[Eeglablist] permutation testing in sLORETA - t-value or z-value

Gary Wilkinson gary.t.wilkinson at gmail.com
Wed Jul 19 08:04:14 PDT 2017

Hi guys,

Not strictly an eeglab question, but hoping someone out there might be able
to help.

I'm using the old sLORETA (http://www.uzh.ch/keyinst/loretaOldy.htm) to
estimate sources of data I've analysed with eeglab, but the output file is
confusing. The sources of differences are consistent with my hypothesis,
and the statistical non-parametric mapping (permutation testing with
t-tests on a voxel by voxel basis)  suggests that they are significant.
However, the output file (a single column of numbers, each row
corresponding to a voxel) is described as being both exact z-values, and
corresponding to t-values.

I'm not seeing how the result of a t-test can be both a t-value and a
z-value, or how a z-value would relate to the critical t-threshold. Is this
something related to the permutation procedure?


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20170719/917af0b3/attachment.html>

More information about the eeglablist mailing list