Format du dépôt |
Fichier |
Type de dépôt |
Article dans une revue |
Titre |
en
Active inference as a unifying, generic and adaptive framework for a P300-based BCI
|
Résumé |
en
Objective. Going adaptive is a major challenge for the field of Brain-Computer Interface (BCI). This entails a machine that optimally articulates inference about the user’s intentions and its own actions. Adaptation can operate over several dimensions which calls for a generic and flexible framework.
Approach. We appeal to one of the most comprehensive computational approach to brain (adaptive) functions: the Active Inference (AI) framework. It entails an explicit (probabilistic) model of the user that the machine interacts with, here involved in a P300-spelling task. This takes the form of a discrete input-output state-space model establishing the link between the machine’s (i) observations – a P300 or Error Potential for instance, (ii) representations – of the user intentions to spell or pause, and (iii) actions – to flash, spell or switch-off the application.
Main results. Using simulations with real EEG data from 18 subjects, results demonstrate the ability of AI to yield a significant increase in bit
rate (17%) over state-of-the-art approaches, such as dynamic stopping.
Significance. Thanks to its flexibility, this one model enables to implement optimal (dynamic) stopping but also optimal flashing (i.e. active sampling), automated error correction, and switching off when the user does not look at the screen anymore. Importantly, this approach enables the machine to flexibly arbitrate between all these possible actions. We demonstrate AI as a unifying and generic framework to implement a flexible interaction in a given BCI context.
|
Auteur(s)
|
Jelena Mladenovic
1
, Jérémy Frey
2
, Mateus Joffily
3
, Emmanuel Maby
4
, Fabien Lotte
1
, Jérémie Mattout
4
1
Potioc -
Popular interaction with 3d content
( 179935 )
- 200, avenue de la Vieille Tour
33405 Talence cedex
- France
-
Laboratoire Bordelais de Recherche en Informatique ( 3102 )
;
-
Université de Bordeaux ( 259761 )
;
-
École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB) ( 300366 )
;
-
Centre National de la Recherche Scientifique UMR5800 / URA1304 ( 441569 )
;
-
Inria Bordeaux - Sud-Ouest ( 104751 )
;
-
Institut National de Recherche en Informatique et en Automatique ( 300009 )
2
Ullo
( 487838 )
- 40 Rue Chef de Baie
17000 La Rochelle
- France
3
GATE Lyon Saint-Étienne -
Groupe d'Analyse et de Théorie Economique Lyon - Saint-Etienne
( 102550 )
- 93, chemin des Mouilles 69130 Écully
6, rue Basse des Rives 42023 Saint-Étienne cedex 02
- France
-
École normale supérieure de Lyon ( 6818 )
;
-
Université Lumière - Lyon 2 ( 33804 )
;
-
Université Claude Bernard Lyon 1 ( 194495 )
;
-
Université de Lyon ( 301088 )
;
-
Université Jean Monnet - Saint-Étienne ( 300284 )
;
-
Centre National de la Recherche Scientifique UMR5824 ( 441569 )
4
CRNL -
Centre de recherche en neurosciences de Lyon - Lyon Neuroscience Research Center
( 139719 )
- Centre Hospitalier Le Vinatier, Bâtiment 462 Neurocampus Michel Jouvet, 95 boulevard Pinel, 69500 Bron
- France
-
Université Claude Bernard Lyon 1 ( 194495 )
;
-
Université de Lyon ( 301088 )
;
-
Université Jean Monnet - Saint-Étienne ( 300284 )
;
-
Université de Lyon ( 301088 )
;
-
Institut National de la Santé et de la Recherche Médicale U1028 ( 303623 )
;
-
Centre National de la Recherche Scientifique UMR5292 ( 441569 )
|
Date de publication |
2020
|
Volume |
17
|
Page/Identifiant |
016054
|
Langue du document |
Anglais
|
Nom de la revue |
|
Vulgarisation |
Non
|
Comité de lecture |
Oui
|
Audience |
Internationale
|
Domaine(s) |
-
Sciences de l'Homme et Société/Méthodes et statistiques
-
Informatique [cs]/Modélisation et simulation
-
Informatique [cs]/Biotechnologie
-
Informatique [cs]/Intelligence artificielle [cs.AI]
-
Informatique [cs]/Apprentissage [cs.LG]
-
Informatique [cs]/Traitement du signal et de l'image [eess.SP]
-
Sciences cognitives/Neurosciences
|
Financement |
-
This work was financially supported by BCI-Lift Inria project, and the French National Research Agency (ANR-11-LABX-0042, ANR-11-IDEX-007 and ANR-18-CE28-0016).The data used in this study were acquired as part of the French ANR project ANR-DEFIS09-EMER-002 CoAdapt
|
Projet(s) ANR |
|
Projet(s) Européen(s) |
-
BrainConquest
- Boosting Brain-Computer Communication with high Quality User Training
Numéro CORDIS :
714567
|
Mots-clés |
en
P300 speller, Brain-Computer Interfaces BCI, Active inference, Electroencelography
|
DOI |
10.1088/1741-2552/ab5d5c |