[Eeglablist] Neighborhood Distance for Mass Univariate

Amie Jeannette Durston ajdurston at edu.uwaterloo.ca
Thu Apr 25 10:07:44 PDT 2019


Hi Eric,

Thanks for your response. Yes I see. I was just thinking about how the distance effects different types of corrections, lets say the clustering mass correction which uses the distance. Mainly I am just nervous that not having the "correct" number will shift the results from what they "should be". Regardless it is comforting that the Type I error will be controlled regardless of the decided distance.

Thanks again,
Amie
________________________________
From: Eric Fields <eric.fields at bc.edu>
Sent: Thursday, April 25, 2019 12:21 PM
To: Amie Jeannette Durston
Cc: eeglablist
Subject: Re: [Eeglablist] Neighborhood Distance for Mass Univariate

Hi Amie,

I don't think there is a "correct" answer to this. The permutation approach will ensure that the Type I error rate is controlled at alpha for any neighborhood distance you use, so the just want to choose a value that maximizes power and/or gives cluster that are make sense. But which value maximizes power may depend on the characteristics of the effects you are analyzing. I would just choose a value that makes intuitive sense in terms of which electrodes it makes neighbors.

Eric

-----
Eric Fields, Ph.D.
Postdoctoral Fellow
Cognitive and Affective Neuroscience Laboratory<https://www2.bc.edu/elizabeth-kensinger/>, Boston College
Aging, Culture, and Cognition Laboratory<http://www.brandeis.edu/gutchess/>, Brandeis University
eric.fields at bc.edu<mailto:eric.fields at bc.edu>


On Thu, Apr 25, 2019 at 11:06 AM Amie Jeannette Durston <ajdurston at edu.uwaterloo.ca<mailto:ajdurston at edu.uwaterloo.ca>> wrote:
Hi Everyone,

Just wondering what neighbourhood distance all of you are using with mass univariate analysis?
We are using a 64 channel biosemi extended (additional PO9 and PO10 electrodes on the cap + 6 facial electrodes for a total 72 electrode montage).
We are looking around 0.4 (0.3759 to be exact), but we aren't sure this is right.
Has anyone found the "perfect" number for a similar system or if there is an easy way to calculate it?

Thanks,
Amie


Amie J. Durston

B.Sc Honours Psychology Candidate
Co-op Research Intern
Face Processing & Social Cognition Lab
University of Waterloo
_______________________________________________
Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu<mailto: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<mailto:eeglablist-request at sccn.ucsd.edu>


More information about the eeglablist mailing list