Investigating the learning of translation equivalent words: A case study of two bilingual infants’ lexicons

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Language, Communication

Abstract

One of the major milestones in early language development is word learning –mapping a word to the meaning. This process has been considered even more complex for children who are exposed to more than one language. Researchers have proposed how the initial word mapping process takes place for infants who are learning words in the bilingual context. It has long been suggested that learning a word (e.g., bunny) prevents another word (e.g., rabbit) to be mapped for the same or similar meaning (Haryu, 1998). However, interestingly, some recent bilingual work indicates potential benefit of learning a word in one language (e.g., bunny) for learning of another word (e.g., conejita) in another language (e.g., Spanish). This faciliatory role has been explained as that the challenging aspect of language acquisition is extracting the meaning from the environment and recognize the meaning independent of the context, so once the meaning is learned via first label, adding another label to that learned meaning is trivial process (Bilson et al., 2015). In the present case study, we aim to evaluate these accounts by using diaries from 2 bilingual (Japanese-English) infants whose parent recorded every word that was produced by the children for over 100 of their first words. These recordings include date of the entry, word, grammatical category, meaning by dictionary, and the context in which each word was used. For each diary data set, we made direct comparisons between the time when meanings/ideas were labeled in each language and documented the word level analyses of how words from two languages are related in time and meaning. The key finding is that learning a label from one language facilitates the word learning in a second language when we look at early learned translation equivalent (TE) words – words spoken in both languages. The correlation for learning in a word in each language was 0.75, which suggests that the second language label was produced fairly quickly after the first language label was produced. Also, when we looked at these TE words’ lexical category type, there were significantly more proportion of non-nominal categories’ (e.g., adverb, verb) entries. This suggests that learning of more abstract meaning may benefit the most from first label learning. Additionally, when translational equivalent words were compared to non-translation equivalent words overtime, the result suggest that the proportion of non-translation equivalent words was greater (84.5%) than the number of translation equivalent words (15.5%). Compared to previously reported rate of TE, 20-25% (Nicoladis & Secco, 2000; Woods, 2013). these rates of translation equivalent words was even slower, indicating the case study participants’ TE is considered to be lower. One important implication is that the difficulty associated with early vocabulary development for young children in bilingual learning context may not necessary because of having multiple labels for a similar/same meaning. A case study approach can further contribute to our understanding of how each word child learn may lead to the next learning—fundamental for mechanism underlying language development.

Authors

• Christina Diaz Griep, University of Houston
  Presenting Author

• Hanako Yoshida, University of Houston
  Non-Presenting Author