MSCA-PF 2025: Sorbonne University is France’s top university for the number of projects selected

This project, ARCADE, will address the pressing issue of digital distraction by delivering a neuro-computational model of attentional capture and task reconnection. It will be the first model of its kind to integrate multidimensional reward with task switching cost and effortful cognitive control.
This model will be built and validated with a novel experimental task that contains realistic distraction, and uses eyetracking with pupillometry to predict with more accuracy than the current approaches the re-engagement time. 
The impact of the project will contribute to the understanding of how attention and decision making interact during task selection and will inform policies for healthier digital environments. 
 

This MSCA-funded project investigates the environmental consequences of increasing lithium release into the ocean, particularly in the context of human activities and environmental change. 
The project at Sorbonne Université focuses on marine biogeochemical processes. 
Two main themes will be addressed:
 

1. The interaction between seawater lithium and marine microorganisms (e.g., diatoms) under controlled laboratory conditions to evaluate potential ecological impacts. 
2. The influence of climatic conditions on lithium incorporation during biomineralization in marine calcifying organisms (e.g., corals).
   The objective is to better understand lithium cycling in marine ecosystems under environmental and climatic changes.

Xenon (Xe) is a noble gas that is used in dark-matter detectors, in propulsion for space exploration, and in effective anaesthetic and other therapeutic properties. 
The CrystXeMin project aims to systematically investigate reactivity of xenon with common minerals under Earth crust conditions while utilizing crystallographic knowledge to mitigate instrumental limitations. The project will unveil natural minerals that can react with xenon and can be used to capture and recycle this gas. Furthermore, an understanding of Xe concentration mechanisms can lead to the location of deep sources of this gas, facilitating its use for technological advances. 

Spotting a tiny moving insect in a cluttered scene is a complex computational challenge. Yet insectivorous animals tackle it with disconcerting ease. 
This project aims to understand which neuronal behaviours enable the detection of insects by these animals, depending on the visual context through the retinal circuit. By studying the mouse’s hunt for crickets—one of its favourite prey items—he aims to uncover the mechanisms that enable it to detect, track, and capture these prey items in the wild.

LIPTO aims to understand how bacteria in the gut microbiota physically interact with dietary fats at the level of lipid droplets.
Through the use of anaerobic microfluidic chips, live confocal imaging and gene expression reads, it will measure and visualise the formation of metabolical hotspots in biofilms. LIPTO will provide direct visual insight into this process and translate them into quantitative, predictive tools to better understand lipid metabolism and its potential health implications.

Immune cells face significant geometric constraints within tissues, which influence the development of their defence function. Their ability to move effectively within these tissues is therefore essential.
This project aims to understand the physical mechanisms that enable them to adapt their behaviour and optimise their dynamics in these complex environments.

OCEAN EARS is a research project which transform decades of recordings into a living, interactive, geospatial map containing immersive sonic information of whale and ocean sounds. It aims to deepen public empathy for marine life—especially whales—through hyperreal immersive listening experiences that simulate being virtually underwater and sensing their acoustic reality. 
By bringing together marine bioacoustics, artificial intelligence and virtual reality audio, it will create an open, searchable archive and a spatialised listening platform, revealing both the richness of whale communication and the growing impact of man-made noise in the oceans.
A novel browser-based application and physical installations will extend the experience through infrasound transducers, allowing audiences to feel the ocean’s vibrational energy. The project contributes to public understanding and engagement with science, biodiversity protection. 

This project reconstructs the intriguing cultural representations shaping relations between human communities and bird populations in the Greco-Roman world, through a detailed study of the Ixeutika, a unique classical text devoted to bird hunting.
By integrating philology with eco-ethology and environmental history, it recreates the cultural ecosystem reflected in the treatise and It recreates the cultural ecosystem of the text and the interplay between knowledge, practices and human–animal dynamics. 
In this way, it contributes to environmental humanities by offering an eco-historical reconstruction of the interactions between humans and the rest of the living world during Antiquity.

The project aims to build the first biologically grounded model of visual cortical responses that incorporates both the nonlinear processing of the retina and the effects that eye movements have on cortical activity during natural vision. 
By combining paired recordings from the retina, thalamus and cortex — collected from macaques and mice — within hierarchical models, it will seek to show how retinal activity shape cortical responses. 
This project will provide a better understanding of visual processing under natural conditions and how visual information is processed in the early stages of the visual system.

SAGL aims to produce a complete study of the sexual vocabulary that Greeks of the past included in their own lexica. 
Through a series of keyword searches in the Thesaurus Linguae Graecae, a comprehensive corpus of glosses on sexual terms will be compiled to analyze simultaneously from two perspectives: on the one hand the origin, history, usage, functions, and implications of each term individually. On the other hand, what our glosses say about the relationship of the lexical tradition with sexuality, about its sources, and its taboos. 

This project, TSADA, aims to enhance the explainability, transparency, and efficiency of Automated Design of Algorithms (ADA). 
The project seeks to develop techniques to improve the robustness of ADA whilst reducing the human effort and computational power required to develop high-performance AI systems.
By providing systematic comparisons and practical recommendations for both expert-guided and LLM-driven algorithm design approaches, as applied to satisfiability solvers, TSADA contributes to a more trustworthy, transparent, and sustainable algorithmic innovation.