Imperceptible grating based Steady-State Motion Visual Evoked Potentials Brain-Computer Interface for spatial navigation

Date

2024-7-5

Author

Atabek, Bartu

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

149
views

25
downloads

Cite This

Brain-computer interfaces (BCIs) offer promising solutions for individuals with motor impairments while enhancing human-computer interaction for non-disabled users in virtual reality and cognitive augmentation. However, widespread adoption is hindered by user fatigue and the unnatural feel of traditional visual stimuli, necessitating more comfortable, intuitive BCI paradigms that maintain high performance. This study aims to (1) develop and evaluate a novel imperceptible steady-state motion visual evoked potential (SSMVEP) stimulus and (2) explore its potential for multi-directional spatial BCI control. The first experiment introduces a paradigm using sinusoidal gratings with high-frequency motion, comparing it against traditional SSVEP and SSMVEP stimuli. Results show the imperceptible stimulus evokes robust cortical responses while reducing user discomfort. The second experiment investigates a single imperceptible SSMVEP stimulus for spatial control. By integrating eye-tracking with EEG and employing advanced machine learning techniques, including a transformer-based model, complex attentional responses are successfully decoded with high accuracy. These findings demonstrate the feasibility of naturalistic, high-dimensional neural interfaces surpassing conventional SSVEP-BCIs’ limitations. This work contributes to developing more user-friendly, high-performance BCIs, with potential applications in assistive technologies and human-computer interaction for all individuals.

Subject Keywords

Brain–computer interface, Imperceptible steady–state motion visual evoked potential, Sinusoidal grating components, Spatial navigation, Visual fatigue reduction

URI

https://hdl.handle.net/11511/110375

Collections

Graduate School of Informatics, Thesis