prompting an LLM with “Who was the 1st person to walk on the Moon?”, we're not reasoning about Moon landing. In reality, we are asking: given the statistical distribution of words in its training data, what is the most likely continuation of the prompt “The first person to walk on the Moon was...”?

New article with Jessica Morley, Claudio Novelli, and David Watson on "What Kind of Reasoning (if any) is an LLM actually doing? On the Stochastic Nature and Abductive Appearance of Large Language Models" OA: papers.ssrn.com/sol3...

Do you need your toes to think? YESS! Happy to see this one out 😎! From your humble toes to your noble neurons, you need all your cells to solve your most important problem : How to stay alive 😎😅

and all of the arguments we made 5 yrs ago in this paper about why the idea of robot rights relies on faulty assumptions about the nature of human cognition and intelligence still hold Robot Rights?: Let's Talk about Human Welfare Instead dl.acm.org/doi/abs/10.1...

By studying the process through which a soil bacterium naturally produces a well-known drug, scientists have discovered a powerful antibiotic that could help to fight drug-resistant infections go.nature.com/4nLbCoC

When autumn cultivates some doubts, results flourish immediately.

humans and technologies are co-constitutive […] Technologies are imbued with human subjectivity and deployed by creative subjects. […] People may use technologies in innovative and creative ways […] Affordance: objects afford but do not determine. How Artifacts Afford, Jenny L. Davis

Outside of the US, we're not very confused about this either.

“same/except”: a visual relationship in which forms are clearly similar, yet slightly varied > brain perceives both sameness and difference at once. The repetition/variation interplay is central to how we perceive structure, rhythm and depth across mediums. thereader.mitpress.mit.edu/why-repetiti...

“You are here to learn to ride a bicycle, not to invent a bicycle.” 🚲🧪 Read my story in Science Magazine on how I dealt with my "Mid-PhD Crisis". www.science.org/content/arti... @science.org @sciencecareers.bsky.social #ScienceWorkingLife #AcademicChatter #AcademicTwitter #ScienceTwitter

Neuroscience needs a new paradigm: the brain is not a machine | By Nicole Rust “The brain is a dynamic complex system—like the weather or a megacity—whose parts interact via feedback loops that are impossible to study in isolation from each other.” iai.tv/articles/neu... #philsky #philsci #neuro

I wrote a little bit about a cool recent paper looking at heritability estimates from very large registry data, and how we still really don't understand why outcomes track in families. A short 🧵:

A short thread on free will and accountability…

There is - education.

Bjorn Bremb’s arguing that most systems in living organisms are neither completely deterministic nor completely stochastic, but in a middle domain: underdetermined

Gemma de les Coves being entirely honest from the start. #Helgoland2025

Zeilinger: One day another theory will replace quantum mechanics, and it will be even more counterintuitive. (And those who now criticise quantum mechanics will want it back.)

Classic NS, & I mean that in a friendly way. No experiments in quantum physics will, or need, resolve this issue. But I'm sad that I can't get at the article all the same. (BTW there is no "partial free will". Either we have it or we don't. Whatever volition we have is obviously highly constrained.)

Seeing as it's getting topical again, my book INNATE lays out how we can think about genetic influences on behavior without being essentialist about it. Barb Kaiser called it "A powerful antidote to genetic determinism"... press.princeton.edu/books/paperb...

A thought for Saturday mornings in spring: Electric cars have sparked great debates, but it can never be too early for electric gardening tools! We need a moratorium on two-stroke gardening engines or it will always be like waking up to an 80s motorbike grand prix.

I can clearly hear some “told ya” from a distance www.newscientist.com/article/2481...

𝗣𝗮𝗿𝗮𝗹𝗹𝗲𝗹 𝗱𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗲𝗱 𝗯𝗿𝗮𝗶𝗻 𝗳𝘂𝗻𝗰𝘁𝗶𝗼𝗻, 𝗵𝗼𝘄 𝗱𝗼𝗲𝘀 𝗶𝘁 𝘄𝗼𝗿𝗸? Don't expect simple mappings between mind and brain. Check out this piece I wrote for Aeon. aeon.co/essays/how-t...

In a living cell, there is not only dependence of activity on activity but of existence on activity [… :] on the very metabolic activity that they enable for their ongoing repair and replacement […] as an existential imperative. 1/2 A Drive to Survive by @kathrynnave.bsky.social

From Michael Muthukrishna's hopeful message for building a better, more sustainable future, to how humans can stay in charge in a world populated by algorithms, we've compiled a collection of the best reads for this year's graduates: mitpress.mit.edu/the-best-rea...

How much dangerous he is won’t ever be fully assessed futurism.com/elon-musk-gr...

KIBRA & PKMζ proteins stabilize each other by forming a bond. If a protein needs to be replaced, the other remains in place. Bond & its location at the synapses activated in learning are preserved: a new partner can slot itself in, perpetuating your memories. www.quantamagazine.org/the-molecula...

Apropos absolutely nothing: Everything you put *into* ChatGPT/Copilot etc becomes a part of their training data, too. That means if you ask it a question like, "is Carl in Accounting a cheat?" it might say no, but also "Carl in Accounting is a cheat" is part of its training data.

“even if you could incorporate every detail about the imaged neurons and their interactions with one another, the connectome would still represent a single moment in time, devoid of information about how these connections change with experience.” www.thetransmitter.org/connectome/c...

Way too many people hear “language model” and think the algorithm understands language. It does not. It reduces letters to numbers and looks for patterns in the numbers. That’s all it does. That’s all it can do. It does not derive meaning; it does not extrapolate consequences.

Scientists must push back against the threat of rising white nationalism and the dangerous and pseudoscientific ideas of eugenics https://go.nature.com/4jGqiUc

Breaking News: Pope Francis is dead at 88. The first Latin American and Jesuit pontiff, he championed a more inclusive Catholic Church.

Train your own model.

I'm aiming to get some thoughtful biologists together (there's no lack of them!) to discuss what kind of narratives might best reflect the nature of their field in the 21st century. Here's a preview of the goals; I'm excited and optimistic about it. www.marginaliareviewofbooks.com/post/after-t...

Biology is going to accept agency, purpose, meaning. Meaning-sifting of information received from environment: useful info can be integrated into a decision, other is ignored. Attribution of meaning: organisms are sensitive to context & their history www.marginaliareviewofbooks.com/post/after-t...

Here is our new podcast, The Blind Spot, with yours truly, @adamfrank4.bsky.social and Marcelo Gleiser at Spotify. open.spotify.com/show/5RWMnV4...

Here's the first episode of our new podcast, The Blind Spot, featuring yours truly with @adamfrank4.bsky.social and Marcelo Gleiser. We're starting out with some topics from our book but we'll move on from those soon, and we plan to invite guests for future episodes. www.youtube.com/watch?v=b3vT...

Harvard redid its whole homepage to push back against the administration’s demands. I mean, this is just a website but I think it’s kind of a great PR move: www.harvard.edu

"The leader principle, or in German Führerprinzip, is the idea that a single individual directly represents the people, and that therefore all of his actions are by definition legal and proper. In discussions in the White House and thereafter, we see this notion being advanced."

The audio recording of our fantastic event with Kevin J. Mitchell and Anil Seth, FREE AGENTS: Did Evolution Give Us Free Will?, is now up on our website and Youtube Channel. Go take a listen now! www.youtube.com/watch?v=VVr-... @wiringthebrain.bsky.social @anilseth.bsky.social

Answers can be invented. Doubts never.

Does a universal law govern how thing evolve over time to create ever more complexity? And does this mean the future is unpredictable, even in principle, from the present? I explore these ideas in my latest piece for Quanta. www.quantamagazine.org/why-everythi...

This one sounds plausible ⬇️ Is this what he named “a concept of a plan”?

"Free Agents" is now out in paperback! 😊 press.princeton.edu/books/paperb... - all about how life evolved the power to choose!

The U.S. authorities denied entry to a French scientist and then deported him because his phone contained message exchanges with colleagues and friends in which he expressed his “personal opinion” on the Trump administration’s science policies, said the French government. nyti.ms/4iAZR2l

“tunnelling nanotubes (TNTs) are lipid tubules that can bridge cells [..]. All kinds of materials, including whole organelles such as mitochondria, can be exchanged between TNT-linked cells, *undermining their identity as individual living systems.*” www.chemistryworld.com/opinion/tunn...

Beyond Mechanism—Extending Our Concepts of Causation in Neuroscience onlinelibrary.wiley.com/doi/10.1111/... - really pleased that this paper with Henry Potter is now published in the European Journal of Neuroscience 😊

pioneerworks.org/broadcast/fr... toddlers say “no” precisely to exercise their own agency. They want to be free &, in that desire, they become free @wiringthebrain.bsky.social defined toddlers: “willful little things with a burning desire to make decisions for themselves” one.npr.org/i/1185005974...

Harvard University, one of the nation’s wealthiest schools, on Monday joined the list of universities across the country imposing hiring freezes, citing the uncertainty created by President Trump’s threats to slash funding for higher education.