and music.
Original article
Comments