Variability in the analysis of a single neuroimaging dataset by many teams

Variability in the analysis of a single neuroimaging dataset by many teams https://t.co/eb7cOAez5Z

References: - Gelman, A. and Hennig, C. (2017), Beyond subjective and objective in statistics. J. R. Stat. Soc. A, 180: 967-1033. https://t.co/MgSvKMXPyB - Weinberger, D. (2009) Transparency is the new objectivity. Everything is miscellaneous blog, July…

what a wonderful opportunity for inquiry on why metascience studies reported with narrative are so important cc @stuartbuck1 @notanastronomer @DefenderOfBasic we have obv known "fMRI can be kinda useless" but we have not communicated "where, exactly, d…

this isn't new, but yeah fMRI data has always been problematic see: https://t.co/KvdiKYPdFL & https://t.co/yOG5t1I8T9 for decent overviews on the limitations and considerations re: fMRI https://t.co/ezN0FeIVtW

@ActAppalledxx @lordesbbqribs Congrats. I’m aware of the paper too. Has your Bachelor’s degree helped you maintain enough interest in the field for you to be aware that this paper inspired an additional paper, specifically because of the unreliability of …

link to the Nature paper: https://t.co/fTViQAaFUM

RT @ItaiYanai: https://t.co/ExkqAdaR8j

https://t.co/ExkqAdaR8j

RT @rotembot: It was a pleasure talking with @fMRI_today about analytical variability #narps https://t.co/mQ9j8ZHanG, promoting open & repr…

This was a wonderful conversation not only about the implications of the seminal paper comparing the results of many different labs analyzing the same data set, but much more!

RT @rotembot: It was a pleasure talking with @fMRI_today about analytical variability #narps https://t.co/mQ9j8ZHanG, promoting open & repr…

It was a pleasure talking with @fMRI_today about analytical variability #narps https://t.co/mQ9j8ZHanG, promoting open & reproducible science (https://t.co/o0XcwCRyPG; https://t.co/jXqMOxpmxx; https://t.co/nOrjh2GXqn), placebo effects https://t.co/P…

@YaakovOphir @EikoFried https://t.co/gGHNOJn3C6

@DrGoblin3 @Lageraemia @robertrea @BibiLynch One swallow does not make a summer, one battle does not win a war! Chk out "Variability in the analysis of a single neuroimaging dataset by many teams" https://t.co/25ARKmC0WE via @tschonberg et al Are we su…

@Aella_Girl You are not the only one (maybe you've seen this one already): https://t.co/5gbJdZzbbp

Another example of what has been reported in #fMRI (https://t.co/xYUHOJv5Fd). Is multiverse analysis the way to go or are there other ways to take this uncertainty into account in the scientific process?

@AnilOza16 @JkayFlake @HannahSFraser @TimParker88 @Elliot_Gould_ @Nicole_C_Nelson @BrianNosek At least the lack of reproducibility replicates 😆 https://t.co/gtGoFYHOh9

This isn't just an issue in one discipline tho. And there are many reasons why it occurs. Some of it is due to QRPs, some of it due to analytic flexibility plus the reliability of the measures used (e.g., https://t.co/LgMzL305oA, https://t.co/6WKw4dcAM4…

@szorowi1 Same here Not sure if you’ve seen but relevant. Some of the variability in analysis and preprocessing here is due to default settings between software https://t.co/XDKCo1QlPE

Had an amazing time at our inaugural @ReproducibiliT meeting - looking forward to more discussions with all of the amazing scientists at @UCSBpsych! Note to self - keep eyes open in future pictures

RT @avamadesousa: A great turnout for the inaugural @ReproducibiliT meeting at @UCSBpsych discussing a spicy fMRI paper*! 🍵🔍🧠 Excited for…

I should definitely smile more next time haha, like @madhurik951

Thank you, @avamadesousa, @madhurik951, and Amber Chen! So psyched we have our own @ReproducibiliT at @UCSBpsych now!

A great turnout for the inaugural @ReproducibiliT meeting at @UCSBpsych discussing a spicy fMRI paper*! 🍵🔍🧠 Excited for more fun discussions and workshops to come 😃 *https://t.co/0g1Wd8iYoS https://t.co/EAy1S9hBq1

@micahgallen We know that there is great model dependency in the multiverse of madness. But that's true everywhere. @rotembot https://t.co/qVCKIO5BD8

RT @sNeuroble: @neurograce There are a number of papers that say there are major challenges that hinder progress, but I think most (incl us…

@neurograce There are a number of papers that say there are major challenges that hinder progress, but I think most (incl us) don’t go so far as to say the field is not progressing altogether. E.g., questioning common neuro designs/methods https://t.co/5…

Recall yesterday's event... 🍕🧠 #SanoScience #neuropizza @psychologyUJ

RT @brain_more: Today's Neuropizza @SanoScience we talked about harmonizing multisite neuroimaging and @lola_dachshund almost got the pizz…

Today's Neuropizza @SanoScience we talked about harmonizing multisite neuroimaging and @lola_dachshund almost got the pizza. https://t.co/gBOPvo6xre #neuroscience #neuroimaging #journalclub https://t.co/0Q0GkH8EHI

And the variance is gigantic too, assuming that scientists will all interpret the scan data the same way. BUT they don't, 70 teams examined the same scan data and came to vastly different conclusions https://t.co/vpfJTL9ySQ