Is Functional Magnetic Resonance Imaging-Inspired Electroencephalogram Feedback the Next New Treatment in Psychiatry?

Ever since Hans Berger recorded the first electroencephalogram (EEG) in 1924, scientists have struggled to understand which exact regions of the brain are causing the signal on the scalp. This inverse problem has proved quite difficult to crack, despite decades of research. Some researchers have used electrodes within the brain (electrocorticograms) simultaneously with surface electrodes to try to understand the source. Others have recorded functional magnetic resonance imaging (fMRI) blood oxygen level–dependent (BOLD) signals with concurrent surface EEG. There have been other partial solutions at finding the source of the surface signal, such as low-resolution electromagnetic tomography ( ). However, the problem is a tricky one. In the article in this issue, Keynan et al. ( ) used a hybrid approach in which they estimate the source of an EEG signal for a large database where they recorded EEGs within an MRI scanner for a group of individuals and then used this to build a database to use on others, outside of the MRI scanner. They coin a new term for this solution to the inverse problem, fMRI-inspired EEG ( ). EEG purists will no doubt have numerous criticisms of this group-average approach. The authors admit that the surface amygdala signal likely comes from multiple sources, not just the amygdala. However, using this probabilistic anatomically inspired approach is certainly easier and less expensive than having each person undergo an EEG-fMRI session or using BOLD fMRI feedback within the scanner ( ).