The impact of goal-oriented task design on neurofeedback learning for brain–computer interface control

Neurofeedback training teaches individuals to modulate brain activity by providing real-time feedback and can be used for brain–computer interface control. The present study aimed to optimize training by maximizing engagement through goal-oriented task design. Participants were shown either a visual...

Full description

Saved in:
Bibliographic Details
Published in:Medical & biological engineering & computing 2018-02, Vol.56 (2), p.201-210
Main Authors: McWhinney, S. R., Tremblay, A., Boe, S. G., Bardouille, T.
Format: Article
Language:English
Subjects:
Biofeedback
Biomedical and Life Sciences
Biomedical Engineering and Bioengineering
Biomedicine
Brain
Computer Applications
Design
EEG
Electroencephalography
Feedback
Hemispheres
Human Physiology
Image manipulation
Imaging
Mental task performance
Optimization
Original Article
Radiology
Robot control
Robots
Training
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Neurofeedback training teaches individuals to modulate brain activity by providing real-time feedback and can be used for brain–computer interface control. The present study aimed to optimize training by maximizing engagement through goal-oriented task design. Participants were shown either a visual display or a robot, where each was manipulated using motor imagery (MI)-related electroencephalography signals. Those with the robot were instructed to quickly navigate grid spaces, as the potential for goal-oriented design to strengthen learning was central to our investigation. Both groups were hypothesized to show increased magnitude of these signals across 10 sessions, with the greatest gains being seen in those navigating the robot due to increased engagement. Participants demonstrated the predicted increase in magnitude, with no differentiation between hemispheres. Participants navigating the robot showed stronger left-hand MI increases than those with the computer display. This is likely due to success being reliant on maintaining strong MI-related signals. While older participants showed stronger signals in early sessions, this trend later reversed, suggesting greater natural proficiency but reduced flexibility. These results demonstrate capacity for modulating neurofeedback using MI over a series of training sessions, using tasks of varied design. Importantly, the more goal-oriented robot control task resulted in greater improvements.
ISSN:0140-0118
1741-0444
DOI:10.1007/s11517-017-1683-1