Brendan Tomoschuk

Present and Past Research Topics

Language interference in third language learning

Many learners report interference between a new language they are trying to learn and other foreign languages they already have some ability in, even if the languages aren't related at all. Is there a cognitive reality behind this claim? If so, where would the interference stem from? Does the brain lump all non-native languages together to prevent from interfering in the native language? Or is it just that a learner doesn't have enough experience switching between second and third languages to keep them separate? I investigate these questions by studying people who speak three languages, and by using a computer program to teach words in a third language to people who speak two languages.

Self assessment in bilingual language proficiency

Cognitive scientists often use a subject's self-assessment in their research, and bilingualism researchers are no different. We frequently ask bilinguals to rate their own proficiency in both of their languages to use as a metric for an experiment, but are self-assessments trustworthy? Would one bilingual rate himself or herself similarly to another? If they do, are they consistent between bilingual populations (e.g Spanish-English and Chinese-English bilinguals)? How do these potential differences impact the results of our bilingualism research? See this work here.

Modeling error production in early second language learners

Using data collected from thousands of Duolingo users, we built a classification model of the types of errors learners make in the early stages of a new language. We've found that many well-studied cognitive and linguistic effects help build a simple memory-effcient model of English speaking learners of Spanish and French, as well as Spanish speaking learners of English. This work was done as part of the Duolingo shared task, and can be found here.

Input diversity in second language word learning

Language classes often teach us words in semantic groups. One chapter will teach us all about food and groceries, while the next will cover clothes and shopping. Evidence actually suggests that semantically blocking words in a new language impairs learning, and that diversifying the semantic information we receive in word learning is beneficial for improving retention. I investigate through use of several lines of experiments where participants learn words in a new language from a computer program. By manipulating things such as speaker diversity (i.e. many different instructors vs. one), and context diversity (many different kinds of sentences vs. few) we can see what other kinds of diversity affect language learning. We can even ask questions like "Does using your native language during learning, as opposed to immersion, impact learning a new one?"

Accented speech and lexical access

People can adapt rapidly to understanding accented speech. Bilinguals, in particular, tend to hear a lot of accented speech within their communities. How does listening to accented speech as a bilingual affect their ability to access one language? Does listening to Spanish accented English make switching to Spanish easier? We investigate these questions with experiments in which participants listen to accented speech in both languages, and try to access their lexicon by naming pictures in one of their two languages.

The neural underpinnings of color perception in bilinguals

There is some evidence that the language a person speaks can affect their perception. For example, Russian speakers can more easily distinguish between shades of blue than English speakers, and this is thought to be because they have two terms for the same color range that English speakers call blue. I used event related potentials (ERPs), a measure of EEG that is time locked to behavior, to investigate the neural underpinnings of color perception processes in speakers of two languages, each with a different set of color terms.