A review of user training methods in brain computer interfaces based on mental tasks

Abstract

Mental-Tasks based Brain-Computer Interfaces (MT-BCIs) allow their users to interact with an external device solely by using brain signals produced through mental tasks. While MT-BCIs are promising for many applications, they are still barely used outside laboratories due to their lack of reliability. MT-BCIs require their users to develop the ability to self-regulate specific brain signals. However, the human learning process to control a BCI is still relatively poorly understood and how to optimally train this ability is currently under investigation. Despite their promises and achievements, traditional training programs have been shown to be sub-optimal and could be further improved. In order to optimize user training and improve BCI performance, human factors should be taken into account. An interdisciplinary approach should be adopted to provide learners with appropriate and/or adaptive training. In this article, we provide an overview of existing methods for MT-BCI user training-notably in terms of environment, instructions, feedback and exercises. We present a categorization and taxonomy of these training approaches, provide guidelines on how to choose the best methods and identify open challenges and perspectives to further improve MT-BCI user training.

Long-Term BCI Training of a Tetraplegic User: Adaptive Riemannian Classifiers and User Training

“…For the competition day, we needed a previously trained classifier to avoid calibration time. Moreover, recalibrating the classifier everyday could lead to an ever-changing feedback which may be detrimental to user training (Perdikis et al, 2018;Perdikis and Millan, 2020;Roc et al, 2020). Therefore, for the transfer phase (session 12 and subsequent) we calibrated our classifier based on the runs from the previous phase sessions that were the least noisy and the least contaminated by artifacts.…”

“…A progressive training seem essential for BCI as its efficiency relies on the users' ability to produce EEG patterns that are stable over time and distinct between the different mental commands (McFarland et al, 2010;Chavarriaga et al, 2016). Although improving users' ability to produce such signals through user training can certainly help the participants in controlling MT-BCI (Lotte and Jeunet, 2015;Perdikis and Millan, 2020;Roc et al, 2020), various sources of variability can lead to large shifts of data distribution between different sessions and consequently between the BCI classifier testing and training sets. Beside the largely unknown phenomena in the activity of neuronal populations which lead to non-stationarity of EEG signal (Kaplan et al, 2005), some variability sources including various environmental noises and changes in users' mental states such as their attention, fatigue or stress level are expected in an actual practice such as the CYBATHLON competition.…”

Experimenters' Influence on Mental-Imagery based Brain-Computer Interface User Training

“…In order to improve BCI reliability, it is thus highly relevant to identify, control and manipulate the factors affecting users' states. Among these many factors (e.g., instructions, feedback or exercise design) our literature review presented above suggests that the experimental environment may have a major influence, notably experimenters [42]. Despite the central role that experimenters have in BCI experimental process and the literature regarding the impact of social presence and emotional feedback, no studies had yet been led to evaluate their influence on MI-BCI experimental outcomes, i.e., performances and user-training.…”

Post-Adaptation Effects in a Motor Imagery Brain-Computer Interface Online Coadaptive Paradigm

“…Intense performance fluctuations during and, especially, across BCI sessions [16] and the inability of a large portion of users to get into control of a BCI have been early identified as [17], and still remain today [18], the main obstacles towards deploying BCI technology in real-world scenarios [19]- [23]. SMR-based BCIs are known to be particularly vulnerable to these issues [18], [24]- [27].…”

“…(10)(11)(12)(13) Hz) β low(16)(17)(18)(19)(20)(21)(22)(23)(24) β high(24)(25)(26)(27)(28)(29)(30)(31)(32) S1 Evidence of sensorimotor rhythm modulation during coadaptive training. (a) ERD/ERS maps of best performer S1 (right-handed) throughout the session.…”