[Eeglablist] STUDY: stats results & display CI or SEM on ERPs

Arnaud Delorme arno at ucsd.edu
Wed Jan 15 18:20:00 PST 2014


Dear Bastien,

to compute confidence intervals, from the command line, you may also reuse the accumulation bootstrap array generated by statcond and enter it as input to the function stat_surrogate_ci.

Also, if you are using the std_plotcurve function (using output from the std_erpplot function for example), there is an option to plot the standard error (using parametric statistics though) for each ERP (command line option only).

Best,

Arno

On Jan 13, 2014, at 8:57 PM, Makoto Miyakoshi <mmiyakoshi at ucsd.edu> wrote:

> Dear Bastien,
> 
> > although I am not using clustering in my study does that change where I am looking for stuff?
> 
> Oh I see.
> 
> Maybe you want to check out mass univariate stat toolbox here?
> http://sccn.ucsd.edu/wiki/Plugin_list_process
> I also recommend you read David Groppe's paper about it. It'll tell you the type of statistics you are interested in and what you want to be careful about.
> 
> > When calling statcond() what do I need to put in the brackets? 
> 
> In the command window type 'help statcond' after launching EEGLAB. It'll tell you what input it requires. If you don't understand it ask us questions again.
> 
> Makoto
> 
> 
> 
> 2014/1/12 Bastien Boutonnet <bastien.b1 at gmail.com>
> Hi Makoto,
> 
> Thanks very much for your replies. I will try and look into what you suggest although I am not using clustering in my study does that change where I am looking for stuff?
> 
> What I really want to do would be to be able to say find the lowest F or t values that are significant and report this in the results section. I wouldn’t want to report all values of course.
> 
> When calling statcond() what do I need to put in the brackets? 
> 
> ERPLAB is indeed good for the classical mean amplitude analyses which are then fed into SPSS or R etc. but I don’t think it implements time point by time point analyses which are what I am interested in. I wouldn’t know how it integrates with the STUDY. I know that the •.erpsets that it creates do not seem to be read by STUDY.
> 
> B
> -- 
> Bastien Boutonnet, Ph.D.
> School of Psychology,
> Bangor University
> bastienboutonnet.com
> 
> On 9 January 2014 at 20:04:28, Makoto Miyakoshi (mmiyakoshi at ucsd.edu) wrote:
> 
>> Dear Bastien,
>> 
>> If you want to report multiple significant points on your ERP (I guess) you need to report bunch of statistical scores... that will make another line plot. To obtain the statistical values for each points, you have to access values under STUDY.cluster(x).icaerp (don't forget that you plot all cluster ERP first, close the edit/plot window by pressing ok, then it will appear) and call EEGLAB function statcond().
>> 
>> I wrote a simple function for testing ERP on STUDY. Download std_ErpCalc() from
>> http://sccn.ucsd.edu/wiki/Plugin_list_process
>> Unzip it, and place the folder under /eeglab/plugins and relaunch EEGLAB. If you encounter a problem let me know.
>> 
>> I also heard that ERPLAB is a good tool to test ERP but I'm not sure if it is compatible with STUDY.
>> 
>> Makoto
>> 
>> 
>> 
>> 
>> 2014/1/6 Bastien Boutonnet <bastien.b1 at gmail.com>
>> Dear all,
>> 
>> First, I have finally managed to run successfully an analysis entirely via the STUDY. It’s great. I’m loving the fact that it gives you stats ms by ms on all electrodes. Found some effects I would have otherwise missed.
>> 
>> Now that I get pretty effects I don’t know how to report them. I would basically like to find where or how to obtain t and F values to report them in a paper. So far all I get is pretty graphs with thresholded p-values.
>> 
>> Second, I was wondering if there was any way in which I could plot SEM or CI 95% on the ERPs produced by the study.
>> 
>> Anyone has a solution for one or both of those issues?
>> 
>> Best,
>> B
>> 
>> -- 
>> Bastien Boutonnet, Ph.D.
>> School of Psychology,
>> Bangor University
>> bastienboutonnet.com
>> 
>> _______________________________________________
>> 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
> 
> 
> 
> -- 
> Makoto Miyakoshi
> Swartz Center for Computational Neuroscience
> Institute for Neural Computation, University of California San Diego
> _______________________________________________
> 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/20140115/32e337c3/attachment-0001.html 


More information about the eeglablist mailing list