[Eeglablist] Epoch data after time-frequency transform?

[Max Cantor]

Hey Eric,

How low are you trying to look? You can change the zero-pad ratio, or
potentially 'reflect' your epochs (although that can create artifacts along
that reflection point as well), which are two ways to give you somewhat
better resolution in addition to increasing the epoch length. Also, are you
using a single window length / number of cycles? You could potentially use
a variable window length / variable number of cycles to minimize the
time-frequency tradeoff across frequencies. Is there some reason why you're
concerned with edge artifacts or epoching? I'm not sure what you mean by
adding 'events' to the matrix. If you write the TFR code from the ground-up
(for instance by following Mike Cohen's Analyzing Neural Time Series book)
you could create custom 'events' which in some way change how the TFR is
produced without epoching per se, but again I'm not sure I totally
understand what you're trying to do to give you more specific advice.

Best,
Max

On Thu, Nov 2, 2017 at 7:13 PM, Eric Rawls <elrawls at email.uark.edu> wrote:

> Thanks Tarik - so far the approach I've been taking is to cut arbitrarily
> long epochs out of the continuous EEG, but this is sometimes irritating
> because it increases the size of the .set file substantially and duplicates
> a bunch of events, which sometimes makes counting trials per condition
> annoying. It gets the job done, but I'm thinking that if I were to simply
> transform the continuous data before epoching none of this would be an
> issue.
>
> Any experience with this approach? I don't really use the gui anyway, so
> that's not an issue. What EEGLAB function would you use to accomplish this?
> I guess a related question is, do you know if it's possible to insert
> events into a time-frequency matrix rather than the usual EEG.data time X
> activation matrix? That would make it easier, if I could actually insert
> the time-locking event of interest into the TF representation directly.
> Thanks,
> Eric
>
>
>
> On Mon, Oct 30, 2017 at 8:44 PM, Tarik S Bel-Bahar <
> tarikbelbahar at gmail.com> wrote:
>
>> Hello Eric,
>>
>> Totally sane idea!
>>
>> One option within eeglab is to cut into longer epochs, do TF,  and then
>> ignore the edges/artifacts time-periods of the epochs.
>>
>> Another option is to review various TF techniques and use ones that
>> create less or little edge artifacts.
>>
>> If you haven't had a chance to yet,  check out Mike X Cohen's excellent
>> books, matlab scripts, and videos about this.
>>
>> Generally speaking you should be able to do this (e.g., epoch continuous
>> time data), but you may have to do so outside of eeglab and/or using some
>> eeglab functions in a code/script.
>>
>> See also chronux toolbox too if you haven't had a chance to yet.
>>
>>
>>
>>
>>
>> On Mon, Oct 30, 2017 at 8:11 AM, Eric Rawls <elrawls at email.uark.edu>
>> wrote:
>>
>>> Hello,
>>>
>>> I am looking for ways to improve the low-frequency resolution of
>>> time-frequency transforms, and I had the idea that it might be possible to
>>> transform the continuous data to a time-frequency representation (using
>>> wavelets or the Hilbert transform) before epoching the transformed data.
>>> This would avoid edge artifacts at all but the first and last trials.
>>> However, I'm not sure if this is possible in EEGLAB - does anyone have
>>> experience with this, or a workaround to cut epochs from continuous
>>> time-frequency domain data, time-locked to an event of interest?
>>>
>>> Thanks
>>>
>>> Eric Rawls, M.S.
>>> Graduate Research Assistant
>>> Department of Psychological Sciences
>>> University of Arkansas
>>>
>>> _______________________________________________
>>> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
>>> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc
>>> sd.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
>



-- 
Max Cantor
Graduate Student
Cognitive Neuroscience of Language Lab
University of Colorado Boulder
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20171106/ac20dbbf/attachment.html>