Feature extraction and calibration of EEG signals in sitting and walking meditation

Meditation benefits physical and mental health, but walking meditation is understudied. This work recorded EEG signals from 7-Day Zen participants in four states: sitting around, strolling, sitting meditation, and walking meditation. Rhythm-based EEG features were extracted and calibrated to classify the states. Personalized calibration methods effectively addressed individual differences in EEG signals, and the system identified the four states with high accuracy, offering insight into neural activity during different meditation forms.

Study at a glance

Abstract

Meditation has been proved beneficial to people's physical and mental health in a large number of scientific reports. However, there are few studies on walking meditation. Walking meditation is a common meditation which involves walking with meditation. In order to investigate the differences in effects between sitting and walking meditation, the EEG signals of 7-Day Zen participants in the following four states are recorded: sitting around, strolling, sitting meditation, and walking meditation. Rhythm-based EEG features are extracted to identify the above four states, and features are calibrated to improve the performance of classification. The experimental results show that the personalized calibration methods effectively solve the individual differences in the EEG signals. The proposed system can identify the four states with high accuracy, which provides a further interpretation of neural actions introduced by meditation.