Neurophysiological Investigation of Spontaneous Correlated and Anticorrelated Fluctuations of the BOLD Signal

• Published in: The Journal of neuroscience Vol. 33; no. 15; pp. 6333 - 6342
• Main Authors: Keller, Corey J., Bickel, Stephan, Honey, Christopher J., Groppe, David M., Entz, Laszlo, Craddock, R. Cameron, Lado, Fred A., Kelly, Clare, Milham, Michael, Mehta, Ashesh D.
• Format: Journal Article
• Language: English
• Published: United States Society for Neuroscience 10.04.2013
• Subjects: Adult; Brain Mapping - methods; Brain Mapping - statistics & numerical data; Brain Waves - physiology; Cerebral Cortex - blood supply; Cerebral Cortex - physiology; Electrodes, Implanted; Electroencephalography - methods; Electroencephalography - statistics & numerical data; Female; Humans; Image Interpretation, Computer-Assisted - methods; Magnetic Resonance Imaging - methods; Magnetic Resonance Imaging - statistics & numerical data; Male; Neural Pathways - physiology; Neurons - physiology; Neurophysiology - methods; ROC Curve
• ISSN: 0270-6474; 1529-2401; 1529-2401
• DOI: 10.1523/JNEUROSCI.4837-12.2013

Analyses of intrinsic fMRI BOLD signal fluctuations reliably reveal correlated and anticorrelated functional networks in the brain. Because the BOLD signal is an indirect measure of neuronal activity and anticorrelations can be introduced by preprocessing steps, such as global signal regression, the neurophysiological significance of correlated and anticorrelated BOLD fluctuations is a source of debate. Here, we address this question by examining the correspondence between the spatial organization of correlated BOLD fluctuations and correlated fluctuations in electrophysiological high γ power signals recorded directly from the cortical surface of 5 patients. We demonstrate that both positive and negative BOLD correlations have neurophysiological correlates reflected in fluctuations of spontaneous neuronal activity. Although applying global signal regression to BOLD signals results in some BOLD anticorrelations that are not apparent in the ECoG data, it enhances the neuronal-hemodynamic correspondence overall. Together, these findings provide support for the neurophysiological fidelity of BOLD correlations and anticorrelations.