Extracting ERPs and Amplitudes

  • MNE version: 1.3.0
  • operating system: macOS-12.2.1


I am using the mne-bids-pipeline to preprocess and analyze my EEG data. I have modified the default config file with my desired parameters, and it seems to be successful for the preprocessing part; however, I am a beginner when it comes to Python and am unsure how I can add to the pipeline to get the results I need thereafter.

Can someone please advise me how I can incorporate the code from this example (EEG analysis - Event-Related Potentials (ERPs) — MNE 1.4.0 documentation) so that I can extract the ERPs from my data? I would like to obtain the average evoked potential across all electrodes for each of my conditions, but currently my results HTML file only displays the time course, topographies, and global field power for each of my conditions and the contrast.

Moreover, I would also like to obtain the peak amplitude values for each condition, so that I can use them in a correlation analysis. It appears, also based on the above link, that I can use the get_peak() method for this, but again I am wondering how to do so. Should I be trying to incorporate these analysis steps/code into the pipeline that I’ve also used for preprocessing, or must I write a separate code file to get the ERPs and amplitudes?

Thank you so much for any and all insight! :slight_smile:

Hi again! Just wanted to bump this post since I’m still wondering how to incorporate this into my pipeline. Thanks!! :slight_smile:

Hello @lz4199 and welcome to the forum!

MNE-BIDS-Pipeline is designed such that you can “eject” the data at any stage and continue processing it outside of the pipeline. If you want to perform some actions on epochs that are currently not offered by MNE-BIDS-Pipeline, simply load the cleaned epochs with MNE-Python and plug the data into your processing script :slight_smile:

Regarding the tutorial you mentioned – almost all of the steps discussed there are already being performed by MNE-BIDS-Pipeline. The only missing bits I can see are peak latency & amplitude and mean amplitude:


Is that what you’re looking for?

Best wishes,

Thanks so much, Richard!