LangDiv: The psycholinguistics of grammatical properties encoded on noun class prefixes

Knowing a language well and communicating successfully requires a sufficient breadth of vocabulary, including a lot of nouns. However, what exactly a person knows about any given noun varies from language to language and includes not just its dictionary meaning but also grammatical information about how that noun can be used. Understanding the multiple ways that languages encode noun-related information and how humans know and use that information during language communication is critical both for advancing cutting-edge, competitive technological tools such as AI chat and writing generators, and for a deeper understanding of how the brain manages language in everyday circumstances and during communication breakdown. In addition to meaning, various types of linguistic information are relevant to know for nouns, whether a noun is singular or plural or animate or inanimate, as well as other types of grammatical classes found across languages that serve to characterize nouns. One aim of this project is to innovate experimental tools for investigating what speakers know about nouns in languages. Developing such replicable scientific resources gives researchers a competitive edge in developing language technology and in mapping human cognition in the domain of spoken and written language. A second aim of this project is to determine whether certain types of noun structures facilitate noun recognition during language processing. Different languages encode important noun information in the suffixes and/or prefixes of the noun. This information is placed later or earlier relative to the noun’s meaning, respectively, and as such is predicted to have a differential effect on how these words are processed. The processing of suffixes has been well studied, but it is the grammatical information provided by a prefix that is the earliest information about a noun that someone encounters during processing. This project uses experimental tracking of human eye-movements to determine whether listeners access the grammatical information of the prefix early during processing and to investigate how listeners are able to use that information to help recognize the upcoming noun itself, thus facilitating and optimizing language processing. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.