Inverse solution statistical group Analysis

[Michaël Mouthon]

Dear Cartool community, 

I have computed the individual inverse solution with the tool "Inverse Solutions->Computing Results of Inverse Solutions". However I am not sure which output file I should take to perform a group statistical analysis. Should I take :

- the output of "Every Subjects’ RIS" (file INV.*.ZScorePO.ris )

- the output of "Z-Score Factors" (file INV.*.ZScorePO.ZScoreFactors.ris)

In the second case, why Cartool generate only one file per subject instead of a file for each subject and condition ? 

[Denis]

Hi Michaël

Definitely the first option, "Every Subjects’ RIS" (file INV.*.ZScorePO.ris). These are indeed the EEG data transformed into the brain via the inverse matrix.

The second list of files, <something>.ZScoreFactors.ris (as well as <something>.ZScoreFactors.sef), contain only the re-calibration factors used at each solution points, for each subject. They can be saved for checking how well the standardization went for each subject, and/or to be reloaded for any later new RIS computation, because we definitely want the same calibration each time.

Another way to check is to simply visualize one or two of these files: the first group has the same number of time points as the input EEG, while the second group is only 2 "time points", namely the centering and spreading (kind of the Mean and Standard Deviation) values, for each solution points.

Best,
Denis

[Michaël Mouthon]

Thank you very much, 

I was not sure because the group results using the  INV.*.ZScorePO.ZScoreFactors.ris look more similar as the result obtain by the old way to perform IS (using EEG File calculator to multiply the .eph with the .is). I am sure now that we need to work on the INV.*.ZScorePO.ris for now one. I hope that the inverse solution computed with the old style (File calculator) is not completely wrong?