Developmental Science Research Article: Input to the Language Learning Infant: The Impact of Other Children

Johanna Schick: Institute for the Interdisciplinary Study of Language Evolution, University of Zurich, Zurich, Switzerland
Moritz M. Daum: Department of Psychology, University of Zurich, Zurich, Switzerland; Jacobs Center for Productive Youth Development, University of Zurich, Zurich, Switzerland
Sabine Stoll: Institute for the Interdisciplinary Study of Language Evolution, University of Zurich, Zurich, Switzerland
Correspondence: Johanna Schick (johanna.schick@uzh.ch)
Received: 23 September 2024
Revised: 15 March 2025
DOI: https://doi.org/10.1111/desc.70045

In urban industrialized cultures, adult child-directed speech is the best predictor for how children acquire their native language.
Many societies, however, exhibit considerably less exposure to adult speech directed at children.
Explored Area: The impact of children's speech on language acquisition, especially in non-Western cultures.
Findings:
- Shipibo-Konibo infants (from the Peruvian Amazon) and Swiss infants showed greater attention to children talking among themselves than adults.
- Swiss infants, despite greater exposure to child-directed speech from adults, attended equally to both child-directed speech and child speech.
- Suggests the importance of children's speech as an input for language acquisition.

Language Acquisition: Children learn to speak their native language in early life.
Child-Directed Speech: Speech characterized by specific acoustic and structural features (e.g., higher pitch, frequent repetitions) and preferred by infants (Fernald 1985; Cooper and Aslin 1990).

Most studies focused on children in Western child-centered cultures (Kidd and Garcia 2022).
Research has shown a notable cultural variation in caregiver-child interactions, highlighting diverse child-rearing practices (Ochs and Schieffelin 1984; Gaskins 2020).
Variation Examples:
- Non-Western cultures may feature less child-directed speech but more speech from other children in infancy (Shneidman and Goldin-Meadow 2012).
Emerging Perspectives: Children learn from both directed and surrounding speech (Floor and Akhtar 2006; Arunachalam 2013).

Language: Shipibo-Konibo, a Panoan language from the Peruvian Amazon, with about 30,000 speakers (Valenzuela 2003).
- Bilingualism is common.
- Child-led alloparenting is prevalent, with independence nurtured early on.
- Data gathering occurred across three interconnected villages.

Language: Swiss German, a continuum of Alemannic dialects.
- Urban setups (e.g., Zurich) predominantly consist of nuclear families with approximately 1.5 children per family (Craviolini 2020).
High rates of children accessing external childcare services (63.9% attend at least weekly) (Bundesamt für Statistik 2022).

3.1.1 Participants

Infants Tested: 127 infants (8–20 months)
- Shipibo-Konibo: 67 infants (mean age: 13.3 months, average siblings: 3.85).
- Swiss: 60 infants (mean age: 13.9 months, average siblings: 0.47).
Data Exclusions: Infants excluded due to fussiness (n=22).

3.1.2 Stimuli

Stimuli: 12 conversation snippets (25 seconds each) from naturalistic recordings, including child and adult speech from various demographic backgrounds.
- Acoustic Measurement: Statistical details of fundamental frequency (F0) provided in Table 1.

Infants tested in a controlled environment, observing audio-visual stimuli while seated on caregiver's lap.
Manual Coding: Infants’ looking times recorded and analyzed for inter-coder reliability (ICC = 0.91) using ELAN software.

Utilized a Bayesian generalized mixed model to analyze differences in looking times between groups and stimuli types, including several fixed and random effects.
- Model Parameters: Included child ID as a random effect, assessed via Monte Carlo simulations for convergence and validity.

Infants averaged over 10.5 trials.
Shipibo-Konibo: Average looking time for adult speech: 5.23 s; for child speech: 8.22 s.
Swiss infants: Adult speech: 9.77 s, child speech: 11.51 s.
Analysis Result: Strong evidence that infants pay more attention to child speech across cultures.

12 speech snippets from different speakers, including both child speech and child-directed speech from adults.
- Controlled Procedure: Consistent design with Study 1

Average looking time across conditions: Adult speech: 11.16 s; Child speech: 11.27 s.
No significant difference in attention between child-directed speech and child speech.

Conclusion: The observed child speech captures more attention than adult speech, underscoring its importance in language learning across diverse cultures.
Suggests child-directed speech is comparable in efficacy to child speech, warranting further investigation into the implications for language acquisition.
Suggests increased cross-cultural research to understand language acquisition dynamics in different sociocultural contexts.

The research utilized a comparative approach across two studies to evaluate infant attention to different speech types:

Study 1: Conducted with $127$ infants ( $8$ - $20$ months) from Shipibo-Konibo ( $n=67$ ) and Swiss ( $n=60$ ) cultures. The methodology involved presenting $12$ naturalistic conversation snippets ( $25$ seconds each) consisting of child and adult speech. Researchers manually coded looking times using ELAN software (inter-coder reliability ICC = $0.91$ ) and applied a Bayesian generalized mixed model for statistical analysis.
Study 2: Focused on $61$ Swiss infants ( $8$ - $20$ months) to compare attention levels between child speech and adult child-directed speech (CDS) using the same controlled experimental design as Study 1.

Cross-Cultural Preference: In Study 1, infants from both cultures showed significantly greater attention to child speech than to adult speech.
- Shipibo-Konibo infants: Average looking time of $8.22$ s for child speech versus $5.23$ s for adult speech.
- Swiss infants: Average looking time of $11.51$ s for child speech versus $9.77$ s for adult speech.
CDS Comparison: Study 2 revealed no significant difference in attention between adult child-directed speech ( $11.16$ s) and child speech ( $11.27$ s).
Final Conclusion: The findings suggest that child speech is a highly salient and important input for language acquisition across diverse sociocultural contexts. It captures more attention than general adult speech and is comparable in efficacy to child-directed speech, indicating that other children play a vital role in the early language-learning environment.