Related Experiment Video
Updated: Mar 31, 2026
05:59
Author Spotlight: Unlocking New Insights in fNIRS Studies - A Novel Framework for Inter-Brain Synchrony Analysis
Published on: October 6, 2023
3.5K
Correlates of Near-Infrared Spectroscopy Brain-Computer Interface Accuracy in a Multi-Class Personalization Framework
1PRISM Laboratory, Bloorview Research Institute, Holland Bloorview Kids Rehabilitation Hospital , Toronto, ON , Canada ; PRISM Laboratory, Institute of Biomaterials and Biomedical Engineering, University of Toronto , Toronto, ON , Canada.
Frontiers in Human Neuroscience
|October 21, 2015
Summary
This study shows a personalized approach can improve brain-computer interface (BCI) control using near-infrared spectroscopy (NIRS). Accuracy correlated with user experience, not verbal IQ, for NIRS-BCI systems.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Human-Computer Interaction
Background:
- Brain-computer interfaces (BCIs) enable computer interaction via neural activity.
- Near-infrared spectroscopy (NIRS) BCIs commonly use predefined tasks for binary control.
- Personalized approaches may enhance NIRS-BCI performance and user experience.
Purpose of the Study:
- To assess a personalized strategy for controlling NIRS-BCIs with 2-5 classes.
- To investigate user characteristics influencing NIRS-BCI accuracy.
- To determine the feasibility of individualized NIRS-BCI task selection.
Main Methods:
- Recruited 10 able-bodied participants for 5 data collection sessions.
- Utilized 6 mental tasks, selecting a personalized subset for optimal discrimination.
- Performed offline cross-validation to evaluate classification accuracies.
Main Results:
- Achieved average accuracies of 78% (2-class), 61% (3-class), 47% (4-class), and 37% (5-class).
- All participants exceeded 70% accuracy for 2-class control; two exceeded 70% for 3-class control.
- Accuracy positively correlated with perceived ease, concentration, and enjoyment (ρ=0.653, 0.634, 0.550).
- Accuracy negatively correlated with verbal IQ (ρ=-0.749).
Conclusions:
- A personalized approach is effective for enhancing NIRS-BCI control.
- User-centric factors like perceived ease and enjoyment significantly impact BCI performance.
- Future NIRS-BCI development should consider individual user characteristics beyond cognitive abilities.
