Lightnews — Scholar-powered news

Thomas F. Varley @thosvarley.bsky.social · 4h

I think this is interesting and raises a lot of questions about how we think about "information processing" in non-neural systems, and what these measures *really* tell us about the organization of life.

Huge thanks to @doctorjosh.bsky.social and @drmichaellevin.bsky.social for their support. 7/7

Thomas F. Varley @thosvarley.bsky.social · 4h

How you interpret this depends a lot on how you think about these measures. It is common in neuroscience to view these measures as reflecting some kind of "information processing" (dare I say: "computation?")

We know that many of these features vary in brains in different clinical conditions. 6/N

1

Thomas F. Varley @thosvarley.bsky.social · 4h

We are NOT claiming that Xenobots are conscious, or will do your tax returns, or anything like that. But they are undeniably "complex systems" in their own right.

Even though the process of making these organoids is quite brutal. Somehow the cells maintain the capacity to self-organize 5/N

1

Thomas F. Varley @thosvarley.bsky.social · 4h

I want to be very clear: we are NOT comparing organoids and brains directly. We are NOT saying that the Xenobots are "as complex" as brains. Instead, we are showing that the patterns of organization within each system show common statistical features. 4/N

1

Thomas F. Varley @thosvarley.bsky.social · 4h

Using calcium-imaging of Xenopus frog embryos, and fMRI scans of adults, we find that every pattern of activity in the fMRI data is also significantly present in the non-neural embryo.
Meso-scale FC networks, higher-order information, transfer entropy, integrated information - it's all there. 3/N

1

Thomas F. Varley @thosvarley.bsky.social · 4h

In this paper, we take analyses from computational neuroscience (functional connectivity, higher-order information theory, etc), and ask "are these patterns present in non-neural tissues?"

The answer is unequivocally "yes". 2/N

1

Thomas F. Varley @thosvarley.bsky.social · 4h

New paper!
This one is going to raise some eyebrows, so I want to explain the details before people get all worked-up about the title.
1/N
www.tandfonline.com/doi/full/10....

Identification of brain-like complex information architectures in embryonic tissue of Xenopus laevis organoids

Understanding how populations of cells collectively coordinate activity to produce the complex structures and behaviors that characterize multicellular organisms, and which coordinated activities, ...

www.tandfonline.com

1

Thomas F. Varley @thosvarley.bsky.social · 1d

New release of Nvim-Lab, with better LSP configuration, and some new QoL plugins.
github.com/thosvarley/n...

GitHub - thosvarley/nvim-lab: Neovim: for scientists

Neovim: for scientists. Contribute to thosvarley/nvim-lab development by creating an account on GitHub.

github.com

1

Reposted by Thomas F. Varley

Standplaats Kraków @standplaatskrk.pl · Sep 2

Deadlifting skeleton meme: Microsoft is a corporation that turns "Microsoft CEO claims 30% of its new code is written by AI" into "Windows II Update KB5063878 Reportedly Causing SSD Failures"

2 210 550

Thomas F. Varley @thosvarley.bsky.social · 1d

This is probably a good look at the future of AI - small, lightweight systems optimized for a given task getting wired together to produce high-level behavior.

Taylor Webb @taylorwwebb.bsky.social · 1d

Very excited to share that our work (together with co-first author Shanka Subhra Mondal and @neuroai.bsky.social ) on a brain-inspired architecture for planning with LLMs is now out in Nature Communications! www.nature.com/articles/s41... (thread below)

A brain-inspired agentic architecture to improve planning with LLMs - Nature Communications

Multi-step planning is a challenge for LLMs. Here, the authors introduce a brain-inspired Modular Agentic Planner that decomposes planning into specialized LLM modules, improving performance across tasks and highlighting the value of cognitive neuroscience for LLM design.

www.nature.com

1 1

Thomas F. Varley @thosvarley.bsky.social · 1d

A political scandal involving shoddy construction of a dam:

Watergategate

1

Thomas F. Varley @thosvarley.bsky.social · 1d

The rate of change is too damn high. AI, politics, climate change...there is too much stuff all coming to a head at once! I feel like it is genuinely starting the wear on my mental health and/or soul.

1

Thomas F. Varley @thosvarley.bsky.social · 1d

I wonder how much of the current AI market is literally being driven by 5-10 men who genuinely think that, if they can pull this off, they will actually rule the world.

I don't just mean "will have tons of money", but actually taking over the world.

1

Thomas F. Varley @thosvarley.bsky.social · 5d

We cannot continue to expect scientists to simultaneously be: scientists, writers, teachers, public communicators, software engineers, data visualization experts, fundraisers, administrators, editors, and mentors.
It's just too much! All of these are skills that need training and expertise.

1

Thomas F. Varley @thosvarley.bsky.social · 5d

I see *so many* Python packages written by grad students who may be fantastic scientists and experts in their field, but who are not software engineers and therefore write buggy, poorly cobbled together code. I worry that putting it all on Github is letting errors propagate.

1

Thomas F. Varley @thosvarley.bsky.social · 5d

I already worry that a lot of independent scientific software is buggy (leading to unknowingly publishing wrong results) -- the introduction of "vibe-coding" is going to make this infinitely worse.
Scientists *need* to get training on software engineering/development.

Jason Moore @moorejh.bsky.social · 5d

We introduce here the emerging concept of vibe coding, in which AI accelerates coding by converting natural language or abstract intent (a vibe) into functioning software modules, shortening development cycles biodatamining.biomedcentral.com/articles/10.... #vibecoding #bioinformatics #coding

Vibe coding: a new paradigm for biomedical software development - BioData Mining

biodatamining.biomedcentral.com

1 3

Reposted by Thomas F. Varley

NetSci 2026 @netsciconf.bsky.social · 8d

📌 Save the Date!

The flagship conference of the Network Science Society - 𝗡𝗲𝘁𝗦𝗰𝗶 𝟮𝟬𝟮𝟲 - is coming to Northeastern University’s Network Science Institute, 𝗝𝘂𝗻𝗲 𝟭-𝟱, 𝟮𝟬𝟮𝟲.
Prepare to share ideas, discoveries & challenges in network science.

Registration opens soon! 🔗 www.netsci2026.com

34 40

Thomas F. Varley @thosvarley.bsky.social · 5d

This is objectively the correct response, but it also feels like a pretty big step down a road that ends in open and hostile conflict between the federal government and the biggest state.

Governor Gavin Newsom @governor.ca.gov · 5d

IF ANY CALIFORNIA UNIVERSITY SIGNS THIS RADICAL AGREEMENT, THEY'LL LOSE BILLIONS IN STATE FUNDING — INCLUDING CAL GRANTS — INSTANTLY.

CALIFORNIA WILL NOT BANKROLL SCHOOLS THAT SELL OUT THEIR STUDENTS, PROFESSORS, RESEARCHERS, AND SURRENDER ACADEMIC FREEDOM.

New York Times: "Trump Administration Asks Colleges to Sign ‘Compact’ to Get Funding Preference."

The White House asked nine top universities to pledge support for President Trump’s agenda to help ensure access to research funding.

Thomas F. Varley @thosvarley.bsky.social · 5d

This shit is actually evil.

Julia Carrie Wong @joolia.bsky.social · 6d

Some parents are letting their kids talk to ChatGPT in the guise of characters. Some are using it to tell bedtime stories or create coloring books.

"My son thinks ChatGPT is the coolest train loving person in the world. The bar is set so high now I am never going to be able to compete with that.”

‘My son genuinely believed it was real’: Parents are letting little kids play with AI. Are they wrong?

Some believe AI can spark their child’s imagination through personalized stories and generative images. Scientists are wary of its affect on creativity

www.theguardian.com

Thomas F. Varley @thosvarley.bsky.social · 5d

I'd pay so much money for a real Linux phone.

1 1

Thomas F. Varley @thosvarley.bsky.social · 5d

Dropped by the Other Site for a moment and hoo boy. I don't know if it's gotten worse in the 3 months since I was last on, or if my tolerance has just waned, but man - I kind of feel like spending too much time there will give you actual brain damage or something.

1

Thomas F. Varley @thosvarley.bsky.social · 5d

Is that unique to fascism, or is it a general feature of how cultures portray their enemies?

1

Thomas F. Varley @thosvarley.bsky.social · 5d

This isn't an endorsement of a carceral, War on Drugs-style policy (we know that doesn't work), but it's also probably a bad idea to create a system where companies are financially incentivized to turn their customers into literal addicts.

1

Thomas F. Varley @thosvarley.bsky.social · 5d

This goes for social media and smart phones, too!

Thomas F. Varley @thosvarley.bsky.social · 5d

Probably the biggest personal change in my views has been the transition from being essentially libertarian on issues like drug legalization, gambling, etc. to being a lot more skeptical that society should be permissive when it comes to products designed to exploit known bugs in human cognition.

John Gramlich @johngramlich.bsky.social · 6d

% of US men under 30 who say legal sports betting is a *bad thing* for society

22% in 2022
47% in 2025

No other demographic group has seen a bigger increase.

Americans increasingly see legal sports betting as a bad thing for society and sports

Today, 43% of U.S. adults say the fact that sports betting is now legal in much of the country is a bad thing for society, up from 34% in 2022.

www.pewresearch.org

2 1