EEG correlation at a distance: A re-analysis of two studies using a machine learning approach

Area of Science:

• Neuroscience
• Computer Science
• Signal Processing

Background:

• Re-analysis of electroencephalogram (EEG) data from two prior studies investigating brain activity correlations between physically separated individuals.
• Utilized machine learning algorithms to examine relationships in EEG data from 45 pairs of participants across two distinct stimulation paradigms.

Purpose of the Study:

• To investigate the potential for non-conventional communication by analyzing correlations in EEG activity between stimulated and isolated participants.
• To apply advanced machine learning techniques to identify subtle connections in brain activity across distances.

Main Methods:

• Employed a linear discriminant classifier on EEG data from two datasets.
• Dataset 1: 25 pairs with one participant receiving 1-second visual/auditory 500 Hz stimulation.
• Dataset 2: 20 pairs with one participant receiving 1-second modulated visual/auditory stimulation (10, 12, 14 Hz).

Main Results:

• Successfully classified 50.74% of non-stimulated participants' EEG activity in Dataset 1, correlating with remote stimulation.
• Achieved classification rates of 51.17% (10 Hz), 50.45% (12 Hz), and 51.91% (14 Hz) in Dataset 2 for non-stimulated partners.

Conclusions:

• Machine learning analysis advances understanding of EEG activity connections between stimulated and isolated partners.
• Findings suggest potential for developing practical applications in non-conventional "mental telecommunications" between separated individuals.