<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;
mso-fareast-language:EN-US;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:#0563C1;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:#954F72;
text-decoration:underline;}
p.MsoListParagraph, li.MsoListParagraph, div.MsoListParagraph
{mso-style-priority:34;
margin-top:0cm;
margin-right:0cm;
margin-bottom:0cm;
margin-left:36.0pt;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri",sans-serif;
mso-fareast-language:EN-US;}
span.EmailStyle17
{mso-style-type:personal-compose;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri",sans-serif;
mso-fareast-language:EN-US;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-AU" link="#0563C1" vlink="#954F72">
<div class="WordSection1">
<p class="MsoNormal">Dear FieldTrip community,<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">I am a psycholinguist at the Australian National University, working with eye-tracking data. I have been writing a small custom cluster-based
<o:p></o:p></p>
<p class="MsoNormal">permutation analysis script, and I wanted to ask for your help in understanding some details of the algorithm (as implemented in FieldTrip).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">1. Computation of cluster p-values:<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal" style="text-indent:36.0pt">In "clusterstat.m" file, strict inequalities are used when comparing the observed test statistic to the permutation distribution, e.g., at line 443 for positive clusters: <o:p></o:p></p>
<p class="MsoNormal" style="text-indent:36.0pt"> prob(j) = (sum(posdistribution>stat(j)) + 1)/(Nrand + 1);
<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">My understanding has been that "greater than or equal to" needs to be used for computing cluster p-values (e.g., p. 187 in Maris & Oostenveld 2007). And since<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">test statistic can be discrete (e.g., cluster length), there can be cases where it will be exactly equal to one or several values in the permutation distribution.<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt"><o:p> </o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">Why are strict inequalities used in the code?
<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">2. Permutation distribution for two-tailed tests: <o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">In FieldTrip, two test statistics are computed on the observed data (for pos and neg clusters), and each is compared to the respective permutation distribution<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">(followed by a Bonferroni corection).
<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt"><o:p> </o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">It makes sense, but I can't figure out how it compares to an alternative, namely, recording the maximum *absolute* value of the test statistic, and comparing it<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">against a single permutation distribution of maximum absolute test statistics (probably, without a Bonferroni correction?). My understanding is that
<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">Maris & Oostenveld (2007) paper describes this option.<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt"><o:p> </o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">Is it just an implementation choice which would give the same results? Or is using two distributions for pos/neg clusters better than using one distribution
<o:p></o:p></p>
<p class="MsoNormal" style="margin-left:36.0pt">of max absolute values -- if so, why?<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">I would be grateful for any advice!<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Best wishes,<o:p></o:p></p>
<p class="MsoNormal">Anton<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">--------------------------<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal"><b><span style="color:#2F5496;mso-fareast-language:EN-AU">Dr Anton Malko<o:p></o:p></span></b></p>
<p class="MsoNormal"><span style="color:#2F5496;mso-fareast-language:EN-AU">Research Fellow<o:p></o:p></span></p>
<p class="MsoNormal"><span style="color:#2F5496;mso-fareast-language:EN-AU">School of Literature, Languages and Linguistics | College of Arts & Social Sciences<o:p></o:p></span></p>
<p class="MsoNormal"><span style="color:#2F5496;mso-fareast-language:EN-AU">The Australian National University, Canberra
<o:p></o:p></span></p>
<p class="MsoNormal"><span style="color:#2F5496;mso-fareast-language:EN-AU"><a href="mailto:anton.malko@anu.edu"><span style="color:#034990">anton.malko@anu.edu</span></a><o:p></o:p></span></p>
<p class="MsoNormal"><o:p> </o:p></p>
</div>
</body>
</html>