Papers

Gabriele Cinà reported in an arXiv preprint posted on September 27, 2026, that Jev, a hosted model, can replace the multi-billion-parameter language model that speech neuroprostheses use to rescore candidate sentences, the only GPU-dependent step in the system. Jev returns a probability for each candidate in one call, which is combined with the decoder's own score. On 978 held-out sentences from a participant with ALS, the published decoder alone had an 8.1% word error rate, versus 7.5% with Jev and 7.8% with both OPT-6.7b and Qwen2.5-7B; after re-tuning decoder weights, the figures were 6.9%, 7.2% and 7.4% respectively. Jev needs no GPU, costs $0.07 per thousand sentences, and has 262 ms end-to-end latency over the internet, versus 27 ms for a 7B model on a local GPU. The work is a preprint and has not been peer reviewed.

A collaboration involving the University of Geneva and the CIBM Center for Biomedical Imaging compared resting-state EEG under dry, semi-dry and gel conditions in 22 healthy participants in one session. Hair-wetting with tap water reduced bad channels from about 40% to 15% and improved mean electrode impedance by 75.7%. Eyes-closed spectral-power similarity with gel recordings increased by 7.7%, while functional-connectivity similarity improved by about 56–70% across frequency bands. Signal quality did not fully match gel systems, and the study did not test BCI task performance.

Researchers at Tianjin University of Technology and Education in northern China built a 64-target visual evoked potential brain-computer interface that encodes stimuli as relative changes among five colors (red, green, blue, yellow and gray) with grouped modulation, replacing the high-contrast black-and-white flicker that tires the eyes and limits the number of options. Their decoder uses task-discriminant component analysis (TDCA) to extract target-related components from EEG, then classifies them with a lightweight multilayer perceptron (MLP). In experiments with 10 participants conducted from April to June 2026, the TDCA-MLP algorithm averaged 93.72% accuracy, outperforming 4 commonly used algorithms. The authors say the system requires less calibration and offers a better subjective viewing experience.

A new review surveys the past decade of signal-processing algorithms and hardware for EEG-based motor imagery brain-computer interfaces, which let users drive external devices or retrain motor function simply by imagining a movement. In these systems, users imagine moving a hand or foot while scalp electrodes record signals from the sensorimotor cortex, and algorithms translate them into commands. The authors survey the more practical algorithms of the past ten years, focusing on how machine learning and deep learning improve performance and efficiency. They also review hardware and supporting technologies, including system-on-chip designs, application-specific chips, programmable chips, wearables, the Internet of Things and augmented and virtual reality, discuss how these pair with algorithms, and outline research directions for better real-time performance. The paper is a preprint and has not been peer reviewed.

A review by a team including Tianjin University sorts ways to slim down deep-learning decoders for motor imagery EEG, which are accurate but too large for wearable devices, into two routes: redesigning convolutional neural networks (depthwise separable, multi-branch and dilated convolutions, attention mechanisms) and compressing existing models (pruning, quantization, knowledge distillation, low-rank decomposition). It compares representative lightweight models on parameter count, computational cost and classification accuracy, and recommends shifting toward automated neural architecture search and hardware-software co-design.

A network meta-analysis of 22 randomized study cohorts with 727 participants compared BCI-coupled robotic rehabilitation, robot-assisted therapy alone, motor imagery and conventional rehabilitation for arm weakness after stroke. BCI-coupled robotic rehabilitation ranked highest on upper-limb motor function scores, averaging 6.80 points above conventional rehabilitation, a point estimate above the 5.25-point minimal clinically important difference, although the lower confidence limit fell short of it. Its advantage over robot-assisted therapy alone was only 2.04 points, which the authors judged unlikely to be clinically important.

Researchers have proposed SICMD, a method that combines two non-invasive brain signals, functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG), to identify speech a person is hearing from brain activity with a single model that works across different people. fMRI pinpoints where activity occurs and MEG captures when, together giving fuller spatiotemporal information. In cross-subject perceived-speech decoding, the method improved Top-1, Top-10 and Rankacc by more than 10.6%, 10.1% and 1.7% over baseline methods, and cut training costs by 88.8% and 60.5% compared with multi-subject and intra-subject decoding respectively. The work is a preprint and has not been peer reviewed.

Researchers have shown that language models do more to help decode imagined speech than heard speech from non-invasive brain recordings. Imagined speech is harder to read than speech a person hears: the brain's response is weaker and noisier, and less data is available. Using a magnetoencephalography (MEG) dataset in which participants both listened to and imagined the same material, the team brought a language model into both training and decoding. Imagined speech gained more from the model than heard speech, and its best results required leaning more heavily on the model. The authors conclude that the weaker the neural evidence, the more useful a language model's contextual prior becomes. The study is a preprint and has not been peer reviewed.

A brain-computer interface that times a nerve stimulus to the moment the brain prepares a wrist movement raised the excitability of the motor pathway from cortex to muscle in 17 able-bodied volunteers, whether they actually moved the wrist or only imagined it. Gains persisted 30 minutes after the session, and the two approaches did not differ significantly. The authors say this gives clinicians flexibility for patients who cannot move, though the study did not include stroke patients and individual responses varied widely.

The study tested seven detection methods to assess whether motor imagery BCIs can identify movement classes absent from training rather than force them into known categories. For many users, model uncertainty was higher for familiar classes than for unfamiliar ones, undermining methods that work well in other machine learning domains. Detection performance was correlated with performance on the trained task. Deep Ensemble and Monte Carlo-Dropout models achieved an on-task AUC of 0.9 and out-of-distribution detection ability up to 0.7. The authors conclude that rejecting unfamiliar cognitive states is feasible for models and participants with high task performance.

A collaboration involving the University of Cambridge analyzed surface electrocorticography recordings from the human ventral anterior temporal lobe. It found significant decoding of graded, multidimensional semantic information when signals across 4-200 Hz were considered together, but not from individual frequency bands. Using representational similarity learning on time-frequency power and phase data, the researchers compared three hypotheses about how semantic information maps onto signal frequencies. The authors interpret the findings as supporting transfrequency coding, in which some information emerges only when multiple frequencies are considered jointly. The study has not been peer reviewed.

A single-author study presents a hybrid algorithm for telling apart spikes from the many neurons an implanted electrode records at once, a step a brain-computer interface (BCI) must take before it can decode intent. A deep neural network assigns each spike to a neuron and estimates its waveform, fuzzy-logic methods detect spikes and score confidence when signal quality is uncertain, and the network architecture was chosen by multi-objective selection. On four datasets from rat cortex, hippocampal CA1 and monkey motor cortex, it reached 96.4% accuracy, 94.8% precision and a mean latency of 45.2 ms with full labels, and 88.9% accuracy with only 10% of data labeled, which the author says is comparable to existing unsupervised methods while also providing confidence and waveform estimates. The author notes that prospective validation is needed before clinical use.

A team including the University of Science and Technology Beijing reviewed the open-source software ecosystem for non-invasive, EEG-based brain-computer interfaces, which read brain signals through scalp electrodes without surgery. The review covers four dimensions, technical foundations, tool taxonomy, scenario adaptation and future evolution, shows how open-source tools span the full workflow from signal acquisition to real-time interaction, classifies them by application such as healthcare, education and entertainment, and identifies the ecosystem's current technical challenges, offering developers a method for choosing tools.

Researchers re-examined a multi-class imagined speech EEG dataset, words spoken silently in the mind, under a strictly subject-independent protocol in which a model is never trained on data from the person it is tested on. Spectral bandpower features reached a mean trial-wise accuracy of 49.03% ± 4.18% in coarse-level classification, clearly ahead of the 37.97% ± 3.79% from time-domain statistical features, and a few frequency bands carried most of the discriminative information. The study is a preprint and has not been peer reviewed.

Researchers tested whether shrinking brain-computer interface (BCI) decoders through pruning or low-precision quantization, common steps for running them on low-power devices, changes how they hold up against adversarial attacks, in which imperceptible perturbations to EEG signals trick a model into misclassifying. Compressing a model while preserving accuracy did not make it more robust: at a perturbation strength of ε=0.005, EEGNet accuracy fell to 22% to 24% in its original, 50% pruned and two INT8 quantized versions alike. Attacks transferred markedly less well between pruned and original models, while quantization left transfer almost unchanged. The study is a preprint and has not been peer reviewed.

Researchers propose DenoFlow, an EEG denoising method designed to remove interference from muscle and eye electrical activity while preserving information that a brain-computer interface can decode. The authors report better signal fidelity and decoding accuracy than seven baseline denoising models in tests on two public steady-state visual evoked potential (SSVEP) datasets with five decoders. They created paired contaminated and original signals by injecting recorded muscle and eye artifacts at controlled signal-to-noise targets. The findings come from these controlled dataset tests, and the study has not been peer reviewed.

Canyang Zhao and four co-authors posted a preprint on arXiv on September 23, 2026, proposing Task-Conditioned Latent Alignment (TCLA) to tackle a long-standing problem with implanted brain-computer interfaces: the neurons an electrode records change from day to day, so a decoder trained yesterday may fail today. TCLA combines neural signal reconstruction with continuous behavioral supervision to learn a low-dimensional, behavior-related shared space from one recording, then holds that space fixed and aligns each new session into it separately by task. Across seven monkey datasets, long-term cross-session tests averaged R² 0.476±0.014 with a 6.8% decoding failure rate; 1,356 within-subject session pairs averaged R² 0.371±0.009; across different animals, R² averaged 0.218±0.004 with a 12.9% failure rate. Source code is public, and the work has not been peer reviewed.

Spanish researchers asked 15 healthy participants to imagine raising an arm while watching a humanoid robot do it, viewed in immersive virtual reality, on 2D video or live in person, and recorded their brain activity with EEG. Immersive VR produced the strongest activation of the sensorimotor cortex, measured as desynchronization, a weakening of EEG rhythms that indicates a more active region. The approach of engaging motor areas by watching and imagining, without moving, underlies rehabilitation for people with limited or lost mobility, and the researchers say immersive VR could help optimize such programs and brain-computer interface applications. The study included only healthy people and has not been tested in patients.

Researchers at Lanzhou Jiaotong University proposed a hierarchical graph learning framework to decode motor imagery EEG recorded while people watch a movement in virtual reality and imagine performing it. The method splits scalp EEG into rhythmic components, selects key features, represents each trial's brain-region connectivity as a graph and classifies it with a hierarchical graph convolutional network. On the team's self-collected VR motor imagery dataset, it outperformed common baselines including EEGNet in three-class decoding. The team also found that observing plus imagining in VR strengthened connectivity among sensorimotor regions, engaging a wider brain network. The abstract does not give specific accuracy figures or the number of participants.

A systematic review of 32 studies asks whether stroke survivors can retrain paralyzed limbs by imagining movement. Most approaches had patients imagine an action, read the signal from scalp EEG, and used it to drive virtual reality, functional electrical stimulation or robotic exoskeletons, closing the feedback loop. Training ranged from small pilots to as many as 80 sessions, averaging about 30. The authors conclude such methods can significantly improve clinical outcomes in stroke survivors, though the benefit depends on the specific modality and stroke characteristics, and there is still no agreed treatment dose.

Purdue University researchers compared sparse and redundant intracortical microstimulation encoding in eight mice. Stimulating more electrodes simultaneously accelerated learning, shortened navigation trials and improved success rates and path efficiency when stimulation was combined with vision. The findings were released as a preprint.

In cross-subject transfer learning for motor imagery BCIs, picking the right 'golden subject', the person whose EEG best serves as a training template for others, directly affects decoding accuracy for other users. An earlier study using the fuzzy decision method FDOSM named S5, but that method did not separate weighting from ranking. Researchers at Imam Ja'afar Al-Sadiq University in Iraq built a unified 9 x 18 golden-subject decision matrix covering nine subjects on 18 training- and evaluation-phase metrics, including accuracy, recall, F1 score, log loss and specificity. Using FWZIC for weighting and VIKOR for compromise ranking, they found the best candidate shifted to S1 (Q = 0.0000), which ranked first in three of four weighting scenarios in a sensitivity analysis. The work is an early methodological exploration involving only nine subjects and has not been validated on larger datasets.

Muscle activity can swamp scalp EEG across a broad frequency range and bury the neural information that analysis and brain-computer interfaces rely on. Researchers built P2P Trans, a conditional adversarial denoiser that combines a U-shaped encoder-decoder, multi-head self-attention at the bottleneck, cross-attention-guided skip fusion and a PatchGAN discriminator. On synthetic data built from EEGdenoiseNet across 10 signal-to-noise levels from -7 to 2 dB, it beat 5 convolutional encoder-decoder baselines with the best mean correlation coefficient (0.8645) and lowest mean spectral error (0.3596), with the margin widening as contamination grew. It did not lead on pointwise waveform reconstruction, pointing to a trade-off between suppressing spectral noise and preserving the waveform. The work is a preprint and has not been peer reviewed.

EEG is easily contaminated by blinks, muscle activity and heartbeats, which can drag down how accurately a brain-computer interface reads intent. Researchers from the School of Automation and the School of Software at Northwestern Polytechnical University in Xi'an, northwestern China, built Diff-ADN, a single-channel denoiser that first reconstructs the signal according to contamination severity, then applies one diffusion-guided residual correction. It handles ocular, muscular, cardiac and mixed artifacts, reaching mean Pearson correlations with clean EEG of 0.839 to 0.936 at contamination levels from -7 to +2 dB. Across 4 public motor-imagery datasets, removing cardiac artifacts recovered 4.88 to 5.59 percentage points of decoding accuracy, up to 21.90 points in a single case, and a 2-second segment took 12.52 to 17.08 ms on the tested CPU. Better waveform reconstruction did not always translate into equal gains in decoding accuracy.

As brain-computer interfaces move toward clinical use, regulators, clinicians and developers need agreed rules on what counts as sufficient evidence. A new paper addresses how a clinical evidence framework for BCI can be built and applied. Only bibliographic metadata is publicly available so far, so the framework's components and intended uses cannot yet be verified.

Decoding individual words from scalp EEG and MEG is hard because word-level neural evidence is weak, spread out in time, and entangled with acoustic, lexical and semantic structure. Deep learning can already decode language from intracranial recordings, but extending that to non-invasive recordings remains unsolved. Hierarchical Dynamic Neural Decoding (HDND), proposed by Yueyang Li, Shuran Chen, Wai Ting Siok and Nizhuan Wang, treats word decoding as step-by-step refinement rather than one-shot label retrieval. Across seven EEG and MEG datasets in English, Dutch, Mandarin and Cantonese, covering nine listening, reading and reading-aloud conditions, HDND beat the matched contextual baseline on participant-averaged balanced Top-10 accuracy in all nine and had the highest mean of all compared methods in eight. The work is a preprint and has not been peer reviewed.

Researchers recorded single neurons in the motor cortex of paralyzed people with intracortical brain-computer interface implants while they watched virtual human hands, robotic hands and hand-like dots perform grasps. The more human-like the effector, the stronger the motor cortex response, both across the neural population and in individual neurons, and the effect tracked participants' own sense of human-likeness. When users attempted hand movements themselves, however, motor cortex activity stayed largely the same regardless of what was on screen, suggesting that intuitive control of external devices may not depend much on whether the device looks human.

Researchers in Portugal had seven people with chronic stroke imagine arm movements while an EEG-based brain-computer interface gave some of them real-time feedback and others sham feedback; both groups also received four weeks of standard physiotherapy. Patients receiving real feedback improved beyond the minimal clinically important difference on at least one functional measure, most clearly grip strength, and their brain activity became more symmetrical between hemispheres. As a small case series, the study shows the method is feasible, and its results are preliminary.

Portable EEG headsets have few channels and capture limited cortical information, making emotion decoding difficult. A team at Xiamen University of Technology in the eastern Chinese province of Fujian developed the Spatiotemporal Spectral Asymmetric Fusion Network (STSANet), which models nonlinear hemispheric lateralization between homologous frontotemporal electrodes. It reached 86.80% accuracy on the public SEED dataset and 87.43% on a self-collected dataset. The team also found consistent spectral energy distributions between the portable Xmuse headset and the professional-grade Enobio, suggesting the method could run on consumer EEG hardware. The work is a preprint and has not been peer reviewed.