When Esti Blanco-Elorrieta’s team asked bilingual Spanish-English speakers to turn “barco” into its plural form, they weren’t just testing vocabulary—they were peering into the brain’s deepest linguistic machinery. Using magnetoencephalography (MEG) at New York University, the researchers captured brain activity with millisecond precision as participants applied grammatical rules to real words, cognates like “idea/idea,” and even entirely made-up pseudowords like “paple.” What they found reshapes our understanding of how bilingual brains work: there isn’t a separate grammar engine for each language. Instead, a single, shared neural system handles grammatical computation across languages, regardless of whether the word is familiar, borrowed, or invented.
This discovery matters because it challenges long-standing assumptions about how our brains organize language. For decades, scientists have debated whether bilinguals maintain parallel linguistic systems—one for each tongue—or rely on overlapping neural architecture. The NYU study, published in JNeurosci, offers compelling evidence for the latter. By tracking how brains process pluralization in both English and Spanish, and extending the test to pseudowords, the researchers showed that grammar isn’t locked to specific languages or sounds. It’s a universal computational tool, flexible enough to apply rules to any linguistic input.
The study involved Spanish-English bilinguals at NYU, with Xuanyi Jessica Chen, a doctoral student, leading data collection and analysis. When participants heard singular nouns—like “boat” or “barco”—their brains activated the same neural patterns to generate plurals, even when the words had no real meaning. That consistency across real and invented words suggests grammar operates as an abstract, reusable function, not a set of language-specific rules etched in separate brain regions. The implications stretch beyond bilingualism: if grammar is a universal computation, it may be part of a deeper cognitive toolkit that supports all human language learning.
Blanco-Elorrieta, an assistant professor of psychology and neural science, sees this as a window into the brain’s efficiency. “The results provide some of the clearest neural evidence to date that grammatical computations are shared across languages in bilingual speakers,” she says. This could help explain why people who know two languages often pick up a third more easily—they’re not building a new system from scratch, but expanding an existing one. Looking ahead, the team hopes to explore how this single grammar engine develops in children and whether it’s affected by age of language acquisition or proficiency level.
In a world where over half the population speaks more than one language, understanding the brain’s unified approach to grammar offers more than scientific insight—it reveals a fundamental unity in human communication.