This study investigated the information encoded in human infant cries. Using a large dataset of 39,201 cries from 24 babies recorded longitudinally (15 days to 3.5 months), researchers found that cries reliably convey information about age and identity. Cries became more tonal and less shrill with age, while individual acoustic signatures were stable. However, neither machine learning nor trained listeners could reliably identify the cause of crying (hunger, discomfort, or isolation).
Publisher
Communications Psychology
Published On
Oct 02, 2023
Authors
Marguerite Lockhart-Bouron, Andrey Anikin, Katarzyna Pisanski, Siloé Corvin, Clément Cornec, Léo Papet, Florence Levréro, Camille Fauchon, Hugues Patural, David Reby, Nicolas Mathevon
Tags
infant cries
acoustic signatures
age
identity
crying causes
machine learning
longitudinal study
Related Publications
Explore these studies to deepen your understanding of the subject.
