Brain-computer Interface Research Articles

The Progress and Prospects of Data Capital for Zero-Shot Deep Brain–Computer Interfaces

The vigorous development of deep learning (DL) has been propelled by big data and high-performance computing. For brain–computer interfaces (BCIs) to benefit from DL in a reliable and scalable manner, the scale and quality of data are crucial. Special emphasis is placed on the zero-shot learning (ZSL) paradigm, which is essential for enhancing the flexibility and scalability of BCI systems. ZSL enables models to generalise from limited examples to new, unseen tasks, addressing data scarcity challenges and accelerating the development of robust, adaptable BCIs. Despite a growing number of BCI surveys in recent years, there is a notable gap in clearly presenting public data resources. This paper explores the fundamental data capital necessary for large-scale deep learning BCI (DBCI) models. Our key contributions include (1) a systematic review and comprehensive understanding of the current industrial landscape of DBCI datasets; (2) an in-depth analysis of research gaps and trends in DBCI devices, data and applications, offering insights into the progress and prospects for high-quality data foundation and developing large-scale DBCI models; (3) a focus on the paradigm shift brought by ZSL, which is pivotal for the technical potential and readiness of BCIs in the era of multimodal large AI models.

Challenges and Opportunities of Gamified BCI and BMI on Disabled People Learning: A Systematic Review

This systematic review explores the potential of the gamified brain–machine interfaces (BMIs) and brain–computer interfaces (BCIs) to enhance the quality of life for individuals with disabilities. These technologies promise to solve complex problems by delivering customized interventions considering individual needs, ethical dilemmas, and practical constraints. This review follows the PRISMA statement. The search process extensively explored multiple registered databases for studies published between 2015 and 2024. Articles were selected based on strict eligibility criteria, focusing on empirical research evaluating gamified BCIs and BMIs in rehabilitation and learning. The final analysis included 56 studies. A thorough examination emphasizes the transformative potential of gamified BCIs and BMIs for people with disabilities, highlighting the need for interdisciplinary collaboration, user-centered design principles, and ethical consciousness for gamified neurotechnology. These technologies mark a significant change by providing enjoyable and effective treatments for disabled individuals. It also delves into how gamification, neurofeedback, and adaptive learning techniques can enhance motivation, engagement, and overall well-being. This evaluation underscores the efficiency of gamified BCIs and BMIs as potential instruments for improving the quality of life and empowering disabled people. However, despite their apparent potential for rehabilitation and learning, more research is needed to validate their effectiveness, accessibility, and long-term benefits.

Long‐Term Neural Recording Performance of PEDOT/CNT/Dexamethasone‐Coated Electrode Array Implanted in Visual Cortex of Rats

Implantable neural electrode arrays can be inserted in the brain to provide single‐cell electrophysiology recording for neuroscience research and brain–machine interface applications. However, maintaining signal quality over time is complicated by inflammatory tissue responses and degradation of electrode materials. Organic electrode coatings offer a solution by enhancing recording and stimulation capabilities, including reduced impedance, increased charge injection capacity, and the ability to incorporate and release anti‐inflammatory drugs. Herein, acid‐functionalized multiwalled carbon nanotubes (CNTs) loaded with dexamethasone (Dex) are incorporated into poly(3,4‐ethylendioxythiophene) (PEDOT) as electrode coatings. The electrochemical stability and recording performance of the PEDOT/CNT/Dex coating over an extended period of ≈18 months are investigated. Cyclic voltammetry (CV) stimulation is used to release Dex in half of the recording sites during the first 11 days of implantation to reduce the acute inflammation. The PEDOT/CNT/Dex‐coated floating microelectrode arrays demonstrate stable in vivo electrode impedance and successful detection of visually evoked neural activity from the rat visual cortex even at chronic time points. Additionally, the CV‐stimulated sites exhibit higher single‐unit (SU) recording yield, amplitudes, and signal‐to‐noise ratio compared to unstimulated sites. These results highlight the potential of anti‐inflammatory treatments to improve the quality and longevity of chronic neural recordings.

A cross-domain-based channel selection method for motor imagery.

Selecting channels for motor imagery (MI)-based brain-computer interface (BCI) systems can not only enhance the portability of the systems, but also improve the decoding performance. Hence, we propose a cross-domain-based channel selection (CDCS) approach, which effectively minimizes the number of EEG channels used while maintaining high accuracy in MI recognition. The EEG source imaging (ESI) technique is employed to map scalp EEG into the cortical source domain. We divide the equivalent dipoles in the source domain into different regions by k-means clustering. Then, we calculate the band energy (5-40Hz) of time series of dipoles in these regions by power spectral density (PSD), and the regions with the highest and lowest band energy are selected as the region of interests (ROIs) in the source domain. Subsequently, Pearson correlation coefficients between the dipole time series in ROIs and scalp EEG are used as the criterion for channel selection and a multi-trial-sorting-based channel selection strategy is proposed. Finally, we propose the CDCS-based MI classification framework, where common spatial pattern is applied to extract features and linear discriminant analysis is used to identify MI tasks. The CDCS method demonstrated significant improvement in decoding accuracy on two public datasets, achieving increases of 18.51% and 13.37% compared to all-channel method, and 10.74% and 3.43% compared to the three-channel method. The experimental resultsvalidated that CDCS is effective in selecting important channels.

The Application of BCI Based on Deep Learning in Stroke Treatment

Stroke is considered a global disease that leads to death and certain neural disabilities. Deep learning techniques have revolutionized EEG-based Brain-Computer Interface(BCI), enhancing reliability and promising a transformative future for BCI applications in stroke rehabilitation. However, BCI's implementation in clinical practice has been restricted due to their low accuracy performance. The objective of this review is to summarize how the integration of deep learning and BCI technologies can contribute to the rehabilitation of stroke patients. This paper compiles studies that evaluated deep learning and BCI intersection for stroke subjects, analyses the methodological quality of these studies, and verifies the relationship between the effects of the interventions and performance achieved in rehabilitation. The deficiencies and the future development direction of stroke rehabilitation in BCI with deep learning are also discussed. The various deep learning techniques combined with BCI technology, will improve people's ability to cope with stroke and provide a way to recover from stroke.

Applications of Implantable Microelectrodes in Brain-Computer Interfaces

Implantable microelectrodes represent a crucial advancement in Brain-Computer Interfaces (BCIs), enabling a wide array of applications across neuroscience and clinical domains. This paper presents a comprehensive overview of implantable microelectrodes, and three kinds of microelectrode arrays are explained in detail. The paper also deals with the integration of microelectrodes with BCI, with an emphasis on critical areas such as neuromodulation, including deep brain stimulation and neural feedback control, motor recovery, and three varieties of clinical applications in people with Parkinson's disease, epilepsy and amputees. Additionally, this work delves into the advanced hardware enhancing application of Implantable microelectrodes. This includes the development of microelectrode arrays, closed-loop systems, and the latest materials and microfabrication techniques. Despite the challenges in data processing, managing immune responses, and optimizing electrode-tissue interfaces, implantable micro-electrodes offer tremendous potential for advances in neuroscience, medicine, and rehabilitation engineering, paving the way for enhanced understanding and treatment of neurological disorders.

Depth-sensor-based shared control assistance for mobility and object manipulation: toward long-term home-use of BCI-controlled assistive robotic devices.



Objective. Assistive robots can be developed to restore or provide more autonomy for individuals with motor impairments. In particular, power wheelchairs can compensate lower-limb impairments, while robotic manipulators can compensate upper-limbs impairments. 
Recent studies have shown that Brain-Computer Interfaces (BCI) can be used to operate this type of devices.
However, activities of daily living and long-term use in real-life contexts such as home require robustness and adaptability to complex, changing and cluttered environments which can be problematic given the neural signals that do not always allow a safe and efficient use. Approach. This article describes assist-as-needed sensor-based shared control methods relying on the blending of BCI and depth-sensor-based control.
The proposed assistance targets the BCI-teleoperation of effectors for tasks that answer mobility and manipulation needs in a at-home context. Main Results. The assistance provided by the proposed methods was evaluated through a wheelchair mobility and reach-and-grasp laboratory-based experiments in a controlled environment, as part of a clinical trial with a quadriplegic patient implanted with a wireless 64-channel ElectroCorticoGram (ECoG) recording implant named WIMAGINE.
Results showed that the proposed methods can assist BCI users in both tasks.
Indeed, the time to perform the tasks and the number of changes of mental tasks were reduced. Moreover, unwanted actions, such as wheelchair collisions with the environment, and gripper opening that could result in the fall of the object were avoided. 
Significance. The proposed methods are steps toward at-home use of BCI-teleoperated assistive robots. Indeed, the proposed shared control methods improved the performance of the two assistive devices. Clinical trial, registration number: NCT02550522.

A Review of the Clinical Application of Endovascular Electrodes and Its Future Prospects

Brain-computer interface (BCI) has been used to treat neurological diseases. Traditional BCI technologies are usually invasive and may damage neural tissue during implantation. However, endovascular electrodes (EE) are a promising solution to the above problem. They represent a minimally invasive neural technique that can reduce the risks associated with surgery and its complications. In this review, we summarized the research progress on EE technology in animal models and human clinical trials. In addition, we reviewed the development of devices that can enhance EE applications, focusing on the representative product StentrodeTM. Finally, we discussed the prospects of EE technology based on current experimental results. Early animal experiments have demonstrated the safety and viability of EE, while recent human trials have shown its potential in treating diseases such as paralysis. Besides, with the development of technology, EE may be applied in a broader range of areas. In conclusion, existing studies indicate that EE can overcome the limitations of traditional BCIs and suggest a wider use of EE in the future.

Stochasticity as a solution for overfitting—A new model and comparative study on non-invasive EEG prospects

The potential and utility of inner speech is pivotal for developing practical, everyday Brain-Computer Interface (BCI) applications, as it represents a type of brain signal that operates independently of external stimuli however it is largely underdeveloped due to the challenges faced in deciphering its signals. In this study, we evaluated the behaviors of various Machine Learning (ML) and Deep Learning (DL) models on a publicly available dataset, employing popular preprocessing methods as feature extractors to enhance model training. We face significant challenges like subject-dependent variability, high noise levels, and overfitting. To address overfitting in particular, we propose using “BruteExtraTree”: a new classifier which relies on moderate stochasticity inherited from its base model, the ExtraTreeClassifier. This model not only matches the best DL model, ShallowFBCSPNet, in the subject-independent scenario in our experiments scoring 32% accuracy, but also surpasses the state-of-the-art by achieving 46.6% average per-subject accuracy in the subject-dependent case. Our results on the subject-dependent case show promise on the possibility of a new paradigm for using inner speech data inspired from LLM pretraining but we also highlight the crucial need for a drastic change in data recording or noise removal methods to open the way for more practical accuracies in the subject-independent case.

Multi-Scale Feature Extraction to Improve P300 Detection in Brain–Computer Interfaces

P300 detection is a difficult task in brain–computer interface (BCI) systems due to the low signal-to-noise ratio (SNR). In BCI systems, P300 waves are generated in electroencephalogram (EEG) signals using various oddball paradigms. Convolutional neural networks (CNNs) have previously shown excellent results for P300 detection compared to different machine learning models. However, current CNN architectures limit P300 detection accuracy because these models usually only extract single-scale features. Aiming to enhance P300 detection accuracy, an inception module-based CNN architecture, namely Inception-CNN, is introduced. Inception-CNN effectively learns discriminative features from both spatial and temporal information to reduce overfitting and computational complexity. Furthermore, it can extract multi-scale features, which effectively improves P300 detection accuracy and increases character spelling accuracy. To analyze the effect of the inception layer, two additional models are proposed: Inception-CNN-S, which uses the inception layer with a spatial convolution layer, and Inception-CNN-T, which uses the inception layer with a temporal convolution layer. The proposed model was evaluated on dataset II of BCI Competition III and dataset IIb of BCI Competition II. The experimental results show that Inception-CNN provides a promising solution for improving the accuracy of P300 detection, with F1 scores of 47.14%, 55.28%, and 78.94% for dataset II of BCI Competition III (Subject A and Subject B) and dataset IIb of BCI Competition II, respectively.

Decoding imagined movement in people with multiple sclerosis for brain–computer interface translation

Objective.Multiple sclerosis (MS) is a heterogeneous autoimmune-mediated disorder affecting the central nervous system, commonly manifesting as fatigue and progressive limb impairment. This can significantly impact quality of life due to weakness or paralysis in the upper and lower limbs. A brain-computer interface (BCI) aims to restore quality of life through control of an external device, such as a wheelchair. However, the limited BCI research in people with MS has been confined to exploring the P300 response and brain signals associated with attempted movement. The current study aims to expand the MS-BCI literature by highlighting the feasibility of decoding MS imagined movement.Approach.We collected electroencephalography data from eight participants with various symptoms of MS and ten neurotypical control participants. Participants made imagined movements of the hands and feet as directed by a go no-go protocol. Binary regularised linear discriminant analysis was used to classify imagined movement vs. rest and vs. movement at individual time-frequency points. The frequency bands which provided the maximal accuracy, and the associated latency, were compared.Main Results.In all MS participants, the classification algorithm achieved above 70% accuracy in at least one imagined movement vs. rest classification and most movement vs. movement classifications. There was no significant difference between classification of limbs with weakness or paralysis to neurotypical controls. Both the MS and control groups possessed decodable information within the alpha (7-13 Hz) and beta (16-30 Hz) bands at similar latency.Significance.This study is the first to demonstrate the feasibility of decoding imagined movements in people with MS. As an alternative to the P300 response, motor imagery-based control of a BCI may also be combined with existing motor imagery therapy to supplement MS rehabilitation. These promising results merit further long term BCI studies to investigate the effect of MS progression on classification performance.

Optimizing the proportion of stimulation area in a grid stimulus for user-friendly SSVEP-based BCIs

Objective.Steady-state visual evoked potentials (SSVEPs) rely on the photic driving response to encode electroencephalogram (EEG) signals stably and efficiently. However, the user experience of the traditional stimulation with high-contrast flickers urgently needs to be improved. In this study, we introduce a novel paradigm of grid stimulation with weak flickering perception, distinguished by a markedly lower proportion of stimulation area in the overall pattern.Approach.In an offline single-target experiment, we investigated the unique characteristics of SSVEPs evoked by varying proportions in grid stimuli within low and medium frequency bands. Based on the analysis of simulation performance across a four-class brain-computer interface (BCI) task and the evaluation of user experience questionnaires, a subset of paradigms that balance performance and comfort were selected for implementation in four-target online BCI systems.Main results.Our results demonstrate that even ultra-low stimulation proportion paradigms can still evoke strong responses within specific frequency bands, effectively enhancing user experience with low and middle frequency stimuli. Notably, proportions of 0.94% and 2.10% within the 3-5 Hz range provide an optimal balance between performance and user experience. For frequencies extending up to 15 Hz, a 2.10% proportion remains ideal. At 20 Hz, slightly higher proportions of 3.75% and 8.43% maintain these benefits.Significance.These findings are crucial for advancing the development of effective and user-friendly SSVEP-based BCI systems.

Diagonal loading common spatial patterns with Pearson correlation coefficient based feature selection for efficient motor imagery classification

The conversion of a person’s intentions into device commands through the use of brain–computer interface (BCI) is a feasible communication method for individuals with nervous system disorders. While common spatial pattern (CSP) is commonly used for feature extraction in BCIs, it has limitations. It is known for its susceptibility to noise and tendency to overfit. Moreover, high-dimensional, and irrelevant features can make it harder for a classifier to learn effectively. To address these challenges, exploring potential solutions is crucial. This paper introduces Regularized CSP with diagonal loading (DL-CSP) and Pearson correlation coefficient (PCC) based feature selection to extract the most discriminative motor imagery EEG (MI-EEG) features. Three classifiers in an ensemble are considered; bidirectional long short-term memory (Bi-LSTM), K-nearest neighbors (KNN) and naïve Bayes (NB). Decision level fusion through majority voting is exploited to leverage diverse perspectives and increase the overall system robustness. Experiments have been implemented using three publicly available datasets for MI classification; BCI competition IV–IIA (data-1), BCI Competition III–IVa (data-2), and a stroke patients’ dataset (data-3). The accuracy achieved, according to the results, is 86.96% for data-1, 91.70% for data-2, and 85.75% for data-3. These percentages outperform the accuracy achieved by any state-of-the-art techniques.

Technological Transformations, Formation of GMS (Global Mental System) and GFS (Global Forecasting System) —“Right-Brain Technologies” Based on Biological Entities with Consciousness, Artificial Intelligence, Quantum Computing, and Blockchain: “Banchenko-Market” (Global Market of Lucid Dreams and Other Transcendental States of Consciousness)

This article is a logical continuation of our previous work [1]. In this article, we continue to explore the possibilities of the economy of states of consciousness. In this continuation, drawing on intuitive insights, we significantly develop the theme outlined in the previous article [1]. Modern research in the field of transmission and fixation of consciousness states opens up new possibilities for studying cognitive processes and their objectification. This article examines methodological approaches to transmitting consciousness states from one subject to another using advanced neurointerface technologies and quantum-resistant blockchain systems. Special attention is paid to the objectification of consciousness states through stable patterns of neural activity, captured using technologies such as electroencephalography (EEG), magnetoencefalografi (MEG), and functional magnetic resonance imaging (fMRI). Algorithms of artificial intelligence (AI) are applied for processing and transmitting data, enabling classification and analysis of neural signals. The article also discusses the theoretical aspects of the quantum nature of consciousness, implying the use of quantum processes in the brain to explain cognitive states. The connection of these processes with quantum principles such as superposition and entanglement provides an opportunity for the application of quantum technologies in transmitting data about the state of consciousness. Within the study, quantum-resistant cryptographic systems are proposed, such as Quantum Resistant Ledger (QRL), which ensure data transmission security and protection against hacking by quantum computers. In this article, we use the term neurointerface as the closest to the goals of our research, which involve seeking technological solutions for transferring states from one biologically conditioned subject to another. As the term brain-computer interface implies interaction between the brain and a computer, which is not the focus of our research, we do not use this term in this article. In conclusion, ethical and philosophical aspects of transmitting consciousness states are discussed, as well as the prospects for further research in the integration of technologies and cognitive processes.

AMEEGNet: attention-based multiscale EEGNet for effective motor imagery EEG decoding

Recently, electroencephalogram (EEG) based on motor imagery (MI) have gained significant traction in brain-computer interface (BCI) technology, particularly for the rehabilitation of paralyzed patients. But the low signal-to-noise ratio of MI EEG makes it difficult to decode effectively and hinders the development of BCI. In this paper, a method of attention-based multiscale EEGNet (AMEEGNet) was proposed to improve the decoding performance of MI-EEG. First, three parallel EEGNets with fusion transmission method were employed to extract the high-quality temporal-spatial feature of EEG data from multiple scales. Then, the efficient channel attention (ECA) module enhances the acquisition of more discriminative spatial features through a lightweight approach that weights critical channels. The experimental results demonstrated that the proposed model achieves decoding accuracies of 81.17, 89.83, and 95.49% on BCI-2a, 2b and HGD datasets. The results show that the proposed AMEEGNet effectively decodes temporal-spatial features, providing a novel perspective on MI-EEG decoding and advancing future BCI applications.

A technological proposal for a robot brain computer interface for neurorehabilitation purposes

A technological proposal for a robot brain computer interface for neurorehabilitation purposes

Towards neuromorphic compression based neural sensing for next-generation wireless implantable brain machine interface

Abstract This work introduces a neuromorphic compression based neural sensing architecture with address-event representation inspired readout protocol for massively parallel, next-gen wireless implantable brain machine interface (iBMI). The architectural trade-offs and implications of the proposed method are quantitatively analyzed in terms of compression ratio and spike information preservation. For the latter, we used metrics such as root-mean-square error and correlation coefficient between the original and recovered signals to assess the effect of neuromorphic compression on the spike shape. Furthermore, we use accuracy, sensitivity, and false detection rate to understand the effect of compression on downstream iBMI tasks, specifically, spike detection. We demonstrate that a data compression ratio of 15-265 per channel can be achieved by transmitting address-event pulses for two different biological datasets. The compression ratio increases to 200-50K per channel, 50× more than in prior works, by selective transmission of event pulses corresponding to neural spikes. A correlation coefficient of ∼0.9 and spike detection accuracy of over 90% were obtained for the worst-case analysis involving 10K-channel simulated recording and typical analysis using 100 or 384-channel real neural recordings. We also analyzed the collision handling capability for up to 10K channels and observed no significant error, indicating the scalability of the proposed pipeline. We also present initial results to show the ability of intention decoders to work directly on the events generated by the neuromorphic front-end.

A Spiking Neural Network Approach for Classifying Hand Movement and Relaxation from EEG Signal using Time Domain Features

High-performance prosthetic and exoskeleton systems based on EEG signals can improve the quality of life of hand-impaired people. Effective controlling of these assistive devices requires accurate EEG signal classification. Although there have been advancements in the assistive Brain-Computer Interface (BCI) systems, still classifying the EEG signals with high accuracy is a great challenge. The objective of this research is to investigate the accuracy of the EEG signal classification of the Spiking Neural Network (SNN) classifier for factual and exact control of prosthetic and exoskeleton systems for individuals with hand impairment. The EEG dataset has been taken from the BNCI Horizon 2020 website, which is for hand movement-relax events of a patient with high spinal cord injury (SCI) to operate a neuro-prosthetic device attached to the paralyzed right upper limb. The fusion of Dispersion Entropy (DE), Fuzzy Entropy (FE), and Fluctuation based Dispersion Entropy (FDE) with mean and skewness features are extracted from the Motor Imagery (MI) EEG signals and applied to the Spiking Neural Network (SNN) classifier. To compare the performance of this algorithm, these same features have been used in Convolutional Neural Network (CNN), Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression (LR) classifiers. It has been found that SNN has given the highest classification accuracy of 80% with a precision of 80.95%, recall of 77.28%, and F1-score of 79.07%. This indicates that SNN with these five features has greater potential in BCI system-based applications.

Generative Diffusion-Based Task Incremental Learning Method for Decoding Motor Imagery EEG

Background/Objectives: Motor neurorehabilitation can be realized by gradually learning diverse motor imagery (MI) tasks. EEG-based brain-computer interfaces (BCIs) provide an effective solution. Nevertheless, existing MI decoding methods cannot balance plasticity for unseen tasks and stability for old tasks. This paper proposes a generative diffusion-based task Incremental Learning (IL) method called GD-TIL. Methods: First, data augmentation is employed to increase data diversity by segmenting and recombining EEG signals. Second, to capture temporal-spatial features (TSFs) from different temporal resolutions, a multi-scale temporal-spatial feature extractor (MTSFE) is developed via integrating multiscale temporal-spatial convolutions, a dual-branch pooling operation, multiple multi-head self-attention mechanisms, and a dynamic convolutional encoder. The proposed self-supervised task generalization (SSTG) mechanism introduces a regularization constraint to guide MTSFE and unified classifier updating, which combines labels and semantic similarity between the augmentation with original views to enhance model generalizability for unseen tasks. In the IL phase, a prototype-guided generative replay module (PGGR) is used to generate old tasks’ TSFs by training a lightweight diffusion model based on the prototype and label of each task. Furthermore, the generated TSF is merged with a new TSF to fine-tune the convolutional encoder and update the classifier and PGGR. Finally, GD-TIL is evaluated on a self-collected ADL-MI dataset with two MI pairs and a public dataset with four MI tasks. Results: The continuous decoding accuracy reaches 80.20% and 81.32%, respectively. The experimental results exhibit the excellent plasticity and stability of GD-TIL, even beating the state-of-the-art IL methods. Conclusions: Our work illustrates the potential of MI-based BCI and generative AI for continuous neurorehabilitation.

Toward the Construction of Affective Brain-Computer Interface: A Systematic Review

Electroencephalogram(EEG)-based affective computing aims to recognize the emotional state, which is the core technology of affective brain-computer interface(aBCI). This concept encompasses aspects of physiological computing, human-computer interaction(HCI), mental health care, and brain-computer interfaces(BCI), presenting significant theoretical and practical value. However, the field reached a bottleneck stage due to EEG individual difference issues, causing various challenges to achieve a fundamental aBCI. In this review, we collected some representative works from 2019 to 2023. Combining the historical exploration process and research approaches of EEG-based emotion recognition, a comprehensive understand of current research status was conducted. Furthermore, we analyzed the main obstacles for emotion recognition modeling. To construct a reasonable aBCI, we envisioned the working scenarios, developmental stages, and key impact factors based on the existing EEG physiology knowledge. From the practical application perspective, we evaluated the theoretical significance, implementation difficulty, and real-world limitations of different approaches. By synthesizing the merits and drawbacks of various techniques, we proposed a theoretically feasible aBCI framework under the restrictions of real-world application scenarios. Finally, we suggested several research topics that have not been thoroughly investigated to broaden the research scope and accelerate the development of aBCIs.