<!DOCTYPE html>
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<style>
div.markdown { white-space: normal; }
div.plaintext { white-space: normal; }
body { font-family: sans-serif; }
h1 { font-size: 1.4em; }
h2 { font-size: 1.2em; }
h3 { font-size: 1.1em; }
blockquote { margin: 0 0 5px; padding-left: 5px; border-left: 2px solid #777777; color: #777777; }
blockquote blockquote { border-left-color: #999999; color: #999999; }
blockquote blockquote blockquote { border-left-color: #BBBBBB; color: #BBBBBB; }
a { color: #3983C4; }
blockquote a { color: #777777; }
blockquote blockquote a { color: #999999; }
blockquote blockquote blockquote a { color: #BBBBBB; }
math[display="inline"] > mrow { padding:5px; }
div.footnotes li p { margin: 0.2em 0; }
</style>
</head>
<body>
<div class="markdown">
<p dir="auto">Hi Mats,</p>
<p dir="auto">Thanks for your reply!</p>
<p dir="auto">Most studies are rest-active-rest-active, so subtracting baseline activity is simple. In my analysis, I want to extract the spectral power for different intervals within the epoch/trial and have them baselined to the rest period before the epoch/trial.
The mean of these becomes the spectral power change for this trial.</p>
<p dir="auto">Here is a diagram. And in case it is removed, <a href="https://www.dropbox.com/s/gpy11n3c5dwcq3t/power%20calculation.png?dl=0">
a link to the image</a>.</p>
<p dir="auto"><img src="cid:DA33D254-5D46-4441-9130-9F366756682C@gatech.edu" alt="" title="power calculation.png"></p>
<p dir="auto">Thanks,<br>
John</p>
<p dir="auto">John T. Johnson<br>
PhD Candidate - Cognitive Motor Control Laboratory<br>
Lab TA NEURO 4001 Capstone<br>
School of Biological Sciences<br>
Georgia Institute of Technology</p>
<p dir="auto">On 2 Aug 2019, at 8:28, Es, M.W.J. van (Mats) wrote:</p>
</div>
<div class="plaintext">
<blockquote>
<p dir="auto">Hi John,<br>
<br>
It isn’t completely clear to me what your question is. Which part of your intended analysis are you struggling with to implement? Are you stuck with the first step (i.e. loading in the data and defining trials based on event triggers)? If that is the case,
I recommend you take a look here: <a href="http://www.fieldtriptoolbox.org/example/making_your_own_trialfun_for_conditional_trial_definition/">
http://www.fieldtriptoolbox.org/example/making_your_own_trialfun_for_conditional_trial_definition/</a> and on FieldTrip’s tutorial page (<a href="http://www.fieldtriptoolbox.org/tutorial/">http://www.fieldtriptoolbox.org/tutorial/</a>), for example “Preprocessing
– Trigger based trial selection”. The tutorial page will be helpful in general, especially for FT beginners.<br>
<br>
Best of luck,<br>
Mats<br>
<br>
PhD candidate<br>
<br>
Dynamic Connectivity<br>
<br>
Donders Institute for Brain,<br>
Cognition and Behaviour<br>
<br>
e:<br>
<br>
m.vanes@donders.ru.nl<<a href="mailto:m.vanes@donders.ru.nl">mailto:m.vanes@donders.ru.nl</a>><br>
<br>
a:<br>
<br>
Kapittelweg 29, 6525 EN Nijmegen<br>
<br>
<br>
p:<br>
<br>
+31(0)24 36 68291<br>
<br>
<br>
<br>
<br>
<br>
From: Johnson, John T. <john.johnson@gatech.edu><br>
Sent: donderdag 1 augustus 2019 16:21<br>
To: fieldtrip@science.ru.nl<br>
Subject: [FieldTrip] Extracting spectral power for intervals within a trial<br>
<br>
My EEG experiment has people rest, then perform 6 movements, then rest again, etc.<br>
I would like to perform spectral power analysis on portions of those 6 movements baselined to the rest period before the trail. That is, the entire trial is baselined to its own rest period, then spectral power for different electrode montages (e.g. IPL, SPL,
etc.) for each interval is extracted, meaned per trial, then stats run between the first 15 trials and the last 15 trials, which are performed under different conditions.<br>
<br>
I have event markers for the different movement phases in my eeg data derived from kinematics data. The task is self-paced, so the intervals vary, but as I said, I have markers for the beginning and ending of the times of interest within the trials.<br>
<br>
I’m new to FieldTrip and I am coming from the EEGLAB world. I have data previously cleaned and ICAed there, but I’m not adverse to redoing anything needed.<br>
<br>
Future directions: source localization using MNI templates, and CoSMoMVPA to find patterns I’m not aware of.<br>
<br>
Thanks so much for any advice or pointers to references.<br>
<br>
—<br>
John T. Johnson<br>
PhD Candidate, School of Applied Physiology<br>
GTA, Neuroscience Methods<br>
he/him/his<br>
_______________________________________________<br>
fieldtrip mailing list<br>
<a href="https://mailman.science.ru.nl/mailman/listinfo/fieldtrip">https://mailman.science.ru.nl/mailman/listinfo/fieldtrip</a><br>
<a href="https://doi.org/10.1371/journal.pcbi.1002202">https://doi.org/10.1371/journal.pcbi.1002202</a></p>
</blockquote>
</div>
<div class="markdown"></div>
</body>
</html>