Back to Projects
signedRhyMED

Multimodal Entrainment and Dyslexia: A cross-linguistic investigation of rhythm using explainable machine learning

Programme: HORIZONScheme: HORIZON-TMA-MSCA-PF-EF
EC Contribution

€260K

Duration

02 Nov 202601 Nov 2028

Consortium Size

2

organizations

Objective

RhyMED investigates whether difficulties in rhythmic synchronisation of gesture are a core feature of dyslexia that complement the rhythmic speech entrainment difficulties already identified, and whether these motor difficulties are similar across languages and modalities. The project tests adults with and without dyslexia in English, Spanish, and Basque, which differ in rhythm shaping features such as syllable structure, stress patterns, and pitch use. It moves beyond mono-language perception studies by examining how rhythms are produced and aligned in speech, co-speech gesture, and motor movements. The central objective is to quantify rhythmic entrainment within and across modalities and to compare patterns between dyslexic and control groups. RhyMED extracts signal-based measures of amplitude and frequency modulations in speech and comparable modulation patterns in gesture and motor movement, capturing rhythm at multiple timescales. It then assesses how closely these patterns align within individuals and across groups. An explainable machine learning approach is used to evaluate how well rhythm metrics distinguish groups and to identify which features matter most, without treating the models as a black box. This project will clarify whether rhythmic differences in dyslexia are language-blind or linguistically constrained, and whether gesture and movement provide converging evidence for rhythmic entrainment deficits in dyslexia alongside speech. The project delivers reproducible methods for analysing rhythmic entrainment in natural production and a robust foundation for future research on early identification and intervention. A secondment at the Basque Center on Cognition, Brain and Language strengthens cross-linguistic testing and training. For the fellow, RhyMED advances training in the neuroscience of rhythm and language, translating existing technical skills into applied studies of dyslexia and delivering impact consistent with MSCA priorities.

AI Analysisclaude haiku

Click “Summarize” to get an AI-powered analysis of this project.

Call Topics

HORIZON-MSCA-2025-PF-01-01