[Eeglablist] Options for bad reference (mastoid) channels

[Benedikt Ehinger]

Hi Becky,
We exclude the bad channels before average reference, else you add a 
small amount of the noise back into all channels. Keep in mind that 
average referencing might reduce the rank of your data (compare the 
discussions on the maillinglist for details).
To your second question: I think all of these steps should not interfere 
with the average referencing.

Best,

Benedikt


Am 26.09.2013 14:28, schrieb Becky Prince:
> Dear Makoto,
>
> Thanks for your response!
>
> I'll be analyzing my data in channel space, so I'll go with average 
> referencing as you suggest.
>
> Can you (or anyone else) please tell me whether to exclude bad 
> channels (including both mastoids and any others) before or after 
> applying a common average reference to the data?
>
> Also, is it ok to re-reference to a common average after having 
> completed the following processing steps: 1) remove DC offset, 2) 0.1 
> high-pass filter, 3) epoch the data?
>
> Many thanks!
> Becky
>
>
> ____________________________________________
>
> Becky Gilbert (nee Prince)
> PhD Researcher
>
> Room A109
> Department of Psychology
> University of York
> Heslington, York, YO10 5DD, UK
>
>
> On 25 September 2013 18:46, Makoto Miyakoshi <mmiyakoshi at ucsd.edu 
> <mailto:mmiyakoshi at ucsd.edu>> wrote:
>
>     Dear Becky,
>
>     > I believe the convention is to use the average of linked ears or
>     mastoids as the reference for optimal measurement of these components,
>
>     It depends on whether you want to use ICA to analyze your data in
>     the source-resolved EEG or not. If you do, re-reference does not
>     affect the ICA results. If you want to use channel EEG for the
>     final analysis, I recommend average referencing. If you need to
>     choose 1 electrode, choose Cz or FCz because that's another
>     'convention' I've seen so far (may not specific to your task
>     though). Cz and FCz are easy to interpolate.
>
>     Makoto
>
>     2013/9/18 Becky Prince <becky.prince at york.ac.uk
>     <mailto:becky.prince at york.ac.uk>>
>
>         Dear EEGLAB list,
>
>         I've run an ERP study on auditory temporal expectations in
>         which I'm interested in the auditory N2b and P3b components.
>          I believe the convention is to use the average of linked ears
>         or mastoids as the reference for optimal measurement of these
>         components, but my mastoid channels are consistently noisy in
>         all of my participants.  The mastoid channels have higher
>         impedances (~15 KOhms) than the other sites, probably due to
>         the fact that the caps we use have built in mastoid sites that
>         don't fit close enough to the skin.
>
>         I'm guessing it's a bad idea to re-reference to noisy
>         channels, so I'm wondering what my options are.  Can anyone
>         explain what they would do in this situation?  Are there any
>         resources that explain the effects of re-referencing on
>         specific components?  I'm wondering if I'll be able to see
>         changes in N2b/P3b if I use another reference, e.g. average of
>         all electrodes, and how the appearance of the components
>         (sites/polarities) will be different from those typically
>         reported with a linked ear or mastoid reference.
>
>         Apologies for the fact that this isn't strictly an EEGLAB
>         question - I'm hoping someone will still offer some advice!
>
>         Thanks,
>         Becky
>         ____________________________________________
>
>         Becky Gilbert (nee Prince)
>         PhD Researcher
>
>         Room A109
>         Department of Psychology
>         University of York
>         Heslington, York, YO10 5DD, UK
>
>         _______________________________________________
>         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>
>
>
>
>
>     -- 
>     Makoto Miyakoshi
>     Swartz Center for Computational Neuroscience
>     Institute for Neural Computation, University of California San Diego
>
>
>
>
> _______________________________________________
> 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

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20130929/ca44786e/attachment.html>