Related Experiment Videos
A second linked-reference issue: possible biasing of power and coherence spectra
1Maharishi International University, Department of Psychology, Fairfield, IA 52557.
The International Journal of Neuroscience
|March 1, 1994
Summary
A linked-ears reference in electroencephalography (EEG) distorts frontal power and coherence spectra. This common reference artificially inflates frontal relative power and coherence, impacting analysis of brain activity.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Signal Processing
Background:
- Scalp-recorded electroencephalography (EEG) is crucial for studying brain activity.
- The choice of reference electrode significantly impacts EEG signal interpretation.
- Previous studies suggest common references may introduce artifacts.
Purpose of the Study:
- To investigate the influence of a linked-ears reference on EEG potentials.
- To compare EEG data acquired with linked-ears, Laplacian, and close-bipolar references.
- To assess the impact on power spectra and coherence.
Main Methods:
- EEG was recorded from six subjects during an eyes-closed counting task.
- Data were collected using a linked-ears reference and an active close-bipolar reference.
- Linked-ears data were transformed to current source densities using the Laplacian.
- Power spectra and coherence were computed for all datasets.
Main Results:
- Linked-ears frontal relative power was significantly higher at 7-Hz compared to Laplacian or close-bipolar.
- Linked-ears coherence spectra were significantly higher and showed a smaller range than other methods.
- These findings indicate distortion of spectral analysis by the linked-ears reference.
Conclusions:
- A linked-ears reference introduces significant distortions in EEG frontal power and coherence.
- These distortions can lead to misinterpretations of brain activity, particularly at specific frequencies.
- Researchers should be cautious when using common references for spectral analysis in EEG studies.