[Eeglablist] SIFT - a suitable method for multiple subjects?

Makoto Miyakoshi mmiyakoshi at ucsd.edu
Sun Nov 16 14:49:05 PST 2014


Dear Joshua,

If you want to contact Tim, send email directly to him.

> Firstly, I would wish to ask if there has been any progress on the
group-analysis module for SIFT?



I've heard there was a progress. Tim said he wants someone (me) to test it,
but I haven't heard from him since then.


> Secondly, is it possible to use the centroid components of each cluster
as the measures for SIFT?



No, that does not make sense at all. Multivariate Granger Causality should
be computed on single-trial data.


> Thirdly, in your opinion, would any of the possible solutions suggested
above be a suitable option for group level analysis with SIFT?

Unfortunately no.

Doesn't SIFT now has some kind of group-level analysis already? I thought
Tim implemented a connection to Nima's measure (or network) projection.
Check SIFT and Measure Projection.

Makoto

On Mon, Nov 10, 2014 at 6:11 AM, Baker, Joshua <joshua.baker at ntu.ac.uk>
wrote:

>
>
> Dear Tim/All
>
>
>
> I am currently wanting to use SIFT on a study with a large number of
> subjects. I understand that there is not currently a facility for group
> level analysis in the latest release of SIFT and I am therefore exploring
> methods in order to analyse multi-subject source information flow. My
> colleagues and I have been thinking of different ways that we could use
> SIFT and with what measures. Firstly we thought of using the centroid
> components of each cluster. Would the computed centroid components be
> accessible for use in SIFT? I appreciate the missing variable issue here in
> that we may have differing number of contributing components to each
> computed centroid component. Our second thought was to simply choose a
> representative component from each cluster (a component that is present in
> all/most of the subjects). Although this option is not ideal, it may be
> viable given it would require no further computation/scripting. A third
> option could be to z-transform all connectivity estimates for all pairs of
> (selected) components for all subjects and then calculate averages across
> our subjects.
>
>
>
> Firstly, I would wish to ask if there has been any progress on the
> group-analysis module for SIFT?
>
>
>
> Secondly, is it possible to use the centroid components of each cluster as
> the measures for SIFT?
>
>
>
> Thirdly, in your opinion, would any of the possible solutions suggested
> above be a suitable option for group level analysis with SIFT?
>
>
>
>
>
> Many thanks for your time and consideration,
>
>
>
> Kindest regards,
>
>
>
>
>
>
>
> --
>
>
>
> Joshua Baker. MSc.
>
> Psychology Technician
>
> Psychophysiology Laboratory
>
> Division of Psychology
>
> Nottingham Trent University
>
> Email:Joshua.baker at ntu.ac.uk <mJoshua.baker at ntu.ac.uk>
>
>
>
>
>  DISCLAIMER: This email is intended solely for the addressee. It may
> contain private and confidential information. If you are not the intended
> addressee, please take no action based on it nor show a copy to anyone. In
> this case, please reply to this email to highlight the error. Opinions and
> information in this email that do not relate to the official business of
> Nottingham Trent University shall be understood as neither given nor
> endorsed by the University. Nottingham Trent University has taken steps to
> ensure that this email and any attachments are virus-free, but we do advise
> that the recipient should check that the email and its attachments are
> actually virus free. This is in keeping with good computing practice.
>
> _______________________________________________
> 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/20141116/203a2abd/attachment-0001.html>


More information about the eeglablist mailing list