cluster permutation t tests question

✉️ Mailing List Archive (read-only)
1

External Email - Use Caution

Hello,

I have a question regarding cluster permutation t tests.

I want to make a 1 sample t-test between 2 conditions, comparing Power Spectral Densities averaged over epochs and frequencies, defined at each source. Hence the clustering would only be spatial, not spatio-temporal.

Do I need to use spatio_temporal_cluster_1samp_test and give it as argument an array of shape (n_subjects, n_vertices) and made of contrasted PSD values?

Thanks in advance,

Fleur

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20200408/1641ada6/attachment-0001.html

2

External Email - Use Caution

The easiest way to do it is just to make the time dimension a singleton
with something like X[:, np.newaxis, :].

Eric

3

External Email - Use Caution

Thank you! Would it render the same result as using permutation_cluster_1samp_test?

4

External Email - Use Caution

I think it should, yeah.

Eric