[Eeglablist] Is it appropriate to use Time Point and Channel as random variables?

Stephen Politzer-Ahles politzerahless at gmail.com
Tue Sep 24 00:42:12 PDT 2019


This is a nice idea, but time point could not be random since it is a
[nearly] continuous variable. Channel could technically be random, but
there are dependencies across channels (e.g., Fz is more similar to F1 than
it is to P6) in a way that wouldn't be accounted for as a random effect;
channel makes more sense as a fixed effect (since a given channel has a
fixed meaning across studies---Fz is always Fz, whereas "participant 1" is
presumably not the same person across studies). For more on what qualifies
as a fixed vs. a random effect, see
https://dynamicecology.wordpress.com/2015/11/04/is-it-a-fixed-or-random-effect/
.

If you don't want to have to choose spatiotemporal windows for averaging, a
better option would be to use a cluster-based permutation test or a mass
univariate non-cluster-based permutation test, both of which allow you to
test every sample and correct for multiple comparisons. See Maris &
Oostenveld 2007, or Groppe et al. 2011, for information.

---
Stephen Politzer-Ahles
The Hong Kong Polytechnic University
Department of Chinese and Bilingual Studies
http://www.mypolyuweb.hk/~sjpolit/
<http://www.nyu.edu/projects/politzer-ahles/>


On Tue, Sep 24, 2019 at 3:36 PM Chao Han <hanchao at udel.edu> wrote:

> Dear all,
>
> This is not about EEGLab, but I would really appreciate your input.
> I'm doing an ERP analysis on a mismatch negativity study. The conventional
> approach is to run analysis on the averaged deviants minus standards wave,
> across time points and channels in some specific time window and spatial
> region. But is it appropriate to, instead of averaging across time points
> and channels, use Time Point and Channel as random variables in a
> mixed-effects model for analyzing MMN? Thank you!
>
> Best regards,
> Chao
> _______________________________________________
> 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
>


More information about the eeglablist mailing list