<div dir="ltr"><span style="font-size:12.8px">>>not sure whether I fully understand the problem. May I rephrase: You record data with Cz implicit/online reference. You want to recover the Cz channel data during/after re-referencing. Correct?</span><br style="font-size:12.8px"><div><span style="font-size:12.8px">Yes. </span></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">>>(a) Adding back the reference channel to the data is the to be preferred option as it will reflect the data actually recorded at Cz electrode location. This applies to any re-referencing operation, common or average (except for Biosemi >>data import).</span><br style="font-size:12.8px"><span style="font-size:12.8px">>>(b) pop_interp will only reconstruct/interpolate the Cz data from adjacent electrode sites, thus it will be less accurate (and have a decreased rank).</span><br style="font-size:12.8px"><span style="font-size:12.8px">>>It is expected that (a) and (b) will result in slightly different data. Both approaches are valid but in general (a) is more accurate and to be preferred over (b).</span><br style="font-size:12.8px"></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">Yes, this was exactly what I wanted to know - which option (a)/(b) is more accurate to recover Cz. Your concise explanation clears things up. </span></div><div><span style="font-size:12.8px"><br></span></div><div><span style="font-size:12.8px">Thank you!</span></div><div class="gmail_extra"><br><div class="gmail_quote">2016-05-03 3:28 GMT-04:00 Andreas Widmann <span dir="ltr"><<a href="mailto:widmann@uni-leipzig.de" target="_blank">widmann@uni-leipzig.de</a>></span>:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear Isaiah,<br>
<br>
not sure whether I fully understand the problem. May I rephrase: You record data with Cz implicit/online reference. You want to recover the Cz channel data during/after re-referencing. Correct?<br>
<br>
(a) Adding back the reference channel to the data is the to be preferred option as it will reflect the data actually recorded at Cz electrode location. This applies to any re-referencing operation, common or average (except for Biosemi data import).<br>
(b) pop_interp will only reconstruct/interpolate the Cz data from adjacent electrode sites, thus it will be less accurate (and have a decreased rank).<br>
<br>
It is expected that (a) and (b) will result in slightly different data. Both approaches are valid but in general (a) is more accurate and to be preferred over (b).<br>
<br>
Hope this helps,<br>
Andreas<br>
<div class="HOEnZb"><div class="h5"><br>
> Am 28.04.2016 um 15:18 schrieb Isaiah Innis <<a href="mailto:isainnis@indiana.edu">isainnis@indiana.edu</a>>:<br>
><br>
> Thank you all for your suggestions; I have used them to create a 'template' dataset for channel locations. By editing the channel locations either manually or using the code provided earlier I can properly put a 'Cz' channel in the dataset.<br>
><br>
> However, I am still running into the issue where Cz has a different waveform depending on whether pop_interp was used or the "add ref channel back to data" option in pop_reref. This issue shows up even with datasets that have been preprocessed/cleaned with ICA, so I don't think it's a matter of bad channels.<br>
><br>
> Therefore, I am still uncertain as to which option is valid. My current working assumption is that pop_reref includes 'Cz' in the average computation, thereby producing different results. The wrinkle in this hypothesis is that activity at all sites except 'Cz' is identical - only at 'Cz' is there a discrepancy. This can be seen in these images:<br>
><br>
> <a href="https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!3143&authkey=!ADQdX1b8aLVm9Vo&v=3&ithint=photo%2cpng" rel="noreferrer" target="_blank">https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!3143&authkey=!ADQdX1b8aLVm9Vo&v=3&ithint=photo%2cpng</a><br>
> <a href="https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!3144&authkey=!AFhRR52ewq1woDc&v=3&ithint=photo%2cpng" rel="noreferrer" target="_blank">https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!3144&authkey=!AFhRR52ewq1woDc&v=3&ithint=photo%2cpng</a><br>
><br>
><br>
> Here is a link containing a sample dataset and the code I use for each of the two options. I would greatly appreciate it if someone could confirm what I'm witnessing and advise on what is wrong here (perhaps I'm missing something obvious).<br>
><br>
><br>
> <a href="https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!3146&authkey=!ALCzmvm0xZVfAJk&ithint=folder%2cpng" rel="noreferrer" target="_blank">https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!3146&authkey=!ALCzmvm0xZVfAJk&ithint=folder%2cpng</a><br>
><br>
> Thank you,<br>
><br>
> 2016-04-20 23:04 GMT-04:00 Makoto Miyakoshi <<a href="mailto:mmiyakoshi@ucsd.edu">mmiyakoshi@ucsd.edu</a>>:<br>
> Dear Isaiah,<br>
><br>
> > 1) There seem to be two options - interpolation using pop_interp and the checkbox to "add ref channel back to data" using pop_reref. They produce different results when used:<br>
> <a href="https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!2206&authkey=!AMJUTORsc9y3ao4&v=3&ithint=photo%2cpng" rel="noreferrer" target="_blank">https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!2206&authkey=!AMJUTORsc9y3ao4&v=3&ithint=photo%2cpng</a><br>
> It seems from the discussion here that interpolation after rereferencing is the correct option - I assume because these functions differ in how they calculate Cz?<br>
><br>
> I thought whichever works is fine...until I clicked the link. Wow. Why are they so different? Are you sure that you rejected all bad channels before the process?<br>
><br>
> > 2) After interpolation I receive the following message: "Warning: some channels have the same label." Checking the chanlocs struct there are now two Cz entries, one marked as a data channel and one not. Is this a problem?<br>
><br>
> I believe you can't have two channels with the same name.<br>
><br>
> > Should I delete the non - data Cz?<br>
><br>
> Yes, why not.<br>
><br>
> > 3) When interpolating Cz, do I need to have a dataset with Cz from the same subject or can it be any dataset with Cz?<br>
><br>
> I thought that's easier if you are using EEGLAB. Have one 'template subject' who has all the channels, and always refer to the one to interpolate whatever is 'missing' compared with it.<br>
><br>
> > Do I have to manually edit pop_chanedit to get Cz in a dataset?<br>
><br>
> Whichever more convenient for you (and works) is fine.<br>
><br>
> Makoto<br>
><br>
><br>
><br>
> On Mon, Apr 18, 2016 at 1:32 PM, Isaiah Innis <<a href="mailto:isainnis@umail.iu.edu">isainnis@umail.iu.edu</a>> wrote:<br>
> ---------- Mensaje reenviado ----------<br>
> From: Makoto Miyakoshi <<a href="mailto:mmiyakoshi@ucsd.edu">mmiyakoshi@ucsd.edu</a>><br>
> To: Fang-Yu Chang <<a href="mailto:hardheard101@gmail.com">hardheard101@gmail.com</a>>, EEGLAB List <<a href="mailto:eeglablist@sccn.ucsd.edu">eeglablist@sccn.ucsd.edu</a>><br>
> Cc:<br>
> Date: Wed, 13 Apr 2016 12:25:54 -0700<br>
> Subject: Re: [Eeglablist] Re-referrence to average and add Cz back<br>
> Dear Fang-Yu,<br>
><br>
> > If I would like to add Cz back with computing average, should I operate " 're-reference'-> 'Compute average reference' with 'add current reference channel back to the data' ?"<br>
><br>
> Compute average reference first and then interpolate Cz next.<br>
><br>
> > I would like to ask more question. You mentioned one(any) channel need to be discarded after computing average. How do I discard one channel? I operated " 'Edit' -> 'Channel locations' -> 'delete channel' " by EEGLAB, but it didn't work. I'm not sure it is right to operate like this.<br>
><br>
> Yes this is confusing. Sorry for that. What you need to do is to 'Select data' and not channel editing. Select a channel to discard from the list, check the box to perform rejection for the selected (otherwise it'll leave the selected).<br>
><br>
> Makoto<br>
><br>
> On Tue, Apr 12, 2016 at 11:44 PM, Fang-Yu Chang <<a href="mailto:hardheard101@gmail.com">hardheard101@gmail.com</a>> wrote:<br>
> Dear Makoto,<br>
><br>
> Thanks for your reply. I think I just confused with adding Cz back and interpolating Cz, and Cz dose exist in my dataset.<br>
> If I would like to add Cz back with computing average, should I operate " 're-reference'-> 'Compute average reference' with 'add current reference channel back to the data' ?"<br>
><br>
> I would like to ask more question. You mentioned one(any) channel need to be discarded after computing average. How do I discard one channel? I operated " 'Edit' -> 'Channel locations' -> 'delete channel' " by EEGLAB, but it didn't work. I'm not sure it is right to operate like this.<br>
> Thanks for your help in advanced.<br>
><br>
> Sincerely,<br>
> Fang-Yu Chang<br>
><br>
><br>
> Hello EEGLab list:<br>
><br>
> I still have some confusion regarding Re-referrencing to average and adding Cz back for an EGI system.<br>
><br>
> 1) There seem to be two options - interpolation using pop_interp and the checkbox to "add ref channel back to data" using pop_reref. They produce different results when used:<br>
><br>
> <a href="https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!2206&authkey=!AMJUTORsc9y3ao4&v=3&ithint=photo%2cpng" rel="noreferrer" target="_blank">https://onedrive.live.com/redir?resid=A1398B0A4C8F85DE!2206&authkey=!AMJUTORsc9y3ao4&v=3&ithint=photo%2cpng</a><br>
><br>
> It seems from the discussion here that interpolation after rereferencing is the correct option - I assume because these functions differ in how they calculate Cz?<br>
><br>
> 2) After interpolation I receive the following message: "Warning: some channels have the same label." Checking the chanlocs struct there are now two Cz entries, one marked as a data channel and one not. Is this a problem? Should I delete the non - data Cz? When importing channel locations for our 64 channel system, there are 68 values in chanlocs; after interpolation there are 69.<br>
><br>
><br>
> 3) When interpolating Cz, do I need to have a dataset with Cz from the same subject or can it be any dataset with Cz? Do I have to manually edit pop_chanedit to get Cz in a dataset? I have been using datasets from the same subject, but this issue has made me question that.<br>
><br>
><br>
><br>
> Thank you in advance,<br>
> --<br>
> Isaiah Innis<br>
> Indiana University '13<br>
> EEG Technician, IUB IRF<br>
><br>
><br>
> _______________________________________________<br>
> Eeglablist page: <a href="http://sccn.ucsd.edu/eeglab/eeglabmail.html" rel="noreferrer" target="_blank">http://sccn.ucsd.edu/eeglab/eeglabmail.html</a><br>
> To unsubscribe, send an empty email to <a href="mailto:eeglablist-unsubscribe@sccn.ucsd.edu">eeglablist-unsubscribe@sccn.ucsd.edu</a><br>
> For digest mode, send an email with the subject "set digest mime" to <a href="mailto:eeglablist-request@sccn.ucsd.edu">eeglablist-request@sccn.ucsd.edu</a><br>
><br>
><br>
><br>
> --<br>
> Makoto Miyakoshi<br>
> Swartz Center for Computational Neuroscience<br>
> Institute for Neural Computation, University of California San Diego<br>
><br>
><br>
><br>
> --<br>
> Isaiah Innis<br>
> Indiana University '13<br>
> EEG Technician, IUB IRF<br>
><br>
> _______________________________________________<br>
> Eeglablist page: <a href="http://sccn.ucsd.edu/eeglab/eeglabmail.html" rel="noreferrer" target="_blank">http://sccn.ucsd.edu/eeglab/eeglabmail.html</a><br>
> To unsubscribe, send an empty email to <a href="mailto:eeglablist-unsubscribe@sccn.ucsd.edu">eeglablist-unsubscribe@sccn.ucsd.edu</a><br>
> For digest mode, send an email with the subject "set digest mime" to <a href="mailto:eeglablist-request@sccn.ucsd.edu">eeglablist-request@sccn.ucsd.edu</a><br>
<br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr">Isaiah Innis<br>Indiana University '13<br>EEG Technician, IUB IRF<br><br></div></div>
</div></div>