[Eeglablist] Epoch length in EEGLab

[Makoto Miyakoshi]

Dear Joseph and Clemens,

Use -1 to 2 sec regardless of the fact that your SOA may be (much) shorter
than this. It does not matter if the epochs overlap. The reason for this
long epoch is because for the later wavelet transform you'll want to use
1.1-sec long sliding window to decompose 3-Hz activity. See my pipeline
wiki
http://sccn.ucsd.edu/wiki/Makoto's_preprocessing_pipeline#Epoch_data_to_-1_to_2_sec

Makoto

On Tue, Jun 28, 2016 at 1:20 AM, Clemens DICKHUT <clemens.dickhut at uni.lu>
wrote:

> Dear Joseph,
>
> generally your trial length and your research question/post-processing. If
> you want to calculate event-related potentials you usually need 1000-500ms
> pre-stimulus baseline activity to extract the mean amplitude (which is then
> subtracted from post-stimulus time points). Depending on the components you
> want to investigate, i.e. N200/P300, consequentially you will need at least
> 500ms post-stimulus, better 1000. A good start probably is -500 1000.
>
> For time-frequency transforms you will need longer epochs, because,
> depending on the wavelet starting frequency, a substantial amount of time
> points will be cut away both pre and post stimulus (this is due to the
> wavelets that take only the middle time point). I.e. epochs from -1000
> 3000ms time-frequency transformed with a wavelet starting from 1Hz will
> leave you with data values from ca. -500 2500. So, be aware of that and go
> for longer epochs of you want to observe time-frequency activity. Some
> additional thing: for time-frequency analyses you may (this is a later step
> in the analysis) indicate the baseline (if) to be subtracted from -800 to
> -200 or something like that, to avoid subtracting post-stimulus activity
> “smearing” into the baseline (due to the reason mentioned above, especially
> relevant for low frequencies).
>
> For both approaches, if you have non-random stimulus overlap or responses
> leaking into the pre-stimulus period, its sometimes better to go for
> shorter pre-stimulus times.
>
> Hope this helps.
>
> Best,
> Clemens
>
>
>
>
> On 25 Jun 2016, at 22:06, Joseph Nuamah <jknuamah at aggies.ncat.edu> wrote:
>
> Dear All,
>
> I am new to EEGLAB.
>
> My question is: what determines the choice of epoch length in EEGLAB.
>
> Thanks.
>
> Joseph.
> _______________________________________________
> 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
>
>
>
> _______________________________________________
> 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
>



-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20160701/3b2d51df/attachment.html>