Skip to main content

Research

Meeting on « Research on Language and Childhood in Grenoble: who does what?"

Encounter/Debate Research On 14 December 2023

Saint-Martin-d'Hères - Domaine universitaire

Complément lieu

room A006 Annie Genovese, Michel Dubois building.

To exchange information and bring together those involved in this area.

Program:
13:30-14:00: Coffee/Tea and first discussions
14:00-16:00: Who's doing what? Each participant presents his or her research topics, projects and collaborations in no more than 3 minutes.
16:00-16:30: Discussion: what collaboration/communication tools should be put in place?
After 16:30: Coffee/tea and further discussions

For further information, please contact:
Anne Vilain: GIPSA-lab
Mathilde Fort, Aude Noiray, Hélène Lœvenbruck: LPNC

Understanding Cognition

Research

In the field of memory, our metacognitive abilities, i.e. our ability to evaluate and predict our memory performance, are clearly the result of a progressive adjustment. For example, we have shown that early childhood is characterised by a highly optimistic increase in memory capacity. This is only modulated when the child is confronted with reality. It is not until adolescence that our metacognition approaches real performance.
We are also exploring metacognition in various neurological pathologies in collaboration with the CMRR at Grenoble University Hospital.
 
Financements : 2 bourses Ecole Doctorale EDISCE
Chercheurs : Chris Moulin, Céline Souchay, Audrey Mazancieux, Méline Devaluez

Comprendre la métacognition

Trans 3

Equipe Développement et Apprentissage, Research

The general challenge is the improvement of fundamental learning in elementary school, namely reading and oral English.

The general challenge is the improvement of fundamental learning in elementary school, namely reading and oral English. The objectives of the TRANS3 project are 1) to finalize the development of the applications created and evaluated within the framework of the FLUENCE project so that they are resolutely adapted to classroom use; 2) to allow them to be widely and sustainably disseminated in French-speaking schools, including those in the departments of Guyana and Mayotte; 3) to promote and support their use in the classroom by creating educational guides, training and additional resources for teachers; 4) to test their usability/acceptability in classes, as well as the impact of their use on student progress and on teachers’ teaching practices; 5) to consolidate the scientific knowledge associated with these tools and disseminate them on a large scale in the teaching world, in order to anchor the evolution of educational practices on theoretical frameworks and solid scientific data. These ambitious objectives will be made possible thanks to the continued collaboration of all the authors of the FLUENCE applications with new university partners, but also with the inclusion in the consortium of a digital publisher (HumansMatter) and an educational publisher. (Éditions Hatier), each having solid experience in their field of expertise and determined to collaborate to achieve these objectives. They will participate in the creation of the tools, ensure their promotion and dissemination on a large scale and beyond the project, according to an economic model that is transparent for all stakeholders. The commitment of the rectorates of Grenoble, Mayotte and Guyana guarantees the participation of these academic territories in the evaluation of the impacts linked to the use of the applications. This project will fit into the favorable framework of the Pégase pilot center (PIA3 funding, pilot center for teacher training and research for education). The first year will essentially be devoted to the co-construction, with a few user-tester teachers integrated into the team, of all the necessary tools: version 1 of the applications, the teacher interface, additional resources and content and training materials. Year 2 will allow us to test the effectiveness and usability of these tools in a set of diverse schools, to continue to adapt them as best as possible to practice and to refine the means of dissemination. The scientific research initiated in FLUENCE will also continue throughout the duration of the project and will be promoted in publications and training for teachers. The wide use of the 3 applications in French-speaking schools should contribute to the improvement of educational differentiation practices and the reading and English skills of students.

Trans 3

Consortium

4 laboratoires : 3 à l'UGA (LPNC, GIPSA-LAB et LIDILEM) et 1 à l'INSA-Lyon (LIRIS)
2 partenaires industriels : Humans Matter et les Editions Hatier
3 rectorats : de Grenoble, de Mayotte, de Guyane

Responsable du projet

Responsable du projet : Marie-Line Bosse (PR, LPNC)

Site internet du projet: https://trans3.cnrs.fr
Porteur du projet : UGA

SEMO

Equipe Langage, Research

Sensory-Motor Integration for Speech Rehabilitation in Patients with Post-Stroke Aphasia

Monica BACIU
Hélène LOEVENBRUCK
Doctorante: Célise HALDIN


During a stroke, language-related areas of the brain can be affected, leading to aphasia. Of particular interest to us is expressive or non-fluent aphasia, which occurs after frontal lesions. These patients have the image of the word to be spoken, but cannot express it correctly, even though the speech organs are functioning perfectly. Aphasia requires re-education to regain the ability to speak. As a first step, the speech therapist will try to restore the failing function, or will propose methods that use the intact skills. Speech therapy rehabilitation reaches its limits fairly quickly after about six months. This explains the need for new programs and methods.

In this context, the new technological approaches developed in recent years, based on illustration and visual feedback, offer innovative therapeutic perspectives. In this project, we propose to evaluate the effectiveness of a rehabilitation program inspired by this type of approach, to improve speech in patients with non-fluent aphasia. Our method is based on the reinforcement of the interaction between perceptual and motor representations, thanks to the innovative Ultraspeech device. We will exploit a fundamental psycholinguistic principle, which postulates that speech is based both on the activation of the system controlling the motricity of effectors related to word articulation (action) and on the auditory or visual representation of words (perception). It is precisely this sensory-motor interaction that is impaired after a stroke in the frontal regions, and which explains the difficulties in producing words.

The sensory-motor interaction method that we propose allows the patient to perceive phonemes and visualize on a computer screen the movements of the tongue and lips previously recorded by a healthy speaker, typically a speech therapist. Through repeated exercises, the patient is trained to produce sounds correctly, using the correct pronunciation and articulatory movements of the reference speaker as a model. We will compare patients who will follow a classical speech and language therapy rehabilitation program to another group who will follow an ‘enriched’ rehabilitation program including rehabilitation based on sensory-motor interaction associated with speech and language therapy. In order to judge the favorable effect of the rehabilitation program including sensory-motor integration, the following measures will be considered: (a) language skills with speech therapy assessment, (b) phonemic quality with speech flow analysis, (c) inner speech abilities with an introspective questionnaire and a behavioral task and (d) cognitive function with neuropsychological assessment. Brain language networks will be evaluated with neuroimaging. The impact and spin-offs of the project are aimed at: basic and clinical research to better understand the basics of speech and aphasia; aphasic patients who need new methods and approaches to improve their disability; and health professionals who need better training and wish to use new methods from experimental research to treat patients.

Projet SEMO

ANR SAM-Guide

Equipe Corps et Espace, Research

The Sam-Guide project is developing assistance systems for blind people, through sensory substitution: moving around, locating a target using sounds and vibrations. It unites the forces of researchers from five laboratories.

Univ. Grenoble-Alpes

LPNC, Grenoble Chambéry
Christian Graff, Psychophysique
GIPSA-Lab, Grenoble
Sylvain Huet, Architecture système
Denis Pellerin, Traitement d'images

Univ. Normandie

LITIS, Rouen
Edwige Pissaloux, Physique
Marc-Aurèle Rivière, Sciences cognitives
CERREV, Caen
Bruno Mantel, Sciences du mouvement
Elise Faugloire, Ergonomie


Ecole Polytechnique

CMAP, Palaiseau
François Alouges, Mathématiques appliquées
Sylvain Ferrand, Sciences de l'ingénieur



In Grenoble, the LPNC (Body and Space team), in collaboration with the GIPSA-Lab (COPERNIC team), we are developing the AdVIS System, a virtual prototype allowing you to reach a target object without vision. The prototype is built on a choice of spatial metrics accessible from an artificial sensor, to be matched with a choice of percepts initiated by sound and tactile stimuli. The effectiveness and comfort of the human-machine interfacing resulting from these choices are tested on typical participants (blindfolded) mainly in a controlled environment, in the laboratory and virtually. In a hot-cold game, they seek to hit a 3-D virtual target in a motion-captured space. The deviations in angle and/or distance from this target are sent in real time according to the position of the index finger seeking to aim and then touch the target

At the University of Normandy, CERREV (Caen) in collaboration with LITIS (Rouen) are developing a tactile belt which indicates through vibrations on the skin around the user's waist their position in relation to beacons placed on their path .

In Palaiseau, the CMAP of the Polytechnique school is developing spatialized sound algorithms making it possible to place a "flying" sound source in front of the person's head as a virtual guide such as a "Tinkerbell" or "Navi" fairy. This fairy replaces a human assistant who normally runs in front of her in the corridor of an athletics stadium track. The culmination of this 24-month project is a festive disabled sports event modeled on the “Laser Run”, which is two combined modern pentathlon events. The person (blind, or deprived of vision) who runs on a circuit, approaches a shooting station with a table on which he must find and take in hand a laser gun, to shoot three times at a target before resume, possibly in relay, the rest of his race on the circuit. Performance requires coordination of the systems developed in the three sites. The main issue is not a hypothetical immediate improvement in the living conditions of these people, but a leisure activity; in fact, competitive play is a situation which stimulates the development of devices, then adapted to everyday life. 

At NU, we guide a person along markers using a tactile belt At UGA, we guide by sound towards a 3D virtual target In Saclay, we guide a person on a racing circuit using spatial sound Bottom: Coordination will allow in Laser Run (modern pentathlon), without eyes, to run on a circuit, find the weapon, aim at the target...

Voir les publications dans le portail HAL ANR

 

Projet SAM Guide

Coordinateur & Partenaire

Coordinateur : Christian Graff (LABORATOIRE DE PSYCHOLOGIE ET NEUROCOGNITION)

Partenaires :
BORELLI Centre Borelli
LITIS LABORATOIRE D'INFORMATIQUE, DE TRAITEMENT DE L'INFORMATION ET DES SYSTÈMES - EA 4108
CesamS CENTRE D'ETUDE SPORT ET ACTIONS MOTRICES
LPNC LABORATOIRE DE PSYCHOLOGIE ET NEUROCOGNITION
GIPSA-lab Grenoble Images Parole Signal Automatique
CMAP Centre de mathématiques appliquées

 

Projet-ANR-21-CE33-0011

à partir d'Octobre 2021 - 48 mois

RASMUSSEN

Equipe Langage, Research

Multimodal Assessment of Neurocognitive Functioning and Brain Reorganization After Hemispherotomy in Patients With Rasmussen Encephalitis

Monica BACIU
Marcela PERRONE-BERTOLOTTI
Christine BULTEAU
Doctorante: Anna BORNE


Rasmussen encephalitis is a chronic autoimmune disease characterized by progressive unilateral hemispheric atrophy. This pathology causes drug-resistant partial epilepsy and is accompanied by progressive, disabling neurocognitive disorders. Due to its drug resistance, the only curative treatment is hemispherotomy, a functional disconnection between the hemispheres. Given the young age of surgery, patients with Rasmussen are expected to benefit from significant brain reorganization that allows significant recovery of cognitive functions, even if it occurs unevenly, depending on the specific developmental trajectories of these functions. and individual cognitive and cerebral reserve. In addition to the cognitive recovery observed in adulthood, different models of brain reorganization have also been described. Therefore, it is of particular interest to understand the strategies recruited by these patients due to neuroplasticity. Overall, this research project aims to evaluate both the cognitive outcome and the reorganization of brain networks in adult patients with Rasmussen encephalitis after hemispherotomy performed during childhood. A multimodal approach will be used, combining experimental psychology and neuropsychology to assess a wide range of cognitive functions (language, executive functions, theory of mind and memory) and clinical scores, as well as fMRI in resting state and DTI-MRI to measure the connectivity of functional and structural networks of the functional hemisphere, respectively. The originality of this work also consists of the development of a new multi-cognitive battery specifically adapted to this population (LEXTOMM, Perrone-Bertolotti, M., 2021). We thus wish to improve our understanding of patients with Rasmussen encephalitis by describing cognitive and cerebral phenotypes, with new directions for cognitive rehabilitation and pre-habilitation to ensure their best possible recovery.

Projet RASMUSSEN

Pégase

Equipe Développement et Apprentissage, Research

Pilot center for teacher training and research for education

The Pégase center aims to transform teaching practices from kindergarten to high school to strengthen the learning of fundamental knowledge (oral and written language, mathematics/computer science, respect for others) and thus contribute to reducing social, territorial and social inequalities. cognitive. The Center was designed as an ecosystem closely associating the laboratories of UGA and USMB, the INSPE and Rectorates of Grenoble and Guyana, and the entire teaching community. This ecosystem is collaborative, distributed, open and learning. It works to place the “evidence-basededucation” approach at the heart of the initial and continuing training of teachers, to promote their professional development by relying on data from research.

 

Read more   

Pôle Pegase

CNRS PRIME : Projet de recherche interdisciplinaires multi-équipes : DeepL_IRMf

Equipe Langage, Research

DeepL_IRMf : Modèle pré-entraîné deep-learning pour l'imagerie par résonance magnétique fonctionnelle

L’imagerie par résonance magnétique fonctionnelle (IRMf) est devenue un outil essentiel pour évaluer l’activité neuronale liée aux fonctions cognitives. La majorité des études d’IRMf comprennent de petits ensembles de données, ce qui empêche d’appliquer de nouvelles approches pour l’analyse des données telles que l’intelligence artificielle (IA), et plus particulièrement les réseaux neuronaux convolutifs (CNN) qui nécessitent de grands ensembles de données pour atteindre une efficacité optimale. Le CNN est une méthode d’IA incluse dans l’apprentissage profond (DL), un type d’apprentissage automatique (ML) qui imite la façon dont les humains acquièrent des connaissances. L’apprentissage par transfert (TL) est récemment apparu comme une solution couramment utilisée pour surmonter la rareté des données d’IRMf tout en utilisant l’apprentissage profond. Le TL peut être réalisé grâce à des modèles pré-entraînés pour extraire des caractéristiques afin de répondre à une tâche particulière. L’objectif de notre projet est de développer un modèle générique pré-entraîné pour l’analyse de données IRMf, entraîné sur de grands ensembles de données, pour réaliser la TL. Nous proposons une approche basée sur un transformeur entraîné avec les valeurs de l’activité des voxels IRMf. Un pipeline utilisant un CNN 3D pour extraire les caractéristiques spatiales des données IRMf 4D, suivi d’un transformateur alimenté par les données enchâssées résultant du CNN pour modéliser l’aspect temporel des données, sera développé

Projet soutenu par la MITI Mission pour les Initiavives Transverses et Interdisciplinaire (CNRS)

 
Partenaires du projet
INSB
Monica BACIU
Laboratoire de Psychologie et Neurocognition
(UMR 5105) Grenoble, France
 
INSB
Martial MERMILLOD
Laboratoire de Psychologie et Neurocognition
(UMR 5105) Grenoble, France
 
INS2I
Sophie ACHARD
Laboratoire Jean Kuntzmann
(UMR 5224) Grenoble, France



En savoir plus : https://miti.cnrs.fr/prime/deepl-irmf/

 

Projet Deepl_IRMf

LUM_INTERMOD

Equipe Langage, Research

Inter-cognitive and transcognitive modeling of multimodal biomarkers, with artificial intelligence methods. Application to the unified language-union-memory framework (L∪M)

Monica BACIU
Martial MERMILLOD
Sophie ACHARD

 

This research work is located at the interface between language and declarative memory and addresses the question of their interactive union in a connectomic, multimodal and integrative perspective, using artificial intelligence methods. Integrating multimodal data to elucidate the processes and neural networks underlying cognition and behavior represents a major challenge in cognitive neuroscience. This is associated with a current paradigm shift, with recent neurocognitive models which consider that human behaviors are made possible by complex interactions between cognitive functions. Indeed, these new models assume that complex cognitive systems are shaped by interactions between processes and that functional integration and specialization are supported by a modular architecture. Given the complexity of these connectomic models which reveal new properties, not detectable if traditional classical models are considered, it is fundamental to combine multiple sources of data, both behavioral, cognitive and cerebral. Indeed, the combination of multimodal biomarkers which reflect the multiple facets of cognition and the brain will allow new properties of the system to emerge, intrinsically multimodal and invisible from a monomodal perspective. This is precisely the objective of this project, to consider language and declarative memory in this new interactive framework.
 
Projet LUM_INTERMOD

Babylab Languages and Music

Equipe Langage, Research

2017-2020

 

Art-Science-Parenting Project

 

Funded by IDEX UGA Scientific and Cultural Outreach, in the wake of a previous project launched in 2017 and entitled “The Languages of Babylab”, Languages and Music of Babylab articulated scientific research and artistic creation to resonate the research of artists and scientists from Grenoble working on the development of language and communication in infants. Scientifically, he proposed evaluating the impact of musical interventions on infant language development. On an artistic level, the artists engaged in creative research nourished by interactions with babies. Finally, in terms of parenting, families participated in discussions with artists and scientists to discuss and understand the links between music, language and awareness. Thanks to the mediation of the Mediarts association, the project made it possible to establish links between artists and researchers and to forge long-term partnerships with early childhood stakeholders.

RÉALISATIONS
- 91 artistic residency sessions in 9 different Isère territories with 4,446 audiences. - 10 concert-workshops bringing together 995 spectators, including events as part of Brain Week 2020, rescheduled for the 2022 edition. - 1 professional training “Creation and imaginary language at the service of little ones” conducted today by the Training Center for Performing Musicians (Université Lumière Lyon 2)

Project members :
Mathilde Fort
Hélène Lœvenbruck (porteuse)
Olivier Pascalis

Other IDEX partners - Academic partners :
Anne Vilain, Nathalie Henrich, GIPSA-lab (CNRS-Grenoble INP-UGA) ;
UGA (Direction de la Culture et de la culture scientifique)

External partners :
Christelle Pillet et Christophe Monge, association Médiarts ; Bertille Puissat et Myriam Roulet (musiciennes) ;
crèches Léa Blain et Mosaïque et auditorium de la Source, ville de Fontaine ;
Espaces Petite Enfance de l’agglomération grenobloise ; Musicien·ne·s du Labobascule (musique) ;
Simon Barral-Baron (vidéo).

languesetmusiquesdebabylab_enfant.jpg
Atelier BabyLab / © Simon Barral-Baron - projet LMBabylab

Other links
http://mediarts38.fr/2019/05/05/festival-musiques-et-langues-de-babylab/

Projet LMBABYLAB
Subscribe to Research