Test on sources with different epochs

Discussions
1

I'm trying to replicate the pipeline "example 4: parametric test on sources" but with a different data structure which looks like this:

Screenshot 2021-06-16 at 14.02.12
Screenshot 2021-06-16 at 14.02.12818×460 59.8 KB

Each file is epoched into 6s windows. Is it possible to add the same files in process 1 and process 2 and select different time epochs for comparison? E.g. in process 1 I want to select epoch 1+ 2 [0.00 - 5.998],[6.00 - 11.998] and in process 2 I want to select epoch 3 [12.00 - 17.998]

2

Is it possible to add the same files in process 1 and process 2 and select different time epochs for comparison? E.g. in process 1 I want to select epoch 1+ 2 [0.00 - 5.998],[6.00 - 11.998] and in process 2 I want to select epoch 3 [12.00 - 17.998]

No, this is not possible.
When running statistical tests, the time definition must match across all the files of both sample sets, and the input option "Time window" applies to all the files.

If you want to test different time windows, you need to re-epoch your data. Right-click on a file > Import in database (or use the process equivalent to batch it). Two possibilities:

  1. Create new events in each file first, and then epoch your data using the option "Use events". The time definition of the new short epochs would be based on the events, therefore it would not matter if they are coming from different timing of the files.
  2. Import time blocks without epochs and then run the process "Standardize > Uniform epoch time" to apply the time definition of the first file to all the the others in the list of files selected in Process1

When you have epochs that all have the same time definition and the same dimensions, you'll be able to run your statistical tests.

3

Thank you. That worked.

But after the process I don't have the option to display the data on the cortex anymore. Why is that?

Screenshot 2021-06-21 at 10.28.11
Screenshot 2021-06-21 at 10.28.111068×354 94.8 KB
Screenshot 2021-06-21 at 10.28.37
Screenshot 2021-06-21 at 10.28.371008×142 47.8 KB

4

But after the process I don't have the option to display the data on the cortex anymore. Why is that?

Because you got rid of the spatial dimension, you don't have cortical maps anymore, but isolated values.

For visualization purposes, you can project these values on the cortex, to get 3D maps that you can review like your initial files, with the process " Simulate > Full source maps from scouts".
https://neuroimage.usc.edu/brainstorm/Tutorials/Simulations#Generate_full_source_maps_from_scouts

Do not use this for further processing, prefer working at the ROI level.