High-speed spelling with a noninvasive brain-computer interface.

The past 20 years have witnessed unprecedented progress in brain-computer interfaces (BCIs). However, low communication rates remain key obstacles to BCI-based communication in humans. This study presents an electroencephalogram-based BCI speller that can achieve information transfer rates (ITRs) up to 5.32 bits per second, the highest ITRs reported in BCI spellers using either noninvasive or invasive methods. Based on extremely high consistency of frequency and phase observed between visual flickering signals and the elicited single-trial steady-state visual evoked potentials, this study developed a synchronous modulation and demodulation paradigm to implement the speller. Specifically, this study proposed a new joint frequency-phase modulation method to tag 40 characters with 0.5-s-long flickering signals and developed a user-specific target identification algorithm using individual calibration data. The speller achieved high ITRs in online spelling tasks. This study demonstrates that BCIs can provide a truly naturalistic high-speed communication channel using noninvasively recorded brain activities.

Exploring the temporal dynamics of sustained and transient spatial attention using steady-state visual evoked potentials.

While the behavioral dynamics as well as the functional network of sustained and transient attention have extensively been studied, their underlying neural mechanisms have most often been investigated in separate experiments. In the present study, participants were instructed to perform an audio-visual spatial attention task. They were asked to attend to either the left or the right hemifield and to respond to deviant transient either auditory or visual stimuli. Steady-state visual evoked potentials (SSVEPs) elicited by two task irrelevant pattern reversing checkerboards flickering at 10 and 15 Hz in the left and the right hemifields, respectively, were used to continuously monitor the locus of spatial attention. The amplitude and phase of the SSVEPs were extracted for single trials and were separately analyzed. Sustained attention to one hemifield (spatial attention) as well as to the auditory modality (intermodal attention) increased the inter-trial phase locking of the SSVEP responses, whereas briefly presented visual and auditory stimuli decreased the single-trial SSVEP amplitude between 200 and 500 ms post-stimulus. This transient change of the single-trial amplitude was restricted to the SSVEPs elicited by the reversing checkerboard in the spatially attended hemifield and thus might reflect a transient re-orienting of attention towards the brief stimuli. Thus, the present results demonstrate independent, but interacting neural mechanisms of sustained and transient attentional orienting.

Bayesian estimation of ERP components from multicondition and multichannel EEG.

Extraction and separation of functionally different event-related potentials (ERPs) from electroencephalography (EEG) is a long-standing problem in cognitive neuroscience. In this paper, we propose a Bayesian spatio-temporal model for estimating ERP components from multichannel EEG recorded under multiple experimental conditions. The model isolates the spatially and temporally overlapping ERP components by utilizing their phase-locking structure and the inter-condition non-stationarity structure of their amplitudes and latencies. Critically, unlike in previous multilinear algorithms, the non-phase-locked background EEGs are modeled as spatially correlated and non-isotropic signals. A variational algorithm was developed for approximate Bayesian inference of the proposed model, with the effective number of ERP components automatically determined as a part of the algorithm. The utility of the algorithm is demonstrated with applications to synthetic data and the EEG data collected from 13 subjects during a face inversion experiment. The results show that our algorithm more accurately and reliably estimates the spatio-temporal patterns, amplitudes, and latencies of the underlying ERP components in comparison with several state-of-the-art algorithms.

Align and Pool for EEG Headset Domain Adaptation (ALPHA) to Facilitate Dry Electrode Based SSVEP-BCI.

OBJECTIVE: The steady-state visual evoked potential based brain-computer interface (SSVEP-BCI) implemented in dry electrodes is a promising paradigm for alternative and augmentative communication in real-world applications. To improve its performance and reduce the calibration effort for dry-electrode systems, we utilize cross-device transfer learning by exploiting auxiliary individual wet-electrode electroencephalogram (EEG). METHODS: We proposed a novel transfer learning framework named ALign and Pool for EEG Headset domain Adaptation (ALPHA), which aligns the spatial pattern and the covariance for domain adaptation. To evaluate its efficacy, 75 subjects performed an experiment of 2 sessions involving a 12-target SSVEP-BCI task. RESULTS: ALPHA significantly outperformed a baseline approach (canonical correlation analysis, CCA) and two competing transfer learning approaches (transfer template CCA, ttCCA and least square transformation, LST) in two transfer directions. When transferring from wet to dry EEG headsets, ALPHA significantly outperformed the fully-calibrated approach of task-related component analysis (TRCA). CONCLUSION: ALPHA advances the frontier of recalibration-free cross-device transfer learning for SSVEP-BCIs and boosts the performance of dry electrode based systems. SIGNIFICANCE: ALPHA has methodological and practical implications and pushes the boundary of dry electrode based SSVEP-BCI toward real-world applications.

A hierarchical Bayesian approach for learning sparse spatio-temporal decompositions of multichannel EEG.

Multichannel electroencephalography (EEG) offers a non-invasive tool to explore spatio-temporal dynamics of brain activity. With EEG recordings consisting of multiple trials, traditional signal processing approaches that ignore inter-trial variability in the data may fail to accurately estimate the underlying spatio-temporal brain patterns. Moreover, precise characterization of such inter-trial variability per se can be of high scientific value in establishing the relationship between brain activity and behavior. In this paper, a statistical modeling framework is introduced for learning spatio-temporal decompositions of multiple-trial EEG data recorded under two contrasting experimental conditions. By modeling the variance of source signals as random variables varying across trials, the proposed two-stage hierarchical Bayesian model is able to capture inter-trial amplitude variability in the data in a sparse way where a parsimonious representation of the data can be obtained. A variational Bayesian (VB) algorithm is developed for statistical inference of the hierarchical model. The efficacy of the proposed modeling framework is validated with the analysis of both synthetic and real EEG data. In the simulation study we show that even at low signal-to-noise ratios our approach is able to recover with high precision the underlying spatio-temporal patterns and the dynamics of source amplitude across trials; on two brain-computer interface (BCI) data sets we show that our VB algorithm can extract physiologically meaningful spatio-temporal patterns and make more accurate predictions than other two widely used algorithms: the common spatial patterns (CSP) algorithm and the Infomax algorithm for independent component analysis (ICA). The results demonstrate that our statistical modeling framework can serve as a powerful tool for extracting brain patterns, characterizing trial-to-trial brain dynamics, and decoding brain states by exploiting useful structures in the data.

Interface, interaction, and intelligence in generalized brain-computer interfaces.

A brain-computer interface (BCI) establishes a direct communication channel between a brain and an external device. With recent advances in neurotechnology and artificial intelligence (AI), the brain signals in BCI communication have been advanced from sensation and perception to higher-level cognition activities. While the field of BCI has grown rapidly in the past decades, the core technologies and innovative ideas behind seemingly unrelated BCI systems have never been summarized from an evolutionary point of view. Here, we review various BCI paradigms and present an evolutionary model of generalized BCI technology which comprises three stages: interface, interaction, and intelligence (I3). We also highlight challenges, opportunities, and future perspectives in the development of new BCI technology.

Optimizing a left and right visual field biphasic stimulation paradigm for SSVEP-based BCIs with hairless region behind the ear.

Objective.Steady-state visual evoked potential (SSVEP) based brain-computer interface (BCI) has the characteristics of fast communication speed, high stability, and wide applicability, thus it has been widely studied. With the rapid development in paradigm, algorithm, and system design, SSVEP-BCI is gradually applied in clinical and real-life scenarios. In order to improve the ease of use of the SSVEP-BCI system, many studies have been focusing on developing it on the hairless area, but due to the lack of redesigning the stimulation paradigm to better adapt to the new area, the electroencephalography response in the hairless area is worse than occipital region.Approach. This study first proposed a phase difference estimation method based on stimulating the left and right visual field separately, then developed and optimized a left and right visual field biphasic stimulation paradigm for SSVEP-based BCIs with hairless region behind the ear.Main results.In the 12-target online experiment, after a short model estimation training, all 16 subjects used their best stimulus condition. The paradigm designed in this study can increase the proportion of applicable subjects for the behind-ear SSVEP-BCI system from 58.3% to 75% and increase the accuracy from 74.6 ± 20.0% (the existing best SSVEP stimulus with hairless region behind the ear) to 84.2±14.7%, and the information transfer rate from 14.2±6.4 bits min-1to 17.8±5.7 bits min-1.Significance.These results demonstrated that the proposed paradigm can effectively improve the BCI performance using the signal from the hairless region behind the ear, compared with the standard SSVEP stimulation paradigm.

Improving the Performance of Individually Calibrated SSVEP-BCI by Task- Discriminant Component Analysis.

A brain-computer interface (BCI) provides a direct communication channel between a brain and an external device. Steady-state visual evoked potential based BCI (SSVEP-BCI) has received increasing attention due to its high information transfer rate, which is accomplished by individual calibration for frequency recognition. Task-related component analysis (TRCA) is a recent and state-of-the-art method for individually calibrated SSVEP-BCIs. However, in TRCA, the spatial filter learned from each stimulus may be redundant and temporal information is not fully utilized. To address this issue, this paper proposes a novel method, i.e., task-discriminant component analysis (TDCA), to further improve the performance of individually calibrated SSVEP-BCI. The performance of TDCA was evaluated by two publicly available benchmark datasets, and the results demonstrated that TDCA outperformed ensemble TRCA and other competing methods by a significant margin. An offline and online experiment testing 12 subjects further validated the effectiveness of TDCA. The present study provides a new perspective for designing decoding methods in individually calibrated SSVEP-BCI and presents insight for its implementation in high-speed brain speller applications.

The Spatio-Temporal Equalization for Evoked or Event-Related Potential Detection in Multichannel EEG Data.

GOAL: Evoked or Event-Related Potential (EP/ERP) detection is a major area of interest within the domain of EEG (electroencephalography) signal processing. While traditional methods of EEG processing have mostly focused on enhancing signal components, few have explored background noise suppression techniques. Optimizing the suppression of background noise can play a critical role in improving the performance of EP/ERP detection. METHODS: In this study, a spatio-temporal equalization (STE) method was proposed based on the Multivariate Autoregressive (MVAR) model, which has been shown to suppress the spatio-temporal correlation of background noise and improve the EEG signal detection performance. RESULTS: For practical applications, two optimization schemes based on the spatio-temporal equalization method were designed to solve two common challenges in EEG signal detection: P300 and steady state visual evoked potentials. Our results demonstrated that the STE method effectively improves recognition performance of evoked or event-related potential detection. Additionally, the STE method offers fewer parameters, lower computational complexity, and easier implementation. CONCLUSION: These attributes allow the STE approach to be extended as a preprocessing method which can be used in combination with existing techniques.

Optimizing a dual-frequency and phase modulation method for SSVEP-based BCIs.

OBJECTIVE: The design of the stimulation paradigm plays an important role in steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI) studies. Among various stimulation designs, the dual-frequency paradigm in which two frequencies are used to encode one target is of importance and interest. However, because the number of possible frequency combinations is huge, the existing dual-frequency modulation paradigms failed to optimize the encoding towards the best combinations. Thus, this work aiming at designing a new dual-frequency and phase modulation paradigm with the best combinations stimuli. APPROACH: This study proposed a dual-frequency and phase modulation method, which can achieve a large number of targets by making different combinations of two frequencies and an initial phase. This study also designed a set of methods for quickly optimizing the stimulation codes for the dual-frequency and phase modulation method. MAIN RESULTS: An online 40-class BCI experiment with 12 subjects obtained an accuracy of 96.06[Formula: see text]4.00% and an averaged information transfer rate (ITR) of 196.09[Formula: see text]15.25 bits min-1, which were much higher than the existing dual-frequency modulation paradigms. Moreover, an offline simulation with a public dataset showed that the optimization method was also effective for optimizing the single-frequency and phase modulation paradigm. SIGNIFICANCE: These results demonstrate the high performance of the proposed dual-frequency and phase modulation method and the high efficiency of the optimization method for designing SSVEP stimulation paradigms. In addition, the coding efficiency of the optimized dual-frequency and phase modulation paradigm is higher than that of the single-frequency and phase modulation paradigm, and it is expected to further realize the BCI paradigm with a large amount of targets.

Impaired neural coordination in hippocampus of diabetic rat.

In vitro electrical neurophysiological and behavioural studies have shown that diabetes mellitus negatively affects hippocampal function. In this study, by using in vivo extracellular recording, the spontaneous neural activity was obtained from hippocampus of anaesthetized rats in both streptozotocin-induced diabetes group and normal control group. Temporal relationship between neuronal firing and slow oscillation (1-4 Hz) of local field potentials (LFPs) in hippocampus was analyzed using coherence and phase locking measurement. Lower coherence value (0.617+/-0.028) was observed in diabetic rats than that in control rats (0.730+/-0.024) (P=0.005). Furthermore, phase-locking measurement using von Mises fitting parameterized by a concentration parameter kappa showed a lower degree (kappa= 0.347+/-0.113) of temporal coordination between neuronal spiking and slow oscillation of LFPs in the hippocampus of diabetic rats than that of normal ones (kappa= 1.174+/-0.134) (P<0.001). Both approaches demonstrated that diabetes can indeed impair the temporal coordination between neuronal spiking and slow oscillation of population activity in hippocampus. This observed neural coordination impairment may serve as a network level mechanism for diabetes-induced memory deterioration.

A novel system of SSVEP-based human-robot coordination.

OBJECTIVE: Human-robot coordination (HRC) aims to enable human and robot to form a tightly coupled system to accomplish a task. One of its important application prospects is to improve the physical function of the disabled. However, the low level of the coordination between human and robot and the limited potential users still hamper the efficiency of such systems. APPROACH: To deal with such challenges, a novel steady-state visual evoked potential (SSVEP) based human-robot coordinated brain-computer interface (BCI) system was proposed to finish a target capturing task. In this system, the robot, by combining the information obtained during the human's natural interaction with itself to capture a target, could optimize the same object capturing task and yield a better performance automatically. The combination of human dealing with the uncertainty problem and the robot dealing with the complexity problem was the key to the system. Meanwhile, an asynchronous BCI based on SSVEP was used as the system interface, and a novel asynchronous recognition algorithm was used to discriminate the electroencephalogram (EEG) signal. MAIN RESULTS: The results show that the proposed system can lower the fatigue level of the subject and simplify the operation of the system. Meanwhile, the signal recognition accuracy and the efficiency of the system were also improved. SIGNIFICANCE: Under the help of the close and natural coordination relationship design between human and robot, and the asynchronous SSVEP based BCI design which requires no limb movement to control a robot, the users would be provided with an accurate and efficient control experience. Moreover, people with severe motor diseases might potentially benefit from such a system. Also, the proposed methods can be easily integrated into other BCI diagrams, which would ameliorate the predicament of the HRC.

High-resolution EEG techniques for brain-computer interface applications.

High-resolution electroencephalographic (HREEG) techniques allow estimation of cortical activity based on non-invasive scalp potential measurements, using appropriate models of volume conduction and of neuroelectrical sources. In this study we propose an application of this body of technologies, originally developed to obtain functional images of the brain's electrical activity, in the context of brain-computer interfaces (BCI). Our working hypothesis predicted that, since HREEG pre-processing removes spatial correlation introduced by current conduction in the head structures, by providing the BCI with waveforms that are mostly due to the unmixed activity of a small cortical region, a more reliable classification would be obtained, at least when the activity to detect has a limited generator, which is the case in motor related tasks. HREEG techniques employed in this study rely on (i) individual head models derived from anatomical magnetic resonance images, (ii) distributed source model, composed of a layer of current dipoles, geometrically constrained to the cortical mantle, (iii) depth-weighted minimum L(2)-norm constraint and Tikhonov regularization for linear inverse problem solution and (iv) estimation of electrical activity in cortical regions of interest corresponding to relevant Brodmann areas. Six subjects were trained to learn self modulation of sensorimotor EEG rhythms, related to the imagination of limb movements. Off-line EEG data was used to estimate waveforms of cortical activity (cortical current density, CCD) on selected regions of interest. CCD waveforms were fed into the BCI computational pipeline as an alternative to raw EEG signals; spectral features are evaluated through statistical tests (r(2) analysis), to quantify their reliability for BCI control. These results are compared, within subjects, to analogous results obtained without HREEG techniques. The processing procedure was designed in such a way that computations could be split into a setup phase (which includes most of the computational burden) and the actual EEG processing phase, which was limited to a single matrix multiplication. This separation allowed to make the procedure suitable for on-line utilization, and a pilot experiment was performed. Results show that lateralization of electrical activity, which is expected to be contralateral to the imagined movement, is more evident on the estimated CCDs than in the scalp potentials. CCDs produce a pattern of relevant spectral features that is more spatially focused, and has a higher statistical significance (EEG: 0.20+/-0.114 S.D.; CCD: 0.55+/-0.16 S.D.; p=10(-5)). A pilot experiment showed that a trained subject could utilize voluntary modulation of estimated CCDs for accurate (eight targets) on-line control of a cursor. This study showed that it is practically feasible to utilize HREEG techniques for on-line operation of a BCI system; off-line analysis suggests that accuracy of BCI control is enhanced by the proposed method.

Structure of the cortical networks during successful memory encoding in TV commercials.

OBJECTIVE: This work intends to evaluate the functional characteristics of the cerebral network during the successful memory encoding of TV commercials. METHODS: We estimated the functional networks in the frequency domain from a set of high-resolution EEG data in a group of healthy subjects during the showing of commercial spots within a neutral documentary. We evaluated the differences in the cortical network associated with later remembered and not-remembered commercials by calculating the global-E(g) and local-efficiency E(l) indexes. RESULTS: Successful encoding of TV spots significantly affects the functional communication among cortical areas irrespectively of the frequency band. During the visualization of the video-clips that will be forgotten (FRG), the cortical network exhibited high values of global- and local-efficiency, reflecting a small-world configuration. During the visualization of the video-clips that will be remembered (RMB), the same indexes appeared significantly lower. This difference was especially evident for E(l) in the alpha band and for E(g) in the beta and gamma bands. CONCLUSIONS: The presence of attentional and semantic processes during the RMB condition leads to a significant reduction of the network communication efficiency. SIGNIFICANCE: Such a reduction could represent a predictive measure of accurate recalls of TV spots.

Bipolar electrode selection for a motor imagery based brain-computer interface.

A motor imagery based brain-computer interface (BCI) provides a non-muscular communication channel that enables people with paralysis to control external devices using their motor imagination. Reducing the number of electrodes is critical to improving the portability and practicability of the BCI system. A novel method is proposed to reduce the number of electrodes to a total of four by finding the optimal positions of two bipolar electrodes. Independent component analysis (ICA) is applied to find the source components of mu and alpha rhythms, and optimal electrodes are chosen by comparing the projection weights of sources on each channel. The results of eight subjects demonstrate the better classification performance of the optimal layout compared with traditional layouts, and the stability of this optimal layout over a one week interval was further verified.

A brain-computer interface using motion-onset visual evoked potential.

This paper presents a novel brain-computer interface (BCI) based on motion-onset visual evoked potentials (mVEPs). mVEP has never been used in BCI research, but has been widely studied in basic research. For the BCI application, the brief motion of objects embedded into onscreen virtual buttons is used to evoke mVEP that is time locked to the onset of motion. EEG data registered from 15 subjects are used to investigate the spatio-temporal pattern of mVEP in this paradigm. N2 and P2 components, with distinct temporo-occipital and parietal topography, respectively, are selected as the salient features of the brain response to the attended target that the subject selects by gazing at it. The computer determines the attended target by finding which button elicited prominent N2/P2 components. Besides a simple feature extraction of N2/P2 area calculation, the stepwise linear discriminant analysis is adopted to assess the target detection accuracy of a five-class BCI. A mean accuracy of 98% is achieved when ten trials data are averaged. Even with only three trials, the accuracy remains above 90%, suggesting that the proposed mVEP-based BCI could achieve a high information transfer rate in online implementation.

Study on connectivity between coherent central rhythm and electromyographic activities.

Whether afferent feedback contributes to the generation of cortico-muscular coherence (CMCoh) remains an open question. In the present study, a multivariate autoregressive (MVAR) model and partial directed coherence (PDC) were applied to investigate the causal influences between the central rhythm and electromyographic (EMG) signals in the process of CMCoh. The system modeling included activities from the contralateral and ipsilateral primary sensorimotor cortex (M1/S1), supplementary motor area (SMA) and the time series from extensor carpi radialis (ECR) muscles. The results showed that afferent sensory feedback could also play an important role for the generation of CMCoh. Meanwhile, significant coherence between the EMG signals and the activities in the SMA was found in two subjects out of five. Connectivity analysis revealed a significant descending information flow which possibly reflected direct recruitment on the motoneurons from the SMA to facilitate motor control.

Classifying single-trial EEG during motor imagery by iterative spatio-spectral patterns learning (ISSPL).

In most current motor-imagery-based brain-computer interfaces (BCIs), machine learning is carried out in two consecutive stages: feature extraction and feature classification. Feature extraction has focused on automatic learning of spatial filters, with little or no attention being paid to optimization of parameters for temporal filters that still require time-consuming, ad hoc manual tuning. In this paper, we present a new algorithm termed iterative spatio-spectral patterns learning (ISSPL) that employs statistical learning theory to perform automatic learning of spatio-spectral filters. In ISSPL, spectral filters and the classifier are simultaneously parameterized for optimization to achieve good generalization performance. A detailed derivation and theoretical analysis of ISSPL are given. Experimental results on two datasets show that the proposed algorithm can correctly identify the discriminative frequency bands, demonstrating the algorithm's superiority over contemporary approaches in classification performance.

A study on dynamic model of steady-state visual evoked potentials.

OBJECTIVE: Significant progress has been made in the past two decades to considerably improve the performance of steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI). However, there are still some unsolved problems that may help us to improve BCI performance, one of which is that our understanding of the dynamic process of SSVEP is still superficial, especially for the transient-state response. APPROACH: This study introduced an antiphase stimulation method (antiphase: phase [Formula: see text]), which can simultaneously separate and extract SSVEP and event-related potential (ERP) signals from EEG, and eliminate the interference of ERP to SSVEP. Based on the SSVEP signals obtained by the antiphase stimulation method, the envelope of SSVEP was extracted by the Hilbert transform, and the dynamic model of SSVEP was quantitatively studied by mathematical modeling. The step response of a second-order linear system was used to fit the envelope of SSVEP, and its characteristics were represented by four parameters with physical and physiological meanings: one was amplitude related, one was latency related and two were frequency related. This study attempted to use pre-stimulation paradigms to modulate the dynamic model parameters, and quantitatively analyze the results by applying the dynamic model to further explore the pre-stimulation methods that had the potential to improve BCI performance. MAIN RESULTS: The results showed that the dynamic model had good fitting effect with SSVEP under three pre-stimulation paradigms. The test results revealed that the parameters of SSVEP dynamic models could be modulated by the pre-stimulation baseline luminance, and the gray baseline luminance pre-stimulation obtained the highest performance. SIGNIFICANCE: This study proposed a dynamic model which was helpful to understand and utilize the transient characteristics of SSVEP. This study also found that pre-stimulation could be used to adjust the parameters of SSVEP model, and had the potential to improve the performance of SSVEP-BCI.

A Dynamic Window Recognition Algorithm for SSVEP-Based Brain-Computer Interfaces Using a Spatio-Temporal Equalizer.

The past decade has witnessed rapid development in the field of brain-computer interfaces (BCIs). While the performance is no longer the biggest bottleneck in the BCI application, the tedious training process and the poor ease-of-use have become the most significant challenges. In this study, a spatio-temporal equalization dynamic window (STE-DW) recognition algorithm is proposed for steady-state visual evoked potential (SSVEP)-based BCIs. The algorithm can adaptively control the stimulus time while maintaining the recognition accuracy, which significantly improves the information transfer rate (ITR) and enhances the adaptability of the system to different subjects. Specifically, a spatio-temporal equalization algorithm is used to reduce the adverse effects of spatial and temporal correlation of background noise. Based on the theory of multiple hypotheses testing, a stimulus termination criterion is used to adaptively control the dynamic window. The offline analysis which used a benchmark dataset and an offline dataset collected from 16 subjects demonstrated that the STE-DW algorithm is superior to the filter bank canonical correlation analysis (FBCCA), canonical variates with autoregressive spectral analysis (CVARS), canonical correlation analysis (CCA) and CCA reducing variation (CCA-RV) algorithms in terms of accuracy and ITR. The results show that in the benchmark dataset, the STE-DW algorithm achieved an average ITR of 134 bits/min, which exceeds the FBCCA, CVARS, CCA and CCA-RV. In off-line experiments, the STE-DW algorithm also achieved an average ITR of 116 bits/min. In addition, the online experiment also showed that the STE-DW algorithm can effectively expand the number of applicable users of the SSVEP-based BCI system. We suggest that the STE-DW algorithm can be used as a reliable identification algorithm for training-free SSVEP-based BCIs, because of the good balance between ease of use, recognition accuracy, ITR and user applicability.