Simulating code-switching using a neural network model of bilingual sentence production

Chara Tsoukala, Mirjam Broersma, Antal van den Bosch, & Stefan L. Frank

Abstract

Code-switching is the alternation from one language to the other during bilingual speech. We present a novel method of researching this phenomenon using computational cognitive modeling.
We trained a neural network of bilingual sentence production to simulate early balanced Spanish-English bilinguals, late speakers of English who have Spanish as a dominant native language, and late speakers of Spanish who have English as a dominant native language. The model produced code-switches even though it was not exposed to code-switched input. The simulations predicted how code-switching patterns differ between early balanced and late unbalanced bilinguals; the balanced bilingual simulation code-switches considerably more frequently. Additionally, we compared the patterns produced by the simulations to two corpora of spontaneous bilingual speech and identified noticeable commonalities and differences. To our knowledge, this is the first computational cognitive model simulating the code-switched production of unbalanced bilinguals and comparing the simulated production of balanced and unbalanced bilinguals to that of human bilinguals.

Keywords: code-switching, computational cognitive modeling, sentence production, Bilingual Dual-path model, recurrent neural network