Open research questions in EEG and Brain-Computer Interfaces

What the literature leaves open

• Finally, our current work focuses on robust EEG representation; further research is needed to extend the model to multimodal scenarios such as Video-EEG, Image-EEG, Text-EEG, and fMRI-EEG for broader applicability. A significant open challenge for the field is developing a truly universal EEG model that can generalize to novel tasks with zero- or few-shot prompting, similar to the capabilities of large language models.

  PRiSE-EEG: A Prior-Guided Foundation Model with Depth-Stratified Experts for Cross-Paradigm EEG Representation Learning · 2026

• Further research is needed to explore the applicability of the proposed SSD method to other datasets, - Future studies could investigate the use of the SSD method in combination with other artifact removal techniques

  SSDLabeler: realistic semi-synthetic data generation for multi-label artifact classification in EEG · 2026 · DOI

• Prior semi-synthetic data methods typically inject single artifact types into clean EEG, reducing realism. Existing ICA-based SSD approaches address the challenges of artifact classification only partially. There is a need for a method that can generate realistic, annotated semi-synthetic data for multi-label artifact classification.

  SSDLabeler: realistic semi-synthetic data generation for multi-label artifact classification in EEG · 2026 · DOI

• The requirement for epochs used in training, validation, and testing to come from different subjects. The need to prevent information leakage from testing or validation data into the training process. The challenge of achieving high accuracy in detecting ADHD using EEG signals.

  Multiscale deep learning convolutional neural network for ADHD detection using EEG · 2026 · DOI

• The dataset consists of only 121 participants, - The model's performance needs to be validated across diverse datasets of mental illness EEGs, - The study used a small sample for validation (4 subjects)

  Multiscale deep learning convolutional neural network for ADHD detection using EEG · 2026 · DOI

• Investigating the long-term reliability of the dry electrode system, - Comparing the dry electrode system to other commercial wet electrode systems, - Exploring the use of the dry electrode system in different clinical and research applications

  Systematic Benchmarking of a Dry Electrode EEG Prototype Against Wet Electrode EEG Systems in Electrophysiological/Cognitive Scenarios · 2026 · DOI

• The paper identifies a gap in the treatment of complete tetraplegia, where there are limited options for functional recovery. The gap is that current spinal cord stimulation techniques are restricted to muscles that show some level of residual volitional control.

  A neuroprosthesis for restoring hand movement and sensation in a person with complete tetraplegia · 2026 · DOI

• The challenges in subject-independent settings come from the unique noisy characteristics of each subject - The real-world applicability of subject-dependent and subject-semidependent settings is somewhat limited - The model learns specific subjects' distributions during training in subject-dependent and subject-semidependent settings

  EF-Net: Mental State Recognition by Analyzing Multimodal EEG-fNIRS via CNN · 2024 · DOI

• Automatically extracting features using deep learning - Learning overarching features while discarding subject-specific noise - Exploring the use of multimodal deep-learning models to analyze brain activity for both EEG and fNIRS

  EF-Net: Mental State Recognition by Analyzing Multimodal EEG-fNIRS via CNN · 2024 · DOI

• However, EEG-based diagnosis faces considerable challenges, including high-dimensional and non-stationary signals, limited data availability, and the computational complexity of deep learning models.

  Diagnosis and characterization of stroke using EEG and deep learning: emerging methods and clinical outlook · 2026 · DOI

• However, it remains unclear how many sensors are needed, where to place them, and what sampling frequency is optimal for wearable MEG devices that require cost-effective, computationally efficient operation.

  Optimal sensor set for MEG-based spontaneous and intended speech decoding toward practical communication · 2026 · DOI

• However, the use of EEG under music and no-music conditions for mental health screening remains insufficiently validated.

  Music-evoked EEG signals for mental health assessment using two- and three-class classification · 2026 · DOI

• Electroencephalography (EEG) preprocessing varies widely between studies, but its impact on classification performance remains poorly understood.

  How EEG preprocessing shapes decoding performance · 2025 · DOI

• However, the dynamic interplay between CNS and ANS rhythmicities in sleep remains unclear.

  Dynamic brain-heart interaction in sleep characterized by variational phase-amplitude coupling framework · 2025 · DOI

• However, recent conflicting evidence challenges this assumption and underscores the need to clarify the relationship between β power and movement.

  Changes in cortical beta power predict motor control flexibility, not vigor · 2025 · DOI

• BACKGROUND: As Artificial Intelligence (AI) tools like ChatGPT gain traction in clinical contexts, their role in neurorehabilitation, particularly in addressing executive function impairments associated with ADHD, remains underexplored.

  Exploring AI-assisted design of executive function rehabilitation programs for individuals with ADHD: a mixed-methods evaluation of prompts and chatgpt outputs · 2025 · DOI

• However, feasibility emerged as a key limitation, with concerns over a lack of personalization, unrealistic resource assumptions, and unvalidated techniques, particularly in adult and older adult profiles.

  Exploring AI-assisted design of executive function rehabilitation programs for individuals with ADHD: a mixed-methods evaluation of prompts and chatgpt outputs · 2025 · DOI

• No consensus exists on what electroencephalogram (EEG) data is helpful nor the optimal gait classification system.

  The classification of walking and phases of gait using EEG: a scoping review protocol. · 2025 · DOI

• BACKGROUND: Limited research exists regarding the effectiveness of electroencephalogram (EEG) neurofeedback training for children with cerebral palsy (CP) and co-occurring attention deficits (ADs), despite the increasing prevalence of these dual conditions.

  The effects of neurofeedback training for children with cerebral palsy and co‐occurring attention deficits: A pilot study · 2024 · DOI

• Cardiac function is under neural regulation; however, brain regions in the cerebral cortex responsible for regulating cardiac function remain elusive.

  Manipulation of Glutamatergic Neuronal Activity in the Primary Motor Cortex Regulates Cardiac Function in Normal and Myocardial Infarction Mice · 2024 · DOI

• Background music is widely used to sustain attention, but little is known about what musical properties aid attention.

  Rapid modulation in music supports attention in listeners with attentional difficulties · 2024 · DOI

• While impressive progress was achieved in decoding performed, perceived and attempted speech, imagined speech remains elusive, mainly due to the absence of behavioral output.

  Imagined speech event detection from electrocorticography and its transfer between speech modes and subjects · 2024 · DOI

• Yet, it remains unclear to what extent intended hand movements can be predicted from brain activity recorded during movement planning.

  Human local field potentials in motor and non-motor brain areas encode upcoming movement direction · 2024 · DOI

• Emerging trends indicate that AI, BCIs, and assistive technologies are poised to further transform neurology, - Future BCIs are expected to integrate speech, motor, sensory, and cognitive domains into unified multimodal systems, - AI-driven predictive models are advancing personalized neurology

  Emergence of Artificial Intelligence in Multiple Domains of Neurology: A Review · 2026 · DOI

• The need for more precise identification of molecular targets and disease signatures, - The need for more sophisticated applications in neurology, - The need for interdisciplinary collaboration, ethical governance, and the continued development of interpretable, equitable, and human-centered technologies

  Emergence of Artificial Intelligence in Multiple Domains of Neurology: A Review · 2026 · DOI

Most-cited papers in EEG and Brain-Computer Interfaces

• Removing electroencephalographic artifacts by blind source separation · Psychophysiology · 2000 · 2,875 citations
• ADJUST: An automatic EEG artifact detector based on the joint use of spatial and temporal features · Psychophysiology · 2010 · 1,162 citations
• Using Electroencephalography to Measure Cognitive Load · Educational Psychology Review · 2010 · 617 citations
• Breaking the silence: Brain–computer interfaces (BCI) for communication and motor control · Psychophysiology · 2006 · 547 citations
• Guidelines for the recording and quantitative analysis of electroencephalographic activity in research contexts · Psychophysiology · 1993 · 538 citations
• How about taking a low‐cost, small, and wireless EEG for a walk? · Psychophysiology · 2012 · 511 citations
• Statistical control of artifacts in dense array EEG/MEG studies · Psychophysiology · 2000 · 495 citations
• Brain–computer communication: Unlocking the locked in. · Psychological Bulletin · 2001 · 460 citations
• Automatic removal of eye movement and blink artifacts from EEG data using blind component separation · Psychophysiology · 2003 · 456 citations
• Monitoring Working Memory Load during Computer-Based Tasks with EEG Pattern Recognition Methods · Human Factors The Journal of the Human Factors and Ergonomics Society · 1998 · 433 citations

Most recent work

• mm-HrtEMO: Non-Invasive Emotion Recognition via Heart Rate Using mm-Wave Sensing in Diverse Scenarios · IEEE Journal of Biomedical and Health Informatics · 2026
• Long-term independent use of an intracortical brain–computer interface for speech and cursor control · Nature Medicine · 2026
• Moving intentions from brains to machines · Trends in Cognitive Sciences · 2026
• Organizing experiences into cognitive maps makes goal-directed action selection flexible, efficient, and explainable · bioRxiv · 2026
• Spontaneous Modulation of Alpha Power During a Neurofeedback Session Without Instructions · Psychophysiology · 2026
• Enhancing Unseen Driver State Detection: An EEG-Based Framework with Brain Connectivity and Depthwise Separable Convolutional Neural Networks · Biomedical Signal Processing and Control · 2026
• Toothy: an interactive platform for dentate spike curation · bioRxiv · 2026
• Blind Source Separation-Embedded Electroencephalogram Microstate Trajectory Modeling for Generalized Anxiety Disorder Identification · IEEE Journal of Biomedical and Health Informatics · 2026
• Multimodal machine learning for major depressive disorder: Integrating EEG functional connectivity and clinical variables to enhance diagnostic accuracy · Psychiatry Research · 2026
• Sensorimotor Frequency Tagging Is Enhanced by Auditory and Audiovisual but Not Visual, Inputs During a Body‐Walking Task · Psychophysiology · 2026

Related topics in Neuroscience