Poster #106 - Examining Semantic Word Knowledge in Spanish-English Monolinguals and Bilinguals

• In Event: 3-053 - Poster Session 10
  In Poster Session: PS 10 Section - Language, Communication

Integrative Statement

INTRODUCTION
By the second year, children leverage semantic knowledge to learn new words: 18-month-olds are more likely to learn a new word from a well-known semantic category (i.e., more likely to learn cat than chair if they know more about animals than furniture; Borovsky et al., 2017). We examine lexical-semantic development as a function of single and dual language exposure. Young bilinguals organize word-to-object mappings across two languages and these mappings emerge in the second year of life (e.g., Singh, 2014; De Anda, Hendrickson, Poulin-Dubois, Zesiger, & Friend, 2016). We extend this research examining the composition of early lexical-semantic networks. Specifically, we ask:

1) Do monolinguals and bilinguals differ in the semantic composition of vocabulary?
2) Does semantic overlap across languages predict vocabulary in the second year?

METHOD
Participants were bilingual Spanish-English (N = 32) and monolingual Spanish (N = 32) and English toddlers (N = 41) at 18 and 24 months. Receptive vocabulary in English and Spanish was assessed on the MacArthur-Bates Communicative Development Inventory (MCDI, Fenson et al., 2006). The MCDI is comprised of six semantic categories for early-acquired nouns: food, drink, clothing, vehicles, body parts, and clothing. We examined the distribution of vocabulary across these categories. In bilinguals, we also calculated semantic overlap (the relative proportions of words understood) in English and Spanish in each category (Figure 1).

RESULTS
A Group (Monolingual/Bilingual) X Language (English/Spanish) X Semantic Category ANOVA yielded a significant Group x Language interaction (F(1, 116) = 9.33, p = .003) at 18 months. Follow up Tukey comparisons showed that monolingual and bilingual children differed in the composition of English, but not Spanish, vocabulary (95% CI = (-.36, -.06), p = .002, see Figure 2). At 24 months, a significant Group x Language (F(1, 81) = 4.66, p= .04) interaction was observed but follow up comparisons did not reach significance.

In an ANCOVA controlling for age and dominant language, semantic overlap was associated with vocabulary in both languages. A three-way Semantic Overlap X Age X Dominant Language interaction (F(1, 25) = 7.89, p = .01) revealed a positive association between semantic overlap and English vocabulary in English-dominant children that was strongest at 18 months. Conversely, semantic overlap was positively associated with Spanish vocabulary in Spanish-dominant children (Semantic Overlap X Dominant Language interaction, F(1,25) = 8.12, p = .009). Semantic overlap also predicted the number of translation equivalents at 24 but not 18 months of age (Semantic Overlap X Age interaction: F(1, 31) = 13.86, p = .0008).

CONCLUSION
Vocabulary composition in English varied as a function of single versus dual language exposure. Further, larger overlap in semantic knowledge across languages predicted larger dominant language vocabulary at 18 and 24 months. Finally, greater semantic overlap predicted a greater likelihood that words mapped onto a single concept by 24 months. These findings corroborate recent bilingual first language acquisition findings (i.e., Singh, 2014; De Anda et al., 2016) and a model of language development in which growth across languages is supported by a shared semantic space modulated by exposure and language dominance.

Authors

• Stephanie De Anda, University of Oregon
  Presenting Author

• Margaret Friend, San Diego State University
  Non-Presenting Author