BREAKING: Police have released a sketch of the alleged ringleader of the Kit Kat heist.

How spatial configuration shapes memory and cognitive load: A behavioural and EEG study of simultaneous spatial and associative memory www.sciencedirect.com/science/arti...

A great entry into the proposals available for physiologically plausible gradient descent! I think the way they use dendrite targeting inhibition in this model is particularly elegant. Time to start testing these ideas folks!!! #neuroscience 🧪 #NeuroAI

I love how in the space of ~10 years the field has moved from “the brain can’t do gradient descent” to “which of the many possible ways of implementing gradient descent is the brain doing, in which circuits?”

🚨 We're very happy to introduce TRIBE v2: a foundation model of the brain's responses to sight, sound & language. 📄 Paper: ai.meta.com/research/pub... ▶️ Demo: aidemos.atmeta.com/tribev2/ 💻 Code: github.com/facebookrese... 🤗 Model: huggingface.co/facebook/tri...

Our latest work, on modeling the human brain responses to sight, sound and language at scale:

war_iran_final-blow_v1.pdf war_iran_final-final-blow-v2.pdf war_iran_final-blow-no edits.pdf war_iran_final-blow-no edits (1).pdf war_iran_final-blow-no edits (1) APPROVED.pdf IRANWARPLANSFINALFINALFINAL.pdf IRANWARPLANSFINALFINALFINAL (1).pdf

Historic levels of insider trading are occurring here,,,and now.

This is awesome - great response to the Eon Systems hype.

🧵 New preprint led by @bingbrunton.bsky.social, @elliottabe.bsky.social, @lawrencehu.bsky.social We gave a worm brain control of a fly body and it walked What did we learn? Nothing, other than deep reinforcement learning is effective We call it the digital sphinx www.biorxiv.org/content/10.6...

What are the systems in neuroscience that we really have something that we can call “explanation” at all relevant levels, other than reflexive feed-forward like circuits. Here are a few that I would argue are getting there. Obviously not complete explanations but genuinely satisfying.

Super interesting - certainly if dopamine signals are multiplexed like this it could explain a lot. Also, really cool to think of acetylcholine as the demixing signal. Such neuromodulatory interactions open so many possibilities! #neuroscience 🧪

Thrilled to share our new paper, which shows that the relative timing of cholinergic and dopamine release dynamically gates whether dopamine acts as an RPE for in vivo plasticity and reinforcement learning. www.nature.com/articles/s41...

Awesome work by @crosenthal.bsky.social's team:

This is an important one. That soma-targeting interneurons can do background subtraction and gain modulation is well understood. But the roles of dendrite-targeting interneurons in normalizing top-down learning signals have not been studied in detail.

Really stoked to see this project in (pre)print. On how to implement normalization-via-inhibition in ANNs, and what it might mean for inhibition in the brain. ⚖️ Great job @royeyono.bsky.social and the team 👏👏👏

How do neural circuits in the brain implement normalization? 🧠 In our new paper, we show that just normalizing sensory input isn't enough. Crucially, we must also normalize the error signals! 🧵👇 Paper: arxiv.org/abs/2603.17676

Cosyne is always the best week of the year but this year it was just spectacular. Congrats @neurokim.bsky.social, @markhisted.org, @glajoie.bsky.social, and Alex Cayco Gajic for one of the best programs ever.

Cosyne invited me to give a long tutorial (4 hours!) on methods to quantify differences high-d neural recordings across animals, brain regions, deep neural nets, etc. The recording is up on youtube. I hope it inspires more research on this fundamental topic! www.youtube.com/watch?v=n44x...

Here's a lovely #blueprint on a new study from our lab led by @royeyono.bsky.social. tl;dr: it implies that there may be interneurons whose role is to normalize credit assignment signals during learning. #neuroscience 🧪

I know I've said this a trillion times but I have to say it periodically to keep from going insane: The right ran a huge, well-funded, & ultimately successful campaign to completely take over political media & the infosphere more generally. This is by far the most important political development...

When I started working on UK energy in 2007, fossil fuels provided ~78% of electricity. Coal alone: 34%. Wind & solar: less than 2%. Today: wind & solar 33%, fossil fuels 28%, coal 0%. The chart below tells the story better than words can.

Congrats Charlie and Giles! Long overdue. You were just kids when you proposed this groundbreaking work. Explore this gift article from The New York Times. You can read it for free without a subscription. www.nytimes.com/2026/03/18/t...

Interesting result, but this looks much more like selection and survivor bias than a causal effect of street design on the hippocampus. The association is between people, within-person change is null, and attrition /compositional change across waves is substantial. www.nature.com/articles/s44...

Pleased to share that our paper "Representation Biases: Variance is Not Always a Good Proxy for Importance" is now out as Theory/New Concepts paper in eNeuro! www.eneuro.org/content/13/3... 1/

New paper! We introduce JEDI, Jointly Embedded Dynamics Inference for neural dynamics. arxiv.org/abs/2603.10489. JEDI flexibly infers dynamical principles (across behaviors/contexts) from neural population data through RNNs constrained at single-neuron resolution to reproduce that data.

Genuinely just bonkers to watch the USA do this to one of the most successful and innovative hubs of scientific research the world has ever seen. All those years of Free Speech On Campus debates and it turns out they actually wanted less cancer research. Absurd.

This is kinda bonkers. In addition, 20 amino acids were found on Ryugu, along with uracil and vitamin B3. These have been found in other instances, but since the samples from Ryugu were collected directly from the asteroid and delivered in sealed capsules, contamination on Earth could be ruled out.

Sharing “Neural Thickets”. We find: In large models, the neighborhood around pretrained weights can become dense with task-improving solutions. In this regime, post-training can be easy; even random guessing works Paper: arxiv.org/abs/2603.12228 Web: thickets.mit.edu 1/

Writing out a conversation I’ve been having a lot at this conference: Things in US science are far, far worse than people know. Far worse than even other scientists know. 1/

📣 We've extended the application deadline for #Neuromatch and #Climatematch Academy Students and Teaching Assistants by one week! 🚨 📅 New deadline: March 22, midnight in your local time. ➡️ Apply now: portal.neuromatchacademy.org/sign-in Tag someone who almost missed out! #TA #SummerSchool

1/3) Small non-science rant: I don't normally comment on this kinda stuff, but putting Timothée Chalamet "in his place" for having stated a simple fact (ballet and opera do in fact require government subsidies to survive, and that does tell you something about their popularity) is just so silly...

New lab preprint! Human Navigation Behaviour and Brain Dynamics in Real-world Contexts with co-authors: @pfvelasco.bsky.social and @antoinecoutrot.bsky.social arxiv.org/pdf/2603.11347

Are you #Coayne2026? Wanna know how we can have a common notion of state in mind and machines and find them in an unsupervised fashion with temporal embeddings based on dynamic similarity check @armanbehrad.bsky.social and @maxschwabe.bsky.social’s poster today, 3-001!

Excited to present my poster at COSYNE 2026! We explore whether Successor Features (SFs) in reinforcement learning exhibit properties similar to hippocampal place cells, using analytical tools from neuroscience. Come say hi if you’re around!

Fabulous keynote session at @cosynemeeting.bsky.social. @neurokim.bsky.social and I kicked off the meeting and introduced Chris Olah. m.youtube.com/watch?v=koGy...

Beautiful paper out now from @mhyaghoubi.bsky.social and @markbrandonlab.bsky.social : www.nature.com/articles/s41... #prediction #neuroscience 🧪

Excited to be at #Cosyne2026! I'll be presenting my poster tomorrow during the Thursday session: [1-106] “A biologically inspired predictive coding algorithm with multiplicative and additive feedback”. With @colin-bredenberg.bsky.social and @tyrellturing.bsky.social

For those traveling to #cosyne2026, sad to miss catching up this year, but here's two posters you should check out: 1) Complementary roles of spatial tuning and population geometry in place- vs. memory-guided navigation w/ @audiodrome.bsky.social @apeyrache.bsky.social @tyrellturing.bsky.social.

I tracked every keyword in 22 years of Cosyne abstracts to map how computational neuroscience evolved — from Bayesian brains to neural manifolds to LLMs — and where it's heading next.

Thanks @natmesanash.bsky.social for covering our new work, in @thetransmitter.bsky.social!

As promised. @annaschapiro.bsky.social

Indeed, both my grandmothers worked outside of the house. My paternal grandmother stopped when my granddad got a big promotion at some point, but my maternal grandmother worked full-time her entire adult life. (And still handled all the cooking, etc. - so unfair it's hard to imagine!)

This didn't happen. The perception that it did is largely based on fictional shows that were about rich people. Much more common was a household where one person (theoretically Dad in a het household) was the higher earner and paid the housing costs and the other paid utilities & food

The most impt change at #NIH and to US science this year is bigger than grant cancellations— it’s how the agency is governed. For 75 years NIH has been largely independent of presidential control. That’s changed this year. New piece from me and @nataliebaviles.bsky.social in @nature.com 🧪

Compositional representation of self, others, and gaze direction in *human* hippocampus - super cool. arxiv.org/abs/2603.04747

You know those moments when you see research that so well aligns with your personal interests that ask yourself if you went into the wrong scientific field... yeah, that's this for me. 😜

Raccoon information seeking! Very cool work on raccoons optimizing information gain in puzzle solving from @sarahba.bsky.social and lab: www.sciencedirect.com/science/arti...

The reason businesses put their money behind liberal democracies is because stable=predictable wealth. That the current batch of wealthy think that they could instead become oligarchs in a fascist dictatorship to make more money was always going to backfire. www.techdirt.com/2026/03/02/a...