[Eeglablist] Comparison of EEG components to depth recordings
mmiyakoshi at ucsd.edu
Tue Nov 5 15:22:22 PST 2019
I agree with Ramesh.
Your scalp sensor data must be (contribution from your
depth-electrode-scale ROI) + (contribution from other cortical mesosource
contribution) + (contribution from anything non-cortical origin).
One way to quantify the direct contribution by your depth-electrode-scale
ROI to the ICA-decomposed scalp sensor projection is to calculate the
following measure called percent variance accounted for (PVAF):
PVAF (%) = 100-100* mean(var(selected_IC_backprojection -
Here, variance is calculated across scalp channels to which backprojection
should be calculated. Then temporal average is calculated. This measure
does not behave well, so be careful.
The suggested calculation above is based on my understanding of your
experimental design and aim, and I could be wrong here.
On Tue, Nov 5, 2019 at 12:15 AM Jenson, David Evans <david.jenson at wsu.edu>
> A reviewer has asked me to acknowledge that my ICA components probably
> contain noise, potentially including additional signal not belonging to the
> signal of interest. I’m hesitant to acknowledge this point in too much
> detail, as I feel like it undercuts the rest of the analyses. However, I’m
> unaware of any published studies that have directly compared ICA decomposed
> EEG data to depth recordings, which is the only method I can think of for
> demonstrating that the components are not contaminated by noise or
> additional signals.
> Is anyone aware of published studies that could help me make this point,
> or has experience responding to similar questions?
> David Jenson, PhD
> Assistant Professor
> Department of Speech and Hearing Sciences
> o: 509-368-6913 | david.jenson at wsu.edu<mailto:david.jenson at wsu.edu>
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
More information about the eeglablist