Bantu language shows that processing of focused information may be universal
In many languages across the world, speakers use intonation to focus on relevant information: ‘Father cooked the sauce’. Speakers may also use a specific sentence structure to mark focus: ‘It was father who cooked the sauce’. Listeners and readers process focused information in a deeper way than non-focused information. For instance, EEG studies show an increased N400 response when words are focused. The N400 is a negative brain response that typically occurs 400 ms after encountering an unexpected word, in sentences such as ‘I eat bottles’.
Makhuwa-Enahara, a Bantu language spoken in northern Mozambique, uniquely marks focus by changing the form of the verb. For instance, by changing the form of the verb ‘ate’ (o-c-aalé vs. o-hoó-cá), speakers can put focus on the noun that follows, as in ‘the woman ate rice’. If focused information is processed in the same way as in other languages, speakers of Makhuwa should show a more pronounced N400 effect when encountering focused words.
To test this hypothesis, the team recorded electrical brain activity in speakers of Makhuwa. To elicit brain responses to unexpected words, participants heard sentences such as ‘I eat bottles’ vs. ‘I eat rice’, with the crucial words in focus or non-focus position, depending on the verb form.
Results show that focused information generated a more negative N400 response than the same information in non-focus position. According to the authors, this demonstrates that regardless of how relevant information is marked, the consequences for processing of relevant information are the same.
According to director Peter Hagoort “our findings point towards a universal pattern where focus marking results in an upregulation of focused information, irrespective of how it is linguistically marked. The universality of focus marking is hence not in its linguistic form, but in the processing consequences it has. This finding is significant for both linguistic theories and cognitive models of language processing, and contributes to the diversification of cognitive neuroscience research on language by including a non-WEIRD (Western, Educated, Industrialized, Rich, and Democratic) language”.
Publication
Rinus G. Verdonschot, Jenneke van der Wal, Ashley Lewis, Birgit Knudsen, Sarah von Grebmer zu Wolfsthurn, Niels O. Schiller & Peter Hagoort (2024). Information structure in Makhuwa: EEG evidence for a universal processing account. PNAS. Doi: 10.1073/pnas.2315438121
Share this page