We are delighted to announce that nine new projects have been awarded seed funding in the latest Incubator Fund round.
Posters on the funded research will be presented at future Cambridge Language Sciences Symposium events. We also hope to feature more information about these projects on the Language Sciences website in due course.
The following two projects are funded directly through the Language Science budget, supporting our core aim of strengthening research collaborations across disciplines.
Measuring and explaining individual differences in language comprehension
Dr Lucy MacGregor, Dr Adam Attaheri (MRC Cognition and Brain Sciences Unit), Dr Kondwani Mughogho (The Psychometrics Centre, Judge Business School), Dr Yasin Karatay (Cambridge University Press and Assessment) and Prof Jenni Rodd (Psychology and Language Sciences, UCL)
This project aims to achieve greater understanding of individual differences in spoken language comprehension, allowing for more targeted educational and therapeutic interventions, as well as greater theoretical insights. The team will develop a psycholinguistically-informed, psychometrically-optimised comprehension test to be administered online. The test will be made freely available.
Investigating the neural correlates of lexical flexibility in early childhood using EEG: "verbing weirds language"
Dr Alex Cairncross, Dr Chrysoula (Lina) Vassiliu, Prof Ianthi Tsimpli (Theoretical and Applied Linguistics), Dr Mirjana Bozic, Dr Jacqueline Phelps (Department of Psychology)
A fundamental question in language acquisition is how children learn and store words in a way that allows them to generalise beyond their initial use (e.g. by making a noun into a verb). Evidence for the age at which children acquire this ability is inconclusive because the design of previous studies has meant that very young children are not able to perform the tasks. This project will exploit neural signatures (ERPs) and pupillometry to examine whether children implicitly detect and process category shifts before they can actively demonstrate this ability. This would provide concrete evidence as to whether or not there is a representational change between the ages 3 and 4, or if the ability is already in place, but existing behavioural paradigms are not sensitive enough to detect it. The project also investigates how the ability to generalise across categories may relate to children’s Cognitive Flexibility (CF) skills.
Projects supported under our AI-deas project in collaboration with AI@Cam
These further seven projects are funded through our ongoing collaboration with AI@Cam, “Improving language equity and inclusion through AI.”
Comparative Evaluation of ASR Adaptation Methods for Neurodivergent Speech Patterns
Dr Calbert Graham (Theoretical & Applied Linguistics), Nathan Roll (Stanford University), Dr Luca Scimeca (Harvard University / Mila AI Institute Quebec), Qian Yang (Shanghai International Studies University)
This study aims to systematically evaluate multiple adaptation approaches for improving ASR performance on neurodivergent speech patterns. We will compare and evaluate both training-free methods (such as in-context learning and prompt engineering) and parameter-efficient training approaches (including LoRA and adapter layers) to develop practical solutions accessible to clinical researchers and practitioners.
Collecting and Annotating an English Learner Corpus for AI-assisted teaching and assessment
Dr Luca Benedetto, Dr Andrew Caines, Dr Shiva Taslimipoor (Computer Science and Technology), Prof Yongcan Liu, Dr Can Jin (Faculty of Education)
The objective is the collection, curation, and release of a corpus containing responses of both EFL (English as a Foreign Language) students and EAL (English as an Additional Language) students to reading comprehension Multiple Choice Questions, as well as an annotation of explanatory text. This corpus will feed into future projects using new technologies and AI to assist assessment and teaching in EFL and EAL settings, and support core educational values centring on the principles of equality and inclusion.
Geographic Disparities in Cognitive and Social Development in England: A Bayesian Network Analysis of the Impact of Socio-Economic Status and speaking English as an Additional Language in North East and Cambridge Children
Dr Margreet Vogelzang, Prof Ianthi Tsimpli (Theoretical and Applied Linguistics), Dr Hannah Roome, Dr Francesca De Petrillo, Dr Elizabeth Price (School of Psychology, Newcastle University)
This project will use Bayesian Networks to analyse the intricate relationships between multiple interacting variables such as SES (socioeconomic status), EAL (English as an Additional Language) and performance in school. The research will take place in 2 contrasting locations – North East England and Cambridge. Understanding the influence of socioeconomic status and language exposure on child development is critical for understanding inequalities, and for subsequently developing effective interventions and policies.
Individualised second language learning
Dr Mirjana Bozic (Psychology), Prof Brechtje Post (Theoretical and Applied Linguistics), Dr Elaine Schmidt (Cambridge University Press and Assessment)
This project aims to use digital technology to tailor language learning to the student’s needs and learning environment. This has the potential to provide accessible, flexible and inclusive access to language learning, to the benefit of both individuals and society.
Inductive and Deductive Reasoning of LLMs on Artificial Language X
Dr Yulong Chen, Prof Andreas Vlachos (Computer Science and Technology), Meiru Zhang (Modern and Medieval Languages and Linguistics), Fenghua Liu, Zhujun Jin (Theoretical and Applied Linguistics,) Hongyi Yang (Faculty of Education)
This project will advance equitable and inclusive research by developing a novel artificial language benchmark to evaluate LLMs’ ability to learn and reason beyond statistical pattern matching from its pre-training data. The team will develop an artificial language X, train a Large Language Model on it, and then analyse the LLMs' reasoning and language understanding abilities (using an artificial language prevents the LLM from leveraging pre-trained knowledge of natural languages). This research will support the evaluation of closed-source LLMs, such as OpenAI’s GPT models.
Automatic detection of developmental language disorders (DLD) with the use of ASR and AI-enabled scoring tools
Dr Penny Karanasou, Dr Kate Knill (Engineering), Dr Matt Davis (MRC Cognition and Brain Sciences Unit)
Automatic speech recognition (ASR) technologies are trained to transcribe lexically and grammatically well-formed speech, and not to record features such as repetitions, hesitations, nonwords, or grammatical errors, typically used in the assessment of Developmental Language Disorders. The team aims to fine-tune pre-trained ASR models for domain-specific data taken from individuals with DLD and typical individuals. This will allow the system to learn the new skill of transcribing speech in a way that preserves speech phenomena indicative of DLD. Once such transcriptions are available, the aim is to develop further statistical tools for detecting and characterising disorders using speech elicited in standard assessment tasks.
Bridging Human and AI Judgement in Writing Assessment: An Eye-Tracking Study
Gabrielle Gaudeau (Computer Science and Technology), Hongyi Yang (Faculty of Education), Dr Andrew Caines (Computer Science and Technology)
This study will use eye-tracking and retrospective interview to provide real-time insights into rater cognition and AI-assisted analytic scoring. By integrating perspectives from language assessment, cognitive science, and AI, the findings will contribute to the development of fairer, more transparent, and more effective analytic writing assessment systems.
About the Incubator Fund
The Incubator Fund is a small grants fund designed to foster innovative interdisciplinary research in the language sciences. It was established by Cambridge Language Sciences with additional funding from the Isaac Newton Trust, Cambridge University Press & Assessment, and the School of Technology.
As well as the opportunity to develop new ideas, collaborations and approaches, Incubator Fund projects can provide proof of concept or evidence of collaboration for larger grant applications. Other positive outcomes include knowledge exchange studentships, publications, fellowships and further career opportunities for researchers involved.
Since the Incubator Fund was established in 2016, over £150,000 of seed funding has been awarded to over 50 projects.
Please visit the Incubator Fund page for more information and to see a full list of Language Sciences Incubator Fund projects.