Babies exposed to a tonal language during their first months of life maintain a unique structural wiring in their brains as adults, even if they completely stop speaking that language before age two. A recent study published in Communications Biology found that international adoptees who forgot their birth language still possessed the neural architecture resembling that of native speakers. These findings demonstrate that early linguistic environments shape brain development in ways that persist entirely independently of ongoing practice or conscious memory.
White matter tracts are the bundles of nerve fibers that connect different regions of the brain, acting like cables that transmit information. Two of these tracts, the arcuate fasciculus and the superior longitudinal fasciculus, play a leading role in how humans process language. The arcuate fasciculus connects areas located in the temporal lobe, which process auditory sounds, to frontal regions responsible for speech production. This pathway is heavily involved in mapping the sounds we hear to the physical movements required to articulate them. The superior longitudinal fasciculus connects similar frontal areas to the parietal lobe, creating a loop that helps individuals monitor and organize the speech sounds they are trying to produce.
In most people who speak non-tonal languages like English or French, these language connections are heavily concentrated in the left hemisphere. However, tonal languages like Mandarin use pitch to determine the meaning of words. Because processing pitch generally engages the right side of the brain, speaking a tonal language requires both hemispheres to work together to integrate the sound’s tone with its linguistic meaning.
Elise Barbeau, a researcher at McGill University, along with neuroscientist Denise Klein and a team of colleagues, wanted to know how this early sensory environment alters physical brain development. Specifically, they sought to determine whether the structural changes associated with learning a tonal language are maintained if the person stops hearing and using that language entirely.
To test this, the researchers compared brain scans from four distinct groups of young people and adults living in Canada. The first group consisted of 36 international adoptees born in China who were exposed to Mandarin early in life but were adopted by French-speaking families between the ages of three months and two years. After adoption, they spoke and heard only French, with no conscious memory of Mandarin.
The second group included 26 Mandarin-French bilinguals who learned Mandarin from birth and French later in childhood. The third group was made up of 33 people who grew up speaking only French. Finally, a fourth group included 25 English-French bilinguals who had never been exposed to a tonal language. All participants were highly proficient in French and used it in their daily lives.
Using an imaging technique called diffusion-weighted magnetic resonance imaging, the team mapped the nerve fiber bundles in the participants’ brains. This specialized scanning method tracks the movement of water molecules along nerve fibers, allowing researchers to reconstruct the shape, direction, and volume of white matter connections. The researchers isolated the specific sections of the arcuate fasciculus and the superior longitudinal fasciculus responsible for language processing. They then measured the total volume of these tracts, which indicates macro-level size. They also assessed their fractional anisotropy, a metric that reveals the microstructure of the tracts, such as how densely packed the nerve fibers are and how thickly they are coated in insulating myelin.
Add PsyPost to your preferred sources The brain structures of the international adoptees closely mirrored those of the Mandarin-French bilinguals. In both of these groups, the language pathways were distributed more symmetrically across the left and right hemispheres. They also exhibited a smaller total volume in the left hemisphere tracts compared to the groups unexposed to tonal languages. The participants who grew up speaking only French, as well as the English-French bilinguals, displayed the classic pattern of highly concentrated, larger pathways strictly in the left hemisphere.
The inclusion of the English-French bilingual group allowed the researchers to isolate general bilingualism as a factor. Because the English-French speakers shared the heavily left-leaning brain structure of the French monolingual speakers, the researchers concluded that the symmetrical brain wiring was a specific response to the demands of processing a tonal language, not just learning multiple languages.
The differences between the groups were strictly related to the overall size and volume of the tracts, as the researchers did not find statistically significant differences in the microstructural density of the nerve fibers. This suggests that the early language experience changed the physical shape and layout of the connections without necessarily altering the internal makeup of the individual fibers.
The team also examined how these brain connections grew over time by comparing the tract volumes across different ages. They found that in the international adoptee and Mandarin-speaking groups, the nerve fibers in both the right and left hemispheres continued to grow in volume as the individuals aged. In contrast, the French-only speakers mostly experienced age-related growth in the left hemisphere.
For the international adoptees, this continued structural development was not tied to the age at which they were adopted, but rather to how many years they had been speaking their new language. The early tonal experience essentially set a bilateral blueprint that the brain continued to follow even as it adapted to speaking only French. This age-related growth pattern was specific to the arcuate fasciculus. The superior longitudinal fasciculus did not show the same continued volume increases over time, likely because different parts of the brain mature at different rates. The pathways connecting the temporal and frontal lobes tend to develop later in childhood than other regions, making them more susceptible to the long-term influence of early childhood environments.
While the results point toward early language exposure as the primary driver of these differences, the researchers note that ethnicity or genetics could play a role in brain anatomy. The groups exposed to Mandarin were of Asian descent, while the other groups were predominantly Caucasian. To address this, the researchers checked the total intracranial volume across all participants and found no disparities between the groups. Past studies have also shown that learning a tonal language later in life induces identical changes in Caucasian learners, making early experience the most probable explanation for the current results. Still, future research comparing genetic differences alongside linguistic backgrounds could help definitively separate these factors.
Additionally, each study group consisted of less than 40 participants, making this a small study. Larger sample sizes in future studies could help confirm the consistency of these anatomical variations across broader populations. Exploring whether this enduring neural architecture gives international adoptees an advantage if they attempt to learn a new tonal language later in life remains an open question. Tracking infant brain development over time in a longitudinal study could also provide direct evidence of exactly when these permanent physical changes take place.
The study, “Early but discontinued exposure to a language exerts lasting effects on white matter architecture in the brain,” was authored by Elise B. Barbeau, Lara Pierce, Stephanie Deschamps, Shanna Kousaie, Annie Gilbert, Jen-Kai Chen, Shari Baum, and Denise Klein.