Development of cross-language lexical influence

Abstract

Bilinguals are often not fully monolingual-like in either language. With respect to the lexicon, recent research demonstrates that their naming patterns for common household objects tend to converge on a common pattern for the two languages. The present study investigates the developmental trajectory of naming of common household objects in Dutch/French bilingual and monolingual children. First, we investigated whether bilingual word diversity for a set of household objects is limited by the demands of learning two languages. We found that children lag behind monolingual controls in terms of vocabulary at young ages, but that they catch up later, ending with as diverse a set of names in each language as the monolinguals. Second, we investigated how the convergence in the adult bilingual lexicon manifests itself over the course of development. We found that naming patterns converge with age following a similarity-driven strategy, a pattern also seen for the monolinguals. However, language-specific exceptions to the similarity principle are acknowledged from age 10 onward by monolinguals, but only from age 14 onward in bilinguals. At all ages, bilinguals show more convergence than monolinguals, and the difference is largest for adults. Together our results indicate that acquisition of naming patterns by bilinguals starts off more or less following the early stages of monolinguals, with separate naming patterns in the two languages, but convergence dominates the later developmental path to a larger extent for bilinguals than for monolinguals.

Keywords:

• bilingualism
• word knowledge
• language development
• monolingual and bilingual lexicon

Acknowledgements

We thank Marieke Buffel, Jochem Raats, and Ellen Seys for their help in gathering the data and Farah Djalal, Loes Stukken, and Steven Verheyen for their help in some of the analyses. Any opinions, findings, and conclusions or recommendations expressed in this material are ours and do not necessarily reflect the views of the National Science Foundation.

Notes

1. The rated proficiency of the bilingual children in both languages is analyzed with a bifactorial ANOVA, including age (5-, 8-, 10-, 12-, and 14-year olds) and language (Dutch and French). The analysis revealed that, on average, the children are more proficient in Dutch than in French, F(1, 202) = 17.72, p < .01, and that the language groups differ in proficiency, averaged over the two languages, F(4, 202) = 3.60, p < .01. The interaction of age and language is not significant, F(4, 202) = 0.32. Tukey's testing further revealed that the proficiency ratings of the 5-year olds are significantly larger than those of the 10- and the 12-year olds, but that no other pairwise comparisons among the age groups yield significance.

Additional information

Funding

This research was partly supported by the Leuven University Research Council [grant number OT/10/024], awarded to Gert Storms, and by National Science Foundation [grant number 1057885] to Barbara Malt and Ping Li.