eeg-self-initiated-attention-shifts

name

eeg-self-initiated-attention-shifts

description

Subject-specific analysis of self-initiated attention shifts from EEG using interpretable machine learning. Demonstrates reliable within-subject classification of preparatory EEG activity distinguishing self-initiated vs externally instructed attention shifts. Uses SHAP feature attribution to identify spectral-spatial contributions. Applicable to: personalized BCI, asynchronous brain-machine interfaces, attention decoding, EEG-based voluntary intent detection. Activation: self-initiated attention, EEG attention shifts, voluntary attention, asynchronous BCI, SHAP EEG analysis, subject-specific EEG, preparatory EEG, attention decoding, internal vs external attention. Based on arXiv:2605.18251 (May 2026).

EEG Self-Initiated Attention Shifts Analysis

Paper Details

• Title: Subject-Specific Analysis of Self-Initiated Attention Shifts from EEG with Controlled Internal and External Attention Conditions
• arXiv: 2605.18251 (2026-05-18)
• Authors: Yuwen Zeng, Dengzhe Hou, Zhang Zhang, Sai Sun, Yongsong Huang et al.
• Categories: eess.SP, cs.LG, q-bio.NC

Core Problem

Self-initiated attention shifts are critical for voluntary behavior but lack explicit temporal markers, making them difficult to study. This paper asks: can preparatory EEG activity distinguish self-initiated from externally instructed attention shifts?

Experimental Paradigm

• Controlled comparison: Task-constrained self-initiated shifts vs externally instructed shifts
• Identical visual stimulation across conditions (eliminates sensory confounds)
• EEG recording during preparatory period before attention shift execution
• Machine learning classification to detect which type of shift is being prepared

Key Findings