A cross-cultural examination of young children’s everyday language experiences

Abstract

Cross-cultural variation in children’s early linguistic experience is vastly understudied. Here we quantify young children’s language exposure across communities speaking North American English, British English, Argentinian Spanish, Yélî Dnye, or Tseltal. Our data includes annotations from 70 daylong, naturalistic recordings of 1–36-month-olds. We focus on what children heard in terms of speaker gender, age, and addressee in randomly selected clips from each recording. We find three key results. First, speech quantity was remarkably stable across age. Second, the bulk of child-directed speech children hear comes from women. Finally, we report on differences and similarities in relative rates of adult- and child-directed speech across communities. Critically, this work provides a much-needed cross-cultural approach to understanding young children’s early language experiences.

Toward a Comparative Approach to Language Acquisition

“…However, a more recent study comparing linguistic environments across industrialized and indigenous communities has revealed a more complex pattern of similarities and differences that largely cuts across the WEIRD/non-WEIRD divide. Bunce et al (2021) analyzed the linguistic environments of children learning languages in WEIRD countries (English in North America and the United Kingdom) and non-WEIRD countries (Spanish in Buenos Aires, Argentina; Tseltal Mayan in a small farming village in Southern Mexico; and Yélî Dnye on Rossel Island in Papua New Guinea). They found that children learning Yélî Dnye heard less speech directly from adults compared with children learning English in North America and the United Kingdom, whereas children learning British English, Argentinian Spanish, and Tseltal heard the same amount of child-directed speech as those learning English in North America.…”

Analysis of children’s everyday language experiences using longform audio: promises and pitfalls (Análisis de las experiencias lingüísticas cotidianas de niños y niñas utilizando audio de formato largo: posibles ventajas y dificultades)

“…Future objectives include features that LENA does not include, such as classification according to language (important for studies of bilingual acquisition) and addressee (i.e., CDS vs ADS), as well as improvements to existing functions and the development of a measure of vocal maturity. Some preliminary findings from this project (Bunce et al, 2020) find both similarities and differences across the studied communities. For example, the prominence of female adult speech is consistent, and child age does not emerge as a key predictor of child speech.…”

A meta‐analysis of infants’ word‐form recognition

“…While the effect of familiarity measured with the CDI was inconclusive in the present analysis, it may hint at a problem with the definition of familiarity used in our own experiments, which could explain the contradictory results obtained in the same laboratory for the same language and age group. Future studies should thus assess other familiarity measures, such as word frequency extracted from adult‐directed speech in CHILDES corpora (as done in Ngon et al, 2013) or from newly emerging corpora based on day‐long recordings of infant input (e.g., Bunce et al, 2020).…”