Impact of inter-language typological difference on metacognition in language tasks by bilinguals

Funding: Alexander von Humboldt Foundation
Principal investigator: Leona Polyanskaya (University of Göttingen)
Funding period: 2022-2023

The project studies how metacognition in language tasks is affected by individual linguistic experience. Metacognition allow tracking one’s own cognitive performance and estimate the likelihood of an error while making decisions. My earlier studies showed that bilinguals have better metacognition than monolinguals in those tasks which activate mechanisms engaged in speech processing. In such tasks, bilinguals can detect the cases when the probability of an error is higher, and adapt their speech processing strategies accordingly. Importantly, bilinguals do not constitute a homogeneous population. Bilinguals differ in age of acquisition, dominance in one of the languages, code switching behavior, literacy in one versus both languages. We still do not know what aspects of bilingualism enhance metacognition. In this project, I will test the hypothesis that typological differences between languages in bilinguals’ inventory modulate metacognition. I predict that phonological differences enhance metacognitive processing on prosodic tasks (e.g., rhythm discrimination), morpho-syntactic differences improve tracking performance on grammar learning (e.g., statistical learning or grammaticality judgment), and vocabulary differences hone metacognitive monitoring in inferential word learning. Bilinguals whose languages do not exhibit typological differences in phonology or syntax do not exhibit enhanced metacognition in language tasks, when all other bilingualism-related factors (age of acquisition, language dominance, etc) or individual differences (e.g., in IQ, in verbal fluency, in socio-economic parameters) are controlled for. It is important to study a modulatory effect of bilingualism on metacognition because metacognition is related to decision making at the group level and thus can influence societal decision. Bilingualism is becoming a norm in contemporary society, and we need to be aware which aspects of bilingualism interact with metacognitive processes.

Belief states of vulnerable groups in crises in Latin America: sociolinguistic and computational assessment (BeVuLa)

Funding: DFG, COVID-19 Focus Funding
Principal investigators: Olga Kellert and Stavros Skopeteas (University of Göttingen)
Funding period: 2021-2023
Indigenous people belong to the particularly vulnerable groups in the COVID-19 era are disproportionally affected by epidemics and other crises, as acknowledged by the United Nations. Beyond the general problems related to the socio-economic marginalization and the concomitant inaccessibility of health-care services (in particular in rural regions and remote communities), a major threat for indigenous people arises through miscommunication, either due to the sparsity of information material in indigenous languages or due to cultural differences hindering the interpretation/application of the recommended health measures; see recent reports on Latin American indigenous people in. Dissemination of reliable COVID-19-related information, adapted to cultural and linguistic background of indigenous peoples is a major priority in the current crisis.

The major aim of the present project proposal is to determine what role the linguistic and cultural background of indigenous people from Latin America plays in dissemination of relevant information about COVID-19 propagated by high-impact news outlets (henceforth the Reference Corpus). This aim entails the development of interdisciplinary methods (including sociolinguistic and computational linguistic metrics) for assessing belief states and measuring the degree to which these belief states mirror the Reference Corpus. Special attention is given to complement questionnaire-based data collections with more socially natural data-gathering methods (e.g. free interview), which are particularly important in order to include individuals who are less accustomed to performing highly controlled tasks.

In order to reach the major aim of the project proposal, we created an interdisciplinary group of experts, already cooperating in various related projects. The group combines expertise in linguistic analysis and sociolinguistics, social and medical anthropology, computational methods in social sciences and sociolinguistics, statistic modelling, bilingualism research in Latin America and research of indigenous languages in Latin America, in particular Ecuador, Peru and Yucatán.

Based on the results of this study, the project partners envisage to submit a larger proposal with the contribution of experts from further regions for the study of the role of indigenous languages and cultures in the dissemination of information with social relevance.

RTG 2636: Form-meaning mismatches

Funding: DFG
Principal investigators: RTG website: participating researchers
Researchers: RTG website: PhD students
Funding period: 2021-2025
Website: RTG website

Ergativity and Information Structure: Comparing Chibchan languages (CHIBERGIS)

Funding: ANR-DFG
Principal investigators: Claudine Chamoreau (CNRS, SeDyL), Stavros Skopeteas (University of Göttingen), Elisabeth Verhoeven (Humboldt-Universität zu Berlin)
Researchers: Marie Benzerrak (University of Göttingen), Jana Bajorat (Humboldt-Universität zu Berlin), Natalia Caceres (CNRS)
Funding period: 2021-2024
Linguistic research in the last decades has established substantial knowledge about complex grammatical phenomena at several layers of the grammar. A challenge for the current linguistic research is to understand the linking between these layers of linguistic structure. This project investigates two layers of clause structure: alignment and information structure as well as the interaction between these layers. The project targets are motivated by the current research and collaboration of a network of French and German linguists sharing a common interest in the creation of empirical and analytical tools for the description of endangered languages. It will provide deeply annotated corpora and empirical studies on a language family that is understudied and underrepresented in linguistic research, namely Chibchan (the members being dispersed in the geographical area from Honduras in Central America to Colombia and Venezuela in South America). The core team of the project will examine three languages in detail (Pesh, Cabécar, Ika) and the project partners will conduct comparative studies in further languages, in order to obtain a detailed picture of the micro-variation within this language family. The methodological approach combines exploratory research based on corpora and confirmatory research through field experiments. This combined approach will provide insights into the different types of field data and will yield a sustainable resource for future study of the Chibchan languages. The research goals include (a) a detailed structural investigation of the syntactic and semantic properties of ergativity, (b) a study on information structure focusing on the role of microvariation in word order possibilities, and (c) a precise examination of the role of information structure in languages with optional ergative marking, which is a phenomenon that has been observed for several languages but is not yet studied or understood in detail. The project teams integrate a large group of linguists in France and Germany and will enhance their existing collaborations with cooperative measures in research and in the qualification of PhD students that will lead to collaborative products and will create a scientific network.

The syntax of ellipsis and information structure in Old Irish: a typological approach

Funding: DFG
Principal investigator: Elliott Lash (University of Göttingen)
Funding period: 2020-2023

Yucatec Maya: Variation in Space and Time

Funding: DFG-CONACyT
Principal investigators: Barbara Blaha Pfeiler (Universidad Nacional Autónoma de México, Merida, Yucatán, Mexico), Stavros Skopeteas (University of Göttingen), Elisabeth Verhoeven (Humboldt-Universität zu Berlin)
Researchers: Flor Canche Teh (UNAM, Merida), Yidong Yu (University of Göttingen), Nico Lehmann (Humboldt-Universität zu Berlin)
Funding period: 2018-2022
Linguistic variation is crucial for our understanding of the dynamic processes that manifest themselves in language use. The development of varieties along different dimensions such as time, geographical or social space reflects different communicative practices of a speech community. Varying linguistic properties are the source of language change while dependencies between them contribute to our understanding of linguistic structures.

The aim of this project is to study variation in the indigenous Mexican language Yucatec Maya with a focus on the spatial dispersion of the language and the current developments in language use. The investigation of variation in well-studied languages such as English, Swedish or German has considerably advanced with the creation of databases, atlases and computational tools and theoretical models. A large-scale study on a language such as Yucatec Maya promises a significant contribution to the understanding of variation in two respects: (a) it will offer insights about the amount of variation that arises if a language is mainly used in oral communication; (b) it will contribute to our understanding of the impact of the transformational processes in modern societies (especially urbanization and generalized bilingualism) on the situation of indigenous languages.

This project will compile a large-scale data collection of a sample of lexical and grammatical features in a sample of rural and urban settlements of currently spoken Yucatec Maya. It will identify dialectal variants, it will create a detailed atlas, and will draw conclusions about the existence and dispersion of varieties in geographical space. The comparison between rural and urban settlements will offer insights about developments in the time axis, such as the recent creation of regiolects in urban centers and/or the impact of generalized bilingualism. The use of current dialectometric methods will allow for precise estimations of the relations between varieties.

In order to understand the roots of variation, the project will investigate the relations between variants in selected phonetic, phonological and syntactic phenomena. The planed studies are devoted to core properties of Yucatec Maya that are in the focus of the current research. At the phonetics/phonology level, we will study the variation in the realization of ejectives, lexical tones, as well as the alignment of enclitics with prosodic domains. In syntax, we will investigate basic word order patterns and argument-focus constructions. These studies are expected to shed light on the linguistic structures that determine the observed variation.

This agenda is expected to make a contribution to the current research on variation by combining knowledge from several research paradigms, including sociolinguistics, dialectometry, studies on micro-variation in phonology and syntax.

Focus and thematic role assignment: A comparison of Hungarian and German in Child Language Comprehension (FoTeRo)

Funding: DFG, SPP 1727: New Pragmatic Theories based on experimental evidence (
Principal investigators: Pia Knoeferle (Bielefeld University, Humboldt-Universität zu Berlin), Stavros Skopeteas (Bielefeld University, University of Göttingen)
Funding period: 2015-2019
The aim of this project is to investigate pragmatic inferences from information structure to thematic role assignment. The fact that information structure contributes to role disambiguation, in addition to syntactic and morphological cues (such as word order and case) is established in psycholinguistic and typological research. However, we know very little about listeners principles in mapping information structural categories, such as topic and focus, to particular arguments and their corresponding thematic roles. This project will collect empirical data with the goal of developing a precise pragmatic account of this process during language comprehension. Two theoretical positions will be compared: (a) Hearers use information structural preferences (e.g., topics are subjects and foci are objects) as defaults for thematic role assignment; (b) alternatively, information structural cues motivate pragmatic inferences about why the speaker uses canonical or non-canonical word orders. In order to examine the predictions of these hypotheses, we will conduct experimental studies (visual word paradigm) in two languages that crucially differ with respect to how focus is interpreted (as an object in German, and as a subject in Hungarian). We examine how (differences in) the syntax and prosody of these languages implicate inferential processes that lead to different thematic role assignments. By examining data from several stages of language development (in both languages), the project will gain insights into (a) the developmental stages in the recognition of particular cues (e.g., prosody and case), (b) the contribution of these cues to the pragmatics of role assignment; and (c) children's emerging pragmatic competence in this process. The results of this project will be integrated into a psycholinguistic model of situated language processing (Coordinated Interplay Account).

Ancient Indo-European languages for the 21st century

Funding: Internationalization of Curricula, University of Göttingen (IdC)
Principal investigators: Götz Keydana (University of Göttingen), Stavros Skopeteas (University of Göttingen)
Researcher: Saverio Dalpedri (University of Göttingen)
Funding period: 2018-2021

Study program “Language in Cultural Context: Documentation, Description, Application”

Funding: DAAD; Fachbezogene Partnerschaft mit Hochschulen in Entwicklungsländern.
Principal investigators: Stavros Skopeteas (Bielefeld University, University of Göttingen), Firmin Ahoua (Université Félix Houphouet Boigny, Abidjan, Côte d'Ivoire)
Funding period: 2015-2018