Question about make_forward_solution

đź’¬ Support & Discussions
,
1
  • MNE version:1.5.0
  • OS: Windows 10
  • IDE: Spyder (MNE)

Hello,

I am looking for help with the make_forward_solution() function. In my project, I am not wanting map anything onto the head. I can guess (and somewhat to my understanding from standard EEG processwith without MNE), I can go from rejecting using annotations to computing the forward solution (in the “typical MEG/EEG workflow” provided by MNE documentation). One of my qestions is, do I need to do anything with the head of regions in order to utilize the make_forward_solution function? It is giving me the following error: TypeError: make_forward_solution() missing 2 required positional arguments: 'src' and 'bem'

The code I am using for the forward solutions is: p01_fwd = mne.make_forward_solution(p01_raw.info, p01_raw)

A few more questions to come as this one is answered. All questions will still pertain to this topic, so I would like to keep it on this discussion post, if possible.

2

Please don’t ask duplicate questions (Questions about make_forward_solution)

If you don’t want to “map anything onto the head” I guess that means you just want to analyze sensor data, not perform inverse imaging to estimate where in the brain the signals are coming from. Is that right?

If that’s right, then I don’t think you even need a forward solution. Why do you think you need one?

3

I am sorry for doing the duplicate questions. And yes, that is what I want to do. I strictly want to analyze sensor data and not map anything to a head.

I was not sure which part of the “typical MEG/EEG workflow” I could skip to if I wanted to go ahead and start analyzing the sensor data. I figured computing the forward solution was the next step in the process, but I guess I was wrong.

If I don’t need a forward solution, then is the next step to start calculating the noise-covariance matrix to analyze the sensor data? Then continuing on until the end of the workflow?

4

I don’t know what the right “next step” is because I don’t know what your ultimate analysis goal is.

If you’re following this page then once you’ve hit the “source localization” section then you’ve gone as far as you need to for sensor-only EEG analysis. I recommend you take a look at our tutorials too… if you follow this tutorial then everything except for the last section “inverse imaging” is relevant for sensor-level data. All the other introductory tutorials are relevant for sensor data. Same with the continuous data tutorials, all the preprocessing tutorials except for the fNIRS and EyeTracklng ones, all the epoching tutorials, all the evoked responses tutorials, and all the sensor-space statistics tutorials.

The main difference you need to look out for is that whenever the example uses “mag” or “grad” or “meg” channels you’ll want to use “eeg”. There will probably be a few things that don’t work; if you hit a snag then please don’t hesitate to come back and ask a more specific question.

1 Like