Dear Duncan,

If you do:

D = spm_eeg_load

and select your TF file in the file selection window you can then use
the syntax D(channels, frequencies, samples, trials) to access the

You can look at:


To find out what indices to use. For instance squeeze(D(1,  :,  :, 1))
will give you a TF matrix for the first channel in the first trial.
Just note that the spm_eeg_load part is quite critical for this to



On Mon, Jul 6, 2009 at 4:17 PM, Duncan Astle<[log in to unmask]> wrote:
> I have conducted a time-frequency analysis on my 36-channel EEG data.
> Everything looks fine. But I really want to get the raw power values that it
> is using to produce the plots. Basically, I just want to compare my four
> conditions over a set of electrodes at a particular bandwidth - without
> drawing fancy pictures. Is there any straightforward way of getting this
> information?
> My first inclination was to just load it into matlab and get it out that
> way, but then I realised that the .mat file is just a header, and the actual
> values are in a separate file that I can't seem to load into matlab.
> Does anyone have any suggestions as to how to do it?
> Thanks,
> Duncan