Threshold for topographic map?

✉️ Mailing List Archive (read-only)
1

External Email - Use Caution

Hello MNE experts,

I created a topographic map based on the difference between the MEG epochs
of two conditions averaged across a group of subjects. Is there any way to
apply threshold on the map to reveal only significant difference between
two conditions at the group level? Also, are there any ways to apply
threshold when plotting group-level topographic map for decoding results (
https://martinos.org/mne/stable/auto_tutorials/plot_sensors_decoding.html
-- this is for single subject)

Suggestions will be appreciated, thanks!

Best,
Veda
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/mne_analysis/attachments/20180626/051299ee/attachment.html

2

External Email - Use Caution

hi,

do you have an example of figure you want to produce?

Alex

3

External Email - Use Caution

Hi Alex,

Please see figure 2 in the following link
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2783912/

Thanks.

Veda

4

External Email - Use Caution

the easy for this is to create your own custom colormap with white
in the middle.

see for example: https://matplotlib.org/tutorials/colors/colorbar_only.html
but I am sure you can find better examples online. Then use the cmap
parameter.

HTH
Alex

5

External Email - Use Caution

Thanks for the information. However, it needs a user-specified threshold
which I don't have.
Just came across the following page. It is more close to what I want. Can
it be applied on the epochs or decoding results from a group of subjects?
https://www.martinos.org/mne/stable/auto_examples/stats/plot_sensor_permutation_test.html

6

External Email - Use Caution

if you can make an Evoked object from your results yes. It's easy.
You need one value per sensor to display.

As you see this example just plots a subset of significant sensors.

HTH
A