Search

Research and development of electroencephalogram (EEG) based brain-computer interfaces (BCIs) have advanced rapidly, partly due to deeper understanding of the brain and wide adoption of sophisticated machine learning approaches for decoding the EEG signals. However, recent studies have shown that machine learning algorithms are vulnerable to adversarial attacks. This paper proposes to use narrow period pulse for poisoning attack of EEG-based BCIs, which makes adversarial attacks much easier to implement. One can create dangerous backdoors in the machine learning model by injecting poisoning samples into the training set. Test samples with the backdoor key will then be classified into the target class specified by the attacker. What most distinguishes our approach from previous ones is that the backdoor key does not need to be synchronized with the EEG trials, making it very easy to implement. The effectiveness and robustness of the backdoor attack approach is demonstrated, highlighting a critical security concern for EEG-based BCIs and calling for urgent attention to address it.

As neuroscience projects increase in scale and cross international borders, different ethical principles, national and international laws, regulations, and policies for data sharing must be considered. These concerns are part of what is collectively called data governance. Whereas neuroscience data transcend borders, data governance is typically constrained within geopolitical boundaries. An international data governance framework and accompanying infrastructure can assist investigators, institutions, data repositories, and funders with navigating disparate policies. Here, we propose principles and operational considerations for how data governance in neuroscience can be navigated at an international scale and highlight gaps, challenges, and opportunities in a global brain data ecosystem. We consider how to approach data governance in a way that balances data protection requirements and the need for open science, so as to promote international collaboration through federated constructs such as the International Brain Initiative (IBI).

The Eighth International Brain-Computer Interface (BCI) Meeting was held June 7-9, 2021 in a virtual format. The conference continued the BC Meeting series' interactive nature with 21 workshops covering the breadth of topics in BCI (also called brain-machine interface) research. Some workshops provided detailed examinations of methods, hardware, or processes. Others focused on BCI applications or user groups. Several workshops continued consensus building efforts designed to create BCI standards and improve comparisons between studies and the potential for meta-analysis and large multi-site clinical trials. Ethical and translational considerations were the primary topic for some workshops or an important secondary consideration for others. The range of BCI applications continues to expand, with more workshops focusing on approaches that can extend beyond the needs of those with physical impairments. This paper summarizes each workshop, provides background information and references for further study, summarizes discussions, and describes the resulting conclusion, challenges, or initiatives.

Coupling behavioral measures and brain imaging in naturalistic, ecological conditions is key to comprehend the neural bases of spatial navigation. This highly integrative function encompasses sensorimotor, cognitive, and executive processes that jointly mediate active exploration and spatial learning. However, most neuroimaging approaches in humans are based on static, motion-constrained paradigms and they do not account for all these processes, in particular multisensory integration. Following the Mobile Brain/Body Imaging approach, we aimed to explore the cortical correlates of landmark-based navigation in actively behaving young adults, solving a Y-maze task in immersive virtual reality. EEG analysis identified a set of brain areas matching state-of-the-art brain imaging literature of landmark-based navigation. Spatial behavior in mobile conditions additionally involved sensorimotor areas related to motor execution and proprioception usually overlooked in static fMRI paradigms. Expectedly, we located a cortical source in or near the posterior cingulate, in line with the engagement of the retrosplenial complex in spatial reorientation. Consistent with its role in visuo-spatial processing and coding, we observed an alpha-power desynchronization while participants gathered visual information. We also hypothesized behavior-dependent modulations of the cortical signal during navigation. Despite finding few differences between the encoding and retrieval phases of the task, we identified transient time-frequency patterns attributed, for instance, to attentional demand, as reflected in the alpha/gamma range, or memory workload in the delta/theta range. We confirmed that combining mobile high-density EEG and biometric measures can help unravel the brain structures and the neural modulations subtending ecological landmark-based navigation.

Abstract Research and development of electroencephalogram (EEG) based brain-computer interfaces (BCIs) have advanced rapidly, partly due to the wide adoption of sophisticated machine learning approaches for decoding the EEG signals. However, recent studies have shown that machine learning algorithms are vulnerable to adversarial attacks, e.g., the attacker can add tiny adversarial perturbations to a test sample to fool the model, or poison the training data to insert a secret backdoor. Previous research has shown that adversarial attacks are also possible for EEG-based BCIs. However, only adversarial perturbations have been considered, and the approaches are theoretically sound but very difficult to implement in practice. This article proposes to use narrow period pulse for poisoning attack of EEG-based BCIs, which is more feasible in practice and has never been considered before. One can create dangerous backdoors in the machine learning model by injecting poisoning samples into the training set. Test samples with the backdoor key will then be classified into the target class specified by the attacker. What most distinguishes our approach from previous ones is that the backdoor key does not need to be synchronized with the EEG trials, making it very easy to implement. The effectiveness and robustness of the backdoor attack approach is demonstrated, highlighting a critical security concern for EEG-based BCIs.

Abstract An electroencephalogram (EEG)-based brain–computer interface (BCI) speller allows a user to input text to a computer by thought. It is particularly useful to severely disabled individuals, e.g. amyotrophic lateral sclerosis patients, who have no other effective means of communication with another person or a computer. Most studies so far focused on making EEG-based BCI spellers faster and more reliable; however, few have considered their security. This study, for the first time, shows that P300 and steady-state visual evoked potential BCI spellers are very vulnerable, i.e. they can be severely attacked by adversarial perturbations, which are too tiny to be noticed when added to EEG signals, but can mislead the spellers to spell anything the attacker wants. The consequence could range from merely user frustration to severe misdiagnosis in clinical applications. We hope our research can attract more attention to the security of EEG-based BCI spellers, and more broadly, EEG-based BCIs, which has received little attention before.

Item does not contain fulltext, Results that do not confirm expectations are generally referred to as 'negative' results. While essential for scientific progress, they are too rarely reported in the literature - Brain-Machine Interface (BMI) research is no exception. This led us to organize a workshop on BMI negative results during the 2018 International BCI meeting. The outcomes of this workshop are reported herein. First, we demonstrate why (valid) negative results are useful, and even necessary for BMIs. These results can be used to confirm or disprove current BMI knowledge, or to refine current theories. Second, we provide concrete examples of such useful negative results, including the limits in BMI-control for complete locked-in users and predictors of motor imagery BMI performances. Finally, we suggest levers to promote the diffusion of (valid) BMI negative results, e.g. promoting hypothesis-driven research using valid statistical tools, organizing special issues dedicated to BMI negative results, or convincing institutions and editors that negative results are valuable.