Tactile-evoked EEG Dataset for Natural Perception Using an Integrated Stimulation-Recording Framework

Area of Science:

• Neuroscience
• Biomedical Engineering
• Human-Computer Interaction

Background:

• Aging societies increase demand for assistive living and medical technologies.
• Traditional Brain Computer Interface (BCI) systems rely on visual or auditory input, limiting accessibility.
• Developing tactile-evoked BCI systems faces challenges in quantifying sensation and data diversity.

Purpose of the Study:

• To present an integrated system for natural tactile perception and simultaneous electroencephalographic (EEG) recording.
• To investigate tactile-evoked EEG signals for improved BCI applications.
• To create a diverse dataset for tactile-related neural mechanism research.

Main Methods:

• Collected 64-channel, 1000 Hz EEG data from 10 healthy participants during natural tactile perception tasks.
• Utilized three distinct materials for tactile stimulation.
• Applied Common Spatial Pattern (CSP) and Support Vector Machine (SVM) for three-class classification of EEG signals.

Main Results:

• Identified significant differences in P300 peak latency and amplitude across different tactile conditions.
• Demonstrated above-chance classification accuracy for tactile-evoked EEG signals.
• Highlighted the unique neural characteristics of tactile perception.

Conclusions:

• The developed tactile-evoked EEG system enables natural tactile perception and signal recording.
• The findings support the potential of tactile BCI for enhanced user experiences in assistive technologies.
• The created dataset serves as a valuable resource for advancing tactile BCI research and applications.