Neural Signatures and Multi-Cognitive Decoding of EEGSignals Induced by Shared Stimulus: A Paradigm Study.

Area of Science:

• Neuroscience
• Cognitive Science
• Biomedical Engineering

Background:

• Multi-task decoding from electroencephalogram (EEG) signals is crucial for brain-computer interface (BCI) development.
• Existing research often overlooks the nuanced cognitive responses to a single stimulus.

Purpose of the Study:

• To introduce a novel paradigm for decoding multiple cognitive states from EEG signals elicited by a common visual stimulus.
• To investigate the potential of EEG-based BCI to differentiate diverse cognitive processes.

Main Methods:

• A novel experimental paradigm presenting a single visual stimulus to elicit five distinct cognitive processes (single, interception, sequence, attention, inhibition reach).
• Analysis of EEG signatures using temporal and spectral methods, including event-related spectral perturbation (ERSP).
• Decoding using a regularized linear discriminant analysis (RLDA) classifier with temporal and ERSP features.

Main Results:

• Significant neural activation differences were observed across tasks and brain regions (p < 0.05).
• The RLDA classifier achieved high decoding accuracy of 91.72% ± 6.10% for classifying the five cognitive states using ERSP features.
• Temporal features enabled classification of normal versus catch trials for the sequence reach task with 77.96% ± 7.03% accuracy.

Conclusions:

• EEG-based BCI can effectively distinguish diverse cognitive states evoked by identical stimuli.
• Findings offer insights for enhancing the naturalness and intelligence of BCI systems.
• Future work will focus on improving decoding performance and enabling online BCI applications.