Connect at the ELLIS Sites ICML Fest...in Tübingen/Stuttgart 🇩🇪 on 3 July at @mpi-is.bsky.social Stuttgart. Meet the local AI/ML community and engage with the ELLIS network with the @ellisinsttue.bsky.social. 🔗 More info → https://bit.ly/4g8DHWN #ICML2026

"The prevalence of depressive symptoms was approximately two to three times higher and anxiety symptoms three to five times higher than in age-matched general-population samples These findings underscore the need for structural reforms and implementation of evidence-based support services for ECRs"

It’s surprising how rarely this perspective is taught

We’re happy to announce 2 releases today: - 🧠Brain2qwerty v1 is published at @Nature Neuroscience - 🚀 Brain2Qwerty v2 is now publicly released Explore how we decode sentences from non-invasive brain recordings: facebookresearch.github.io/brain2qwerty/ Code, papers, data below:

Updated preprint and code from my lab. If you want to explore more realistic neural dynamics in models that can still perform visual tasks, check it out! www.biorxiv.org/content/10.1...

Some of the first work related to my ERC is now available as a preprint 👉 arxiv.org/pdf/2606.20623 We model the learning of real-world melodies as resource-constrained inference over generative programs, with new implications for Rate-distortion theory: learning changes what is cheap to encode! 🧵👇

My new book! Written w my pal Alex Townsend, it's about the math behind enormous systems (medical scans, AI, etc.) and what happens when they outgrow our ability to understand them. Pre-order at @barnesandnoble.com for 25% off -> www.barnesandnoble.com/w/big-math-s... @littlebrown.bsky.social

when the PI tries out the postdoc's latest technique 🧪

Everybody should have an Open Mind (paper) 👇👇👇

I've been writing about power-law probability distributions, and man -- we do NOT spend enough time worrying about really powerful solar storms. Not crazy to think that a once-every-500-years event would basically knock out all electrical technology on Earth. en.wikipedia.org/wiki/List_of...

—— Open Mind is MIT Press’s Diamond Open Access Cognitive Science journal. We now have our first Impact Factor: 2.9 as of June 2026 Submit your papers to Open Mind direct.mit.edu/opmi

Second preprint, with Elizabeth Bonawitz, one of my favorite studies children search the hypothesis space more broadly and variably than adults. osf.io/preprints/ps...

We are hosting a special issue on "Is There More to Consciousness Than Computation?" in @nconsc.bsky.social. Full information here: academic.oup.com/nc/pages/cal... With @krysdolega.bsky.social and Yair Pinto.

Good morning #OHBM26 Interested in transient functional networks - i.e. with millisecond precision - and their role in #decisionmaking and #attention ? Then come to see my poster and we can chat concepts, cognition, caveats and co. Poster 3063 Wed & Thu 12:45-13:45 #MEG #oscillation

new paper, #NeuroAI 📣📜 Can measured cortical organization be used as an inductive bias for artificial recurrent neural networks? In this work, we ask whether cortical geometry, wiring, and function can push RNNs learn. Not as metaphor, but as measurable structure! 1/n🧵👇 arxiv.org/abs/2606.14975

RFK Jr demanded the *retraction* of a 20 yr Danish study that showed aluminums in vaccines do not cause health issues. The journal REFUSED saying the study has been peer-reviewed and found to have no significant flaws. Imagine asking science to un-science itself.

What matters most for childhood brain organization? We analyzed 649 variables. The answer: Socioeconomics (SES); with brain patterns pointing at sleep & stress as drivers. Even brain-IQ associations were better explained by SES. In Science today: www.science.org/doi/10.1126/...

Thanks to @agreco.bsky.social and Jannik Reichert for organizing the #NeuroAISymposium2026 and inviting us, this was a lot of fun, hope this series continues in the next years!

Our new paper is out this week in Nature Neuroscience! www.nature.com/articles/s41... We built a BCI that works with the brain's natural geometry — and we found that people could learn to play a video game with their brains in <1 hr of training. This efficiency is groundbreaking & here's why:

Doing formal/computational theory in a multidisciplinary intersection is wonderful because neuroscientists say it’s not biological enough, it’s too theoretical for AI folks, too applied for theoretical computer scientists, not empirical enough for cognitive scientists, & too formal for philosophers💫

NEW PAPER. Why do larger networks train better? "Because they contain more candidate *sub*networks that can learn the task" → lottery tickets This popular explanation uses an appealing but misleading metaphor🧵 We propose an intuitive alternative grounded in theory: escape dimensions

New paper with @doctorjosh.bsky.social, @drmichaellevin.bsky.social, and co-first author Vaibhav Pai. This one is a doozy - in it we show that we can modulate the degree of emergent, multicellular organization in a Xenobot with an exogenous compound (ATP). 1/N www.biorxiv.org/content/10.6...

1/11 Happy to share our TICS paper on using the flexibility of one of the most basic cognitive functions, perception, to understand one of the most complex cognitive dysfunctions, psychiatric conditions (also my first formal work in computational psychiatry 🎉) 📄: www.cell.com/trends/cogni... 🧵 : 👇

“My interest isn't really open science; it's high-quality, rigorous science..often that happens to be open.” Sense from @casey.greenelab.com that aligns with how serious scientists see research and the idea it should be "open as possible, closed as necessary" www.springernature.com/gp/researche...

Looking forward to giving a talk tomorrow at the Computational Cognitive Science Colloquium at TU Darmstadt! 🗣️ The informational architecture of predictive processing in intelligent systems 🧠🤖 👇 www.tu-darmstadt.de/cogsci/ccs_n... #NeuroAI #CompCogSci #Neuroskyence

Meet Eyewire II: a new connectomic resource for the mouse retina. ~1 mm² of retina at nanometer resolution, with synapses and circuits of ca. 100,000 neurons, plus visual responses from the ~400 neurons shown in the video! Preprint: doi.org/10.64898/202... Data: eyewire.ai 🧠📈 🧠💻🧪 #VisionScience

New results! The prefrontal cortex uses a compact, low-dimensional “coding space”. Information can be captured in just 3–6 dimensions. The brain compresses what we remember. Low-dimensional prefrontal representations of objects during working memory doi.org/10.64898/202... #neuroscience

The cortex is not best understood as a set of fully separate, self-contained modules. Instead, regions work together in a more integrated, network-like way. Brain-wide distributed processing underlying natural vision and audition doi.org/10.64898/202... #neuroscience

I recently heard people calling the new generation graduating thanks to AI “Summa Cum Claude” and it’s as funny as it is terrifying 🎓💀🤖

yesterday we launched the Gemma 4 12B model and reception by the community has been great! 🎉 did you know it has vision and audio capabilities, but it doesn't have the vision and audio dedicated encoders How? we call this architecture Encoder Free newsletter.maartengrootendorst.com/p/a-visual-g...

@manybabies.org in Nature! www.nature.com/articles/d41...

New MEG preprint out! Huge congrats to postdoc Rohit Yadav. We (among other things…) decoded explore-exploit decisions in the human brain, revealing the lateral frontopolar cortex kicks off the shift to exploration hundreds of milliseconds before the choice itself. www.biorxiv.org/content/10.6...

Task diversity is supposedly key to generalization in RL. But what does it do to continual RL, where agents face one new task distribution after another? We find that past a point, more diversity actually inhibits continual reinforcement learning 🧵

Arguing if the brain is really a computer sounds a bit silly but the analogy is as old as the computer itself

Neuropixels + Optogenetics = Neuropixels Opto Combining high-resolution electrophysiology and optogenetics. 960 sites, 28 emitters, 2 colors. Today in @natmethods.nature.com doi.org/10.1038/s415... Thanks to @wellcometrust.bsky.social, @alleninstitute.org, @hhmijanelia.bsky.social & team.

First lab posters! 🤯

What is the relationship between memorization and generalization in AI? Is there a fundamental tradeoff? In infinitefaculty.substack.com/p/memorizati... I’ve reviewed some of the evolving perspectives on memorization & generalization in machine learning, from classic perspectives through LLMs.

Am I in a minority in having a broad view of what "computation" means? Example: to me, low-pass filtering is a computation. It can have analog or digital implementation. Some seem to insist that "computation" means "digital implementation". But if so, doesn't "analog computer" become an oxymoron?

Brain-wide distributed processing underlying natural vision and audition www.biorxiv.org/content/10.6... "prefrontal cortex coordinated brain dynamics through synergistic interactions independently of sensory modality and previous sensory experience"

🔴 New results 🔴 How distributed is the brain-wide network that is recruited for perception? 👁️👂🧠 Thrilled to share our new results on brain-wide distributed processing underlying natural vision and audition! Link and 🧵 👇

Indeterminism and the physics of open neural dynamics www.cell.com/trends-open/... - the brain is not a deterministic system

We’re hosting a hybrid symposium on Neurophysiological mechanisms supporting attention and prediction. 📅 11 June 2026, 09:00–14:00 ECT 📍 Psychology Department, University of Oslo + online www.uio.no/ritmo/englis... All welcome! @unioslo-ritmo.bsky.social @uio.no

Ever wondered why we fixate some locations longer than others? We initially thought that it had to do with recognition complexity, and were quite wrong. In Nature Neuroscience we present an alternative take: the need to memorise governs fixation durations.

"Our work shows that these sophisticated behaviours can emerge from the modulation of self-sustained oscillations coupled by diffusion, providing a physically grounded mechanism for information processing in non-neural organisms" Physarum polycephalum, my favorite non-neural intelligent system 💛

🚀 PhD position in #NeuroAI & neurodevelopment 🚀 Co-supervised by Sarah Lippé and myself, to investigate visual processing & cognition abnormalities in children with neurodevelopmental disorders in a neuroAI framework. Full project details and how to apply here: tinyurl.com/kbuyntpn 🧠🤖 📈

beautiful perspective article on the unification of a science of intelligence across physics, neuroscience and AI! #NeuroAI 👇👇👇

kids are just adults tripping

Now out in Nature Neuroscience: "Fixation duration on natural scenes is explained by memory encoding not processing demand". www.nature.com/articles/s41... Our eyes don't linger because recognition is hard; they linger to remember. Let me take you on a quick tour. 🧵

Complexity is ubiquitous in the world around us. But different stimuli are complex for different reasons. Does the mind nevertheless represent a ‘unified’ notion of complexity across domains? In @nathumbehav.nature.com, @chazfirestone.bsky.social & I show the answer is: yes! osf.io/preprints/ps...

We've updated the preprint of our Naturalistic Computational Cognitive Science paper (arxiv.org/abs/2502.20349) — we've tried to clarify and streamline the arguments, and added some new examples: 1/5