{"id":517,"date":"2024-01-29T10:35:57","date_gmt":"2024-01-29T08:35:57","guid":{"rendered":"https:\/\/benediktehinger.de\/blog\/science\/?p=517"},"modified":"2024-01-29T10:38:40","modified_gmt":"2024-01-29T08:38:40","slug":"on-the-onsets-of-clusters-a-replication-of-rousselet-2023","status":"publish","type":"post","link":"https:\/\/benediktehinger.de\/blog\/science\/on-the-onsets-of-clusters-a-replication-of-rousselet-2023\/","title":{"rendered":"On the onsets of clusters: A replication of Rousselet 2023 \/ adding ClusterDepth"},"content":{"rendered":"\n

I recently read this paper, on the measurement of onsets of clusters in EEG data from Guillaume Rousselet<\/a> and later got asked to be a reviewer of the paper. The main point it raises is a different one to what I adress in the paper: Most people do not explicitly test for an onset, they fall victim to the interaction-fallacy. In principle, you need an explicit test, testing e.g. timepoint 100 vs. 150 to check whether the activity changed significantly.<\/p>\n\n\n\n

But because I recently implemented the ClusterDepth <\/a>algorithm in Julia<\/a>, I thought it would be nice to add this to the papers algorithms to test onsets (and I thought it should do great).<\/p>\n\n\n\n

Let’s start with simulating data – I used our toolbox UnfoldSim.jl <\/a>which made simulation relatively easy<\/p>\n\n\n\n

\n
\"\"<\/a>
Theirs<\/figcaption><\/figure>\n\n\n\n
\"\"<\/a>
Mine<\/figcaption><\/figure>\n<\/div>\n\n\n\n
design = SingleSubjectDesign(;conditions=Dict(:condA=>[\"noise\",\"test\"])) |> x->RepeatDesign(x,50);\n\n\t\nsignal = LinearModelComponent(;\n        basis=basis_signal,\n        formula = @formula(0~1+condA),\n        \u03b2 = [0,1]\n);\n\nnoise = PinkNoise(;noiselevel=1);\nsim(seed) = simulate(MersenneTwister(seed), design, signal, UniformOnset(1,1), noise; return_epoched=true);<\/code><\/pre>\n\n\n\n
Next, I ran clusterdepth a 100 times (the paper uses many more repetitions) using<\/p>\n\n\n\n
cls = clusterdepth(d,statFun=ttest,permFun = permute_fun,nperm=2000)\nonsettimes.clusterdepth[k] = times[findfirst(<=(0.05),cls)]<\/code><\/pre>\n\n\n\n
And plotted everything<\/p>\n\n\n\n
\n
\"\"<\/a>
Theirs<\/figcaption><\/figure>\n\n\n\n
\"\"<\/a>
Mine<\/figcaption><\/figure>\n<\/div>\n\n\n\n
As you can see, Cluster-sum is earliest, then FDR but with many false positives in the bsaaeline, and then maxT \/ clusterdepth.  My implementation contains clusterdepth, but I did not implement the change-point detection. I think the results very nicely replicate Rousselet 2023!<\/p>\n\n\n\n
I am surprised though, that the clusterdepth did not perform better, and I’m currently investigating why this is the case.<\/p>\n","protected":false},"excerpt":{"rendered":"
I recently read this paper, on the measurement of onsets of clusters in EEG data from Guillaume Rousselet and later got asked to be a reviewer of the paper. The main point it raises is a different one to what I adress in the paper: Most people do not explicitly test for an onset, they fall victim to the interaction-fallacy. In principle, you need an explicit test, testing e.g. timepoint 100 vs. 150 to check whether the activity changed significantly. But because I recently implemented the ClusterDepth algorithm in Julia, I thought it would be nice to add this to…<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5],"tags":[],"class_list":["post-517","post","type-post","status-publish","format-standard","hentry","category-blog"],"_links":{"self":[{"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/posts\/517","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/comments?post=517"}],"version-history":[{"count":0,"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/posts\/517\/revisions"}],"wp:attachment":[{"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/media?parent=517"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/categories?post=517"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/benediktehinger.de\/blog\/science\/wp-json\/wp\/v2\/tags?post=517"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}