[Homer-users] parameters for identifying motion artifact and the influences of bad channels

[Jacey Chen]

Dear Hommer-users
I have a question regarding the amp thresh on the motion artifact
detection.
As I remembered from the Homer training and some other articles, it was
usually set as 0.4 or 0.5. However, in our data, I tried different numbers
from 0.5 to 3 in the ampthresh and the motion artifact detection always
marked the whole data set as a motion artifact and excluded all of the stim
marks. (Although when I changed the ampthresh number to 5 and the marks
were gone, I don't know how to justify the parameter we set).

When I took at the individual channel separately, I saw the motion artifact
marks only appeared in the bad channels which we have already excluded. I
do not understand why the excluded channels still influence the stim mark
rejection, and would like to see if anyone got the similar problem and how
to solve that.

I am attaching the screen shot here: Capture1.png shows that all the stim
marks were excluded after processing motion_artifacts_by_Chanel (dashed dot
for the stim mark). The second graph (motion by channel good channel) shows
when the "motion by channel" option was checked, no motion artifact was
marked in these good channels (but the stim mark was still in dashed dot).
The third graph (motion by channel bad channel) further revealed that the
motion artifacts only happened in this bad channel which may result in
excluding all the stim marks.

To sum up, although the bad channels have been excluded, the motion
artifact and stim mark rejection seems to still depend on all the channels
even though we chose "motion_artifacts_by Chanel" for the motion artifact
detection. Is there anyway to have the bad channel out of all the analyses
even for the motion artifact detection? Or is there anything I did wrong?

Thank you!!

-- 
Postdoctoral Research Fellow
Center for Human Growth and Development
Department of Psychology
University of Michigan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20130710/c1eb4ac2/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Capture1.PNG
Type: image/png
Size: 183461 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20130710/c1eb4ac2/attachment.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Motion by Channel good Channel.PNG
Type: image/png
Size: 179932 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20130710/c1eb4ac2/attachment-0001.png 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Motion by Channel bad Channel.PNG
Type: image/png
Size: 190852 bytes
Desc: not available
Url : http://mail.nmr.mgh.harvard.edu/pipermail/homer-users/attachments/20130710/c1eb4ac2/attachment-0002.png