[FieldTrip] Fwd: Question about unit of the wavelet analysis

[Roemer van der Meij]

Hi Theresa,

The answer to question #1 is a bit difficult because of scaling, I thought
for a long time about it but I couldn't define a definite convincing
answer, so I'll leave that to someone else. Without all the scaling that is
going on, power for EEG is simply V^2. However, the fft/ifft in
ft_specest_wavelet and the scaling by the square root of 2 over the
sampling rate in line 176 of the same function complicate matters for me. I
hope someone else can shine some light on it.

For your second question, even though the total wavelet energy is the same
at every frequency, the energy of the frequencies present in your data are
not, and their energy decays with a 1/f pattern. This 1/f decrease found in
EEG data is a pattern that is found in many natural
phenomena<http://en.wikipedia.org/wiki/Pink_noise>.
It is often referred to as 1/f noise, but I personally don't like that
term, as it indicates something that is not interesting in any way.
While doing some hand-waving, my intuition for neuronal recordings has
always been that the time constants of lower frequencies are much further
away from the 'noise' of EPSPs, dendritic and somatic currents, etc of
neurons (and therefore have a higher amplitude) and that lower frequencies
are thought to be generated by larger groups of neurons, which of course
increases the total energy at those frequencies.
This is just a vague notion of me however, surely there are more clever
people on the mailinglist to give their insights, and give a more
physically founded explanation of what the 1/f pattern means for neuronal
recordings. But since nobody gave an answer, I'd thought I'd give it a shot
anyway.

Hope it helps,
Cheers,
Roemer



On Sat, May 5, 2012 at 9:15 PM, jan-mathijs schoffelen <
jan.schoffelen at donders.ru.nl> wrote:

> Dear Theresa,
>
> I am forwarding your questions to the discussion list, as this is the
> forum on which we discuss this type of issues. Other people may benefit
> from it too, or may feel inclined to answer.
>
> Cheers,
>
> Jan-Mathijs
>
> Begin forwarded message:
>
> *From: *Theresa Götz <theresa at biomag.uni-jena.de>
> *Date: *April 25, 2012 9:52:33 AM GMT+02:00
> *To: *j.schoffelen at donders.ru.nl
> *Subject: **Question about unit of the wavelet analysis*
>
> Dear Jan-Mathijs Schoeffelen,
>
> we have a question about the unit of the time-frequency power spectrum:
> maybe you can help me with that.
>
> 1. In fieldtrip, we performed ft_frequanalysis with method 'wavelet' and
> output 'power'. What is the exact unit of the resulting power spectrum
> (e.g. V/sqrt(hz) or V^2/Hz or V^2 or V)?
>
> 2. Why does the amplitude of power spectrum decrease toward higher
> frequencies even though we keep the amplitude of the input sine wave (in
> time domain) constant?
>
> Thank you very much.
>
> Best,
> Theresa
>
>
>
>
> --
> Bitte denken Sie an die Umwelt, bevor Sie diese e-Mail ausdrucken.
>
>
> Dr. Theresa Götz  Phone: +49-3641-9325780
> University Hospital Jena         Fax:   +49-3641-9325772
> Dept. for Neurology
> Biomagnetic Center
>
> Erlanger Allee 101
> D-07747 Jena
> Germany
>
>
>
>
> Universitätsklinikum Jena
> Körperschaft des öffentlichen Rechts und
> Teilkörperschaft der Friedrich-Schiller-Universität Jena
> Bachstraße 18, 07743 Jena
>
> Verwaltungsratsvorsitzender:  Prof. Dr. Thomas Deufel
> Medizinischer Vorstand:       Prof. Dr. Klaus Höffken
> Wissenschaftlicher Vorstand:  Prof. Dr. Klaus Benndorf
> Kaufmännischer Vorstand und
> Sprecher des Klinikumsvorstandes:  Dr. Brunhilde Seidel-Kwem
>
> Bankverbindung: Sparkasse Jena
> BLZ: 830 530 30, Kto.: 221
>
> Gerichtsstand Jena
> Steuernummer: 161/144/02978  USt.-IdNr. : DE 150545777
>
>
>    Jan-Mathijs Schoffelen, MD PhD
>
> Donders Institute for Brain, Cognition and Behaviour,
> Centre for Cognitive Neuroimaging,
> Radboud University Nijmegen, The Netherlands
>
> Max Planck Institute for Psycholinguistics,
> Nijmegen, The Netherlands
>
> J.Schoffelen at donders.ru.nl
> Telephone: +31-24-3614793
>
>
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>



-- 
Roemer van der Meij M.Sc.
PhD student
Donders Institute for Brain, Cognition and Behaviour
Centre for Cognition
P.O. Box 9104
6500 HE Nijmegen
The Netherlands
Tel: +31(0)24 3655932
E-mail: r.vandermeij at donders.ru.nl
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20120511/773cb127/attachment-0002.html>