BILINGUAL LEXICAL AND SYNTACTIC REPRESENTATIONS AND LANGUAGE CONTROL

Most people nowadays have notions of more than one language. Many think that being a bilingual means being fluent in two languages, but bilingualism is in fact a continuum, from only knowing a few words in another language to being fully proficient in two languages.

But any knowledge of other languages means knowing more words, more grammar and more sounds than speakers of a single language. How are the two languages represented in the mind? For words, some theories say that there are separate lexicons for each language, and other say that there is a single lexicon; in both cases, words would be tagged to indicate language membership. There are also words that are similar across languages (cognates), and they seem easier to say and understand. But where exactly does this facilitation come from?

For grammar, theories say that it is shared across the two languages of bilinguals, and it contains structures that are the same across languages and structures that are different across languages (and those are tagged for language membership).

But how do the two language systems interact in processing? Bilinguals switch languages whenever they want to, but also manage to speak in a single language without saying words in the wrong language. Such mistakes (they happen very rarely) are prevented by language control mechanisms that are there to ensure that bilinguals don't say anything in the wrong language because they won't be understood. Such control can happen by making less accessible (for example, by inhibiting) the language that is not being used in the situation, or making more accessible the relevant  language.  In either case, it seems that the more dominant the language that is not needed, the more it suffers from the application of control (because the better we speak a language, the more likely it is to interfere when it is not needed, so more care needs to be taken to make sure it doesn't interfere). 

There are many open questions about bilingual language control. What is the mechanism? Is it applied once or over and over? Is it global (operating over a whole language) or local (operating only over translation equivalents, for example, over "perro" [dog in Spanish] when a speaker wants to say "dog")? (Some evidence suggests that both options happen to some extent.) How does its application depend on the conversation partner, the speaker's proficiency in the two languages, and their language history (such as which language they speak more often, how long ago they spoke each language, which language they speak in what situation, and how often they code-switch)? Is the control mechanism unique to bilinguals, or is it used to control which word is said when speaking a single language (for example, to avoid saying "chair" instead of "table", or using the wrong register)? How does control of lexical selection interface with control of grammar? For example, if a language is inhibited as a whole, what happens to the shared structures? 

And, this is all about language production, but what about comprehension? Much evidence suggests that both languages of bilinguals are activated during comprehension (e.g., comprehension is language non-selective). Consistent with this, we have shown that hearing a word in the wrong language doesn't throw off bilinguals as much as hearing something that doesn't make sense, in either language. 

Projects in progress:

Relevant publications: