<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
<title></title>
</head>
<body text="#000000" bgcolor="#ffffff">
Thanks all for your explanations,<br>
<br>
<br>
<br>
Le 24/03/11 14:01, jan-mathijs schoffelen a écrit :
<blockquote
cite="mid:A6FD52BD-9149-402D-8A41-3128E0C6DC8D@donders.ru.nl"
type="cite">Dear JM et al,
<br>
<br>
<blockquote type="cite">
<br>
see for example in version 20110222 ft_sourceanalysis from line
783
<br>
elseif Ntrials>1
<br>
% average the single-trial covariance matrices
<br>
Cy = reshape(mean(data.cov,1), [Nchans Nchans]);
<br>
% select the average ERF
<br>
avg = data.avg;
<br>
Nrepetitions = 1;
<br>
<br>
of course developers of fieldtrip would be the experts to
comment on this.
<br>
</blockquote>
<br>
I am not sure what kind of comment you are looking for here. I
think the previous discussion with much appreciated contributions
by Yuval and Michael was quite conclusive.
<br>
<br>
The difference between keeptrials='yes' and keeptrials='no' is
that the single replicates are kept in memory when
keeptrials='yes' whereas the covariance will be averaged (across
the single trial covariances) if you specify keeptrials='no'. The
default behaviour of sourceanalysis is to average across single
replicate covariances before computing the spatial filters. This
is done to ensure a relatively robust estimate of the covariance
matrix. In general, covariances estimated from single trials are
just very noisy and don't lead to good quality source
reconstructions. </blockquote>
It is clear.<br>
Part of my misunderstanding comes from a very fast overview of the
code and/or documentation
and I did not realize that the option covariance = 'yes' would be
enough to calculate the covariance over the wholes trials and there
is no need to specify keeptrials = 'yes'.<br>
<blockquote
cite="mid:A6FD52BD-9149-402D-8A41-3128E0C6DC8D@donders.ru.nl"
type="cite">In other words, if you would want to have a single
trial estimate of the reconstructed source time series, you still
need to compute the spatial filters based on a covariance averaged
across multiple trials. Once you have the spatial filter, you can
project the single trial sensor-level data through them.
<br>
<br>
I believe this is most easily achieved by a 2-step procedure.
<br>
<br>
tlck = ft_timelockanalysis(cfg, data); (with keeptrials = 'no')
<br>
<br>
<br>
tlck2 = ft_timelockanalysis(cfg, data); (with keeptrials = 'yes')
<br>
<br>
cfg = [];
<br>
...
<br>
source = ft_sourceanalysis(cfg, tlck); (with keepfilters = 'yes')
<br>
<br>
cfg.grid.filter = source.avg.filter;
<br>
cfg.rawtrial = 'yes';
<br>
source = ft_sourceanalysis(cfg, tlck2);
<br>
<br>
This should (I think) give you the single trial time courses.
<br>
</blockquote>
Ok it is an other step further<br>
<blockquote
cite="mid:A6FD52BD-9149-402D-8A41-3128E0C6DC8D@donders.ru.nl"
type="cite">
<br>
But perhaps I am totally missing the point and this is not the
information you were looking for.
<br>
</blockquote>
I do not think so.<br>
Thanks again<br>
<br>
Jean-Michel<br>
<blockquote
cite="mid:A6FD52BD-9149-402D-8A41-3128E0C6DC8D@donders.ru.nl"
type="cite">
<br>
BW,
<br>
<br>
JM
<br>
<br>
<br>
<br>
<br>
Dr. J.M. (Jan-Mathijs) Schoffelen
<br>
Donders Institute for Brain, Cognition and Behaviour,
<br>
Centre for Cognitive Neuroimaging,
<br>
Radboud University Nijmegen, The Netherlands
<br>
<a class="moz-txt-link-abbreviated" href="mailto:J.Schoffelen@donders.ru.nl">J.Schoffelen@donders.ru.nl</a>
<br>
Telephone: 0031-24-3614793
<br>
<br>
_______________________________________________
<br>
fieldtrip mailing list
<br>
<a class="moz-txt-link-abbreviated" href="mailto:fieldtrip@donders.ru.nl">fieldtrip@donders.ru.nl</a>
<br>
<a class="moz-txt-link-freetext" href="http://mailman.science.ru.nl/mailman/listinfo/fieldtrip">http://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a>
<br>
</blockquote>
<br>
<br>
<div class="moz-signature">
<address>-- </address>
<meta http-equiv="CONTENT-TYPE" content="text/html;
charset=ISO-8859-1">
<title></title>
<meta name="GENERATOR" content="NeoOffice 2.2 (Unix)">
<meta name="CREATED" content="20080916;13142500">
<meta name="CHANGED" content="20080916;14013400">
<style type="text/css">
<!--
@page { size: 21cm 29.7cm; margin: 2cm }
P { margin-bottom: 0.21cm }
-->
</style>
<address style="margin-bottom: 0cm;">Jean-Michel Badier</address>
<address style="margin-bottom: 0cm;"><br>
</address>
<address style="margin-bottom: 0cm;" align="LEFT"><font
color="#000000"><font face="Times Roman, serif"><font size="3">Laboratoire
de
MagnétoEncéphaloGraphie</font></font></font></address>
<address style="margin-bottom: 0cm;" align="LEFT"><font
color="#000000"><font face="Times Roman, serif"><font size="3">INSERM
U751. Aix
Marseille Université</font></font></font></address>
<address style="margin-bottom: 0cm;" align="LEFT"><font
color="#000000"><font face="Times Roman, serif"><font size="3">33
(0)4
91 38 55 62 </font></font></font></address>
<address style="margin-bottom: 0cm;" align="LEFT"><a
href="mailto:jean-michel.badier@univmed.fr"><font
color="#000af1"><font face="Times Roman, serif"><font
size="3"><u>jean-michel.badier@univmed.fr</u></font></font></font></a></address>
<address style="margin-bottom: 0cm;" align="LEFT"><br>
</address>
<address style="margin-bottom: 0cm;" align="LEFT"><font
color="#000000"><font face="Times Roman, serif"><font size="2"><font
color="#000000"><font face="Times Roman, serif"><font
size="2">Service
de Neurophysiologie Clinique. </font></font></font>CHU
Timone</font></font></font></address>
<address style="margin-bottom: 0cm;" align="LEFT"><font
color="#000000"><font face="Times Roman, serif"><font size="2">264
Rue
Saint-Pierre, 13005 Marseille-France</font></font></font></address>
<address style="margin-bottom: 0cm;" align="LEFT"><br>
</address>
<p style="margin-bottom: 0cm;"><br>
</p>
</div>
</body>
</html>