Pipelines for 2x2 repeated-measures design for ERP/TFR data

Support & Discussions
1

Hello everyone!

I’m currently trying to find examples of published research that used MNE-Python to perform TFR or ERP analyses in a 2x2 repeated-measures design, ideally with statistical comparisons.

Currently I am working with this tutorial: Mass-univariate twoway repeated measures ANOVA on single trial power — MNE 1.9.0 documentation

So far, I haven’t had any luck locating articles where the pipeline from the tutorial was used (or any other MNE pipeline for such experimental design). I would be grateful for any references or examples you could share — either your own work or papers you know of.

Has anyone here published using this type of analysis with MNE-Python? Or do you know of any peer-reviewed articles that demonstrate this approach?

Any tips or resources would be greatly appreciated!

2

I don’t know of any but maybe some article by Denis Engemann since IIRC he helped write that tutorial?

Also have you looked at the FieldTrip docs? They often contain complementary information, for example:

3

Hi, the code itself implements a standard repeated measures ANOVA - as documented translated from Rik Henson’s MATLAB code and tested vs R and SPSS results. The examples / tutorials explore using the code but do not implement a specific analysis from a peer reviewed article. Even if things look plausible here, I would double check using simulations if the permutation procedure always gives correct results for interaction terms, which I vaguely remember to be more tricky based on conversations over the years. For main effects I am quite confident that the procedure is fine. But perhaps someone here has looked deeper into interaction effects with our rANOVA code over the past 10 years? You could also test vs uncorrected mass-univariate analytical p values returned for the interaction effect. Hope this helps.