Related Experiment Video
Updated: Jul 10, 2026
09:37
Measurement of Neurophysiological Signals of Ignoring and Attending Processes in Attention Control
Published on: July 5, 2015
9.0K
Oscillating Mindfully: Using Machine Learning to Characterize Systems-Level Electrophysiological Activity During
Noga Aviad1, Oz Moskovich2, Ophir Orenstein2
1Observing Minds Laboratory, School of Psychological Science, University of Haifa, Haifa, Israel.
Biological Psychiatry Global Open Science
|February 6, 2025
Summary
Machine learning accurately distinguished meditation from mind-wandering states using electroencephalography (EEG) data. This approach identified key EEG features, like high-frequency oscillations, characterizing focused attention meditation.
Area of Science:
- Neuroscience
- Cognitive Science
- Computational Neuroscience
Background:
- Neuroelectrophysiological studies of mindfulness and meditation are rapidly growing.
- Traditional analysis methods struggle with the complex, nonlinear nature of meditation's neurophysiology.
- Understanding the brain's activity during meditation requires advanced analytical techniques.
Purpose of the Study:
- To reveal the complex, systemic neuroelectrophysiology of meditation states.
- To apply machine learning for analyzing electroencephalography (EEG) data during meditation.
- To identify specific EEG features that characterize focused attention meditation.
Main Methods:
- Applied an extreme gradient boosting classification algorithm to EEG data.
- Utilized 4 complementary feature importance methods.
- Recorded EEG from 26 experienced meditators during focused attention meditation and mind-wandering states.
Main Results:
- The machine learning algorithm achieved 83% accuracy in classifying meditation versus mind-wandering states.
- Area under the ROC curve was 79%, and F1 score was 74%.
- Identified 10 EEG features, including increased high-frequency power and coherence, associated with meditation.
Conclusions:
- The findings delineate the complex systemic oscillatory activity characterizing meditation.
- Machine learning provides a powerful tool for analyzing neurophysiological data in meditation research.
- Specific EEG patterns are linked to the focused attention state in experienced meditators.
