Are you at #OHBM2024? Come check out my talk on Localized Cluster Enhancement - TFCE revisited with valid error control, in the oral session on advanced statistical methods on Wednesday at 11.30. Or poster 1871, on Wednesday/Thursday to see how to improve the validity of TFCE (1)

RT @mbeisen: It's an absolute scandal and tragedy that India is forced to spend $250m/year to provide access to the latest science to its p…

@ValerioCapraro I get strongly disagree for both man/woman for both gpt4o and claude.

RT @MFarajtabar: 1/ Can Large Language Models (LLMs) truly reason? Or are they just sophisticated pattern matchers? In our latest preprint,…

RT @sTeamTraen: Just submitted this review. Is my request reasonable? The paper consists mostly of hierarchical regression analysis, with a…

RT @JuanEugenioIgl1: We have updated NextBrain with: - new atlas with improved brainstem. - updated segmentation code. - new labeled ex viv…

RT @PaulNaish78: Big News! On behalf of the ASA, Taylor & Francis are launching a new journal: 𝗦𝘁𝗮𝘁𝗶𝘀𝘁𝗶𝗰𝘀 𝗮𝗻𝗱 𝗗𝗮𝘁𝗮 𝗦𝗰𝗶𝗲𝗻𝗰𝗲 𝗶𝗻 𝗜𝗺𝗮𝗴𝗶𝗻𝗴 The…

RT @lpachter: Challenge accepted. Here are a few comments on the paper after starting to wade through its massive content. The paper in que…

If you're not interested in the null hypothesis of zero effect then you shouldn't be testing it. But then you have to specify a minimum effect size of interest. If you can't do that maybe you should be testing the null.

@PessoaBrain If that's the case then it substantially limits the Bayesian's ability to criticise the null hypothesis testing framework. I.e. if you can't say when an effect is too small to matter then you shouldn't have a problem with testing the null (directionally).

RT @lakens: Just a reminder that if you can not specify which effects are too small to matter, and 'any effect matters', no effect can fals…

@Yifanyu12 In what sense does it account for more variation than the ALE reshuffling null?

RT @Yifanyu12: We introduce a new coordinate-based meta-regression (CBMR) framework for coordinate-based meta-analysis (CBMA), which uses a…

RT @TimothyRaben: The UK Biobank (UKB) has been an amazing field changing resource for the past decade. Unfortunately, and very suddenly, t…

@benedikt_sun @afni_pt @sNeuroble @CyrilRPernet @ten_photos @AGerlachPhD Right but that's my point, the type of see-through thresholding that @afni_pt is advocating cannot distinguish signal from noise. And I'm not just talking about the global null but also local nulls around the brain.

@afni_pt @sNeuroble @benedikt_sun @CyrilRPernet @ten_photos @AGerlachPhD But I come back to my earlier question which is what would you plot if you just have smooth noise? (Something you can't actually know). The plots would be very similar it would seem meaning that these plots can be rather misleading! And thus this does not count as evidence.

@afni_pt @sNeuroble @benedikt_sun @CyrilRPernet @ten_photos @AGerlachPhD What do you mean by "evidence of the results" that sounds good but I don't think it's well defined. You have no probabilistic guarantees on those findings.

@edwinhayward The libdems got about 10 percent of the vote and just over 10 percent of the seats so their seats are basically what they would have got under proportional representation. So I don't see why he shouldn't be celebrating.

@CyrilRPernet @benedikt_sun @afni_pt @ten_photos @AGerlachPhD @sNeuroble Thanks for the reference, it looks interesting! I'll give it a proper read when I have a chance a bit later.

@benedikt_sun @afni_pt So we should have a pre-print on how to do this in neuroimaging coming out at the end of the summer hopefully. However to see the idea in a different context (2D climate data) which has the same principles see e.g. this reference: