Nonlinear EEG Analysis for Distinguishing Mind Wandering and Focused Attention: A Machine Learning Approach
Sundaram, S.; Kheiri, F.; Bragin, A.
Show abstract
This study uses nonlinear analysis techniques to distinguish between mind wandering (MW) and focused attention (FA) states using EEG data. EEG recordings from 21 sessions were segmented into intervals of 2, 3, 5, 6, 10, and 15 seconds, and seven nonlinear features were extracted to capture the brains dynamic complexity. Machine learning models, including gradient boosting trees, were applied to classify MW and FA states, with the highest accuracy of 75% achieved using 5-second segments. Frequency-related features, particularly mean frequency and global frequency, were the most important in distinguishing between MW and FA. These findings emphasize the role of nonlinear EEG analysis in understanding the chaotic brain patterns underlying cognitive states. Future work should focus on temporal dynamics and personalized models to improve classification accuracy, with potential applications in cognitive enhancement and mental health.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Fat tails, fat earnings, fat mistakes: On the need to disclose distribution parameters of qEEG databases 94%
- Estimating person-specific neural correlates of mental rotation: A machine learning approach 94%
- Increased signal diversity/complexity of spontaneous EEG in humans given sub-anaesthetic doses of ketamine 93%
Similar papers in this journal
- Classification of complex emotions using EEG and virtual environment: proof of concept and therapeutic implication 95%
- Brain Functional Connectivity Correlates of Anomalous Interaction Between Sensorily Isolated Monozygotic Twins 94%
- Disrupted information flow in resting-state in adolescents with sports related concussion 94%
Similar papers in this journal
- Direct comparison of EEG resting state and task functional connectivity patterns for predicting working memory performance using connectome-based predictive modeling 94%
- Brain network dynamics correlate with personality traits 93%
- A classification-based approach to estimate the number of resting fMRI dynamic functional connectivity states 92%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.