[Eeglablist] Truncating time valuesl when calculating low frequency ERSP/ITC

[Arnaud Delorme]

Dear John,

this is a common problem people encounter. To perform time-frequency analysis, the epoch time range is divided into overlapping windows, and spectral decompositions (FFT or wavelets) is to be applied to these windows. When your final time range is -136 to 136, it means that -136 ms is the center of the first time windows (ranging from -250 ms to -22 ms with -136 being the center of the window). 

Best,

Arno

On May 20, 2011, at 6:59 AM, J Saunders wrote:

> I searched through the list for an answer and saw something related
> (http://sccn.ucsd.edu/pipermail/eeglablist/2010/003231.html).
> 
> However, when I implement the suggestion, increasing the pre-stimulus
> time, I still get a time values removed warning.
> 
> Specific case:
> 
> I imported epochs from Spike from -0.25 to 0.25 s into matlab
> 
> I then used the following parameters for :
> 
> 'cycles', [2 10], 'freqs',[10
> 80],'nfreqs',70,'freqscale','linear','timesout',[-200:4:250]
> 
> and EEGlab gives me the warning:
> 
> Warning: 82 out of 151 time values were removed (now -136.00 to 136.00
> ms) so the lowest frequency could be computed with the requested
> accuracy
> 
> Why is this happening?  I'm confused about how EEGlab is making this
> decision, and what is driving it to remove time points.
> 
> Thanks so much for your help.
> 
> John
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu