Script differences hinder language switching in bilinguals

Researchers at the HSE Centre for Language and Brain used eye-tracking to examine how bilinguals switch between languages in response to context shifts. Script differences were found to slow down this process. When letters appear unfamiliar—such as the Latin alphabet in a Russian-language text—the brain does not immediately switch to the other language, even when the person is aware they are in a bilingual setting. The article has been published in Bilingualism: Language and Cognition.

Bilinguals can differentiate between the target and non-target languages based on the global language context. In an environment where everyone speaks Russian, a bilingual fluent in both Russian and English will experience suppression of their English. In this context, Russian becomes the target language, while English remains the non-target language.

The researchers aimed to investigate how switching between target and non-target languages occurs when the global language context changes and to test the hypothesis of proactive gain control in bilinguals. This hypothesis suggests that as exposure to a non-target language increases, bilinguals activate its vocabulary to facilitate faster use. It is important that such control is exercised proactively rather than in response to already received information, and this helps prevent processing delays.

To test this hypothesis, the researchers conducted an experiment with a sample of 50 adult Russian–English bilinguals. Participants were asked to read several sentences on various topics displayed on a computer screen and answer related questions. Initially, participants' target language was Russian, but the researchers gradually altered the language context to make English the target language. Initially, Russian was the only language spoken to participants, and they were given task instructions in Russian. In the second stage, English sentences were introduced alongside Russian. In the third stage, English sentences were removed, but an English-speaking instructor joined the participants. In the final stage, participants were exposed solely to English sentences, and the experimenter spoke only English to them.

As participants performed the tasks, their eye movements were recorded, specifically the duration of gaze fixation on a word, the number of regressions to previous words, and any skipped words. It is known that longer fixations indicate greater difficulty in processing a word, while regressions suggest a need to revisit the word to better understand its meaning.

'The invisible boundary paradigm was used to measure access to the non-target language. While a participant was reading a sentence in Russian, an English translation of a word in that sentence briefly appeared before the target Russian word was displayed,' the authors explain.

For example, in the Russian sentence 'You will need to complete certain training to obtain a permit,' the English word training appeared before the Russian word for ‘training.’

The researchers hypothesised that if access to the English language was activated, the duration of fixation on the Russian word after the appearance of its English translation would be shorter. However, this hypothesis was not confirmed. Despite the gradual introduction of English elements in the experiment, changes in the linguistic context did not influence early lexical access.

'Probably, the difference between the Cyrillic and Latin scripts is too significant, causing the brain to immediately recognise the mismatch and automatically suppress the non-target script. Additionally, it is possible that the immersion in the non-target language was not prolonged enough for its activation,' the researchers suggest.

Thus, the study findings confirmed the key predictions of the Multilink and BIA+ models, namely that lexical processing in a bilingual environment is influenced by both bottom-up and top-down factors. However, bottom-up factors dominate when the scripts differ.

The bottom level involves automatic input processing, where the brain first recognises letters, followed by words, and then their meanings. In the case of different scripts, such as Cyrillic and Latin, the brain may struggle to recognise the letters of the non-target script. The difference significantly impacts how quickly and efficiently a person can switch between languages.

The top level involves conscious input processing, which relies on the global context and prior experience. For example, if a person knows they are in a bilingual environment, this awareness can activate an expectation in their brain of encountering a word in the second language. However, this process requires more time and effort.

The researchers plan to conduct experiments involving deeper immersion in a non-target language. 'We expect that after a certain point, we will observe an increase in the speed of cross-language switching,' explain the authors.

The findings from this and potential future studies may be valuable for developing strategies to teach foreign languages, particularly reading skills, by considering the cognitive load involved in suppressing the native language, script differences, and the duration of immersion in the language environment.