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 Video

Updated: Jul 10, 2025

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

603

Multimodal pre-screening can predict BCI performance variability: A novel subject-specific experimental scheme.

Seyyed Bahram Borgheai1, Alyssa Hillary Zisk2, John McLinden3

  • 1Department of Electrical, Computer, and Biomedical Engineering, University of Rhode Island, Kingston, RI, United States; Neurology Department, Emory University, Atlanta, GA, United States.

Computers in Biology and Medicine
|November 20, 2023
PubMed
Summary

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

Cross-frequency bispectral EEG analysis of reach-to-grasp planning and execution.

Computers in biology and medicine·2026
Same author

Multimodal analysis of hemodynamic and electrocortical causal interactions during auditory processing.

Journal of neural engineering·2026
Same author

Case Series: Feasibility of Longitudinal Assessment of the Sciatic Nerve in CMT1A Using High-Resolution 7T MRI.

Muscle & nerve·2026
Same author

A Comprehensive Overview of Neurophysiological Correlates of Cognitive Impairment in Amyotrophic Lateral Sclerosis.

Cells·2026
Same author

Living with adrenoleukodystrophy: adult patient and caregiver perspectives.

Orphanet journal of rare diseases·2026
Same author

Longitudinal Assessment of Quality of Life and Functionality by CAP-PRI in Patients With Active Chronic Inflammatory Demyelinating Polyradiculoneuropathy.

Journal of the peripheral nervous system : JPNS·2025

This study introduces a personalized method to improve the reliability of brain-computer interfaces. By using a short pre-screening phase with brain imaging, researchers successfully predicted which task settings would work best for individual users, leading to improved performance accuracy.

Area of Science:

  • Neuroengineering and Brain-Computer Interface research
  • Multimodal fNIRS and EEG signal processing

Background:

Reliable long-term operation of brain-computer interfaces remains elusive due to significant fluctuations in user performance. That uncertainty drove the need for methods that account for individual differences in brain activity. Prior research has shown that standard interface setups often fail to adapt to changing user states. No prior work had resolved how to effectively integrate multiple imaging modalities for real-time performance optimization. This gap motivated the development of a personalized approach to stabilize system output. Current systems frequently struggle with both inter-subject and intra-subject variability during daily usage. Researchers have long sought ways to identify factors that interfere with successful device control. This study addresses these challenges by proposing a novel scheme to predict and compensate for performance-related fluctuations.

Purpose Of The Study:

The study aims to develop a personalized scheme for a multimodal brain-computer interface system to address performance variability. This research seeks to identify, predict, and compensate for factors that influence user competence. The investigators focus on integrating functional near-infrared spectroscopy and electroencephalography to enhance system robustness. This effort addresses the lack of reliability in current interfaces during long-term daily usage. The team intends to create predictive platforms that adapt to individual user needs. By analyzing pre-screening data, the researchers hope to determine the optimal task variations for each participant. This work is motivated by the need to support users with motor deficits who rely on these technologies. The study ultimately explores whether a multimodal predictive framework can effectively stabilize performance across different sessions.

Keywords:
Electroencephalography (EEG)Functional near-infrared spectroscopy (fNIRS)General linear model (GLM)Multivariate linear regression (MLR) modelPredictive modelsneurotechnologysignal processingpredictive modelingmotor impairment

Frequently Asked Questions

The researchers propose a personalized scheme using fNIRS and EEG to predict task performance. By analyzing pre-screening data with multivariate linear regression, the system identifies optimal task variations, achieving an average performance gain of 5.18% for participants.

The study utilizes a visuo-mental protocol during training sessions. This specific task structure allows for the extraction of features that inform the predictive models, enabling the system to determine the most effective workload for each individual user.

A pre-screening phase is necessary to capture baseline neural data. This initial step provides the required information to construct predictive platforms, which are then used to adjust task variations and maximize accuracy during subsequent test sessions.

More Related Videos

Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.1K
Cross-Modal Multivariate Pattern Analysis
13:51

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

20.0K

Related Experiment Videos

Last Updated: Jul 10, 2025

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

603
Assessment and Communication for People with Disorders of Consciousness
07:37

Assessment and Communication for People with Disorders of Consciousness

Published on: August 1, 2017

9.1K
Cross-Modal Multivariate Pattern Analysis
13:51

Cross-Modal Multivariate Pattern Analysis

Published on: November 9, 2011

20.0K

Main Methods:

The research team implemented a personalized experimental design involving eleven participants, including five individuals with motor deficits. Review Approach framing focuses on the use of functional near-infrared spectroscopy and electroencephalography for data collection. Subjects completed multiple training sessions featuring a short pre-screening phase followed by a novel visuo-mental protocol. The investigators utilized stepwise multivariate linear regression models to construct predictive platforms based on extracted features. During test sessions, the team employed a task-correction phase to select the ideal variation for each user. An interference-correction phase followed to further refine the system response. The study evaluated the associations between predicted and actual performance metrics across different task variations. Finally, the researchers assessed the overall outcome of the correction strategies by comparing performance gains.

Main Results:

Key Findings From the Literature indicate that the predictive models achieved adjusted R-squared values of 0.942, 0.724, and 0.939 for the three task variations. Statistical analysis revealed significant associations between predicted and actual performance for the first two variations. These associations yielded rho values of 0.7289 and 0.6970, respectively. The task-correction stage successfully determined the optimal task variation for 81.82% of the participants. Applying these correction strategies resulted in an average performance gain of 5.18% across the cohort. The researchers observed that the multimodal approach effectively identified factors affecting competence during the sessions. These results demonstrate the potential of the framework to compensate for performance variability. The data confirm that personalized models can improve accuracy for users with and without motor deficits.

Conclusions:

The researchers propose that their personalized scheme effectively mitigates performance variability in brain-computer interfaces. Their findings suggest that integrating multiple imaging modalities provides a robust framework for real-time task optimization. The study demonstrates that predictive models can successfully identify optimal task variations for a majority of participants. These results imply that such adaptive frameworks could enhance the usability of interfaces for individuals with motor impairments. The authors suggest that their approach offers a viable path toward more reliable long-term device operation. Their evidence indicates that task-correction strategies lead to measurable performance gains across different user groups. The study highlights the potential for future systems to automatically adjust based on individual pre-screening data. This work provides a foundation for developing more personalized and adaptive neurotechnological solutions.

The researchers employ stepwise multivariate linear regression models to process the extracted features. These statistical platforms enable the prediction of ideal task variations, which are then applied to compensate for performance fluctuations observed in the participants.

The study measures performance through adjusted R-squared values and correlation coefficients. For the first two task variations, the researchers observed significant associations between predicted and actual performance, with rho values of 0.7289 and 0.6970, respectively.

The authors suggest that this integrated framework could improve device reliability for individuals with severe motor deficits. By compensating for performance variability, the proposed method offers a pathway toward more consistent and effective long-term interface usage.