Science1 publisher3 min readPublished
A two-decade review puts the foundations of beat perception in the newborn brain
Gábor Háden and Henkjan Honing read twenty years of studies on how a sense of beat develops and evolved, and they report that newborns already have the capacity in rough form while the field's stronger claims go beyond the evidence.
The Scientist · Science desk

What happened
- Gábor Háden of HUN-REN and Henkjan Honing of the University of Amsterdam have published a critical review of more than two decades of research on how beat perception develops and evolved, in the Annals of the New York Academy of Sciences.
- They test four claims the field makes routinely: that beat processing is near-universal across cultures, develops spontaneously, supports a distinctive form of predictive timing, and is rare across species.
- Their conclusion is that the foundations of beat perception are present at birth and refined by experience, movement and culture, and that the available evidence does not justify every stronger claim made in the field.
- Macaques detect regular timing and will adjust their tapping tempo to music after extensive training, but generally lack the stable phase alignment and flexible predictive synchronization humans manage without it.
- The authors add reward to the Gradual Audiomotor Evolution hypothesis, proposing that human beat perception came from tighter integration of old auditory, motor, timing and reward systems.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- contradiction phys.org headlined the review "Brain's ability to hear a musical beat remains distinctly human, study suggests", which is a firmer position than the paper it summarizes takes; anyone citing the headline inherits a claim the review hedges.
- constraint Because the newborn evidence is electrophysiological, the developmental claim can only reach as far as neural sensitivity, and the stronger version of it needs behavioral measures that infants that young cannot supply.
- precedent If the reward-integration account holds, the experiment that settles what is human about beat perception is a measurement of coupling strength across species, and a hunt for a human-only circuit would be looking for the wrong object.
In the newborn work the review leans on, the informative trial is the silent one. Recordings of newborn brain activity show responses to sounds omitted at metrically important positions, even when nothing louder marked those positions [6]. A loudness-detection account predicts no response there. The rival explanation is that the infant has learned which sound tends to follow which, and newer experiments indicate the responses cannot be explained by that statistical order alone [7]. What is left is an early sensitivity to periodic structure and to beat-based prediction [8].
The authors set the limit themselves. Neural sensitivity is not the same thing as conscious experience, and early timing capacities go on developing, so the review declines to credit newborns with fully mature beat or meter perception [9]. "Newborns are not miniature adult listeners, but their brains already seem prepared to use rhythmic structure to predict when something should happen," Honing said. "Experience does not create this capacity from scratch; it calibrates and enriches it" [10]. Across infancy and childhood, auditory exposure, motor development and musical enculturation improve the precision, flexibility and hierarchical organization of rhythmic processing [12].
Outside primates the record is thin. Synchronization has been observed in a small number of species, including some birds and a sea lion, and the review describes the evidence from nonhuman animals as sparse [15]. Rarity measured on a literature that small is partly a statement about how many species have been tested.
The human-versus-other-primate difference the review defends is one of degree and of integration among systems both groups have, since its GAER proposal has human beat perception emerging from tighter coupling of old auditory, motor, predictive-timing and reward systems [16][14]. A circuit humans alone possess would be a different claim. "The difference between humans and other primates may be not only whether they can match the tempo of the music, but whether synchronizing is intrinsically motivating and whether auditory, motor and reward systems work together strongly enough to stabilize prediction," Honing said [17].
Of the four claims the review examines, the phys.org account reports evidence bearing on three: spontaneous development, predictive timing, and distribution across species. It does not say what the authors concluded about near-universality across human cultures [20]. The summary verdict quoted there, that the available evidence does not justify every stronger claim made in the field, is not attached to any particular one of the four [5].
The authors ask for closely matched experiments across newborns, adults and other species [18]. They also want behavioral, neurophysiological, genetic and computational work alongside the neuropsychological methods that dominate the field, so that passive tracking of a stimulus can be told apart from internally generated prediction of a beat [18].
What to watch
- An independent replication of the newborn omission responses with the statistical-order control built in; the review's developmental claim rests on that result.
- Direct measurement of reward-system involvement during synchronization in macaques, which is what would make GAER testable rather than plausible.
- A new positive species. The list is currently a handful of birds and one sea lion, and a second mammal would change how rare the trait looks.