[Eeglablist] ICA component

[Tarik S Bel-Bahar]

Hello Dorian, if you're new to eeg analysis, please go thorugh the whole
eeglab tutorial using the provided sample data. Also you would benefit from
other basic resources (such as Luck's book Introduction to Event-related
potentials and Onton & Makeig's chapter in Luck's Handbook of Event-Related
Components). Please also be sure to search for and review past postings on
the eeglablist, and make use of the many tutorials and introductions for
beginner students that can be found online to EEG and ERP techniques.

"Trials" in the paragraph refers to "segments" or "epochs", the "pieces"
that the continuous data is often broken into, usually locked to the onset
of a stimulus in experiments with many "trials" of "stimulus-presentation".

The approximate number of necessary samples of EEG is a rule of thumb, and
biased towards longer data. ICA needs enough information about what's going
on to separate valid spatial patterns. ICA has a big stomach.

>From what you are saying, it does not seem you have long enough data to do
a proper ICA decompisition. However, there are many published articles
where people do ICA on very short periods and consider them valid. You can
find these by searching on Google Scholar.

On Tue, Oct 27, 2015 at 7:02 AM, Dorian Grelli <dorian.grelli at gmail.com>
wrote:

> Hi guys,
> I am very new with eeg data analysis and it would be great to have some
> support from you!
>
> I found the quotation below from this tutorial:
> http://sccn.ucsd.edu/wiki/Chapter_09:_Decomposing_Data_Using_ICA
>
> *"Very important note: We usually run ICA using many more trials that the
> sample decomposition presented here. As a general rule, finding Nstable
> components (from N-channel data) typically requires more than kN^2 data
> sample points (at each channel), where N^2 is the number of weights in the
> unmixing matrix that ICA is trying to learn and k is a multiplier. In our
> experience, the value of k increases as the number of channels increases.
> In our example using 32 channels, we have 30800 data points, giving
> 30800/32^2 = 30 pts/weight points. However, to find 256 components, it
> appears that even 30 points per weight is not enough data. In general, it
> is important to give ICA as much data as possible for successful training.
> Can you use too much data? This would only occur when data from radically
> different EEG states, from different electrode placements, or containing
> non-stereotypic noise were concatenated, increasing the number of scalp
> maps associated with independent time courses and forcing ICA to mixture
> together dissimilar activations into the N output components. The bottom
> line is: ICA works best when given a large amount of basically similar and
> mostly clean data. When the number of channels (N) is large (>>32) then a
> very large amount of data may be required to find N components. When
> insufficient data are available, then using the 'pca' option to jader.m
> <http://sccn.ucsd.edu/eeglab/locatefile.php?file=jader.m>** to find fewer
> than N components may be the only good option.*"
>
> I don't know if each of my datasets has enough datapoints for performing
> an ICA. Each dataset has 20 channels, last 2 minutes and is 4 seconds
> epoched, baseline corrected and pass band filtered. I also reject bad
> epochs.
>
> Which is the meaning of "trials" in the quotation above? Would be better
> to have longer registrations?
> When I run ICA I got 20 components. Why are there some examples with 256
> components?
>
> Dorian
>
> Dorian
>
>
> _______________________________________________
> 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
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20151027/00c68e9d/attachment.html>