Investigating the robustness of infants’ preference for infant-directed speech: A large-scale, multi-site collaboration in Africa

• In Event: Updates from ManyBabies: Collaborative efforts to promote best practices and improve robustness in developmental science

Abstract

Numerous past studies with convenience-sampled participant populations have shown that young infants prefer listening to infant-directed speech (IDS) over adult-directed speech (ADS). Infants’ IDS preference is also hypothesized to be related to infants’ better general attention and later language outcomes (e.g., Spinelli, Fasolo, & Mesman, 2017), pointing to the importance of studying infants’ IDS preference across infants with different backgrounds. Recently, a large-scale, multisite collaborative study found that infants aged 3-15 months and from various language backgrounds (specifically, 16 languages) have an overall preference for North-American English IDS over ADS (ManyBabies Consortium, 2020). However, almost all infants in this study were from western, educated, industrialized, rich, and democratic (WEIRD) populations (e.g., from North America or Europe), with only a small subset of participants from non-WEIRD populations (Figure 1). It remains unknown whether infants’ preference for IDS will be observed in samples of participants from non-WEIRD populations around the world. This sampling bias makes it difficult to build general, broadly applicable theories on infants’ IDS preference to infants across different cultures and languages. Our study addresses this problem by extending the ManyBabies Consortium (2020) paradigm to understudied African populations.
We are planning a large-scale, multisite study that will investigate whether infants show a preference for IDS over ADS in countries across the African continent. Extending western research methodology to other cultures requires careful adaption. We have discussed the challenges of adapting our paradigm in Africa (e.g., adapting questionnaires to ensure they are culturally appropriate) and our strategies for supporting a research team with a broad range of resource availabilities (e.g., providing experimental equipment). In order to adapt our methodology to fit the cultures and resources, we gathered a team of researchers from 9 countries in Africa for a multi-day training workshop in Nairobi, Kenya in January 2020, and worked with laboratories to set up the experimental paradigm. Currently, 11 labs in sub-Saharan Africa (Figure 2) have joined and are projected to collect data from over 300 infants in total. These labs cover regions in East Africa (e.g., Kenya), the west (e.g., Senegal) and the south (e.g., South Africa). We will collect data from infants across a number of diverse regions in Africa, and will explore whether variations in infants’ demographic factors within non-WEIRD societies may affect their IDS preferences. Further, with a larger sample size, our study can also provide a reliable estimate of the IDS preference effect size in Africa, enabling comparison of infants’ IDS preference in existing WEIRD samples.
In sum, this study will measure the robustness and generalizability of infants’ IDS preference in a more demographically diverse population, providing important insights into how variation in infant’s demographics in these societies may affect their IDS preference. This work represents the ManyBabies Consortium's goals and efforts to provide a more diverse, global and comprehensive picture of infancy research, and sets the stage for future collaborations with colleagues on the African continent.

Group Authors

The ManyBabies Africa team

Author

• Sin Mei Angeline Tsui, Stanford University
  Presenting Author