Non-invasive EEG-based BCI spellers from the beginning to today: a mini-review

Abstract

The defeat of the central motor neuron leads to the motor disorders. Patients lose the ability to control voluntary muscles, for example, of the upper limbs, which introduces a fundamental dissonance in the possibility of daily use of a computer or smartphone. As a result, the patients lose the ability to communicate with other people. The article presents the most popular paradigms used in the brain-computer-interface speller system and designed for typing by people with severe forms of the movement disorders. Brain-computer interfaces (BCIs) have emerged as a promising technology for individuals with communication impairments. BCI-spellers are systems that enable users to spell words by selecting letters on a computer screen using their brain activity. There are three main types of BCI-spellers: P300, motor imagery (MI), and steady-state visual evoked potential (SSVEP). However, each type has its own limitations, which has led to the development of hybrid BCI-spellers that combine the strengths of multiple types. Hybrid BCI-spellers can improve accuracy and reduce the training period required for users to become proficient. Overall, hybrid BCI-spellers have the potential to improve communication for individuals with impairments by combining the strengths of multiple types of BCI-spellers. In conclusion, BCI-spellers are a promising technology for individuals with communication impairments. P300, MI, and SSVEP are the three main types of BCI-spellers, each with their own advantages and limitations. Further research is needed to improve the accuracy and usability of BCI-spellers and to explore their potential applications in other areas such as gaming and virtual reality.

A method for synchronized use of EEG and eye tracking in fully immersive VR

“…This study distinguishes itself because it combines a commercial EEG and a commercial VR headset with eye-tracking capabilities within a hybrid BCI Speller, set in a fully immersive VR environment. Both the accuracy and ITR of the presented BCI speller are lower to comparable stateof-the-art BCI spellers (Wen et al, 2021;Maslova et al, 2023). However, the primary intention of the study is to display the feasibility of the synchronized ecosystem, and this was also the main concern when implementing the speller.…”

Brain-computer interfaces and human factors: the role of language and cultural differences—Still a missing gap?

“…Likewise, the user’s traits or states [e.g., the personality or motivation, current mood, emotion or affect or the cognitive skills (e.g., attention, memory or imagery) of the user] and their impacts on BCI performance have been examined among healthy users or patients. This included well-established BCI systems such as the P300-BCI 2 (for a recent reviews, e.g., Maslova et al, 2023 ) or SMR-BCI (for an overview, e.g., Grosse-Wentrup and Schölkopf, 2013 ). For a joint review of psychological human factors and performance in P300- and SMR-BCIs of healthy and vulnerable BCI users, see Herbert (2024) in the Research Topic.…”

A toolbox for decoding BCI commands based on event-related potentials

“…Here we extend the variety of publicly available algorithm implementations and introduce the ERPCCA toolbox 1 , which implements CCA as a tool for extracting canonical ERPs. The strength of the toolbox is the detection of a user’s intention that results from a series of stimuli, such as in P300-based speller applications ( Rezeika et al, 2018 ; Maslova et al, 2023 ). Likewise, the ERPCCA toolbox is suitable for binary single event decoding such as the detection of error-related potentials.…”