Acoustic Indicators of Cognitive Load during Language Use in Local Multilingual Speech Communities − An Investigation of Saterland Frisian, Low German, and High German in Saterland

• Applicant: Professor Dr. Jörg Peters
• Subject Area: General and Comparative Linguistics, Experimental Linguistics, Typology, Non-European Languages
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 553833753

Saterland Frisian, spoken in the northwest of the district of Cloppenburg in the German state of Lower Saxony, is the last remaining variety of East Frisian and one of the most endangered minority languages in Europe. Historically, it never had significantly more than 2500-3000 speakers. The current threat arises from the fact that Saterland Frisian, which is spoken almost exclusively with family members, neighbors, and close friends, competes with Low and High German and shows a dramatic decline in its intergenerational transmission. The project is looking for new ways to assess the vitality of Saterland Frisian, focusing on factors that may hinder its adoption as a family language. It starts from the assumption that, for a minority language acquired in a multilingual context, language competence alone does not guarantee its intergenerational transmission. The cognitive load associated with its use can be considered an equally important factor. If speakers of a minority language remain proficient in this language but encounter a higher cognitive load when using it compared to other languages, they are less likely to choose this language as their family language. To measure the cognitive load associated with using Saterland Frisian and its contact languages, acoustic measures of voice quality and oral fluency are collected, which have been shown to be reliable indicators of cognitive load in psychological research and in research on multilingual and foreign language use. Special attention is given to the relationship between cognitive load and the speakers’ age, language competence, language dominance, and social and linguistic ties to other members of the local speech community. Another innovative aspect of this project is its approach to data collection. Instead of conducting a representative survey and recruiting participants through public calls, which primarily attract speakers who are confident that they have sufficient language competence, the aim is to systematically tap into family networks that may be spread across one or more households. This procedure allows for capturing a broader range of linguistic competence levels. Additionally, it helps to identify structural characteristics of the family networks and of the communicative networks that emerge from the interactions among network members in one or more languages. Through this research, the project will provide valuable information for future language protection measures. Additionally, it will provide an extensive speech corpus that can subsequently be used to train AI-supported dialogue systems for educational and revitalization purposes.

• DFG Programme: Research Grants
• Co-Investigator: Dr. Heike Schoormann