Dear Amir,
On Mon, Nov 16, 2009 at 12:12 PM, Amir H Javadi <[log in to unmask]> wrote:
> I want to analyse the frequency
> components of the signal in theta (4-8 Hz) and gamma (28-64 Hz) bands.
>
> Which method should I use to frequency analyse the data?
> Once i used "spm_eeg_tf" method to analyse the data. But it asks for
> individual frequencies, rather than a frequency band.
You can specify separate frequencies there but usually you would
specify something like 28:64.
> On the other hand, I
> used "convert to image" option and it asked me for a band. I entered [35 37]
> and it took a long time, about 10', to generate some images. Later I could
> display the images, but I don't know how to interpret them. Is [35 37] the
> frequency band? how can I put an axis on the image?
I don't know why it ran 10'. If you have single trials it might indeed
take some time to write them all out. 35 to 37 is indeed the frequency
band. But I don't know what you entered before for the TF analysis and
how many frequency bins (if any) you really have within that band.
> The previous time that I used "spm_eeg_tf", I used it for epoched data, not
> for averaged data. I assume that I should analyse the averaged data, right?
Depends on what you want to do. You can also do statistics on single
trials within subjects or you can do statistics on averaged TF images
across subjects. You might also want to baseline-correct or rescale
your TF data (Other/Rescale TF) before exporting it to images.
> When I want to perform the analysis on an averaged data, it gives me the
> following error message. Once I entered [35 37 40 45 60] for frequencies and
> once I entered just 35, but still I get the following error message.
> ??? Operands to the || and && operators must be convertible to logical
> scalar values.
> Error in ==> meeg.frequencies at 33
> if length(ind)~=length(f) || max(ind)>size(this, 2)
> Error in ==> meeg.subsref at 28
> varargout = {feval(subs(1).subs, this, subs(2).subs{:})};
> Error in ==> spm_eeg_tf at 120
> Dtf = Dtf.frequencies(:, tf.frequencies);
>
I suspect it has something to do with non-uniform frequency axis, but
if you keep getting it, send me the file you used as the input for TF
and the exact values you used for other inputs.
Best,
Vladimir
|