Neurophysiological Markers for Passive Brain–Computer Interfaces

Raphaëlle N. Roy 1, 2 Jérémy Frey 3, 4
2 GIPSA-SAIGA - SAIGA
GIPSA-DA - Département Automatique, GIPSA-DIS - Département Images et Signal
4 Potioc - Popular interaction with 3d content
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest
Abstract : "Passive" BCIs use brain signals involuntarily generated by individuals. Those signals can be used for mental state monitoring in divers work situations such as hazardous ones (e.g. driving, plant monitoring, etc.), or more generally in order to improve human-computer interaction, notably in the entertainment industry. Through several neuroimaging techniques, the present chapter will detail what the common cerebral markers used by passive BCIs are, and which associated mental states are currently assessed. Thanks to their ease-of-use outside the laboratory, electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) are the two techniques that are mainly used to build passive BCI systems. EEG is a direct measure of brain activity that allows us to record onto the scalp the fluctuations of neuronal electrical activity. On the other hand, fNIRS is an indirect brain activity recording method that measures neuronal metabolic activity as it probes blood oxygenation. Passive BCIs aim to monitor various mental states, such as attention, mental fatigue, workload, error detection and emotions to name a few. For that purpose, BCIs make use of signal characteristics of different kinds. For example, there are spectral makers (e.g. power in a frequency band), or temporal ones (e.g. potentials triggered by certain events), and even spatial ones (e.g. relative activation of specific brain areas). There are different ways to put passive BCIs into practice depending on the assessed mental state. For instance, the recording of drivers' lapses of attention can help to prevent accidents, while measuring workload - which increases with the amount of data to process will allow to adjust the difficulty of a task. Error-related potentials, generated when we realize that a mistake was made, could be used by machines to get to know when they acted wrong. Using passive BCIs, it is also possible to characterize the emotions that are produced by media, and their intensity. Whether it be to prevent hazard or facilitate communication, the growing use of passive BCIs promises significant technological progress.
Type de document :
Chapitre d'ouvrage
Maureen Clerc; Laurent Bougrain; Fabien Lotte. Brain–Computer Interfaces 1: Foundations and Methods, Wiley-ISTE, 2016, 978-1-84821-826-0. 〈10.1002/9781119144977.ch5〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01413462
Contributeur : Jérémy Frey <>
Soumis le : vendredi 9 décembre 2016 - 17:54:24
Dernière modification le : vendredi 24 novembre 2017 - 15:37:59

Identifiants

Citation

Raphaëlle N. Roy, Jérémy Frey. Neurophysiological Markers for Passive Brain–Computer Interfaces. Maureen Clerc; Laurent Bougrain; Fabien Lotte. Brain–Computer Interfaces 1: Foundations and Methods, Wiley-ISTE, 2016, 978-1-84821-826-0. 〈10.1002/9781119144977.ch5〉. 〈hal-01413462〉

Partager

Métriques

Consultations de la notice

300