Methodological Excellence in Data-Driven Approaches to Linguistics

Project Information

MEDAL

Grant agreement ID: 101079429

DOI

10.3030/101079429

Project closed

EC signature date 21 July 2022

Start date 1 January 2023

End date 31 December 2025

Funded under

• Widening participation and spreading excellence

Total cost

€ 1 050 000,00

EU contribution

€ 1 050 000,00

Coordinated by

TARTU ULIKOOL
Estonia

Periodic Reporting for period 1 - MEDAL (Methodological Excellence in Data-Driven Approaches to Linguistics)

Reporting period: 2023-01-01 to 2025-12-31

Summary of the context and overall objectives of the project

Methodological Excellence in Data-Driven Approaches to Linguistics (MEDAL) created a consortium comprising four highly acclaimed research organisations: the University of Tartu (UTARTU) in Estonia; the Max Planck Institute for Psycholinguistics (MPI), a German research institute located in the Netherlands; the Donders Institute at Radboud University (RU), also in the Netherlands; together with an associated partner, the University of Birmingham (UB) in the United Kingdom. The consortium, led by UTARTU, aimed to increase the global competitiveness of Linguistics in Europe by bringing together world-class pioneers and early-career researchers (ECRs) to develop joint excellence in methodological training and promoting a model of intensive, international collaboration in cross-linguistic, cross-modal and cross-disciplinary research.

Knowledge transfer between the partners and between different generations of researchers was facilitated via numerous training events in linguistic methods, focussed on three distinct but interconnected topics: (1) corpus linguistics, (2) experimental methods, and (3) computational modelling, as well as training events developing skills in Open Science practices and academic career-building, and events targeting institutional capacity building. Alongside these activities, ECRs were given the opportunity to implement this knowledge in the GoLD (Generalising over Language Data) MEDAL research project, which followed the same three main topics in linguistic methods. By the end of the project in December 2025, the MEDAL consortium had established a sustainable joint training and research programme and was poised to seamlessly continue this collaboration. In addition, UT had gained the recognition of an international Linguistics research excellence hub.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

By the end of the project, MEDAL (Methodological Excellence in Data‑Driven Approaches to Linguistics) delivered a highly visible programme of international collaboration and capacity building that strengthened competence in data-driven linguistic methodology, with a particular focus on ECRs, via three annual Summer Schools and 24 training events such as workshops and master classes in research and career skills.

Flagship Summer Schools covered corpus linguistics, experimental linguistics, and computational modelling, creating a structured methods training programme across the project lifetime in order to equip current PhD students and early career researchers with crucial skills for driving research and innovation in the field for years to come. In the first year of the project, the lead partner UTARTU hosted a Summer School in Corpus Linguistics, and in the second year, the partners in Nijmegen (MPI in Psycholinguistics and RU) co-hosted a Summer School in Experimental Linguistics. The Summer School in Computational Methods took place in Birmingham (organised by the associated partner UB) in the project's final year. MEDAL finished with a conference: “MEDAL Final Conference: New Challenges, Novel Approaches” in Tartu on 9-10 October 2025, hosted by UTARTU. Training events were organised across the consortium, with an effort to make as many workshops as possible available to a wide group of interested ECRs within and beyond the consortium. Training workshops on methodological tools and skills for linguists, statistical analysis, and career skills, such as writing grant proposals, practising Open Science, and responding to reviewers, were organised as in-person, online and hybrid events.

Project-related training materials were disseminated as widely as possible on OSF (project MEDAL) and YouTube (#MEDAL2325). Engagement was promoted on social media platforms, such as Instagram (#MEDAL2325).

MEDAL also delivered a structured package of mobility and mentoring that ensured training translated into real collaboration, research practice, and long-term networks. Altogether, 130 mobility activities took place, supporting attendance at training events, mentoring sessions, collaboration stays, and institutional development exchanges, and 27 guest lecture visits. In order to establish a sustainable mobility exchange program between UTARTU and the other MEDAL partners beyond the scope of the EC-funded Twinning project, an Erasmus+ exchange agreement (with RU) and a memorandum of understanding (with UB) were established.

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

State-of-the-art research studies were carried out in the GoLD (Generalising over Language Data) MEDAL research project on three key topics: (1) optional ‘redundant’ subject pronoun expression, which was investigated through a cross-linguistic corpus study (involving Estonian, Finnish, Russian, and Polish datasets) followed by two experimental eye-tracking reading studies, (2) a cross-linguistic investigation of the mechanisms underlying language acquisition, or more specifically, the separability of vocabulary, morphology and syntax in acquisition of Estonian and English, with pre-registered hypotheses, and (3) a cross-linguistic comparison of negation in multimodal corpora. Datasets for all the studies are available on OSF and all publications will be made available through Open Access journals.

Each of these studies brought together ECRs with pioneers in the field to tackle currently relevant issues with novel methods. Redundancy in communication was examined from a multi-methodological vantage point, using both cross-linguistic corpus analysis and two eye-tracking reading experiments, one tracking participants with Estonian as a first language, and the other with Estonian as a second language. The language acquisition strand employed a novel statistical method to investigate the separability of vocabulary, morphology and syntactic knowledge in English and Estonian, a highly inflected language, combining parametric and non-parametric item response theory models applied to existing data from communicative development inventories. Finally, cross-linguistic comparison was used to probe differences in co-speech gestures in multimodal corpora for Estonian and Turkish. Open dissemination of research, training and outputs extend the benefits of MEDAL beyond the immediate participants and support reuse.

project-of-the-year-in-humanities-in-university-of-tartu-f-and-t.jpg