[FieldTrip] time frequency analysis - basic question

["Jörn M. Horschig"]

Dear Beth,

 From all my knowledge I gathered so far, I can try to answer your 
question (and hope that people from this list correct me if I am wrong):
As often, the truth lies somewhere in between. The frequency smoothing 
is determined by the type and length of the taper you apply - this is 
similar to wavelet analysis. If you use multitapers, you can specify 
cfg.tapsmofrq, which determines the frequency band you want to smooth 
with. The FT freqanalysis method uses multiplication in the time domain 
instead of convolution in the frequency domain (which is equivalent). 
Any taper will always have a specific frequency spectrum itself, and 
thus it will not give you the value of a single frequency (though the 
output looks as if). So if you specify foi as [5 10 15], then the power 
at 5Hz will not be from 2.5 to 7.5Hz, but neither will it be at exactly 
and only 5Hz (btw, the exact center frequency might not be 5Hz, this 
depends on the (Rayleigh) frequency resolution of your signal).

I do not know any more specifics concerning this, maybe the help of 
ft_freqanalysis can tell you some additonal things: 
http://fieldtrip.fcdonders.nl/reference/ft_freqanalysis

Best,
Jörn

Btw, if this is not correct, else from the mailinglist should correct me 
asap ;)



On 8/4/2011 3:23 PM, Belluscio, Beth (NIH/NINDS) [E] wrote:
>
> Hi all-
>
>   When calling ft_freqanalysis you specify the foi limits and 
> intervals -- eg 5:5:30.  The resulting data set appears to generate 
> the power at discreet frequencies (5, 10, 15, 20, 25, 30).  Does this 
> actually reflect an estimate of the power over a range of frequencies 
> (2.5-7.5, 7.5-12.5, 12.5-17.5,....) or the power at discreet 
> frequencies with those in between not evaluated?
>
>   Thanks for your help,
>
> Beth.
>
> Beth Belluscio MD-PhD
>
> Clinical Fellow
>
> Human Motor Control Section
>
> NINDS, NIH
>
> 301-402-3495
>
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip


-- 
Jörn M. Horschig
PhD Student
Donders Institute for Brain, Cognition and Behaviour
Centre for Cognitive Neuroimaging
Radboud University Nijmegen
Neuronal Oscillations Group

P.O. Box 9101
NL-6500 HB Nijmegen
The Netherlands

Contact:
E-Mail: jm.horschig at donders.ru.nl
Tel:    +31-(0)24-36-68493
Web: http://www.ru.nl/donders

Visiting address:
Trigon, room 2.30
Kapittelweg 29
NL-6525 EN Nijmegen
The Netherlands

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20110805/20197c6b/attachment.html>