[Eeglablist] MPT study on ERP for three groups and three conditions

Makoto Miyakoshi mmiyakoshi at ucsd.edu
Wed Nov 25 20:19:13 PST 2015


Dear Mohammed (Nima please comment if possible),

> I am however not sure if I can have a domain I which I have twice as many
healthy people (and their dipoles) than patients. (MPT ignored dipoles of
half of patient group).

Since it's data-driven, it tends to reveal such biases across groups (I
know it is often discussed negatively, which is unfortunate! All the
geneticists know we are inhomogeneous.)

> What is the exact criteria for grouping dipoles (base on ERP)?
If I understand it correctly, it's based on the similarity in both spatial
location and the selected measure. I've never read his paper in detail to
figure it out though.

Makoto



On Fri, Oct 16, 2015 at 3:35 AM, Mohammed Jarjees <
m.jarjees.1 at research.gla.ac.uk> wrote:

> Dear EEGLab team
>
>
>
> I am doing MPT study on ERP. I have 3 groups and 3 conditions. Groups are
> healthy people and two groups of patients and conditions are different
> types of movements. I have equal number of participant n each group. So I
> use MPT to find differences between healthy people and patients and need to
> be sure that what I see is a consequence of pathology rather than
> unbalanced dipoles per group.
>
>
>
> From MPT theory I know that if we have one group and several conditions,
> it is not necessary to have every single subject in each domain. One might
> argue that the same should imply if we have different number of groups. I
> am however not sure if I can have a domain I which I have twice as many
> healthy people (and their dipoles) than patients. (MPT ignored dipoles of
> half of patient group). I tried changing correlation from 0.6 to 0.8 but
> this did not affect the number of dipoles.
>
>
>
> What is the exact criteria for grouping dipoles (base on ERP)?
>
> Paper of Bigdely-Shamlo et al. 2014 Neuroimage, explains that the method
> is based on significant location selection.
>
> I just wonder if that can be a strong argument to explain why MPT removed
> data of half patient group and whether a comparison between patients and
> healthy is still valid?
>
>
>
> Best Regards
>
>
>
> Mohammed
>
>
>
> _______________________________________________
> 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/20151125/b7286766/attachment.html>


More information about the eeglablist mailing list