ERPs statistic analysis

💬 Support & Discussions
,
1

:question: If you have a question or issue with MNE-Python, please include the following info:

  • MNE version: latest one
  • operating system: Windows 11

I have two processed epochs that comes from two different experimental design: rest-state and stimulated-state.

Now, I want to compare these two epochs and analyze when the data has a significant correlation.

Best regards
kerwin

2

I’m not exactly sure what you are trying to do, but you can compute correlations with numpy.corrcoef — NumPy v1.26 Manual, and cross-correlation with numpy.correlate — NumPy v1.26 Manual. These functions take two 1D NumPy arrays, which you can generate from your epochs with epochs.get_data(). You have to pick one channel, and you might have to flatten the resulting 2D array.

3

ok thank you！I am trying to do cluster permutation test. Do you have any ideas

4

There’s a very good notebook about data science for neuroscience which has some good tutorials on analysing ERP data.

https://neuraldatascience.io/7-eeg/erp_group_stats.html#permutation-t-test-for-multiple-comparison-correction

I’ve linked you to a chapter that has a section on permutation t-testing. Hopefully you find this useful.

1 Like
5

wow！ Thank you vert much！