[Eeglablist] General Continuous Data Analysis Questions

[Makoto Miyakoshi]

Dear Evan,

> I am trying to meet a deadline for my research project (next week) and I
am really struggling with data analysis.

Sorry I read it now.

> My professors are completely unfamiliar with EEGLAB so my knowledge only
encompasses what I could gather from tutorials.

So were my former mentors.

> Please let me know what I am missing/what else you think I should be
doing in order to efficiently get valid data that I can start comparing and
writing about.

In your case, data cleaning before ICA is missing. This is a very important
stage. See this page for future info. Read Nima's paper too (referenced
there)
http://sccn.ucsd.edu/wiki/Makoto's_preprocessing_pipeline

> 7.  Spectral analysis of the epochs to then compare frontal alpha
asymmetry
-How would I calculate a significant difference?

You should use EEGLAB STUDY to run the stats on this. Check also NBT by
Simon Shlomo-Poil.

Makoto

On Thu, Jul 9, 2015 at 7:06 PM, Evan Wilson <erw007 at bucknell.edu> wrote:

> Dear List,
>
> I am a novice in need of some general guidance.  I am trying to meet a
> deadline for my research project (next week) and I am really struggling
> with data analysis.  My professors are completely unfamiliar with EEGLAB so
> my knowledge only encompasses what I could gather from tutorials.
>
> I currently have a raw .edf file (I used the Emotiv EPOC research
> edition).  I am trying to examine frontal alpha asymmetry at specific
> points in an experience and compare it across two separate conditions (one
> where subjects were walking around so I assume there are quite a lot of
> artifacts to remove).  These are the steps I know how to do so far:
>
> 1. I imported the continuous data
>
> 2. I added channel locations (although EEGLAB gives me trouble, saying
> there can’t be empty channels, but I found a way around this)
>
> 3. Remove the baseline (I was told this is likely unnecessary)
>
> 4. Run ICA (do I do this before or after adding high and low pass filters?)
> - Also, what does this do as far as filtering the data? I am still a
> little confused about this but I am under the impression that it is a
> pivotal step.
>
> 5. It was suggested that I could then just filter the data down to ~7 - 14
> Hz and this will filter out most artifacts and preserve the alpha bands,
> but this seems fishy to me.
>
> 6. Epoch the data to just the three or four instances I wish to examine
> -I am unsure about this part.  I do not know if this is the correct next
> step nor do i know how to epoch the data, but I figure I can refer back to
> the tutorials for the latter.
>
> 7.  Spectral analysis of the epochs to then compare frontal alpha asymmetry
> -How would I calculate a significant difference?
>
> Please let me know what I am missing/what else you think I should be doing
> in order to efficiently get valid data that I can start comparing and
> writing about.  Sorry if these are really basic questions, guys, but I am
> looking down the barrel right now and nobody else has been able to help me
> so far.
>
> All the best,
>
> Evan
> _______________________________________________
> 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/20150720/99c22153/attachment.html>