iiinova — where your digital self keeps evolving

Grace Lindsay @neurograce.bsky.social · 25m

I'd settle for a taxonomy of the computations *claimed* as important for any given task

220

Grace Lindsay @neurograce.bsky.social · 26m

N psychologists and N+1 different answers is my guess

100

Reposted by Grace Lindsay

Blake Richards @tyrellturing.bsky.social · 2h

I signed! This is common sense, IMO, and would be obvious in most other industries with the potential to have major impacts on other parts of the economy.

2143

Grace Lindsay @neurograce.bsky.social · 2h

It's probably not that there is *no* structure here. But yea, we are inflating how much brain/behavior can be neatly carved by these concepts

000

Grace Lindsay @neurograce.bsky.social · 2h

This has been a benefit of LLM-guided paper search. Instead of looking for "attention" papers, I can look for papers that show stimulus X and do manipulation Y, whatever they call it.

110

Reposted by Grace Lindsay

Matthias Nau @matthiasnau.bsky.social · 3h

If you're interested, we discuss this issue extensively in this paper: www.nature.com/articles/s41..., arguing that we need to understand task demands, how they are shared or unique across tasks, and how they constrain neurons & behavior, independent of which mental process a task is assumed to test

nature.com

Centering cognitive neuroscience on task demands and generalization - Nature Neuroscience

Task demands are a primary determiner of behavior and neurophysiology. Here the authors discuss how understanding their influence through multitask studies and tests of generalization is the key to ar...

391

Reposted by Grace Lindsay

Shahab Bakhtiari @shahabbakht.bsky.social · 4h

We can probably isolate most cognitive processes in experimental designs, but how ecologically valid is that? If our tasks force artificial independence between processes, do the conclusions neatly generalize to real-world scenarios where those processes are deeply entangled?

151

Grace Lindsay @neurograce.bsky.social · 5h

There's no anthropomorphization risk for those of us studying humans. But, in any case, to me the question is not if these things 'exist' but rather if they are the most useful way of approaching the question of how the brain (or an ANN) work. I don't know that we can say it is.

110

Grace Lindsay @neurograce.bsky.social · 5h

I don't do clinical work so I'm afraid I wouldn't know

010

Grace Lindsay @neurograce.bsky.social · 5h

When pre & post cues are the same, attention is aligned & we'd expect performance to be better. But part of this process relies on how subjects compress the visual scene and store it in memory, to be able to provide the post-cue info after the scene is gone. This part is under-discussed.

211

Grace Lindsay @neurograce.bsky.social · 5h

Some attention tasks provide a pre-cue, telling the subject what might be important in the visual scene they about to see, and then, afterwards, a post-cue which tells them what information about the scene they saw they need to provide.

111

Grace Lindsay @neurograce.bsky.social · 5h

Now I would also argue that trying to divide up what the brain does in terms of words like 'memory' and 'attention' is ill-guided, so maybe the task names aren't really the problem here after all...

2120

Grace Lindsay @neurograce.bsky.social · 5h

So many tasks used to study visual attention depend critically on processes of memory/compression. It would probably be better if we had more words to clearly label the types of tasks we're doing, to help tease about the functions they test/rely on. #psychscisky #neuroskyence

4326

Reposted by Grace Lindsay

Alyson Smith @pedsaly.bsky.social · 08/10/2026

“It’s a wholly unethical trial, and it shouldn’t be allowed,” says Dr. Paul Offit. “It saddens me to think that our country would in any way participate in a trial where half of the children would be receiving substandard care and thousands will be doomed to develop cirrhosis or liver cancer.”

065

Grace Lindsay @neurograce.bsky.social · 07/10/2026

Right, the differential pricing. When that is done transparently it can be fine, but when done behind closed doors it makes me put my guard up.

020

Grace Lindsay @neurograce.bsky.social · 07/10/2026

It's odd to me that that can't be transparent. Plenty of companies have educational discounts that they post on their sites.

130

Grace Lindsay @neurograce.bsky.social · 07/10/2026

Are there any legitimate reasons why a research equipment website would make you 'request a quote' for a specific piece of hardware instead of just listing the price? Or is it all for upselling/differential pricing?

580

Reposted by Grace Lindsay

Ryan Stauffer @ryanswx.bsky.social · 06/10/2026

Haha, this fucking sucks.

281369443

Reposted by Grace Lindsay

Jason Karsh @jkarsh.bsky.social · 06/10/2026

This is an important statement people, and especially elected Dems need to really understand. The current Republican project is one of destruction, by design. Yes, they have other goals, in RFK’s case he’s objectively pro-disease, but their primary metric of success is breaking government.

26949392

Reposted by Grace Lindsay

Steve Fleming @smfleming.bsky.social · 05/10/2026

Excited to be launching our new PhD program in Minds and Machines at UCL with co-Directors Tali Sharot and @nadinedijkstra.bsky.social To apply, find out more here: www.ucl.ac.uk/brain-scienc... Our launch event is on Nov 10th, with a cracking lineup of speakers: luma.com/cikzcth0

lnkd.in

This link will take you to a page that’s not on LinkedIn

210152

Grace Lindsay @neurograce.bsky.social · 05/10/2026

Are you ready to see how many requests are haunting your inbox 🎃 oooOoOOoooo

240

Grace Lindsay @neurograce.bsky.social · 05/10/2026

We've finally reached the spookiest time of year 👻 ...by which I mean letter of recommendation season

1140

Reposted by Grace Lindsay

Shahab Bakhtiari @shahabbakht.bsky.social · 03/10/2026

I had a feeling recently that seems to be corroborated by chats with colleagues, so I decided to share it here. A lot of systems neuroscientists driven by algorithmic-level questions are going through what looks and feels like a research depression. 🧵 (1/n) 🧠📈 #NeuroAI

139223

Reposted by Grace Lindsay

Vinny Costa @vincentcostaphd.bsky.social · 30/09/2026

Please stop telling high school students to cold email university professors. Really, just stop. If you are a high school student looking for research opportunities, search and apply to the myriad of formal training programs that cater specifically to your education level.

4246

Reposted by Grace Lindsay

Sainsbury Wellcome Centre @sainsburywellcome.bsky.social · 21/09/2026

Applications are now open for the SWC Systems Neuroscience PhD Programme. Join us in London! 🎓 World-class neuroscience training 💰 Fully funded 4-year programme 🖥️Close links to @gatsbyucl.bsky.social Apply by 9 Nov: www.sainsburywellcome.org/web/content/...

02723

Grace Lindsay @neurograce.bsky.social · 24/09/2026

Surya Ganguli kicking off the meeting for this collaboration today, presenting slides (from Lenka Zdeborova) showing the trajectory of how society gains technology before having a full scientific understanding of how it works. Watchable here: simonsfoundation.zoom.us/j/9679556946...

Steam engine prototypes started in 1670s, Steam Engine itself in 1760s, full understanding not until late 19th c. AI still doesn't have full understanding.

1214

Reposted by Grace Lindsay

David Kaye @davidakaye.bsky.social · 24/09/2026

to be clear, the admin is alleged to have deported AT LEAST 25,447 individuals TO COUNTRIES THAT ARE NOT THEIR OWN with, evidently, ZERO PROTECTIONS. this is a scandal and crime of historic proportions.

forbiddenstories.org

The Deportation Project Map

Between 2025 and August 2026, the Trump administration deported over 25,000 people to countries other than their country of origin. Very few of these removals were publicly announced by the US. The De...

3822411169

Grace Lindsay @neurograce.bsky.social · 24/09/2026

Ooo that paper ChatGPT demo is good!

100

Grace Lindsay @neurograce.bsky.social · 24/09/2026

Anyone know of any good short videos or visualizations that cover how modern AI is built and trained in a way that is suitable for a college gen ed class? It can lean a bit technical. The stuff I've surveyed so far on this has been...bleak.

7226

Reposted by Grace Lindsay

AGU Science Policy @aguscipolicy.bsky.social · 21/09/2026

📣 2027 Cohort Applications are OPEN NOW! AGU's Local Science Partners program offers skills training for scientists looking to make an impact in science policy and work with their legislators Join our team and apply by 18 October to make an impact: www.agu.org/honors/local...

Make an Impact with AGU's Local Science Partners

023

Reposted by Grace Lindsay

Kate Marvel @drkatemarvel.bsky.social · 19/09/2026

My hot take on AI and climate change is that I would not like to be killed by either of them

1343392

Reposted by Grace Lindsay

Grace @gracekind.net · 19/09/2026

Frog built a wet lab for the AI model. "There," he said. "Now it can do its own experiments." "What the fuck?" said Toad

13819129

Grace Lindsay @neurograce.bsky.social · 19/09/2026

and it actually takes you to a ReadCube page 😱

130

Grace Lindsay @neurograce.bsky.social · 19/09/2026

What is the disease that journal website designers have where they must make the thing I want most (the actual text of the paper) the smallest part of the page

2306

Grace Lindsay @neurograce.bsky.social · 16/09/2026

It's more that I have no desire to try to revise a paper to convince a reviewer that I've addressed concerns they may have not even raised themselves to begin with.

0150

Grace Lindsay @neurograce.bsky.social · 16/09/2026

Note that this isn't about whether or not AI can raise valid review points (though I usually do still find AI reviews to be missing something about the broader perspective or situatedness of the research).

140

Grace Lindsay @neurograce.bsky.social · 16/09/2026

The thing about using AI to 'clean up' your ideas when writing a paper review is that the review still reads as AI-written. And that makes it very hard for the author to be motivated to respond to it imo

2280

Reposted by Grace Lindsay

Aran Nayebi @anayebi.bsky.social · 14/09/2026

If you're interested in the discussions around the merits of NeuroAI for understanding the mind & brain, check out our #CCN2026 GAC Debate recording! www.youtube.com/watch?v=kYES...

youtube.com

CCN 2026 | GAC: NeuroAI Methods & Frameworks

YouTube video by Cognitive Computational Neuroscience

1173

Reposted by Grace Lindsay

The Washington Post @washingtonpost.com · 04/09/2026

As the first ballots are being mailed out to voters for November’s midterm election, there is an enormous amount of uncertainty and confusion about how Americans will vote. Here are some clarifications:

washingtonpost.com

Analysis | What to watch for when voting in the midterms

The Justice Department plans to send 1,000 monitors to polls and voting by mail could be tricky if you don’t plan ahead.

81810

Grace Lindsay @neurograce.bsky.social · 04/09/2026

The last time I taught the course the investigations were still ongoing actually, so this semester will be the first where it really seems like Duke is totally fine with him

190

Reposted by Grace Lindsay

Nicole Rust @nicolecrust.bsky.social · 03/09/2026

Not stated strongly enough.

Without notifying Congress or the American people, DoD has signed an interagency agreement that could give the Department access to billions of dollars that Congress provided to NIH for critical biomedical research. These funds were provided to support research to develop treatments or cures for infectious diseases such as influenza, malaria, tuberculosis, HIV, Ebola, and other dangerous pathogens.

5210104

Grace Lindsay @neurograce.bsky.social · 04/09/2026

"*you* will get kicked out of the university if you commit fraud, but a professor won't"

1313

Grace Lindsay @neurograce.bsky.social · 04/09/2026

I teach about Ariely's fraud in my methods class and it is probably not the best message to undergrads that he still has his job

521339

Reposted by Grace Lindsay

Philip N Cohen @philipncohen.com · 03/09/2026

Part 2. There is no bottom to Ariely's malfeasance. The Arday tragedy is so raw now the comparison is unavoidable. Ariely's fraud cost millions. Books were sold, careers altered. People and organizations changed policies and practices based on his fake results...

25613