Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Defining and quantifying users' mental imagery-based BCI skills: a first step.

Fabien Lotte1,2, Camille Jeunet3,4

  • 1Inria Bordeaux Sud-Ouest, Talence, France.

Journal of Neural Engineering
|May 18, 2018
PubMed
Summary

Classification accuracy is a poor metric for assessing brain-computer interface (BCI) user skills. New metrics are proposed to better quantify electroencephalography (EEG) pattern self-modulation for improved BCI training and reliability.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Evolution of users' subjective experience over three training sessions with an EEG Motor-Imagery Brain-Computer Interface (MI-BCI).

Brain research·2025
Same author

Quantifying Inter- and Intra-Subject Variability of Sensorimotor Desynchronization Induced by Median Nerve Stimulation and Motor Imagery for BCI.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Reliable predictor of BCI motor imagery performance using median nerve stimulation.

Journal of neural engineering·2025
Same author

The NERVE-ML (neural engineering reproducibility and validity essentials for machine learning) checklist: ensuring machine learning advances neural engineering<sup></sup>.

Journal of neural engineering·2025
Same author

Master classes of the tenth international brain-computer interface meeting: showcasing the research of BCI trainees.

Journal of neural engineering·2025
Same author

Large scale investigation of the effect of gender on mu rhythm suppression in motor imagery brain-computer interfaces.

Brain computer interfaces (Abingdon, England)·2024

Area of Science:

  • Neuroscience
  • Biomedical Engineering
  • Signal Processing

Background:

  • Electroencephalography (EEG)-based brain-computer interfaces (BCIs) show promise but suffer from low reliability, hindering widespread adoption.
  • Current BCI research often relies on classification accuracy (CA) to evaluate both algorithms and user performance.
  • CA is insufficient for accurately assessing a user's ability to self-modulate EEG patterns, a critical skill for BCI control.

Purpose of the Study:

  • To address the limitations of CA in evaluating BCI user skills, particularly for mental imagery (MI) BCIs.
  • To propose a new definition and metrics for quantifying MI-BCI user skills independently of classification algorithms.
  • To provide tools for more accurate assessment of BCI user proficiency and training.

Main Methods:

Related Experiment Videos

  • Demonstrated the limitations of CA, highlighting its dependence on training data and classifiers.
  • Defined BCI skills based on the user's capacity for self-modulation of EEG patterns.
  • Introduced novel metrics (classDis, restDist, classStab) to measure the distinctness and stability of EEG patterns.

Main Results:

  • Re-analysis of EEG datasets confirmed that CA can obscure improvements or deficiencies in user skill.
  • New metrics successfully revealed skill enhancements and identified instances where user performance did not differ from baseline EEG.
  • The proposed metrics offer a more nuanced understanding of user self-modulation capabilities.

Conclusions:

  • CA should be used cautiously when assessing MI-BCI user skills and complemented by more specific metrics.
  • BCI user training requires re-evaluation to incorporate distinct BCI subskills and their objective measures.
  • Open-source Matlab code is provided to facilitate the adoption of these new metrics for enhanced BCI research and development.