EEG-Based Differentiation of Forest and Urban Walking Environments Using Machine Learning.
Summary
Walking in forests significantly alters brain activity compared to urban walks, as shown by electroencephalography (EEG) and machine learning. Nature exposure offers distinct cognitive benefits, impacting neural patterns.
Area of Science:
- Environmental Neuroscience
- Cognitive Neuroscience
- Neuroscience
Background:
- Understanding environmental impacts on neural activity is crucial for cognitive research.
- Previous research has not specifically analyzed post-walk neural changes in forest environments.
Purpose of the Study:
- To investigate the effects of forest versus urban walking on brain activity using electroencephalography (EEG).
- To differentiate neural patterns associated with different environments using machine learning.
Main Methods:
- Collected EEG data from 30 participants before and after walks in forest and urban settings.
- Performed spectral analysis on Theta, Alpha, Beta, and Gamma frequency bands.
- Utilized Random Forest, Support Vector Machine, and Multi-Layer Perceptron classifiers to analyze EEG data.
Main Results:
- The Low Beta frequency band showed the highest classification accuracy (0.78) in distinguishing between forest and urban walking environments.
- Distinct EEG patterns were observed after walks in natural versus urban settings.
Conclusions:
- This study is the first to demonstrate the neural effects of post-walk forest exposure.
- Findings support the cognitive benefits of nature exposure and highlight the utility of EEG and machine learning in environmental neuroscience.
