EEG Coherence and Visual Evoked Potentials: Investigation of Neural Energy Transfer Between Human Subjects

Todd L. Richards, PhD, Professor, Department of Radiology, University of Washington, Seattle, WA
Leanna J. Standish, ND, PhD, Director, Research Institute, Bastyr University, Kenmore, WA 
L. Clark Johnson, PhD, Research Asst. Prof., School of Nursing, University of Washington, Seattle,WA

Primary Subject Classification: 05.00 Experiential and Anomalous Approaches
Secondary Subject Classification: 02.01 Neural Correlates of Consciousness


Clinicians and scientists of mind-body complementary and alternative medicine (CAM) have suggested that intention and conscious states can influence biological systems both within the same organism and, more controversially, between two organisms at a distance. The publication of a neuroscientific paper by Grinberg-Zylberbaum, Delaflor, Attie and Goswami (Physics Essays, volume 7, pages 422-428, 1994) has been claimed as landmark event in mind-body medicine and the emerging new field of consciousness science because it is held as the first demonstration of remote nonlocal influence in human neurophysiology.

The results of this study appear to indicate that neural events (evoked potentials) in one human brain produced by visual stimulation to a single human subject can induce similar visual evoked potentials in the brain of another person located several meters away who is not visually stimulated.

In preparation to replicate the Grinberg-Zylberbaum experiment, we conducted a pilot study to better understand the nature of the evoked potential signal and the baseline characteristics of electroencephalographic (EEG) coherence. The purpose of this study was to determine the random level of coherence or correlation between two EEG scans acquired at different times from two people that did not know each other other and that did not know their scans were going to be compared. One electrode was placed at the CZ position (10-20 electrode system) and two reference electrodes were placed on the right and left ear lobes. EEG was acquired at a sampling rate 512 points/second using the J&J Engineering physiologic monitoring device (model DSP 12). Software was written to sequentially calculate a cross-correlation coefficient on two EEG data segments from the two people where the signal was averaged over 100 sets of 1 second epochs (as is done for evoked potential averaging) before calculating the cross-correlation. Three sets of EEG coherence measurements (12 cross-correlations in each set) were made using data from three subjects (taken two at a time). In 36 cross-correlations performed, we observed correlations coefficients that ranged from -0.37 to +0.46. The mean was close to zero (0.03) but the standard deviation was substantial (0.21).

From this experiment, we conclude that random EEGs recorded from two unrelated people cannot be assumed to have zero correlation. While our simulation experiment does not invalidate the results or conclusions of Grinberg-Zylberbaum et als paper, our data show that it is incorrect to base statistical tests under the null hypothesis that there is zero correlation between EEGs from two humans. The proper statistical model must show that the cross-correlation between EEG activity from two human subjects is statistically different from random correlation that we now know can range from -0.37 to +0.46.

Corresponding Author:

Todd L. Richards, PhD Phone: 206-598-6725
University of Washington, Dept. of Radiology FAX: 206-543-3495
Box 357115, Seattle, WA 98195-7115 Email: