Alpha neurofeedback training improves SSVEP-based BCI performance

Exploring the effects of head movements and accompanying gaze fixation switch on steady-state visual evoked potential

In a realistic steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI) application like driving a car or controlling a quadrotor, observing the surrounding environment while simultaneously gazing at the stimulus is necessary. This kind of application inevitably could cause head movements and variation of the accompanying gaze fixation point, which might affect the SSVEP and BCI’s performance. However, few papers studied the effects of head movements and gaze fixation switch on SSVEP response, and the corresponding BCI performance. This study aimed to explore these effects by designing a new ball tracking paradigm in a virtual reality (VR) environment with two different moving tasks, i.e., the following and free moving tasks, and three moving patterns, pitch, yaw, and static. Sixteen subjects were recruited to conduct a BCI VR experiment. The offline data analysis showed that head moving patterns [F(2, 30) = 9.369, p = 0.001, effect size = 0.384] resulted in significantly different BCI decoding performance but the moving tasks had no effect on the results [F(1, 15) = 3.484, p = 0.082, effect size = 0.188]. Besides, the canonical correlation analysis (CCA) and filter bank canonical correlation analysis (FBCCA) accuracy were better than the PSDA and MEC methods in all of the conditions. These results implied that head movement could significantly affect the SSVEP performance but it was possible to switch gaze fixation to interact with the surroundings in a realistic BCI application.

Neurofeedback Training of the Control Network Improves Children's Performance with an SSVEP-based BCI

In the past 20 years, neural engineering has made unprecedented progress in the interpretation of brain information (e.g., brain-computer interfaces) and in neuromodulation (e.g., electromagnetic stimulation and neurofeedback). However, there has been little research aiming to improve the performance of brain-computer interfaces (BCIs) using neuromodulation. The present study presents a novel design for a neurofeedback training (NFT) method to improve the operation of a steady-state visual evoked potential (SSVEP)-based BCI and further explores its underlying mechanisms. The use of NFT to upregulate alpha-band power in the user's parietal lobe is presented in this study as a new neuromodulation method to improve SSVEP-based BCI in this study. After users completed this NFT intervention, the signal-to-noise ratio (SNR), accuracy, and information transfer rate (ITR) of the SSVEP-based BCI were increased by 5.8%, 4.7%, and 15.6%, respectively. However, no improvement was observed in the control group in which the subjects did not participate in NFT. Moreover, a general reinforcement of the information flow from the parietal lobe to the occipital lobe was observed. Evidence from a network analysis and an attention test further indicates that NFT improves attention by developing the control capacity of the parietal lobe and then enhances the above SSVEP indicators. Upregulating the amplitude of parietal alpha oscillations using NFT significantly improves the SSVEP-based BCI performance by modulating the control network. The study validates an effective neuromodulation method and possibly contributes to explaining the function of the parietal lobe in the control network.

Selective Subject Pooling Strategy to Improve Model Generalization for a Motor Imagery BCI

Brain–computer interfaces (BCIs) facilitate communication for people who cannot move their own body. A BCI system requires a lengthy calibration phase to produce a reasonable classifier. To reduce the duration of the calibration phase, it is natural to attempt to create a subject-independent classifier with all subject datasets that are available; however, electroencephalogram (EEG) data have notable inter-subject variability. Thus, it is very challenging to achieve subject-independent BCI performance comparable to subject-specific BCI performance. In this study, we investigate the potential for achieving better subject-independent motor imagery BCI performance by conducting comparative performance tests with several selective subject pooling strategies (i.e., choosing subjects who yield reasonable performance selectively and using them for training) rather than using all subjects available. We observed that the selective subject pooling strategy worked reasonably well with public MI BCI datasets. Finally, based upon the findings, criteria to select subjects for subject-independent BCIs are proposed here.

Effects of Transcranial Alternating Current Stimulation and Neurofeedback on Alpha (EEG) Dynamics: A Review

Transcranial alternating current stimulation (tACS) and neurofeedback (NFB) are two different types of non-invasive neuromodulation techniques, which can modulate brain activity and improve brain functioning. In this review, we compared the current state of knowledge related to the mechanisms of tACS and NFB and their effects on electroencephalogram (EEG) activity (online period/stimulation period) and on aftereffects (offline period/post/stimulation period), including the duration of their persistence and potential behavioral benefits. Since alpha bandwidth has been broadly studied in NFB and in tACS research, the studies of NFB and tACS in modulating alpha bandwidth were selected for comparing the online and offline effects of these two neuromodulation techniques. The factors responsible for variability in the responsiveness of the modulated EEG activity by tACS and NFB were analyzed and compared too. Based on the current literature related to tACS and NFB, it can be concluded that tACS and NFB differ a lot in the mechanisms responsible for their effects on an online EEG activity but they possibly share the common universal mechanisms responsible for the induction of aftereffects in the targeted stimulated EEG band, namely Hebbian and homeostatic plasticity. Many studies of both neuromodulation techniques report the aftereffects connected to the behavioral benefits. The duration of persistence of aftereffects for NFB and tACS is comparable. In relation to the factors influencing responsiveness to tACS and NFB, significantly more types of factors were analyzed in the NFB studies compared to the tACS studies. Several common factors for both tACS and NFB have been already investigated. Based on these outcomes, we propose several new research directions regarding tACS and NFB.

An Exploratory Study of Training Intensity in EEG Neurofeedback

Neurofeedback training has shown benefits in clinical treatment and behavioral performance enhancement. Despite the wide range of applications, no consensus has been reached about the optimal training schedule. In this work, an EEG neurofeedback practical experiment was conducted aimed at investigating the effects of training intensity on the enhancement of the amplitude in the individual upper alpha band. We designed INTENSIVE and SPARSE training modalities, which differed regarding three essential aspects of training intensity: the number of sessions, the duration of a session, and the interval between sessions. Nine participants in the INTENSIVE group completed 4 sessions with 37.5 minutes each during consecutive days, while nine participants in the SPARSE group performed 6 sessions of 25 minutes spread over approximately 3 weeks. As a result, regarding the short-term effects, the upper alpha band amplitude change within sessions did not significantly differ between the two groups. Nonetheless, only the INTENSIVE group showed a significant increase in the upper alpha band amplitude. However, for the sustained effects across sessions, none of the groups showed significant changes in the upper alpha band amplitude across the whole course of training. The findings suggest that the progression within session is favored by the intensive design. Therefore, based on these findings, it is proposed that training intensity influences EEG self-regulation within sessions. Further investigations are needed to isolate different aspects of training intensity and effectively confirm if one modality globally outperforms the other.

Stress-Induced Effects in Resting EEG Spectra Predict the Performance of SSVEP-Based BCI

Most research in Brain-Computer-Interfaces (BCI) focuses on technologies to improve accuracy and speed. Little has been done on the effects of subject variability, both across individuals and within the same individual, on BCI performance. For example, stress, arousal, motivation, and fatigue can all affect the electroencephalogram (EEG) signals used by a BCI, which in turn impacts performance. Overcoming the impact of such user variability on BCI performance is an impending and inevitable challenge for routine applications of BCIs in the real world. To systematically explore the factors affecting BCI performance, this study embeds a Steady-State Visually Evoked Potential (SSVEP) based BCI into a "game with a purpose" (GWAP) to obtain data over significant lengths of time, under both high- and low-stress conditions. Ten healthy volunteers played a GWAP that resembles popular match-three games, such as Jewel Quest, Zoo Boom, or Candy Crush. We recorded the target search time, target search accuracy, and EEG signals during gameplay to investigate the impacts of stress on EEG signals and BCI performance. We used Canonical Correlation Analysis (CCA) to determine whether the subject had found and attended to the correct target. The experimental results show that SSVEP target-classification accuracy is reduced by stress. We also found a negative correlation between EEG spectra and the SNR of EEG in the frontal and occipital regions during gameplay, with a larger negative correlation for the high-stress conditions. Furthermore, CCA also showed that when the EEG alpha and theta power increased, the search accuracy decreased, and the spectral amplitude drop was more evident under the high-stress situation. These results provide new, valuable insights into research on how to improve the robustness of BCIs in real-world applications.

Indoor Simulated Training Environment for Brain-Controlled Wheelchair Based on Steady-State Visual Evoked Potentials

Brain-controlled wheelchair (BCW) has the potential to improve the quality of life for people with motor disabilities. A lot of training is necessary for users to learn and improve BCW control ability and the performances of BCW control are crucial for patients in daily use. In consideration of safety and efficiency, an indoor simulated training environment is built up in this paper to improve the performance of BCW control. The indoor simulated environment mainly realizes BCW implementation, simulated training scenario setup, path planning and recommendation, simulated operation, and scoring. And the BCW is based on steady-state visual evoked potentials (SSVEP) and the filter bank canonical correlation analysis (FBCCA) is used to analyze the electroencephalography (EEG). Five tasks include individual accuracy, simple linear path, obstacles avoidance, comprehensive steering scenarios, and evaluation task are designed, 10 healthy subjects were recruited and carried out the 7-days training experiment to assess the performance of the training environment. Scoring and command-consuming are conducted to evaluate the improvement before and after training. The results indicate that the average accuracy is 93.55% and improves from 91.05% in the first stage to 96.05% in the second stage (p = 0.001). Meanwhile, the average score increases from 79.88 in the first session to 96.66 in the last session and tend to be stable (p < 0.001). The average number of commands and collisions to complete the tasks decreases significantly with or without the approximate shortest path (p < 0.001). These results show that the performance of subjects in BCW control achieves improvement and verify the feasibility and effectiveness of the proposed simulated training environment.

Combined Lateral and Longitudinal Control of EEG Signals-Based Brain-Controlled Vehicles

Using brain signals rather than limbs to control a vehicle may not only help persons with disabilities to acquire driving ability, but also provide healthy persons with a new alternative way to drive. In this paper, we propose a combined lateral and longitudinal control system for electroencephalogram (EEG) signals-based brain-controlled vehicles (BCVs). The proposed system is designed by integrating a user interface, a brain-computer interface (BCI), a control interface model, a lateral controller, and a longitudinal controller. We conduct driver-and-hardware-in-the-loop experiments under two control conditions (i.e., the brain- and manual-control conditions) with different subjects and three driving tests (i.e., the lane-changing, path-selection, and car-following tests). Experimental results show the feasibility of using brain signals to continuously perform both the lateral and longitudinal control of a vehicle. This study not only promotes the development of BCVs, but also provides some insights on how to apply BCIs in conjunction with assistant controllers to control other dynamic systems.

P300 Speller Performance Predictor Based on RSVP Multi-feature

Brain-computer interface (BCI) systems were developed so that people can control computers or machines through their brain activity without moving their limbs. The P300 speller is one of the BCI applications used most commonly, as is very simple and reliable and can achieve satisfactory performance. However, like other BCIs, the P300 speller still has room for improvements in terms of its practical use, for example, selecting the best compromise between spelling accuracy and information transfer rate (ITR; speed) so that the P300 speller can maintain high accuracy while increasing spelling speed. Therefore, seeking correlates of, and predicting, the P300 speller's performance is necessary to understand and improve the technique. In this work, we investigated the correlations between rapid serial visual presentation (RSVP) task features and the P300 speller's performance. Fifty-five subjects participated in the RSVP and conventional matrix P300 speller tasks and RSVP behavioral and electroencephalography (EEG) features were compared in the P300's speller performance. We found that several of the RSVP's event-related potential (ERP) and behavioral features were correlated with the P300 speller's offline binary classification accuracy. Using these features, we propose a simple multi-feature performance predictor (r = 0.53, p = 0.0001) that outperforms any single feature performance predictor, including that of the conventional RSVP T1% predictor (r = 0.28, p = 0.06). This result demonstrates that selective multi-features can predict BCI performance better than a single feature alone.

EEG-Based Brain-Computer Interfaces

Brain-computer interfaces (BCIs) provide a direct communication channel between human brain and output devices. Due to advantages such as non-invasiveness, ease of use, and low cost, electroencephalography (EEG) is the most popular method for current BCIs. This chapter gives an overview of the current EEG-based BCIs for the main purpose of communication and control. This chapter first provides a taxonomy of the EEG-based BCI systems by categorizing them into three major groups: (1) BCIs based on event-related potentials (ERPs), (2) BCIs based on sensorimotor rhythms, and (3) hybrid BCIs. Next, this chapter describes challenges and potential solutions in developing practical BCI systems toward high communication speed, convenient system use, and low user variation. Then this chapter briefly reviews both medical and non-medical applications of current BCIs. Finally, this chapter concludes with a summary of current stage and future perspectives of the EEG-based BCI technology.

EEG Signals-Based Longitudinal Control System for a Brain-Controlled Vehicle

Directly using brain signals to drive a vehicle may not only help persons with disabilities to regain driving ability but also provide a new alternative way for healthy people to control a vehicle. In this paper, we propose a new longitudinal control system based on electroencephalogram signals for brain-controlled vehicles (BCVs) by combining a user interface, a brain-computer interface (BCI) system, and a longitudinal control module. Driver-in-the-loop experiments were conducted by using two driving tests (i.e., the destination-approaching and car-following tests) with different subjects under two control conditions, i.e., the brain and manual control conditions. Experimental results show the feasibility of alone using brain signals to continuously perform the longitudinal control of a vehicle at a relatively high speed, at least for some users. This paper not only promotes the development of BCVs but also provides some insights into the research on how to apply BCIs to control other high-speed dynamic systems.

Eyes-Closed Resting EEG Predicts the Learning of Alpha Down-Regulation in Neurofeedback Training

Neurofeedback training, which enables the trainee to learn self-control of the EEG activity of interest based on online feedback, has demonstrated benefits on cognitive and behavioral performance. Nevertheless, as a core mechanism of neurofeedback, learning of EEG regulation (i.e., EEG learning) has not been well understood. Moreover, a substantial number of non-learners who fail to achieve successful EEG learning have often been reported. This study investigated the EEG learning in alpha down-regulation neurofeedback, aiming to better understand the alpha learning and to early predict learner/non-learner. Twenty-nine participants received neurofeedback training to down-regulate alpha in two days, while eight of them were identified as non-learners who failed to reduce their alpha within sessions. Through a stepwise linear discriminant analysis, a prediction model was built based on participant's eyes-closed resting EEG activities in broad frequency bands including lower alpha, theta, sigma and beta 1 measured before training, which was validated in predicting learners/non-learners. The findings would assist in the early identification of the individuals who would not likely reduce their alpha during neurofeedback.

Mathematical Modeling of EEG Signals-Based Brain-Control Behavior

Brain-control behaviors (BCBs) are behaviors of humans that communicate with external devices by means of the human brain rather than peripheral nerves or muscles. In this paper, to understand and simulate such behaviors, we propose a mathematical model by combining a queuing network-based encoding model with a brain-computer interface model. Experimental results under the static tests show the effectiveness of the proposed model in simulating real BCBs. Furthermore, we verify the effectiveness and applicability of the proposed model through the dynamic experimental tests in a simulated vehicle. This paper not only promotes the understanding and prediction of BCBs, but also provides some insights into assistive technology on brain-controlled systems and extends the scope of research on human behavior modeling.

Training in Use of Brain-Machine Interface-Controlled Robotic Hand Improves Accuracy Decoding Two Types of Hand Movements

Objective: Brain-machine interfaces (BMIs) are useful for inducing plastic changes in cortical representation. A BMI first decodes hand movements using cortical signals and then converts the decoded information into movements of a robotic hand. By using the BMI robotic hand, the cortical representation decoded by the BMI is modulated to improve decoding accuracy. We developed a BMI based on real-time magnetoencephalography (MEG) signals to control a robotic hand using decoded hand movements. Subjects were trained to use the BMI robotic hand freely for 10 min to evaluate plastic changes in the cortical representation due to the training.

Method: We trained nine young healthy subjects with normal motor function. In open-loop conditions, they were instructed to grasp or open their right hands during MEG recording. Time-averaged MEG signals were then used to train a real decoder to control the robotic arm in real time. Then, subjects were instructed to control the BMI-controlled robotic hand by moving their right hands for 10 min while watching the robot's movement. During this closed-loop session, subjects tried to improve their ability to control the robot. Finally, subjects performed the same offline task to compare cortical activities related to the hand movements. As a control, we used a random decoder trained by the MEG signals with shuffled movement labels. We performed the same experiments with the random decoder as a crossover trial. To evaluate the cortical representation, cortical currents were estimated using a source localization technique. Hand movements were also decoded by a support vector machine using the MEG signals during the offline task. The classification accuracy of the movements was compared among offline tasks.

Results: During the BMI training with the real decoder, the subjects succeeded in improving their accuracy in controlling the BMI robotic hand with correct rates of 0.28 ± 0.13 to 0.50 ± 0.11 (p = 0.017, n = 8, paired Student's t-test). Moreover, the classification accuracy of hand movements during the offline task was significantly increased after BMI training with the real decoder from 62.7 ± 6.5 to 70.0 ± 11.1% (p = 0.022, n = 8, t₍₇₎ = 2.93, paired Student's t-test), whereas accuracy did not significantly change after BMI training with the random decoder from 63.0 ± 8.8 to 66.4 ± 9.0% (p = 0.225, n = 8, t₍₇₎ = 1.33).

Conclusion: BMI training is a useful tool to train the cortical activity necessary for BMI control and to induce some plastic changes in the activity.

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 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.