Are users' traits informative enough to predict/explain their mental-imagery based BCI performances ?

Camille Benaroch 1 Camille Jeunet 2 Fabien Lotte 1
1 Potioc - Popular interaction with 3d content
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest
Abstract : Mental-Imagery based Brain-Computer Interfaces (MI-BCIs) make use of brain signals produced during mental imagery tasks to control a computerised system. The current unreliability of MI-BCIs could be due, at least in part, to the use of inappropriate user-training procedures. In order to improve these procedures , it is necessary first to understand the mechanisms underlying MI-BCI user-training, notably through the identification of the factors influencing it. Thus, this paper aims at creating a statistical model that could explain/predict the performances of MI-BCI users using their traits (e.g., personality). We used the data of 42 participants (i.e., 180 MI-BCI sessions in total) collected from three different studies that were based on the same MI-BCI paradigm. We used machine learning regressions with a leave-one-subject-out cross validation to build different models. Our first results showed that using the users' traits only may enable the prediction of performances within one multiple-session experiment, but might not be sufficient to reliably predict MI-BCI performances across experiments.
Complete list of metadatas

Cited literature [30 references]  Display  Hide  Download

https://hal.inria.fr/hal-02111581
Contributor : Camille Benaroch <>
Submitted on : Friday, April 26, 2019 - 10:15:28 AM
Last modification on : Thursday, June 27, 2019 - 4:36:04 PM

File

GBCIC2019_Paper48.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02111581, version 1

Collections

Citation

Camille Benaroch, Camille Jeunet, Fabien Lotte. Are users' traits informative enough to predict/explain their mental-imagery based BCI performances ?. 8th Graz BCI Conference 2019, Sep 2019, Graz, Austria. ⟨hal-02111581⟩

Share

Metrics

Record views

134

Files downloads

295