<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'>Dear experts,<div><br></div><div>I got some weird results when running MVPA (to distinguish two different stimulus categories) on two fMRI datasets with different smoothing using three different ROIs. I would like to know your opinion on why this could happen.</div><div><br></div><div>I used linear SVM.</div><div>The two datasets are from the same data acquisition but with different spatial smoothing: (1) without any spatial smoothing and (2) Gaussian smoothing with sigma=5mm</div><div>Three ROIs: (A) brain area A containing 357 voxels, (B) brain area B containing 30432 voxels, and (C) brain areas A+B containing <span style="font-size: 12pt;">357+</span><span style="font-size: 12pt;">30432=30789 voxels.</span></div><div><span style="font-size: 12pt;"><br></span></div><div><span style="font-size: 12pt;">The classification accuracies when using dataset#1: </span></div><div><span style="font-size: 12pt;"> 0.750 for A, </span></div><div><span style="font-size: 12pt;"> 0.792 for B,</span></div><div><span style="font-size: 12pt;"> 0.792 for A+B</span></div><div><span style="font-size: 12pt;"><br></span></div><div><span style="font-size: 12pt;">The c</span><span style="font-size: 12pt;">lassification accuracy when using dataset#2: </span></div><div><span style="font-size: 12pt;"> 0.875 for A, </span></div><div><span style="font-size: 12pt;"> 0.667 for B,</span></div><div><span style="font-size: 12pt;"> 0.583 for A+B</span></div><div><span style="font-size: 12pt;"><br></span></div><div><span style="font-size: 12pt;">so, using unsmoothed data, combining A and B did not change the classification accuracy. However, using smoothed data, combining A and B reduced the classification accuracy considerably and the accuracy was not significantly higher than chance level any more (all other accuracies were significantly higher than chance level according to permutation test).</span></div><div><span style="font-size: 12pt;"><br></span></div><div><span style="font-size: 12pt;">I would be grateful if anyone could let me know your thoughts why changing the ROI size has different effect on smoothed and unsmoothed data.</span></div><div><span style="font-size: 12pt;"><br></span></div><div>Best,</div><div>Meng</div> </div></body>
</html>