Information about the project
Decoding Bilingual Speech Perception (Deco-B) explores how the bilingual brain represents meaning across two structurally different languages: Spanish and Basque. The central question is whether bilinguals use separate language-specific semantic systems or a shared, unified semantic representation. To investigate this, the project combines advanced neuroimaging techniques (MEG and fMRI) with transformer-based multilingual AI models. Twenty young adult Spanish–Basque bilinguals will listen to natural speech in both languages while their brain activity is recorded across multiple sessions. A specially designed bilingual speech corpus will support high-resolution neural decoding. The decoding models will be trained on brain activity from one language (e.g., Spanish) using semantic features extracted from pre-trained language models (such as GPT-2, XLM-R, mT5, and LaBSE). The models will then be tested on the other language (Basque) to assess cross-language generalization. The reverse direction (training on Basque, testing on Spanish) will also be examined to evaluate symmetry in bilingual semantic mapping. To integrate the spatial precision of fMRI and the temporal resolution of MEG, the project will use advanced multimodal fusion techniques, including cross-modal transformers, shared latent space models (e.g., Deep CCA, multimodal VAEs), and contrastive learning approaches inspired by CLIP. Through this interdisciplinary approach, Deco-B aims to uncover whether meaning in the bilingual brain is language-specific or shared, offering new insights into the neural architecture of bilingual semantic representation.
Job description
One full-time predoctoral researcher with a background in computational neuroscience will be hired for a four-year period. The researcher will work under the direct supervision of the Principal Investigator and will play a key role in implementing and adapting machine learning models for the decoding of bilingual speech and neural data.
Primary responsibilities will include developing and optimizing multimodal analysis pipelines integrating MEG and fMRI data, extracting and aligning linguistic features from transformer-based language models, and contributing to cross-language decoding and model evaluation. The researcher will also support data preprocessing, quality control, and documentation, and collaborate with other members of the project to integrate shared decoding frameworks and explainability tools.
The position offers training in advanced neuroimaging analysis, multilingual AI modeling, and interdisciplinary research at the intersection of cognitive neuroscience and machine learning. The researcher will be encouraged to contribute to scientific publications, conference presentations, and open science practices associated with the project.
The contract is full time and includes an end-of-contract payout in accordance with institutional and funding regulations.
PI and research group:
Nicola Molinaro is Group Leader of the Brain Rhythms and Cognition at the Basque Center on Cognition, Brain and Language (BCBL). During his career he has spearheaded cutting-edge research that bridges neurobiology and language, leveraging advanced EEG and MEG neuroimaging techniques to unravel the temporal dynamics of speech and cognition. The Brain Rhythms and Cognition group explores how the brain encodes visual, auditory and linguistic rhythms by focusing on neural oscillatory activity. Based on this approach we investigate predictive processing in language comprehension, music processing and visual and attentional processes. These research lines merge into the more general goal of detecting oscillatory neural components that lead to the development of language disorders across the lifespan.
Required skills:
Desirable skills:
Develops integrated language, communication and environment skills, especially in an international context.
Salary: 21.000€ gross annual salary
https://www.bcbl.eu/en/join-us/what-is-like-to-work-bcbl
Training opportunities and Career development plan:
Researchers at any stage of their career, regardless of their contractual situation, are given an opportunity for professional development and for improving their employability through access to a Personal Career Development Plan which includes
(1) Training through individually personalized research projects under senior supervision
(2) Exchanging knowledge with the scientific community and the general public
(3) Network-wide training in theory and methods
(4) Complementary training courses
(5) Involvement in proposal writing, task coordination
(6) Development of skills for the organization of training and scientific events
BCBL seeks to foster an environment where all talents can flourish, regardless of gender, sex, sexual orientation, age, sociocultural background, nationality or impairments. We are proud to have in place both an LGTBIQ+ Plan and a Gender Equality Plan to actively promote inclusion and equal opportunities. If you have any questions relating to accessibility or support, please contact us.
Language policy
Submission of the application and documentation:
To submit your application, please follow this link: applying for “Artificial Intelligence" and attach the following documentation: by May 10th, 2026:
Learn more about the BCBL’s OTM-R policy
Application process timetable:
Contact details for enquiries:
For technical problems please contact: hr@bcbl.eu
For inquiries or questions related to the content of the position and/or information about the research group please contact: n.molinaro@bcbl.eu
Inicia sesión para acceder a opiniones reales, valoraciones anónimas e información sobre sueldos antes de enviar tu solicitud.