About this srcd poster session

---

Panel information
Panel 11. Language, Communication

Abstract

---

Children typically rely on mutual exclusivity—the assumption that each object has only one label—to learn new words (Markman, 1989; Markman, Wasow, & Hansen, 2003). Previous research has shown that the speaker’s identity, such as race (e.g., Weatherhead et al., 2021) or language (e.g., Au & Glusman, 1990), can influence children's flexibility in this assumption. Building on such prior findings, the present research examines whether children’s integrated knowledge of a speaker’s race and language can shape their willingness to override mutual exclusivity and accept two labels, one from a native and the other from a foreign language.

Forty 5-year-old Korean children participated in the current research. All children were native Korean speakers of a single racial background (Asian) with no prior exposure to Spanish. They were randomly assigned to either the same-race or different-race speakers condition. Children in both conditions watched videos of a native-language (Korean) speaker and a foreign-language (Spanish) speaker labeling various objects. The native-language speaker was always an Asian woman, whereas the foreign-language speaker was either an Asian (same-race speakers condition) or Hispanic (different-race speakers condition) woman. During the familiarization trial, the native and foreign speakers introduced themselves and labeled three familiar objects (a shoe, a teddy bear, and a bus) in Korean and Spanish, respectively. In the test trial, a novel object was presented, and the native-language speaker labeled it with a novel Korean word (i.e., “Muppi”) while the foreign-language speaker labeled it with a novel Spanish word (i.e., “Chisa”). Children were asked if they would accept only one label or both (i.e., “Which do you think is the right name for this toy? Muppi? Chisa? Or are both OK?”).

The results revealed a significant difference in children’s willingness to accept both labels between the conditions (t(38) = 2.33, p = .03). Children in the different-race speakers condition (60% of the participants) were more willing to accept both labels from the native- and foreign-language speakers than the children in the same-race speakers condition (25% of the participants). One-sample t-tests against chance showed that in the same-race speakers condition, the rate of endorsing both labels was below chance (t(19) = 2.52, p = .02), suggesting that the children more readily accepted the novel label from the native-language speaker. Such preference for the native-language label was not shown in the different-race speakers condition: children’s response did not significantly differ from chance in the different-race condition (t(19) = 0.89, p = .38).

These findings suggest that 5-year-old monoracial Korean children have race-based linguistic expectations that lead them to avoid accepting a novel label used by a same-race foreign-language speaker, while being more willing to accept the same novel label used by a different-race foreign-language speaker. This underscores the importance of understanding how mutual exclusivity can be influenced by social factors such as group membership determined by race or language, contributing to our understanding of the social cognitive processes underlying word learning. Future research will examine the developmental trajectory of these race-based expectations and how exposure to racial diversity might influence them.

Author information

Author	Role
Hyuna Lee, Yonsei University	Presenting author
Hyun-joo Song, Yonsei University	Non-presenting author

⇦ Back to session