Baby Talk improves child speech learning

Summary: A new meta -analysis confirms that adults exaggerate vocal sounds when they talk to babies, a characteristic known as vocal hyperticulation in speech directed by babies (IDS). When analyzing 55 studies in at least 10 languages, researchers found consistent evidence of this speech pattern, although their strength varied according to the language, method and sample size.

The findings suggest that vocal exaggeration can play a significant role to help babies perceive and learn language. However, methodological inconsistencies and small sample sizes limit the wider generalizations and require a stronger intercultural investigation.

Key facts:

Cross language confirmation: Vocal hyperticulation in IDS was found in at least 10 languages. Meta-Alcance allysis: Combined results of 55 studies to strengthen statistical power.

Source: University of Tokyo

There are many factors that contribute to the development of children’s language, among which is the baby’s talk: modified speech mothers and other adults often use when they talk to babies.

Baby Talk is more formally known as speech led by infants (IDS) often differs from the discourse led by adults (ADS) both in tone and vocabulary. However, a component of the ID that remains controversial is the exaggeration or hyperticulation of vocal sounds.

Some experts argue that vocal exaggeration produces lighter speech sounds that are easier to process for babies, but studies conflict in whether vocal hyperticulation is really present in IDS.

A group of scientists from the International Center for Neurointelligence Research (WPI-CIVN) at the University of Tokyo, the University of Amsterdam, the University of PSL and the University of Aarhus analyzed the existing research on vocal hyperticulation in IDS to determine if this specific type of exaggeration of speech occurs in IDS.

The team published its research on 02-06-2025 in the Psychological Bulletin magazine.

Before metaanalysis, a group of studies suggested that adults particularly exaggerate the vowels “A”, “I” and “U” when talking to babies, which makes the vowels easier to perceive and potentially facilitate words learning.

On the contrary, other studies had not detected the consistent exaggeration of the vowels in the IDs, which suggests that the baby’s conversation can be a byproduct of smiles or some other positive effect.

“(W) decided to perform (a) meta-analysis in all studies on this topic to understand if vocal exaggeration is a characteristic of discourse aimed at babies, and if so, what factors led to the differences observed in existing research,” said Irena Lovčević, postdoctoral researcher of the International Research Center for Neurointell The University of The Research.

The researchers conducted meta -analysis in 20 IDS vocal hyperticulation studies that used a similar methodology and 35 studies that used any research method. Meta -analysis techniques combine the results of more than a similar research study to improve the statistical power of the results.

“Our meta -analysis confirms that mothers exaggerate vowels in speech aimed at babies in at least 10 languages. However, our results also point to the diversity of methods used in existing research.

“Therefore, we want readers, especially speech acquisition researchers, carefully consider and document methodological decisions in their research and avoid generalization from one language to another or one method to another,” said Lovčević.

In general, the team observed that IDS vocal hyperticulation studies have lower sample sizes than what would be expected for well -enhanced statistical analysis and recommend that the study size increase.

The researchers also pointed out that the differences in hyperticulation between languages ​​were difficult to detect due to smaller sample sizes for less represented languages ​​and that more intercultural and intercultural research is needed to generalize the results of language acquisition research.

In addition, the size of vocal hyperticulation could also be influenced by the measurement scale (eg, Hertz, Mel) used by the research team, illustrating how intimate methodology can influence experimental results and the ability to compare those results in studies.

After establishing the presence of vocal hyperticulation in IDS, the research team has an eye on learning more on other aspects of children’s language development.

“The next step is a more detailed examination of different factors that influence the development of language in babies, with a special approach to modifiable factors of the immediate environment of babies, such as the amount of discourse to which babies and the qualities of interactions with the care of the vocal caregiver are exposed.

“How babies learn to speak without so much effort is still a great puzzle, so my goal is to contribute to solve … this puzzle,” Lovčević said.

Titia Benders of the University of Amsterdam, Netherlands; Sho Tsuji of the International Center for Neurointelligence Research (WPI-CIVN) at the University of Tokyo for advanced studies in Japan and the cognitive Laborate of Sciences et de psycholinguistique, Département d ′ Cognitives, Ash, Ehes, CNRs in PSL University in France; and Riccardo Fusaroli of the department of linguistics in cognitive science and semiotics and the center of interactive minds of the University of Aarhus in Denmark also contributed to this research.

FINANCING: This work was supported by an initiative for initiative for International Research Research MEXT, JSPS Kakenhi number 20H05617 and 20H05919, and seed funds of the center of interactive minds.

About this language and learning news

Author: Kazuyo Okada
Source: University of Tokyo
Contact: Kazuyo Okada – University of Tokyo
Image: The image is accredited to Neuroscience News

Original research: closed access.
“Acoustic exaggeration of vowels in the discourse led by babies: a multiple meta -analytical review of Irena Lovčević et al. Psychological Bulletin

Abstract

Acoustic exaggeration of vowels in the discourse led by babies: a multiple meteanalithic review of method

There is a long -standing debate about the extent to which vocal hyperticulation, the production of acoustically exaggerated vowels, occurs in speech directed by babies (IDS).

It has been argued that this exaggeration results in lighter speaking sounds that are easier to process for babies and that they could be positively related to the linguistic results of babies.

However, the previous findings regarding the presence of vocal hyperticulation in IDS are apparently inconsistent and contradictory, which makes it difficult to advance in our understanding of the role of hyperticulation in language development and, consequently, the possible functions/roles of ID in the acquisition of language.

Therefore, we adopt a systematic review and a meta -alitical approach to investigate the robustness of vocal hyperticulation in IDs and identify sources of heterogeneity in literature.

We use four complementary meta -alitical approaches and evaluate the robustness of the results in the different options.

First we carry out traditional (frequentist) and Bayesian meta -analysis in methodologically consistent studies (20 studies, 42 effect sizes) and then in all studies of vocal hyperticulation in ID, regardless of the method (35 studies, 80 effect sizes).

The results indicate the presence of vocal hyperticulation in IDS compared to speech aimed at adults (effect sizes that vary from 0.41 to 0.69), as well as systematic and non -systematic variability due to, for example, a variability and interlinguistic methods used, which makes it difficult to identify specific factors associated with the hyperthericulation of stronger vowels.

The quantitative results combined with a systematic review of literature also allow important methodological information, which we summarize in recommended practices, such as the expansion of sample sizes and explicitly incorporating sources of heterogeneity in the analysis.