JosetAEtzel

@JosetAEtzel@fediscience.org

Working to increase the likelihood we can trust the what comes out of our fMRI analyses.

Staff scientist (PhD) at Washington U in St Louis (#wustl #WashU #STL #Missouri USA), Cognitive Control and Psychopathology lab. Very keen on #baseR #rstats (esp. for graphing).

I'll occasionally post about (in no particular order): task #fMRI, QC/QA, MVPA, permutation tests robust stats, #knitr, #OpenScience, and research ethics, particularly in the current "red-state" context. Plus a few cats.

she/her

This profile is from a federated server and may be incomplete. Browse more on the original instance.

JosetAEtzel, to stlouis

Good article on the non-cleanup of #radioactive contamination near the #StLouis airport. ("Good" in the sense of interesting and detailed, not the actual situation.)

https://www.popularmechanics.com/military/weapons/a46599173/st-louis-nuclear-fallout/

#STL #nuclear

ai6yr, to random
@ai6yr@m.ai6yr.org avatar

Hmm, well, went to check on the house and the contractors left a giant dump (#2) unflushed and clogged in the toilet, if you're wondering how my remodeling is going.

JosetAEtzel,

@ai6yr Drastically smaller scale project than yours, but same surprise ... unfortunately the plumbing fairy failed to appear before we did, requiring us to make an urgent shopping trip for a more advanced plunger.

JosetAEtzel, to stlouis

A pair of sweet adoptable kittens at Tenth Life Cat Rescue. The siblings are nearly identical, and were busy wrestling just the minute before I took the picture. #StLouis #CatsOfMastodon #kittens

andrew, to random

WELP the 8-yo has it somehow

JosetAEtzel,

@andrew oh no!

JosetAEtzel, to linuxmint

Our trial has been a success: I installed it on an old computer, and had my swears-at-computers husband use it for a few weeks. He found it much less annoying than , and asked me to put Linux mint onto his laptop instead of windows.

I configured it a bit for him, but not much - everything he needs came in the basic install, and he finds the GUI fairly interpretable (biggest confusion was logging out vs powering down).

1/2

JosetAEtzel,

I definitely suggest folks give linux mint a try ... in fact I did suggest it to my father. I was able to talk him through making a bootable USB stick, but not getting the computer to boot from the USB.

I've booted from USB on three windows computers now: one did immediately, one required bios changes, and one required finding the asus-specific key press. This seems to be the hardest part; booting from USB is unfamiliar to many people, and sometimes requires specialized knowledge to enable.
2/2

atomicpoet, to random

Everyone needs to accept that liars exist. Particularly, that liars tend to lie about who they are.

If you don’t allow for the possibility that liars exist and that they may lie to you, then you’re an easy mark.

JosetAEtzel,

@atomicpoet And you will often be wrong.

Not intended as snark - if you are not a lying psychopath yourself it is very difficult to comprehend or accurately predict what one will do.

lakens, to random Dutch
@lakens@mastodon.social avatar

A scientific paper is a necessary but not sufficient summary of the scientific research. Fully reporting your work means making data (or anonymized summaries) and the analysis code that produces all results in the paper transparently available.

JosetAEtzel,

@lakens Yes, and the experimental procedures (e.g., task instructions) as well.

JosetAEtzel, to random

I left my monitor on the table in a hotel conference room during a meeting today. A fairly large room, around 30 people, but at least twice that would have fit comfortably. The room was overly air conditioned in the morning, and I suspect someone asked the staff to adjust the thermostat for the afternoon session ... which apparently further reduced the ventilation.

I was the only attendee masked.

elduvelle, to Neuroscience
@elduvelle@neuromatch.social avatar

Important to be aware of if you’re doing #fMRI experiments or reading fMRI papers!⬇️ #Neuroscience

From: @knutson_brain
https://sfba.social/

JosetAEtzel,

@tdverstynen @elduvelle @knutson_brain Yup ... but not by everyone and often underestimated.

JosetAEtzel,

@knutson_brain @Andrewpapale @tdverstynen @elduvelle Subcortical definitely takes a major hit with MB acceleration. Certainly not impossible; we're doing ok with an MB4 AP 3 mm iso 1 sec TR acquisition in a current study. We tested several acquisitions with a reward task before starting, and I looked very closely at subcortical signal.

If you care most about primary sensory cortices you can get more aggressive, though apparent motion, ghosting, etc. can still be major issues.

JosetAEtzel,

@jonny @knutson_brain @Andrewpapale @tdverstynen @elduvelle RE "to what degree you can tune fMRI acquisition to target specific regions" ... I'd characterize it more as balancing tradeoffs.

Subcortical is extra difficult because of the g-factor penalty, which I understand from physicists is unavoidable (something about heads not being cubes).

Scanners & head coils are also not as interchangeable as you'd think; certain things might work better on one than another. 1/2?

knutson_brain, to random
@knutson_brain@sfba.social avatar

Warning: acquisition can decrease signal to noise ratio due to smaller voxels, shorter echo time, and nonlinear noise induction (thanks to Matt Wall for the mention):
https://apertureneuro.org/article/91292-multiband-acquisition-sequences-for-fmri-proceed-with-caution

JosetAEtzel,

@knutson_brain Amusing in a black humor sort of way that we now have enough material for review articles on some of the ways multiband acquisition is not a panacea. (and I can think of several artifacts & troublesome aspects he didn't mention.)

... and I wish yet again that folks would always carefully test their sequences and do some QC & control analyses before starting studies. (I really am behind on blog posts; I started one after OHBM on this very topic.)

neuralreckoning, to science
@neuralreckoning@neuromatch.social avatar

"Brain Inspired" podcast interview with Mike Frank got me thinking about open science advocates for a "compliance" box checking approach, whereas he thinks we should do it to share the joy of discovery. I agree but would add: we should share our data to make it as easy as possible to prove us wrong. This ought to be part of the culture of science but instead we actively disincentivise it by an adversarial approach to publishing combined with extreme competition for jobs. That's the real problem we need to solve.

JosetAEtzel,

@neuralreckoning @jonny Haven't heard the podcast ... but spent this afternoon outlining a "sales pitch" talk about data sharing practicalities. (Working title, "The Joy of BIDS".)

Structuring and describing datasets and analyses in enough detail for reuse (or checking) to actually be possible is still a non-trivial problem. But I do think standards like BIDS (and the NDA data dictionary for questionnaires) are a step in the right direction; journals like Scientific Data as well.

albertcardona, to Neuroscience
@albertcardona@mathstodon.xyz avatar

Periodic reminder of what fMRI's BOLD signal is measuring, and its temporal dynamics:

"blood oxygenation level–dependent (BOLD) contrast. [...] increased signal in a voxel measured with an EPI [echo planar imaging] sequence indicates recent neuronal activity because of the relative increase in local blood oxygenation that accompanies such activity. The temporal profile of this BOLD response, known as the hemodynamic response function, looks like a bell curve with a long tail, peaking around 4 to 5 seconds after local neural activity and returning to baseline after 12 to 15 seconds."

From "Principles of Neural Science", Kandel et al. 6th edition, page 115.

No matter how fast the EPI imaging is (~100 ms), the BOLD dynamics makes GCaMP look lighting fast. Temporally deconvolving BOLD is possible, to a point, but remember its spatial resolution is measured in millimetres, whereas neuronal somas measure ~0.025 millimetres.

#fMRI #BOLD #neuroscience

JosetAEtzel,

@albertcardona I'm surprised this could be read as a critique of fMRI; the quote strikes me as quite a conventional definition. .... which I guess underscores the need for periodic reminders and precise language (a personal cringe-worthy phrase is "neural correlates).

ai6yr, to random
@ai6yr@m.ai6yr.org avatar

This morning's task. Double check moisture levels in walls before approving mitigation contractor's work. Replace Corsi Rosenthal filter fans.

JosetAEtzel,

@ai6yr "Trust but verify."

Reminds me of the time my husband painted little tick marks on the tires before I took the car in for an oil change and tire rotation. It came back ... with tick marks in the exact same spots. Crew scrambled when I pointed it out, and we haven't been back since.

jonny, to random
@jonny@neuromatch.social avatar

One of the best feelings in the world is to find someone thinking about exactly the same things as you at exactly the same time but further along. Then you get to see into the future, only its a different future where you are also a different person that can see the problem in a different way than you can. It seems like the healthy response is to want to talk to that person and hear what they think and tell them what you think, since you are both thinking about the same things.

How sad is it that many of our friends in academia feel exactly the opposite: you need to constantly be watching out for someone thinking about the same things as you to beat them, cut them down, or coopt them. The last thing you would do is talk to them, because they might "steal" your precious ideas.

The prestige cult is depressing, and if that's an inroad to help our colleagues help themselves to not do DoD supersoldier research or pay for another RELX/ICE surveillance apparatus it seems worth taking

JosetAEtzel,

@jonny Definitely too many of these information-hoarding folks around, but they're not the only type.

I've found that the amount of data and code sharing can be a proxy for likelihood of collaboration; less sharing is a red flag. I see federally-funded research as a form of public service; unethical not to help others build on our work.

arstechnica, to random
@arstechnica@mastodon.social avatar

Toxic toddler fruit pouches: “Extremely high” lead levels sicken 7 in 5 states

Three brands of apple cinnamon fruit pouches have now been recalled.

https://arstechnica.com/health/2023/11/toxic-toddler-fruit-pouches-extremely-high-lead-levels-sicken-7-in-5-states/?utm_brand=arstechnica&utm_social-type=owned&utm_source=mastodon&utm_medium=social

JosetAEtzel,

@arstechnica Note for St Louis folks: is one of the brand of recalled applesauce.

JosetAEtzel, to stlouis

Julia wasn't going to let the fact that the basket is full of toys interfere with her nap. She's adoptable, at Tenth Life Cat Rescue in .

bcmFietser, to random

Someone posted here today that Mastodon was superior for News than Bsky. I've had little problem finding News accounts over on Bsky - but here? How does one go about it? I mean, I've stumbled upon a few, but...what am I missing?

JosetAEtzel,

@bcmFietser

For the US a few to try are
@chrisgeidner
@GottaLaff
@w7voa
@ai6yr
@RollingStone

plus the FollowFriday hashtag.

thadryanjs, to datascience

(1/n) Heads up/PSA/reminder for stats folks.

Almost misinformed my PI about a key variable the other day after stumbling into this little bit of computational profanity:

#rstats #data #datascience #research #stats

@academicsunite

JosetAEtzel,

@thadryanjs @Lluis_Revilla @academicsunite

How about these?
length(which(df$A == 64))
length(which(df[,“B"] == 64))

jonny, to random
@jonny@neuromatch.social avatar

I keep getting emails from my employer about how they are the "#1 public university," and ive only been here a few months, but the widespread consensus and every interaction ive had with it suggests that it is broken as fuck. So ya those metrics are meaningless.

JosetAEtzel,

@jonny Perhaps a second order version of ? Some of the issues arise from "improvements" to the administrative software and procedures that are anything but. Then the long-term staff's institutional knowledge gets less valuable, eventually no one knows how to do one-routine bureaucratic tasks efficiently.

NicoleCRust, to random
@NicoleCRust@neuromatch.social avatar

Provocative

As a young psychologist, this chills me to my bones. Apparently is possible to reach the stratosphere of scientific achievement, to publish over and over again in “high impact” journals, to rack up tens of thousands of citations, and for none of it to matter. Every marker of success, the things that are supposed to tell you that you're on the right track, that you're making a real contribution to science—they might mean nothing.

(I agree: time to rethink the idea that any individual is how this works).

I’m so sorry for psychology’s loss, whatever it is

https://www.experimental-history.com/p/im-so-sorry-for-psychologys-loss

JosetAEtzel,

@tdverstynen @NicoleCRust Theory can also be less than helpful, such as when it takes the form of highly abstracted latent factor analyses (sort of an exponential version of the author's "noun" issue). I'm not convinced that any psychological dataset could meet the statistical assumptions of such models (even things like multiple regression are often a stretch), and wonton use of them can just muddy the waters further.

kinozhao, to random

A longshot cry for help from statisticians/ philosophers of stats:

I vividly remember reading a paragraph quoted from some famous frequentist statistician (I'd like to say Fisher, but I'm not entirely sure) about how we should only use statistical analysis on a data when we don't have other information to go on. And there's a story about the probability of him forgetting to put stamp on his letter is different to a stranger than to himself.

I cannot for the life of me remember where I saw that. Anyone happens to know? 🙏

JosetAEtzel,

@kinozhao The part about only using stats when the answer is unclear reminds me of Di Cook, @visnut. No idea about the second.

Private
JosetAEtzel,

@bojacobs @sts Reminds me of the studies described by Kate Brown in "Manual for Survival".

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • JUstTest
  • mdbf
  • ngwrru68w68
  • InstantRegret
  • magazineikmin
  • thenastyranch
  • rosin
  • khanakhh
  • tacticalgear
  • Youngstown
  • slotface
  • Durango
  • kavyap
  • DreamBathrooms
  • provamag3
  • ethstaker
  • GTA5RPClips
  • modclub
  • tester
  • Leos
  • osvaldo12
  • cisconetworking
  • everett
  • cubers
  • normalnudes
  • anitta
  • megavids
  • lostlight
  • All magazines
    •